Direct osteolysis induced by metastatic murine melanoma cells: role of matrix metalloproteinases

Osteosarcoma and Metastasis Associated Bone Degradation-A Tale of Osteoclast and Malignant Cell Cooperativity

Cancer-induced bone degradation is part of the pathological process associated with both primary bone cancers, such as osteosarcoma, and bone metastases originating from, e.g., breast, prostate, and colon carcinomas. Typically, this includes a cancer-dependent hijacking of processes also occurring during physiological bone remodeling, including osteoclast-mediated disruption of the inorganic bone component and collagenolysis. Extensive research has revealed the significance of osteoclast-mediated bone resorption throughout the course of disease for both primary and secondary bone cancer. Nevertheless, cancer cells representing both primary bone cancer and bone metastasis have also been implicated directly in bone degradation. We will present and discuss observations on the contribution of osteoclasts and cancer cells in cancer-associated bone degradation and reciprocal modulatory actions between these cells. The focus of this review is osteosarcoma, but we will also include relevant observations from studies of bone metastasis. Additionally, we propose a model for cancer-associated bone degradation that involves a collaboration between osteoclasts and cancer cells and in which both cell types may directly participate in the degradation process.

Soluble RANKL is physiologically dispensable but accelerates tumour metastasis to bone

Receptor activator of NF-κB ligand (RANKL) is a multifunctional cytokine known to affect immune and skeletal systems, as well as oncogenesis and metastasis^1-4. RANKL is synthesized as a membrane-bound molecule, and cleaved into its soluble form by proteases^5-7. As the soluble form of RANKL does not contribute greatly to bone remodelling or ovariectomy-induced bone loss⁸, whether soluble RANKL has a role in pathological settings remains unclear. Here we show that soluble RANKL promotes the formation of tumour metastases in bone. Mice that selectively lack soluble RANKL (Tnfsf11^ΔS/ΔS)^5-7,9 have normal bone homoeostasis and develop a normal immune system but display markedly reduced numbers of bone metastases after intracardiac injection of RANK-expressing melanoma and breast cancer cells. Deletion of soluble RANKL does not affect osteoclast numbers in metastatic lesions or tumour metastasis to non-skeletal tissues. Therefore, soluble RANKL is dispensable for physiological regulation of bone and immune systems, but has a distinct and pivotal role in the promotion of bone metastases.

Bone metastasis-related signaling pathways in breast cancers stratified by estrogen receptor status

Background: Breast cancer bone metastasis (BCBM)-specific genes have been reported without considering biological differences based on estrogen receptor (ER) status. The aims of this study were to identify BCBM-specific genes using our patient dataset and validate previously reported BCBM-specific genes, and to determine whether ER-status-related biological differences matter in identification of BCBM-specific genes.

Methods: We used Affymetrix GeneChips to analyze 365 primary human epidermal growth factor receptor 2 (HER2)-negative invasive breast cancer specimens. Genes that were differentially expressed between patients who developed bone metastasis and those who developed non-bone metastasis were identified using Cox proportional hazards model, and differential expression of gene sets was assessed using gene set analysis. We performed gene set analysis to determine whether biological function associated with bone metastasis were different by ER status using 2,246 functionally annotated gene sets assembled from Gene Ontology data base.

Results: Among 16,712 probe sets, 592 were overexpressed in the bone metastasis cohort compared to the non-bone-metastasis cohort (false discovery rate ≤ 0.05). However, no BCBM-specific genes met our significance tests when the cancers were stratified by ER status. In ER-positive and ER-negative breast cancers, 151 and 125 gene sets, respectively, were overexpressed for BCBM and the majority of BCBM-related pathways were different. Of significant gene sets, only 13 gene sets were overlapped between ER-positive and -negative cohorts.

Conclusion: ER-positive and ER-negative breast cancers have different biological pathways in BCBM development. We have yet to explore BCBM-related biomarkers and targets considering the biological features associated with BCBM depending on the ER status.

Preparation of tetrazine-containing [2 + 1] complexes of 99mTc and in vivo targeting using bioorthogonal inverse electron demand Diels–Alder chemistry

A new ^99mTc-labelled tetrazine for targeted imaging using bioorthogonal chemistry was developed and evaluated in vivo using a trans-cyclooctene derived bisphosphonate targeting regions of high bone turnover and bone lesions.

Synchronous disruption of anisotropic arrangement of the osteocyte network and collagen/apatite in melanoma bone metastasis

Cancer metastasis to bones increases the risk of fragility fracture by altering bone metabolism and disrupting bone structure. Osteocytes, which organize a dense network that is closely linked with the circumambient matrix, play a key role in regulation of bone microstructure and material properties. The aim of this study was to elucidate the influence of cancer metastasis on the organization of the osteocyte network and collagen/biological apatite (BAp) microstructure in the context of osteocyte/matrix coupling. Using a mouse model intracardially injected with B16F10 melanoma cells or vehicle, the geometric and metabolic changes to osteocytes were analyzed by nano-computed tomography (nano-CT) and histology, and the alignment of collagen fibrils and BAp was analyzed by birefringence measurement and microbeam-X-ray diffraction, respectively. The material properties of bones were further analyzed with nanoindentation method. These experiments revealed that the osteocyte network was markedly disorganized in cancer-bearing bone tissues. The osteocytes showed a variety of residing states in the lacunae; some lacunae were osteolytic while some were replete with immature matrix, suggesting significant disruption in osteocyte/matrix coupling. Collagen/BAp microstructure was also disorganized in cancer-bearing bones as observed by significant decreases in the preferential alignment of both collagen fibrils and BAp; the latter was further shown to be significantly correlated with Young's modulus. The present study revealed that the disruption in the arrangement of the osteocyte network and collagen/BAp microstructure and the deterioration of mechanical function occurred synchronously during cancer bone metastasis.

Passive Entrapment of Tumor Cells Determines Metastatic Dissemination to Spinal Bone and Other Osseous Tissues

During the metastatic process tumor cells circulate in the blood stream and are carried to various organs. In order to spread to different organs tumor cell—endothelial cell interactions are crucial for extravasation mechanisms. It remains unclear if tumor cell dissemination to the spinal bone occurs by passive entrapment of circulating tumor cells or by active cellular mechanisms mediated by cell surface molecules or secreted factors. We investigated the seeding of three different tumor cell lines (melanoma, lung and prostate carcinoma) to the microvasculature of different organs. Their dissemination was compared to biologically passive microbeads. The spine and other organs were resected three hours after intraarterial injection of tumor cells or microbeads. Ex vivo homogenization and fluorescence analysis allowed quantification of tumor cells or microbeads in different organs. Interestingly, tumor cell distribution to the spinal bone was comparable to dissemination of microbeads independent of the tumor cell type (melanoma: 5.646% ± 7.614%, lung: 6.007% ± 1.785%, prostate: 3.469% ± 0.602%, 7 μm beads: 9.884% ± 7.379%, 16 μm beads: 7.23% ± 1.488%). Tumor cell seeding differed significantly between tumor cells and microbeads in all soft tissue organs. Moreover, there were significant differences between the different tumor cell lines in their dissemination behaviour to soft tissue organs only. These findings demonstrate that metastatic dissemination of tumor cells to spinal bone and other osseous organs is mediated by passive entrapment of tumor cells similar to passive plugging of microvasculature observed after intraarterial microbeads injection.

Three-Dimensional in Vitro Model to Study Osteobiology and Osteopathology

The bone is an amazing organ that grows and remodels itself over a lifetime. It is generally accepted that bone sculpting in response to stress and force is carried out by groups of cells contained within bone multicellular units that are coordinated to degrade existing bone and form new bone. Because of the nature of bone and the extensiveness of the skeleton, it is difficult to study bone remodeling in vivo. On the other hand, because the bone contains a complex environment of many cell types, is it possible to study bone remodeling in vitro? We propose that one can at minimum study the interaction between osteoblasts (bone formation) and osteoclasts (bone degradation) in a three dimensional (3D) "bioreactor". Furthermore, one can add bone degrading metastatic cancer cells, and study how they contribute to and take part in the bone degradation process. We have primarily cultured and differentiated MC3T3-E1 osteoblasts for long periods (2-10 months) before addition of bone marrow osteoclasts and/or metastatic (MDA-MB-231), metastasis suppressed (MDA-MB-231BRMS1) or non-metastatic (MCF-7) breast cancer cells. In the co-culture of osteoblasts and osteoclasts there was clear evidence of matrix degradation. Loss of matrix was also evident after co-culture with metastatic breast cancer cells. Tri-culture permitted an evaluation of the interaction of the three cell types. The 3D system holds promise for further studies of cancer dormancy, hormone, and cytokine effects and matrix manipulation.

Advances in the biology of bone metastasis: how the skeleton affects tumor behavior

It is increasingly evident that the microenvironment of bone can influence the cancer phenotype in many ways that favor growth in bone. The ability of cancer cells to adhere to bone matrix and to promote osteoclast formation are key requirements for the establishment and growth of bone metastases. Several cytokine products of breast cancers (e.g. PTHrP, IL-11, IL-8) have been shown to act upon host cells of the bone microenvironment to promote osteoclast formation, allowing for excessive bone resorption. The increased release of matrix-derived growth factors, especially TGF-β, acts back upon the tumor to facilitate further tumor expansion and enhance cytokine production, and also upon osteoblasts to suppress bone formation. This provides a self-perpetuating cycle of bone loss and tumor growth within the skeleton. Other contributing factors favoring tumor metastasis and colonization in bone include the unique structure and stiffness of skeletal tissue, along with the diverse cellular composition of the marrow environment (e.g. bone cells, stromal fibroblasts, immune cells), any of which can contribute to the phenotypic changes that can take place in metastatic deposits that favor their survival. Additionally, it is also apparent that breast cancer cells begin to express different bone specific proteins as well as proteins important for normal breast development and lactation that allow them to grow in bone and stimulate bone destruction. Taken together, these continually emerging areas of study suggest new potential pathways important in the pathogenesis of bone metastasis and potential areas for targeting therapeutics.

Future directions of bone-targeted therapy for metastatic breast cancer

Bone is the most common metastatic site for breast cancer, and bone metastases can cause pain as well as risk of pathological fractures. Emerging treatments for metastatic bone disease have arisen from advances in our understanding of the unique cellular and molecular mechanisms that contribute to bone metastasis. The interaction between tumor cells and the bone microenvironment results in a 'vicious cycle' that increases both bone destruction and tumor burden. The tumor secretes factors, such as parathyroid hormone-related peptide, that stimulate osteoclastogenesis. Similarly, the bone stroma produces growth factors, such as transforming growth factor β, that promote tumor growth in bone. Therapeutic targeting of these microenvironmental factors is under intensive investigation. Other attractive therapeutic targets include signaling molecules, such as receptor activator of nuclear factor κB ligand, Src kinase, and cathepsin K, all of which regulate osteoclast function, and chemokine receptor 4, which is involved in the homing of tumor cells to bone. In this Review, we describe the progress and future directions of novel bone-targeted therapies that may reduce or prevent destructive bone metastasis from breast cancer. Novel modalities for predicting and monitoring treatment response will also be described.

Bone-resorbing cells in multiple myeloma: osteoclasts, myeloma cell polykaryons, or both?

Myeloma bone disease (MBD) leads to progressive destruction of the skeleton and is the most severe cause of morbidity in multiple myeloma. Its pathogenetic mechanisms are not fully understood, though the current evidence points to osteoclast (OC) hyperactivity coupled with defective osteoblast function unable to counteract bone resorption. OCs are generated in bone marrow by myeloid progenitors through increased levels of receptor activator of nuclear factor kappaB ligand and M-CSF, whose intracellular pathways propagate signals that activate sequential transcription factors, resulting in the production of major OC enzymes that drive specific functions such as acidification and degradation of the bone matrix. Osteolytic lesions, however, are not characterized by massive OC content, whereas malignant plasma cells, which are usually present in a high number, may occur as large multinucleated cells. The possibility that myeloma cells fuse and generate polykaryons in vivo is suggested by the in vitro formation of multinuclear cells that express tartrate-resistant acid phosphatase and produce pits and erosive lacunae on experimental osteologic substrates. Further, the detection in vivo of polykaryons with chromosome translocations typical of myeloma cells lends support to the view that myeloma polykaryons may act as functional OCs and participate in the skeletal destruction by resorbing bone.

A role for leukemia inhibitory factor in melanoma-induced bone metastasis

Melanoma is commonly associated with multi-organ metastasis, and bone is a frequent metastatic site for melanoma. However, the mechanism responsible for such melanoma-induced bone metastasis is still poorly understood. In the present study, the intracardiac inoculation of leukemia inhibitory factor (LIF)-producing human melanoma-derived cells (SEKI) developed osteolytic bone destruction in male BALB/cA-nu/nu nude mice. To elucidate the role of LIF in melanoma-induced osteolysis, cells were prepared in which the expression of LIF was reduced using a siRNA technique from the parent SEKI cells. Osteoclastogenesis was induced in the co-culture of LIF and/or SEKI cells with osteoblastic stromal cells in vitro, whereas the LIF-reduced SEKI cells did not induce osteoclastogenesis. The intracardiac inoculation of LIF-reduced SEKI cells resulted in a significant reduction in the incidence and number of bone metastasis in comparison to those in the mice inoculated with the parent SEKI cells. The expression of LIF was found in seven of nine human melanoma-derived cell lines, suggesting that LIF expression is a universal event in melanoma. These findings suggest that a potential role for LIF in the melanoma-induced bone metastasis possibly through the stimulation of osteoclastogenesis. LIF might therefore be a potentially effective drug target in the treatment of bone metastasis in melanoma.

Metastatic breast cancer cells colonize and degrade three-dimensional osteoblastic tissue in vitro

Metastatic breast cancer cells (BCs) colonize a mineralized three-dimensional (3D) osteoblastic tissue (OT) grown from isolated pre-osteoblasts for up to 5 months in a specialized bioreactor. Sequential stages of BC interaction with OT include BC adhesion, penetration, colony formation, and OT reorganization into "Indian files" paralleling BC colonies, heretofore observed only in authentic pathological cancer tissue. BCs permeabilize OT by degrading the extra-cellular collagenous matrix (ECM) in which the osteoblasts are embedded. OT maturity (characterized by culture age and cell phenotype) profoundly affects the patterns of BC colonization. BCs rapidly form colonies on immature OT (higher cell/ECM ratio, osteoblastic phenotype) but fail to completely penetrate OT. By contrast, BCs efficiently penetrate mature OT (lower cell/ECM ratio, osteocytic phenotype) and reorganize OT. BC colonization provokes a strong osteoblast inflammatory response marked by increased expression of the pro-inflammatory cytokine IL-6. Furthermore, BCs inhibit osteoblastic bone formation by down-regulating synthesis of collagen and osteocalcin. Results strongly suggest that breast cancer disrupts the process of osteoblastic bone formation, in addition to upregulating osteoclastic bone resorption as widely reported. These observations may help explain why administration of bisphosphonates to humans with osteolytic metastases slows lesion progression by inhibiting osteoclasts but does not bring about osteoblast-mediated healing.

Cathepsin K in melanoma invasion

Cathepsin K (catK) is a lysosomal cysteine protease with strong collagenolytic activity that mediates bone resorption in osteoclasts. Recently, catK expression has been reported in skin and lung fibroblasts, which suggests a role in maintaining homeostasis of the extracellular matrix outside of bone. Matrix degradation is a pivotal step in tumor invasion and metastasis. As other proteases, in particular matrix metalloproteinases and some cathepsins, but not catK, have been described to mediate melanoma invasion, we studied catK in melanoma. Immunostaining revealed strong catK expression in most primary melanomas and all cutaneous melanoma metastases. Melanocytic nevi also demonstrated catK expression, but it was less intense than in melanomas. Melanoma lines express both the pro- and the active form of catK and internalize extracellular collagen into lysosomes. Inhibition of catK greatly reduced melanoma cell invasion through Matrigel basement membrane matrix and increased detection of internalized collagen. We suggest that catK may play an important role in melanoma invasion and metastasis by mediating intracellular degradation of matrix proteins after phagocytosis. Clinical use of catK inhibitors, a class of medication currently in clinical trials for the treatment of osteoporosis, may be a promising avenue for the treatment of melanoma.

Early detection and quantification of murine melanoma bone metastases with magnetic resonance imaging

OBJECTIVE

Bone metastases occur in approximately 80% of patients with advanced cancer and cause significant morbidity. There are currently no established means by which to identify the early growth of micro-metastatic cells or their effects on bone at a time when curative therapy might be initiated. We postulated that high-resolution magnetic resonance imaging (MRI) could detect and quantify the growth and destructive effects of bone micrometastases.

DESIGN

Using a mouse model for metastasis of malignant melanoma, we have examined the ability of MRI to quantify cortical bone destruction and the percentage of the medullary cavity occupied by tumour, trabecular bone, and marrow. The results from MRI were compared to histomorphometry (the reference standard) and to radiographs.

RESULTS

In vivo gradient-echo and spin-echo MRI demonstrated that metastatic melanoma replaced the marrow space but that the cortical bone integrity was preserved (P < or = 0.001). The smallest detectable micrometastasis had an area of 0.323 mm(2). In contrast, we observed no trends after quantifying the radiograph data.

CONCLUSION

These approaches delineated the limits of MRI in its ability to quantify tumour burden and the effect on bone in this model. Given the increasing use of MRI as a non-invasive clinical diagnostic method, the present findings may be applicable in detecting bone metastases in the clinical setting at an early and potentially treatable stage.

Osteoclast nuclei of myeloma patients show chromosome translocations specific for the myeloma cell clone: a new type of cancer-host partnership?

A major clinical manifestation of bone cancers is bone destruction. It is widely accepted that this destruction is not caused by the malignant cells themselves, but by osteoclasts, multinucleated cells of monocytic origin that are considered to be the only cells able to degrade bone. The present study demonstrates that bone-resorbing osteoclasts from myeloma patients contain nuclei with translocated chromosomes of myeloma B-cell clone origin, in addition to nuclei without these translocations, by using combined FISH and immunohistochemistry on bone sections. These nuclei of malignant origin are transcriptionally active and appear fully integrated amongst the other nuclei. The contribution of malignant nuclei to the osteoclast population analysed in this study was greater than 30%. Osteoclast-myeloma clone hybrids contained more nuclei than normal osteoclasts and their occurrence correlated with the proximity of myeloma cells. Similar hybrid cells were generated in myeloma cell-osteoclast co-cultures, as revealed by tracing myeloma nuclei using translocations, bromo-deoxyuridine, or the Y chromosome of male myeloma cells in female osteoclasts. These observations indicate that hybrid cells can originate through fusion between myeloma cells and osteoclasts. In conclusion, malignant cells contribute significantly to the formation of bone-resorbing osteoclasts in multiple myeloma. Osteoclast-myeloma clone hybrids reflect a previously unrecognized mechanism of bone destruction in which malignant cells participate directly. The possibility that malignant cells corrupt host cells by the transfer of malignant DNA may have been underestimated to date in cancer research.

Enhanced metastatic dissemination to multiple organs by melanoma and lymphoma cells in timp-3-/- mice

Identifying versatile inhibitors of metastasis that operate in multiple sites against distinct cancer cell types is important for designing novel therapeutics for metastasis. We show that multiple tissues of timp-3-/- mice are more susceptible to metastatic colonization. Overall, a 5-14-fold increase in liver and kidney colonization occurred by EL-4 lymphoma cells, and a twofold increase upon targeting B16F10 melanoma cells to the bone or lung of timp-3-/- mice. There was a general lack of macrophage or neutrophil localization to metastases in the liver, kidney and lung, and of osteoclasts to bone in both genotypes. Analysis of lung showed that proliferation or angiogenesis were unaltered within the metastatic colonies. Lung-trap assays revealed that initial tumor cell trapping was similar in the lung vasculature of timp-3-/- and wild-type mice. However, more tumor cells were found in timp-3-/- lungs at 48 and 96 h after tumor cell injection indicating more efficient extravasation and initial proliferation. Activation of pro-MMP-2 was greater in timp-3-/- lungs at these time points. These data demonstrate TIMP-3 functions to inhibit metastatic dissemination of diverse cancer cells to multiple organs. TIMP-3 regulates MMP-2 activation to limit tumor cell extravasation and subsequent colonization of the lung, without augmenting inflammatory cell response.

Regulation of cancer cell migration and bone metastasis by RANKL

Bone metastases are a frequent complication of many cancers that result in severe disease burden and pain. Since the late nineteenth century, it has been thought that the microenvironment of the local host tissue actively participates in the propensity of certain cancers to metastasize to specific organs, and that bone provides an especially fertile 'soil'. In the case of breast cancers, the local chemokine milieu is now emerging as an explanation for why these tumours preferentially metastasize to certain organs. However, as the inhibition of chemokine receptors in vivo only partially blocks metastatic behaviour, other factors must exist that regulate the preferential metastasis of breast cancer cells. Here we show that the cytokine RANKL (receptor activator of NF-kappaB ligand) triggers migration of human epithelial cancer cells and melanoma cells that express the receptor RANK. RANK is expressed on cancer cell lines and breast cancer cells in patients. In a mouse model of melanoma metastasis, in vivo neutralization of RANKL by osteoprotegerin results in complete protection from paralysis and a marked reduction in tumour burden in bones but not in other organs. Our data show that local differentiation factors such as RANKL have an important role in cell migration and the tissue-specific metastatic behaviour of cancer cells.

Human breast cancer cell-mediated bone collagen degradation requires plasminogen activation and matrix metalloproteinase activity

BACKGROUND: Breast cancer cells frequently metastasize to the skeleton and induce extensive bone destruction. Cancer cells produce proteinases, including matrix metalloproteinases (MMPs) and the plasminogen activator system (PAS) which promote invasion of extracellular matrices, but whether these proteinases degrade bone matrix is unclear. To characterize the role that breast cancer cell proteinases play in bone degradation we compared the effects of three human breast cancer cell lines, MDA-MB-231, ZR-75-1 and MCF-7 with those of a normal breast epithelial cell line, HME. The cell lines were cultured atop radiolabelled matrices of either mineralized or non-mineralized bone or type I collagen, the principal organic constituent of bone. RESULTS: The 3 breast cancer cell lines all produced significant degradation of the 3 collagenous extracellular matrices (ECMs) whilst the normal breast cell line was without effect. Breast cancer cells displayed an absolute requirement for serum to dissolve collagen. Degradation of collagen was abolished in plasminogen-depleted serum and could be restored by the addition of exogenous plasminogen. Localization of plasmin activity to the cell surface was critical for the degradation process as aprotinin, but not alpha2 antiplasmin, prevented collagen dissolution. During ECM degradation breast cancer cell lines expressed urokinase-type plasminogen activator (u-PA) and uPA receptor, and MMPs-1, -3, -9,-13, and -14. The normal breast epithelial cell line expressed low levels of MMPs-1, and -3, uPA and uPA receptor. Inhibitors of both the PAS (aprotinin and PA inhibitor-1) and MMPs (CT1166 and tisue inhibitor of metalloproteinase) blocked collagen degradation, demonstrating the requirement of both plasminogen activation and MMP activity for degradation. The activation of MMP-13 in human breast cancer cells was prevented by plasminogen activator inhibitor-1 but not by tissue inhibitor of metalloproteinase-1, suggesting that plasmin activates MMP-13 directly. CONCLUSIONS: These data demonstrate that breast cancer cells dissolve type I collagen and that there is an absolute requirement for plasminogen activation and MMP activity in the degradation process.

Different tumors in bone each give rise to a distinct pattern of skeletal destruction, bone cancer-related pain behaviors and neurochemical changes in the central nervous system

Pain is the most common presenting symptom in patients with bone cancer and bone cancer pain can be both debilitating and difficult to control fully. To begin to understand the mechanisms involved in the generation and maintenance of bone cancer pain, we implanted 3 well-described murine tumor cell lines, 2472 sarcoma, B16 melanoma and C26 colon adenocarcinoma into the femur of immunocompromised C3H-SCID mice. Although each of the tumor cell lines proliferated and completely filled the intramedullary space of the femur within 3 weeks, the location and extent of bone destruction, the type and severity of the pain behaviors and the neurochemical reorganization of the spinal cord was unique to each tumor cell line injected. These data suggest that bone cancer pain is not caused by a single factor such as increased pressure induced by intramedullary tumor growth, but rather that multiple factors are involved in generating and maintaining bone cancer pain.

Morphological, histomorphometric, and microstructural alterations in human bone metastasis from breast carcinoma

Bone is one of the most common sites of breast cancer metastasis. Metastases are often associated with bone destruction and are a major cause of morbidity. We examined structural bone changes induced by metastatic tumor in bone biopsies from 33 patients with metastatic breast carcinoma (20 from patients with pathological femoral fracture and 13 with no fracture) and 20 normal controls. In all metastatic biopsies bone remodeling was shown to be tumor volume-dependent. Bone resorption and bone formation were biphasic with both increasing at earlier stages of metastatic bone disease and decreasing later on. A comparison of patients with fracture and no fracture did not reveal statistically significant differences in the extent of bone destruction or trabecular thinning. Bone histomorphometry showed limited ability to explain the higher bone volume loss in fracture patients (decreases of 42% and 25%, respectively, in fracture and nonfracture patients compared with controls). However, changes in bone quality, including increased disconnectivity and decreased connectivity, as evaluated by node-strut analysis, suggested that there were more structural changes in the fracture compared with the nonfracture group. The nonfracture group included six patients with no radiological evidence of bone metastasis (occult metastasis). They showed a higher tumor volume and a twofold lower eroded surface compared with the rest of the group. The decrease in bone volume (14% lower than controls) was below the limit of X-ray detection. Because we observed no increase in osteoclast-related parameters and no correlation between osteoclast surface and eroded surface, we believe that, in occult metastasis, osteoclastic bone resorption is not an important factor in overall bone resorption. Quantitatively, the eroded surface in direct contact with tumor cells was threefold higher than the osteoclast surface in occult metastasis, whereas the rest of the metastatic group (27 of 33) showed predominantly osteoclast-mediated eroded surface. Node-strut analysis on occult metastasis revealed a significant increase in disconnectivity without a concomitant significant decrease in bone volume and trabecular thinning. We conclude that, in occult metastasis, bone resorption may be more osteoclast-independent and other mechanisms involving the tumor cells may be more prevalent.

A matrix metalloproteinase inhibitor, batimastat, retards the development of osteolytic bone metastases by MDA-MB-231 human breast cancer cells in Balb C nu/nu mice

Bone resorption is a dominant feature of many bone metastases and releases factors from the bone matrix that can promote the expression of the metastatic phenotype in cancer cells. Since proteolytic enzymes, including matrix metalloproteinases (MMPs) contribute to bone destruction by metastatic tumour cells and host cells, we have examined the effect of a MMP inhibitor, batimastat, on the ability of MDA-MB-231 cells to degrade bone in vitro and to form bone metastases in BalbC nu/nu mice. In vitro, the neoplastic cells produced MMP-2 and MMP-9, degraded [3H]-proline-labelled osteoblast matrices, and formed resorption pits in cortical bone. These phenomena were inhibited by < or = 20 microM batimastat. To induce vertebral and long bone metastases in vivo, 1x10(5) MDA-MB-231 cells were injected into the arterial circulation of BalbC nu/nu mice. Test groups were also given 30 mg/kg batimastat intraperitoneally (i.p.). After 21 days, the long bone metastases were characterised by a 67% reduction of metaphyseal medullary bone and complete replacement of marrow by tumour. In tumour-bearing mice that had been treated with 30 mg/kg batimastat i.p., the tumour volume decreased 8-fold, osteolysis was inhibited by 35%, and replacement of the bone marrow by tumour was inhibited by 65%. Similar effects were observed in the vertebral metastases. These data provide evidence that MDA-MB-231 cells can degrade osteoblast matrices and mineralised bone in vitro and support the hypothesis that MMPs are involved in the pathogenesis of osteolytic bone metastases in vivo. They demonstrate that an agent which inhibits proteolysis can retard the development of osteolytic bone metastases in this model.

Pathophysiologic interactions in skeletal metastasis

BACKGROUND

This review summarizes evidence that the formation of bone metastases is the result of multiple synergistic cellular and molecular interactions between metastatic cells and the unique microenvironment in bone.

METHODS

Molecular technologies have been used to detect cancer cells in bone and to define their genotypic and phenotypic properties. Bone organ cultures have been employed to analyze the ability of tumor cells to modulate bone resorption and to study the effects of resorption products on the phenotypic properties of cancer cells. Experimental models of bone metastasis provide the ability to examine the effects of modulating specific host or tumor properties in vivo by quantifying their effects on the formation of bone tumors.

RESULTS

By means of the blood stream, cells from many common neoplasms seed bone marrow as an early clinical event. The subsequent growth of these cells into clinically significant metastatic lesions is associated with their ability to stimulate bone resorption through osteoclasts and macrophages or through a direct action on bone. In turn, the products of bone resorption, which include matrix-derived growth factors, act on the tumor cells to stimulate the expression of properties that promote their metastatic competence. These include the induction of integrin adhesion molecules, the stimulation of cell motility and chemotaxis, the enhanced expression of matrix metalloproteinases, and the stimulation of tumor cell growth.

CONCLUSIONS

The interdependency of tumor cells and bone was recognized by Steven Paget over 100 years ago, and it provides a rational basis for the development of current therapeutic strategies against bone metastasis.

Osteoclasts are required for bone tumors to grow and destroy bone

It has been hypothesized that bone resorption during tumor osteolysis is performed by osteoclasts. Data supporting this hypothesis have been provided from analysis of human biopsy specimens obtained from sites of tumor osteolysis, as well as from experimentation with in vivo animal models. Experiments in this report take this concept one step further by testing the hypothesis that osteoclasts are required for bone tumors to grow and destroy bone. To test this hypothesis, the influence of an osteolytic sarcoma tumor, NCTC clone 2472 (2472), on bone was studied in animals that are osteoclast deficient (microphthalmic, strain B6C3Fe-a/a-Mitf(mi)) but whose osteoclast deficiency can be reversed following bone marrow transplantation. Femora of these mice and unaffected wild-type siblings were injected with 10(5) 2472 cells, and after 14 days the femora were analyzed by radiographic and histomorphometric analysis. Macroscopic tumor, tumor-induced osteolysis, and increased osteoclast number were noted in femora of normal mice but not in femora of osteoclast-deficient mice (p < 0.001). Bone marrow transplantation converted osteoclast-deficient mice to mice with femora that contained osteoclasts in 4 weeks. Femora of these mice were then injected with 10(5) 2472 tumor cells; after 14 days, in contrast to the findings in the original osteoclast-deficient mice, macroscopic tumor was present, tumor-induced osteolysis was noted on roentgenograms, and osteoclast number was increased when tumor-bearing limbs were compared with sham-injected limbs (p < 0.001). These data prove the hypothesis that osteoclasts are required for 2472 tumor-induced osteolysis, and they introduce the exciting possibility that osteoclasts are also required for tumors to grow in bone.