Paracrine Regulation of Osteoclast Activity in Metastatic Breast Cancer

By A. Shaw and M. J. Oursler, PhD

Bone is the most common site of breast cancer metastasis. Breast tumors in the bone marrow cavity recruit osteoclasts to degrade bone, forming localized areas where bone is considerably weakened. Metastatic breast cancer-induced bone loss is termed osteolysis. Osteolysis leads to hypercalcemia and bone fractures, resulting in considerable pain for patients with metastatic breast cancer. Elevated bone resorption by osteoclasts is responsible for tumor-induced osteolysis. The mechanism of the increase in osteoclast activity that causes osteolysis is believed to involve soluble growth factors secreted by the tumor. Recently, the importance of these paracrine influences on osteoclast activity have been recognized, and it is the purpose of this review to discuss the results of investigations into the mechanism of tumor-derived growth factor actions on osteoclasts.

Introduction

Paracrine factors are implicated as mediators of bone loss associated with tumor osteolysis in patients with metastatic breast cancer. Osteolytic lesions result from an increase in osteoclast bone resorption activity at sites adjacent to a tumor in the marrow cavity.1 Bone loss during tumor osteolysis may result from any of the following mechanisms: 1) increased formation of osteoclasts; 2) increased resorption activity by mature osteoclasts; or 3) increased survival of mature osteoclasts.

Osteolytic bone loss may result from increased formation of mature osteoclasts from osteoclast precursors. Mature osteoclasts form by differentiation from hematopoietic stem cells found in the marrow activity. This process requires direct cell-cell contact between osteoclast precursors and marrow stromal cells. Later stages in osteoclast differentiation include fusion of mononuclear osteoclast precursors to form multinucleated cells, and activation of multinucleated osteoclasts to induce adhesion to bone and secretion of bone-degrading enzymes.2

Osteolytic bone loss may result from an increase in the resorption activity of each individual osteoclast. Bone resorption activity is related to the ability of an osteoclast to adhere to bone, secrete bone-degrading lysosomal enzymes, and migrate to form more and/or larger resorption pits on bone slices. A mature osteoclast may be activated by a number of stimuli. Upon activation, an osteoclast will adhere to bone, forming a tight-sealing zone around the periphery of the bone resorption compartment, which is analogous to a secondary lysosome. Bone-degrading proteases are secreted into this compartment by fusion of lysosomal vesicles with the plasma membrane adjacent to the bone surface. Fusion of these vesicles increases the amount of membrane at this interface and this portion of the plasma membrane becomes highly convoluted, forming a structure called the ruffled border of the osteoclast.

The bone resorption compartment becomes acidified by the action of numerous proton pumps on the ruffled border. The combined action of acid and proteases degrades the bone surface enclosed by the sealing zone. An active osteoclast may detach, migrate, and reattach at another site to form multiple resorption pits. The important role that soluble growth factors play in activation of osteoclasts in metastatic breast cancer was first alluded to by a study that examined the conditioned medium of an osteolytic breast cancer cell line, MDA MB 231.3

This study revealed that 231 cell conditioned medium was capable of increasing the activity of mature osteoclasts, suggesting that the medium contained soluble factors that could act on osteoclasts. Analysis of the 231 cell conditioned medium revealed that it contained many growth factors that have been implicated as regulators of osteoclast activity. This conditioned medium was fractionated over a sizing column and used to treat osteoclasts. Some fractions stimulated, while some fractions inhibited osteoclast activity. This observation suggests that stimulatory factors overcome the actions of inhibitory factors to increase osteoclast activity. Paracrine stimulation of osteoclast activity may be a primary mechanism by which metastatic breast tumors are able to degrade bone.

Osteolytic bone loss may result from increased survival of mature osteoclasts. It is believed that mature osteoclasts are removed from the bone surface by a signal to undergo apoptosis. Soluble factors secreted by tumors may delay the apoptotic signal, allowing osteoclasts to continue with their bone resorption program for a longer period of time. Below, several paracrine factors that may be important in the development of osteolytic lesions are discussed.

OPG/RANKL/RANK

Osteoprotegerin (OPG) is a recently discovered protein that plays an important role in osteoclast formation. OPG is a soluble factor secreted by bone marrow stromal cells and osteoblasts. OPG functions as an inhibitor of osteoclast formation. OPG interferes with a critical interaction between the osteoclast precursor receptor-activator of NFkB (RANK) and its cognate ligand (RANKL). RANK is present on the plasma membrane of osteoclast precursors and mature osteoclasts, while RANKL is expressed on the plasma membrane of stromal cells. In addition, the interaction between RANKL and RANK gives rise to signaling events that inhibit apoptosis in mature osteoclasts. In this way, OPG reduces overall osteoclast numbers by blocking osteoclast formation and stimulating apoptosis of mature osteoclasts. Conversely, increased levels of RANKL are correlated with increased osteoclast numbers. One might expect that osteolytic tumors would express RANKL or induce RANKL expression. However, we and others have shown that most primary breast cancers, metastatic breast cancers, and breast cancer cell lines do not express RANKL.4 When breast cancer cells are co-cultured with bone marrow stromal cells, expression of RANKL is induced and OPG is inhibited.5 These results indicate that another factor produced by breast cancer cells within the bone marrow cavity induces RANKL expression and inhibits OPG secretion by bone marrow stromal cells.

Macrophage-Colony Stimulating Factor

Macrophage-colony stimulating factor (M-CSF) is absolutely required for osteoclast formation. M-CSF is expressed by bone marrow stromal cells and osteoblasts. The essential role of M-CSF is evident in mutant op/op mice that express nonfunctional M-CSF. These mice show a complete lack of mature osteoclasts, a condition that is reversed by infusion of M-CSF. M-CSF stimulates proliferation of early osteoclast precursors and induces migration of actively resorbing osteoclasts. Estrogen deficiency enhances M-CSF production, which results in the overall increase in osteoclast number seen in osteoporosis. M-CSF also promotes the survival of mature osteoclasts by delaying the onset of apoptosis. M-CSF and RANKL have been shown to be essential factors for osteoclast formation in vitro.6

Granulocyte-Macrophage Colony Stimulating Factor

Granulocyte-macrophage colony stimulating factor (GM-CSF) has opposing effects on osteoclast formation. Effects of GM-CSF appear to depend on the differentiation state of the target cell. GM-CSF stimulates proliferation of early osteoclast precursors, but potently inhibits late stages of osteoclast differentiation. GM-CSF inhibits formation of osteoclasts from mouse bone marrow, which contains mid-stage osteoclast precursors. In an in vivo nude mouse model of osteolysis, expression of GM-CSF declined as osteolytic lesions appeared.7 Local repression of GM-CSF to allow osteoclast formation may provide a mechanism by which an osteolytic tumor can mediate bone loss.

Tumor Necrosis Factor Alpha

Tumor necrosis factor alpha (TNF-a) is a soluble protein secreted by many bone marrow cells and is known to play a role in the development of other bone loss pathologies, including periodontitis and orthopedic implant loosening. Antibody blockade of TNF-a results in decreased osteoclast numbers and reduced pit formation activity of mature osteoclasts.8 TNF-a is secreted in large quantities by osteolytic breast cancer cell lines and by breast tumors in bone. We have used a mouse model of breast cancer osteolysis to examine the timing of TNF-a expression as it relates to the appearance of osteolytic lesions. TNF-a expression by mouse marrow cells increases as tumor size increases and as osteolytic lesions appear in the mice. We have examined the effects of TNF-a added to osteoclast precursors during differentiation. TNF-a treatment results in increased osteoclast numbers and larger osteoclasts as compared to untreated osteoclasts. These larger TNF-a-stimulated osteoclasts secrete more bone-degrading enzymes and form more resorption pits per cell than unstimulated osteoclasts. TNF-a also appears to prolong the lifespan of mature osteoclasts by inhibiting apoptosis.9 These results indicate that elevated TNF-a levels may play a role in metastatic breast cancer-induced osteolysis. TNF-a secreted by tumor cells and marrow cells may induce formation of larger osteoclasts that destroy bone at a rate faster than it can be replaced by osteoblasts.

Insulin-Like Growth Factors

Insulin-like growth factors (IGFs) are secreted by osteoblasts and stimulate proliferation of osteoblasts. Mature osteoclasts express type I IGF receptors and IGF stimulates resorption activity of mature osteoclasts in the presence of osteoblasts.10 IGF induction is responsible for growth hormone and parathyroid hormone-induced osteoclast formation.11 This suggests that IGF may contribute to osteoclast formation indirectly by inducing expression of another osteoclast-promoting growth factor. IGF is secreted by breast cancer cell lines and metastatic breast tumors in bone. IGF treatment of osteoclast precursors induces differentiation into mature osteoclasts. IGF secreted by metastatic breast tumors may contribute to elevated osteoclast activity by inducing differentiation of osteoclast precursors and stimulating resorption activity of mature osteoclasts.

Parathyroid Hormone-Related Peptide Expression

Parathyroid hormone-related peptide (PTHrP) expression is correlated with increased metastasis of breast cancer cells, which leads to increased frequency of osteolytic lesions in vivo. PTHrP is expressed by mature osteoclasts and 92% of breast tumors showing bone metastases.12 PTHrP is the main causative agent of hypercalcemia that is associated with a variety of cancers. Hypercalcemia is mediated through an endocrine mechanism involving PTH receptor-mediated effects on kidney and bone metabolism. Paracrine actions of PTHrP are dual, with different portions of the PTHrP molecule giving rise to opposing effects. PTHrP(1-34) mediates stimulatory PTH-like actions by binding to the PTH receptor expressed on osteoblasts. These indirect effects include stimulation of osteoclast formation and increased bone resorption activity. However, PTHrP(107-139) mediates inhibitory actions of PTHrP, including direct inhibition of osteoclastic bone resorption, by an unknown PTH receptor-independent mechanism.13 PTHrP expression appears to be permissive for formation of bone metastases, although its role in the induction of osteolysis remains unclear.

Interleukins

Interleukin-1 (IL-1) stimulates bone resorption primarily by prolonging survival of mature osteoclasts. IL-6 also stimulates bone resorption, but its mode of action is stimulation of osteoclast formation. IL-6 induces proliferation of osteoclast precursors and induces these precursors to commit to the osteoclast lineage. Infusion of IL-1 together with IL-6 in mice causes marked bone loss that results in hypercalcemia. Bone loss due to elevated levels of parathyroid hormone or 1, 25-dihydroxyvitamin D3 is due to the ability of these hormones to induce IL-6 secretion from osteoblasts.6 Many breast tumors express IL-1 and IL-6, which makes these factors possible mediators of bone loss associated with metastatic breast cancer.

Leukemia Inhibitory Factor

Leukemia inhibitory factor (LIF) is so named because of its ability to inhibit proliferation and induce differentiation to the macrophage line of a myeloid leukemic cell line.14 Since its discovery, a variety of systemic effects have been attributed to LIF. The overall effect of LIF on bone metabolism is to increase the rate of bone turnover by increasing both osteoblast and osteoclast activity. LIF overexpression in mice causes splenic enlargement due to the excessive proliferation of hematopoietic stem cells, the cells that give rise to osteoclasts. LIF prolongs survival of osteoclast precursors and induces the proliferation of osteoblasts. LIF expression is induced by cytokines known to induce bone loss, including TNF-a, IL-1, and IL-6.15 LIF is expressed by 78% of primary breast tumors and stimulates the proliferation of breast cancer cell lines in vitro.16 Expression of LIF by breast cancers in bone could potentially stimulate osteoclast activity to induce bone loss.

Transforming Growth Factor

Transforming growth factor (TGF-b) has biphasic effects on osteoclast activity. At high doses, TGF-b inhibits osteoclast formation. However, at low doses, TGF-b stimulates formation and survival of osteoclasts. We have shown that osteoclasts formed in the presence of TGF-b become TGF-b-dependent. Withdrawal of TGF-b from these TGF-b-dependent osteoclasts induces immediate apoptosis.17 Osteoclasts formed in the environment adjacent to a TGF-b-secreting breast tumor may become TGF-b-dependent and may survive longer than osteoclasts formed in the absence of TGF-b. Enhanced survival of TGF-b-dependent osteoclasts may provide a mechanism by which breast tumors can induce osteolysis.

Summary

Breast tumors in the bone marrow cavity can induce bone loss by inducing the formation, activity, and survival of mature osteoclasts. These effects may be due to the secretion of growth factors in the area adjacent to the tumor. RANKL, M-CSF, TNF-a, IGF, IL-1, IL-6, LIF, and TGF-b can enhance formation of mature osteoclasts by exerting effects on various stages of osteoclast differentiation. RANKL, M-CSF, TNF-a, IGF, and LIF can promote osteoclast resorption activity by increasing adhesion to bone, secretion of lysosomal proteases, and migration. M-CSF, TNF-a, IL-1, and TGF-b increase the lifespan of mature osteoclasts by delaying apoptosis. Additionally, repression of factors that block osteoclast formation, such as GM-CSF, may enable tumors to recruit osteoclasts. In order to clarify the role of tumor-derived growth factors in development of osteolytic lesions, future investigations should explore the combined effects of growth factors on osteoclast activity. (Ms. Shaw is a research assistant and Dr. Oursler is an Assistant Professor, Biology Department, University of Minnesota, Duluth.)

References

1. Clohisy DR, Palkert D, Ramnaraine MLR, et al. Human breast cancer induces osteoclast activation and increases the number of osteoclasts at sites of tumor osteolysis. J Bone Joint Surg 1996;14:396-402.

2. Reddy S, Roodman GD. Control of osteoclast differentiation. Crit Rev Euk Gene Exp 1998;8:1-17.

3. Pederson L, Winding B, Foged N, et al. Identification of breast cancer cell line-derived paracrine factors that stimulate osteoclast activity. Cancer Res 1999;59: 5849-5855.

4. Thomas R, Guise T, Yin J, et al. Breast cancer cells interact with osteoblasts to support osteoclast formation. Endocrinol 1999;140:4451-4458.

5. Chikatsu N, Takeuchi Y, Tamura Y, et al. Interactions between cancer and bone marrow cells induce osteoclast differentiation factor expression and osteoclast-like cell formation in vitro. Biochem Biophys Res Commun 2000;67:632-637.

6. Martin TJ, Romas E, Gillespie M. Interleukins in the control of osteoclast differentiation. Crit Rev Euk Gene Exp 1998;8:107-123.

7. Shaw A, Jakobsen M, Rock M, et al. Metastatic breast cancer: Identification of factors involved in tumor progression in bone. In: Second Joint Meeting of the American Society for Bone and Mineral Research and the International Bone and Mineral Society, San Francisco, CA; November 1998:S280.

8. Tani-Ishii N, Tsunoda A, Teranaka T, et al. Autocrine regulation of osteoclast formation and bone resorption by IL-1 alpha and TNF alpha. J Dent Res 1999;78: 1610-1623.

9. Shaw A, Gingery A, Maki J, et al. TNF alpha stimulates osteoclast maturation and inhibits apoptosis of mature osteoclasts: A model for tumor osteolysis. In: 21st Annual Meeting of the American Society for Bone and Mineral Research: St. Louis, MO; September 1999:S484.

10. Conover C. The role of insulin-like growth factors and binding proteins in bone cell biology. Principles of Bone Biology. Academic Press; 1996:607-618.

11. Chihara K, Sugimoto T. The action of GH/IGF-I/IGFBP in osteoblasts and osteoclasts. Horm Res 1997;48: 45-49.

12. Powell G, Southby J, Danks J, et al. Localization of parathyroid hormone-related protein in breast cancer metastases: Increased evidence in bone compared with other sites. Cancer Res 1991;51:3059-3061.

13. Rankin W, Grill V, Martin TJ. Parathyroid hormone-related protein and hypercalcemia. Cancer 1997;80: 1564-1571.

14. Van Beek E, Van der Wee-Pals L, Van de Ruit M, et al. Leukemia inhibitory factor inhibits osteoclastic resorption, growth, mineralization, and alkaline phosphatase activity in fetal mouse metacarpal bones in culture. J Bone Miner Res 1993;8:191-198.

15. Kurzrock R, Estrov Z, Wetzler M, et al. LIF: Not just a leukemia inhibitory factor. Endocrine Rev 1991;12: 208-217.

16. Dhingra K, Sahin A, Emami K, et al. Expression of leukemia inhibitory factor and its receptor in breast cancer: A potential autocrine and paracrine growth regulatory mechanism. Breast Cancer Res Treat 1998;48:165-174.

17. Gingery A, Shaw A, Holy J, et al. TGF beta is a survival factor for TGF beta-induced osteoclasts. In: 21st Annual Meeting of the American Society for Bone and Mineral Research: St. Louis, MO; September 1999:S300.