
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Advances in Treating Metastatic Bone Cancer |
Authors' Affiliations: 1 University of Rochester Medical Center, Rochester, New York and 2 ARIAD Pharmaceuticals, Cambridge, Massachusetts
Requests for reprints: Brendan F. Boyce, Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, 601 Elmwood Avenue, Box 626, Rochester, NY 14642. Phone: 585-275-5837; Fax: 585-273-3637; E-mail: brendan_boyce{at}urmc.rochester.edu.
| Abstract |
|---|
|
|
|---|
15 years ago. Since then, our understanding of the regulation of bone catabolism has advanced significantly with the identification of other key enzymes that regulate osteoclast formation, activation, and survival after their knockout in mice or recognition of mutations in them in humans. This led to the discovery or development of specific inhibitors of some of these key enzymes, including Src, as proof-of-concept lead compounds or potential clinical candidates for the prevention of diseases associated with increased bone resorption, such as osteoporosis and metastatic bone disease. Although bisphosphonates have been prescribed with proven and improving efficacy for the prevention of bone loss for >30 years, adverse effects, such as upper gastrointestinal tract symptoms, and the requirement to take them at least 2 hours before food have limited patient compliance. Thus, with growing knowledge of the pathways regulating osteoclast function and the appreciation that some of these are active also in tumor cells, drug companies have made efforts to identify small-molecular lead compounds for development into new therapeutic agents for the prevention of bone loss with efficacy that matches or supersedes that of bisphosphonates. In this article, we review our current understanding of the signaling pathways that regulate osteoclast formation, activation, and survival with specific reference to the role of Src tyrosine kinase and downstream signaling and highlight in a variety of models of increased bone resorption the effects of Src kinase inhibitors that have been targeted to bone to limit potential adverse effects on other cells.
B ligand (RANKL)/RANK signaling pathway was identified as the key regulatory mechanism for osteoclastogenesis (36). RANKL/RANK interaction on the cell surface of osteoclasts and their precursors triggers signaling through several enzymatic pathways, which leads to the activation of the transcription factors nuclear factor-
B, c-Fos, and nuclear factor of activated T cells (36), resulting in activation of the enzymes that mediate the secretion of protons (7) and chloride ions (8) for dissolution of the mineral and degradation of the matrix of bone (refs. 9, 10; reviewed in ref. 11). These studies led to the development or identification from screens of compound libraries of several small molecules that specifically inhibit these enzymes, including cathepsin K, vacuolar adenosine triphosphatase, the mammalian target of rapamycin, and Src tyrosine kinase. In addition to studies that describe the effects of synthetic small-molecule agents on bone, several studies have suggested that the active ingredients in a variety of natural product-related herbal remedies or foods have osteoprotective effects. These include tanshinone, the main active diterpene quinone of a herbal medicine used to treat coronary heart disease and also present in Chinese green tea (12), the water extract of Dioscorea spongiosa (13), and the soybean isoflavones, genistein, and daidzein (14). Most of these have been studied in patients with osteoporosis and not in those with metastatic bone disease. However, the osteoprotective effects of these compounds and phytoestrogens remain controversial (15), and double-blind controlled clinical trials will be required to prove efficacy of all herbal compounds in humans.
| Src Tyrosine Kinase |
|---|
|
|
|---|
In addition to these osteoclast-activating actions, Src also seems to mediate osteoclast survival in response to RANKL in vitro. In studies in which RANKL and macrophage colony-stimulating factor were withdrawn from src/ or wild-type (WT) osteoclasts, the WT cells survived in the presence of RANKL alone, whereas src/ osteoclasts died by apoptosis (17, 21). Src seems to mediate this protective effect by recruiting phosphatidylinositol 3-kinase to the intracellular domain of TRAF6 and activating the Akt/mammalian target of rapamycin pathway. This promotes osteoclast survival by phosphorylating the apoptosis inducer caspase-3 and thus inactivating it (22), although recent studies have questioned this role for Akt (23). Therefore, Src expression in osteoclasts could influence bone mass by regulating not only osteoclast activation but also osteoclast life span. In contrast, the src/ cells survived, albeit in slightly reduced numbers, when they were treated with macrophage colony-stimulating factor alone, indicating that Src does not mediate macrophage colony-stimulating factor survival signaling. Interestingly, osteoclast apoptosis is not increased in src/ mice in vivo, and indeed, osteoclast numbers are actually higher in src/ mice than in WT controls (21). Thus, either these in vitro findings reflect a phenomenon that does not apply in vivo or macrophage colony-stimulating factor or perhaps other Src family kinase members substitute for Src in vivo in the src/ mice to enhance their survival. Further studies will be required to determine which of these theories is correct.
There are three major structural domains in Src that could mediate its functions in osteoclasts: the kinase domain, which phosphorylates tyrosine residues in molecules that bind to the Src homology 2 domain, and the Src homology 3 domain, which binds proline-rich portions of interacting molecules. Studies designed to determine which domain(s) of the molecule mediates its resorptive function in osteoclasts suggested surprisingly that, although the kinase domain is required, Src kinase activity is not essential for bone resorption in vivo (24). In these studies, the TRAP promoter was used to drive expression of WT and various mutated src genes in crossed src knockout/transgenic mice. As expected, expression of WT Src in the src/ transgenic mice rescued the defect in ruffled border formation and resorption in the mice, which had normal bone volumes and tooth eruption. However, a surprising observation was that, in 25% of the src/ mice expressing the kinase-inactive mutant K295M in which the ATP-binding site of Src is mutated to prevent phosphorylation of tyrosine residues on interacting molecules, bone volumes were normal, and the mice also had normal tooth eruption and ruffled border formation. Equally surprising was the observation that the osteopetrosis became more pronounced in 25% of the mice, which had increased osteoclast apoptosis, whereas no phenotypic change was observed in the remaining 50% of the mice (24).
We found it difficult to explain these divergent findings definitively, but based on our previous experience that expression levels in osteoclasts of SV40 large T antigen driven by this same TRAP promoter in transgenic mice correlated with the number of apoptotic osteoclasts and the severity of osteopetrosis (25), we speculated that K295M expression levels varied in the src/ transgenic mice. We further hypothesized that at high levels of expression Src K295M was acting in a dominant-negative fashion to inhibit recruitment of interacting molecules, whereas optimal expression levels rescued the defect, and that there was low expression in most mice, which thus had no change in the intensity of their osteopetrosis. Unfortunately, we were unable to assess the levels of K295M protein expression in these mice with the antibodies available at the time. Nevertheless, our interpretation was supported by our observation that the percentage of apoptotic osteoclasts correlated with the severity of the osteopetrosis in the Src K295M transgenic mice. Furthermore, expression of the Src homology 2 and Src homology 3 domains of the molecule without the kinase domain also induced more pronounced osteopetrosis in src/ transgenic mice and in WT mice due to increased osteoclast apoptosis (21). Osteoclasts from these mice had reduced phosphorylated Akt and phosphatidylinositol 3-kinase, consistent with a failure of Src activation of the survival pathway.
Our conclusion that Src kinase activity is not essential for osteoclastic resorption in vivo has been challenged recently (26). These authors overexpressed the Src K295M mutant and other Src mutant molecules in osteoclasts in vitro using adenoviral infection and concluded that kinase activity was essential for resorption based on profoundly reduced resorptive activity by osteoclasts expressing Src K295M. This reduced resorptive activity was not rescued by concomitant expression of the constitutively active Src Y527F src mutant, which rescued the defective resorption of the kinase-deficient mutant Src Y416M. Although these are interesting and provocative findings, we believe that, based on their reported findings, there is an alternative explanation to the authors' conclusions. Specifically, the level of expression of the Src K295M mutant assessed by Western blot analysis was slightly higher in the src/ osteoclasts than that of native Src in the control src/ mice. Thus, we propose that, in these experiments, the authors had achieved a level of expression of Src K295M in osteoclasts in vitro that was equivalent to the levels we had achieved in vivo in osteoclasts in our Src K295M mice, which had a slight increase in the degree of their osteopetrosis. The osteoclasts infected with the K295M retroviral construct did not have increased apoptosis; thus, the level of expression of this construct is likely to have been lower than that which induces detectably increased osteoclast apoptosis in vivo (0.5 ± 0.1% osteoclasts apoptotic in rescued K295M transgenic mice versus 4.3 ± 1.2% in osteopetrotic K295M transgenic mice).3 Further studies will be required to determine the role of the kinase domain of Src in osteoclasts for their resorptive activity.
In addition to its role in osteoclasts, Src tyrosine kinase also has important functions in malignant cells, including regulation of cell division, growth factor signaling, and movement (27). Its activity is up-regulated in certain cancers. Thus, in patients with metastatic bone disease, Src inhibitors could potentially have negative effects not only on osteoclasts but also on tumor cells and their interactions with osteoclasts. Because of these roles for Src in osteoclasts and tumor cells, several pharmaceutical companies have advanced Src inhibitors as potential therapeutic agents for use in osteoporosis, osteolytic bone metastasis, and related diseases with increased resorption. Example of such Src inhibitors include the ATP-based lead compounds NVP-AAK980 (28) and AP23451 (see below); the pyrazolopyrimidines PP1 and PP2 (29), the pyrrolopyrimidines CGP-76775 and CGP-76030 (30), the pyridopyrimidinones PD166326 and PD18097 (31), the quinazoline AZM475271 (32), the quinoline SKI-606 (33), and the indolinone SU6656 (34). Hereinafter, we will briefly describe the in vitro and in vivo effects of Src kinase inhibitors AP23451 and AZD0530.
| Src Tyrosine Kinase Inhibitors |
|---|
|
|
|---|
|
|
| Summary |
|---|
|
|
|---|
| Open Discussion |
|---|
|
|
|---|
B signaling pathway.
Dr. Boyce: There's a paper published showing that an NF-
B inhibitor is able to prevent bone loss in an animal model of rheumatoid arthritis (Jimi, E, Aoki K, Saito H, et al. Selective inhibition of NF-
B blocks osteoclastogenesis and prevents inflammatory bone destruction in vivo. Nat Med 2004;23:617-624). It is important to have drugs that can specifically target the osteoclast and tumor cells.
Dr. Berenson: We're publishing data on an oncogene that's a potent inhibitor of osteoclast development and also knocks out the tumor.
Dr. Boyce: Do you have evidence that TRAF-6 signaling is important in tumor cell growth?
Dr. Berenson: We've used IL-1 in this model to induce and then we can block that with the TRAF-6.
Dr. Boyce: Do you think you are hitting the osteoclast only with your TRAF-6 antibody or antagonist?
Dr. Berenson: We attenuate growth of the tumor cells in both cell lines and fresh tumor cells with the construct. It's a common pathway for both and then leads down to NF-
B, so that's why we thought it was a good target.
Dr. Suva: You mentioned that the phosphonate group had no effect in vitro. Do you have any data on that molecule bringing something, for example, an inactive Src inhibitor or some other molecule? Are you seeing effects of the phosphonate group on resorption and the Src inhibitor, perhaps on the tumor cells? Are you confident that the compound is affecting both of those processes and that it's not a reflection of the activity of the compound on one cell and the phosphonate group, perhaps, on the osteoclast?
Dr. Boyce: The phosphonate group itself has no efficacy in vitro at inhibiting osteoclasts.
Dr. Suva: I wouldn't expect it to have much activity in vitro, but what about in vivo?
Dr. Boyce: We haven't tested that. Based on the lack of effects in vitro, we're assuming the phosphonate group isn't having any effect on either the tumor cells or osteoclasts but we believe that you need it to get it to the bone.
Dr. Guise: Does Src having any role in osteoblast function, and do the Src inhibitors affect that at all?
Dr. Boyce: A few years ago, we published data showing that Src is a negative regulator of osteoblast activity with Anna Teti (Marzia M, Sims NA, Voit S. Decreased c-Src expression enhances osteoblast differentiation and bone formation. J Cell Biol 2000;151:311-320). Osteoblasts from Src knockout mice have increased alkaline phosphatase activity and increased nodule formation, and the mice have increased bone formation rate in vivo. When we looked at the effects of this particular Src inhibitor on osteoblasts, to our chagrin it killed osteoclasts in vitro. We then tested other inhibitors similar to this, which also had an antiosteoclastic activity in vitro. In fact, we screened many compounds to look for one that inhibited osteoblasts and stimulated osteoblasts in vitro and identified one that stimulated alkaline phosphatase. It's interesting the way that small changes in the molecular structure of these compounds can have a significant affect on their efficacy on cells. So structure-function relationships are vital in this regard. There is reason to be hopeful that a Src tyrosine kinase inhibitor could inhibit bone resorption and potentially stimulate bone formation in osteoporotic patients.
Dr. Vessella: Do you have data comparing the effect of the Src compounds compared with zoledronic acid? Which one has a more pronounced effect?
Dr. Boyce: They had similar effects in inhibiting bone resorption in vivo. There was a significant reduction in tumor volume only with the Src inhibitor. However, when we measured tumor volume in all of the bone sites, the reduction did not reach significance in the zoledronic acid-treated mice. In other studies we've done using PC3 cells with zoledronic acid, given once a week for 4 weeks, 3 days after the tumor cells were injected into the tibia, the mice developed osteopetrosis, but there is clearly significant resorption of the bone. Once you have established metastases, zoledronic acid and other bisphosphonates are not able to fully stop aggressive resorption in this regimen.
Dr. Roodman: Do you have any feel for what is the contribution of blocking bone destruction by the osteoclast?
Dr. Boyce: In the first paper we published showing that bisphosphonates inhibit osteoclastic resorption and osteolytic lesions, there was some necrosis of tumors growing within those bones and there certainly was increased intraosseous pressure that could have contributed to that (Sasaki A, Boyce BF, Story B, et al. Bisphosphonate risedronate reduces metastatic human breast cancer burden in bone in nude mice. Cancer Res 1995;55:3551-3557). If the tumor cells can get out of the bone, then some of the tumors survive. We found that the bone could become necrotic as well. I haven't seen that with zoledronic acid or with the Src tyrosine kinase inhibitor. Generally, if bone resorption is stopped while the tumor cells are in the bone, the bone erosion is inhibited, but tumor cells can still fill the marrow cavity.
Dr. Weilbaecher: We have done experiments where we took the B16 melanoma line, a very aggressive line, and injected it into the Src knockout mice. There were tumor metastases but no tumor-associated bone destruction.
Dr. Roodman: If you get resorption, growth factors are released. The question is whether this release of growth factors is purely mechanical or if it's impacting on tumor growth. Osteopetrotic animal experiments won't help because you don't get resorption.
Dr. Boyce: I've not done those experiments in mice. I've seen enough metastases in humans in whom cortical bone gets remodeled. This cortical remodeling doesn't happen in mice, so having spaces for osteoclasts to resorb in cortical bone and not have the extensive necrosis that you can see in mice made osteopetrotic by osteoclast inhibitors may not be an issue. Metastases tend to grow at sites where there is red marrow, and they can then spread down the marrow cavity of long bones, for example, or burst out through the sides of vertebral bodies. The challenge in mice is because of failure of cortical bone to remodel. It may be easier to answer these questions in humans, but that's difficult.
| Footnotes |
|---|
Received 4/21/06; revised 7/13/06; accepted 8/ 2/06.
| References |
|---|
|
|
|---|
B1 and NF-
B2. Nat Med 1997;3:12859.[CrossRef][Medline]
B in osteoclast and B-cell development. Genes Dev 1997;11:348296.
(v)ß(3) integrin-mediated signaling, cell adhesion, and osteoclast motility. J Cell Biol 2001;152:18195.
, and RANK ligand promote osteoclast survival by signaling through mTOR/S6 kinase. Cell Death Differ 2003;10:116577.[CrossRef][Medline]This article has been cited by other articles:
![]() |
H.-J. Kim, K. Zhang, L. Zhang, F. P. Ross, S. L. Teitelbaum, and R. Faccio The Src family kinase, Lyn, suppresses osteoclastogenesis in vitro and in vivo PNAS, February 17, 2009; 106(7): 2325 - 2330. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Guan, Z. Zhou, G. E. Gallick, S.-F. Jia, J. Morales, A. K. Sood, S. J. Corey, and E. S. Kleinerman Targeting Lyn inhibits tumor growth and metastasis in Ewing's sarcoma Mol. Cancer Ther., July 1, 2008; 7(7): 1807 - 1816. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q. Zhang, R. Guo, Y. Lu, L. Zhao, Q. Zhou, E. M. Schwarz, J. Huang, D. Chen, Z.-G. Jin, B. F. Boyce, et al. VEGF-C, a Lymphatic Growth Factor, Is a RANKL Target Gene in Osteoclasts That Enhances Osteoclastic Bone Resorption through an Autocrine Mechanism J. Biol. Chem., May 9, 2008; 283(19): 13491 - 13499. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. J. H. Chu and P. L. Yang c-Src protein kinase inhibitors block assembly and maturation of dengue virus PNAS, February 27, 2007; 104(9): 3520 - 3525. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |