Cancer Cell Biology
Inhibition of human prostate cancer growth, osteolysis and angiogenesis in a bone metastasis model by a novel mechanism-based selective gelatinase inhibitor
Article first published online: 27 DEC 2005
Copyright © 2005 Wiley-Liss, Inc.
International Journal of Cancer
Volume 118, Issue 11, pages 2721–2726, 1 June 2006
How to Cite
Bonfil, R. D., Sabbota, A., Nabha, S., Bernardo, M. M., Dong, Z., Meng, H., Yamamoto, H., Chinni, S. R., Lim, I. T., Chang, M., Filetti, L. C., Mobashery, S., Cher, M. L. and Fridman, R. (2006), Inhibition of human prostate cancer growth, osteolysis and angiogenesis in a bone metastasis model by a novel mechanism-based selective gelatinase inhibitor. Int. J. Cancer, 118: 2721–2726. doi: 10.1002/ijc.21645
- Issue published online: 14 MAR 2006
- Article first published online: 27 DEC 2005
- Manuscript Accepted: 26 SEP 2005
- Manuscript Received: 30 JUN 2005
- U.S. Department of Defense, Consortium. Grant Number: PC 21004
- National Institutes of Health-NCI. Grant Numbers: RO1 CA88028, RO1 CA100475
- NIDDK. Grant Number: 067687
- U.S. Department of Defense. Grant Number: DAMD 17-02-1-0157
- matrix metalloproteinases;
- gelatinase inhibitor
Metastasis to the bone is a major clinical complication in patients with prostate cancer (PC). However, therapeutic options for treatment of PC bone metastasis are limited. Gelatinases are members of the matrix metalloproteinase (MMP) family and have been shown to play a key role in PC metastasis. Herein, we investigated the effect of SB-3CT, a covalent mechanism-based MMP inhibitor with high selectivity for gelatinases, in an experimental model of PC bone metastases. Intraperitoneal (i.p.) treatment with SB-3CT (50 mg/kg) inhibited intraosseous growth of human PC3 cells within the marrow of human fetal femur fragments previously implanted in SCID mice, as demonstrated by histomorphometry and Ki-67 immunohistochemistry. The anti-osteolytic effect of SB-3CT was confirmed by radiographic images. Treatment with SB-3CT also reduced intratumoral vascular density and bone degradation in the PC3 bone tumors. A direct inhibition of bone marrow endothelial cell invasion and tubule formation in Matrigel by SB-3CT in vitro was also demonstrated. The use of the highly selective gelatinase inhibitors holds the promise of effective intervention of metastases of PC to the bone. © 2005 Wiley-Liss, Inc.
PC is the second leading cause of cancer death in males in the United States.1 Bone metastasis represents a major clinical complication of PC and is usually associated with pain, fractures and other life-impairing conditions.2 In spite of the extent of this complication, the therapeutic options for the treatment of PC bone metastasis are very limited and are mostly palliative in nature. Therefore, new approaches to treat PC bone metastasis are urgently needed.
Members of the matrix metalloproteinase (MMP) family of zinc-dependent endopeptidases, in particular gelatinases A and B, also known as MMP-2 and MMP-9, have been associated with the development of bone metastasis by PC cells.3, 4, 5, 6, 7, 8 Therefore, MMPs constitute an attractive target for intervention in PC bone metastasis. A limited number of preclinical studies reported the ability of synthetic MMP inhibitors to inhibit primary and metastatic PC growth in animal models.9, 10, 11 We previously reported that administration of the broad-spectrum MMP inhibitor batimastat (BB-94) reduced tumor burden and bone degradation by human PC3 cells growing within human bone in severe combined immunodeficiency (SCID) mice.8 These studies demonstrated the potential benefit of targeting MMP activity in PC bone metastasis. In spite of the success of broad-spectrum MMP inhibitors in preclinical studies, these inhibitors failed to demonstrate therapeutic efficacy in clinical trials in patients with advanced cancer, and also produced undesired side effects.12, 13, 14 Lack of selectivity has been considered one of the main reasons for the disappointing performance of broad-spectrum MMP inhibitors in clinical trials.12, 14 Thus, targeting of specific and relevant MMPs in cancer progression remains an important goal.
Accumulating evidence indicates that the expression and activity of gelatinases is associated with PC progression.3, 4, 5, 6, 7, 8, 15, 16, 17 Inhibition of MMP-9 expression reduced lung metastases and tumor growth of PC cells in mice.18, 19 Recently, we have also shown that MMP-9 activity is upregulated in PC3 tumors growing within human bone implanted in SCID mice.20 Together, these data suggest that specific inhibition of MMP-9 in PC bone metastasis may be therapeutically beneficial.
We reported on the development of the first and only mechanism-based MMP inhibitor that is selective for gelatinases.21, 22 This inhibitor, referred to as SB-3CT, was recently shown to reduce liver metastasis and to increase survival in a murine model of T-cell lymphoma23 characterized by high MMP-9 expression and activity.24 In this study, we examined the effect of SB-3CT on the growth of human PC3 cells within human bone and showed that SB-3CT is a potent inhibitor of intraosseous tumor growth, bone degradation, and intratumoral angiogenesis.
Material and methods
Preparation of SB-3CT
SB-3CT ((4-phenoxyphenylsulfonyl)methylthiirane, MW = 306.04 Da), was designed specifically to bind to the active sites of gelatinases, MMP-2 and MMP-9, and the thiirane moiety was expected to coordinate to the active-site zinc ion. The chemical structure, synthesis, and characterization of SB-3CT has been previously described.21, 22 The inhibitor was synthesized in the laboratory of Dr. Mobashery (University of Notre Dame, Notre Dame, IN). A 10-mM stock solution of SB-3CT was prepared in 100% dimethylsufoxide (DMSO) (Sigma, St. Louis, MO) and stored at −20°C until used. For in vitro studies, the stock solution was diluted in culture medium with 1% DMSO as needed. For in vivo experiments, the stock solution was diluted to a final concentration of 1.25 mg/mL SB-3CT in 10% DMSO in normal saline.
The prostate cancer PC3 cell line25 and the human fibrosarcoma HT1080 cell line26 were both purchased from the American type culture collection (Manassas, VA). The cell lines were cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS) (Invitrogen). BMEC-1, a human bone marrow microvascular endothelial immortalized cell line,27 was kindly provided by Dr. Francisco Candal (Centers for Disease Control and Prevention, Atlanta, GA). These cells were cultured in Medium-199 (Gibco BRL Life Technologies, Grand Island, NY) supplemented with 20% FBS, 2 mmol/L L-glutamine, 5 U/mL heparin, 100 IE/ml penicillin, 100 μg/ml streptomycin, and 50 μg/ml endothelial cell growth supplements (ECGS; Biomedical Technologies, Inc., Stoughton, MA).
Cell proliferation assay
PC3 cells were seeded in 35-mm dishes (5 × 104 cells/dish) in complete culture medium. The next day, the medium was replaced with complete medium supplemented with 1% DMSO alone (vehicle) or SB-3CT (final concentrations 0.1–50 μM) in 1% DMSO. At various times, the cells were harvested with trypsin and counted.
Five-week-old male C.B.-17.SCID mice were purchased from Taconic Farms (Germantown, NY). Mice were maintained under aseptic conditions, according to the NIH standards established in the Guidelines for the Care and Use of Experimental Animals, and the Animal Investigation Committee of Wayne State University approved all of the experimental protocols.
In situ gelatin zymography
Frozen tissue sections were obtained from HT1080 tumors grown subcutaneously in SCID mice, which were intraperitonially (i.p.) treated for two consecutive days before sacrifice either with 1 mL vehicle (10% DMSO in PBS) or 1 ml containing 1.25 mg SB-3CT in 10% DMSO (equivalent to 50 mg/kg of mouse weight). In situ gelatin zymography was performed in 8-μm thick unfixed cryostat tumor sections incubated for 1 h with 100 μg/ml DQ™-gelatin (EnzChek; Molecular Probes, Eugene, OR) and 1 μg/mL DAPI (Molecular Probes), as described previously.23, 28
Establishment of PC3 human bone tumors and experimental treatment
One fourth human fetal femur fragments were implanted subcutaneously in SCID mice as described previously.29 Four weeks later, 1 × 105 PC3 cells were injected through the mouse skin directly into the marrow of the previously implanted bone, as described.29 Twenty-four h after tumor cell inoculation, the mice were injected i.p. with either vehicle (10% DMSO) or SB-3CT in 10% DMSO (50 mg/kg of mouse weight) every other day. Each experimental group contained 9 animals.
Five weeks after tumor cell inoculation, the mice were killed and bone implants harvested, weighed, fixed overnight in 10% buffered formalin, and then X-ray imaged using a Lo-Rad M-IV mammography unit (Lorad, Danbury, CT) with a magnified specimen technique. Images were developed using a Kodak 2000 screen and radiography film (Kodak, Rochester, NY). For histomorphometrical and histological analyses, bone tumors were decalcified with 10% ethylenediaminetetraacetic acid (EDTA) (pH 6.5) in PBS, dehydrated, infiltrated and paraffin-embedded.
Paraffin sections (5 μm) derived from bone tumors were immunostained for cytokeratin and counterstained with hematoxylin. Digital photomicrographs of the entire histological section were captured at 5× magnification and stored as jpeg files. The entire image was then reconstructed using Adobe Photoshop® 7.0 (Adobe Systems, Mountain View, CA). Tumor tissue (cytokeratin positive areas) and trabecular bone were isolated into separate layers and separately thresholded to black. The whole tissue cross sectional area (considered 100%) was then highlighted and the area occupied by either tumor or bone was automatically calculated using the software for each reconstructed figure.
Tissue sections were deparaffinized, pretreated in Ag Citrus Plus Retrieval Solution (BioGenex, San Ramon, CA) in a microwave to achieve antigen retrieval and incubated with either monoclonal antibodies against pan cytokeratin (Sigma), Ki-67 (BioGenex), or human CD34 (DAKO, Carpinteria, CA). The antigens were visualized by the Mouse on Mouse (MOM™) immunodetection peroxidase kit (Vector, Burlingame, CA) or DAKO EnVision™ + system, and sections were counterstained by light Mayer's hematoxylin. To assess proliferation index (Ki-67) and intratumoral microvascular density (CD34), “hot spots” at high magnification (×400) were considered. Negative controls were obtained by replacing the primary antibodies with non-immune mouse immunoglobulin.
Effect of SB-3CT on BMEC-1 cell viability
BMEC-1 cells were seeded in 96-well culture plates (104 cells/well) in complete culture medium. Twenty-four h later, the medium was replaced with serum-free, phenolred-free media supplemented with either vehicle (1% DMSO) or SB-3CT (1 nM–50 μM final concentrations). After 72 h, 10 μL of WST-1 (Roche Diagnostics GmBH, Mannheim, Germany) were added to each well, and the optical density was measured at 450 nm, according to the manufacturer's instructions.
Capillary-like tubule formation assay
Twenty-four-well plates were coated with 300 μL of an ice-cold Matrigel (Becton Dickinson, Franklin Lakes, NJ) solution (10 mg/mL). The plates were then incubated for 30 min at 37°C to allow Matrigel polymerization, and then 5 × 104 BMEC-1 cells were placed onto the Matrigel-coated wells in the presence of complete medium supplemented with various amounts of SB-3CT (0.1–1 μM) or vehicle (1% DMSO). After overnight incubation at 37°C, digital photographs of three randomly selected areas from each well were taken at 10× magnification, using an Olympus® DP12 Microscope Camera. The area occupied by the capillary-like structures was calculated using Adobe Photoshop 7.0.
Endothelial cell invasion assay
BMEC-1 cells suspended in Medium-199 with 0.1 % bovine serum albumin supplemented with either SB-3CT (0.1–1 μM) or 1% DMSO (vehicle) were seeded (2 × 105 cells per insert) onto Transwell inserts (8-μm pore size, Becton Dickinson) coated with 25 μg/filter Matrigel. Culture medium supplemented with 5% FBS was placed in the lower chamber as a chemoattractant. After 24 h incubation at 37°C, the cells that migrated to the lower side of the filter were stained with Diff-Quik® (Dade Behring, Newark, DE) and counted under 200× magnification.
Data were statistically analyzed by one-way ANOVA followed by Tukey-Kramer multiple comparison post-test, Student's t test or Mann-Whitney test, depending on the number of groups involved in each experiment and whether the values had a Gaussian distribution or not. All the statistical analyses were carried out using the GraphPad InStat® version 3.0 (GraphPad Software, San Diego, CA), and the differences were considered to be statistically significant at p < 0.05.
SB-3CT inhibits gelatinolytic activity in vivo
We first examined the ability of SB-3CT to inhibit gelatinase activity in vivo. To this end, we used in situ gelatin zymography together with DAPI counterstaining of frozen sections of human tumor xenografts harvested from mice treated either with SB-3CT or with vehicle alone. This technique was chosen for analysis of inhibition of gelatinase activity in vivo, because gelatin zymography of tumor extracts cannot detect gelatinase inhibition probably because of the dissociation of the enzyme-inhibitor complex during sample preparation and/or electrophoresis of the samples (Price et al.30 and our unpublished results). As a tumor model, we used HT1080 human fibrosarcoma cells, which are known to produce both MMP-2 and MMP-9 in vivo and to readily form tumors when inoculated subcutaneously into immunodeficient mice.31 Also, frozen sections can be obtained from HT1080 tumors, a procedure that cannot be accomplished with PC3 bone tumor xenografts. Subcutaneous tumors of PC3 cells were not appropriate for in situ gelatin zymography, because these cells do not express gelatinases when inoculated in this site, as we have recently described.20 However, SB-3CT inhibited the gelatinolytic activity of MMP-9 secreted by cultured PC3 cells, as determined by gelatin zymography (data not shown).
As shown in Figure 1a, HT1080 tumor sections from mice treated with vehicle alone exhibited intense green fluorescence, indicative of gelatinolytic activity. This activity was completely abrogated by 1,10-o-phenanthroline, a zinc-ion chelator, consistent with metalloproteinase activity (data not shown). SB-3CT treatment significantly reduced gelatinolytic activity in the tumor sections (Fig. 1b). Figure 1c shows a representative image of a frozen section obtained from an HT1080 fibrosarcoma grown subcutaneously in a SB-3CT-treated mouse (hematoxylin–eosin stain). These results indicate that SB-3CT administration systemically inhibits gelatinase activity in vivo, in agreement with previous results in a T-cell lymphoma tumor model23 and in a mouse model of cerebral ischemia.32
Effect of SB-3CT on intraosseous tumor growth
The SCID-human model of PC growth within human fetal bone29 was used to examine the effect of SB-3CT on intraosseous tumor growth. To this end, SCID mice bearing implants of human fetal bone fragments were inoculated with PC3 cells into the bone fragments, and a day later, treatment began with either SB-3CT or vehicle (10% DMSO in PBS). Gross morphology of human bones injected with PC3 cells revealed that SB-3CT treatment resulted in smaller bone tumors when compared to those harvested from vehicle-treated mice (Fig. 2). When weighed, the bone tumors from SB-3CT-treated mice were significantly lighter (median, 160 mg; range, 130–260 mg) than the bone tumors from vehicle-treated mice (median, 268 mg; range, 131–634 mg) (p = 0.03, Mann-Whitney test). By histomorphometry, tumor cells occupied a significantly smaller proportion of the bone tissue area in SB-3CT-treated mice compared with vehicle-treated mice (Fig. 3a). Histomorphometric analysis also demonstrated that the area of the tissue comprised of trabecular bone was significantly higher in SB-3CT-treated mice (Fig. 3b), suggesting that gelatinase inhibition helps preserve bone integrity. X-ray imaging (Fig. 2, lower panels) confirmed smaller and less osteolytic lesions in bone tumors harvested from SB-3CT-treated mice. The reduced tumor burden in the SB-3CT-treated mice was consistent with a significant decrease in the percentage of cells staining positively for the nuclear proliferative antigen Ki-67 (Fig. 3c). However, SB-3CT had no direct effect on PC3 tumor cell proliferation in vitro in doses up to 10 μM (data not shown). Taken together, these studies demonstrated that SB-3CT is a potent inhibitor of intraosseous PC growth and cancer-associated bone degradation.
Effect of SB-3CT on tumor vascularization and distinct stages of angiogenesis
We previously showed that in the SCID-human tumor model, the majority of blood vessels present within PC3 bone tumors are of human origin.33 Therefore, we examined the expression of CD34, an established marker of angiogenesis, in untreated and SB-3CT-treated mice bearing PC3 tumors, using an antibody that specifically recognizes human CD34.34 As shown in Figure 4, CD34 immunostaining indicated a significant reduction in the number of human blood vessels in the intraosseous prostate tumors in SB-3CT-treated animals compared to tumors in vehicle-treated controls. Furthermore, the size of the blood vessels in the tumors of SB-3CT-treated mice was significantly smaller (Fig. 4b), suggesting that SB-3CT treatment inhibits tumor angiogenesis.
To further examine the anti-angiogenic activity of SB-3CT, we carried out an in vitro assay of tube formation using BMEC-1 cells cultured within Matrigel. Figure 5 shows that SB-3CT inhibited capillary-like tubule formation by BMEC-1 cells in a dose-dependent fashion. SB-3CT also inhibited the ability of BMEC-1 cells to invade filters coated with Matrigel (Fig. 6). In contrast, SB-3CT had no effect on BMEC-1 cell proliferation when tested at doses up to 10 μM (data not shown).
Previous studies suggested that MMPs play a role in the development and growth of PC skeletal metastasis.3, 4, 5, 6, 7, 8, 15, 16, 17 However, the relative importance of gelatinases in this process in a relevant tumor model could not be addressed because of the lack of selective gelatinase inhibitors. Here we report for the first time the anti-tumor effect of a new class of MMP inhibitor, SB-3CT, which is a mechanism-based and selective inhibitor for gelatinases in the SCID-human model of PC bone metastasis.29 SB-3CT was shown recently to inhibit experimental liver metastasis in an aggressive model of mouse T-cell lymphoma and to increase mice survival.23 In this study we show that SB-3CT administration effectively inhibits tumor growth of human prostate cancer PC3 cells within human bone tissue. The reduced intraosseous growth of PC3 tumor cells as a result of SB-3CT treatment correlated with a decrease in intratumoral vascularity, as determined by CD34 immunostaining.35, 36, 37 Consistently, we found that SB-3CT inhibited invasion of Matrigel and formation of tube-like structures by cultured human bone marrow endothelial cells. Therefore, SB-3CT's effects on intraosseous tumor growth may be mediated by a direct action on endothelial cell invasion and neovessel formation, two processes that were shown to be dependent in part on gelatinase activity.36, 38, 39 Thus, inhibition of angiogenesis by SB-3CT could be a key aspect of its anti-tumor activity within the bone microenvironment. This is consistent with recent studies showing that inhibition of angiogenesis by a tyrosine kinase inhibitor targeting platelet derived growth factor receptor reduced tumor growth of PC3 cells within the bone.40
Inhibition of PC3 tumor growth by SB-3CT could also be a direct consequence of reduced extracellular matrix degradation within the bone tissue by the tumor cells themselves and/or by osteoclasts. Indeed, SB-3CT treatment was associated with a reduced osteolytic response, indicating that SB-3CT helps to preserve bone integrity. We recently reported that in the SCID-human model of PC used herein, MMP-9 activity is increased in PC3 tumor-bearing bones and that this increase in activity is associated with osteoclasts recruitment and bone degradation.20 This finding is consistent with a role for MMP-9 in osteoclast migration to sites of bone remodeling,41 which may contribute to the process of bone resorption. Previously, we showed that BB-94, a broad-spectrum MMP inhibitor, reduced PC3 bone tumor growth, osteoclast recruitment and bone degradation.8 Since SB-3CT is a highly selective MMP-9 inhibitor, the reduced osteolytic response observed here may also be ascribed to an inhibitory effect of SB-3CT on MMP-9-mediated osteoclasts recruitment.
No clinical trials have been conducted to date directly testing the therapeutic value of MMP inhibitors in metastatic bone disease, in spite of the accumulating evidence implicating MMP proteolytic activity in bone metastasis. In general, MMP inhibition was not successful in clinical trials.12, 14, 42 However, MMP inhibitor clinical trials had many deficiencies, including poor inhibitor selectivity, which produced undesired side effects, lack of tumor MMP expression/activity profiling, and the advanced stage of the cancer patient population, just to mention a few.12, 14, 43, 44 Although MMPs have not been targeted for therapeutic intervention in PC bone metastatic disease, there is evidence that bone anti-resorptive therapies that target osteoclast activity can delay the progression of symptoms from bone metastasis in PC patients.45 The results presented in the current study using a relevant model of PC growth within bone tissue suggest that targeting gelatinases in PC with inhibitors such as SB-3CT has the potential to become an effective approach for the treatment of bone metastasis.
This work was supported by the U.S. Department of Defense, Consortium Grant PC 21004 (MLC), National Institutes of Health-NCI (grants RO1 CA88028 to MLC and RO1 CA100475 to RF), and NIDDK Grant 067687 (MLC), and U.S. Department of Defense, grant no. DAMD 17-02-1-0157 (MLC).
- 41Stromal cell-derived factor-1 (SDF-1) recruits osteoclast precursors by inducing chemotaxis, matrix metalloproteinase-9 (MMP-9) activity, and collagen transmigration. J Bone Miner Res 2003; 18: 1404–18., , , .