help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2005-1502
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
147/5/2171    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Valta, M. P.
Right arrow Articles by Härkönen, P. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Valta, M. P.
Right arrow Articles by Härkönen, P. L.
Endocrinology Vol. 147, No. 5 2171-2182
Copyright © 2006 by The Endocrine Society

Regulation of Osteoblast Differentiation: A Novel Function for Fibroblast Growth Factor 8

Maija P. Valta, Teuvo Hentunen, Qiang Qu, Eeva M. Valve, Anna Harjula, Jani A. Seppänen, H. Kalervo Väänänen and Pirkko L. Härkönen

Institute of Biomedicine, Department of Anatomy, University of Turku (M.P.V., T.H., Q.Q., E.M.V., A.H., J.A.S., H.K.V., P.L.H.), 20520 Turku, Finland; and Department of Laboratory Medicine, Tumor Biology, Malmö University Hospital, Lund University (P.L.H.), 20502 Malmö, Sweden

Address all correspondence to: Dr. Maija P. Valta, Department of Anatomy, Institute of Biomedicine, University of Turku, Kiinamyllynkatu 10, 20520 Turku, Finland. E-mail: maija.valta{at}utu.fi.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Several members of the fibroblast growth factor (FGF) family have an important role in the development of skeletal tissues. FGF-8 is widely expressed in the developing skeleton, but its function there has remained unknown. We asked in this study whether FGF-8 could have a role in the differentiation of mesenchymal stem cells to an osteoblastic lineage. Addition of FGF-8 to mouse bone marrow cultures effectively increased initial cell proliferation as well as subsequent osteoblast-specific alkaline phosphatase production, bone nodule formation, and calcium accumulation if it was added to the cultures at an early stage of osteoblastic differentiation. Exogenous FGF-8 also stimulated the proliferation of MG63 osteosarcoma cells, which was blocked by a neutralizing antibody to FGF-8b. In addition, the heparin-binding growth factor fraction of Shionogi 115 (S115) mouse breast cancer cells, which express and secrete FGF-8 at a very high level, had an effect in bone marrow cultures similar to that of exogenous FGF-8. Interestingly, experimental nude mouse tumors of S115 cells present ectopic bone and cartilage formation as demonstrated by typical histology and expression of markers specific for cartilage (type II and IX collagen) and bone (osteocalcin). These results demonstrate that FGF-8 effectively predetermines bone marrow cells to differentiate to osteoblasts and increases bone formation in vitro. It is possible that FGF-8 also stimulates bone formation in vivo. The results suggest that FGF-8, which is expressed by a great proportion of malignant breast and prostate tumors, may, among other factors, also be involved in the formation of osteosclerotic bone metastases.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
MAMMARY CARCINOMAS OCCASIONALLY contain ectopic cartilage and bone-like formations. Metaplastic bone formation is detected in less than 0.2% of human breast carcinomas (1), and it is considered to originate in the underlying breast carcinoma, not in a simultaneously growing sarcoma (1, 2). The clinical features of metaplastic mammary carcinomas seem to be similar to those of invasive mammary carcinomas (2, 3, 4).

The mechanisms of these metaplastic changes are not clear. However, induction of tumor or stromal cell differentiation toward an osteoblastic phenotype in breast tumors is obviously required. These mechanisms are of considerable interest, because they may represent not only increased or uncontrolled production of osteogenic factors in malignancy, but also factors involved in normal bone formation. They may also have a role in bone metastasis, another pathophysiological process involving bone formation in cancer. Breast and prostate cancer commonly metastasize in bone and cause local stimulation of osteoblasts, leading to osteosclerotic bone lesions (5). Metaplastic tumors thus offer a possibility to study the regulatory factors of osteoblastic differentiation.

Metaplastic cartilage and bone formation have previously been demonstrated by us and others in nude mouse tumors (6) produced by S115 cells and their genetic variant cell lines (7) and in syngenic S115 tumors (8, 9, 10). S115 cells produce and secrete considerable amounts of fibroblast growth factor 8 (FGF-8) (11). The 23 members of the FGF family have diverse roles in regulating cell proliferation, differentiation, and migration, and they mediate their effects by binding to specific high-affinity tyrosine kinase receptors, designated FGFR1 to -4 and to low affinity heparan sulfate proteoglycans (12). FGFs are important modulators of tumor growth and angiogenesis (12, 13), and they have been implicated in tumor formation, for example, in breast and prostate cancer (14, 15). The expression of FGFRs is also distorted in various types of cancer (12).

FGF-8, initially identified as androgen-induced growth factor, was first cloned from Shionogi mouse mammary tumor-derived SC-3 cells (11). Alternative splicing of the mouse FGF-8 gene allows transcription of eight different isoforms, of which FGF-8b has the highest potential in transforming NIH-3T3 cells (16) and the greatest ability to activate FGFRs in mitogenic assays (17). FGF-8 is highly conserved in evolution and has important roles in embryogenesis (18). It has been suggested to function as a key regulator of limb development (19) and the central nervous system (20). In normal adult tissues, FGF-8 is not known to be expressed, except in certain cell types involved in spermatogenesis and oogenesis (21). It is found, however, in human breast, ovarian, and prostate cancer (22, 23, 24), and the results of tumor analyses and experiments with cell models (25, 26) suggest that FGF-8 is an important autocrine or paracrine factor in these carcinomas.

FGF signaling pathways also have important roles in bone. Activating FGFR mutations have been linked to several forms of human dwarfism and craniosynostosis syndromes (27). The FGFR3 has been identified as a negative regulator of bone growth (28), whereas activating mutations in FGFR2 (27) and FGFR1 (29) contribute to increased osteoblast differentiation. Of ligands of FGFRs, at least FGF-1 and/or FGF-2 are expressed in chondroblasts and differentiating and mature osteoblasts during skeletal development (30, 31, 32). The effects of FGFs on osteoblasts depend on their differentiation stage. FGFs induce immature osteoblastic cells to proliferate, whereas mature osteoblasts respond by unaltered or decreased proliferation (27). In a study by Izbicka et al. (33), a human amniotic tumor (WISH) was found to induce extensive new bone formation by way of FGF-mediated mechanisms. Recently, FGF-18 has also been shown to be involved in bone formation (34). FGFs regulate osteoclastogenesis (35) and have been reported to inhibit chondrocyte proliferation (36). Together, evidence from studies carried out in vitro and in vivo indicates that FGFs are able to regulate both the proliferation and differentiation of bone cells by way of complex mechanisms.

In this communication we aimed to study whether FGF-8 has a role in osteoblast differentiation, because S115 cells, which express FGF-8 at a high level, induce ectopic bone and cartilage in nude mouse tumors. We found that FGF-8 regulates different stages of mesenchymal stem cell differentiation toward osteogenic lineage in mouse bone marrow culture. This factor considerably increased the osteogenic capacity of bone marrow cells at the early stage of their differentiation and may have a role in bone formation.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animal experiments
Animal experiments were performed as previously described (6). Briefly, 63 6- to 7-wk-old athymic male mice (NCr-{nu}/{nu}) were purchased from Bomholtgård (Rye, Denmark). The animal experiments were carried out at the central animal laboratory of Turku University, which is managed according to the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes. The experimental procedures were reviewed by the local ethics committee on animal experimentation at University of Turku and approved by the local provincial state office of Western Finland. The mice were randomly grouped, and 5 x 106 S115 mouse breast cancer tumor cells and their genetic variants (clones 21, 22, 27, and 33) (7) were inoculated sc into intact mice. These cell lines have an increased rate of proliferation and expression of FGF-8b at a high level in the presence of androgens (25). The tumors were allowed to grow for 7–13 wk to an average weight of 900 mg. At the end of this period, the animals were killed in a CO2 chamber. The tumors were quickly removed and divided into two parts. One half was fixed for histological examination, and the other half was frozen for RNA analysis.

In situ hybridization
Type II and IX collagen antisense and sense probes were prepared by in vitro transcription of linearized pBluescript KS (Stratagene, La Jolla, CA) containing the coding region of mouse pro-{alpha}1(II)-collagen cDNA (405 bp) and COL1 domain, and parts of the NC1 and NC2 domains of mouse {alpha}1(IX)-collagen (444 bp; the probes were provided by Dr. Eero Vuorio, University of Turku). In situ hybridization of nude mouse tumors formed by S115 cells or S115 cell clones was performed as previously described (21).

Immunohistochemistry
Serial sections (5 µm) of paraffin-embedded S115 and S115 cell clone tumor samples were cut on silane-coated glass slides. The sections were deparaffined, rehydrated, and decalcified with EDTA treatment. Samples were then digested with ficin (Digest-All 1, Zymed Laboratories, Inc., San Francisco, CA) for 10 min at 37 C. The endogenous peroxidase activity was blocked by incubating the slides in 3% peroxide in methanol. Nonspecific binding of IgG was minimized by a 30-min preincubation of the slides in normal horse serum (Vector Laboratories, Inc., Burlingame, CA) at room temperature. The primary polyclonal goat antimouse osteocalcin antibody (33 mg/ml; Paesel+Lorei, Hanau, Germany) was applied at a dilution of 1:20,000 in PBS at 4 C overnight. Controls for specificity of osteocalcin staining included incubation of tumor sections without the primary antibody in PBS (negative control) and incubation of femur sections from a young mouse with osteocalcin antibody (positive control). After washes in PBS, the samples were reacted with secondary biotinylated horse antigoat IgG secondary antibody (Vector Laboratories, Inc.) at a 1:200 dilution for 1 h at room temperature. Samples were then incubated with avidin-biotin (Vector Laboratories, Inc.) for 1 h at room temperature. The detection was performed with a DAB kit (Vector Laboratories, Inc.). The samples were reacted with diaminobenzidine, washed, counterstained with Mayer’s hematoxylin, dehydrated, treated with xylene, and mounted.

RNA isolation, RT-PCR, and Southern blotting for RT-PCR
Total RNA was extracted from the cells and tumor homogenates using the guanidium isothiocyanate method (37). The specific antisense primer for FGF-8 (RB418) was designed according to Ghosh et al. (38). RT-PCR for bone morphogenetic protein 4 (BMP-4) was performed according to the method described by Xiao et al. (39), and RT-PCR for FGFR1IIIc, FGFR2IIIc, and FGFR4 was performed according to the procedure reported by Ittman and Mansukhani (40). The specific sense primer for the FGF-8b isoform was 5'-GCC TCC AAG CCC AGG TAA CT-3'. The specific primers were: for BMP-2: sense, 5'-TAT GAA ATC ATA AAA CCT GCA-3'; and antisense 5'-TGT TCA TCT TGG TGC AAA GAC-3'; for BMP-6: sense, 5'-GAT CCT CTG TAC AAC GCC CTG-3'; and antisense, 5'-TCC GCG TCG TTG AGG AAA GCG-3'; for core binding factor 1 (cbfa1): sense, 5'-CCC AGC CAC CTT TAC CTA CA-3'; and antisense, 5'-TAT GGA GTG CTG CTG GTC TG-3'; and for FGFR3IIIc: sense, 5'-ACC CTA CGT TAC CGT GCT CAA-3'; and antisense, 5'-CCG CCA GGC AGG TGT ACT-3'. The predicted sizes of RT-PCR amplification products of mouse FGF-8b, BMP-2, BMP-4, BMP-6, cbfa1, FGFR1IIIc, FGFR2IIIc, FGFR3IIIc, and FGFR4 were 333, 247, 574, 230, 150, 673, 752, 116, and 637 bp, respectively.

For RT-PCR analysis, 2 µg total RNA was used for the RT to make cDNA via reverse transcriptase (Moloney murine leukemia virus, Promega Corp., Madison, WI). cDNA was amplified by Dynazyme II DNA polymerase (Finnzymes, Espoo, Finland). Thirty cycles of annealing and extension [1 min at 94 C, 3 min at 55 C (for BMP-2), 60 C (for FGFR2IIIc), 61 C (for FGFR1IIIc and FGFR4), 62 C (for cbfa1, BMP-4 and FGF-8b), 65 C (for BMP-6), or 66 C (for FGFR3IIIc), and 3 min at 72 C] were carried out with an Eppendorf Mastercycler Gradient (Eppendorf AG, Hamburg, Germany). The products were visualized on 1% NuSieve (3:1) agarose gels (FMC BioProducts, Rockland, ME) and transferred to GeneScreen Plus nylon membranes (DuPont, NEN Life Science Products, Boston, MA). The RT-PCR products of BMP-2, BMP-4, and BMP-6 were confirmed by hybridization with 32P-labeled BMP-2, BMP-4, and BMP-6 cDNAs (gifts from John M. Wozney, Genetics Institute, Inc., Cambridge, MA) under standard conditions. The RT-PCR products of FGF-8b, FGFR2IIIc, and FGFR4 were confirmed by hybridization with 32P-labeled FGF-8 (25), FGFR2IIIc (41), and FGFR4 (42) cDNAs under standard conditions. The RT-PCR-products of cbfa1, FGFR1IIIc, and FGFR3IIIc were confirmed with the oligonucleotide probes 5'-GGT ACG TGT GGT AGT GAG TGG TGG CGG AC-3', 5'-TGT GTA AGG TGT ACA GTG A-3', and 5'-TTC TCT CCT TGC ACA ACG TCA CCT TTG A-3', respectively. The probes were 3' end labeled with {gamma}-32P, and hybridization was carried out overnight at room temperature in standard conditions.

Cell culture
The androgen-regulated mouse breast cancer cell line S115 was maintained in DMEM supplemented (4%) with heat-inactivated fetal bovine serum (i-FBS) and 10 nM testosterone (4-androsten-17-ol-3-one; Sigma-Aldrich Corp., St. Louis, MO). The MG63 cell line was maintained in DMEM supplemented with i-FBS at 10%. The culture media were supplemented with 1 mM L-glutamine (Fluka, Riedel de-Haën, Germany). The DMEM culture medium and FBS were purchased from Invitrogen Life Technologies, Inc. (Paisley, Scotland). The human osteosarcoma cell line MG63 was obtained from the American Type Culture Collection (Manassas, VA).

The cells were grown as monolayer cultures in 100-mm diameter plastic tissue culture dishes (10 ml medium/dish; Nunc, Roskilde, Denmark) at 37 C in a humidified atmosphere of 95% air and 5% CO2 and subcultured at 4- to 6-d intervals.

Preparation of the heparin-binding growth factor fraction (HBGF)
The HBGF was prepared from medium conditioned by S115 cells grown in the presence of testosterone and shown to stimulate the proliferation of S115 cells, as described previously (25). The presence of FGF-8b protein in HBGF was verified in a standard Western blot by FGF-8b antibody (anti-FGF-8b neutralizing antibody, R&D Systems, Inc., Minneapolis, MN) and the secondary antibody, horseradish peroxidase-labeled antigoat IgG (affinity-purified from Dako, Glostrup, Denmark). Protein bands were visualized using the enhanced chemiluminescence detection system (ECL, Amersham Biosciences, Piscataway, NJ). The total FGF-8b protein content in HBGF was approximately 1 µg/ml, as quantified by MicroComputer Imaging Device (Imaging Research, Inc., Ontario, Canada) on Western blot bands of ascending FGF-8b protein concentrations and HBGF.

Measurement of cell growth
To determine the effects of FGF-8b and HBGF on the proliferation of MG-63 cells, [3 H]thymidine incorporation was used. The cells were seeded at a density of 3 x 103/well in 96-well dishes in DMEM and 10% i-FBS. The next day, the medium was changed to serum-free DMEM-Ham’s F-12 containing 0.1% BSA with various concentrations of FGF-8b (R&D Systems, Inc.) or HBGF. The DMEM-Ham’s F-12 culture medium was purchased from Invitrogen Life Technologies, Inc. The neutralizing antibody against FGF-8b was used to identify which FGFs were responsible for the proliferation of osteosarcoma cells in HBGF. The antibody (5 µg/ml) was preincubated with HBGF and FGF-8b for 1 h at 37 C in the presence of 0.1 µg/ml heparin. [3H]Thymidine incorporation was performed as previously described (43).

Bone marrow cell culture and osteoblastic differentiation
Bone marrow cells were maintained in phenol red-free {alpha}-MEM (Invitrogen Life Technologies, Inc.) supplemented with fetal calf serum [15%; Bioclear UK Ltd. (Wilts, UK) or Invitrogen Life Technologies, Inc.], 10 nM dexamethasone (Sigma-Aldrich Corp.), ascorbic acid (50 µg/ml; Merck, Darmstadt, Germany), 10 mM sodium ß-glycerophosphate (Fluka, Buchs, Switzerland), and antibiotics (Invitrogen Life Technologies, Inc.) in a humidified atmosphere of 95% air and 5% CO2 at 37 C. This medium has been shown to favor osteogenic differentiation of mesenchymal stem cells (44, 45). In this article, growth factors (FGF-2, FGF-8b, HBGF, and BMP-2) were added to this culture medium with all the above-mentioned supplements.

Bone marrow cells were obtained from tibiae and femora of 8- to 12-wk-old female NMRI mice bred in the animal center of the University of Turku. The method for studying bone marrow cell proliferation was developed by Qu et al. (44). The animals were euthanized. Soft tissues were detached aseptically from the long bones. Metaphyses from both bone ends were resected and bone marrow cells were flushed from the marrow cavity using a 10-ml syringe, a 27G needle and culture medium.

Nucleated cells were counted in a hemocytometer and cells were dispersed into 6-well plates (Nunc, Roskilde, Denmark) or in T75 tissue culture flasks (Nunc, Roskilde, Denmark; or Becton Dickinson, Franklin Lakes, NJ) at a density of 1 x 104 or 1 x 106 cells/cm2 (proliferation culture; see Fig. 5Go).


Figure 5
View larger version (25K):
[in this window]
[in a new window]
  		FIG. 5. Scheme for bone marrow cultures leading to osteoblastic differentiation. Mouse bone marrow cells were first flushed from the marrow cavity and then plated in six-well plates or tissue culture flasks (proliferation culture). After 6 d of culture, cell numbers were determined, and cells were subcultured in 24-well plates (differentiation culture). After culture periods of 7–21 d, the cultures were analyzed for ALP activity by histochemical staining, for the presence of calcified bone nodules by von Kossa staining, and for the level of calcium extracted from the bone nodules.

 
To determine the effects FGF-8b, HBGF or an FGFR inhibitor PD 173074 (46) (PD 173074 was a gift from Dr. Raz Dewji, Oncology, Pfizer, UK) on bone marrow cell proliferation, these factors were added at various concentrations to the six-well plates in the proliferation culture. Half of the culture medium was replaced on d 3. On d 6, the adherent cells were washed with warm PBS and detached using trypsin-EDTA (Invitrogen Life Technologies, Inc.), or they were scraped using a rubber policeman (Nunc). Cell numbers were counted using a hemocytometer. The assay was repeated twice.

The adherent bone marrow cells grown in flasks during proliferation culture were detached and counted on d 6 of the proliferation culture. The cells were then plated in 24-well plates at a density of 1 x 106 or 2 x 106 cells/well (differentiation culture; see Fig. 5Go). The differentiation cultures were continued for 7–21 d by replacing half the medium every 3 d. For studying the effects of HBGF, FGF-8b, FGF-2 (human recombinant FGF-2, Roche), and BMP-4 (R&D Systems, Inc.) on osteoblast differentiation, the compounds were added to cultures at various concentrations for different time periods. Proliferation culture was performed either with or without FGFs or HBGF to assess the effects of these factors on osteogenic development. To study the effects of these factors on the differentiation of mesenchymal stem cells, differentiation culture was performed with and without the growth factors for d 1–3, 1–7, 1–14, or 1–21 of the differentiation culture.

For alkaline phosphatase (ALP) staining, the differentiation cultures were continued for 6–7 d, and after fixation with 3% paraformaldehyde for 20 min, the cells were stained for ALP using a commercial kit (Sigma-Aldrich Corp.). Cellular ALP activity was assayed using p-nitrophenylphosphate as substrate and in parallel, determining the protein contents of the wells by a protein assay (Bio-Rad Laboratories, Inc., Richmond, CA) as previously described (44). For detection of bone nodules, the differentiation cultures were continued for an additional 14–21 d and fixed as described above. Bone nodules were detected by means of von Kossa staining for calcium. For quantitative measurement of calcium, the differentiation cultures were continued for 14 d. The cells were washed three times with Ca2+- and Mg2+-free PBS and incubated overnight at room temperature in 0.6 N HCl. Extracts (50 µl) were complexed with 1 ml o-cresol-phthalein-complex (Test Combination Calcium, Roche). The colorimetric reaction was read at 570 nm in a 1420 Victor2-multilabel counter (Wallac Perkin-Elmer, Turku, Finland). Absolute calcium concentrations were determined by comparison with a calibrated standard provided by the vendor.

Statistical analyses
Statistical analyses were carried out using Statistica 6.0 software. Normality of distribution was tested by Shapiro-Wilks W test, and statistical significances of differences were tested by means of independent t tests. If the groups were not normally distributed, the Mann-Whitney U test was applied. Results were considered statistically significant at P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Presence of metaplastic bone and cartilage in S115 tumors
Mouse breast cancer cell tumors (S115 cells) and their variant cell line tumors grown sc in nude mice showed bone- or cartilage-like formation inside the tumor in 21% (n = 13) of the cases (n = 63; data combined from two studies) (6). The metaplastic tumors were mostly formed by S115 cells (69% of all metaplastic tumors), but clone 27 (23%) and clone 33 (8%) cells also formed metaplastic tumors.

Histology of ectopic cartilage and expression of cartilage markers in S115 tumors
S115 cell tumors and their variant cell line tumors grown in nude mice contained chondrocyte- or hypertrophic chondrocyte-like cells and cartilaginous matrix as shown by hematoxylin-eosin (Fig. 1AGo) and toluidine blue staining (Fig. 1BGo) of the tumors. The expression of specific markers (type II and IX collagen) for cartilage was colocalized with cartilage-like matrix by in situ hybridization of nude mouse tumors produced by S115 mouse breast cancer cells (Fig. 1Go, C–H). The expression of collagen II (Fig. 1Go, C and D) was found in six of 12 tumors, and that of collagen IX (Fig. 1Go, E and F) was found in two of eight tumors studied. Type II collagen is the major component of the multicomponent collagen fibril network of hyaline cartilage, and types IX and XI are minor components. Type II collagen is known to be expressed later and at a higher level than collagen type IX during mouse embryo development (47), which might explain the higher probability of observing type II collagen expression at a given point of cartilage formation. These findings suggest that cartilage formation is an actual event in certain nude mouse tumors produced by S115 cells and their variant cell lines.


Figure 1
View larger version (111K):
[in this window]
[in a new window]
  		FIG. 1. Ectopic cartilage formation in sc nude mouse tumors of S115 mouse breast cancer cells. Hematoxylin-eosin (A) and toluidine blue (B) staining of S115 tumor containing ectopic cartilage was performed. In situ hybridization of S115 nude mouse tumors with probes specific for type II (C and D) and type IX collagen (E and F) was conducted. Bright (C and E) and dark (D and F)-field photomicrographs show the localization of type II and type IX collagen mRNA. No specific labeling was observed in parallel sections hybridized with type collagen II sense RNA (G) or type IX collagen sense RNA (H). Bar, 100 µm; insets, 10 µm.

 
Histology of ectopic bone and expression of bone markers in S115 tumors
Calcification and actual bone formation were also observed in S115 tumors (Fig. 2Go, A and B), which suggests either induction of endochondral bone formation or direct bone formation. Immunohistochemical staining of EDTA-treated tumor sections for bone-specific osteocalcin (48) showed that bone-like formation inside the tumors most likely represented bone and not, for example, nonspecific calcification due to necrosis (Fig. 2Go, C and D).


Figure 2
View larger version (145K):
[in this window]
[in a new window]
  		FIG. 2. Ectopic bone formation in sc nude mouse tumors of S115 mouse breast cancer cells. Hematoxylin-eosin staining of S115 tumor containing both cartilage and calcified areas (A) and primary bone-like spiculae (B), and immunohistochemical staining of osteocalcin in EDTA-treated tumor sections (C) were performed. A control section without the primary antibody (D) was negative. Bar, 100 µM; insets, 10 µM.

 
BMPs are known to induce ectopic bone in vivo (49). RT-PCR analysis of mRNA from S115 cells and tumors showed that BMP-2, BMP-4, and BMP-6 were expressed in almost all samples studied (Fig. 3Go). All samples expressed mRNA for BMP-2. All except one (Fig. 3Go, lane 6, S115 clone 27 tumor) expressed mRNA for BMP-4, and all except one (lane 10, S115 clone 33 tumor) expressed mRNA for BMP-6. All samples but two (Fig. 3Go, lane 9, S115 cells, and lane 10, S115 clone 33 cells) also expressed mRNA for cbfa1 (also called Runx2), a transcription factor found to be essential in osteogenesis (50). All samples also expressed mRNA for FGF-8b.


Figure 3
View larger version (57K):
[in this window]
[in a new window]
  		FIG. 3. Expression of FGF-8b and various genes associated with bone formation in S115 tumors. Tumors induced by same clones formed metaplastic or nonmetaplastic tumors. Total RNA from sc grown tumors of S115 cells and from S115 clone 22, S115 clone 27, and S115 clone 33 cells was examined by RT-PCR for expression of mRNA for FGF-8b, BMP-2, BMP-4, BMP-6, and cbfa1. Lanes 1–5, Tumors with no bone or cartilage formation; lanes 1–3, S115 clone 27 cells; lane 4, S115 cells; lane 5, S115 clone 22 cells; lanes 6–10, sc grown S115 tumors with bone or cartilage formation; lanes 6 and 7, S115 clone 27 cells; lanes 8 and 9, S115 cells; lane 10, S115 clone 33 cells; lane 11, S115 cell line grown in vitro; lane 12, PC-3 cells (positive control for BMP-2), MC3T3 osteoblastic cells (positive control for BMP-4 and cbfa1), or mouse liver (positive control for BMP-6). Southern hybridization of RT-PCR products was performed using specific cDNA probes for FGF-8b (333 bp), BMP-2 (247 bp), BMP-4 (574 bp), and BMP-6 (230 bp) and a specific oligonucleotide probe for cbfa1 (150 bp). Reactions that were considered negative are marked with an arrowhead.

 
Expression of FGF-8b-specific receptors in osteosarcoma cells
Experiments carried out in vitro have indicated that FGF-8b preferentially signals through the FGFR splice forms FGFR2IIIc, FGFR3IIIc, and FGFR4 (17, 51) and at higher concentrations also through FGFR1IIIc (52). We studied the expression of mRNAs for FGFR-8b in the human osteosarcoma cell line MG63 by means of RT-PCR. The cells expressed the isoforms FGFR1IIIC, FGFR2IIIc, FGFR3IIIc, and FGFR4 (Fig. 4AGo).


Figure 4
View larger version (26K):
[in this window]
[in a new window]
  		FIG. 4. Expression of FGF-8 receptors in MG63 osteosarcoma cells and the effect of FGF-8b and HBGF containing secreted FGF-8b on MG63 cell proliferation. A, Total RNA from MG63 cells was examined by RT-PCR. Products of FGFR1IIIc (673 bp) and FGFR3IIIc (116 bp) were confirmed with specific oligonucleotide probes and products of FGFR2IIIc (752 bp) and FGFR4 (637 bp) were confirmed by hybridization with specific cDNAs. Products were from a negative control (lane 1, H2O), MG63 cells (lane 2), and positive controls (lane 3, MCF-7 cells for FGFR1IIIc and FGFR4, ROS cells for FGFR2IIIc, and a clinical prostate cancer sample for FGFR3IIIc). B, Incorporation of [3H]thymidine by MG63 osteosarcoma cells grown with and without different concentrations of FGF-8b. C, Western blot analysis of HBGF. Proteins were separated by 15% SDS-PAGE and detected by FGF-8b antibody. Lanes 1 and 2, Recombinant FGF-8b protein (lane 1, 2.5 ng; lane 2, 25 ng), which served as a positive control (molecular mass, 24 x103); lane 3, HBGF (total protein content, 1.5 µg) containing FGF-8b in its secreted glycosylated form (molecular mass, 34 x103). D, Incorporation of [3H]thymidine by MG63 cells grown with and without different concentrations of HBGF, and with 1% HBGF with and without a neutralizing antibody against FGF-8b or a purified irrelevant polyclonal goat IgG as a control for the neutralizing antibody. B and D, Mean ± SD (counts per minute per well) of four to six replicate wells cultured in 96-well plates. Normality of distribution was tested by Shapiro-Wilks W test, and an independent t test was used to test the statistical significance of differences between control cultures (no additions) and cultures treated with different concentrations of FGF-8b. Asterisks indicate significant differences between controls and treated cells; *, P < 0.05; **, P < 0.01; ***, P < 0.001. The assay was carried out three times, and similar results were achieved.

 
Effects of FGF-8b and heparin-Sepharose-enriched fraction of medium conditioned by S115 cells (HBGF) on osteosarcoma cell proliferation
The ability of osteoblastic cells to respond to FGF-8b or HBGF was tested by studying the effects of these growth factors on the proliferation of osteosarcoma cells. Recombinant mouse FGF-8b repeatedly enhanced the incorporation of [3H]thymidine into DNA in a dose-dependent manner in the MG63 cell line (Fig. 4BGo). FGF-8b (50 and 100 ng/ml) significantly stimulated the proliferation of MG63 cells on d 2 and 3.

Western blot analysis revealed the presence of FGF-8b protein (molecular mass, ~34 x 103) in HBGF (Fig. 4CGo). Next, the effect of HBGF on MG63 cells was studied (Fig. 4DGo). HBGF also repeatedly increased the incorporation of [3H]thymidine into DNA in MG63 cells dose dependently. The effect was statistically significant with 1% HBGF on d 1 and with all concentrations used on d 2 and 3 (0.1%, 0.5%, and 1% HBGF; P < 0.001). To study whether FGF-8b has a role in HBGF stimulation of osteosarcoma cells, a neutralizing antibody to FGF-8b (R&D Systems) was used with a purified irrelevant polyclonal goat IgG antibody as a control. The FGF-8b antibody almost totally abolished HBGF stimulation of proliferation of MG63 osteosarcoma cells (P < 0.001), which suggests that FGF-8b has a major role in HBGF stimulation of osteoblastic cells. This effect was more significant compared with the nonspecific effect of IgG on HBGF stimulation of the cells on d 2 and 3 (P < 0.05). The antibody or IgG alone did not have any significant effect on the proliferation of MG63 cells. The FGF-8b antibody was also able to block FGF-8b (100 ng/ml) stimulation of MG63 cell proliferation (data not shown).

Effects of FGF-8b and HBGF on bone marrow cell proliferation
It has been postulated that the effects of FGFs are dependent on the stage of osteoblast maturation (27). Therefore, we tested the effects of FGF-8b and HBGF at various stages of osteoblastic differentiation using cultures of normal bone marrow cells (44). In our model (Fig. 5Go), bone marrow cell proliferation is first induced by various factors (proliferation culture), followed by osteoblastic differentiation and matrix formation (differentiation culture). We first studied the effects of FGF-8b and HBGF on bone marrow cell proliferation. Addition of FGF-8b to the proliferation culture stimulated the proliferation of bone marrow cells (Fig. 6AGo). A maximum effect was obtained with a concentration of 20 ng/ml (P < 0.01). Addition of HBGF also resulted in dose-dependent stimulation of bone marrow cell proliferation in the proliferation culture (Fig. 6BGo). Even a low concentration of HBGF (0.25%) was able to stimulate proliferation compared with the control, and more marked stimulation was obtained with 1% HBGF.


Figure 6
View larger version (12K):
[in this window]
[in a new window]
  		FIG. 6. Effects of HBGF, FGF-8b, and FGFR inhibitor PD 173074 on the proliferation of mouse bone marrow cells in culture. The numbers of mouse bone marrow cells grown with and without different concentrations of FGF-8b (A), HBGF (B), or PD 173074 (C) in proliferation culture for 6 d were counted using a hemocytometer. Shown are the mean cell number per well ± SD of three to eight replicate wells cultured in 24-well plates. Normality of distribution was tested by means of Shapiro-Wilks W test, and an independent t test was used to test the statistical significance of differences between control cultures (no additions) and cells treated with different concentrations of FGF-8b or HBGF. C, The groups were not normally distributed, and thus, a Mann-Whitney U test was applied to test for statistical significances between control cultures (no PD 173074) and PD 173074. Asterisks indicate statistically significant differences between controls and treated cells; *, P < 0.05; **, P < 0.01; ***, P < 0.001. The assay was carried out three times, and similar results were achieved.

 
Osteoblasts secrete FGFs, for example, FGF-2, which probably act as autocrine or paracrine mitogens for bone cells (30). Next we wanted to determine whether HBGF-induced proliferation of bone marrow cells could be affected by inhibiting FGFR signaling. PD 173074, a low molecular weight compound that selectively blocks tyrosine kinase activity and autophosphorylation of FGFR (46), was used at a 1-µM concentration. This compound has been previously shown to have no effect on the proliferation of cells that lack FGFRs when used at a 2-µM concentration (53). The repression of FGFR activity caused a significant inhibition of 68% of bone marrow cells compared with control (P < 0.01; Fig. 6CGo). The effect was not likely to be toxic, because viable colony formation was observed in both groups. When the cells were stimulated with 1% HBGF or 25 ng/ml FGF-8b, the inhibition was 42% (P < 0.05) and 70% (P < 0.01), respectively, suggesting that FGFs could play a role in HBGF-induced stimulation of bone marrow cells.

Effects of FGF-8b and HBGF on osteoblastic differentiation in bone marrow cell cultures
The effects of FGF-8b and HBGF on different stages of osteoblast differentiation and bone matrix formation were next studied by adding them to mouse bone marrow cultures for different time periods (Fig. 5Go). When the cells were treated continuously with various concentrations of FGF-8b or HBGF through both the proliferation (6 d) and differentiation (14–21 d) culture phases, bone nodule formation was inhibited in a dose-dependent manner, as judged on the basis of von Kossa and ALP staining (data not shown).

Next, HBGF (1%) was added to proliferation culture only (not to differentiation culture at all) or to both proliferation culture and the first 3 d of differentiation culture (Fig. 7Go). At the end of the experiment, on d 14 of the differentiation culture, bone nodules were detected by means of von Kossa staining for calcium. HBGF clearly increased the formation of bone nodules vs. controls, in which macroscopic bone nodules were not yet visible. If HBGF was present all the time, no nodule formation was observed. These results suggest that HBGF containing secreted FGF-8 increases bone formation if it is applied at an early phase of osteoblast differentiation.


Figure 7
View larger version (42K):
[in this window]
[in a new window]
  		FIG. 7. Effect of HBGF (1%) containing secreted FGF-8 on matrix mineralization in cultures of mouse bone marrow cells. HBGF was present on d 1–6 of proliferation culture only or in both proliferation culture and d 1–3 or 1–14 of differentiation culture. Bone nodule formation was visualized by von Kossa staining on d 14 of the differentiation cultures. The results are representative of three different experiments.

 
For a more detailed and quantitative analysis of the role of FGF-8 in bone formation, the calcium contents extracted from the bone nodules were next determined (Fig. 8Go). FGF-2 and BMP-4 were used as positive controls, because they have been reported to stimulate the differentiation of osteoblasts (54, 55). First, FGF-8b or FGF-2 at 25 ng/ml was added only to proliferation cultures of bone marrow cells, and differentiation culture was subcultured without FGF-8b for 14 d, similar to the conditions shown in Fig. 7Go. Addition of FGF-8b resulted in an almost 15-fold increase in the calcium content of mouse osteoblast culture vs. controls (Fig. 8AGo; P < 0.001). This effect was even stronger than that produced by FGF-2 (P < 0.001). FGF-8b clearly induced the osteoblastic capacity of bone marrow cells.


Figure 8
View larger version (21K):
[in this window]
[in a new window]
  		FIG. 8. Effects of FGF-8b (25 ng/ml), FGF-2 (25 ng/ml), and BMP-4 (10 ng/ml) on matrix mineralization in mouse bone marrow-derived cultures. The total calcium content of the cultures was measured on d 14 of the differentiation culture. FGF-2 and BMP-4 were used as positive controls. A, No additions in differentiation cultures. A and B, FGF-8b was added in proliferation culture. A and C, FGF-2 was added in proliferation culture. B–D, BMP-4 and FGF-8b were added for d 1–3 of differentiation culture, and FGF-8b was also added for d 1–14 of differentiation culture. D, No additions in proliferation culture. The controls containing no FGF for B and C can be found in A. Columns show the mean ± SD (millimoles per well) of six replicate wells cultured in 24-well plates. Normality of distribution was tested by means of Shapiro-Wilks W test, and an independent t test was used to test the statistical significance of differences between control cultures (no additions) and cells treated with FGF-8b, FGF-2, or BMP-4. Asterisks indicate statistically significant differences between controls and treated cells; *, P < 0.05; **, P < 0.01; ***, P < 0.001. Similar results were obtained from two separate experiments.

 
When FGF-8b or BMP-4 was also added to differentiation cultures for 1–3 d after pretreatment with FGF-8b or FGF-2 in the proliferation culture, the induction of matrix mineralization was further increased. Figure 8Go, B and C, shows an approximately 11% increase with FGF-8b and a 24–26% increase with BMP-4. The stimulatory effects of FGF-8b and BMP-4 did not differ statistically from each other in either FGF-8b- or FGF-2-pretreated cultures. Matrix mineralization, as measured by calcium accumulation, was decreased if FGF treatment was continued throughout the entire proliferation and differentiation phases (Fig. 8Go, B and C). FGF-8b inhibited calcium accumulation of osteoblasts by 53% vs. the controls when the growth factor was present during the total length of culture (Fig. 8BGo). The situation was similar with FGF-2-pretreated osteoblasts; FGF-8b in differentiation culture inhibited calcium accumulation by 28% vs. the controls (Fig. 8CGo). No significant changes were observed if FGF-8b was added for the entire length of the differentiation phase only (Fig. 8DGo).

Finally, FGF-8b (at 25 ng/ml) or BMP-4 was added on d 1–3 of mouse osteoblast differentiation culture, but not to the proliferation culture (Fig. 8DGo). Matrix mineralization, measured by the calcium content of the cultures, was increased by both FGF-8b (4-fold; P < 0.05) and BMP-4 (2-fold; P < 0.05). There was no statistically significant difference between the effects of FGF-8b and BMP-4.

To study whether FGF-8 and HBGF induce osteoblast-specific ALP production, we then added FGF-8b (at 25 ng/ml) or HBGF (1%) to proliferation cultures of bone marrow cells on d 1–6, but not to the osteoblast differentiation cultures. The ALP activity of osteoblasts in this mouse osteoblast culture model has been shown to peak on d 7–8 of the differentiation culture (44). ALP activity was thus measured on d 7 of the differentiation culture by staining the wells with a commercial stain for ALP or by measuring the cellular ALP activities. The results clearly showed that the presence of these growth factors in the proliferation culture increased osteoblast-specific ALP production at later stages of osteoblast differentiation (Fig. 9AGo). The presence of FGF-8b or HBGF in the proliferation culture increased cellular ALP activity of osteoblasts by 33% (P < 0.01) and 37% (P < 0.01) compared with the control, respectively (Fig. 9BGo).


Figure 9
View larger version (20K):
[in this window]
[in a new window]
  		FIG. 9. Effects of FGF-8b (25 ng/ml) or HBGF (1%) on specific activity of ALP in mouse bone marrow-derived cultures. The growth factors were added to the proliferation culture, and ALP activity of the cultures was measured on d 7 of the differentiation culture by staining the cells with a commercial kit (A) or by measuring the enzyme activity colorimetrically using p-nitrophenylphosphate as substrate (B). In parallel, the protein contents of the wells were determined. The columns show the mean ± SD (units per milligram of protein) of six replicate wells cultured in 24-well plates. B, The groups were not normally distributed, and thus, a Mann-Whitney U test was applied to test the statistically significant differences between control cultures (no additions) and cells treated with FGF-8b or HBGF. Asterisks indicate statistically significant differences between controls and treated cells; **, P < 0.01.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Breast tumors occasionally contain ectopic cartilage and bone-like differentiation (1). Metastasized breast cancer, another pathological process involving bone formation in cancer, is almost invariably associated with bone lesions, which are mostly osteolytic, but can also be osteosclerotic (5). The mechanisms behind osteolytic lesions have been characterized to some extent, but those behind osteosclerotic changes are much less understood. The occurrence of ectopic cartilage and bone formation in experimental S115 mouse tumors prompted us to ask which S115 cell-produced factors are involved in osteoblast differentiation. We especially looked into the possible role of FGF-8, which is expressed at a high level in S115 cells and regulates the malignant behavior and tumorigenesis of these cells in many ways (11, 25).

The histological and histochemical analyses showed that both osseous and chondroid metaplasia observed in S115 tumors represented real hyaline-type cartilage and bone formation. This was confirmed by the results of immunohistochemistry, which showed that the bone-specific osteocalcin (48) was expressed in the osseous areas of the tumors. In addition, in situ hybridization studies demonstrated that the chondroid areas expressed type II and IX collagen, which are specific markers of cartilage (47). The presence of cartilaginous matrix also suggests that the ectopic bone observed could be formed by endochondral-type ossification. Ectopic cartilage and bone were observed in tumors formed by several S115 cell sublines expressing various malignant phenotypes (7). A common feature of all those lines is an increased rate of proliferation and expression of FGF-8b at a high level in the presence of androgens (25). Terada et al. (9) and Nagoshi et al. (10) previously reported chondrous metaplasia in original S115 tumors in androgen-depleted mice primarily, but we found metaplastic changes in tumors implanted into intact mice.

Several hypotheses for ectopic bone formation inside tumors have been proposed. Most investigators support a primarily monoclonal origin of the various components (56). Some have claimed that cartilage and bone develop from epithelium-derived (1, 2) or myoepithelium-derived (3) cancer cells. Even S115 tumor cells have been suggested to turn chondrogenic in metaplastic tumors (8, 9, 10). Recently, Huang et al. (57) have shown that conditioned media collected from selected human prostate cancer cells lines induce prostate cancer cells to produce bone sialoprotein and human osteocalcin; they call this phenotype osteomimicry. It is also possible that mesenchymal stem cells normally present in soft connective tissues (58) are induced to differentiate in a chondrocytic or osteoblastic direction in a tumor environment. Although these metaplastic breast tumors with ectopic bone are rare, microcalcifications occur in 30–50% of breast cancers (59). Both states are associated with a poor survival (2, 59), but there have been no published studies that would link the occurrence of breast cancer microcalcifications to a later risk of developing a metaplastic tumor in the breast.

BMPs have generally been considered to be major factors controlling ectopic ossification (60). In this study, S115 cells and tumors were found to express mRNA for BMP-2, BMP-4, BMP-6, and cbfa1, which all are important in normal cartilage and/or bone formation. Interestingly, FGF-8b, which is secreted abundantly by S115 cells, was overexpressed by all S115 tumors compared with the expression of other factors studied. Considering that FGF-8 is expressed in embryonic skeletal tissues (61), we hypothesized that it might contribute to osteoblast differentiation.

The bone-forming capacity of FGF-8 was tested using osteosarcoma cell lines and primary mouse bone marrow cultures. HBGF from S115 cells containing secreted FGF-8, and exogenous FGF-8b both increased the proliferation of human MG63 cells, which also expressed FGFRs (FGFR1IIIc, FGFR2IIIc, FGFR3IIIc, and FGFR4) known to be activated by FGF-8b. Importantly, HBGF-induced stimulation of proliferation was opposed by the addition of a neutralizing anti-FGF-8b antibody, which strongly suggests that it was FGF-8b that was responsible for a major part of the stimulatory effect of HBGF on MG63 proliferation.

Mesenchymal stem cells from the bone marrow are known to differentiate to several specific lineages, including osteoblasts, chondroblasts, adipocytes, etc., in vitro depending on culture conditions (62). We decided to study the possible osteogenic potential of FGF-8 using a well-defined in vitro model to study differentiation of mesenchymal stem cells toward the osteogenic lineage. We found that both FGF-8b and HBGF regulated cell proliferation and subsequent bone formation in primary cultures of mouse bone marrow cells. The net outcome of the response, however, was very much dependent on the timing of FGF-8b and HBGF treatments. A marked stimulation of bone formation was obtained only if FGF-8b or HBGF were added during the first phase of bone marrow culture (proliferation cultures). A specific FGFR inhibitor efficiently blocked FGF-mediated proliferation of bone marrow cells in the proliferation culture. Stimulation of bone formation was also achieved by adding FGF-8b to the second phase of bone marrow cultures (differentiation cultures) for the first 3 d. If FGF-8b was present in culture medium throughout the whole of the differentiation culture period (14 d), no stimulation was observed at all, and its presence through both proliferation and differentiation culture periods (20 d) inhibited bone formation. These results suggest that the effect of FGF-8 is targeted to early stages of osteoblast differentiation, and that this effect is a prerequisite for later differentiation, which, in contrast, can be inhibited by the continuous presence of this growth factor. One explanation for a culture stage-dependent biological response could be different profiles of FGFRs at various stages of osteoblast differentiation.

Corresponding results were also obtained with FGF-2, which is in line with previous reports (55). Addition of BMP-4 to second phase cultures augmented both FGF-8b- and FGF-2-induced bone formation. It is notable, however, that FGF-8b was clearly more effective than FGF-2 and at least as effective as BMP-4 in the induction of bone formation in these bone marrow cell cultures. Both FGF-2 and BMP-4 have been previously reported to be strong inducers of bone formation (54, 55).

The importance of FGF-8 in bone formation has been difficult to assess, because FGF-8 knockout mice die before skeletal development (19). Conditional knockouts targeted to the apical ectodermal ridge have revealed, however, an important role of FGF-8 in limb bud development (19). Recently, FGF-8 was found to regulate cartilage formation in the vertebrate skull (63) and to influence rib development (64). There are also reports on stimulation by FGF-8 of avian chondrocytes (65) and cultured dental mesenchyme (66) in vitro. The wide expression of FGF-8 in the developing skeleton suggests that FGF-8 along with FGF-17 (61) are closely involved in cartilage and bone formation. Interestingly, FGF-8 and FGF-2 have been shown to induce the expression of an osteoblast-specific transcription factor, cbfa1, in fibroblasts; this induction may be mediated by up-regulation of MAPK, an important pathway for FGF function (29). Our results, however, are the first to demonstrate the effects of FGF-8 on osteoblast differentiation and bone formation in vitro. In the light of these results, it would be interesting to observe the effects of conditional knockout of FGF-8 in bone.

In summary, our results show that FGF-8 induces osteoblast differentiation. Our results also demonstrate that FGF-8 is able to stimulate the proliferation of cultured mouse bone marrow cells efficiently and to induce their early stage differentiation as a prerequisite for later bone nodule formation and osteoblast-specific ALP production. Continued exposure of osteoblastic cultures to FGF-8, however, leads to inhibition of bone formation. It is conceivable that FGF-8 stimulates the proliferation of hypothetical osteogenic stem cells in bone marrow and tumors and promotes their osteogenic potential, leading to osteoblastic differentiation and bone formation.


   Acknowledgments
 
Paula Hakala and Soili Jussila are warmly thanked for their excellent technical assistance.


   Footnotes
 
This work was supported by the Academy of Finland, the Finnish Cancer Societies, and the Sigrid Juselius Foundation (to P.L.H.). M.P.V. is a graduate student at Turku Graduate School in Biomedical Sciences.

The authors have nothing to declare.

First Published Online January 26, 2006

Abbreviations: ALP, Alkaline phosphatase; BMP, bone morphogenetic protein; cbfa1, core binding factor 1; FGF, fibroblast growth factor; FGFR, FGF receptor; HBGF, heparin-binding growth factor fraction of S115 cell medium; i-FBS, heat-inactivated fetal bovine serum.

Received November 28, 2005.

Accepted for publication January 19, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Kaufman MW, Marti JR, Gallager HS, Hoehn JL 1984 Carcinoma of the breast with pseudosarcomatous metaplasia. Cancer 53:1908–1917[CrossRef][Medline]
  2. Chhieng C, Cranor M, Lesser ME, Rosen PP 1998 Metaplastic carcinoma of the breast with osteocartilaginous heterologous elements. Am J Surg Pathol 22:188–194[CrossRef][Medline]
  3. Wargotz ES, Norris HJ 1989 Metaplastic carcinomas of the breast. I. Matrix-producing carcinoma. Hum Pathol 20:628–635[CrossRef][Medline]
  4. Bellino R, Arisio R, D’Addato F, Attini R, Durando A, Danese S, Bertone E, Grio R, Massobrio M 2003 Metaplastic breast carcinoma: pathology and clinical outcome. Anticancer Res 23:669–673[Medline]
  5. Mundy GR 2002 Metastasis to bone: causes, consequences and therapeutic opportunities. Nat Rev Cancer 2:584–593[CrossRef][Medline]
  6. Vainiomaki M, Qu Q, Tuomela J, Vaananen K, Harkonen P, Role of FGF-8 in S115 breast cancer cell-induced cartilage and bone-like differentiation in nude mouse tumors. Program of Keystone Symposium: Molecular Targets for Cancer Therapy, Banff, Alberta, Canada, 2003, p 90 (Abstract)
  7. Harkonen PL, Laaksonen EI, Valve EM, Solic N, Darbre PD 1990 Temperature-sensitive mutants for steroid-sensitive growth of S115 mouse mammary tumor cells. Exp Cell Res 186:288–298[CrossRef][Medline]
  8. Kitamura Y, Okamoto S, Hayata I, Uchida N, Yamaguchi K, Matsumoto K 1979 Development of androgen-independent spindle cell tumors from androgen-dependent medullary Shionogi carcinoma 115 in androgen-depleted nude mice. Cancer Res 39:4713–4719[Abstract/Free Full Text]
  9. Terada N, Yamamoto R, Uchida N, Takada T, Ishiguro S, Taniguchi H, Takatsuka D, Tsujimoto M, Li W, Matsumoto K, Kitamura Y 1987 Development of cartilage-like tissue from androgen-dependent Shionogi carcinoma 115 in androgen-depleted hosts. Lab Invest 57:186–192[Medline]
  10. Nagoshi J, Nomura S, Uchida N, Hirota S, Ito A, Nakase T, Hirakawa K, Shiozaki H, Mori T, Kitamura Y 1994 Expression of genes encoding connective tissue proteins in androgen-dependent SC115 tumors after androgen removal. Lab Invest 70:210–216[Medline]
  11. Tanaka A, Miyamoto K, Minamino N, Takeda M, Sato B, Matsuo H, Matsumoto K 1992 Cloning and characterization of an androgen-induced growth factor essential for the androgen-dependent growth of mouse mammary carcinoma cells. Proc Natl Acad Sci USA 89:8928–8932[Abstract/Free Full Text]
  12. Powers CJ, McLeskey SW, Wellstein A 2000 Fibroblast growth factors, their receptors and signaling. Endocr Relat Cancer 7:165–197[Abstract]
  13. Brem H, Klagsbrun M 1992 The role of fibroblast growth factors and related oncogenes in tumor growth. Cancer Treat Res 63:211–231[Medline]
  14. Dickson RB, Lippman ME 1995 Growth factors in breast cancer. Endocr Rev 16:559–589[CrossRef][Medline]
  15. Giri D, Ropiquet F, Ittmann M 1999 Alterations in expression of basic fibroblast growth factor (FGF) 2 and its receptor FGFR-1 in human prostate cancer. Clin Cancer Res 5:1063–1071[Abstract/Free Full Text]
  16. MacArthur CA, Lawshe A, Shankar DB, Heikinheimo M, Shackleford GM 1995 FGF-8 isoforms differ in NIH3T3 cell transforming potential. Cell Growth Differ 6:817–825[Abstract]
  17. Blunt AG, Lawshe A, Cunningham ML, Seto ML, Ornitz DM, MacArthur CA 1997 Overlapping expression and redundant activation of mesenchymal fibroblast growth factor (FGF) receptors by alternatively spliced FGF-8 ligands. J Biol Chem 272:3733–3738[Abstract/Free Full Text]
  18. Lewandoski M, Meyers EN, Martin GR 1997 Analysis of Fgf8 gene function in vertebrate development. Cold Spring Harbor Symp Quant Biol 62:159–168[Abstract/Free Full Text]
  19. Moon AM, Capecchi MR 2000 Fgf8 is required for outgrowth and patterning of the limbs. Nat Genet 26:455–459[CrossRef][Medline]
  20. Crossley PH, Martinez S, Martin GR 1996 Midbrain development induced by FGF8 in the chick embryo. Nature 380:66–68[CrossRef][Medline]
  21. Valve E, Penttila TL, Paranko J, Harkonen P 1997 FGF-8 is expressed during specific phases of rodent oocyte and spermatogonium development. Biochem Biophys Res Commun 232:173–177[CrossRef][Medline]
  22. Tanaka A, Furuya A, Yamasaki M, Hanai N, Kuriki K, Kamiakito T, Kobayashi Y, Yoshida H, Koike M, Fukayama M 1998 High frequency of fibroblast growth factor (FGF) 8 expression in clinical prostate cancers and breast tissues, immunohistochemically demonstrated by a newly established neutralizing monoclonal antibody against FGF 8. Cancer Res 58:2053–2056[Abstract/Free Full Text]
  23. Valve E, Martikainen P, Seppanen J, Oksjoki S, Hinkka S, Anttila L, Grenman S, Klemi P, Harkonen P 2000 Expression of fibroblast growth factor (FGF)-8 isoforms and FGF receptors in human ovarian tumors. Int J Cancer 88:718–725[CrossRef][Medline]
  24. Gnanapragasam VJ, Robinson MC, Marsh C, Robson CN, Hamdy FC, Leung HY 2003 FGF8 isoform b expression in human prostate cancer. Br J Cancer 88:1432–1438[CrossRef][Medline]
  25. Ruohola JK, Valve EM, Vainikka S, Alitalo K, Harkonen PL 1995 Androgen and fibroblast growth factor (FGF) regulation of FGF receptors in S115 mouse mammary tumor cells. Endocrinology 136:2179–2188[Abstract]
  26. Mattila MM, Ruohola JK, Valve EM, Tasanen MJ, Seppanen JA, Harkonen PL 2001 FGF-8b increases angiogenic capacity and tumor growth of androgen-regulated S115 breast cancer cells. Oncogene 20:2791–2804[CrossRef][Medline]
  27. Hurley M, Marie P, Florkiewicz R 2002 Fibroblast growth factor (FGF) and FGF receptor families in bone. In: Bilezikian J, ed. Principles of bone biology, 2nd ed. New York: Academic Press; 825–851
  28. Deng C, Wynshaw-Boris A, Zhou F, Kuo A, Leder P 1996 Fibroblast growth factor receptor 3 is a negative regulator of bone growth. Cell 84:911–921[CrossRef][Medline]
  29. Zhou YX, Xu X, Chen L, Li C, Brodie SG, Deng CX 2000 A Pro250Arg substitution in mouse Fgfr1 causes increased expression of Cbfa1 and premature fusion of calvarial sutures. Hum Mol Genet 9:2001–2008[Abstract/Free Full Text]
  30. Globus RK, Plouet J, Gospodarowicz D 1989 Cultured bovine bone cells synthesize basic fibroblast growth factor and store it in their extracellular matrix. Endocrinology 124:1539–1547[Abstract]
  31. Gonzalez AM, Hill DJ, Logan A, Maher PA, Baird A 1996 Distribution of fibroblast growth factor (FGF)-2 and FGF receptor-1 messenger RNA expression and protein presence in the mid-trimester human fetus. Pediatr Res 39:375–385[Medline]
  32. Rice DP, Aberg T, Chan Y, Tang Z, Kettunen PJ, Pakarinen L, Maxson RE, Thesleff I 2000 Integration of FGF and TWIST in calvarial bone and suture development. Development 127:1845–1855[Abstract]
  33. Izbicka E, Dunstan C, Esparza J, Jacobs C, Sabatini M, Mundy GR 1996 Human amniotic tumor that induces new bone formation in vivo produces growth-regulatory activity in vitro for osteoblasts identified as an extended form of basic fibroblast growth factor. Cancer Res 56:633–636[Abstract/Free Full Text]
  34. Ohbayashi N, Shibayama M, Kurotaki Y, Imanishi M, Fujimori T, Itoh N, Takada S 2002 FGF18 is required for normal cell proliferation and differentiation during osteogenesis and chondrogenesis. Genes Dev 16:870–879[Abstract/Free Full Text]
  35. Chikazu D, Katagiri M, Ogasawara T, Ogata N, Shimoaka T, Takato T, Nakamura K, Kawaguchi H 2001 Regulation of osteoclast differentiation by fibroblast growth factor 2: stimulation of receptor activator of nuclear factor {kappa}B ligand/osteoclast differentiation factor expression in osteoblasts and inhibition of macrophage colony-stimulating factor function in osteoclast precursors. J Bone Miner Res 16:2074–2081[CrossRef][Medline]
  36. Sahni M, Ambrosetti DC, Mansukhani A, Gertner R, Levy D, Basilico C 1999 FGF signaling inhibits chondrocyte proliferation and regulates bone development through the STAT-1 pathway. Genes Dev 13:1361–1366[Abstract/Free Full Text]
  37. Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ 1979 Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18:5294–5299[CrossRef][Medline]
  38. Ghosh AK, Shankar DB, Shackleford GM, Wu K, T’Ang A, Miller GJ, Zheng J, Roy-Burman P 1996 Molecular cloning and characterization of human FGF8 alternative messenger RNA forms. Cell Growth Differ 7:1425–1434[Abstract]
  39. Xiao G, Gopalakrishnan R, Jiang D, Reith E, Benson MD, Franceschi RT 2002 Bone morphogenetic proteins, extracellular matrix, and mitogen-activated protein kinase signaling pathways are required for osteoblast-specific gene expression and differentiation in MC3T3–E1 cells. J Bone Miner Res 17:101–110[CrossRef][Medline]
  40. Ittman M, Mansukhani A 1997 Expression of fibroblast growth factors (FGFs) and FGF receptors in human prostate. J Urol 157:351–356[CrossRef][Medline]
  41. Dionne CA, Crumley G, Bellot F, Kaplow JM, Searfoss G, Ruta M, Burgess WH, Jaye M, Schlessinger J 1990 Cloning and expression of two distinct high-affinity receptors cross-reacting with acidic and basic fibroblast growth factors. EMBO J 9:2685–2692[Medline]
  42. Partanen J, Makela TP, Eerola E, Korhonen J, Hirvonen H, Claesson-Welsh L, Alitalo K 1991 FGFR-4, a novel acidic fibroblast growth factor receptor with a distinct expression pattern. EMBO J 10:1347–1354[Medline]
  43. Siren MJ, Vainiomaki M, Vaananen K, Harkonen P 2004 {alpha}-Trinositol inhibits FGF-stimulated growth of smooth muscle and breast cancer cells. Biochem Biophys Res Commun 325:691–697[CrossRef][Medline]
  44. Qu Q, Perala-Heape M, Kapanen A, Dahllund J, Salo J, Vaananen HK, Harkonen P 1998 Estrogen enhances differentiation of osteoblasts in mouse bone marrow culture. Bone 22:201–209[Medline]
  45. Friedenstein AJ, Chailakhyan RK, Gerasimov UV 1987 Bone marrow osteogenic stem cells: in vitro cultivation and transplantation in diffusion chambers. Cell Tissue Kinet 20:263–272[Medline]
  46. Mohammadi M, Froum S, Hamby JM, Schroeder MC, Panek RL, Lu GH, Eliseenkova AV, Green D, Schlessinger J, Hubbard SR 1998 Crystal structure of an angiogenesis inhibitor bound to the FGF receptor tyrosine kinase domain. EMBO J 17:5896–5904[CrossRef][Medline]
  47. Perala M, Savontaus M, Metsaranta M, Vuorio E 1997 Developmental regulation of mRNA species for types II, IX and XI collagens during mouse embryogenesis. Biochem J 324:209–216[Medline]
  48. Price PA, Poser JW, Raman N 1976 Primary structure of the {gamma}-carboxyglutamic acid-containing protein from bovine bone. Proc Natl Acad Sci USA 73:3374–3375[Abstract/Free Full Text]
  49. Urist MR 1965 Bone: formation by autoinduction. Science 150:893–899[Abstract/Free Full Text]
  50. Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T 1997 Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89:755–764[CrossRef][Medline]
  51. Ornitz DM, Xu J, Colvin JS, McEwen DG, MacArthur CA, Coulier F, Gao G, Goldfarb M 1996 Receptor specificity of the fibroblast growth factor family. J Biol Chem 271:15292–15297[Abstract/Free Full Text]
  52. MacArthur CA, Lawshe A, Xu J, Santos-Ocampo S, Heikinheimo M, Chellaiah AT, Ornitz DM 1995 FGF-8 isoforms activate receptor splice forms that are expressed in mesenchymal regions of mouse development. Development 121:3603–3613[Abstract]
  53. Koziczak M, Holbro T, Hynes NE 2004 Blocking of FGFR signaling inhibits breast cancer cell proliferation through downregulation of D-type cyclins. Oncogene 23:3501–3508[CrossRef][Medline]
  54. Hughes FJ, Collyer J, Stanfield M, Goodman SA 1995 The effects of bone morphogenetic protein-2, -4, and -6 on differentiation of rat osteoblast cells in vitro. Endocrinology 136:2671–2677[Abstract]
  55. Zhang X, Sobue T, Hurley MM 2002 FGF-2 increases colony formation, PTH receptor, and IGF-1 mRNA in mouse marrow stromal cells. Biochem Biophys Res Commun 290:526–531[CrossRef][Medline]
  56. Lien HC, Lin CW, Mao TL, Kuo SH, Hsiao CH, Huang CS 2004 p53 overexpression and mutation in metaplastic carcinoma of the breast: genetic evidence for a monoclonal origin of both the carcinomatous and the heterogeneous sarcomatous components. J Pathol 204:131[CrossRef][Medline]
  57. Huang WC, Xie Z, Konaka H, Sodek J, Zhau HE, Chung LW 2005 Human osteocalcin and bone sialoprotein mediating osteomimicry of prostate cancer cells: role of cAMP-dependent protein kinase A signaling pathway. Cancer Res 65:2303–2313[Abstract/Free Full Text]
  58. Caplan AI 1994 The mesengenic process. Clin Plast Surg 21:429–435[Medline]
  59. Morgan MP, Cooke MM, McCarthy GM 2005 Microcalcifications associated with breast cancer: an epiphenomenon or biologically significant feature of selected tumors? J Mammary Gland Biol Neoplasia 10:181–187[CrossRef][Medline]
  60. Maroulakou IG, Shibata MA, Anver M, Jorcyk CL, Liu M, Roche N, Roberts AB, Tsarfaty I, Reseau J, Ward J, Green JE 1999 Heterotopic endochondrial ossification with mixed tumor formation in C3(1)/Tag transgenic mice is associated with elevated TGF-ß1 and BMP-2 expression. Oncogene 18:5435–5447[CrossRef][Medline]
  61. Xu J, Lawshe A, MacArthur CA, Ornitz DM 1999 Genomic structure, mapping, activity and expression of fibroblast growth factor 17. Mech Dev 83:165–178[CrossRef][Medline]
  62. Gregory CA, Prockop DJ, Spees JL 2005 Non-hematopoietic bone marrow stem cells: molecular control of expansion and differentiation. Exp Cell Res 306:330–335[CrossRef][Medline]
  63. Walshe J, Mason I 2003 Fgf signalling is required for formation of cartilage in the head. Dev Biol 264:522–536[CrossRef][Medline]
  64. Huang R, Stolte D, Kurz H, Ehehalt F, Cann GM, Stockdale FE, Patel K, Christ B 2003 Ventral axial organs regulate expression of myotomal Fgf-8 that influences rib development. Dev Biol 255:30–47[CrossRef][Medline]
  65. Praul CA, Ford BC, Leach RM 2002 Effect of fibroblast growth factors 1, 2, 4, 5, 6, 7, 8, 9, and 10 on avian chondrocyte proliferation. J Cell Biochem 84:359–366[CrossRef][Medline]