The authors have no conflict of interest.
Basic Fibroblast Growth Factor Stimulates Osteoclast Recruitment, Development, and Bone Pit Resorption in Association With Angiogenesis In Vivo on the Chick Chorioallantoic Membrane and Activates Isolated Avian Osteoclast Resorption In Vitro†
Article first published online: 1 OCT 2002
Copyright © 2002 ASBMR
Journal of Bone and Mineral Research
Volume 17, Issue 10, pages 1859–1871, October 2002
How to Cite
Collin-Osdoby, P., Rothe, L., Bekker, S., Anderson, F., Huang, Y. and Osdoby, P. (2002), Basic Fibroblast Growth Factor Stimulates Osteoclast Recruitment, Development, and Bone Pit Resorption in Association With Angiogenesis In Vivo on the Chick Chorioallantoic Membrane and Activates Isolated Avian Osteoclast Resorption In Vitro. J Bone Miner Res, 17: 1859–1871. doi: 10.1359/jbmr.2002.17.10.1859
- Issue published online: 2 DEC 2009
- Article first published online: 1 OCT 2002
- Manuscript Accepted: 23 MAY 2002
- Manuscript Revised: 27 MAR 2002
- Manuscript Received: 7 SEP 2001
- basic FGF;
- bone resorption;
- chorioallantoic membrane
Increased local osteoclast (OC)-mediated bone resorption coincides with angiogenesis in normal bone development and fracture repair, as well as in pathological disorders such as tumor-associated osteolysis and inflammatory-related rheumatoid arthritis or periodontal disease. Angiogenic stimulation causes recruitment, activation, adhesion, transmigration, and differentiation of hematopoietic cells which may therefore enable greater numbers of pre-OC to emigrate from the circulation and develop into bone-resorptive OCs. A chick chorioallantoic membrane (CAM) model, involving coimplantation of a stimulus in an agarose plug directly adjacent to a bone chip was used to investigate if a potent angiogenic stimulator, basic fibroblast growth factor (bFGF), could promote OC recruitment, differentiation, and resorption in vivo. Angiogenesis elicited by bFGF on the CAM was accompanied by increased OC formation and bone pit resorption (both overall and on a per OC basis) on the bone implants in vivo. In complementary in vitro assays, bFGF did not directly stimulate avian OC development from bone marrow mononuclear cell precursors, consistent with their low mRNA expression of the four avian signaling FGF receptors (FGFR)-1, FGFR-2, FGFR-3, and FGFR-like embryonic kinase (FREK). In contrast, bFGF activated isolated avian OC bone pit resorption via mechanisms inhibited by a selective cyclo-oxygenase (COX)-2 prostaglandin inhibitor (NS-398) or p42/p44 MAPK activation inhibitor (PD98059), consistent with a relatively high expression of FGFR-1 by differentiated avian OCs. Thus, bFGF may sensitively regulate local bone resorption and remodeling through direct and indirect mechanisms that promote angiogenesis and OC recruitment, formation, differentiation, and activated bone pit resorption. The potential for bFGF to coinduce angiogenesis and OC bone remodeling may find clinical applications in reconstructive surgery, fracture repair, or the treatment of avascular necrosis. Alternatively, inhibiting such bFGF-dependent processes may aid in the treatment of inflammatory-related or metastatic bone loss.
Basic FGF (bFGF) or FGF-2 is a prototypical member of a family of structurally related growth factors that exert important pleiotropic effects on cell differentiation and organ development in vitro and in vivo.(1–3) In addition to playing a key role in normal development, morphogenesis, and tissue regeneration, FGFs such as bFGF significantly contribute to the pathogenesis of various diseases that involve tumor growth, abnormal neovascularization, or inflammation.(4–6) Recent clinical, genetic, and basic research studies have revealed that FGFs, including bFGF, also function as important regulators of bone development, remodeling, and repair.(7–10) FGFs signal through receptor tyrosine kinases that exhibit varying affinities and specificities for FGF members and are differentially expressed in specific spatial and temporal patterns during embryonic development and intramembranous or endochondral ossification.(11,12) Autosomal dominant mutations in any of three of the four genes encoding FGF signaling receptors lead to crucial alterations in chondrocyte or osteoblast (OB) proliferation, differentiation, or function that result in a characteristic group of hereditary skeletal and cranial disorders of varying severity in either humans or mice.(13,14) Thus, FGFs serve as key spatiotemporal signals governing bone morphogenesis. After embryonic skeletal development, bFGF is among the FGFs that continue to be expressed and serve postnatal modulatory influences within bone tissue. bFGF is produced by various cells including OBs, deposited as a stabilized complex with heparin sulfate proteoglycans into the bone matrix, and can be released and activated from this tissue reservoir during osteoclast (OC) bone resorption or in response to specific signals.(15–18) In vitro, bFGF acts via autocrine and paracrine mechanisms to variably promote pre-OB proliferation, stimulate or inhibit OB differentiation, and regulate bone nodule formation and calcification in isolated cell or organ cultures in a complex manner.(19–22) Recently, bFGF was also shown to indirectly regulate osteoclastogenesis and bone resorption in bone marrow or organ cultures by OB-mediated mechanisms and to directly stimulate mature OC bone pit resorption, via signal transduction pathways involving RANKL, cyclo-oxygenase (COX)-2-mediated prostaglandin production and p42/p44 MAPK activation.(23–27)
In vivo, bFGF administration via local injection or systemic infusion into animals exerts anabolic actions on bone tissue that culminate in increased endosteal bone formation and improved fracture healing.(28–30) These anabolic effects have been attributed primarily to bFGF stimulation of immature pre-OB cell proliferation at hematopoietic sites, thereby providing an increased precursor pool from which bone-forming OBs can be recruited.(28–32) Infusion of bFGF also restores cancellous bone mass in osteopenic ovariectomized rats via stimulating bone formation on preexisting bone surfaces as well as by causing the de novo formation of new bone spicules within the bone marrow.(31) During bone fracture repair, endogenous bFGF concentrations locally rise and significantly contribute to the regeneration process through stimulating both initial callus formation and its subsequent resorption.(29,33,34) Overall, the ability of bFGF to increase specifically cancellous bone mass and speed fracture healing has made it an attractive target for possible therapeutic use in alleviating conditions of osteopenia and/or inducing local bone regeneration. In particular, because few bone anabolic agents promote the formation of new trabecular structures, FGFs have garnered much interest for their potential ability to restore the diminished trabecular connectivity that increases bone fragility and fracture risk in severe osteoporosis.(28,31–36) However, the bone anabolic effects of bFGF in vivo may not be realized fully unless bFGF is subsequently withdrawn, and benefits may not be sustained unless concomitant antiresorptive measures are used.(28,29,31,32) Currently, how bFGF may influence in vivo bone resorption is not entirely clear. Thus, in vivo bFGF administration has led to either negligible bone resorption, resorption that accompanied bone formation because of an overall stimulation of coupled bone remodeling, or resorption that predominated over bone formation.(26,29,31,33,37) Furthermore, in a study involving rheumatoid arthritic patients, bFGF was the only inflammatory cytokine exhibiting elevated levels in the synovial fluid that directly correlated with the extent of OC formation, degree of joint destruction, and severity of the disease.(38)
Increased local bone resorption and remodeling coincide with angiogenesis in normal bone development and fracture healing, as well as in multiple pathological disorders such as inflammatory-related rheumatoid arthritis, periodontal disease, tumor-associated osteolysis, and osteoporosis.(6,17,39–43) bFGF plays an important role in many of these processes and represents one of the most potent angiogenic inducers known, functioning as an autocrine and paracrine factor to stimulate vascular endothelial cell proliferation, migration, and expression of specific proteases, growth factors, and integrins involved in angiogenesis.(44–46) Stimulated angiogenesis facilitates increased delivery of immune and hematopoietic precursor cells to the affected region, while it causes endothelial cells to display on their cell surfaces multiple key regulatory signals that trigger the activation, transmigration, differentiation and/or function of such cells.(6,41,47,48) Thus, angiogenic stimulation may enable greater numbers of OC precursor cells to emigrate from the peripheral circulation into the bone tissue and develop into resorptive OCs. Previously, we showed that vascular endothelial cells could promote the formation, differentiation, and bone pit-resorptive function of OCs in vitro.(49,50) Furthermore, using a modified chick chorioallantoic membrane (CAM) model, we showed that OC recruitment, formation, and bone pit resorption was increased in vivo in parallel with angiogenesis induced by the nitric oxide synthase inhibitor aminoguanidine.(51) Here, we investigated whether a classical angiogenesis stimulator, bFGF, could promote the in vivo recruitment, development, and resorptive function of OCs formed on bone implanted onto the chick CAM. Potential angiogenesis-dependent effects in vivo were distinguished from possible direct actions of bFGF to stimulate OC formation or bone-resorptive activity through the use of two well-characterized in vitro assays that independently examined avian bone marrow OC-like cell development or isolated avian OC bone pit resorption.
MATERIALS AND METHODS
CAM model of in vivo osteoclast development
A modified chick CAM model that enables angiogenic and osteoclastogenic responses to various modulators to be evaluated simultaneously was used here and has been described previously in detail.(51) Briefly, fertile White Leghorn chicken eggs (Kenroy Hatchery, Berger, MO, USA, or Spafas, Inc., Roanoke, IL, USA) were incubated on arrival (day 0) in a horizontal position in a forced draft incubator (37°C, 70% relative humidity), the eggs were windowed by gentle sanding on day 3 to expose a 3- to 4-cm2 opening on the CAM, the opening was sealed with UV-sterilized adhesive tape, and the eggs were incubated further until day 7 when a small chip of devitalized bovine cortical bone and a tiny agarose plug containing 0.54 ng (0.1 nM final concentration) human recombinant bFGF (R & D Systems, Minneapolis, MN, USA) were coimplanted directly next to one another on a lightly vascularized and slightly abraded site. Although relatively avascular regions are typically chosen for angiogenic (short-term) studies alone, such sites do not support sufficient OC recruitment, even in the presence of bone and angiogenic stimulators, to enable accurate assessments of either OC development or bone pit-resorptive activity in this uniquely modified CAM assay model system. Therefore, lightly vascularized CAM regions have proven to be the most suitable sites for evaluating both angiogenic responses and OC formation and activity.(51) The eggs were resealed and incubated until day 16, and the CAMs were scored and photographed with a Nikon 55-mm color camera to document angiogenenic responses. The relative capillary growth in the immediate vicinity of the bone/agarose implants was scored (poor, moderate, or high) and the number of eggs exhibiting substantially more angiogenesis at the implant site relative to control CAMs was determined for the surviving eggs (typically, >90%) from each trial. Bone chips were removed from the CAMs, rinsed, fixed in 1% paraformaldehyde in Hanks' balanced salt solution (HBSS), stained for TRAP activity as a marker of OCs, and analyzed for OC formation and bone pit resorption. Previous studies documented that the relative levels of angiogenesis, OC formation, and OC resorptive activity were unaffected by the presence of control agarose plugs when compared with CAMs implanted with a bone chip but lacking agarose plugs.(51)
OC formation and bone pit resorption in vivo
The number of bone-associated multinucleated TRAP stained OC was determined for a constant number of random fields per bone chip. Cells were subsequently removed and the bone pit-resorptive activity was quantified within the same fields using a computer-linked dark-field reflective light image analysis system.(50,51) The total area of bone resorbed, number of pits formed, and size of each excavation were assessed. Results were also normalized to OC number to compare the mean area of bone resorbed per OC (area/OC) and number of pits formed per OC (pits/OC). Data are presented as mean ± SEM of six independent CAM trials, having up to six replicate CAM eggs per condition. More than 2200 OCs formed on the CAM bone implants and their associated resorption pits were analyzed to yield the quantitative in vivo OC developmental and resorption pit activity data presented here.
Bone marrow-derived OC-like cell development in vitro
Mononuclear cells were isolated from the bone marrow of White Leghorn hatchlings maintained 28 days on a low-calcium diet and Ficoll-separated as described previously.(51,52) All animals were handled in accordance with the institutional Animal Care and Use Committee and standards approved by the National Institutes of Health (NIH) Guide for the Care and Use of Experimental Animals. Cells were plated on day 0 at 7 × 106 cells/well of a 6-well cluster dish in phenol red-free α-minimal essential medium (α-MEM) containing 10% fetal bovine serum (FBS) (GIBCO BRL, Gaithersburg, MD, USA) and 1% antibiotic/antimycotic (GIBCO BRL). Medium was changed every other day, mononuclear cell fusion commenced on days 1–2 and peaked on days 2–4, and large multinucleated cells (MNCs) had formed and typically covered the dishes by days 6–8. The cells and culture-conditioned medium were harvested in parallel for analysis. Developmental modulators promote an OC-like phenotype in this bone marrow mononuclear cell system when given between days 0 and 2 before cell fusion; induction of antigens and other OC characteristics are optimally measured in the MNCs that have formed by days 6–8 of culture; and OC differentiative properties persist after withdrawal of the developmental modulator.(52) bFGF was freshly diluted into culture medium for each feeding and was administered to the cultures on days 1, 2, and 4 at concentrations ranging from 0.001 to 10 nM. Thereafter, bFGF was discontinued on postdevelopmental day 6 refeeding of cultures, and the cells and medium were harvested on day 7 for analysis. A similar regimen was followed for administration of the OC developmental modulators 1,25-dihydroxyvitamin D3 (VD3; a generous gift from Hoffman-La Roche, Nutley, NJ, USA), human recombinant macrophage colony-stimulating factor (M-CSF; R & D Systems), or aminoguanidine (AG; Sigma Chemicals, St. Louis, MO, USA).
ELISA and TRAP analysis
Harvested cell pellets were fixed onto 96-well dishes and analyzed via an established ELISA protocol to determine the relative protein expression levels of two characteristic OC developmental markers reactive with monoclonal antibodies (MAb) 121F or 75B as described previously.(51) Parallel cell lysates were used to measure TRAP enzymatic activity levels by microplate assay.(51,52) ELISA and TRAP assay results were each normalized for cell protein using bovine serum albumin (BSA) as a standard in either the Lowry(53) or the bicinchoninic acid (BCA) protein assay (Pierce, Rockford, IL, USA), respectively. ELISA and TRAP measurements, which were assayed in duplicate from at least triplicate independent culture wells per trial over at least three independent trials, were corrected for nonspecific background values and expressed as the mean ± SEM.
RNA isolation from primary isolated or cultured cells
Because only limited amounts of RNA are obtained from primary isolated chick OCs (from 30 chicks), TRAP+ multinucleated bone pit-resorptive chick OCs were generated in vitro from avian bone marrow precursor cells to provide sufficient RNA for evaluating chick OC FGF receptor (FGFR) expression. Based on methods optimized in our laboratory, freshly isolated avian bone marrow mononuclear cells were placed into culture (40 × 106 cells/100-mm dish) on day 0 in α-MEM containing 10% FBS and 1% antibiotic/antimycotic and treated with 25 ng/ml of human recombinant M-CSF (R & D Sytems) and 100 ng/ml of murine recombinant RANKL on day 1 to initiate OC development. RANKL is routinely prepared from bacterial cultures in our laboratory using a murine RANKL clone kindly provided by Dr. Beth Lee (Renal Division, Washington University, St. Louis, MO, USA), and each preparation is evaluated for yield, concentration, purity (SDS-PAGE and Western blots), negligible endotoxin levels, and osteoclastogenic bioactivity toward human monocytes, murine bone marrow cells, RAW 264.7 cells, and chicken marrow cells (producing in all cases TRAP+ multinucleated bone pit-resorptive cells with hallmark OC characteristics). M-CSF and RANKL were resupplemented during refeeding of the chick marrow cultures on days 2, 4, and 6. RNA was harvested from the in vitro formed OCs on day 7 or 8 using RNA STAT-60 (Tel-Test, Inc., Friendswood, TX, USA). For comparative purposes, freshly isolated chick bone marrow mononuclear cells (day 0) were also extracted with RNA STAT-60 to provide RNA from an untreated initial population containing OC precursor cells. RNA from chick calvarial cell populations highly enriched for OBs was obtained by sequentially scraping, using a sterile pipette tip, the well-rinsed (α-MEM) periosteal and endosteal stripped calvariae that were freshly dissected from 10 chicks (the same ones used for bone marrow and isolated OC preparations mentioned previously) into 2 ml of RNA STAT-60.
Reverse transcription-polymerase chain reaction analysis
Semiquantitative reverse-transcription polymerase chain reaction (RT-PCR) amplifications for each of the four avian FGFRs (FGFR-1, FGFR-2, FGFR-3, and FREK) and GAPDH were performed in parallel using RNA isolated from chick OCs, OBs, or bone marrow mononuclear cells (as mentioned previously), together with forward and reverse primers specific for these receptors as described by Mitchell et al.(54) and Ready-to-Go RT-PCR beads (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Conditions were chosen based on initial trials performed using these RNA samples to determine optimal cycle numbers and RNA amounts for linear RT-PCR amplification of each FGFR and GAPDH. Thus, PCR reactions were heated initially to 95°C for 1 minute and then run at 95°C for 45 s, 47°C for 30 s, and 72°C for 1 minute for 34 cycles (FGFRs) or 30 cycles (GAPDH). Products were separated by agarose gel electrophoresis, visualized by UV illumination after ethidium bromide staining, photographed using a Polaroid camera, and quantified in a scanner (Scanjet II; Hewlett-Packard, Palo Alto, CA, USA) computer-linked to a Quantimet image analysis system (Leica Microsystems, Cambridge, UK).(50) Control reactions exhibited no product formation in the absence of active reverse transcriptase. Amplicons of 463 bp for FGFR-1, 469 bp for FGFR-2, 384 bp for FGFR-3, 251 bp for FREK, and 444 bp for GAPDH were of the expected sizes and directly sequenced using an ABI Prism Cycle Sequencing kit (Perkin Elmer Life Sciences, Boston, MA, USA), and DNA sequences were compared with published sequences to confirm their identity using computation performed at the National Center for Biotechnology Information (NCBI) and the basic local alignment search tool (BLAST) network service.
Isolated avian OC bone pit resorption
OCs were isolated according to previously published methods from the tibias and humeri of the same White Leghorn chick hatchlings (maintained 28 days on a low-calcium diet) that were used to obtain the bone marrow cells.(51,52) Briefly, OCs were released by collagenase/trypsin digestion and shaking, enriched by filtration and Percoll fractionation, and resuspended in OC culture medium which consists of phenol red-free medium 199 Earle's salts supplemented with 8.3 mM NaHCO3, 100 mM HEPES (pH 6.8), 5% twice charcoal-stripped FBS, and 2.5% antibiotic/antimycotic. OCs were seeded onto devitalized bovine cortical bone or ivory discs (2.2 × 106 OC in 1 ml of OC medium, 25 discs covering a 35-mm dish) and allowed to attach for 2.5 h; each bone was then moved using sterilized tweezers to an individual well of a 48-well dish containing OC medium, and human recombinant bFGF (1 to 100 nM) was added immediately (250 μl final volume in wells). For resorption inhibitor studies, 1 μM of the COX-2 selective inhibitor NS-398 (1 mM stock solution in DMSO; Calbiochem, San Diego, CA, USA) or 10 μM of the p42/p44 MAPK (ERK 1/ERK 2) inhibitor PD98059 (10 mM stock solution in DMSO; Calbiochem) was added to appropriate wells just before the addition of bFGF to the wells. Resorption cultures were incubated for an additional 42–45 h, after which the cells on bone were rinsed, fixed, and stained for TRAP activity.(51,52) The numbers of OCs and resorption pits, size of pits, and total areas resorbed were quantified and normalized to OC numbers as was performed for the bones harvested from the CAM studies. Data were compiled from three or five independent OC preparations (using 30 chicks per preparation), having four to seven replicate wells per condition per trial, for the bFGF concentration or inhibitor resorption studies, respectively. Each individual trial exhibited a dose-dependent stimulation of bone pit-resorptive activity in response to bFGF. More than 100 ivory slices with 71,000 OCs and their associated resorption pits were analyzed to yield the quantitative resorption pit data presented here.
Data are presented as the mean ± SEM of three to six independent trials having at least three replicates per condition and assayed at least in duplicate. Differences between treatments were analyzed using single-factor ANOVA. For simultaneous comparisons between multiple treatments, significant differences were determined using the post-ANOVA Bonferroni test. Differences were considered significant for p < 0.05.
bFGF promotes localized angiogenesis on the CAM
Focal application of bFGF onto the chick CAM led to marked increases observed 9 days later in the degree of angiogenesis proximal to the implantation site (Figs. 1C and 1D) when compared with control CAMs not receiving bFGF (Figs. 1A and 1B). Thus, the proportion of CAMs exhibiting substantial angiogenic sprouting from vessels in the immediate vicinity of the implantation site significantly increased by nearly 50% in response to bFGF relative to control CAMs (Fig. 1E). The CAM region encircling the site of bFGF application actually overgrew in some instances to entirely envelop the implanted bone slice; in all cases, the outgrowing CAM tissue was also highly vascularized (data not shown). Such overgrowth of CAM tissue to enclose the bone chip was never observed in over 50 control CAMs implanted with bone that did not receive bFGF treatment.
bFGF stimulates the recruitment, formation, and development of resorptive OCs on bone slices implanted onto the CAM
In both the control and the bFGF-treated CAMs, precursor cells capable of developing into TRAP+ multinucleated bone-resorptive OCs were recruited to the devitalized bone chips. OCs that formed on bone in the bFGF CAMs were as intensely stained for TRAP enzymatic activity as those on control CAMs, and both OC populations excavated typical well-formed lacunar resorption pits (data not shown). However, bFGF markedly increased the number of OCs that formed in vivo on the CAM bone implants in association with its localized stimulation of angiogenesis at those sites (Fig. 2A). Thus, the number of hematopoietic precursors recruited to the implant site that fused and differentiated into TRAP+ multinucleated OCs was significantly greater by 75% in response to bFGF when compared with the number of OCs that formed on the implanted bone chips of control CAMs (Fig. 2A).
bFGF promotes the bone pit-resorptive function of OCs formed on CAM bone implants in vivo
In addition to stimulating angiogenesis and OC formation in vivo on the CAM, bFGF also promoted OC function in vivo. Significant and dose-dependent increases in OC-mediated bone pit-resorptive activity were observed both overall and on an individual cell basis (Figs. 2B–2F). Thus, increased numbers of OCs formed on the bone implants of bFGF-treated CAMs were associated with significant 2.5-fold and 3-fold increases in the mean number of pits formed (Fig. 2B) and overall area of bone resorbed (Fig. 2C), respectively. Although such stimulation could simply reflect the bFGF-induced formation of greater numbers of OCs in vivo, normalization of the resorption data for OC numbers revealed that bFGF also promoted the resorptive activity of individual OCs formed in vivo. Thus, bFGF activated OCs for bone pit resorption by significantly increasing the mean area of bone resorbed per OC by 66% (Fig. 2D) and the mean number of pits formed per OC by 30% (Fig. 2E), relative to OCs formed on control CAM bone implants. Because the resorptive activity of an OC (total area of bone resorbed per OC) is a composite function of both the number and the size of the pits that it forms, bFGF must have activated OCs for bone pit resorption in vivo primarily by causing OCs to initiate more resorption sites, rather than excavate much larger lacunae, because bFGF significantly increased the number of pits formed per OC (Fig. 2E) and only slightly increased the mean size of pits excavated (Fig. 2F). The dual actions of bFGF to increase both the number of OCs formed on the CAM bone implants (1.8-fold, Fig. 2A) and the resorptive activity of individual OCs (1.6-fold, Fig. 2D) led to a net 3-fold stimulation in the overall extent of bone resorption that occurred in vivo over that in the absence of bFGF (Fig. 2C). Therefore, bFGF stimulated bone degradation in vivo on the CAM by promoting increased recruitment and formation of bone pit-resorptive OCs and by activating individual OCs to initiate more sites of resorption.
FGFR expression by avian OC and bone marrow precursor cell populations
bFGF-induced increases in OC formation and resorption in vivo on CAM bone implants could result from direct or indirect mechanisms targeting OCs and their precursor cells. Therefore, FGFR expression was examined by RT-PCR in bone pit-resorptive avian OCs formed in vitro from M-CSF/RANKL differentiated bone marrow mononuclear cells, as well as in the undifferentiated bone marrow mononuclear cell precursor population (Fig. 3). Because OBs of other species express FGFR-1 through FGFR-4, we analyzed chick calvarial OB populations in parallel to OCs for their relative expression of avian FGFR-1, FGFR-2, FGFR-3, and FREK. All four avian FGFRs were detected to varying degrees in the chick OC and OB populations, as well as in undifferentiated bone marrow mononuclear cell populations (Fig. 3A). In independent trials, avian OCs and OBs consistently exhibited relatively higher expression of the signaling FGFR-1 compared with the other three FGFRs, and FGFR-1 levels typically were comparable to or slightly greater in OBs than OCs and lower in marrow cells (Figs. 3A and 3B). FGFR-2 and FGFR-3 expression was more readily detected in OBs than either OCs or marrow cells, whereas FREK expression was relatively weak in all three cell populations (Fig. 3A). Thus, chick OCs and their bone marrow precursor cell populations express FGFRs known to mediate bFGF actions in vivo and in vitro.
bFGF does not directly promote OC development in avian bone marrow mononuclear cell cultures in vitro
Increased OC formation in the CAM in vivo model could be caused by the stimulated recruitment of OC precursor cells to the site of bFGF-induced angiogenesis. However, because OC bone marrow precursor populations express FGFRs, it was possible that bFGF might also directly promote osteoclastogenesis. This was investigated using a well-characterized in vitro chick bone marrow mononuclear cell developmental culture system in which precursor cells fuse to form MNCs by days 6–8 that exhibit induced levels of multiple phenotypic and functional OC characteristics if differentiation promoting agents have been administered (e.g., on day 1) to the marrow cultures before cell fusion (days 2–4). Although TRAP+ MNCs arose by day 4 in both control and bFGF (0.001–10 nM)-treated cultures and became more numerous in all cultures by day 8, no differences were noted in the size, shape, morphology, membrane ruffling, association of small cells with MNCs, or intensity of TRAP staining for bFGF-treated versus control cultures (data not shown). Furthermore, the OC-like nature of MNCs that formed was not increased by the presence of bFGF (0.001–10 nM) during their development based on quantifying TRAP biochemical activity (not shown) or the relative protein expression levels of two characteristic OC markers recognized by MAb's 121F or 75B by ELISA (Table 1). This contrasts with the ability of various other osteoclastogenesis-promoting agents (e.g., VD3, M-CSF or AG) to significantly increase these parameters in this model system (Table 1). Thus, bFGF does not appear to directly promote OC-like cell differentiation in avian bone marrow mononuclear cell cultures that contain OC precursors and express FGFRs.
bFGF stimulates isolated avian OC bone pit resorption in vitro via mechanisms involving p42/p44 MAPK activation and prostaglandin production
Because OCs that were formed in vivo on bFGF-treated CAMs exhibited greater activation for resorption than those on control CAMs and FGFRs were expressed by OC populations formed in vitro, it was possible that bFGF targeted OCs and/or other cells to stimulate OC bone resorption independently from the proangiogenic-related actions of bFGF. To investigate this, avian OCs formed in vivo were isolated and cultured on bone or ivory slices for 2 days in the presence or absence of varying concentrations of bFGF. Significant, biphasic dose-dependent increases were elicited by bFGF in the mean area of bone resorbed per OC (Fig. 4A), number of pits formed per OC (Fig. 4B), and (to a lesser but statistically significant degree) size of excavated pits (Fig. 4C), in comparison with OCs cultured in the absence of bFGF. Maximal stimulation of OC in vitro bone pit resorption occurred with 10 nM bFGF, and weaker responses were evoked by either 1 nM or 100 nM bFGF (Figs. 4A–4C). The number of OCs remaining on the bone slices at the end of the resorption culture period was not significantly different between the control and bFGF-treated cultures (Fig. 4D), suggesting that bFGF did not influence OC survival (apoptosis) or adhesion to bone (detachment). Thus, bFGF dose-dependently stimulated bone pit resorption by isolated avian OCs in culture primarily via activating OCs to initiate more, slightly larger, resorption lacunae.
Signal transduction mechanisms by which bFGF might have stimulated bone pit resorption in isolated avian OC cultures were investigated in a series of large trials using specific inhibitors for COX-2-induced prostaglandin production or activation of p42/p44 MAPK, two key intracellular pathways of bFGF signaling (Figs. 5 and 6). As shown in Fig. 5, avian OC bone pit resorption stimulated by 10 nM bFGF was reduced substantially (to levels not statistically different from control) by the presence of 1 μM NS-398, a selective inhibitor of COX-2-mediated prostaglandin production. Increases in the mean area of bone resorbed per OC (Fig. 5A), number of pits formed per OC (Fig. 5B), and size of pit excavations (Fig. 5C) due to bFGF were inhibited by 73, 50, and 100%, respectively, in cultures that were cotreated with NS-398. In contrast, this compound did not influence basal OC resorption (Figs. 5A–5C), and it did not alter the number of OCs remaining attached to bone in either the presence or absence of bFGF (Fig. 5D). OC bone pit resorption stimulated by 10 nM bFGF was also inhibited by the presence of 10 μM PD98059, a specific p42/p44 MAPK inhibitor (Figs. 6A–6C). In contrast to the COX-2 inhibitor, resorption was fully inhibited to levels significantly below basal in both unstimulated and bFGF-treated OC cultures (Figs. 6A–6C). However, like NS-398, PD98059 did not alter the number of OCs remaining attached to bone either in the presence or absence of bFGF (Fig. 6D). Therefore, bFGF caused a dose-dependent increase in bone pit resorption by isolated avian OCs and such stimulation was dependent on bFGF-induced COX-2-mediated prostaglandin production as well as p42/p44 MAPK activation signal transduction pathways.
Here, we have shown for the first time that the potent proangiogenic growth factor bFGF promotes the localized recruitment, formation, differentiation, and activity of bone-resorptive OCs at sites of stimulated angiogenesis in vivo on the chick CAM. Not only were more OCs formed at such sites, but they exhibited an increased activation for bone pit resorption. Using in vitro assays to measure OC development or function, bFGF did not directly promote OC differentiation whereas it stimulated isolated OC bone pit resorption via mechanisms involving COX-2-mediated prostaglandin production and p42/p44 MAPK activation. Thus, bFGF appears to regulate multiple aspects of OC recruitment, development, and function in diverse ways. Our results also lend further support to the growing number of studies documenting a close relationship between the bone vasculature and OC formation/resorption, processes that are crucial for normal bone development, remodeling, and repair but can be linked to pathological consequences when inappropriately stimulated.(40,51,55,56)
Angiogenesis is a highly orchestrated process by which new capillaries arise from existing blood vessels during embryonic development, reproduction, and wound healing.(6,41,44) Increased local bone resorption and remodeling coincide with angiogenesis in normal bone development and repair.(7,8,17,39–43) However, increased blood vessel formation is also a critical feature of the pathology associated with tumor progression, osteoporosis, and skeletal or inflammatory disorders (e.g., rheumatoid arthritis or periodontal disease) that often are associated with bone or tissue destruction.(6–8,17,18,39–43,55–58) One of the most potent angiogenic substances known is bFGF, and the mechanisms by which it is released from activated cells or the extracellular matrix and induces angiogenic sprouting have been well studied in various in vitro and in vivo systems including the chick CAM.(1–6,18,59–63) Thus, bFGF (up to 500 ng) incorporated into either acetate discs or gelatin sponges has promoted neoangiogenesis in vivo on the developing chick CAM.(59–62,64) Here, low physiological concentrations of bFGF (0.54 ng or 0.1 nM) locally applied in agarose plugs to the chick CAM similarly stimulated focal microcapillary formation in a potent manner.
Stimulated angiogenesis leads to an increased delivery of nutrients, transport of immune and hematopoietic cells into the affected region, and exposure of such cells to regulatory signals displayed on the vessel surface that trigger the activation, transmigration, differentiation, and function of these cells.(6,41,47–51,55) Similarly, OC hematopoietic precursors may be delivered to sites of neoangiogenesis and subsequently develop into bone-resorptive OC under the influence of local cues in the microenvironment. As part of a series of studies using our modified chick CAM assay to examine the relationship between angiogenesis, OC formation, and bone resorption, we have shown that four diverse proangiogenic signals and three antiangiogenic signals caused a stimulation or inhibition, respectively, in the formation and bone pit resorption of OC on CAM bone implants(51) (Collin-Osdoby P and coworkers, unpublished results, 1997–2000). Here, bFGF significantly increased the recruitment and development of bone pit-resorptive OCs formed on CAM bone implants in vivo, in parallel with promoting microcapillary formation. These in vivo bFGF-induced responses could involve (1) increased pre-OC recruitment via enhanced angiogenesis; (2) direct effects of bFGF on pre-OCs or OCs to promote their formation, development, and/or resorptive activity; or (3) bFGF effects on other cells that indirectly stimulate OC formation or function. Therefore, complementary in vitro studies were performed to help clarify the cellular mechanisms by which bFGF increased OC formation and bone pit resorption in vivo on the chick CAM. In isolated avian bone marrow mononuclear cell cultures containing OC precursors (but relatively few OB or stromal cells), bFGF did not elicit OC-like cell differentiation. These data suggest that bFGF indirectly stimulated OC formation in vivo on the chick CAM. Consistent with this, marrow mononuclear cells expressed relatively low levels of the four avian FGFRs FGFR-1, FGFR-2, FGFR-3, and FREK. In contrast, isolated chick calvarial OBs expressed relatively high levels of FGFR-1, FGFR-2, and FGFR-3 (and low FREK). Similar to our avian findings, bFGF only indirectly regulated osteoclastogenesis in murine whole bone marrow or organ cultures by OB-mediated mechanisms, and murine OB expressed all four high-affinity signaling FGFRs (FGFR-1 to FGFR-4).(23–27)
Bone pit resorption in vivo also was enhanced markedly in bFGF-treated CAMs, both because of the formation of greater numbers of OCs as well as the activation of a larger proportion of OCs for bone pit resorption. Thus, bFGF significantly increased the mean area of bone resorbed per OC formed on CAM bone implants, primarily by increasing the number of pits initiated per OC (OC activation). Together, the combined effects in vivo of bFGF to stimulate OC recruitment, development, and activity culminated in a remarkable overall 3-fold rise in the extent of bone resorption occurring in bFGF-treated CAMs compared with control CAMs lacking angiogenic stimulation. In vitro, bFGF also activated in a biphasic fashion the bone pit-resorptive activity of isolated avian OCs formed in vivo, primarily by increasing the number of pits excavated per OC. Two key signal transduction pathways by which bFGF exerts its biological actions in other systems involve COX-2 induction of prostaglandin production and p42/p44 MAPK (ERK1/ERK2) activation.(5,23–27,63,65) In our isolated avian OC cultures, resorption stimulated by bFGF was reduced substantially by a selective inhibitor of COX-2-mediated prostaglandin production (NS-398) and fully inhibited (to below basal levels of resorption) by an inhibitor of p42/p44 MAPK (ERK1/ERK2) activation (PD98059). Therefore, both COX-2-mediated prostaglandin production and p42/p44 MAPK activation are required for bFGF to stimulate bone pit resorption in isolated avian OC cultures. Moreover, because avian OCs formed in vitro expressed relatively high levels of FGFR-1 (and lower FGFR-2, FGFR-3, or FREK), this signaling receptor may mediate the regulatory effects of bFGF on avian OCs bone resorption. Consistent with our findings, bFGF has been reported to enhance bone pit resorption by isolated rabbit or mouse OCs in vitro directly through activating FGFR-1 and p42/p44 MAPK and indirectly via inducing a COX-2-mediated prostaglandin pathway.(25,26) On a broader level, bFGF effects to promote local angiogenesis, pre-OB proliferation/differentiation, and OC formation and bone resorption may be amplified further via autocrine/paracrine circuits because of the mobilization and activation of endogenous bFGF stored in the underlying matrix by OCs during their resorption of bone.
Although brief exposure to bFGF can exert anabolic effects on bone, chronically elevated levels of bFGF in various pathological disorders may amplify and exacerbate such diseases through stimulating blood vessel endothelial cells and angiogenesis, promoting excessive recruitment and activation of inflammatory cells, restraining chondrocyte and OB maturation and function, and enhancing OC formation and bone resorption, thereby contributing to a localized osteopenia.(17,39,43,56) Elevated bFGF levels have been detected in the serum and urine of patients with various cancers, as well as in the rheumatoid synovial joint fluid of animals or human patients, and appear intimately linked with the progression and tissue destruction associated with these diseases.(38,56,66) Sustained high bFGF levels in inflammatory disorders such as rheumatoid arthritis or periodontal disease may costimulate angiogenesis and OC-mediated bone resorption.(6,17,39,42–47) For example, overexpression of bFGF in a rat model of rheumatoid arthritis worsened disease symptoms, increased bone destruction (and OC numbers), and enhanced vascularity, whereas a neutralizing antibody against bFGF attenuated all of these arthritic responses (and also did so in the absence of induced bFGF overexpression).(56) Moreover, in a human study involving patients with rheumatoid arthritis, bFGF was the only inflammatory cytokine exhibiting elevated levels in the synovial fluid that directly correlated with the extent of OC formation, degree of joint destruction, and severity of the disease.(38) Thus, bFGF appears to have a prominent role in the pathogenesis of this and other diseases. Because angiogenesis and inflammation are codependent processes that act to amplify one another, they may function together to elicit local osteoclastogenesis and bone loss; consequently, angiogenesis inhibitors may provide a valuable approach for efficiently blocking such an alliance.(6,17,39,42–47,56,67)
In conclusion, we have shown that bFGF regulates multiple aspects of in vivo OC-mediated bone resorption via promoting the localized recruitment, formation, differentiation, and function of bone-resorptive OCs at sites of stimulated angiogenesis on the chick CAM. bFGF did not directly influence in vitro OC development from bone marrow mononuclear cell precursors. In contrast, bFGF activated bone pit resorption by isolated avian OCs expressing FGFR-1 by COX-2 and p42/p44 MAPK-dependent signal mechanisms. The potential to regulate in vivo local bone metabolism through modulating bFGF levels and angiogenesis may have beneficial clinical applications such as in the treatment of rheumatoid arthritis, avascular necrosis of bone, and fracture repair, or as an alternative to bone vascular grafting in skeletal reconstructive surgery.(56,68,69)
This work was supported by NIH grants DK46547 (to P.C-O.) and AR32927 (to P.O.).
- 11989 The fibroblast growth factor family: Structural and biological properties. Prog Growth Factor Res 1:207–235.
- 21994 The ins and outs of fibroblast growth factors. Cell 78:547–552.
- 31997 Biological roles of fibroblast growth factor-2. Endocr Rev 18:26–45., , ,
- 41992 The FGF family of growth factors and oncogenes. Adv Cancer Res 59:115–165.,
- 51997 Fibroblast growth factors as angiogenesis factors: New insights into their mechanism of action. Experientia Suppl 79:159–192., ,
- 61997 The codependence of angiogenesis and chronic inflammation. FASEB J 11:457–465., , , ,
- 72000 New factors controlling bone remodeling. Joint Bone Spine 67:150–156., , ,
- 81999 Recent advances in bone biology provide insight into the pathogenesis of bone diseases. Lab Invest 79:83–94., , , ,
- 91996 Functions of fibroblast growth factors in vertebrate development. Cytokine Growth Factor Rev 7:311–325.
- 101999 Fgfr2 is required for limb outgrowth and lung-branching morphogenesis. Proc Natl Acad Sci USA 96:11895–11899., , ,
- 111998 Spatio-temporal expression of FGFR 1, 2, and 3 genes during human embryo-fetal ossification. Mech Dev 77:19–30., , , , , ,
- 121999 Differential expression of fibroblast growth factor receptor-1, -2, and -3 and syndecan-1, -2, and -4 in neonatal rat mandibular condyle and calvaria during osteogenic differentiation in vitro. Bone 24:337–347., , ,
- 131997 FGFR activation in skeletal disorders: Too much of a good thing. Trends Genet 13:178–182.,
- 142000 The molecular and genetic basis of fibroblast growth factor receptor 3 disorders: The achondroplasia family of skeletal dysplasias, Muenke craniosynostosis, and Crouzon syndrome with acanthosis nigricans. Endocr Rev 21:23–39., ,
- 151989 Cultured bovine bone cells synthesize fibroblast growth factor and store it in their extracellular matrix. Endocrinology 124:1539–1547., ,
- 161999 FGF-2/fibroblast growth factor receptor/heparin-like glycosaminoglycan interactions: A compensation model for FGF-2 signaling. FASEB J 13:1677–1687., , , ,
- 171999 Cellular and molecular regulation of bone turnover. Bone 24:35S–38S.
- 181997 A secreted FGF-binding protein can serve as the angiogenic switch in human cancer. Nat Med 3:1137–1140., , , , ,
- 191989 Opposing effects of fibroblast growth factor and pertussis toxin on alkaline phosphatase, osteopontin, osteocalcin, and type I collagen mRNA levels in ROS/17.2 cells. J Biol Chem 264:19934–19941., , ,
- 201998 The effects of fibroblast growth factor-2 on human neonatal calvaria osteoblastic cells are differentiation stage specific. J Bone Miner Res 13:645–654., , ,
- 211998 Basic fibroblast growth factor enhances the growth and expression of the osteogenic phenotype of dexamethasone-treated human bone marrow-derived bone-like cells in culture. Bone 23:111–117., , ,
- 221999 Basic fibroblast growth factor in the presence of dexamethasone stimulates colony formation, expansion, and osteoblastic differentiation by rat bone marrow stromal cells. Calcif Tissue Int 64:69–77.,
- 231999 Basic fibroblast growth factor induces osteoclast formation by reciprocally regulating the production of osteoclast differentiation factor and osteoclastogenesis inhibitory factor in mouse osteoblastic cells. Biochem Biophys Res Commun 265:158–163., , , , , , , , , ,
- 241996 Basic fibroblast growth factor inhibits osteoclast-like cell formation. J Cell Physiol 168:395–402., , , , ,
- 252000 Direct and indirect actions of fibroblast growth factor 2 on osteoclastic bone resorption in cultures. J Bone Miner Res 15:466–473., , , ,
- 262000 Fibroblast growth factor (FGF)-2 directly stimulates mature osteoclast function through activation of FGF receptor 1 and p42/p44 MAP kinase. J Biol Chem 275:31444–31450., , , , , , , ,
- 271998 Basic fibroblast growth factor induces osteoclast formation in murine bone marrow cultures. Bone 22:309–316., , , ,
- 282000 Anabolic effects of basic fibroblast growth factor in the tibial diaphysis of ovariectomized rats. Bone 27:197–202., ,
- 291998 Local application of basic fibroblast growth factor minipellet induces the healing of segmental bony defects in rabbits. Calcif Tissue Int 63:490–495., , , , , , , ,
- 301998 Stimulation of bone formation by intraosseous injection of basic fibroblast growth factor in ovariectomised rats. Int Orthop 22:49–54., , , , ,
- 311999 Bone anabolic effects of basic fibroblast growth factor in ovariectomized rats. Endocrinology 140:5780–5788., ,
- 322001 Decreased bone anabolic effect of basic fibroblast growth factor at fatty marrow sites in ovariectomized rats. Bone 28:220–226., , , ,
- 331998 Recombinant human basic fibroblast growth factor accelerates fracture healing by enhancing callus remodeling in experimental dog tibial fracture. J Bone Miner Res 13:942–949., , , , , ,
- 342000 Basic fibroblast growth factor may improve devascularized sternal healing. Ann Thorac Surg 70:824–828., , , , , ,
- 351999 Systemic administration of acidic fibroblast growth factor (FGF-1) prevents bone loss and increases new bone formation in ovariectomized rats. J Bone Miner Res 14:953–959., , , , , ,
- 361999 Anabolic effect of aminoterminally truncated fibroblast growth factor 4 (FGF4) on bone. Bone 25:431–437., , , , ,
- 371998 Morphological examination of bone synthesis via direct administration of basic fibroblast growth factor into rat bone marrow. Microsc Res Tech 41:313–322., , , , , , ,
- 381999 Involvement of fibroblast growth factor-2 in joint destruction of rheumatoid arthritis patients. Rheumatology 38:714–720., , , , ,
- 391994 Role of vascular endothelial cells in bone biology. J Cell Biochem 55:304–309.
- 401999 VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation. Nat Med 5:623–628., , , , ,
- 411998 Angiogenesis: Implications for rheumatoid arthritis. Arthrit Rheum 41:951–962.
- 421998 Mechanisms of alveolar bone destruction in periodontitis. Periodontol 2000 14:158–172., , ,
- 431996 Dynamics of bone remodeling: Biochemical and pathophysiological basis. Eur J Clin Chem Clin Biochem 34:729–734.
- 441997 Vascular development: Cellular and molecular regulation. FASEB J 11:365–373.,
- 451998 Fibroblast growth factor-2 (FGF-2) induces vascular endothelial growth factor (VEGF) expression in the endothelial cells of forming capillaries: An autocrine mechanism contributing to angiogenesis. J Cell Biol 141:1659–1673., , , , , , , , ,
- 461999 Interaction of fibroblast growth factor-2 (FGF-2) with free gangliosides: Biochemical characterization and biological consequences in endothelial cell cultures. Mol Biol Cell 10:313–327., , , , , ,
- 471993 The adhesion molecules of inflammation. Arthritis Rheum 36:147–157.,
- 481994 Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm. Cell 76:301–314.
- 491995 In vitro structural and functional relationships between preosteoclastic and bone endothelial cells: A juxtacrine model for migration and adhesion of osteoclast precursors. J Cell Physiol 162:199–212., , , , , , , , ,
- 502001 Receptor activator of NF-kB ligand and osteoprotegerin expression by human microvascular endothelial cells, regulation by inflammatory cytokines, and role in human osteoclastogenesis. J Biol Chem 276:20659–20672., , , , ,
- 512000 Decreased nitric oxide levels stimulate osteoclastogenesis and bone resorption both in vitro and in vivo on the chick chorioallantoic membrane in association with neoangiogenesis. J Bone Miner Res 15:474–488., , , ,
- 521995 Osteoclast 121F antigen expression during osteoblast conditioned medium induction of osteoclast-like cells in vitro: Relationship to calcitonin responsiveness, tartrate resistant acid phosphatase levels, and bone resorptive activity. J Bone Miner Res 10:45–58., , , , ,
- 531951 Protein measurement with the folin phenol reagent. J Biol Chem 193:265–275., , ,
- 541999 Expression of fibroblast growth factor family during postnatal skeletal muscle hypertrophy. J Appl Physiol 86:313–319., ,
- 551995 Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1:27–31.
- 562002 Fibroblast growth factor-2 determines severity of joint disease in adjuvant-induced arthritis in rats. J Immunol 168:450–457., , , , , , , , ,
- 571996 Advances in bone biology: The osteoclast. Endocr Rev 17:308–332.
- 581992 Cytokines and local factors which affect osteoclast function. Int J Cell Cloning 10:215–222.
- 591997 Basic fibroblast growth factor-induced angiogenesis in the chick embryo chorioallantoic membrane: An electron microscopy study. Microvasc Res 53:187–190., , , ,
- 602000 Fibroblast growth factor-2 selectively stimulates angiogenesis of small vessels in arterial trees. Arterioscler Thromb Vasc Biol 20:1250–1256., , ,
- 611995 Definition of two angiogenic pathways by distinct alpha v integrins. Science 270:1500–1502., , , , ,
- 621998 Integrin alphavbeta3 requirement for sustained mitogen-activated protein kinase activity during angiogenesis. J Cell Biol 140:1255–1263., , ,
- 632000 Differential ability of heparin sulfate proteoglycans to assemble the fibroblast growth factor receptor complex in situ. FASEB J 14:137–144., , ,
- 642000 Osteogenic protein-1, a bone morphogenetic protein, induces angiogenesis in the chick chorioallantoic membrane and synergizes with basic fibroblast growth factor and transforming growth factor-β1. Anat Rec 259:97–107.,
- 651999 Basic fibroblast growth factor induces cyclooxygenase-2 expression in endothelial cells derived from bone. Biochem Biophys Res Commun 254:259–263., , , , ,
- 661997 bFGF and tumor angiogenesis—back in the limelight? Nat Med 3:1083–1084.,
- 672001 Angiogenesis in the pathogenesis of inflammatory joint and lung diseases. Arthritis Res 3:147–153.,
- 681997 Avascular necrosis of bone following an intensified chemotherapy regimen including high dose steroids. Clin Radiol 52:607–612., , ,
- 691998 Angiogenesis in fracture repair. Clin Orthop 355(Suppl):S82–S89.