Chondrosarcoma, a common malignant tumour, develops in bone. Effective adjuvant therapy remains inadequate for treatment, meaning poor prognosis. It is imperative to explore novel remedies. Angiogenesis is a rate-limiting step in progression that explains neovessel formation for blood supply in the tumour microenvironment. Numerous studies indicate that EPCs (endothelial progenitor cells) promote angiogenesis and contribute to tumour growth. bFGF (basic fibroblast growth factor), a secreted cytokine, regulates biological activity, including angiogenesis, and correlates with tumorigenesis. However, the role of bFGF in angiogenesis-related tumour progression by recruiting EPCs in human chondrosarcoma is rarely discussed. In the present study, we found that bFGF induced VEGF (vascular endothelial growth factor) expression via the FGFR1 (fibroblast growth factor receptor 1)/c-Src/p38/NF-κB (nuclear factor κB) signalling pathway in chondrosarcoma cells, thereby triggering angiogenesis of endothelial progenitor cells. Our in vivo data revealed that tumour-secreted bFGF promotes angiogenesis in both mouse plug and chick CAM (chorioallantoic membrane) assays. Xenograft mouse model data, due to bFGF-regulated angiogenesis, showed the bFGF regulates angiogenesis-linked tumour growth. Finally, bFGF was highly expressed in chondrosarcoma patients compared with normal cartilage, positively correlating with VEGF expression and tumour stage. The present study reveals a novel therapeutic target for chondrosarcoma progression.

CLINICAL PERSPECTIVES

  • VEGF and bFGF were the most prominent tumour-related angiogenesis-secreted factors produced by cancer cells. VEGF expression correlated with chondrosarcoma angiogenesis. However, a clinical correlation between bFGF and chondrosarcoma is unknown.

  • Higher levels of bFGF expression is associated with higher clinical pathological stages: expression of bFGF correlates with VEGF expression in human chondrosarcoma specimens. We have shown that bFGF promotes EPC-primed angiogenesis by regulating VEGF expression in chondrosarcoma cells, with FGFR1/c-Src/p38/NF-κB signal pathway activation.

  • A concept of tumour neoangiogenesis has been verified. We cite evidence of bFGF-regulated VEGF expression and angiogenesis in chondrosarcoma via FGFR1. Recently, brivanib, a specific receptor tyrosine kinase inhibitor targeting the key angiogenesis receptors VEGFR2, FGFR1 and FGFR2, has been devised under clinical evaluation. Our findings with regard to bFGF and VEGF may provide a novel target for brivanib as a synergic inhibition drug.

INTRODUCTION

Chondrosarcoma, a common malignant tumour developing in bone, shows a refractory phenotype to chemo- and radio-therapy, posing a complicated challenge [1]. Surgical resection remains the most common mode of therapy for chondrosarcoma. Since adjuvant therapy remains inadequate and prognosis poor, we must explore novel remedies [2].

Angiogenesis, the state of cancer that promotes neovessel formation, has been proposed as a regulator in tumour progression. VEGF (vascular endothelial growth factor) plays a key role in the angiogenesis of tumour progression [3]. EPCs (endothelial progenitor cells) in peripheral blood were first discovered as CD34+/VEGFR2+ (VEGF receptor 2) mononuclear cells in 1997; they have the capacity to differentiate into an endothelial phenotype, express endothelial markers and form new vessels at sites of ischaemia [4]. A large number of reviews have discussed the capacity of EPCs to participate in endothelial repair and neoangiogenesis due to their ability to differentiate into endothelial cells [57]. Researchers cite EPCs in promoting angiogenesis-related growth [8], and targeting EPCs in search of anti-angiogenesis drugs.

bFGF (basic fibroblast growth factor)/FGF-2 (fibroblast growth factor 2), a secreted cytokine, encodes heparin-binding proteins with growth, anti-apoptotic, differentiation promotion and angiogenic activity. Its correlation with progression is certified in many cancers, e.g. expressed in OSCC (oral squamous cell carcinoma), correlated with lymph node metastasis and prognosis in OSCC [9]. Expression of bFGF is associated with tumour recurrence and reduced survival after surgical resection of oesophageal cancer [10]. This secreted cytokine regulates biological activities, including angiogenesis, and expression of bFGF correlates with tumorigenesis [911].

Chondrosarcoma shows poor prognosis caused by an absence of effective adjuvant therapy. Previous study indicates that angiogenesis plays crucial roles in biological processes and rate-limiting steps in tumour progression, with VEGF a paramount factor in angiogenesis while participating in tumorigenesis [12]. bFGF notably regulates MMP-13 (matrix metalloproteinase 13) expression and promotes human chondrosarcoma [11]. Moreover, bFGF has been proposed to accelerate the growth of six tumour cell lines, including SW1353 (chondrosarcoma) [13]. In addition, in vivo results indicate that intralesional administration of mouse monoclonal anti-β-FGF (DG-2) antibody significantly inhibited rat chondrosarcoma growth and vascularization [14]. Studies indicate that EPCs expedite both angiogenesis and angiogenesis-related growth [8]. Whether bFGF promotes angiogenesis by affecting EPCs in human chondrosarcoma is still unclear. We show evidence of bFGF accelerating EPC-primed angiogenesis by VEGF expression in chondrosarcoma cells: bFGF treatment induces VEGF expression in chondrosarcoma cells and promotes EPC tube formation in a dose-dependent manner. Pre-treatment with VEGF-neutralizing antibody dramatically reduced this effect. We also proved the involvement of the FGFR1 (FGF receptor 1)/c-Src/p38/NF-κB (nuclear factor κB) signalling pathway in VEGF expression and EPC-primed angiogenesis. Both chick CAM (chorioallantoic membrane) assay and mouse xenograft in vivo models also revealed the role of bFGF in angiogenesis-related tumour growth. Finally, correlation between bFGF, VEGF and tumour stage in clinical specimens was consistent with our in vitro and in vivo results, which suggests that bFGF promotes angiogenesis by VEGF expression within the chondrosarcoma microenvironment.

MATERIALS AND METHODS

Materials

Horseradish peroxidase-conjugated anti-mouse and anti-rabbit IgG, rabbit polyclonal antibodies specific for p-c-Src, c-Src, p-p38, p38, p-p65, p65, β-actin, CD31, CD34, CD133 and bFGF were purchased from Santa Cruz Biotechnology. Anti-VEGF antibody was from Abcam. Recombinant human VEGF was purchased from R&D Systems. DMEM (Dulbecco's modified Eagle's medium), α-MEM (α-minimum essential medium), FBs and all other cell culture reagents were from Gibco-BRL Life Technologies. ON-TARGETplus siRNAs were purchased from Dharmacon Research, pSV-β-galactosidase vector and luciferase assay kit were from Promega. All other chemicals were from Sigma–Aldrich.

Cell culture

The human chondrosarcoma cell line JJ012 was donated by the laboratory of Dr Sean P. Scully (University of Miami School of Medicine, Miami, FL, U.S.A.). Cells were cultured in complete medium containing DMEM/α-MEM with 10% (v/v) FBS supplement.

EPC culture

The protocol for EPC culture was approved by the Institutional Review Board of Mackay Medical College, New Taipei City, Taiwan (reference number P1000002). All subjects gave informed written consent before enrolling. Peripheral blood (80 ml) was taken from healthy donors, mononuclear cells were isolated by Ficoll-Paque PLUS centrifugation (GE Healthcare), according to the manufacturer's instructions. CD34-positive mononuclear cells were isolated from mononuclear cell fraction by CD34 MicroBead kit and MACS Cell Separation System (Miltenyi Biotec), and their EPCs were maintained and characterized as detailed previously [15]. EPCs were defined as UEA-1-positive, CD34-positive, KDR-positive and CD31-positive [16,17]. Human CD34-positive EPCs were cultured in MV2 complete medium containing MV2 basal medium and growth supplement (PromoCell), supplemented with 20% (v/v) defined FBS (HyClone). Cultures were seeded on 1% gelatin-coated plasticware and maintained at 37°C in a humidified 5% CO2 atmosphere.

Western blotting

Cellular lysates were prepared, proteins were resolved by SDS/PAGE and transferred on to a Immobilon PVDF membrane. Blots were blocked with 4% (w/v) BSA for 1 h at room temperature and then probed with antibodies (1:1000 dilution) for 1 h at room temperature. After three washes, blots were incubated with horseradish peroxidase-conjugated secondary antibody (1:1000 dilution) for 1 h at room temperature and visualized by ECL using X-OMAT LS film (Eastman Kodak). Data were quantified using a computing densitometer and ImageQuant software (Molecular Dynamics).

qPCR (quantitative real-time PCR)

qPCR analysis used Taqman® one-step PCR Master Mix (Applied Biosystems). A 100 mg volume of total cDNA was added per 25-μl reaction volume with sequence-specific primers and Taqman® probes. Sequences for all target gene primers and probes were purchased commercially (β-actin was used as an internal control) (Applied Biosystems). The qPCR assay was carried out in triplicate using the StepOnePlus sequence detection system. Cycling conditions were 10-min of polymerase activation at 95°C followed by 40 cycles at 95°C for 15 s and 60°C for 60 s. The threshold was set above non-template control background and within the linear phase of target gene amplification to calculate the cycle number at which transcript was detected (denoted as CT).

Reporter assay

Cells were transfected for 24 h with NF-κB reporter plasmid by Lipofectamine™ 2000 (Invitrogen), as per the manufacturer's recommendations, and extracts prepared. Activities of luciferase and β-galactosidase were then measured.

Transwell migration assay

The migration assay used Transwell inserts (8-μm pore size; Corning Costar) in 24-well dishes. Chondrosarcoma cells were pre-treated for 30 min with the designated inhibitor or vehicle concentration (0.1% DMSO) and CM (conditioned medium) was collected after 24 h. EPCs were seeded in the upper Transwell chamber and 300 μl of CM was placed in the lower chamber. After 24 h, cells on the upper side of the filters were removed with cotton-tipped swabs and the filters were washed with PBS. Cells on the underside of the filters were examined and counted under a microscope. Each experiment was performed in triplicate and repeated at least three times.

ELISA

All cells were pre-treated for 30 min with various concentrations of inhibitors [PP2, SB203580, PDTC (pyrrolidine dithiocarbamate) or TPCK (tosylphenylalanylchloromethane)] or vehicle (0.1% DMSO) and then incubated with bFGF (30 ng/ml) for 24 h at 37°C. Medium removed was stored at −80°C, VEGF was determined using ELISA kits (Biocompare), as per the manufacturer's instructions.

Tube formation

Matrigel™ (BD Biosciences) was dissolved at 4°C, aliquots of 150 μl/well were added to 48-well plates, which were incubated at 37°C for 30 min. EPCs were resuspended at 5×104 cells in 100 μl of culture medium (50% EGM-MV2 medium and 50% CM from JJ012 cells) and added to wells. VEGF (20 ng/ml) and culture medium were used as positive and negative controls respectively. After 6 h of incubation at 37°C, EPC tube formation was assessed by microscopy and each well was photographed. The number of tube branches and total tube length were calculated using MacBiophotonics ImageJ software (NIH).

Animal models and imaging

Experimental procedures were approved by the Institutional Animal Care and Use Committee. Male nu/nu mice (6–8 weeks of age) were subcutaneously injected with 5×105 cells suspended in 100 μl of medium. Tumour growth, local invasion and metastasis were monitored using the IVIS Imaging System.

IHC (immunohistochemistry)

Human chondrosarcoma tissue arrays were purchased from Cybrdi and Biomax. We explored eight cases of normal cartilage, as well as 26 of Grade I, 12 of Grade II and 18 of Grade III chondrosarcoma. Sections (5-μm thick) of paraffin-embedded tissue were placed on glass slides, rehydrated, incubated with 3% hydrogen peroxide to quench endogenous peroxidase activity, then blocked by incubation in 3% (w/v) BSA in PBS. Sections were incubated with the primary mouse polyclonal anti-human bFGF and anti-VEGF antibodies at 1:50 dilution and incubated at 4°C overnight. After three PBS washes, samples were incubated with a 1:50 dilution of biotin-labelled goat anti-mouse IgG secondary antibody, bound antibodies were detected using an ABC Kit (Vector Laboratories). Slides were stained with the chromogen diaminobenzidine, washed, counterstained with Delafield's haematoxylin, dehydrated, treated with xylene, then mounted.

For in vivo bFGF, CD31, CD34 and CD133 IHC assays in the xenograft model, plug and tumour samples collected from killed mice were fixed in 4% (w/v) paraformaldehyde in PBS for at least 72 h, dehydrated in increasing concentrations of ethanol, then embedded in paraffin. Serial sections of 5-μm thickness were cut longitudinally and incubated with anti-bFGF (1:50 dilution), anti-CD31 (1:50 dilution), anti-CD34 (1:100 dilution) or anti-CD133 (1:100 dilution) antibody at 4°C overnight. After three PBS washes, samples were incubated with 1:50 dilution of biotin-labelled goat anti-mouse IgG secondary antibody, with bound antibodies detected using the ABC Kit. The slides were stained with the chromogen diaminobenzidine, washed, counterstained with Delafield's haematoxylin, dehydrated, treated with xylene, then mounted. To quantify MVD (microvessel density), we evaluated the degree of microvessel coverage in three random fields per plug (×200 magnification). MVD was calculated using quantification of CD31-positive microvessels per field of view.

Statistical analysis

Results are presented as means±S.E.M. Statistical comparisons between two samples were performed using Student's t test. Statistical comparisons of more than two groups were performed using one-way ANOVA with Bonferroni's post-hoc test. A P-value of less than 0.05 was considered statistically significant.

RESULTS

bFGF regulates EPC-primed angiogenesis by increasing VEGF expression in chondrosarcoma cells

Emerging evidence indicates that angiogenesis contributes to tumour metastasis as a vital step in tumour progression [3]. VEGF is the most important angiogenic factor involved in angiogenesis and participating in tumorigenesis [12]. To learn whether bFGF promotes angiogenesis by VEGF expression, we found that VEGF expression increased with bFGF treatment in chondrosarcoma cells (JJ012) (Figures 1A–1C). Prior study indicates that EPCs contribute to both angiogenesis and angiogenesis-related growth. The data confirmed that tumours recruit EPCs to the tumour microenvironment, followed by differentiation into endothelial cells and promotion of vessel formation [8]. To investigate whether bFGF-induced VEGF secretion in chondrosarcoma cells could recruit EPCs and promote EPCs angiogenesis, we treated JJ012 cells with recombinant bFGF and collected CM after 24 h. The JJ012 CM was assessed for EPC migration and tube formation assays. The Transwell migration assay indicated that CM collected from bFGF-treated JJ012 increased EPC migration. Pre-treatment with VEGF-neutralizing antibody negated the effect (Figure 1D), suggesting that EPCs were recruited to the tumour microenvironment by bFGF-regulated VEGF expression. We also examined the angiogenic function of recruited EPCs, noting that CM collected from bFGF-treated chondrosarcoma cells increased EPC tube formation and was inhibited by VEGF-neutralizing antibody treatment (Figures 1E and 1F). Cell-surface receptors are key mediators to co-ordinate the extracellular response into cells; previous studies indicate that bFGF could regulate cell signalling via FGFR1 [18,19]. Transfection with FGFR1 siRNA inhibited bFGF-enhanced VEGF expression in chondrosarcoma and as well as EPC-primed angiogenesis (Figures 1G–1I). These results point to bFGF promoting VEGF expression in chondrosarcoma cells and regulating EPC-primed angiogenesis in chondrosarcoma.

bFGF regulates EPC-primed angiogenesis by raising VEGF expression in chondrosarcoma

Figure 1
bFGF regulates EPC-primed angiogenesis by raising VEGF expression in chondrosarcoma

(AC) JJ012 cells were incubated with bFGF (0–100 ng/ml) for 24 h, and VEGF expression examined by qPCR, ELISA and Western blotting. (DF) JJ012 cells were incubated with bFGF (0–100 ng/ml) for 24 h, and medium was collected as CM. EPCs were pre-treated for 30 min with anti-IgG control antibody or anti-VEGF antibody (1 μg/ml) and incubated with CM for 24 h, and cell migration was examined by Transwell assay (D). EPCs were incubated with CM for 6 h, and cell capillary-like structure formation in EPCs was photographed and counted (E and F). (G and H) JJ012 chondrosarcoma cells were transfected with FGFR1 siRNAs for 24 h, followed by stimulation with bFGF (30 ng/ml) for 24 h. VEGF expression was examined by qPCR and ELISA. (I and J). JJ012 cells were treated as in (G), and medium was collected as CM. The EPCs were incubated with CM for 24 h, and cell migration was examined by Transwell assay (I). Moreover, EPCs were incubated with CM for 6 h, and cell capillary-like structure formation in EPCs was examined by the tube formation assay (J). Results are means±S.E.M. *P<0.05 compared with control; #P<0.05 compared with bFGF-treated group. Ab, antibody.

Figure 1
bFGF regulates EPC-primed angiogenesis by raising VEGF expression in chondrosarcoma

(AC) JJ012 cells were incubated with bFGF (0–100 ng/ml) for 24 h, and VEGF expression examined by qPCR, ELISA and Western blotting. (DF) JJ012 cells were incubated with bFGF (0–100 ng/ml) for 24 h, and medium was collected as CM. EPCs were pre-treated for 30 min with anti-IgG control antibody or anti-VEGF antibody (1 μg/ml) and incubated with CM for 24 h, and cell migration was examined by Transwell assay (D). EPCs were incubated with CM for 6 h, and cell capillary-like structure formation in EPCs was photographed and counted (E and F). (G and H) JJ012 chondrosarcoma cells were transfected with FGFR1 siRNAs for 24 h, followed by stimulation with bFGF (30 ng/ml) for 24 h. VEGF expression was examined by qPCR and ELISA. (I and J). JJ012 cells were treated as in (G), and medium was collected as CM. The EPCs were incubated with CM for 24 h, and cell migration was examined by Transwell assay (I). Moreover, EPCs were incubated with CM for 6 h, and cell capillary-like structure formation in EPCs was examined by the tube formation assay (J). Results are means±S.E.M. *P<0.05 compared with control; #P<0.05 compared with bFGF-treated group. Ab, antibody.

bFGF promotes VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis by the c-Src/p38 signalling pathway

Intracellular signalling pathways that activate transcription factors and subsequently up-regulate gene expression are crucial to cell biological functions. A previous study found that c-Src and p38 signalling cascades were involved in VEGF expression [20]. We investigated the signalling pathway involved in bFGF-regulated VEGF expression. Pre-treatment or transfection with c-Src and p38 inhibitors or siRNAs reversed bFGF-induced VEGF expression in chondrosarcoma cells (Figures 2A–2C). EPC migration and tube formation induced by bFGF-treated JJ012 CM were abolished (Figures 2D and 2E). Finally, bFGF pre-treatment of chondrosarcoma cells (30 ng/ml) increased phosphory-lation of c-Src and p38 signalling proteins (Figure 2F); pre-treatment with c-Src inhibitor (PP2) decreased p38 phosphorylation (Figure 2G). These indicated that bFGF regulates VEGF expression and EPC-primed angiogenesis through the c-Src/p38 signalling pathway.

The Src/p38 signalling pathway is involved in bFGF-promoted VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis

Figure 2
The Src/p38 signalling pathway is involved in bFGF-promoted VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis

(AC) JJ012 cells were pre-treated with Src inhibitor PP2 (1 μM) or p38 inhibitor SB203580 (1 μM) for 30 min, or transfected with c-Src or p38 siRNAs for 24 h, followed by bFGF (30 ng/ml) stimulation for 24 h, with VEGF expression examined by qPCR, ELISA and Western blotting. (D and E) JJ012 cells were treated as in (A), and medium was collected as CM. EPCs were incubated with CM for 24 h, and cell migration was examined by Transwell assay (D). EPCs were incubated with CM for 6 h, and the tube formation assay was used to examine capillary-like structure formation in EPCs (E). (F) JJ012 cells were incubated with bFGF (30 ng/ml) for the designated times, and c-Src and p38 phosphorylation were determined by Western blotting. (G) JJ012 cells were pre-treated with the Src inhibitor PP2 (1 μM) for 30 min, followed by stimulation with bFGF (30 ng/ml) for 60 min, and p38 phosphorylation was determined by Western blotting. Results are means±S.E.M. *P<0.05 compared with control; #P<0.05 compared with bFGF-treated group.

Figure 2
The Src/p38 signalling pathway is involved in bFGF-promoted VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis

(AC) JJ012 cells were pre-treated with Src inhibitor PP2 (1 μM) or p38 inhibitor SB203580 (1 μM) for 30 min, or transfected with c-Src or p38 siRNAs for 24 h, followed by bFGF (30 ng/ml) stimulation for 24 h, with VEGF expression examined by qPCR, ELISA and Western blotting. (D and E) JJ012 cells were treated as in (A), and medium was collected as CM. EPCs were incubated with CM for 24 h, and cell migration was examined by Transwell assay (D). EPCs were incubated with CM for 6 h, and the tube formation assay was used to examine capillary-like structure formation in EPCs (E). (F) JJ012 cells were incubated with bFGF (30 ng/ml) for the designated times, and c-Src and p38 phosphorylation were determined by Western blotting. (G) JJ012 cells were pre-treated with the Src inhibitor PP2 (1 μM) for 30 min, followed by stimulation with bFGF (30 ng/ml) for 60 min, and p38 phosphorylation was determined by Western blotting. Results are means±S.E.M. *P<0.05 compared with control; #P<0.05 compared with bFGF-treated group.

bFGF promotes VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis by the NF-κB signalling pathway

HIF-1α (hypoxia-induced factor 1α) has been suggested to be the canonical transcription factor involved in VEGF expression [21], but other such factors bind to the VEGF promoter. NF-κB is often activated in tumours and is associated with progression of cancer. Previous study indicates the involvement of NF-κB in up-regulation of VEGF mRNA in breast cancer [22], suggesting possible regulation of VEGF expression by NF-κB in chondrosarcoma. Pre-treatment or transfection with NF-κB pathway inhibitors or p65 siRNAs reversed bFGF-induced VEGF expression (Figures 3A–3C). As expected, EPC migration and tube formation induced by bFGF-treated JJ012 CM were abolished (Figures 3D and 3E); bFGF (30 ng/ml) pre-treatment of chondrosarcoma cells increased, whereas pre-treatment with c-Src and p38 inhibitors impeded, p65 phosphorylation (Figures 3F and 3G). Luciferase reporter assay analysis of NF-κB activation revealed that stimulation of JJ012 cells with bFGF increased luciferase reporter activity and was negated by pre-treatment with c-Src, p38 and NF-κB inhibitor or siRNA (Figures 3H and 3I). These data hint that bFGF regulates VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis via NF-κB activation.

NF-κB transcription factor is involved in bFGF-promoted VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis

Figure 3
NF-κB transcription factor is involved in bFGF-promoted VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis

(AC) JJ012 cells were pre-treated with the NF-κB inhibitor PDTC (10 μM) or TPCK (1 μM) for 30 min, or transfection with p65 siRNAs for 24 h, followed by stimulation with bFGF (30 ng/ml) for 24 h. VEGF expression was examined by qPCR, ELISA and Western blotting. (D and E) JJ012 cells were treated as in (A), and medium was collected as CM. EPCs were incubated with CM for 24 h, and cell migration was tested by Transwell assay (D). EPCs were incubated with CM for 6 h, and the tube formation assay was used to examine capillary-like structure formation (E). (F) JJ012 cells were incubated with bFGF (30 ng/ml) for the designated times, and p65 phosphorylation was assessed by Western blotting. (G) JJ012 chondrosarcoma cells were pre-treated with the Src inhibitor PP2 (1 μM) and the p38 inhibitor SB203580 (1 μM) for 30 min, followed by stimulation with bFGF (30 ng/ml) for 24 h, with p65 phosphorylation determined by Western blotting. (H) JJ012 cells were transfected for 24 h with NF-κB promoter reporter plasmid, treated with the Src inhibitor PP2 (1 μM), the p38 inhibitor SB203580 (1 μM), the NF-κB inhibitors PDTC (10 μM) or TPCK (1 μM) for 30 min, followed by bFGF (30 ng/ml) stimulation for 24 h, then analysed for luciferase activity. (I) JJ012 cells were co-transfected for 24 h with NF-κB promoter reporter plasmid and c-Src, p38 or p65 siRNAs for 24 h, followed by stimulation with bFGF (30 ng/ml) for 24 h, and analysed for luciferase activity. Results are means±S.E.M. *P<0.05 compared with control; #P<0.05 compared with bFGF-treated group.

Figure 3
NF-κB transcription factor is involved in bFGF-promoted VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis

(AC) JJ012 cells were pre-treated with the NF-κB inhibitor PDTC (10 μM) or TPCK (1 μM) for 30 min, or transfection with p65 siRNAs for 24 h, followed by stimulation with bFGF (30 ng/ml) for 24 h. VEGF expression was examined by qPCR, ELISA and Western blotting. (D and E) JJ012 cells were treated as in (A), and medium was collected as CM. EPCs were incubated with CM for 24 h, and cell migration was tested by Transwell assay (D). EPCs were incubated with CM for 6 h, and the tube formation assay was used to examine capillary-like structure formation (E). (F) JJ012 cells were incubated with bFGF (30 ng/ml) for the designated times, and p65 phosphorylation was assessed by Western blotting. (G) JJ012 chondrosarcoma cells were pre-treated with the Src inhibitor PP2 (1 μM) and the p38 inhibitor SB203580 (1 μM) for 30 min, followed by stimulation with bFGF (30 ng/ml) for 24 h, with p65 phosphorylation determined by Western blotting. (H) JJ012 cells were transfected for 24 h with NF-κB promoter reporter plasmid, treated with the Src inhibitor PP2 (1 μM), the p38 inhibitor SB203580 (1 μM), the NF-κB inhibitors PDTC (10 μM) or TPCK (1 μM) for 30 min, followed by bFGF (30 ng/ml) stimulation for 24 h, then analysed for luciferase activity. (I) JJ012 cells were co-transfected for 24 h with NF-κB promoter reporter plasmid and c-Src, p38 or p65 siRNAs for 24 h, followed by stimulation with bFGF (30 ng/ml) for 24 h, and analysed for luciferase activity. Results are means±S.E.M. *P<0.05 compared with control; #P<0.05 compared with bFGF-treated group.

Knockdown of bFGF expression decreases VEGF expression in chondrosarcoma cells and EPC-primed angiogenesis

We saw that bFGF promoted VEGF expression in chondrosarcoma and enhanced EPC-primed angiogenesis. It is critical to pinpoint the role of bFGF in vivo. To confirm its regulatory role in VEGF expression, we utilized JJ012 cells stably expressing bFGF shRNA. Results showed that the expression level of bFGF was decreased in bFGF shRNA stable clone. Moreover, bFGF knockdown significantly reduced VEGF expression (Figures 4A and 4B). CM collected from JJ012/control shRNA promoted EPC cell migration and tube formation, but this decreased when incubating with CM collected from JJ012/bFGF shRNA (Figures 4C and 4D). Finally, the bFGF in vivo role was examined using a chick embryo CAM assay. As expected, CM amassed from JJ012/control shRNA cells obviously enhanced CAM angiogenesis, whereas bFGF shRNA completely reduced angiogenesis in CAMs (Figure 4E). For in vivo Matrigel plug formation assay by subcutaneous implantation in mice, Matrigel mixed with CM from JJ012/control shRNA increased blood vessel growth; CM from JJ012/bFGF shRNA starkly reduced neovascularization (Figure 4F, top row). Vascular formation in Matrigel declined in CD31 IHC and haemoglobin content assay (Figure 4F, middle and bottom rows); these results indicate that bFGF promoted angiogenesis through VEGF expression in vivo.

Knockdown of bFGF expression in chondrosarcoma cells lowers VEGF expression and EPC-primed angiogenesis

Figure 4
Knockdown of bFGF expression in chondrosarcoma cells lowers VEGF expression and EPC-primed angiogenesis

(A and B) The mRNA and protein expressions of bFGF and VEGF in JJ012/control shRNA and JJ012/bFGF shRNA were examined by qPCR, Western blotting and ELISA. (C and D) EPCs were incubated with CM collected from JJ012/control shRNA or JJ012/bFGF shRNA for 24 h, and cell migration or tube formation was examined by Transwell assay or photographed under a microscope. (E) Top and middle panels: PBS, VEGF, JJ012/control shRNA CM or JJ012/bFGF shRNA CM suspended in Matrigel were placed on chick CAMs. CAM arteriosus branches in groups were photographed on developmental day 12. Bottom panel: quantification of the branch number in the CAM assay (n=7). (F) Mice were subcutaneously injected with Matrigel mixed with PBS, JJ012/control shRNA CM or JJ012/bFGF shRNA CM for 7 days. Plugs excised from mice were photographed and stained with CD31, quantification of MVD is shown as mean±S.E.M. percentages of three random fields (n=5). #P < 0.05 compared with JJ012/control sh-RNA group.

Figure 4
Knockdown of bFGF expression in chondrosarcoma cells lowers VEGF expression and EPC-primed angiogenesis

(A and B) The mRNA and protein expressions of bFGF and VEGF in JJ012/control shRNA and JJ012/bFGF shRNA were examined by qPCR, Western blotting and ELISA. (C and D) EPCs were incubated with CM collected from JJ012/control shRNA or JJ012/bFGF shRNA for 24 h, and cell migration or tube formation was examined by Transwell assay or photographed under a microscope. (E) Top and middle panels: PBS, VEGF, JJ012/control shRNA CM or JJ012/bFGF shRNA CM suspended in Matrigel were placed on chick CAMs. CAM arteriosus branches in groups were photographed on developmental day 12. Bottom panel: quantification of the branch number in the CAM assay (n=7). (F) Mice were subcutaneously injected with Matrigel mixed with PBS, JJ012/control shRNA CM or JJ012/bFGF shRNA CM for 7 days. Plugs excised from mice were photographed and stained with CD31, quantification of MVD is shown as mean±S.E.M. percentages of three random fields (n=5). #P < 0.05 compared with JJ012/control sh-RNA group.

Knockdown of bFGF decreases angiogenesis-related tumour growth in vivo

To evaluate that bFGF contributes to VEGF expression and EPC-primed angiogenesis in the chondrosarcoma microenvironment, as well as promoting tumour growth, our experiments used bFGF shRNA stable cell line in a mouse xenograft model. Bioluminescence images indicated that JJ012/control shRNA profoundly induces tumour mass formation, but knockdown of bFGF reduced tumour growth in mice (Figures 5A–5D). We quantified the level of angiogenesis by examining haemoglobin content of tumours to find that knockdown of bFGF expression impedes chondrosarcoma-related angiogenesis in vivo (Figure 5E). Haemoglobin content also positively correlated with tumour volume (Figure 5F). Finally, IHC results of bFGF and CD31 confirmed bFGF-promoted angiogenesis (Figure 5G). Moreover, detection of positive staining of EPC markers (CD-34 and CD-133) in the control shRNA group but not the bFGF shRNA group showed that bFGF recruited EPCs in the chondrosarcoma microenvironment (Figure 5G). In summary, data revealed that bFGF promotes VEGF expression, recruits EPCs and promotes EPC-primed angiogenesis and tumour growth in vivo.

Knockdown of bFGF expression decreases angiogenesis-related tumour growth in vivo

Figure 5
Knockdown of bFGF expression decreases angiogenesis-related tumour growth in vivo

(A) JJ012/control shRNA and JJ012/bFGF shRNA cells were mixed with Matrigel and injected into flank sites of mice for 28 days, and tumour growth was monitored using the IVIS Imaging System (n=6). (BE) Fluorescent imaging quantified data at weeks 0–6, after which mice were killed. Tumours were resected, photographed by a microscope, measured for weight and volume, and haemoglobin quantified. (F) Correlation between tumour volume and haemoglobin level. (G) Tumours were paraffin-embedded, and anti-bFGF, anti-CD31, anti-CD34 and anti-CD133 antibodies were used for IHC assay. Results are means±S.E.M. #P < 0.05 compared with JJ012/control sh-RNA group.

Figure 5
Knockdown of bFGF expression decreases angiogenesis-related tumour growth in vivo

(A) JJ012/control shRNA and JJ012/bFGF shRNA cells were mixed with Matrigel and injected into flank sites of mice for 28 days, and tumour growth was monitored using the IVIS Imaging System (n=6). (BE) Fluorescent imaging quantified data at weeks 0–6, after which mice were killed. Tumours were resected, photographed by a microscope, measured for weight and volume, and haemoglobin quantified. (F) Correlation between tumour volume and haemoglobin level. (G) Tumours were paraffin-embedded, and anti-bFGF, anti-CD31, anti-CD34 and anti-CD133 antibodies were used for IHC assay. Results are means±S.E.M. #P < 0.05 compared with JJ012/control sh-RNA group.

bFGF is highly expressed in chondrosarcoma and is correlated with VEGF expression

Our in vitro and in vivo results showed that bFGF promotes EPC-primed angiogenesis by regulating VEGF expression in chondrosarcoma cells. It is important to investigate the correlation between bFGF and VEGF in clinical specimens. We tabulated bFGF and VEGF expression levels in specimens collected from human chondrosarcoma cases, using IHC to find that bFGF and VEGF were nearly undetectable in normal cartilage, but were associated with higher clinical pathological grades (Figure 6A–6C). Quantitative data show that bFGF expression correlates with VEGF expression in our specimens (Figure 6D), suggesting that bFGF is linked to VEGF expression and tumour progression in chondrosarcoma.

Correlation of bFGF, VEGF and tumour grades in human chondrosarcoma tissues

Figure 6
Correlation of bFGF, VEGF and tumour grades in human chondrosarcoma tissues

Tumour specimens were stained for IHC with anti-bFGF and anti-VEGF antibodies, with the staining intensity scored as 1–5. IHC quantified bFGF and VEGF expression. (A) IHC photography. (BD) Quantitative results and correlation between bFGF, VEGF and clinical grade of chondrosarcoma. (E) Diagrammatic model of the role of bFGF in chondrosarcoma tumour microenvironment. (1) bFGF induces VEGF expression and secretion in JJ012 chondrosarcoma through FGFR1/c-Src/p38/NF-κB signal pathway activation. (2) The bFGF-induced secretion of VEGF subsequently recruits EPCs to the chondrosarcoma tumour microenvironment and promotes neoangiogenesis.

Figure 6
Correlation of bFGF, VEGF and tumour grades in human chondrosarcoma tissues

Tumour specimens were stained for IHC with anti-bFGF and anti-VEGF antibodies, with the staining intensity scored as 1–5. IHC quantified bFGF and VEGF expression. (A) IHC photography. (BD) Quantitative results and correlation between bFGF, VEGF and clinical grade of chondrosarcoma. (E) Diagrammatic model of the role of bFGF in chondrosarcoma tumour microenvironment. (1) bFGF induces VEGF expression and secretion in JJ012 chondrosarcoma through FGFR1/c-Src/p38/NF-κB signal pathway activation. (2) The bFGF-induced secretion of VEGF subsequently recruits EPCs to the chondrosarcoma tumour microenvironment and promotes neoangiogenesis.

DISCUSSION

Unlike other mesenchymal malignancies such as osteosarcoma and Ewing's sarcoma, which saw dramatically longer survival with the advent of systemic chemotherapy, chondrosarcoma continues to show poor prognosis without effective adjuvant therapy [23]. In the last decade, it has become a general concept that angiogenesis could serve as a therapeutic target for blocking cancer growth; VEGF has been implicated as a pivotal target of anti-angiogenic therapy [24]. Previous observation indicates that high microvessel density is associated with higher histological grade and poor prognosis in chondrosarcoma [25], prompting the search for a means to prevent angiogenesis. We found evidence that bFGF promoted EPC-primed angiogenesis by regulating VEGF expression in chondrosarcoma cells. Also, bFGF regu-lated VEGF expression through the c-Src/p38/NF-κB signalling pathway in chondrosarcoma cells. The IHC results from clinical specimens of chondrosarcoma validated our findings. In summary, our clinical specimens, cell experiments and animal models suggest that bFGF boosts VEGF expression and expedites EPC-primed angiogenesis.

Owing to the pro-angiogenic function and tumour angiogenesis effects of bFGF, up-regulation of bFGF is discussed in various types of cancer such as OSCC, oesophageal cancer, haematological malignancies and glioblastoma [9,10,26,27]. However, controversial results were found in breast cancer. bFGF expression in breast cancer cells shows a less malignant phenotype by decreasing motility and invasion [28]. Another study also provides the evidence of pro-apoptotic effects of bFGF in breast cancer by regulating expression of Bcl-2 family members [29]. In contrast, some studies indicate that bFGF promotes cell proliferation and treatment with bFGF antagonist peptide (P7) inhibits breast cancer cell growth [30,31]. These observations may be caused by cell-type-specific effects and heterogeneity of cancer cells. However, the role of bFGF in chondrosarcoma is poorly discussed.

Previous study indicates that bFGF regulates MMP-13 expression to promote chondrosarcoma progression [11]. Moreover, bFGF increases proliferation of the SW1353 chondrosarcoma cell line [13]. Our in vivo results showed that inhibition of tumour growth by bFGF knockdown may be due to proliferation inhibition and decreasing angiogenesis-related tumour growth (Figures 5A–5D). In spite of this, the angiogenic regulation by bFGF shows a novel therapeutic opportunity. A single angiogenic inhibitor is not sufficient to inhibit angiogenesis and tumour growth. The pivotal role of bFGF in chondrosarcoma may provide a combination treatment to inhibit the bFGF and VEGF pathways.

Since EPCs were first described a decade ago, emerging evidence indicates their contribution to angiogenesis-mediated growth of certain tumours in mice and humans [8]. Recruitment of EPCs is regulated by growth factors, chemokines and cytokines secreted during tumour growth [32]. One of them is VEGF; it modulates the bone marrow microenvironment from a quiescent to a highly pro-angiogenic state, thus promoting mobilization of both vascular and haematopoietic progenitors to peripheral circulation, which are recruited to primary tumours or metastatic lesions [33]. The role of bFGF, a pivotal component involved in angiogenesis, in EPC regulation is little discussed. We found tumour-secreted bFGF recruiting EPCs and promoting angiogenesis through VEGF expression in chondrosarcoma cells (Figures 1A–1F). It has been proposed that bFGF and VEGF are the most important tumour-secreted angiogenic cytokines in lung cancer [34], but the angiogenic cytokine involved in chondrosarcoma has not been discussed. Our study is the first to confirm the role of these cytokines in angiogenesis, warranting investigation of their clinical significance in chondrosarcoma.

Cell-surface receptors are important mediators to co-ordinate extracellular responses into cells. Yet the receptor activated by bFGF is little discussed. We classified FGFR1 as the binding receptor activated by bFGF treatment. Our results showed that transfection with FGFR1 siRNA abolishes bFGF-induced VEGF expression in chondrosarcoma (Figures 1G–1J). These results reveal a vital role for FGFR1 in bFGF activation. We have first discussed the involvement of FGFR1 in bFGF activation, meriting further examination to reveal its clinical relevance. Likewise, downstream effectors and transcription factors of FGFR1 by bFGF activation prove critical for exerting cell biological functions. We found that bFGF activates c-Src and p38 signalling proteins to regulate VEGF expression (Figure 2). Previous studies indicated that FGFRs elicited different signal cascades, e.g. Src, PLCγ (phospholipase γ), PI3K (phosphoinositide 3-kinase), Ras and MAPK (mitogen-activated protein kinase) signalling proteins [3539]. Our results showed that bFGF promotes VEGF expression in chondrosarcoma via c-Src/p38 signal cascade. c-Src, a non-receptor tyrosine kinase, is a co-transducer of mitogenic signals emanating from numerous tyrosine kinase growth factor receptors, including EGFR (epidermal growth factor receptor), PDGFR (platelet-derived growth factor receptor), CSF-1R (colony-stimulating factor 1 receptor) and bFGFR (bFGF receptor) [40]. Prior study shows that the Src signalling protein is highly activated in chondrosarcoma, playing a role in chemoresistance [41,42]. Dasatinib, a small-molecule tyrosine kinase inhibitor targeting Src, decreases viability in chondrosarcoma cell lines. Whether or not it can abolish bFGF-regulated angiogenesis in chondrosarcoma will be determined in the future.

AUTHOR CONTRIBUTION

Huey-En Tzeng: conception and design, collection of data, data analysis and interpretation; Po-Chun Chen: data analysis and interpretation, and writing the paper; Kai-Wei Lin and Chih-Yang Lin: conception and design, collection of data, data analysis and interpretation; Chun-Hao Tsai, Shao-Min Han, Chieh-Lin Teng, Wen-Li Hwang and Shih-Wei Wang: provision of study material; Chih-Hsin Tang: conception and design, data analysis and interpretation, writing the paper and provision of study material.

FUNDING

This study was supported by the National Science Council of Taiwan [grant number NSC 102-2632-B-039-001-MY3), China Medical University [grant number CMU 103-S-06], Ministry of Science and Technology [grant number MOST 103-2628-B-039-002] and Taichung Veterans General Hospital [grant number TCVGH-1043701C].

Abbreviations

     
  • bFGF

    basic fibroblast growth factor

  •  
  • CAM

    chorioallantoic membrane

  •  
  • CM

    conditioned medium

  •  
  • DMEM

    Dulbecco’s modified Eagle’s medium

  •  
  • EPC

    endothelial progenitor cell

  •  
  • FGF

    fibroblast growth factor

  •  
  • FGFR

    FGF receptor

  •  
  • IHC

    immunohistochemistry

  •  
  • α-MEM

    α-minimum essential medium

  •  
  • MMP-13

    matrix metalloproteinase 13

  •  
  • MVD

    microvessel density

  •  
  • NF-κB

    nuclear factor κB

  •  
  • OSCC

    oral squamous cell carcinoma

  •  
  • PDTC

    pyrrolidine dithiocarbamate

  •  
  • qPCR

    quantitative real-time PCR

  •  
  • TPCK

    tosylphenylalanylchloromethane

  •  
  • VEGF

    vascular endothelial growth factor

  •  
  • VEGFR2

    VEGF receptor 2

References

References
1
Terek
 
R.M.
Schwartz
 
G.K.
Devaney
 
K.
Glantz
 
L.
Mak
 
S.
Healey
 
J.H.
Albino
 
A.P.
 
Chemotherapy and P-glycoprotein expression in chondrosarcoma
J. Orthop. Res.
1998
, vol. 
16
 (pg. 
585
-
590
)
[PubMed]
2
Yuan
 
J.
Dutton
 
C.M.
Scully
 
S.P.
 
RNAi mediated MMP-1 silencing inhibits human chondrosarcoma invasion
J. Orthop. Res.
2005
, vol. 
23
 (pg. 
1467
-
1474
)
[PubMed]
3
Baeriswyl
 
V.
Christofori
 
G.
 
The angiogenic switch in carcinogenesis
Semin. Cancer Biol.
2009
, vol. 
19
 (pg. 
329
-
337
)
[PubMed]
4
Asahara
 
T.
Murohara
 
T.
Sullivan
 
A.
Silver
 
M.
van der Zee
 
R.
Li
 
T.
Witzenbichler
 
B.
Schatteman
 
G.
Isner
 
J.M.
 
Isolation of putative progenitor endothelial cells for angiogenesis
Science
1997
, vol. 
275
 (pg. 
964
-
967
)
[PubMed]
5
Zhao
 
Y.H.
Yuan
 
B.
Chen
 
J.
Feng
 
D.H.
Zhao
 
B.
Qin
 
C.
Chen
 
Y.F.
 
Endothelial progenitor cells: therapeutic perspective for ischemic stroke
CNS Neurosci. Ther.
2013
, vol. 
19
 (pg. 
67
-
75
)
6
Patenaude
 
A.
Parker
 
J.
Karsan
 
A.
 
Involvement of endothelial progenitor cells in tumor vascularization
Microvasc. Res.
2010
, vol. 
79
 (pg. 
217
-
223
)
[PubMed]
7
Kirton
 
J.P.
Xu
 
Q.
 
Endothelial precursors in vascular repair
Microvasc. Res.
2010
, vol. 
79
 (pg. 
193
-
199
)
[PubMed]
8
Gao
 
D.
Nolan
 
D.
McDonnell
 
K.
Vahdat
 
L.
Benezra
 
R.
Altorki
 
N.
Mittal
 
V.
 
Bone marrow-derived endothelial progenitor cells contribute to the angiogenic switch in tumor growth and metastatic progression
Biochim. Biophys. Acta
2009
, vol. 
1796
 (pg. 
33
-
40
)
9
Hase
 
T.
Kawashiri
 
S.
Tanaka
 
A.
Nozaki
 
S.
Noguchi
 
N.
Kato
 
K.
Nakaya
 
H.
Nakagawa
 
K.
 
Correlation of basic fibroblast growth factor expression with the invasion and the prognosis of oral squamous cell carcinoma
J. Oral Pathol. Med.
2006
, vol. 
35
 (pg. 
136
-
139
)
[PubMed]
10
Barclay
 
C.
Li
 
A.W.
Geldenhuys
 
L.
Baguma-Nibasheka
 
M.
Porter
 
G.A.
Veugelers
 
P.J.
Murphy
 
P.R.
Casson
 
A.G.
 
Basic fibroblast growth factor (FGF-2) overexpression is a risk factor for esophageal cancer recurrence and reduced survival, which is ameliorated by coexpression of the FGF-2 antisense gene
Clin. Cancer Res.
2005
, vol. 
11
 (pg. 
7683
-
7691
)
[PubMed]
11
Uria
 
J.A.
Balbin
 
M.
Lopez
 
J.M.
Alvarez
 
J.
Vizoso
 
F.
Takigawa
 
M.
Lopez-Otin
 
C.
 
Collagenase-3 (MMP-13) expression in chondrosarcoma cells and its regulation by basic fibroblast growth factor
Am. J. Pathol.
1998
, vol. 
153
 (pg. 
91
-
101
)
[PubMed]
12
Bergers
 
G.
Benjamin
 
L.E.
 
Tumorigenesis and the angiogenic switch
Nat. Rev. Cancer
2003
, vol. 
3
 (pg. 
401
-
410
)
[PubMed]
13
Sugimoto
 
H.
Nishino
 
H.
 
[Effect of recombinant human basic fibroblast growth factor (bFGF) on the growth of human tumor cell lines]
Hum. Cell.
1996
, vol. 
9
 (pg. 
129
-
140
)
[PubMed]
14
Coppola
 
G.
Atlas-White
 
M.
Katsahambas
 
S.
Bertolini
 
J.
Hearn
 
M.T.
Underwood
 
J.R.
 
Effect of intraperitoneally, intravenously and intralesionally administered monoclonal anti-β-FGF antibodies on rat chondrosarcoma tumor vascularization and growth
Anticancer Res.
1997
, vol. 
17
 (pg. 
2033
-
2039
)
[PubMed]
15
Wang
 
H.H.
Lin
 
C.A.
Lee
 
C.H.
Lin
 
Y.C.
Tseng
 
Y.M.
Hsieh
 
C.L.
Chen
 
C.H.
Tsai
 
C.H.
Hsieh
 
C.T.
Shen
 
J.L.
, et al 
Fluorescent gold nanoclusters as a biocompatible marker for in vitro and in vivo tracking of endothelial cells
ACS Nano
2011
, vol. 
5
 (pg. 
4337
-
4344
)
[PubMed]
16
Wang
 
H.H.
Su
 
C.H.
Wu
 
Y.J.
Li
 
J.Y.
Tseng
 
Y.M.
Lin
 
Y.C.
Hsieh
 
C.L.
Tsai
 
C.H.
Yeh
 
H.I.
 
Reduction of connexin43 in human endothelial progenitor cells impairs the angiogenic potential
Angiogenesis
2013
, vol. 
16
 (pg. 
553
-
560
)
[PubMed]
17
Chung
 
C.H.
Chang
 
C.H.
Chen
 
S.S.
Wang
 
H.H.
Yen
 
J.Y.
Hsiao
 
C.J.
Wu
 
N.L.
Chen
 
Y.L.
Huang
 
T.F.
Wang
 
P.C.
, et al 
Butein inhibits angiogenesis of human endothelial progenitor cells via the translation dependent signaling pathway. Evid. Based Complement
Alternat. Med.
2013
, vol. 
2013
 pg. 
943187
 
18
Lam
 
K.
Rao
 
V.S.
Qasba
 
P.K.
 
Molecular modeling studies on binding of bFGF to heparin and its receptor FGFR1
J. Biomol. Struct. Dyn.
1998
, vol. 
15
 (pg. 
1009
-
1027
)
[PubMed]
19
Gu
 
Y.
Xue
 
C.
Zhu
 
J.
Sun
 
H.
Ding
 
F.
Cao
 
Z.
Gu
 
X.
 
Basic fibroblast growth factor (bFGF) facilitates differentiation of adult dorsal root ganglia-derived neural stem cells toward Schwann cells by binding to FGFR-1 through MAPK/ERK activation
J. Mol. Neurosci.
2014
, vol. 
52
 (pg. 
538
-
551
)
[PubMed]
20
Zetser
 
A.
Bashenko
 
Y.
Edovitsky
 
E.
Levy-Adam
 
F.
Vlodavsky
 
I.
Ilan
 
N.
 
Heparanase induces vascular endothelial growth factor expression: correlation with p38 phosphorylation levels and Src activation
Cancer Res.
2006
, vol. 
66
 (pg. 
1455
-
1463
)
[PubMed]
21
Ahluwalia
 
A.
Tarnawski
 
A.S.
 
Critical role of hypoxia sensor HIF-1α in VEGF gene activation: implications for angiogenesis and tissue injury healing
Curr. Med. Chem.
2012
, vol. 
19
 (pg. 
90
-
97
)
[PubMed]
22
Shibata
 
A.
Nagaya
 
T.
Imai
 
T.
Funahashi
 
H.
Nakao
 
A.
Seo
 
H.
 
Inhibition of NF-κB activity decreases the VEGF mRNA expression in MDA-MB-231 breast cancer cells
Breast Cancer Res Treat.
2002
, vol. 
73
 (pg. 
237
-
243
)
[PubMed]
23
Fong
 
Y.C.
Yang
 
W.H.
Hsu
 
S.F.
Hsu
 
H.C.
Tseng
 
K.F.
Hsu
 
C.J.
Lee
 
C.Y.
Scully
 
S.P.
 
2-Methoxyestradiol induces apoptosis and cell cycle arrest in human chondrosarcoma cells
J. Orthop. Res.
2007
, vol. 
25
 (pg. 
1106
-
1114
)
[PubMed]
24
Welti
 
J.
Loges
 
S.
Dimmeler
 
S.
Carmeliet
 
P.
 
Recent molecular discoveries in angiogenesis and antiangiogenic therapies in cancer
J. Clin. Invest.
2013
, vol. 
123
 (pg. 
3190
-
3200
)
[PubMed]
25
Kubo
 
T.
Shimose
 
S.
Fujimori
 
J.
Arihiro
 
K.
Ochi
 
M.
 
Diversity of angiogenesis among malignant bone tumors
Mol. Clin. Oncol.
2013
, vol. 
1
 (pg. 
131
-
136
)
[PubMed]
26
Haley
 
E.M.
Kim
 
Y.
 
The role of basic fibroblast growth factor in glioblastoma multiforme and glioblastoma stem cells and in their in vitro culture
Cancer Lett.
2014
, vol. 
346
 (pg. 
1
-
5
)
[PubMed]
27
Ribatti
 
D.
Vacca
 
A.
Rusnati
 
M.
Presta
 
M.
 
The discovery of basic fibroblast growth factor/fibroblast growth factor-2 and its role in haematological malignancies
Cytokine Growth Factor Rev.
2007
, vol. 
18
 (pg. 
327
-
334
)
[PubMed]
28
Korah
 
R.M.
Sysounthone
 
V.
Golowa
 
Y.
Wieder
 
R.
 
Basic fibroblast growth factor confers a less malignant phenotype in MDA-MB-231 human breast cancer cells
Cancer Res.
2000
, vol. 
60
 (pg. 
733
-
740
)
[PubMed]
29
Maloof
 
P.
Wang
 
Q.
Wang
 
H.
Stein
 
D.
Denny
 
T.N.
Yahalom
 
J.
Fenig
 
E.
Wieder
 
R.
 
Overexpression of basic fibroblast growth factor (FGF-2) downregulates Bcl-2 and promotes apoptosis in MCF-7 human breast cancer cells
Breast Cancer Res. Treat.
1999
, vol. 
56
 (pg. 
153
-
167
)
[PubMed]
30
Li
 
Q.
Gao
 
S.
Yu
 
Y.
Wang
 
W.
Chen
 
X.
Wang
 
R.
Li
 
T.
Wang
 
C.
Li
 
X.
Wu
 
X.
 
A novel bFGF antagonist peptide inhibits breast cancer cell growth
Mol. Med. Rep.
2012
, vol. 
6
 (pg. 
210
-
214
)
[PubMed]
31
Yang
 
Z.L.
Cheng
 
K.
Han
 
Z.D.
 
Effect of bFGF on the MCF-7 cell cycle with CD44+/CD24−: promoting the G0/G1→G2/S transition
J. Breast Cancer
2012
, vol. 
15
 (pg. 
388
-
392
)
[PubMed]
32
Hillen
 
F.
Griffioen
 
A.W.
 
Tumour vascularization: sprouting angiogenesis and beyond
Cancer Metastasis Rev.
2007
, vol. 
26
 (pg. 
489
-
502
)
[PubMed]
33
Kopp
 
H.G.
Avecilla
 
S.T.
Hooper
 
A.T.
Rafii
 
S.
 
The bone marrow vascular niche: home of HSC differentiation and mobilization
Physiology
2005
, vol. 
20
 (pg. 
349
-
356
)
[PubMed]
34
Bremnes
 
R.M.
Camps
 
C.
Sirera
 
R.
 
Angiogenesis in non-small cell lung cancer: the prognostic impact of neoangiogenesis and the cytokines VEGF and bFGF in tumours and blood
Lung Cancer
2006
, vol. 
51
 (pg. 
143
-
158
)
[PubMed]
35
Mohammadi
 
M.
Honegger
 
A.M.
Rotin
 
D.
Fischer
 
R.
Bellot
 
F.
Li
 
W.
Dionne
 
C.A.
Jaye
 
M.
Rubinstein
 
M.
Schlessinger
 
J.
 
A tyrosine-phosphorylated carboxy-terminal peptide of the fibroblast growth factor receptor (Flg) is a binding site for the SH2 domain of phospholipase C-γ1
Mol. Cell. Biol.
1991
, vol. 
11
 (pg. 
5068
-
5078
)
[PubMed]
36
Chen
 
Y.
Li
 
X.
Eswarakumar
 
V.P.
Seger
 
R.
Lonai
 
P.
 
Fibroblast growth factor (FGF) signaling through PI 3-kinase and Akt/PKB is required for embryoid body differentiation
Oncogene
2000
, vol. 
19
 (pg. 
3750
-
3756
)
[PubMed]
37
Weinstein
 
D.C.
Marden
 
J.
Carnevali
 
F.
Hemmati-Brivanlou
 
A.
 
FGF-mediated mesoderm induction involves the Src-family kinase Laloo
Nature
1998
, vol. 
394
 (pg. 
904
-
908
)
[PubMed]
38
O’Neill
 
E.
Kolch
 
W.
 
Conferring specificity on the ubiquitous Raf/MEK signalling pathway
Br. J. Cancer
2004
, vol. 
90
 (pg. 
283
-
288
)
[PubMed]
39
Lu
 
Q.
Wang
 
C.
Pan
 
R.
Gao
 
X.
Wei
 
Z.
Xia
 
Y.
Dai
 
Y.
 
Histamine synergistically promotes bFGF-induced angiogenesis by enhancing VEGF production via H1 receptor
J. Cell. Biochem.
2013
, vol. 
114
 (pg. 
1009
-
1019
)
[PubMed]
40
Belsches
 
A.P.
Haskell
 
M.D.
Parsons
 
S.J.
 
Role of c-Src tyrosine kinase in EGF-induced mitogenesis
Front. Biosci.
1997
, vol. 
2
 (pg. 
d501
-
d518
)
[PubMed]
41
Schrage
 
Y.M.
Briaire-de Bruijn
 
I.H.
de Miranda
 
N.F.
van Oosterwijk
 
J.
Taminiau
 
A.H.
van Wezel
 
T.
Hogendoorn
 
P.C.
Bovee
 
J.V.
 
Kinome profiling of chondrosarcoma reveals SRC-pathway activity and dasatinib as option for treatment
Cancer Res.
2009
, vol. 
69
 (pg. 
6216
-
6222
)
[PubMed]
42
van Oosterwijk
 
J.G.
van Ruler
 
M.A.
Briaire-de Bruijn
 
I.H.
Herpers
 
B.
Gelderblom
 
H.
van de Water
 
B.
Bovee
 
J.V.
 
Src kinases in chondrosarcoma chemoresistance and migration: dasatinib sensitises to doxorubicin in TP53 mutant cells
Br. J. Cancer
2013
, vol. 
109
 (pg. 
1214
-
1222
)
[PubMed]

Author notes

1

These authors contributed equally to this work and share first authorship.