FGF5 as an oncogenic factor in human glioblastoma multiforme: autocrine and paracrine activities

Abstract

Fibroblast growth factor 5 (FGF5) is widely expressed in embryonic but scarcely in adult tissues. Here we report simultaneous overexpression of FGF5 and its predominant high-affinity receptor (FGFR1 IIIc) in astrocytic brain tumour specimens (N=49) and cell cultures (N=49). The levels of both ligand and receptor increased with enhanced malignancy in vivo and in vitro. Furthermore, secreted FGF5 protein was generally present in the supernatants of glioblastoma (GBM) cells. siRNA-mediated FGF5 downmodulation reduced moderately but significantly GBM cell proliferation while recombinant FGF5 (rFGF5) increased this parameter preferentially in cell lines with low endogenous expression levels. Apoptosis induction by prolonged serum starvation was significantly prevented by rFGF5. Moreover, tumour cell migration was distinctly stimulated by rFGF5 but attenuated by FGF5 siRNA. Blockade of FGFR1-mediated signals by pharmacological FGFR inhibitors or a dominant-negative FGFR1 IIIc protein inhibited GBM cell proliferation and/or induced apoptotic cell death. Moreover, rFGF5 and supernatants of highly FGF5-positive GBM cell lines specifically stimulated proliferation, migration and tube formation of human umbilical vein endothelial cells. In summary, we demonstrate for the first time that FGF5 contributes to the malignant progression of human astrocytic brain tumours by both autocrine and paracrine effects.

Figure 1

Expression of fibroblast growth factor 5 (FGF5) and FGFR1 in astrocytic brain tumour specimens and cell cultures. (a) FGF5 mRNA expression in tissue specimens (upper panel) and cell cultures (middle panel) was quantified by real-time RT-PCR. FGF5 mRNA expression was normalized to the housekeeping gene β-2 microglobulin. Data are given relative to an internal control (the GBM cell line MGC) included in all experiments and set arbitrarily as 1. Means are indicated by horizontal bars. (Lower panel) Representative examples of FGF5 and β-actin reverse transcription (RT)–PCR products amplified from the indicated tissue specimens are shown. NM, non-malignant brain; AC II, low-grade astrocytoma; AC III, anaplastic astrocytoma; GBM, glioblastoma. (b) Formalin-fixed, paraffin-embedded tissue specimens from AC II, AC III and GBM were immunostained using FGF5- and FGFR1-specific antibodies. Staining was evaluated as described under ‘Materials and methods’ and results given as quantitative scores (QS, means±s.d.) of all samples analysed. Significant differences are indicated (Student’s t-test, *P<0.05; ***P<0.0001). (c) Examples of FGF5 and FGFR1 immunohistochemical stainings in the indicated tissue specimens are shown. (d) FGF5 secretion by selected GBM cell lines was investigated by western blots. Serum-free culture medium (CM) and medium spiked with 150 ng ml⁻¹ rFGF5 (rFGF5) were used as controls. (e, f) Immunofluorescence staining of FGF5 (e) and FGFR1 (f) was performed in the indicated GBM cell lines grown on chamber slides. DAPI (46-diamidino-2-phenyl indole) was used for staining of DNA. For control purposes, primary antibodies were either omitted (control panels) or replaced by isotype-specific unrelated control antibodies (data not shown).

Figure 2

Effects of fibroblast growth factor 5 (FGF5) on growth and survival of human glioblastoma (GBM) cells. (a) Mitogenic effects of FGF5 were determined by treating the indicated GBM cell lines (after 48 h of serum starvation) for 72 h with increasing concentrations of rFGF5. DNA synthesis was measured by [³H]-thymidine incorporation assays and data are given relatively to serum-starved controls. (b) Cell death rescue effects of exogenous FGF5 were determined in MGC cells serum starved for 8 days (leading to around 50% cell death in the control group). The impact of increasing concentrations of rFGF5 on cell survival was measured by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay and data are given relatively to the control group without growth factor addition. (c) Downmodulation of endogenous FGF5 expression was performed using siRNA oligonucleotides as described in ‘Materials and methods’ and efficacy was determined after 72 h by western blot (left panel) and real-time PCR (right panel). Data for LN140 cells are shown representatively. (d) The indicated GBM cell lines were transfected with FGF5 or control siRNA, cultured for 72 h and analysed by MTT assay. Data are given relative to uninfected controls. Experiments were performed at least three times in triplicates. Significant differences are indicated as follows: Student’s t-test, *P<0.05; **P<0.01

Figure 3

Effects of fibroblast growth factor 5 (FGF5) on the migratory potential of human glioblastoma (GBM) cells. (a) GBM cell migration in culture medium containing 1% fetal calf serum (FCS) without and with rFGF5 at the indicated concentrations was studied by transwell chamber assays. After a migration period of 12 h, cells at the lower side of the membranes and in the lower chambers were fixed and counted microscopically. Data are given relative to the 1% FCS control without FGF5 set as 1. At least two experiments were performed delivering comparable results. (b) Migration ability of GBM cells treated with FGF5 or control siRNA was determined as described under (a). Data are shown relative to uninfected controls. Three experiments were performed and significant differences are indicated as follows: Student’s t-test, **P<0.01.

Figure 4

Blockade of FGFR1-mediated signal transduction in glioblastoma (GBM) cells. (a) Effects of a dnFGFR1 IIIc/GFP adenovirus on the survival of GBM cell lines were evaluated microscopically. Phase-contrast and fluorescence photomicrographs of U373 cells infected with the dnFGFR1 IIIc/GFP as compared to the green fluorescent protein (GFP) control adenovirus are shown representatively (upper panel). The number of dead cells induced by infection with the indicated adenoviruses was determined after 48 h by Trypan blue exclusion assay (lower panel). (b) The impact of infection with the indicated adenoviruses on p53 expression and caspase substrate cleavage (poly(ADP-ribosyl)polymerase (PARP), caspase 7) is representatively shown for U373 cells. Proteins were isolated 48 h after infection. (c) Effects of a 72 h exposure to the specific FGFR1 inhibitor PD166866 (left panel) and the general FGFR inhibitor SU5402 (right panel) on viability of GBM cell lines were determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay performed at least three times in triplicates. (d) The impact of PD166866 on phosphorylation of extracellular signal-regulated kinase (ERK), S6, and STAT3 in serum-starved MGC cells stimulated with recombinant fibroblast growth factor 5 (rFGF5, 30 min) was determined by western blot.

Figure 5

Impact of recombinant and glioblastoma (GBM) cell-derived fibroblast growth factor 5 (FGF5) on human umbilical vein endothelial cells (HUVEC). (a) Effects of rFGF5 (left panel) as compared to rVEGF (middle panel) at the indicated concentrations and the respective neutralizing antibodies (Abs) on HUVEC growth were measured by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Additionally, HUVEC were exposed to culture supernatants (SN) of highly FGF5-positive MGC cells with and without antibody for FGF5 and vascular endothelial growth factor (VEGF, right panel). (b) The effects of FGF5, VEGF and MGC-conditioned medium (24 h exposure) with and without blocking antibodies on HUVEC migration were studied by micropore transwell chamber assays. To investigate a possible cross-reactivity of the neutralizing antibody, rFGF5 was combined with VEGF antibody. Cells at the lower side of the membranes were counted and data are given relative to the control (1% fetal calf serum, FCS). (c) HUVEC were suspended in matrigel and tube formation was induced with rFGF5 as well as with SN of MGC cells at the indicated dilutions with and without FGF5 blocking antibody. Tube formation was quantified as described in ‘Materials and methods’ and means±s.d. of at least three experiments are given. Significant differences to controls are indicated as follows: Student’s t-test, *P<0.05; **P<0.01; ***P<0.0001. In case of the antibody-containing groups ( + Ab) in panels a–c, the statistical comparison was performed to the respective groups without antibody.