Skip Navigation


Carcinogenesis Advance Access originally published online on September 22, 2007
Carcinogenesis 2008 29(1):15-24; doi:10.1093/carcin/bgm202
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
29/1/15    most recent
bgm202v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Sonvilla, G.
Right arrow Articles by Grusch, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sonvilla, G.
Right arrow Articles by Grusch, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
The online version of this article has been published under an open access model. Users are entitled to use, reproduce, disseminate, or display the open access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and Oxford University Press are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact journals.permissions@oxfordjournals.org

FGF18 in colorectal tumour cells: autocrine and paracrine effects

Gudrun Sonvilla, Sigrid Allerstorfer, Stefan Stättner2, Josef Karner2, Martin Klimpfinger3, Hendrik Fischer, Bettina Grasl-Kraupp, Klaus Holzmann, Walter Berger, Friedrich Wrba1, Brigitte Marian* and Michael Grusch

Department of Medicine I, Institute of Cancer Research
1 Institute of Clinical Pathology, Medical University Vienna, Vienna, Austria
2 Department of Surgery, Social Medical Center South, Vienna 1100, Austria
3 Department of Pathology, Social Medical Center South, Vienna 1100, Austria

* To whom correspondence should be addressed. Tel: +43 1 4277 65241; Fax: +43 1 4277 9651; Email: brigitte.marian{at}meduniwien.ac.at


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Funding
 References
 
Fibroblast growth factors (FGFs) and their high-affinity receptors contribute to the autocrine growth stimulation in several human malignancies. Here, we describe that FGF18 expression is up-regulated in 34/38 colorectal tumours and is progressively enhanced during colon carcinogenesis reaching very high levels in carcinoma. Moreover, our data suggest that FGF18 affects both tumour cells and tumour microenvironment in a pro-tumorigenic and pro-metastatic way. Addition of recombinant FGF18 to the culture media of slowly growing colorectal tumour cell lines LT97 and Caco-2 stimulated proliferation. Phosphorylation of externally regulated kinase 1/2 and S6 was increased already 5 min after growth factor addition. SW480 cells, endogenously producing large amounts of FGF18, were not affected in this setting, but recombinant FGF18 supported tumour cell survival under conditions of serum starvation. Down-modulation of endogenous FGF18 production by small interference RNA (siRNA) significantly reduced clonogenicity of SW480 cells and restored sensitivity to exogenous FGF18. With respect to the tumour microenvironment, both recombinant and tumour-derived FGF18 stimulated growth of colon-associated fibroblasts at 0.1 ng/ml and migration at 10 ng/ml. In addition, recombinant FGF18 (10 ng/ml) induced tube formation in human umbilical vein endothelial cells. siRNA knock down demonstrated that tube-forming activity of colon cancer cell supernatants depended to a large part on tumour cell-derived FGF18. In summary, this study demonstrates that FGF18 is almost generally over-expressed in colon cancer and exerts pro-tumorigenic effects both in the epithelial and the stromal compartments by stimulating growth and survival of tumour cells, migration of fibroblasts and neovascularization. Together, these data strongly support an oncogenic role of FGF18 in colorectal cancer.

Abbreviations: ERK, externally regulated kinase; FCS, fetal calf serum; FGF, fibroblast growth factor; GAPDH, glyceraldehyde phosphate dehydrogenase; HUVEC, human umbilical vein endothelial cell; IGF, insulin-like growth factor; MAPK, mitogen-activated protein kinase; mRNA, messenger RNA; RT–PCR, reverse transcription–polymerase chain reaction; siRNA, small interference RNA; TGF{alpha}, transforming growth factor alpha


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Funding
 References
 
Colorectal cancer develops through a series of pre-malignant (adenomatous) lesions. Molecular analysis has shown the accumulation of tumour-specific mutations during this process along with, and responsible for, progressive de-regulation of growth (1,2). A prominent mechanism of de-regulation is the establishment of autocrine growth stimulation by transforming growth factor alpha (TGF{alpha}) that is induced at the adenoma stage and increases with tumour progression (3–5). Recent studies suggest that also members of the fibroblast growth factor (FGF) family are de-regulated in colon carcinogenesis (6–8).

FGFs are a family of 22 peptide growth factors with essential functions in normal cell growth, morphogenesis, tissue repair and angiogenesis (9). In many tumour types, however, FGFs and their tyrosine kinase receptors are inappropriately expressed (10–12). Like other FGFs, FGF18 is an essential mitogen in embryonic limb development and its loss causes severe skeletal defects (13,14). It has been shown that FGF18 plays a crucial role in the development of bone and cartilage, but over-expression in tumours has also been demonstrated (6,15,16). FGF18 expression is up-regulated through the constitutive activation of the Wnt pathway observed in most colorectal carcinomas (6) and, together with FGF16, FGF20 and the FGF signal regulator sprouty 4, it represents a β-catenin target gene (17). FGF18 stimulates growth in NIH3T3 cells, osteoblasts, chondrocytes and glial cells (6,18–20). As a secreted protein, FGF18 can thus affect both the tumour and the connective tissue cells of the tumour microenvironment. To assess the pathophysiological impact of FGF18 in colorectal cancer, we have performed a comprehensive study of gene expression in tissue specimens and cell cultures of normal, pre-malignant and malignant colons. In addition, we have characterized autocrine and paracrine effects of FGF18 on colorectal cancer cells, fibroblast and endothelial cells and suggest an oncogenic function of this FGF family member in human colon cancer.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Funding
 References
 
Tissue specimen
Tissue specimens of colorectal carcinoma and normal mucosa were obtained from patients undergoing surgery for colorectal cancer. Informed consent was obtained from all the patients. Immediately after surgery, tissue specimens were frozen in liquid N2 until the extraction of RNA.

Immunohistochemistry
Formalin-fixed and paraffin-embedded tissue specimens of adenomas, carcinomas and adjacent normal tissue from the resection margin were used to prepare serial sections from the same tissue block (1–3 µm). They were deparaffinized, rehydrated and stained according to the standard methods with a polyclonal goat anti-human FGF18 antibody diluted 1:50 (FGF18 C-16, sc-16830, Santa Cruz, Biotechnology, Santa Cruz, CA) and a biotinylated polyclonal rabbit anti-goat second antibody diluted 1:400 (E 0466, Dakocytomation, Glostrup, Denmark).

Cell lines
SW480, SW620, T84 and Caco-2 colon carcinoma cell lines were obtained from the American Type Culture Collection (Manassas, VA). The cell lines were kept under standard tissue culture conditions using minimal essential medium containing 10% fetal calf serum (FCS). The LT97 colon adenoma cell line has been established in our laboratory (21). This cell line was kept under standard tissue culture conditions using Ham F-12 medium containing 20% L-15 medium, 2% FCS, 10 µg/ml insulin, 2 x 10–10 M triiodotyronine, 2 µg/ml transferrin, 1 µg/ml hydrocortisone, 5 x 10–9 M Na-selenite and 30 ng/ml epidermal growth factor. WI38, a normal embryonic lung fibroblast cell line, and F331, a colon-associated fibroblast cell line, were kept in Dulbecco's minimal essential medium containing 10% FCS. Human umbilical vein endothelial cells (HUVECs) were provided by DI.T.Mohr (Institute of Cancer Research, Vienna) and cultured in EBM-2 MV medium (Cambrex, East Rutherford, NJ) containing 10% FCS, EGM-2MV SingleQuots (Cambrex) and 60 µg/ml endothelial cell growth supplement. Serum starvation experiments in these cells were done in M199 medium (PromoCell, Heidelberg, Germany) with 1% FCS and 50 U heparin/ml.

Isolation of RNA and cDNA synthesis
Total RNA was isolated from sub-confluent cultures or frozen colon tissue specimens using Trifast reagent according to the instructions of the manufacturer (PeqLab, Erlangen, Germany). First, strand cDNA was synthesized using mouse Moloney leukaemia virus reverse transcriptase (Fermentas, Burlington, Ontario, Canada) and random hexamer primers (GE Healthcare, Piscataway, NJ).

Quantitative reverse transcription–polymerase chain reaction analysis of FGF18 messenger RNA
The expression of FGF18 and glyceraldehyde phosphate dehydrogenase (GAPDH) messenger RNAs (mRNAs) was determined by quantitative real-time reverse transcription–polymerase chain reaction (RT–PCR) with Taqman assays from Applied Biosystems (Foster City, CA). Reactions were carried out on an ABI Prism 7000 system (Applied Biosystems) using the standard Taqman assay conditions. FGF18 gene expression in the tumours was calculated as x-fold change compared with the corresponding normal mucosa using GAPDH as a control gene for normalization. Selected RT–PCR experiments for the evaluation of FGF18 gene expression were done using cycles of 40 s denaturation at 94°C, 40 s annealing at 53°C and 40 s extension at 72°C. FGF18 primer sequences were sense 5'-ACTTGCCTGTGTTTACACTTCC-3' and antisense 5'-CCAGAACCTTCTCGATGAAC-3'. For GAPDH, primers were sense 5'-CGGGAAGCTTGTGATCAATGG-3' and antisense 5'-GGCAGTGATGGCATGGACTG-3'. Amplifications were done for 35 and 23 cycles in the case of FGF18 and GAPDH, respectively. PCR products were separated on 6% acrylamide gels. Bands were stained with ethidium bromide and quantified using a GelDoc system (Bio-Rad, Hercules, CA) and ImageQuant 5.0 (GE Healthcare) software.

Luciferase assay
To analyse β-catenin-dependent gene expression, we used reporter constructs that contained wild-type (TOP) or mutant (FOP) Tcf4-binding sites to determine the transactivational activity of endogenous β-catenin/Tcf4 (22). Higher ratios of these two reporter activities (TOP:FOP) indicated a higher β-catenin/Tcf4 activity. A total of 2 x 104 cells per well were seeded into 24-well plates, and after 48 h were co-transfected with 1 µg of TOP-plasmid DNA, respectively, FOP-plasmid DNA and 1 ng Renilla-plasmid DNA using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). After 24 h, cell lysates were obtained by using passive lysis buffer provided with the Dual Luciferase Assay (Promega, Madison, WI) and luciferase activity determined according to the manufacturer's instructions. Quantification was calculated using the activity of Renilla as an internal control.

Modulation of FGF18 expression
To achieve over-expression of FGF18, a FGF18 cDNA German Resource Center for Genome Research (RZPD, Berlin, Germany, IRAU p69B0732D6; NM 003862) was sub-cloned into an adenoviral construct expressing the gene under the control of a cytomegalovirus (CMV) promoter. SW480 and Caco-2 cells were infected with the virus at 1 and 20 MOI (multiplicity of infection), respectively, 24 h after plating. Growth assays of the infected cultures were initiated 48 h after infection and cell number determined another 48 h later.

For knock down of FGF18 expression, cells were seeded at a density of 2 x 105 cells per well into six-well plates in a medium containing 10% serum. After 24 h, they were transfected with 5 µl Lipofectamine 2000 (Invitrogen) and 100 pmol FGF18-small interference RNA (siRNA) (Ambion, Applied Biosystems) per well in a culture medium without serum. A scrambled siRNA without significant sequence homology to any known human gene served as negative control. After 48 h, total RNA was isolated, cDNA was synthesized and real-time RT–PCR was performed to measure the mRNA level of FGF18.

Cell cycle distribution and apoptosis induction determined by flow cytometry
Cell cycle distribution was determined as published (23). Early apoptosis detection was done by incubation with 10 µg/ml JC-1 (Sigma, St Louis, MO) followed by flow cytometry (23) on a fluorescence-activated cell sorter CALIBUR flow cytometer (Becton Dickinson, Sunnyvale, CA).

Protein isolation and western blotting
Protein was extracted from sub-confluent cultures using Radioimmunoprecipitation (RIPA) buffer (50 mM Tris–HCl pH 7.4, 500 mM NaCl, 1% Igepal CA-630, 0.5% Na-Desoxycholat, 0.1% sodium dodecyl sulphate and 0.05% NaN3) containing protease inhibitor cocktail 1:50 (Complete, Roche Diagnostics, Mannheim, Germany) and the phosphatase inhibitors sodium orthovanadate 1 mM and NaF 10 mM. Protein concentration of cell extracts was determined with the bicinchoninic acid (BCA) Protein Assay Kit (Pierce, Rockford, IL). In the case of secreted FGF18 detection, conditioned supernatants of sub-confluent SW480 cells were collected after 48 h in FCS-free medium. Purification was carried out by adding cell supernatant to 50 µl of pre-washed Heparin–Sepharose CL-6B (Pharmacia, Uppsala, Sweden). The slurry was incubated for 24 h at 4°C with rotation, washed thrice with 0.15 M NaCl, 0.01 M Tris–HCl, pH 7.5, followed by washing twice with 0.45 M NaCl, 0.02 mM Tris–HCl (24). Supernatant proteins were dissolved in 30 µl 4x Laemmli buffer/lane, whereas in the case of cell extracts, 30 µg of total protein/lane were separated on 12% polyacrylamide gels and transferred to a polyvinylfluoride membrane (Pall Corporation, East Hills, NY) by electroblotting. Western blots were done by using the following antibodies: phospho-S6 (#2215) and S6 1:1000 (#2212, Cell Signalling, Boston, MA), phospho-externally regulated kinase (ERK) 1/2 [sampler kit: phospho-p44/42 MAPK (mitogen-activated protein kinase) Thr202/Tyr204, Cell Signalling], ERK1/2 1:5000 (#06-182, Upstate, Lake Placid, NY), β-actin 1:5000 (AC-15, Sigma), goat anti-human FGF18 1:200 (FGF18 C-16, sc-16830, Santa Cruz), goat anti-human FGF2 1:500 (FGF basic, AB-233-NA, R&D Systems, Minneapolis, MN). Bound antibodies were detected by appropriate peroxidase-coupled secondary antibodies, and blots were developed using enhanced chemoluminescence (Pierce) followed by exposure to X-Omat film (Kodak, Stuttgart, Germany). Band intensities were quantified using ImageQuant 5.0 software.

Cell proliferation assays
Cells were seeded at a density of 3 x 103 cells per well into 96-well plates. After 24 h, the cells were serum starved for another 24 h and then stimulated with recombinant FGF18 (hFGF18-25, Strathmann Biotec, Bovenau, Germany) for at least 3 days. Viable cell number was determined with 3' (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assays (EZ4U; Biomedica, Vienna, Austria) and DNA synthesis by [3H]-thymidine incorporation as described (25).

Colony formation assay
SW480 cells were transfected with FGF18 or control siRNA as described above. After 48 h, transfectants were plated at a density of 100 cell/cm2 onto six-well plates in a medium containing 1% FCS. After 7 days, the cells were washed twice with phosphate-buffered saline and then fixed with methanol for 20 min. After two times washing with phosphate-buffered saline, they were stained with 0.01% of crystal violet solution to assess colony formation.

Cell migration assay
Confluent monolayers of WI38 and F331 cells were serum starved for 24 h and ‘scratch’ wounded using a yellow pipette tip. After rinsing with phosphate-buffered saline, 10 ng/ml of recombinant FGF18 were added. Control cells received no FGF18. Scratches were photographed at appropriate time points. For each time point, four measurements for each of two scratches were performed. For quantification and statistical analysis, the individual scratch width (micrometre, mean and standard deviation) was measured using the MetaMorph software (Meta Imaging Series 6.1, Universal Imaging Corporation Downington, PA).

Tube formation assay
HUVECs were serum starved for 48 h and then 1.5 x 104 cells were seeded in 75 µl starvation medium into 96-well plates coated with 50 µl growth factor-reduced BD Matrigel (Becton Dickinson, Franklin Lakes, NJ). Recombinant FGF18 was added (10–50 µg/ml) and the extent of tube formation was quantified after 6 h by digitizing tube images followed by measuring the tube length using ImageJ software (National Institutes of Health, Bethesda, MD).


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Funding
 References
 
FGF18 up-regulation during tumour progression in the colon
Total RNA was isolated from tissue specimen of 38 colorectal carcinomas and the respective normal mucosa and expression of FGF18 was determined by quantitative real-time RT–PCR relative to GAPDH. In the majority of the tumour specimen (34 of 38), growth factor expression was found to be up-regulated compared with the normal mucosa, whereas only three tumours had decreased FGF18 mRNA levels (Figure 1A). The average expression as compared with normal tissue was 23.9 ± 11.3-fold (range 0.1- to 155-fold) when calculated relative to GAPDH (Figure 1A). Use of either β2-microglobulin or β-actin as housekeeping genes for standardization did not significantly change the result. Within the carcinomas, no correlation between expression of FGF18 mRNA and staging could be found (data not shown). FGF18 protein was detected by immunohistochemistry in formalin-fixed tissue sections from six normal mucosa, six adenoma, six carcinoma and two liver metastases. FGF18 protein staining increased from absent in normal non-cancerous mucosa to low in adenomas with positive areas at the bottom of the crypts and further to strongly positive in malignant tumour including both primary and metastatic lesions. Typical examples are shown in Figure 1B.


Figure 1
View larger version (60K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Expression of FGF18 and Wnt-pathway activity. (A) Total RNA was extracted from 38 colorectal carcinomas and the corresponding normal mucosa and expression of FGF18 determined by real-time RT–PCR as described in Materials and Methods. Data are given as fold expression of FGF18 in tumour compared with normal tissue. (B) FGF18 was detected in formalin-fixed, paraffin-embedded tissue specimen by immunohistochemistry and representative examples of normal mucosa, adenoma, primary tumour and metastasis are shown. (C) Expression of FGF18 mRNA in the indicated colorectal tumour cell lines was determined by RT–PCR. PCR products were semi-quantified using ImageQuant software and normalized to GAPDH mRNA. Results are presented as means ± SDs of two experiments performed in duplicates (upper panel) and representative examples for FGF18 (380 bp) and GAPDH (360 bp) PCR products are shown (lower panel). (D) Wnt-pathway activity was determined as Tcf4-dependent promoter activity using Top/Fop and Renilla luciferase reporter constructs as described in Materials and Methods. The results represent means ± SDs of two experiments performed in triplicates. (E) Correlation of FGF18 mRNA levels and Wnt-pathway activity was determined by linear regression and was found to be highly significant at P = 0.0001 and r2 = 0.6959.

 
To clarify whether this increase is also reflected by in vitro models for colorectal carcinogenesis, RNA was isolated from five cell lines with increasing malignant potential—LT97, Caco-2, T84, SW480 and SW620. FGF18 expression analysis by RT–PCR (Figure 1C) and by quantitative real-time RT–PCR (data not shown) yielded comparable results. FGF18 expression was low in the adenoma cell line LT97 and in the well-differentiated, slowly growing carcinoma cell lines Caco-2 and T84, whereas it was highest in SW620 cells that have been isolated from a metastatic lesion.

As FGF18 is a known β-catenin target gene, activity of a luciferase-coupled Tcf-promotor was analysed in these cell lines (Figure 1D). Activity of this promoter was found to correlate with the level of FGF18 mRNA expression in a highly significant manner, with P = 0.0001 and r2 = 0.6959, calculated for linear regression (Figure 1E).

Growth stimulation by exogenous FGF18
To investigate the cellular effects of FGF18, cells were kept in serum-reduced culture medium (1% FCS, LT97 and Caco-2) or serum-free medium (SW480) for 2 days. Then recombinant FGF18 (1–10 ng/ml) was added and cell number was measured by 3' (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay 6 days later. FGF18 stimulated growth in the slowly growing, well-differentiated tumour cell lines LT97 and Caco-2 in a dose-dependent manner (Figure 2A). Growth stimulation ranged from 40 (LT97, 1 ng/ml FGF18) to 100% (Caco-2, 10 ng/ml FGF18) in these cell lines. The poorly differentiated cell line SW480, which expressed large amounts of endogenous FGF18, was not growth stimulated by exogenous FGF18 in this setting (Figure 2A). To study the effects of endogenous FGF18 production, Caco-2 and SW480 cells were infected with an adenovirus expressing FGF18 (FGF18 virus). Cell number was increased in Caco-2 cells infected with FGF18 virus compared with uninfected Caco-2 cells or Caco-2 cells infected with an adenovirus-expressing green fluorescence protein virus. Exogenous addition of FGF18 did not further increase the cell number of FGF18v-infected Caco-2 cells but increased the cell number of uninfected or green fluorescence protein virus infected Caco-2 cells to the level of FGF18v-infected cells. Cell number of SW480 cultures was not affected by either over-expression or exogenous addition of FGF18 (Figure 2B). Growth stimulation with time was analysed in LT97 and Caco-2 cell lines, which produced low endogenous FGF18 and found to be progressively increased with prolonged exposure time (Figure 2C and D). The cell line SW480 showed no growth factor-dependent increase of cell number even after 6 days (data not shown).


Figure 2
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Growth and survival effects of FGF18 in colon tumour cells. (A) Influence of increasing doses of recombinant FGF18 on the indicated tumour cell lines was measured by 3' (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay after 6 days exposure. Media contained no FCS for SW480 and 1% FCS for LT97 and Caco-2. Results are expressed as percentage of control cells (growth without FGF18) and represent means ± SDs of two experiments using quintuplicates. (B) Cell number was determined from Caco-2 and SW480 cells infected with FGF18 Virus (FGF18v), green fluorescence protein virus as well as uninfected cells, with or without the addition of recombinant FGF18 to medium without serum (sfr). Results are expressed as fold increase over uninfected control cells and represent means ± SDs of three to five cultures. (C and D) Time-dependent growth stimulation of the colon tumour cell models LT97 and Caco-2 by the indicated concentrations of FGF18 in a medium containing 1% FCS was determined by 3' (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. The results represent means ± SDs of two experiments in quintuplicates. (E) The impact of FGF18 (10 ng/ml) on the cell fractions in S phase of the cell cycle was determined under conditions of prolonged serum starvation in SW480 cells by fluorescence-activated cell sorter analysis. Results represent means ± SDs of two experiments performed in duplicates. (F) Influence of FGF18 on induction of mitochondrial depolarization as a measure of early apoptosis due to prolonged serum starvation was determined by fluorescence-activated cell sorter analysis of the mitochondrial membrane potential using JC-1 as described in Materials and Methods. Results represent means ± SDs of two experiments using triplicate cultures. * and #, significantly increased or decreased as compared with control at P ≤ 0.05, respectively.

 
Measuring DNA synthesis by [3H]-thymidine incorporation checked the impact of FGF18 on SW480 cells after prolonged serum starvation. Under these settings and also in the case of SW480 cells, DNA synthesis was increased by FGF18 up to ~50% (data not shown). Accordingly, cell cycle analysis of SW480 cells by fluorescence-activated cell sorter demonstrated that after prolonged starvation (5 days) and growth factor exposure (4 days), FGF18 significantly increased the fraction of cells in S phase (Figure 2E).

The consequences of FGF18 on apoptosis induction by complete serum starvation were measured by determination of the mitochondrial membrane potential using JC-1. After 5 days of serum starvation, apoptosis was induced in 13% of the cells, while 10 ng/ml FGF18 reduced this value to 8% (Figure 2F; P ≤ 0.05).

The data indicate that under low serum conditions colonic tumour cells require FGF18 for growth and survival and that it may come from exogenous or endogenous sources. Consequently, the FGF18 insensitivity in SW480 cells at early time points might be caused by the high production of endogenous growth factor. Thus, we have used siRNA to knock down FGF18 expression and checked for resensitization to FGF18. The two different siRNAs reduced FGF18 mRNA expression to 22 and 35%, respectively, whereas a scrambled siRNA had no effect on FGF18 expression (Figure 3A). These results were also reflected at the level of FGF18 secreted into the culture supernatant as determined by western blot (Figure 3B). Consequently, for all further experiments the most potent siRNA (si-FGF18-1) was employed. Overall, growth of SW480 under standard conditions was not affected in cultures transfected with si-FGF18, but sensitivity to growth stimulation by recombinant FGF18 under total FCS reduction was restored (Figure 3C). Scrambled siRNA had no effect on stimulation of growth by FGF18.


Figure 3
View larger version (36K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Consequences of FGF18 knock down in SW480 cells. (A) siRNA oligonucleotides complementary to FGF18 mRNA (si-FGF18) or scrambled siRNA (si-scrambled) were introduced into SW480 cells by lipofection and expression of FGF18 mRNA after 48 h was determined by real-time RT–PCR. Data are normalized to the scrambled siRNA control set as 1. (B) Effects of the siRNA molecules described under (A) on FGF18 secretion by SW480 cells were detected by western analysis (upper panel) and semi-quantified using ImageQuant software (lower panel). Lane 1, 100 ng recombinant FGF18 and lane 2, 100 ng recombinant FGF18 recollected from growth medium by heparin–sepharose. FGF18 secreted by SW480 cells within 48 h treated with scrambled siRNA (lane 3), si-FGF18-2 (lane 4) and si-FGF18-1 (lane 5) were prepared by heparin–sepharose affinity chromatography. (C) The impact of FGF18 knock down by the indicated siRNAs on growth stimulation by recombinant FGF18 was determined by 3' (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay after 48 h. Data are given normalized to the respective control groups without FGF18 set as 100%. (D) Impact of FGF18 knock down on colony-forming ability was determined by plating 100 transfectants/cm2 into six-well plates in a medium containing 1% serum. After 7 days, cells were fixed and stained with crystal violet (upper panel) and colonies counted (lower panel). Results are means ± SDs of three experiments in duplicates. * indicates a significant increase and # a decrease as compared with control at P ≤ 0.05 by Student's t-test.

 
The effect of endogenous FGF18 on colony formation was investigated in SW480 cells plated at low density (100 cells per cm2; Figure 3D, upper panel). Compared with scrambled siRNA, FGF18 siRNA resulted in a 30% decrease in the number of colonies formed (Figure 3D, lower panel), demonstrating autocrine growth stimulation by FGF18 under conditions of low-density plating.

Activation of FGF18 downstream pathways
To analyse the signalling pathways that mediate FGF18-induced cell proliferation, Caco-2 carcinoma cells were serum starved for 24 h and then stimulated with FGF18 10 ng/ml for 5–30 min before total protein was harvested. Phosphorylation of ERK and S6 was analysed by western blotting as a measure of the MAPK and the phosphoinositide-3 kinase pathways, respectively. ERK was strongly phosphorylated 5 min after growth factor addition (a 3-fold increase was obtained, P = 0.0068), but phosphorylation was lost within 30 min (Figure 4A). Weak phosphorylation of S6 was observed at 5 min of FGF18 exposure (Figure 4B). Similar results were obtained using LT97 adenoma cells (data not shown). Reflecting the lack of growth response in SW480 cells, no phosphorylation of signalling kinases could be observed (data not shown). However, FGF18 response in terms of downstream pathway activation was again restored when endogenous FGF18 was knocked down by si-FGF18. The resulting ERK stimulation was pronounced (55%, P = 0.003; Figure 4C), while phosphorylation of S6 was increased only by ~15% (P = 0.039; Figure 4D).


Figure 4
View larger version (52K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Stimulation of downstream signal pathways by FGF18. Activation of the MAPK and phosphoinositide-3 kinase/Akt pathways were determined by immunoblot analysis of phosphorylations of ERK. (A) and S6 (B) in Caco-2 cells serum starved for 24 h before addition of 10 ng/ml FGF18. (C and D) The impact of FGF18 knock down by the indicated siRNAs on the responsiveness of SW480 cells to FGF18-mediated downstream pathway activation was determined by analyzing the phosphorylation status of ERK (C) and S6 (D). The figures show representative gels and quantification of band intensities using ImageQuant software. Results are presented as means of two experiments using duplicates. *, significantly increased above control at P ≤ 0.05.

 
Paracrine effects of FGF18 on mesenchymal cells
To analyse possible paracrine effects of FGF18 on mesenchymal cells, fibroblasts obtained from the colon (F331) or the lung (WI38) were exposed to the growth factor. These cell lines express very little (WI38) or no (F331) FGF18 on mRNA level (data not shown). Both cell lines grew slowly with doubling times of 5 and 2 days, respectively. Addition of FGF18 at concentrations 0.05–1 ng/ml significantly stimulated cell growth in both cell lines, with WI38 being significantly more responsive as compared with F331 (Figure 5A).


Figure 5
View larger version (45K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Impact of FGF18 on growth and migration of fibroblasts. (A) Cultures of colonic F331 and lung WI38 fibroblasts were exposed to increasing concentrations of FGF18 diluted in a medium containing 0.2% FCS. Cell viability was determined by 3' (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. Results represent means ± SDs of three experiments using triplicates. (B–D) Impact of FGF18 on migration of colon F331 fibroblast (C) as well as lung WI38 fibroblast (D) cultures was determined by scratch assays. Control cultures received medium without FCS (control) or 10% FCS as indicated. Four positions in each scratch as representatively shown in (B) were measured at the indicated time points and migration distance was calculated from the differences of scratch width. The results are given in micrometre migrated distance and represent the means ± SDs of two respective experiments. (E) F331 cells were exposed to conditioned media obtained from Caco-2 and SW480 cells with and without substitution by recombinant FGF18. (F) FGF18 production in SW480 cells was knocked down by transfection with si-FGF18. Control cells received si-scrambled as indicated and migration determined by scratch assays. Conditioned media with or without recombinant FGF18 were used. * and #, significantly increased or decreased above control at P ≤ 0.05, respectively.

 
In addition, cell migration induced by FGF18 in fibroblast cultures was analysed by scratch tests (Figure 5B–D). F331 cells were characterized by a high-intrinsic migration potential so that scratches healed in <24 h (20% of the population doubling time). Serum starvation reduced the scratch closure capability from 100 to 50%. A total of 10 ng/ml of recombinant FGF18 restored migration, leading to ~80% scratch closure (Figure 5C). In WI38 cells, serum starvation completely abolished migration capability and treatment with FGF18 did not increase migration within 24 h. However, induction of migration by 10% FCS was observed (Figure 5D). Conditioned media were obtained from SW480 cells (high FGF18) and Caco-2 cells (low FGF18) and transferred onto F331 cultures. The SW480-conditioned medium caused migration to about twice the distance as compared with Caco-2-conditioned media reflecting the differences in FGF18 production. Substitution of the media with recombinant FGF18 enhanced activity in Caco-2 but not SW480 groups (Figure 5E). Knock down of FGF18 production in SW480 cells by siRNA decreased the migration-inducing activity of supernatant, demonstrating induction of migration by tumour-derived FGF18. Addition of FGF18 prevented this decrease (Figure 5F).

To analyse angiogenic activity of FGF18, HUVECs were investigated. Addition of FGF18 to HUVECs, which have a very low expression of endogenous FGF18 on mRNA level, did not significantly affect growth (data not shown). However, it stimulated tube formation in matrigel (Figure 6A), an indicator for angiogenic activity. A total of 10 ng/ml FGF18 increased tube length ~2.5-fold, which is comparable with the known angiogenic factor, vascular endothelial growth factor. A total of 50 ng/ml FGF18 did not increase this effect further (Figure 6B). Conditioned media obtained from Caco-2 cells induced a 1.5-fold increase in tube formation; SW480 and SW620 culture supernatants caused a doubling (Figure 6B). Addition of FGF18 to the conditioned media further increased induction of tubes by Caco-2 but not by SW480 supernatants.


Figure 6
View larger version (38K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6. Effect of FGF18 on tube formation by HUVECs. (A) HUVECs were suspended in 50 µl matrigel and tube formation was induced with 10 ng/ml FGF18. Respective photomicrographs are shown. (B) Tube formation of HUVECs (control) was stimulated by the indicated growth factors (solid bars) or conditioned media of the indicated colon carcinoma cell lines with or without FGF18. (C) Impact of FGF18 knock down by siRNA on tube formation capacity of SW480 supernatants with or without FGF18 was determined in comparison to scrambled siRNA. The extent of tube formation in all experiments was quantified by measuring tube lengths from digitized photomicrographs using ImageJ software. The results (B) and (C) represent means ± SDs of two experiments. * and #, significantly increased or decreased above control at P ≤ 0.05, respectively.

 
Media obtained from SW480 cultures with knocked down FGF18 had significantly reduced tube induction, which could also be restored by the addition of recombinant FGF18. This demonstrates angiogenic activity of colon cancer cell-derived FGF18 (Figure 6C).


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Funding
 References
 
FGF18 is a secreted heparin-binding polypeptide growth factor essential for normal embryogenesis and development. Although its function in normal skeletal development has been extensively studied (26), its role in malignant growth and progression is largely unexplored. FGF18 has been found up-regulated as a β-catenin target gene in colorectal tumours (6). Our study confirms these results using a larger number of tissue specimens and demonstrates progressive FGF18 over-expression during colon carcinogenesis from adenoma to carcinoma. In vitro cell models perfectly mirrored the clinical situation with highest FGF18 expression in carcinoma-derived cell lines. In the in vitro models, FGF18 expression correlated with the activity of a β-catenin-dependent promoter measured with Top/Fop reporter constructs (22). Levels of free β-catenin, as the major effectors of the Wnt pathway, are limited in normal cells by both ubiquitination and degradation via the proteasome and sequestration at the cell periphery due to interaction with E-cadherin (27). The hyperactivation of the Wnt pathway early in colon carcinogenesis is mainly based on the characteristic adenomatous polyposis coli (APC) mutation (28). Binding of β-catenin in a complex with APC, axin and glycogen synthase kinase 3β is essential for degradation of β-catenin. Accordingly, our adenoma cell model established from a patient suffering from familial adenomatous polyposis harbours a mutated APC gene (21). In addition, loss of E-cadherin (27,29) in poorly differentiated colorectal cancer cells may further contribute to pathway activation explaining the gradual increase of Wnt-pathway activity and hence also target gene products like FGF18 from adenoma to carcinoma.

In addition, our results demonstrate growth stimulatory effects of both recombinant FGF18 and endogenous over-expression of FGF18 in colorectal tumour cell lines with low endogenous production of the growth factor. This corresponds well to the proliferative effect of recombinant FGF18 on hepatic and intestinal epithelia of the mouse (18). In contrast, carcinoma-derived cell lines with high endogenous FGF18 production were insensitive to exogenous growth factor under standard cell culture conditions. Specific survival effects were, however, observed under prolonged serum starvation, where recombinant FGF18 normalized cell cycle distribution and prevented apoptosis induction. These observations suggest that carcinoma cells like SW480 support their own growth and survival by autocrine FGF18 production. Comparable findings have been reported with regard to other FGFs, for example in lung cancer (30). Insensitivity due to endogenous growth factor production is believed to be based on saturation of available receptor growth factor-binding sites on the cell surface and subsequent receptor internalization (31,32). To prove whether this concept also holds true in our models, we knocked down endogenous FGF18 production in SW480. Indeed, sensitivity to recombinant FGF18 was restored not only in terms of growth stimulation but also of downstream signal pathway activation.

This implies that colon carcinoma cell lines like SW480 utilize endogenous FGF18 in an autocrine manner. Accordingly, knock down of FGF18 by siRNA significantly reduced colony-forming capability of SW480 cells confirming a growth- and survival-promoting effect of endogenous FGF18 under more stringent culture conditions. Analysis of the intracellular signalling pathways demonstrated significant phosphorylation of ERK and S6 as indicators of the growth-promoting MAPK and the survival-promoting phosphoinositide-3 kinase/Akt pathways (33,34), respectively. Phosphorylation of S6 was relatively weak as compared with ERK, suggesting that FGF18 signalling is mainly mediated by the MAPK pathway. Together, these results indicate that FGF18 is capable of exerting autocrine stimulation in colorectal tumour cells enhancing both cell proliferation and survival. Until now, autocrine TGF{alpha} was the only growth factor that has been unequivocally shown to support colorectal tumour cell growth and survival (35,36). It is produced by the tumour in large amounts, and in adenoma tissue sections TGF{alpha} clearly correlates with the level of apoptosis (3). Inhibition of TGF{alpha} signalling induced cell loss by apoptosis in cell line models (5,21). In addition, for insulin-like growth factor (IGF) 1 and IGF2, over-expression has been observed in colorectal tumour cells (37,38). Both factors provide autocrine stimulation via the IGF1 receptor. Tumour cells also synthesize proteases capable of degrading IGF-binding proteins to release biologically active IGFs (39). With respect to the FGF system, β-catenin-mediated up-regulation of FGF-binding protein, a chaperon for locally stored FGFs enhancing their activities, was shown during colon carcinogenesis (40). This suggests that several pro-tumorigenic members of the FGF family, including FGF18, FGF20 and FGF-binding protein, are up-regulated in a coordinated manner based on hyperactivation of the Wnt-pathway (6). Up-regulation of non-Wnt target FGFs has not been observed in colorectal tumours. For FGF2 even down-modulation has been reported (8). Together, these findings indicate that FGF-mediated signals are substantial factors contributing to the tumorigenic functions of the Wnt pathway in colon cancer (6,7,40).

During embryogenesis and wound healing, FGFs play an important role in tissue interactions, stimulating connective tissue fibroblasts and endothelial cells (9,26). FGF18 mainly functions as a mesenchymal growth factor by stimulating fibroblasts and chondrocytes via the FGF receptor 3 variant IIIc (15,41,42). It is increasingly recognized that the stroma has distinct functions in tumour growth and spread (43). In colorectal tumours, myofibroblasts have been shown to accumulate at the invasion front where they provide pro-invasive signals (44) and produce an altered extracellular matrix (45) and the matrix degrading protease urokinase (46). Our work used colon-associated fibroblasts that have previously been shown to interact with normal colonic epithelial cells and provide them with survival signals (47,48). FGF18 produced by the tumour cells exerts paracrine effects on these fibroblasts stimulating both proliferation and cell migration, which is in agreement with a previous study in the mouse (18). Fibroblasts of lung origin (WI38), in contrast, were not induced to migrate, even though their growth response to FGF18 was much stronger. These data suggest organ-specific responsiveness of mesenchymal cells to tumour-derived growth factors (49,50). Tumour cell supernatants also stimulated migration of F331 fibroblasts and the differences between Caco-2 and SW480-derived conditioned media as well as knock-down experiments demonstrate that FGF18 contributes a considerable part of this activity and consequently participates in the reorganization of the tumour-specific tissue (49,51,52).

In addition to fibroblasts, HUVECs were used as a model for the endothelial compartment of the microenvironment. Recombinant FGF18 did not induce growth of HUVECs; however, it stimulated vessel tube formation comparably to vascular endothelial growth factor, a well-known vascularization factor in several tumour types including colon cancer (53). Accordingly, Antoine et al. (50) found enhanced migration but not growth of HUVECs in response to FGF18. The contribution of tumour-derived FGF18 in the conditioned medium from metastasizing colorectal tumour cells (SW620 and SW480) was shown by comparison with the media from Caco-2 cells, producing low FGF18 or cells transfected with FGF18 siRNA. The obtained data demonstrate the capability of secreted FGF18 to stimulate vascularization as a prerequisite of tumour progression (54). Together, these data indicate a role of FGF18 in neoangiogenesis. Thus, the paracrine effects of FGF18 can produce a condition reminiscent of wound healing and tissue repair—a process that has been shown to exert pro-tumour effects (43,55).

In summary, our results show that FGF18 is increasingly up-regulated during malignant progression in the colon. It affects both the epithelial cell compartment of colorectal tumours by stimulating tumour cell growth and the connective tissue compartment by promoting growth and migration of fibroblasts as well as vessel formation in vascular endothelial cells. This suggests that FGF18 exerts autocrine and paracrine oncogenic functions in colorectal cancers.


   Funding
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Funding
References
 
Austrian Science Fund project (P17630 [GenBank] -B12).


   Acknowledgments
 
Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Funding
 References
 

  1. Fearon ER, et al. Progressing toward a molecular description of colorectal cancer development. FASEB J. (1992) 6:2783–2790.[Abstract]
  2. Weinstein IB. The role of protein kinase C in growth control and the concept of carcinogenesis as a progressive disorder in signal transduction. Adv. Second Messenger Phosphoprotein Res. (1990) 24:307–316.[ISI][Medline]
  3. Habel O, et al. Tissue localization of TGFalpha and apoptosis are inversely related in colorectal tumors. Histochem. Cell Biol. (2002) 117:235–241.[CrossRef][ISI][Medline]
  4. Coffey RJ Jr, et al. Differential regulation of transforming growth factor alpha autoinduction in a nontransformed and transformed epithelial cell. Cell Growth Differ. (1992) 3:347–354.[Abstract]
  5. Partik G, et al. Inhibition of epidermal-growth-factor-receptor-dependent signalling by tyrphostins A25 and AG1478 blocks growth and induces apoptosis in colorectal tumor cells in vitro. J. Cancer Res. Clin. Oncol. (1999) 125:379–388.[CrossRef][ISI][Medline]
  6. Shimokawa T, et al. Involvement of the FGF18 gene in colorectal carcinogenesis, as a novel downstream target of the {beta}-catenin/T-cell factor complex. Cancer Res. (2003) 63:6116–6120.[Abstract/Free Full Text]
  7. Kirikoshi H, et al. Molecular cloning and characterization of human FGF-20 on chromosome 8p21.3-p22. Biochem. Biophys. Res. Commun. (2000) 274:337–343.[CrossRef][ISI][Medline]
  8. Mathonnet M, et al. Quantitative analysis using ELISA of vascular endothelial growth factor and basic fibroblast growth factor in human colorectal cancer, liver metastasis of colorectal cancer and hepatocellular carcinoma. World J. Gastroenterol. (2006) 12:3782–3783.[ISI][Medline]
  9. Ornitz DM, et al. Fibroblast growth factors. Genome Biol. (2001) 2. REVIEWS3005.
  10. Cappellen D, et al. Frequent activating mutations of FGFR3 in human bladder and cervix carcinomas. Nat. Genet. (1999) 23:18–20.[ISI][Medline]
  11. Ricol D, et al. Tumour suppressive properties of fibroblast growth factor receptor 2-IIIb in human bladder cancer. Oncogene (1999) 18:7234–7243.[CrossRef][ISI][Medline]
  12. Sakamoto H, et al. Transforming gene from human stomach cancers and a noncancerous portion of stomach mucosa. Proc. Natl Acad. Sci. USA (1986) 83:3997–4001.[Abstract/Free Full Text]
  13. Hajihosseini MK, et al. Expression patterns of fibroblast growth factors-18 and -20 in mouse embryos is suggestive of novel roles in calvarial and limb development. Mech. Dev. (2002) 113:79–83.[CrossRef][ISI][Medline]
  14. Martin G. Making a vertebrate limb: new players enter from the wings. Bioessays (2001) 23:865–868.[CrossRef][ISI][Medline]
  15. Davidson D, et al. Fibroblast growth factor (FGF) 18 signals through FGF receptor 3 to promote chondrogenesis. J. Biol. Chem. (2005) 280:20509–20515.[Abstract/Free Full Text]
  16. Ohbayashi N, et al. FGF18 is required for normal cell proliferation and differentiation during osteogenesis and chondrogenesis. Genes Dev. (2002) 16:870–879.[Abstract/Free Full Text]
  17. Katoh Y, et al. FGF signaling inhibitor, SPRY4, is evolutionarily conserved target of WNT signaling pathway in progenitor cells. Int. J. Mol. Med. (2006) 17:529–532.[ISI][Medline]
  18. Hu MC, et al. FGF-18, a novel member of the fibroblast growth factor family, stimulates hepatic and intestinal proliferation. Mol. Cell. Biol. (1998) 18:6063–6074.[Abstract/Free Full Text]
  19. Shimoaka T, et al. Regulation of osteoblast, chondrocyte, and osteoclast functions by fibroblast growth factor (FGF)-18 in comparison with FGF-2 and FGF-10. J. Biol. Chem. (2002) 277:7493–7500.[Abstract/Free Full Text]
  20. Hoshikawa M, et al. FGF-18 is a neuron-derived glial cell growth factor expressed in the rat brain during early postnatal development. Brain Res. Mol. Brain Res. (2002) 105:60–66.[Medline]
  21. Richter M, et al. Cells obtained from colorectal microadenomas mirror early premalignant growth patterns in vitro. Eur. J. Cancer (2002) 38:1937–1945.[CrossRef][ISI][Medline]
  22. Veeman MT, et al. Zebrafish prickle, a modulator of noncanonical Wnt/Fz signaling, regulates gastrulation movements. Curr. Biol. (2003) 13:680–685.[CrossRef][ISI][Medline]
  23. Hausott B, et al. Flavaglines: a group of efficient growth inhibitors block cell cycle progression and induce apoptosis in colorectal cancer cells. Int. J. Cancer (2004) 109:933–940.[CrossRef][ISI][Medline]
  24. Kornmann M, et al. Fibroblast growth factor-5 stimulates mitogenic signaling and is overexpressed in human pancreatic cancer: evidence for autocrine and paracrine actions. Oncogene (1997) 15:1417–1424.[CrossRef][ISI][Medline]
  25. Heffeter P, et al. Anticancer activity of the lanthanum compound [tris(1,10-phenanthroline)lanthanum(III)]trithiocyanate (KP772; FFC24). Biochem. Pharmacol. (2006) 71:426–440.[CrossRef][ISI][Medline]
  26. Haque T, et al. A review of FGF18: its expression, signaling pathways and possible functions during embryogenesis and post-natal development. Histol. Histopathol. (2007) 22:97–105.[ISI][Medline]
  27. Cavallaro U, et al. Cell adhesion and signalling by cadherins and Ig-CAMS in cancer. Nat. Rev. Cancer (2004) 4:118–132.[ISI][Medline]
  28. Polakis P. Wnt signaling and cancer. Genes Dev. (2000) 14:1837–1851.[Free Full Text]
  29. Pierce M, et al. Overexpression of the beta-catenin binding domain of cadherin selectively kills colorectal cancer cells. Int. J. Cancer (2003) 107:229–237.[CrossRef][ISI][Medline]
  30. Berger W, et al. Evidence for a role of FGF-2 and FGF receptors in the proliferation of non-small cell lung cancer cells. Int. J. Cancer (1999) 83:415–423.[CrossRef][ISI][Medline]
  31. Reilly JF, et al. Ligand dependent and independent internalization and nuclear translocation of fibroblast growth factor (FGF) receptor 1. DNA Cell Biol. (2004) 23:538–548.[CrossRef][ISI][Medline]
  32. Barbieri MA, et al. Receptor tyrosine kinase signaling and trafficking—paradigms revisited. Curr. Top. Microbiol. Immunol. (2004) 286:1–20.[ISI][Medline]
  33. Seger R, et al. The MAPK signaling cascade. FASEB J. (1995) 9:726–735.[Abstract]
  34. Testa JR, et al. AKT signaling in normal and malignant cells. Oncogene (2005) 24:7391–7393.[CrossRef][ISI][Medline]
  35. Barnard JA, et al. Auto- and cross-induction within the mammalian epidermal growth factor-related peptide family. J. Biol. Chem. (1994) 269:22817–22822.[Abstract/Free Full Text]
  36. Untawale S, et al. Transforming growth factor-alpha production and autoinduction in a colorectal carcinoma cell line (DiFi) with an amplified epidermal growth factor receptor gene. Cancer Res. (1993) 53:1630–1636.[Abstract/Free Full Text]
  37. Dignass AU, et al. Peptide growth factors in the intestine. Eur. J. Gastroenterol. Hepatol. (2001) 13:763–770.[CrossRef][ISI][Medline]
  38. Thompson MA, et al. Autocrine regulation of human tumor cell proliferation by insulin-like growth factor II: an in-vitro model. Endocrinology (1990) 126:3033–3042.[Abstract]
  39. Durai R, et al. The role of the insulin-like growth factor system in colorectal cancer: review of current knowledge. Int. J. Colorectal Dis. (2005) 20:203–220.[CrossRef][ISI][Medline]
  40. Ray R, et al. Up-regulation of fibroblast growth factor-binding protein, by beta-catenin during colon carcinogenesis. Cancer Res. (2003) 63:8085–8089.[Abstract/Free Full Text]
  41. Kapadia RM, et al. Glycogen synthase kinase 3 controls endochondral bone development: contribution of fibroblast growth factor 18. Dev. Biol. (2005) 285:496–507.[CrossRef][ISI][Medline]
  42. Ozasa A, et al. Complementary antagonistic actions between C-type natriuretic peptide and the MAPK pathway through FGFR-3 in ATDC5 cells. Bone (2005) 36:1056–1064.[Medline]
  43. van Kempen LCLT, et al. The tumor microenvironment: a critical determinant of neoplastic evolution. Eur. J. Cell Biol. (2003) 82:539–548.[CrossRef][ISI][Medline]
  44. De Wever O, et al. Tenascin-C and SF/HGF produced by myofibroblasts in vitro provide convergent pro-invasive signals to human colon cancer cells through RhoA and Rac. FASEB J. (2004) 18:1016–1018.[Abstract/Free Full Text]
  45. Martin M, et al. Role of stromal myofibroblasts infiltrating colon cancer in tumor invasion. Pathol. Res. Pract. (1996) 192:712–717.[ISI][Medline]
  46. Grondahl-Hansen J, et al. Localization of urokinase-type plasminogen activator in stromal cells in adenocarcinomas of the colon in humans. Am. J. Pathol. (1991) 138:111–117.[Abstract]
  47. Kalabis J, et al. Stimulation of human colonic epithelial cells by leukemia inhibitory factor is dependent on collagen-embedded fibroblasts in organotypic culture. FASEB J. (2003) 17:1115–1117.[Abstract/Free Full Text]
  48. Schorkhuber M, et al. Survival of normal colonic epithelial cells from both rats and humans is prolonged by coculture with rat embryo colonic fibroblasts. Cell Biol. Toxicol. (1998) 14:211–223.[CrossRef][ISI][Medline]
  49. Kataoka H, et al. Role of cancer cell-stroma interaction in invasive growth of cancer cells. Hum. Cell (2003) 16:1–14.[Medline]
  50. Antoine M, et al. Fibroblast growth factor 16 and 18 are expressed in human cardiovascular tissues and induce on endothelial cells migration but not proliferation. Biochem. Biophys. Res. Commun. (2006) 346:224–233.[CrossRef][ISI][Medline]
  51. Haass NK, et al. Adhesion, migration and communication in melanocytes and melanoma. Pigment Cell Res. (2005) 18:150–159.[CrossRef][ISI][Medline]
  52. DeClerck YA. Interactions between tumour cells and stromal cells and proteolytic modification of the extracellular matrix by metalloproteinases in cancer. Eur. J. Cancer (2000) 36:1258–1268.[CrossRef][ISI][Medline]
  53. Hanahan D, et al. Transgenic mouse models of tumour angiogenesis: the angiogenic switch, its molecular controls, and prospects for preclinical therapeutic models. Eur. J. Cancer (1996) 32A:2386–2393.[CrossRef]
  54. Steeg PS. Tumor metastasis: mechanistic insights and clinical challenges. Nat. Med. (2006) 12:895–904.[CrossRef][ISI][Medline]
  55. Marks F, et al. Mechanisms in tumor promotion: guidance for risk assessment and cancer chemoprevention. Toxicol. Lett. (1995) 82-83:907–917.[CrossRef]
Received May 16, 2007; revised August 10, 2007; accepted September 5, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
29/1/15    most recent
bgm202v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Sonvilla, G.
Right arrow Articles by Grusch, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sonvilla, G.
Right arrow Articles by Grusch, M.
Social Bookmarking
Add to CiteULike