Inhibition of human bladder tumour cell growth by fibroblast ... - Nature

2 downloads 124 Views 330KB Size Report
Oct 25, 2004 - Inhibition of human bladder tumour cell growth by fibroblast growth factor receptor 2b is independent of its kinase activity. Involvement of the.
Oncogene (2004) 23, 9201–9211

& 2004 Nature Publishing Group All rights reserved 0950-9232/04 $30.00 www.nature.com/onc

Inhibition of human bladder tumour cell growth by fibroblast growth factor receptor 2b is independent of its kinase activity. Involvement of the carboxy-terminal region of the receptor Isabelle Bernard-Pierrot1, David Ricol1, Andrew Cassidy2, Alexander Graham2, Paul Elvin2, Aure´lie Caillault1, Se´verine Lair1, Philippe Broe¨t3, Jean-Paul Thiery1 and Franc¸ois Radvanyi*,1 1 UMR144, CNRS Institut Curie, Section de Recherche, 26 rue d’Ulm, 75248 Paris Cedex, France; 2Cancer and Infection Research Area, AstraZeneca, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK; 3Hoˆpital Paul Brousse, 16 Avenue Paul Vaillant Couturier, 94807 Villejuif Cedex, France

The b isoform of fibroblast growth factor receptor 2, FGFR2b/FGFR2-IIIb/Ksam-IIC1/KGFR, a tyrosine kinase receptor, is expressed in a wide variety of epithelia and is downregulated in several human carcinomas including prostate, salivary and urothelial cell carcinomas. FGFR2b has been shown to inhibit growth in tumour cell lines derived from these carcinomas. Here, we investigated the molecular mechanisms underlying the inhibition of human urothelial carcinoma cell growth following FGFR2b expression. Using a nylon DNA array, we analysed the gene expression profile of the T24 bladder tumour cell line, transfected or not with a construct encoding FGFR2b. The expression of FGFR2b in T24 cells decreased insulin-like growth factor (IGF)-II mRNA levels. This decrease was correlated with a decrease in IGF-II secretion and may have been responsible for the observed inhibition of cell growth because the addition of exogenous IGF-II restored growth rates to normal levels. Using SU5402, an inhibitor of FGFR tyrosine kinase activity, and a kinase dead mutant of the receptor, FGFR2b Y659F/Y660F, we also demonstrated that the growth inhibition and decrease in IGF-II secretion induced by FGFR2b did not require tyrosine kinase activity. Finally, we demonstrated the involvement of the distal carboxy-terminal domain of the receptor in decreasing IGF-II expression and inhibiting T24 cell growth, as Ksam-IIC3, a variant of FGFR2b carrying a short carboxy-terminus, neither downregulated IGF-II nor inhibited cell proliferation. Our data suggest that FGFR2b inhibits the growth of bladder carcinoma cells by reducing IGF-II levels via its carboxy-terminal domain, independent of its tyrosine kinase activity. Oncogene (2004) 23, 9201–9211. doi:10.1038/sj.onc.1208150 Published online 25 October 2004 Keywords: bladder; fibroblast growth factor receptor; urothelial cell carcinoma; IGF-II; tumour suppressor gene; DNA array

*Correspondence: F Radvanyi; E-mail: [email protected] Received 20 February 2004; revised 28 July 2004; accepted 29 July 2004; published online 25 October 2004

Introduction Fibroblast growth factor receptors (FGFRs) are transmembrane tyrosine kinase receptors involved in signal transduction regulating cell growth, differentiation, migration, wound healing and angiogenesis, depending on target cell type and developmental stage. They are encoded by four different genes, FGFR1–4 (Chellaiah et al., 1994; McKeehan et al., 1998; Powers et al., 2000; Ornitz and Itoh, 2001), and among the 22 different fibroblast growth factor (FGF) genes identified, 18 have been shown to encode FGFR ligands (Basilico and Moscatelli, 1992; Ornitz and Itoh, 2001; Ware and Matthay, 2002; for reviews see Wilkie et al., 1995; Powers et al., 2000). FGFRs are glycoproteins with an extracellular domain composed of two or three immunoglobulin (Ig)-like domains, which mediates ligand binding and interaction with heparan sulphate proteoglycans, a hydrophobic transmembrane region and an intracellular region containing a juxtamembrane domain, a tyrosine kinase catalytic domain and a carboxyterminal domain with multiple tyrosine phosphorylation sites. The diversity of the FGFR family is increased by alternative mRNA splicing. In humans, the second half of the third Ig-like domain of FGFR1–3 is encoded by either exon IIIb or IIIc, generating receptors with different ligand-binding specificities and affinities. Hence, FGFR2b, the isoform containing the second half of the third Ig-like domain encoded by exon IIIb (also called FGFR2-IIIb, KGFR or Ksam-IIC1), binds FGF1, FGF3, FGF7/KGF and FGF10 with high affinity. In contrast, FGFR2c, the isoform containing the second half of the third Ig-like domain encoded by exon IIIc (FGFR2-IIIc/Bek), binds FGF1 and FGF2 but not FGF7/KGF or FGF10 (Basilico and Moscatelli, 1992; Finch et al., 1995; Bellusci et al., 1997; McKeehan et al., 1998; Zetter, 1998). FGFR2b expression is tightly restricted to epithelial cells, whereas FGFR2c is located primarily in the mesenchyme (Finch et al., 1989, 1995; Miki et al., 1992; Ornitz et al., 1996; Yamasaki et al., 1996). Other alternative splicing can generate isoforms of FGFR2b with altered carboxy-termini (Johnson et al., 1991; Miki et al., 1992; Yayon et al., 1992; Johnson and

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9202

Williams, 1993; Itoh et al., 1994). Herein, FGFR2b refers to the most common form of FGFR2b – the isoform with a C-terminus encoded by the C1 exon. The variant with a C-terminus encoded by the C3 exon is referred to as Ksam-IIC3 (Itoh et al., 1994). The Cterminus of Ksam-IIC3 is shorter than the C-terminus of FGFR2b and lacks the putative phospholipase c-g1 binding site, Tyr-769. Ksam-IIC3 is preferentially expressed in certain poorly differentiated gastric carcinomas (Itoh et al., 1994). In many cases, FGFRs have been implicated as oncogenes in tumorigenesis. Translocations involving the kinase domain of FGFR1 have been demonstrated in haematopoietic malignancies (Xiao et al., 1998), and the activation of FGFR3 by point mutations is a frequent event in bladder cancer (50% of cases) (Cappellen et al., 1999; Billerey et al., 2001). The activation of FGFR1 by overexpression has been suggested in a wide variety of cancers, including astrocytoma (Morrison et al., 1994; Yamaguchi et al., 1994), salivary adenocarcinoma (Myoken et al., 1996) and lung carcinoma (Volm et al., 1997; Berger et al., 1999). Splice forms with transforming activity, not expressed in normal tissues, have been reported in gastric cancer for FGFR2 and in pituitary tumours for FGFR4 (Itoh et al., 1994; Ishii et al., 1995; Lorenzi et al., 1997; Ezzat et al., 2002). In contrast to this role of FGFRs as positive regulators of tumorigenesis, FGFR2 has been reported to be downregulated in several cancers, suggesting that FGFR2 may also have tumour suppressor properties. The downregulation of FGFR2c is closely associated with malignant progression in astrocyte-derived tumours (Yamaguchi et al., 1994). The expression of FGFR2b is reduced in human salivary adenocarcinoma (Rubin et al., 1995), in prostate tumour epithelial cells (Feng et al., 1997; Matsubara et al., 1998, Naimi et al., 2002) and in transitional cell bladder carcinoma, in which this loss of FGFR2b expression is associated with a poor prognosis (Diez de Medina et al., 1997). Studies in vitro have confirmed that FGFR2b has tumour suppressor properties as this receptor inhibits the growth of prostate tumour cells (Feng et al., 1997, Matsubara et al., 1998), human urothelial carcinoma cells (Ricol et al., 1999) and human salivary adenocarcinoma cells (Zhang et al., 2001). FGFR2b inhibits the growth of human salivary adenocarcinoma cells by inducing differentiation and apoptosis. However, the detailed downstream signalling pathway responsible for the inhibition of FGFR2b-mediated tumour cell proliferation is unknown. T24 is a cell line derived from a human bladder tumour that does not express FGFR2b. We previously showed that FGFR2b inhibits T24 growth both in vitro and in vivo. Interestingly, the growth inhibitory effect was still observed in the presence of FGF7, an FGFR2b-specific ligand (Ricol et al., 1999). In this study, we investigated the molecular mechanisms underlying the inhibition of T24 growth by FGFR2b. We used nylon DNA macroarrays to compare gene expression profiles between untransfected T24 cells and T24 cells Oncogene

transfected with a construct encoding FGFR2b. This led to the identification of the insulin-like growth factor (IGF)-II gene as a gene downregulated by FGFR2b. The addition of exogenous IGF-II to the culture medium of T24 cells expressing FGFR2b restored the proliferation rate to normal levels and the addition of an anti-IGF-II blocking antibody to the culture medium of T24 cells inhibited growth to a similar level as that obtained following FGFR2b expression, demonstrating that the down-regulation of IGF-II was involved in growth inhibition induced by FGFR2b. We also characterized the domain of FGFR2b responsible for growth inhibition. Using an FGFR tyrosine kinase inhibitor and a kinase dead mutant of FGFR2b, FGFR2b Y659F/Y660F, we showed that the tyrosine activity of the receptor was not required for growth inhibition. Using Ksam-IIC3, a variant of FGFR2b carrying a short carboxy-terminus, we demonstrated the involvement of the distal carboxy-terminal portion of the receptor.

Results Decreased IGF-II expression in FGFR2b-expressing T24 cells In order to understand the molecular mechanisms of FGFR2b-induced inhibition of T24 human bladder carcinoma cell proliferation, we compared the gene expression profiles of one FGFR2b-expressing T24 clone (T24K10) with that of a control T24 clone (T24N4), using a nylon DNA array spotted with 4608 different cDNAs. Results were analysed using a Bayesian hierarchical model to identify the cDNAs that were significantly differentially expressed between the two cell lines (T24N4 and T24K10) (Broe¨t et al., 2002). Based on the results of the stochastic algorithms, we focused on a five-component mixture model, which was a good trade-off between obtaining a good fit and allowing sufficient flexibility for fine description of the distribution of the data across its whole range. Most of the analysed cDNAs belonged to component 3, corresponding to unregulated genes; 48 cDNAs corresponded to 40 genes that were significantly more expressed in T24K10 than in T24N4 (groups 1 and 2), including seven that were strongly upregulated (group 1); 40 cDNAs corresponded to 36 genes significantly less expressed in T24K10 than in T24N4 (groups 4 and 5), including seven that were strongly downregulated (group 5) (Table 1). As a useful check of the method, a list of 29 housekeeping genes was analysed. All these genes belonged to the unmodified component. As expected, the FGFR2 gene was strongly upregulated (group 1) in FGFR2b-expressing T24K10 cells and the various cDNAs corresponding to a given gene were generally in the same class. For example, the IL-13 and FGFR2 genes were represented by three and two different cDNAs, respectively, with all these cDNAs belonging to group 1. Some of the genes differentially expressed between T24K10 and T24N4 may be involved

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9203 Table 1 Genes significantly regulated following FGFR2b expression by T24K10 cells according to nylon macroarray analysis GenBank Acc. no.

Gene name

Z71929 Z71929 U70981 U70982 U70983 NM_003179 M15330 BC050674 AF217542 BC051899 U14588 M14221 AF307851 AB063322 D86550

Fibroblast growth factor receptor k-sam #2 Fibroblast growth factor receptor k-sam #2 IL-13R a chain Interleukin-13 receptor IL-13R a chain Homo sapiens synaptophysin IL-1b Homo sapiens hypothetical protein MGC 2655, mRNA Tissues factor pathway inhibitor Homo sapiens ras-related C3 botulinum toxin substrate 2 Human paxillin mRNA Human cathepsin B proteinase mRNA p53 Aminolevulinic acid (ALA1) synthetase Human mRNA for serine/threonine protein kinase cDNA clone IMAGE 148426 TGF-beta superfamily variant member containing morphogenetic protein Human mRNA for epiregulin Home sapiens prepro-major basic protein homolog mRNA Fibronectin Homo sapiens lysosmal pepstatin insensitive protease (CLN2) Homo sapiens mRNA for KIAA1723 protein Urokinase cDNA clone IMAGE 148105 Homo sapiens mRNA for zinc finger protein (ZAC gene) FGFR2 Human 5T4 gene for 5T4 oncofoetal antigen Human oligodendrocyte myelin glycoprotein Human cathepsin jB proteinase mRNA Tissue factor pathway inhibitor Human ferritin Heavy subunit mRNA Human interleukin 3 receptor Fibronectin Human mRNA for ZNF185 gene Fibronectin Homo sapiens mRNA similar to serine/threonine kinase 4 Fibronectin SPARC/osteonectin Tissue factor pathway inhibitor Human cathepsin B proteinase mRNA Homo sapiens hemoglobin, alpha 2 Human cathepsin B proteinase mRNA Human apoptosis inhibitor surviving gene Human ferritin heavy subunit mRNA Human KIP2 gene for Cdk-inhibitor p57KIP Homo sapiens major histocompatibility complex, class I, B Urokinase Fibronectin Homo sapiens transforming growth factor, beta 1 Homo sapiens cDNA FLJ13068 fis, clone NT2RP3001739 BRCA1 Cyclin G1 Human heparin-binding EGF-like growth factor Caveolin Secretory granule proteoglycan peptide Novel gene and novel gene fragment cloned in human neuroblastomas Thrombospondin-1 Human ribosomal protein S7 mRNA, 30 end Human ribosomal protein S7 mRNA, 30 end Human v-erba related era-3 gene Human sterol carrier protein X/sterol carrier protein 2 mRNA Human DNA sequence from clone RP 11-533E19 on chromosome 1 Homo sapiens acetyl-CoA carboxylase 1 (ACC1) mRNA Human ribosomal protein L21 mRNA Homo sapiens eukaryotic translation initiator factor 3 Human placental alkaline phosphatase type 3 mRNA Homo sapiens ERK activator kinase Homo sapiens proteoglycan 1, secretory granule Human mRNA for aminopeptidase N

BD237178 D30783 AF132209 X02761; K00055 AF039704 AB051510 BC013575 AJ311395 Z71929 Z29083 AC004526 M14221 AF217542 M12937 U70982 E01162 Y09538 E01162 BC005231 E01162 J03040 AF217542 M14221 BC050661 M14221 U75285 M12937 U22398 BC013187 BC013575 E01162 BC001180 AK023130 U14680 X77794 L17032 BT007143 BC015516 BD018438 X14787 BC002866 BC002866 X12795 M75883 AL353708 AY315627 U14967 BC001571 M14170 L11284 BC015516 M22324

Group

P-value

1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2

0.9301 0.9297 0.9235 0.9132 0.8968 0.8315 0.8007 0.6046 0.6002 0.8343 0.8342 0.8335 0.833 0.8297 0.8293 0.8264 0.8287

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

0.8215 0.8201 0.8191 0.8143 0.8124 0.8029 0.8004 0.8006 0.7918 0.791 0.7699 0.7538 0.7491 0.7413 0.732 0.7255 0.7242 0.6877 0.6537 0.6514 0.6339 0.6311 0.622 0.6145 0.6119 0.6107 0.5953 0.5512 0.5471 0.521 0.4899 0.7563 0.7559 0.7544 0.7536 0.7519 0.7518 0.7467 0.7464 0.746 0.7445 0.7431 0.742 0.7379 0.7332 0.7302 0.7174 0.7165 0.7091 0.7073 0.6973 0.6964 Oncogene

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9204 Table 1

(continued )

GenBank Acc. no.

Gene name

AK125220 AJ401609 BC052558 BD205155 X14787 BC003158 X77794 AK023362 D50374 AF168956 X56468 M15518 BC005929 BC51195 AF17226 AF17226 AF008551 AY228337

Connective tissue growth factor Homo sapiens mRNA for 30 50 cyclic nucleotide phosphodiesterase 1B2 Annexin II Human nucleic acid sequence originating in normal cystic tissue Thrombospondin Lysosomal-associated protein transmembrane 4 alpha Cyclin G Human rpS8 gene for ribosomal protein S8 Homo sapiens mRNA for calcium binding protein Homo sapiens amyloid precursor protein homolog HSD-2 mRNA Human mRNA for 14.3.3 protein, a protein kinase regulator Tissue plasminogen activator Homo sapiens proteoglycan 2, bone marrow Homo sapiens, clone IMAGE:5276612 Insulin-like growth factor-II IGF II (somatomedinA) Human aurora-related kinase 1 (ARK1) mRN Homo sapiens TNF receptor-associated factor 6 (TRAF6)

Group

P-value

4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5

0.6936 0.6901 0.689 0.6445 0.6435 0.6383 0.6357 0.623 0.6191 0.5292 0.5094 0.8973 0.7841 0.7049 0.6985 0.6906 0.6302 0.5864

Genes were ordered into five groups: groups 1 and 2, corresponding, respectively, to genes strongly or slightly significantly overexpressed in T24K10 cells; group 3 corresponding to genes unregulated following KGFR expression by T24 cells; and groups 4 and 5 corresponding, respectively, to genes slightly or strongly significantly downregulated in T24K10 cells. For each regulated gene, the group number is given and P-value represents the probability of this gene to belong to the group

in inhibition by FGFR2, given the functions of the proteins they encode. We therefore used reverse transcription–polymerase chain reaction (RT–PCR) to determine the mRNA levels of some of these genes (IL13 receptor, IL1b (group 1); survivin, epiregulin (group 2); CTGF, TGFa, HB-EGF (group 4); tissue plasminogen activator (tPA), IGF-II (group 5)) in the T24K10 and T24N4 cell lines, and in other FGFR2b-expressing T24 clones and control clones. Only the IGF-II gene was differentially expressed in all FGFR2b-expressing clones versus control clones. Indeed, although, as expected, all the other genes tested were differentially expressed between T24K10 and T24N4 cells, this differential expression was not observed when other FGFR2bexpressing clones and T24 control clones were compared (Figure 1a and b and data not shown). Furthermore, the downregulation of IGF-II mRNA induced by FGFR2b was correlated with a decrease in IGF-II protein secretion in the culture medium (Figure 1c). The decrease in IGF-II expression induced by FGFR2b is responsible for FGFR2b-expressing T24 cells growth inhibition We investigated whether the decrease in IGF-II expression in FGFR2b-expressing cells was responsible for the observed decrease in cell growth in vitro by studying the effects of a blocking anti-IGF-II antibody or recombinant IGF-II on the proliferation of FGFR2b-expressing T24 cells (T24K6, T24K8 and T24K10) and T24 control cells (T24N1 and T24N4) (Figure 2). We determined the proliferation rate of these cells by measuring [3H]thymidine incorporation after 4 days of culture in the presence of 15 mg/ml IGF-II monoclonal blocking antibody or 15 mg/ml mouse IgG, used as a control (Figure 2a). We investigated the mitogenic activity of exogenous IGF-II Oncogene

in these serum-starved cell lines by measuring [3H]thymidine incorporation after 18 h of stimulation with various concentrations of this growth factor (ranging from 1 to 20 ng/ml) (Figure 2b). The neutralizing antiIGF-II antibody reduced the proliferation rate of vector-transfected cells (T24N1 and T24N4) to the same rate as FGFR2b-transfected cells (T24K6, T24K8 and T24K10), whereas the antibody did not significantly inhibit the growth of FGFR2b-transfected cells. Furthermore, the addition of exogenous IGF-II to the culture medium led to an increase in the proliferation of FGFR2b-expressing T24 cells (T24K10 and T24K8). This increase in proliferation rate was dose-dependent until a plateau was reached from 10 ng/ml of IGF-II. The proliferation rate at the plateau corresponded to that of T24 control cells in the absence of exogenous IGF-II. In contrast, exogenous IGF-II inhibited [3H]thymidine incorporation by T24 control cells (clones T24N1 and T24N4) in a dose-dependent manner. This suggests that the endogenous level of IGF-II in T24 control cells corresponds to the plateau of stimulation of these cells by IGF-II and that, as observed for many growth factors, higher concentrations of IGF-II inhibit cell proliferation, giving a bell-shaped dose–response curve. Taken together, these results strongly suggested that the observed decrease in IGF-II expression in T24 cells following FGFR2b expression was at least partly responsible for the inhibition of cell growth by FGFR2b. The inhibition of T24 cell proliferation by FGFR2b is independent of the kinase activity of the receptor We investigated whether the growth inhibitory properties of FGFR2b were linked to its tyrosine kinase

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9205

Figure 2 Effect of a blocking anti-IGF-II antibody and of recombinant IGF-II on proliferation of T24 vector-transfected or FGFR2b-expressing cells. [3H]thymidine incorporation by serumstarved FGFR2b-expressing T24 clones and T24 control clones treated for 4 days with 10% FCS and 15 mg/ml anti-IGF-II antibody or 15 mg/ml mouse IgG (a) or for 24 h with various concentrations of recombinant IGF-II (b). (a, b) Results are the means of two independent experiments carried out in triplicate and the standard errors are indicated Figure 1 tPA and IGF-II expression by FGFR2b-expressing T24 clones. tPA (a) and IGF-II (b) mRNA levels in various FGFR2bexpressing T24 clones (T24K: T24 cells transfected with the pcDNA1 expression vector encoding FGFR2b under the control of the CMV promoter), T24 control clones (T24N: T24 cells transfected with the empty pcDNA1-Neo expression vector) and T24 parental cells (T24) were evaluated by real-time RT–PCR, as described in Materials and methods.(c) T24 control and FGFR2bexpressing cells were cultured for 72 h and IGF-II protein levels in the culture medium were determined by ELISA. (a–c) Results are the means of two independent experiments carried out in triplicate and the standard errors are indicated

activity by analysing the phosphorylation status of the receptor in FGFR2b-expressing T24 cells (T24K10). FGFR2b immunoprecipitation from a T24K10 lysate with an anti-FGFR2 antibody followed by immunoblot analysis with an anti-phosphotyrosine antibody (a-pY) showed that the receptor was not or was only very slightly phosphorylated in the absence of exogenous ligand (Figure 3a). As expected, tyrosine phosphorylation of the receptor was observed in T24K10 cells stimulated with FGF1 in the presence of heparin (Figure 3a) or FGF7 (data not shown). These data suggest that tyrosine phosphorylation of the receptor

was not required for inhibition of the growth of T24 cells by FGFR2. To confirm this result, we evaluated the growth inhibitory activity of FGFR2b in the presence of an inhibitor of FGFR2 tyrosine kinase activity. SU5402 is an ATP mimetic known to inhibit FGFR1 tyrosine phosphorylation by binding to the ATP-binding site. As the catalytic domains of FGFRs display considerable sequence and structure similarity, we thought it likely that SU5402 would also inhibit FGFR2b phosphorylation. We found that SU5402 did indeed inhibit FGF1induced FGFR2b phosphorylation (Figure 3a). We therefore evaluated the proliferation (Figure 3b) and anchorage-independent growth (data not shown) of FGFR2b-expressing clones in the presence of this inhibitor. The inhibition of proliferation induced by FGFR2b in T24K10 and T24K8 cells was not affected by 30 mM SU5402 (Figure 3b). Similarly, 30 mM SU5402 had no effect on the ability of cells to form colonies on soft agar, with T24K10 cells forming far fewer colonies than control T24N4 cells (data not shown). These results strongly reinforced the hypothesis that the growth inhibitory properties of FGFR2b were independent of Oncogene

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9206

the kinase activity of this receptor. However, we could not exclude the possibility that the inhibitor did not totally block FGFR2b phosphorylation. Hence, we tested the ability of a kinase dead mutant, FGFR2b Y659F/Y660F, to inhibit T24 cell growth (Figure 3d) and to decrease IGF-II secretion (Figure 3e). Mohammadi et al. (1996) demonstrated that the autophosphorylation of both tyrosines 653 and 654 is important for the activation of FGFR1 tyrosine kinase activity because the mutant receptor, FGFR1 Y653F/Y654F, was not phosphorylated when stimulated with FGF1 and failed to mediate FGF1 FGFR1-mediated biological response in L6-cells or PC12 cells. Hence, it was tempting to speculate that the corresponding tyrosine residues Y659 and Y660 were required for FGFR2b tyrosine kinase activity. We constructed a mutated receptor (FGFR2b Y659F/Y660F) and generated stable FGFR2b Y659F/Y660F T24 clones as described in the Materials and methods section. Three clones expressing FGFR2b Y659F/Y660F protein to a similar or slightly lower level than FGFR2b in the T24K6 and T24K10 clones were selected (clones KD2, KD3 and KD5) (data not shown). These cell lines and T24K10 cells were treated with FGF1, lysed and immunoprecipitated with anti-FGFR2 antibodies, followed by immunoblot analysis with an anti-phosphotyrosine antibody (a-pY) or with an anti-FGFR2 antibody (Figure 3c). As expected, the phosphorylation of the receptor induced by FGF1 was abolished in cells expressing FGFR2b Y659F/ Y660F. However, like FGFR2b, the kinase dead receptor still inhibited cell growth (Figure 3d) and decreased IGF-II secretion (Figure 3e), unambiguously demonstrating that the growth inhibitory properties of FGFR2b are independent of the kinase activity of this receptor. Figure 3 FGFR2b tumour suppressive activity is independent of kinase activity. (a) Total cellular lysate (1 mg) from FGFR2bexpressing cells (T24K10) were immunoprecipitated with an antiFGFR2 antibody and immunoblotted with an anti-phosphotyrosine antibody (a-PY, upper blot). The membrane was then stripped and immunoblotted with an anti-FGFR2 antibody (lower blot). Where indicated, cells were pretreated or not for 1 h with an FGFR tyrosine kinase inhibitor (SU5402 at 30 mM) then stimulated or not by incubation with 10 ng/ml FGF1 in the presence of 50 mg/ ml heparin for 5 min before harvesting. (b) Effect of SU5402 on FGFR2b tumour suppression activity. [3H]thymidine incorporation by serum-starved FGFR2b-expressing clones (T24K8, T24K10) and T24 control clones (T24N1, T24N4) stimulated with 10% FCS, with or without a 1 h pretreatment with 30 mM SU5402. (c) Anti-FGFR2 immunoprecipitates from FGF1-stimulated T24 cells expressing wild-type FGFR2b (T24K10) or the double mutant FGFR2b Y659F/Y660F (T24KD2, T24KD3, T24KD5) were analysed by immunoblotting with either an anti-phosphotyrosine antibody (a-PY, lower blot) or an anti-FGFR2 antibody (upper blot). (d) [3H]thymidine incorporation by serum-starved T24 control clones (T24N1, T24N4), FGFR2b Y659F/Y660F-expressing clones (T24KD2, T24KD3, T24KD5) and FGFR2b-expressing clones (T24K10) were stimulated with 10% FCS. (e) T24 control cells (clones T24N1 and T24N4), FGFR2b Y659F/Y660Fexpressing clones (clones T24KD2, T24KD3, T24KD5) or T24 cells expressing FGFR2b (clone T24K10) were cultured for 72 h and IGF-II levels in the culture medium were determined by ELISA assay. (b, d and e) Experiments were repeated twice and representative results are shown

Oncogene

The distal carboxy-terminal portion of FGFR2b is involved in growth inhibition As the tyrosine kinase activity of FGFR2b did not seem to be responsible for the growth inhibitory properties of this receptor, we tried to identify the structural determinant of the receptor required for growth inhibition. Two isoforms of the receptor, FGFR2b and KsamIIC3, which differ in the carboxy-terminal domain, have been described and have very different growth-activating properties (Ishii et al., 1995; Lorenzi et al., 1997; Sakaguchi et al., 1999). We therefore investigated the growth inhibitory properties of Ksam-IIC3, the isoform with the short carboxy-terminal domain (Figure 4a). Stable Ksam-IIC3 T24 clones were obtained as described in the Materials and methods section. Three clones expressing FGFR2 protein to a level similar to that in the T24K6 and T24K10 clones were selected (clones KC3-1, KC3-2, KC3-3) (data not shown). Immunofluorescence analysis showed that the receptor was distributed similarly in the clones transfected with the Ksam-IIC3 construct and the clones transfected with the FGFR2b construct (data not shown). Unlike the FGFR2bexpressing T24 clones, the Ksam-IIC3-expressing T24

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9207

Figure 5 IGF-II expression by T24 Ksam-IIC3-expressing cells. (a) IGF-II mRNA levels in T24 control clones (clones T24N1, T24N2 and T24N4), T24 FGFR2b-expressing cells (clone T24K6) and T24 Ksam-IIC3-expressing cells (clones T24KC3-1, KC3-2 and KC3-3) were evaluated by real-time RT–PCR, as described in the Materials and methods. (b) T24 control cells (clones T24N1 and T24N4), T24 FGFR2b-expressing cells (clone T24K6) and T24 Ksam-IIC3-expressing cells (clones T24KC3-1 and KC3-2) were cultured for 72 h and IGF-II protein levels in the culture medium were determined by ELISA assay. (a and b) Results are the means of two independent experiments and the standard errors are indicated Figure 4 Proliferation and soft agar growth of Ksam-IIC3expressing T24 cells. (a) Schematic representation of FGFR2b isoforms expressed by stably transfected T24 clones, showing the carboxy-terminal portions of Ksam-IIC3 and FGFR2b (also named Ksam-IIC1). (b) [3H]thymidine incorporation by serumstarved Ksam-IIC3-expressing clones (T24KC3-1, T24KC3-2) and T24 control clones (T24N1, T24N4) stimulated with 10% FCS. (c) Soft agar assay of T24KC3-1 cells and T24N4 control cells. After 3 weeks, colonies with diameters greater than 50 mm were scored positive. (b and c) Experiments were repeated twice and representative results are shown

clones displayed neither lower rates of proliferation (Figure 4b) nor lower levels of anchorage-independent growth (Figure 4c) than control T24 cells. Thus, KsamIIC3 displayed no growth inhibitory properties, indicating that the C-terminal portion of FGFR2b is required for its inhibitory function. To confirm that the FGFR2b-induced inhibition of T24 cell proliferation was linked to a decrease in IGF-II gene expression in these cells, we evaluated IGF-II mRNA and protein levels in Ksam-IIC3-expressing T24 cells. Consistent with its lack of growth inhibitory properties, Ksam-IIC3 did not decrease levels of IGF-II mRNA (Figure 5a) or protein (Figure 5b).

Discussion We previously reported an association between the downregulation of FGFR2b and a poor prognosis in bladder cancer (Diez de Medina et al., 1997). We have

also shown that FGFR2b inhibits growth in the T24 bladder tumour cell line (Ricol et al., 1999). We show here that this growth inhibition is a consequence of the downregulation of IGF-II induced by FGFR2b, and is dependent on the C-terminal part of the receptor but not on the kinase activity of the receptor. We also demonstrated that, in the T24 bladder tumour cell line, FGFR2b induces growth inhibition but not differentiation (as assessed by measuring various differentiation markers, uroplakins and keratins) or apoptosis (data not shown). Thus, depending on the type of cell, FGFR2b exerts various tumour suppressive properties. Indeed, in rat prostatic carcinoma cells, FGFR2b inhibits proliferation and induces differentiation; in salivary carcinoma cells (HSY cells), FGFR2b induces differentiation and apoptosis, and was recently shown to inhibit the growth of mouse prostatic carcinoma cells in vivo but not in vitro (Feng et al., 1997; Matsubara et al., 1998; Zhang et al., 2001; Freeman et al., 2003). It would therefore be interesting to investigate, in these models, whether the various tumour suppressor effects induced by FGFR2b are also due to the downregulation of IGFII, dependent on the C-terminal part of the receptor, and independent of receptor kinase activity, as in the bladder tumour cell line. Growth inhibition is probably independent of kinase activity in the mouse prostatic tumour model, as growth inhibition was observed in the absence of receptor activation due to dimerization (Freeman et al., 2003). However, in contrast to what is Oncogene

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9208

observed in T24 cells (Ricol et al., 1999), in the rat prostate tumour model and in HSY cells, growth inhibition is observed in the absence of ligand but is stronger if exogenous ligand is added (Matsubara et al., 1998, Zhang et al., 2001). In an attempt to account for tumour suppressor activity in the absence of ligand, it has been suggested that FGFR2b may heterodimerize with FGFR1, thereby inhibiting its activity (Feng et al., 1997; Ricol et al., 1999). However, our results, showing no inhibition of the proliferation of T24 control cells by SU5402, an inhibitor of FGFR1 tyrosine kinase activity, demonstrate that FGFR1 is not involved in T24 cell proliferation. This implies that FGFR2b does not exert its tumour suppression activity by inhibiting FGFR1 activity. A kinase-independent effect (apoptosis) has been reported for the tyrosine kinase receptor ret. However, this proapoptotic effect was inhibited by the presence of the ret ligand, GDNF, suggesting that this receptor behaves as a dependence receptor, like DCC (deleted in colorectal cancer; Mehlen et al., 1998; Bordeaux et al., 2000). However, in the T24 bladder tumour cell line, the growth inhibitory properties of FGFR2b were not affected by the presence of the FGFR2b ligand, FGF7. Our results, demonstrating that the Ksam-IIC3 receptor, which lacks 54 aa of the C-terminal domain, does not inhibit growth, demonstrate the importance of this domain for the growth inhibitory activity of FGFR2b and confirm the results of previous studies showing that the two receptors may display different biological roles in cell growth, cell transformation and differentiation (Ishii et al., 1995; Lorenzi et al., 1997). These studies showed that the short carboxy-terminal form of the receptor had stronger transformation activity. In this study, we demonstrated that the long carboxy-terminal form of FGFR2b had tumour suppressive activity not displayed by the short carboxyterminal form of the receptor. This domain contains the Tyr-769 residue, which is well conserved in other FGFRs and corresponds to Tyr-766 in FGFR1. In FGFR1, this residue, when phosphorylated, has been shown to bind to phospholipase C-g, resulting in the activation of phosphatidylinositol turnover (Mohammadi et al., 1991, 1992, Peters et al., 1992). We found that the phosphorylation of FGFR2b was not required for tumour suppressive activity, suggesting that this residue is not crucial for FGFR2b activity and that it is a nonphosphotyrosine-containing region within the Cterminal part of FGFR2b that plays the key role in the FGFR2b-induced inhibitory signal. In terms of treatment, the fact that the tumour suppressor activity of FGFR2b is independent of tyrosine kinase activity is of particular interest. Indeed, FGFR1 and FGFR3 have been shown to play oncogenic roles in several cancers and thus to constitute therapeutic targets. However, as the catalytic domains of FGFRs display high levels of sequence and structure similarity, it may be difficult to obtain tyrosine kinase inhibitors able to discriminate between FGFR2 and Oncogene

FGFR1 or FGFR3. This lack of specificity could have constituted a barrier for the use of FGFR1 or FGFR3 inhibitors because, in addition to inhibiting the oncogenic activity of these receptors, these inhibitors could also have inhibited the tumour suppressor activity of FGFR2b. However, this is not probably the case, as the FGFR2b tumour suppressor activity does not depend on tyrosine kinase activity. In conclusion, we demonstrate here that FGFR2b inhibited the growth of human bladder cancer cells by downregulating IGF-II. Surprisingly, this growth inhibition was independent of kinase activity but dependent on the C-terminal domain of the protein, suggesting that the observed inhibition resulted from a direct interaction of FGFR2b with another protein via this domain. The protein interacting with FGFR2b and the molecular mechanism underlying the inhibition of IGF-II mRNA synthesis following FGFR2b expression remain to be investigated. Materials and methods Cell culture, constructs and transfection T24 cells were cultured at 371C under an atmosphere containing 5% CO2 in DMEM/F12 (Dulbecco’s modified Eagle’s medium) supplemented with 2 mM glutamine, 100 U/ml penicillin, 100 mg/ml streptomycin, 10 mM HEPES, 50 nM hydrocortisone, 5 mg/ml apotransferrin and 5 nM sodium selenium (hereafter referred to as T24 medium) plus 10% heat-inactivated foetal calf serum (FCS). Transfected cells were cultured in this medium in the presence of 400 mg/ml G418. Some transfectants of T24 human urothelial carcinoma cells have been described in previous studies (Ricol et al., 1999). The transfectants expressing FGFR2b are T24K6, T24K8 and T24K10; those transfected with the pcDNAI-Neo plasmid (control clones) are T24N1, T24N2 and T24N4. To obtain T24 cells expressing the Ksam-IIC3 isoform and the mutant FGFR2b Y659F/Y660F, we transfected T24 cells (obtained from ATCC, Rockeville, MD, USA) by means of a liposomemediated method, using Lipofectamine (Invitrogen, Cergy Pontoise, France) according to the manufacturer’s protocol, respectively, with the pcDNAI-Neo-Ksam-IIC3 expression vector (Invitrogen, Cergy Pontoise, France) provided by Drs M Terada and T Yoshida (National Cancer Center Research Institute, Tokyo, Japan) (Itoh et al., 1994) and with the pcDNAI-Neo-FGFR2bY659/600F expression vector generated using QuickChange XL Site-Directed Mutagenesis Kit (Stratagene, Saint-Quentin en Yvelines, France) according to the manufacturer’s protocol using the oligonucleotides: 50 CAACAATATAGACTTTTTCAAAAAGACCACCAATG GGCGGCTTCC30 and 50 GGAAGCCGCCCATTGGTGG TCTTTTTGAAAAAGTCTATATTGTTG30 . Cells were selected for G418 resistance (400 mg/ml) 48 h after transfection. The medium was replaced every 2 days until colonies formed. Each G418-resistant clonal population was scraped off the medium with a pipette tip, resuspended and cultured in the selective medium. Clones were screened by Western blotting with a polyclonal anti-FGFR2 antibody (Santa-Cruz, Le Perray en Yvelines, France). Three clones with the highest levels of Ksam-IIC3 and FGFR2b Y659F/Y660F receptor expression were selected and named T24KC3-1, T24KC3-2 and T24KC3-3 for Ksam-IIC3, and T24KD2, T24KD3 and Y24KD5 for FGFR2b Y659F/Y660F.

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9209 Thymidine incorporation assay T24 cells (2  104) were seeded in 24-well plates in T24 medium supplemented with 10% FCS. Cells were incubated for 24 h and were then serum starved for 24 h. We then added 10% FCS or FGF1 and incubated the plates for a further 18 h. Alternatively, we added 10% FCS and 15 mg/ml anti-IGF-II antibody (Research diagnostics Inc., Flanders, NJ, USA) or 15 mg/ml mouse IgG for 4 days. We then added 1 mCi of [3H]thymidine (ICN, Orsay, France), incubated the plates for a further 6 h, fixed the cells in 10% trichloroacetic acid, washed them with water and lysed them with 0.1 N NaOH. Total incorporated radioactivity was counted with a micro-beta scintillation counter (LKB, Perkin-Elmer Life Sciences, Courtaboeuf, France). Soft agar assay We seeded 12-well plates containing agar and T24 medium supplemented with 10% FCS with or without 30 mM SU5402 (FGFR tyrosine kinase inhibitor) (Merck Eurolab, Fontenay Sous Bois, France) with 3  104 T24 cells per well. The inhibitor was added to the culture medium once per week. After 21 days, colonies with diameters greater than 50 mm were scored as positive, using a phase-contrast microscope equipped with a measuring grid. Immunoprecipitation and immunoblot analysis T24 cells expressing wild-type or kinase dead mutant FGFR2 cultured in selective medium were washed once with PBS and serum starved for 24 h. Untreated cells and cells pretreated for 1 h with 30 mM SU5402 were left unstimulated or were stimulated for 5 min with 10 ng/ml FGF1 in the presence of 50 mg/ml heparin at 371C. Cells were then lysed in 50 mM TrisHCl (pH 7.4), 150 mM NaCl, 1% NP-40, 0.25% sodium deoxycholate, 1 mM EDTA, 1 mM Na3VO4, 5 mM NaF, 1 mg/ ml aprotinin, 1 mg/ml leupeptin and 1 mg/ml pepstatin. Protein concentration was determined with the Biorad Bradford Protein Assay Kit (BioRad, Ivry sur Seine, France). Lysates (1 mg) were incubated with 5 mg/ml rabbit polyclonal antiFGFR2 antibody (Santa-Cruz) for immunoprecipitation. Immunocomplexes were purified by incubation with protein A-agarose beads. Immunoprecipitated proteins were resolved on a 7.5% polyacrylamide SDS–PAGE and electrotransferred to Immobilon-P membrane in 25 mM Tris, pH 8.3, 200 mM glycine, 10% ethanol. Nonspecific binding was prevented by incubating the membrane with 0.2% (v/v) Tween-20 in PBS (PBS-T) supplemented with 3% (w/v) bovine serum albumin (BSA) (Sigma, Saint-Quentin Fallavier, France) for 1 h at 371C. To detect phosphorylated FGFR2, the membrane was then incubated overnight at 41C with 1 mg/ml monoclonal anti-phosphotyrosine antibody coupled to horseradish peroxidase (RC20) (BD Transduction Laboratories, Saint-Quentin en Yvelines, France) diluted in PBS-T supplemented with 1% (w/v) BSA. The membrane was thoroughly washed with PBS-T and peroxidase activity was detected with the BM chemiluminescence reagent (Roche Mannheim, Meylan, France). Quantitative real-time RT–PCR with SYBR green detection Total RNA was obtained with the RNeasy mini-Kit (Qiagen, Courtaboeuf, France) according to the manufacturer’s instructions, including DNAse I treatment to prevent genomic DNA contamination. We used 1 mg of total RNA for reverse transcription, with random hexamer (20 pmol) and 200 U MMLV reverse transcriptase. The levels of mRNA for various

genes displaying differential expression in the T24K10 and T24N4 cell lines (insulin-like growth factor-II, tPA, IL-13 receptor, HB-EGF, TGF-b, epiregulin, IL-1) were determined by real-time PCR, using TBP (TATA-binding protein) as an internal standard. PCR was performed in an ABI PRISM 7700 real-time thermal cycler, using the SYBR Green kit (Applied Biosystems, Foster City, CA, USA). The sequences of the various primers used are available on request. The PCR conditions consisted of denaturation at 951C for 10 min, followed by 40 cycles of denaturation at 951C for 15 s and annealing/extension at 601C for 1 min. A dissociation curve was performed at the end of each PCR run, to check that a single product was amplified. A negative control without cDNA template and a negative control with a nonreverse transcribed RNA sample were included, in triplicate, in every assay. For each target gene and for the control gene (TBP), a standard curve was performed, using dilutions of cDNA obtained from T24 cells. The standard curves established for the target gene and the control gene were used to calculate the relative level of mRNA for the target gene. IGF-II ELISA T24 cells were cultured for 72 h in T24 medium supplemented with 10% FCS. IGF-II levels in the conditioned media were determined using the nonextraction IGF-II ELISA kit from Diagnostic System Laboratories (Cergy Pontoise, France) according to the manufacturer’s instructions. Nylon DNA array hybridization and analysis For each cell line (T24N4 and T24K10), total RNA was obtained with the RNeasy mini-kit (Qiagen) according to the manufacturer’s instructions, including DNAse I treatment to prevent genomic DNA contamination. We converted 10 mg of total RNA to 33P-labelled cDNA using [a-33P]dATP (Amersham), Superscript II reverse transcriptase (Life Technologies, Cergy Pontoise, France), random hexamer primer and the Klenow fragment of DNA polymerase I. We hybridized 25 mg/ ml of radiolabelled cDNA probe for 16 h at 621C, in parallel, with eight nylon microarrays containing 4608 spotted cDNA clones (four microarrays with labelled extracts from T24N4 and four microarrays with labelled extracts from T24NK10). We minimized variation by using nylon microarrays from the same batch (printed the same day on the same batch of nylon membrane with the same PCR products). Furthermore, the same investigator labelled the cDNA, and carried out the prehybridization, hybridization and washing procedures for both cell lines. The arrays were washed in stringent conditions and placed against a Kodak Phosphorimager screen. Hybridization signals were acquired with a Storm 860 Phosphorimager apparatus (Molecular Dynamics, Sunnyvale, CA, USA). We quantified the intensity of the radioactive signal on scanned images (indicating, for each gene, the abundance of the corresponding mRNA) without background subtraction, using ArrayVision software (Imaging Research Inc., St Catharines, Ontario, Canada). For each reporter, we calculated an arithmetic mean (for duplicate spots). Spots of buffer only (no sequence), referred to here as ‘blanks’, were excluded from analyses. The data were log transformed, and an additive model was applied on data to extract the parameter of interest: gene–cell line interaction. From this model, a statistic analysis was performed to quantify the differential expression of gene–cell line interaction (between T24N4 and T24K10 cell lines). We then .calculated the posterior probabilities (for each gene) of belonging to the nonmodified group and applied a Oncogene

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9210 Bayesian hierarchical model to identify changes in gene expression (between the two cell lines). More details on the statistical procedures used are provided in the work of Broe¨t et al. (2002). Additional microarray information The description of this microarray study followed the Minimum Information About Microarray Experiment (MIAME) guidelines (Brazma et al.. 2001). The original data and detailed protocols for RNA isolation, amplification, labelling and hybridization are available from: http://microarrays.curie.fr/publications/oncologie_moleculaire/ (user/utilisateur: ibp2004, password/mot de passe: dataT24)

Abbreviations FGFR, fibroblast growth factor receptor; FGF, fibroblast growth factor; IGF, insulin-like growth factor; RT–PCR, reverse transcription–polymerase chain reaction; FCS, foetal calf serum. Acknowledgements We would like to thank Dr Yutaka Hattori, Dr Masaaki Terada and Dr Teruhiko Yoshida for providing the KsamIIC1/FGFR2b and Ksam-IIC3 plasmids, and Dr Sylvia Richardson for fruitful discussion. This work was supported by the CNRS, the Bio-inge´nierie 2001 programme from the French Ministry of Education and Research (Biogen74 project) and the Ligue Contre le Cancer (laboratoire associe´).

References Basilico C and Moscatelli D. (1992). Adv. Cancer Res., 59, 115–165. Bellusci S, Grindley J, Emoto H, Itoh N and Hogan BL. (1997). Development, 23, 4867–4878. Berger W, Setinek U, Mohr T, Kindas-Mugge I, Vetterlein M, Dekan G, Eckersberger F, Caldas C and Micksche M. (1999). Int. J. Cancer, 3, 415–423. Billerey C, Chopin D, Aubriot-Lorton MH, Ricol D, Gil Diez de Medina S, Van Rhijn B, Bralet MP, Lefrere-Belda MA, Lahaye JB, Abbou CC, Bonaventure J, Zafrani ES, van der Kwast T, Thiery JP and Radvanyi F. (2001). Am. J. Pathol., 6, 1955–1959. Bordeaux MC, Forcet C, Granger L, Corset V, Bidaud C, Billaud M, Bredesen DE, Edery P and Mehlen P. (2000). EMBO J., 15, 4056–4063. Brazma A, Hingamp P, Quackenbush J, Sherlock G, Spellman P, Stoeckert C, Aach J, Ansorge W, Ball CA, Causton HC, Gaasterland T, Glenisson P, Holstege FC, Kim IF, Markowitz V, Matese JC, Parkinson H, Robinson A, Sarkans U, Schulze-Kremer S, Stewart J, Taylor R, Vilo J and Vingron M. (2001). Nat. Genet., 4, 365–371. Broe¨t P, Richardson S and Radvanyi F. (2002). J. Comput. Biol., 4, 671–683. Cappellen D, De Oliveira C, Ricol D, de Medina S, Bourdin J, Sastre-Garau X, Chopin D, Thiery JP and Radvanyi F. (1999). Nat. Genet., 23, 18–20. Chellaiah AT, McEwen DG, Werner S, Xu J and Ornitz DM. (1994). J. Biol. Chem., 15, 11620–11627. Diez de Medina SG, Chopin D, El Marjou A, Delouvee A, LaRochelle WJ, Hoznek A, Abbou C, Aaronson SA, Thiery JP and Radvanyi F. (1997). Oncogene, l4, 323–330. Ezzat S, Zheng L, Zhu XF, Wu GE and Asa SL. (2002). J. Clin. Invest., 1, 69–78. Feng S, Wang F, Matsubara A, Kan M and McKeehan WL. (1997). Cancer Res., 23, 5369–5378. Finch PW, Cunha GR, Rubin JS, Wong J and Ron D. (1995). Dev. Dyn., 2, 223–240. Finch PW, Rubin JS, Miki T, Ron D and Aaronson SA. (1989). Science, 4919, 752–755. Freeman KW, Gangula RD, Welm BE, Ozen M, Foster BA, Rosen JM, Ittmann M, Greenberg NM and Spencer DM. (2003). Cancer Res., 19, 6237–6243. Ishii H, Yoshida T, Oh H, Yoshida S and Terada M. (1995). Mol. Cell. Biol., 7, 3664–3671. Itoh H, Hattori Y, Sakamoto H, Ishii H, Kishi T, Sasaki H, Yoshida T, Koono M, Sugimura T and Terada M. (1994). Cancer Res., 12, 3237–3241. Johnson DE, Lu J, Chen H, Werner S and Williams LT. (1991). Mol. Cell. Biol., 9, 4627–4634. Oncogene

Johnson DE and Williams LT. (1993). Adv. Cancer Res., 60, 1–41. Lorenzi MV, Castagnino P, Chen Q, Chedid M and Miki T. (1997). Oncogene, 7, 817–826. Matsubara A, Kan M, Feng S and McKeehan WL. (1998). Cancer Res., 7, 1509–1514. McKeehan WL, Wang F and Kan M. (1998). Prog. Nucleic Acid Res. Mol. Biol., 59, 135–176. Mehlen P, Rabizadeh S, Snipas SJ, Assa-Munt N, Salvesen GS and Bredesen DE. (1998). Nature, 6704, 801–804. Miki T, Bottaro DP, Fleming TP, Smith CL, Burgess WH, Chan AM and Aaronson SA. (1992). Proc. Natl. Acad. Sci. USA, 1, 246–250. Mohammadi M, Dikic I, Sorokin A, Burgess WH, Jaye M and Schlessinger J. (1996). Mol. Cell. Biol., 16, 977–989. Mohammadi M, Dionne CA, Li W, Li N, Spivak T, Honegger AM, Jaye M and Schlessinger J. (1992). Nature, 6388, 681–684. Mohammadi M, Honegger AM, Rotin D, Fischer R, Bellot F, Li W, Dionne CA, Jaye M, Rubinstein M and Schlessinger J. (1991). Mol. Cell. Biol., 10, 5068–5078. Morrison RS, Yamaguchi F, Saya H, Bruner JM, Yahanda AM, Donehower LA and Berger M. (1994). J. Neurooncol., 3, 207–216. Myoken Y, Myoken Y, Okamoto T, Kan M, McKeehan WL, Sato JD and Takada K. (1996). Int. J. Cancer, 5, 650–657. Naimi B, Latil A, Fournier G, Mangin P, Cussenot O and Berthon P. (2002). Prostate, 3, 245–252. Ornitz DM and Itoh N. (2001). Genome Biol., 3, Reviews 3005.1–3005.12. Ornitz DM, Xu J, Colvin JS, McEwen DG, MacArthur CA, Coulier F, Gao G and Goldfarb M. (1996). J. Biol. Chem., 25, 15292–15297. Peters KG, Marie J, Wilson E, Ives HE, Escobedo J, Del Rosario M, Mirda D and Williams LT. (1992). Nature, 6388, 678–681. Powers CJ, McLeskey SW and Wellstein A. (2000). Endocr. Relat. Cancer, 7, 165–197. Ricol D, Cappellen D, El Marjou A, Gil-Diez-de-Medina S, Girault JM, Yoshida T, Ferry G, Tucker G, Poupon MF, Chopin D, Thiery JP and Radvanyi F. (1999). Oncogene, 51, 7234–7243. Rubin JS, Bottaro DP, Chedid M, Miki T, Ron D, Cheon G, Taylor WG, Fortney E, Sakata H, Finch PW and LaRochelle WJ. (1995). Cell Biol. Int., 5, 399–411. Sakaguchi K, Lorenzi MV, Matsushita H and Miki T. (1999). Oncogene, 40, 5497–5505. Volm M, Koomagi R, Mattern J and Stammler G. (1997). Eur. J. Cancer, 4, 691–693.

Mechanism of FGFR2b-induced growth inhibition I Bernard-Pierrot et al

9211 Ware LB and Matthay MA. (2002). Am. J. Physiol. Lung Cell. Mol. Physiol., 5, 924–940. Wilkie AO, Morriss-Kay GM, Jones EY and Heath JK. (1995). Curr. Biol., 5, 500–507. Xiao S, Nalabolu SR, Aster JC, Ma J, Abruzzo L, Jaffe ES, Stone R, Weissman SM, Hudson TJ and Fletcher JA. (1998). Nat. Genet., 1, 84–87. Yamaguchi F, Saya H, Bruner JM and Morrison RS. (1994). Proc. Natl. Acad. Sci. USA, 2, 484–488.

Yamasaki M, Miyake A, Tagashira S and Itoh N. (1996). J. Biol. Chem., 27, 15918–15921. Yayon A, Zimmer Y, Shen GH, Avivi A, Yarden Y and Givol D. (1992). EMBO J., 5, 1885–1890. Zetter BR. (1998). Annu. Rev. Med., 49, 407–424. Zhang Y, Wang H, Toratani S, Sato JD, Kan M, McKeehan WL and Okamoto T. (2001). Proc. Natl. Acad. Sci. USA, 20, 11336–11340.

Oncogene