• FGFR3b;
  • oral squamous cell carcinomas;
  • mutation


  1. Top of page
  2. Abstract
  3. Acknowledgements
  4. References

A G to T mutation at nucleotide position 2128 in the human FGFR3b coding region resulting in a Cys for Gly substitution (G697C) in the tyrosine kinase domain was observed in 62% (44/71) of oral squamous cell carcinomas (OSCC) examined. Immunostained FGFR3b was found in the cytoplasm of prickle cells in normal epithelia, and FGFR3b was localized in the cytoplasm and nucleus in non-FGFR3b mutant OSCC. Overexpressed FGFR3b protein on plasma membranes was noted in OSCC bearing the FGFR3b mutation. Enhanced tyrosine kinase activity of G697CFGFR3b was confirmed. Our results indicate that G697C is an activating mutation causing constitutive ligand-independent FGFR3b signaling. This mutation may be involved in the progression of OSCC and thus the FGFR3b coding sequence may have diagnostic or prognostic value for OSCC. © 2005 Wiley-Liss, Inc.

The fibroblast growth factor (FGF) family currently includes 22 different gene products that bind to 4 distinct FGF receptors (FGFR1-FGFR4).1, 2 The activities of FGF are mediated by high-affinity FGFR located on the cell surface, which share a common structure of 2 or 3 extracellular immunoglobulin-like loops, a transmembrane domain, and an intracellular split tyrosine kinase domain.3, 4 It is believed that FGFR, when activated by bound FGF in the presence of heparin or heparan sulfate proteoglycan, forms a dimer, which undergoes phosphorylation on tyrosine residues, and transduces signals of cell growth and differentiation.5, 6 It was shown recently that mutations in FGFR cause various congenital skeletal and chondral dysplasia.7 In these congenital diseases, constitutively active tyrosine kinase activity due to ligand-independent receptor dimerization was observed, and the degree of receptor tyrosine kinase activity correlated with the severity of disease. Squamous cell carcinoma (SCC) is a cancer originating from stratified squamous epithelia that mainly cover skin, oral cavity, esophagus, vagina and bronchus. We have reported previously that all 4 FGFR, including FGFR1 (c isoform), FGFR2 (b isoform), FGFR3 (b isoform) and FGFR4 were expressed exclusively in cells derived from normal oral epithelia and oral squamous cell carcinomas (OSCC).8 The growth of normal epithelial cells was stimulated by FGF whereas OSCC cell proliferation was not.9 Mutations in FGFR genes have been reported in human bladder, cervical, gastric and colorectal carcinomas,10, 11 however no such mutations have been reported in OSCC.

We examined the mutational status of FGFR3b in a series of human OSCC to determine whether FGFR3b is involved in oral squamous tumorigenesis. We found that 44 OSCC DNAs showed band shifts in exon 17 of the FGFR3b gene by SSCP analysis (Fig. 1a). The DNA sequence of exon 17 in each sample was investigated by direct nucleotide sequencing. A heterozygous missense mutation (G2128T) was observed in exon 17 in all cases (44/71) that exhibited a band shift in PCR-SSCP (Fig. 1b). This mutation resulted in a glycine-to-cysteine substitution at position 697 (G697C) in the second part of kinase domain of FGFR3b. Sequencing analysis of available blood genomic DNAs obtained from 12 OSCC patients with G2128T mutation showed that all of them exhibited normal sequence indicating G2128T mutation is somatic.

thumbnail image

Figure 1. Detection of mutation in exon 17 of FGFR3b gene. Sections were prepared from archival material from 71 OSCC cases. The specimens consisted of formalin-fixed, paraffin-embedded tissues from the files of the Department of Molecular Oral Medicine and Maxillofacial Surgery, Hiroshima University Dental Hospital. The Ethics Committee in Hiroshima University approved specimen collection. The tissues were trimmed to contain at least 50% tumor cells and then DNA was extracted using TakaRa DEXPAT Kit (TakaRa, Osaka, Japan) following the manufacturer's instruction. PCR amplifications of the FGFR3b intracellular region were carried out using primers described previously.12 PCR was carried out for 35 cycles (denaturation at 94°C for 30 sec, annealing at 55–68°C for 15 sec and extension at 72°C for 60 sec). Each exon was investigated for the presence of mutations by PCR-SSCP. In brief, 3.5 μl of PCR product was mixed with an equal volume of denaturation solution (95% formamide, 0.05% xylene cyanol, 0.04% bromophenol blue). After denaturing at 95°C for 5 min, the samples were loaded onto SSCP gel (GeneGel Excel 12.5/24 Kit, Amersham Pharmacia Biotech), and electrophoresis was conducted at 600 V, 25 mA at 15°C. After the electrophoresis, the gel was stained with a PlusOne DNA Silver Staining Kit (Amersham Pharmacia Biotech) to detect DNA bands. When a band exhibited a mobility shift in the SSCP analysis, the PCR product was sequenced in both directions using the dRhodamine Terminator Cycle Sequencing Ready Reaction (Applied Biosystems) on an ABI PRISM 310 Genetic Analyzer (Perkin-Elmer Cetus Instrument). (a) Representative PCR-SSCP analysis. PCR product from OSCC DNA (lane 2) was run alongside product from normal DNA (lane 1) isolated from non-neoplastic gingiva. Arrow indicates band with altered mobility. (b) Representative sequence histogram of exon 17 from OSCC specimen and normal control. Arrow indicates the G2128T mutation.

Download figure to PowerPoint

Mutations in several oncogenes and tumor suppressor genes have been found in SCC. According to previous reports, the frequencies of these mutations were 10% in H-ras,13 20% in MTS1,14 10% in ING115 and 55% in p53.16 The G2128T mutation of FGFR3b in OSCC was observed at a high frequency (62%), which strongly suggested that the mutation was closely related to the pathogenesis of OSCC.

It was reported that cysteine is an important amino acid for maintaining biological activity of proteins because it affects protein conformation through the formation of intra- and inter-molecular disulfide bonds.17 The substitution of glycine with cysteine might cause ligand-independent dimerization of FGFR molecules, which results in alterations in receptor function including increased levels of membrane-localized FGFR3b and enhanced autophosphorylation activity. In our present study, the expression of FGFR3b protein was observed in the cytoplasm of prickle cells in normal epithelial tissue (Fig. 2a). In the OSCC having no G2128T mutation, the expression of FGFR3b protein was observed in the cytoplasm and nucleus of the cancer cells (Fig. 2b). By contrast, in OSCC bearing the G2128T mutation strong expression of FGFR3b protein was observed on the cell membrane of cancer cells (Fig. 2c). The difference in localization of receptor proteins is thought to indicate differential receptor expression or differential receptor activity. It was reported that in COS cells FGFR2 and FGFR3 were localized in the cytoplasm and nucleus, whereas when the cells over-expressed wtFGFR1, wtFGFR2, wtFGFR3 and wtFGFR4, all 4 FGFR protein were detected on the cell surface.18 In OSCC bearing the G2128T mutation, the overexpressed G697CFGFR3b protein is similarly localized to the cell surface. In non-malignant cells FGFR molecules on the cell surface are difficult to detect immunohistochemically because the expression levels of these receptors is normally low.

thumbnail image

Figure 2. Immunostaining of FGFR3b protein in normal epithelia and OSCC tissue. Four μm formalin-fixed, paraffin-embedded tissue sections were deparaffinized, rehydrated and pretreated with 0.05% Proteinase K for 5 min at room temperature before anti-FGFR3 polyclonal antibody (Santa Cruz Biotechnology, Inc.) was applied overnight at 4°C. Immunostaining of FGFR3b was visualized by using the standard avidin-biotin peroxidase technique. For the evaluation of FGFR3b staining, 10 random 0.039-mm2 fields at 200× were selected, and assessed by 2 independent observers. (a) Expression of FGFR3b was observed in prickle cells in normal epithelia (200×). (b) Expression of FGFR3b in cytoplasm and nucleus of OSCC with no G2128T mutation (400×). (c) Strong expression of FGFR3b on the cell surface of OSCC with the G2128T mutation (400×).

Download figure to PowerPoint

To investigate the functional significance of G697C amino acid substitution of FGFR3b in OSCC, tyrosine kinase activity was compared between cells expressing the wtFGFR3b and G697CFGFR3b. When a Western blot of Sf9 cell lysate was probed with 4G10 anti-phosphotyrosine antibody, a weak band of 65 kDa corresponding to the FGFR3b band was detected in the wtFGFR3b sample. The equivalent band in the G697CFGFR3b sample was highly autophosphorylated. To control for sample loading the blot was probed with an anti-FGFR3 antibody, which yielded bands of similar intensities in the 2 samples (Fig. 3a). Densitometric analysis was carried out by using NIH image to quantitate the phosphorylation activity of wtFGFR3b and G697CFGFR3b. The band intensities of phosphorylation were normalized to the intensity of FGFR3b protein band, and relative ratio of phosphorylated-FGFR3b protein was calculated as increase of phosphorylation in G697CFGFR3b over that in wtFGFR3b. The densitometric analysis confirmed that G697CFGFR3b had 23 times as much activity as wtFGFR3b (Fig. 3b). These results indicated that the G697C amino acid substitution in the kinase domain of FGFR3b resulted in an enhanced tyrosine kinase activity, which would promote FGFR3b signaling. Su et al.19 expressed a K650E mutant FGFR3 in 293T cells and demonstrated that the mutant FGFR3 exhibited constitutive tyrosine kinase activity. Onose et al.20 found that thyroid cancer cells overexpressing FGFR3 did not differ from control cells in growth rate, however, they continued to proliferate even after the control cells ceased growing. This experiment suggests that FGFR3 is involved in the control of contact inhibition.

thumbnail image

Figure 3. Tyrosine autophosphorylation of wtFGFR3b and G697CFGFR3b. To generate wild-type FGFR3b, cDNA encoding the intracellular region of FGFR3b was inserted into the baculovirus expression vector 6xHis-Tag pAcHLT-A (Pharmingen, San Diego, CA) (wtFGFR3b). To construct mutant FGFR3b, cDNA encoding the intracellular region of FGFR3b was inserted into the pALTER-MAX vector (Promega, Madison, WI), and the G2128T mutation was produced using the Altered Sites II Mammal Mutagenesis System (Promega). The mutant FGFR3b cDNA was cloned into the 6xHis-Tag pAcHLT-A vector (G697CFGFR3b). Sf9 cells were maintained in BaculoGold Protein-free Insect Medium (Pharmingen) at 27°C. The wtFGFR3b and G697CFGFR3b were individually transfected into Sf9 cells according to the manufacturer's Baculovirus Expression Vector System protocol (Pharmingen). After transfection, cell lysates from the same number of wtFGFR3b-expressing Sf9 and G697CFGFR3b-expressing cells were individually prepared and incubated with Ni-NTA agarose (QIAGEN) at 4°C for l hr. The Ni-NTA agarose was recovered by centrifugation, and it was further incubated with phosphorylation reaction buffer (50 mM Tris-HCl, pH = 6.8, 200 mM KCl, 0.1 mM ATP, 1 mM MgCl2, 2 mM DTT and 1 mM PMSF) at 37°C for 15 min. The kinase reaction was resolved by 10% SDS-PAGE and transferred to PVDF membrane (Millipore). The membrane was incubated with 1,000-fold diluted anti-phosphotyrosine antibody (monoclonal antibody 4G10; Upstate Biotechnology, Lake Placid, NY) at 4°C overnight. Bound antibody was detected with an NBT-BCIP detection system (Kirkegaard & Perry Laboratories). Anti-FGFR3 antibody was used as a control to quantify of each protein sample. Densitometric analysis was carried out by using NIH image to quantitate the difference in phosphorylation activity of the FGFR3b band. Phosphorylation levels were normalized to the intensity of FGFR3b protein band; i.e., the intensity of phosphorylated band was divided by that of FGFR3b protein in each lane. Relative ratio of phosphorylated-FGFR3b protein was calculated as increase of phosphorylation in G697CFGFR3b over that in wtFGFR3b. The results represent the mean from 3 independent experiments. Error bars = SEM. (a) Equal amounts of total protein were subjected to SDS-PAGE and immunoblotted with 4G10 anti-phosphotyrosine antibody (anti-pTyr). The autophosphorylation activity of the G697CFGFR3b was increased compared to that of wtFGFR3b. Anti-FGFR3 was used as a control for protein loading. (b) Densitometric analysis confirmed that G697CFGFR3b had 23 times as much activity as wtFGFR3b.

Download figure to PowerPoint

In our present study, the G697C substitution caused constitutively active ligand-independent FGFR3b kinase activity. The apparent over-expression of G697CFGFR3b in 62% of the OSCC cases investigated strongly suggests that this particular mutation is involved in the progression of OSCC. FGFR3b may therefore be an important diagnostic and prognostic marker for OSSC and a molecular target for future therapies.


  1. Top of page
  2. Abstract
  3. Acknowledgements
  4. References

This work was supported in part by a grant-in-aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan and Smoking Research Foundation to T.O. and J.D.S was supported by grant P20 RR16463 from the National Center for Research Resources, National Institutes of Health.


  1. Top of page
  2. Abstract
  3. Acknowledgements
  4. References
  • 1
    Shimada T, Mizutani S, Muto T, Yoneya T, Hino R, Takeda S, Takeuchi Y, Fujita T, Fukumoto S, Yamashita T. Cloning and characterization of FGF23 as a causative factor of tumor-induced osteomalacia. Proc Natl Acad Sci USA 2001; 98: 65005.
  • 2
    Burke D, Wilkes D, Blundell TL, Malcom S. Fibroblast growth factor receptors: lessons from the genes. Trends Biochem Sci 1998; 23: 5962.
  • 3
    Johnson DE, Williams LT. Structural and functional diversity in the FGF receptor multigene family. Adv Cancer Res 1993; 60: 141.
  • 4
    Jaye M, Schlessinger J, Dionne CA. Fibroblast growth factor receptor tyrosine kinases: molecular analysis and signal transduction. Biochem Biophys Acta 1992; 1135: 18599.
  • 5
    Heldin CH. Dimerization of cell surface receptors in signal transduction. Cell 1995; 80: 21323.
  • 6
    Ornitz DM. FGFs, heparan sulfate and FGFRs: complex interactions essential for development. Bio Essays 2000; 22: l0812.
  • 7
    Webster MK, Donoghue DJ. FGFR activation in skeletal disorders: too much of a good thing. Trends in Genetics 1997; 13: 17882.
  • 8
    Yamanaka T, Sakamoto A, Tanaka Y, Zhang Y, Hayashido Y, Toratani S, Akagawa Y, Okamoto T. Isolation and serum-free culture of epithelial cells derived from epithelial rests of Malassez in human periodontal ligament. In Vitro Cell Dev Biol 2000; 36: 54853.
  • 9
    Myoken Y, Myoken Y, Okamoto T, Sato JD, Takada K. Immunocytochemical localization of fibroblast growth factor-1 (FGF-1) and FGF-2 in oral squamous cell carcinoma (SCC). J Oral Pathol Med 1994; 23: 4516.
  • 10
    Cappellen D, De Oliveira C, Ricol D, de Medina S, Bourdin J, Sastre-Garau X, Chopin D, Thiery JP, Radvanyi F. Frequent activating mutations of FGFR3 in human bladder and cervix carcinomas. Nat Genet 1999; 23: 1820.
  • 11
    Jang JH, Shin KH, Park JG. Mutations in fibroblast growth factor receptor 2 and fibroblast growth factor receptor 3 genes associated with human gastric and colorectal cancers. Cancer Res 2001; 61: 35413.
  • 12
    Perez-Castro AV, Wilson J, Altherr MR. Genomic organization of the human fibroblast growth factor receptor 3 (FGFR3) gene and comparative sequence analysis with the mouse Fgfr3 gene. Genomics 1997; 41: 106.
  • 13
    Shibata MA, Shirai T, Ogawa K, Takahashi S, Wild CP, Montesano R, Tsuda H, Ito N. DNA methylation adduct formation and H-ras gene mutations in progression of N-butyl-N-(4-hydroxybutyl) nitrosamine-induced bladder tumors caused by a single exposure to N-methyl-N-nitrosourea. Carcinogenesis 1994; 15: 29658.
  • 14
    Dawson CD, Chang KW, Solt DB. MTSl gene mutations in archival oral squamous cell carcinoma. J Oral Pathol Med 1996; 25: 5416.
  • 15
    Gunduz M, Ouchida M, Fukushima I, Hanafusa H, Etani T, Nishioka S, Nishizaki K, Shimizu K. Genomic structure of the human INGI gene and tumor-specific mutations detected in head and neck squamous cell carcinoma. Cancer Res 2000; 60: 31436.
  • 16
    Sakai E, Rikimaru K, Ueda M, Matsumoto Y, Ishii N, Enomoto S, Yamamoto H, Tsuchida N. The p53 tumor-suppressor gene and ras oncogene mutations in oral squamous cell carcinoma. Int J Cancer 1992; 52: 86772.
  • 17
    Wijk T, Overvoorde J, Hertog J. COR1 H2O2-induced intermolecular disulfide bond formation between receptor protein-tyrosine phosphatases. J Biol Chem 2004; 279: 4435561.
  • 18
    Johnston CL, Cox HC, Gomm JJ, Coombes RC. Fibroblast growth factor receptors (FGFRs) localize in different cellular compartment. J Biol Chem 1995; 270: 3064350.
  • 19
    Su WS, Kitagawa M, Xue N, Xie B, Garofalo S, Cho J, Deng C, Horton WA, Fu XY. Activation of STAT1 by mutant fibroblast growth-factor receptor in thanatophoric dysplasia type II dwarfism. Nature 1997; 386: 28892.
  • 20
    Onose H, Emoto N, Sugihara H, Shimizu K, Wakabayashi I. Overexpression of fibroblast growth factor receptor 3 in a human thyroid carcinoma cell line results in overgrowth of the confluent cultures. Eur J Endocrinol 1999; 40: 16973.