Oncogene (2001) 20, 5059 ± 5061 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc
No evidence of somatic FGFR3 mutation in various types of carcinoma Mehdi Karoui1,2,3, HeÂleÁne Hofmann-Radvanyi3,4, Ute Zimmermann5, Anne Couvelard6, Claude Degott6, Laetitia Faridoni-Laurens7, Jean-Charles Ahomadegbe7, Sylvie Gazzeri8, Elisabeth Brambilla8, Thierry Clerici5, Peggy Charbonnier3, Christophe Tresallet1,3,4, Emmanuel Mitry1, Christophe Penna1, Philippe Rougier1, Catherine Boileau4, Jean-Paul Thiery*,3, Bernard Nordlinger1, Brigitte Franc5 and FrancËois Radvanyi3 1
FeÂdeÂration des SpeÂcialiteÂs Digestives, HoÃpital Ambroise PareÂ, UniversiteÂ Paris V, 92104 Boulogne Cedex, France; 2Service de Chirurgie Digestive et de Transplantation, HoÃpital Claude Huriez, CHRU de Lille, 59037 Lille Cedex, France; 3UMR 144, CNRS/ Institut Curie, 75248 Paris Cedex 05, France; 4Laboratoire de Biochimie et de GeÂneÂtique MoleÂculaire, HoÃpital Ambroise PareÂ, UniversiteÂ Paris V, 92104 Boulogne Cedex, France; 5Service d'Anatomie et de Cytologie Pathologiques, HoÃpital Ambroise PareÂ, UniversiteÂ Paris V, 92104 Boulogne Cedex, France; 6Service d'Anatomie et de Cytologie Pathologiques, HoÃpital Beaujon, 92110 Clichy, France; 7DeÂpartement de Biologie Clinique, Institut Gustave Roussy, 94800 Villejuif, France; 8Service de Pathologie Cellulaire, CHU de Grenoble, 38700 Grenoble, France
Germline speci®c point mutations in the gene encoding ®broblast growth factor receptor 3 (FGFR3) are associated with autosomal dominant human skeletal dysplasia and craniosynostosis syndromes. Mutations identical to the germinal activating mutations found in severe skeletal dysplasias have been identi®ed in certain types of cancer: at low frequency in multiple myeloma and cervix carcinoma and at high frequency in bladder carcinoma. We analysed, by SSCP and sequencing, the prevalence of FGFR3 mutations in 116 primary tumours of various types (upper aerodigestive tract, oesophagus, stomach, lung and skin). The regions analysed encompassed all FGFR3 point mutations previously described in severe skeletal dysplasia and cancers. No mutations were detected in the tumour types examined, suggesting that FGFR3 mutations are restricted to a few tumour types, the evidence to date suggesting that they are very speci®c to bladder carcinomas. Oncogene (2001) 20, 5059 ± 5061. Keywords: FGFR3; growth factor receptor; carcinoma; oncogene Fibroblast growth factor receptor 3 (FGFR3) belongs to a family of structurally related tyrosine kinase receptors (FGFR1 ± 4) that regulate growth, dierentiation, migration, wound healing and angiogenesis (Johnson and Williams, 1993; Robertson et al., 2000). These receptors are glycoproteins composed of two or three extracellular immunoglobulin-like domains, a transmembrane domain, and a split tyrosine-kinase domain. Binding to members of the ®broblast growth factor family induces FGFR dimerization, resulting in autophosphorylation of the kinase domain and interaction with and phosphorylation of signalling proteins.
*Correspondence: J-P Thiery; E-mail: [email protected]
Received 9 March 2001; revised 18 April 2001; accepted 23 May 2001
Alternative mRNA splicing mechanisms generate many receptor isoforms, diering in ligand speci®city. Isoforms FGFR3b and FGFR3c have dierent tissue distributions; for example, FGFR3b is the main form in epithelial cells whereas FGFR3c is the form found in chondrocytes (Delezoide et al., 1998). Speci®c point mutations in dierent domains of FGFR3 are associated with autosomal dominant dwar®sm and craniosynostosis syndromes such as hypochondroplasia, achondroplasia, severe achondroplasia with developmental delay and acanthosis nigricans (SADDAN), thanatophoric dysplasia, Crouzon syndrome with acanthosis nigricans, and coronal craniosynostosis (Tavormina et al., 1999; Robertson et al., 2000). Several reports have demonstrated that these mutations lead to constitutive activation of the receptor (Naski et al., 1996; Webster and Donoghue, 1997; Tavormina et al., 1999). In contrast with this inhibitory role on bone growth, an oncogenic role has been proposed for FGFR3 in multiple myeloma. A t(4;14) (p16.3; q32.3) chromosomal translocation with breakpoints on 4p16 located 50 ± 100 kb centromeric to FGFR3 is present in a subset of multiple myelomas and is associated with overexpression of FGFR3. In rare cases (one primary tumour and two cell lines) point mutations activating FGFR3 have been found to be associated with the translocation (Chesi et al., 1997; Richelda et al., 1997; Fracchiolla et al., 1998). We recently provided the ®rst evidence that FGFR3 is involved in epithelial malignancies by identifying, in bladder and cervical carcinomas, somatic mutations of FGFR3 identical to the activating mutations responsible for thanatophoric dysplasia (Cappellen et al., 1999). In this study, we investigated the oncogenic role of FGFR3 in other carcinomas by assessing the prevalence of FGFR3 somatic mutations in a series of 116 primary tumours consisting of 27 upper aerodigestive tract, 32 oesophagus, 15 stomach, 21 lung and 21 skin carcinomas (Table 1).
FGFR3 mutations in carcinomas M Karoui et al
Tumour cells were isolated by manual microdissection from formalin-®xed and paran-embedded tissue. We used single strand conformation polymorphism (SSCP) and sequencing to analyse the four regions of FGFR3 located in exons 7, 10, 15 and 19, known to harbour the point mutations (R248C, S249C, G370C, S371C, Y373C, K650M/E, J807L/G/R/C/W) previously described in multiple myeloma, bladder and cervix carcinomas, thanatophoric dysplasia and SADDAN (Figure 1). None of these mutations of the FGFR3 gene were detected in the 116 carcinomas analysed. Controls carrying known mutations were included in our Table 1 Tumour type Upper aerodigestive tract oral cavity oropharynx hypopharynx Oesophagus Stomach Lung
analysis to ensure that it was possible to detect the dierent FGFR3 mutations in the SSCP conditions used. In the course of our mutational study, we detected four single nucleotide polymorphisms in the FGFR3 regions examined. Two of these polymorphisms had been reported before: the T?C substitution at the ®rst position of codon 384 in exon 10 leading to a Phe?Leu aminoacid substitution (Fracchiolla et al., 1998), and the silent A?G substitution at the third position of codon 651 in exon 15 (Wu et al., 2000). The two remaining single-nucleotide polymorphisms had not been identi®ed before: a silent G?A substitution
Types of carcinoma studied Histology
Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Adenocarcinoma of Barrett's oesophagus Adenocarcinoma Signet ring cell carcinoma Squamous cell carcinoma Adenocarcinoma Small cell carcinoma Squamous cell carcinoma Basal cell carcinoma
Tumours analysed (n=116) 27 12 3 12 32 24 8 15 10 5 21 11 5 5 21 11 10
Figure 1 Location of activating mutations of FGFR3 associated with severe developmental syndromes and cancer. (a) A schematic diagram of the FGFR3 protein is presented. Immunoglobulin-like domains (I to III), transmembrane domain (TM), proximal tyrosine kinase domain (TK1) and distal tyrosine kinase domain (TK2) are indicated. Point mutations associated with severe autosomal dominant skeletal disorders (thanatophoric dysplasia, TD; severe achondroplasia with developmental delay and acanthosis nigricans, SADDAN) and cancer (multiple myeloma, MM; bladder cancer, BC; cervix carcinoma, CC) are shown. TD mutations are displayed in red, SADDAN mutations in blue, MM mutations in yellow, BC mutations in green, CC mutations in purple. (b) The FGFR3 exons 7, 10, 15 and 19 are indicated by white boxes and the introns are indicated by lines. The 3' untranslated region is indicated by the dashed box. The approximate locations of the primers and approximate positions of codons 248, 249, 370, 371, 373, 650, 807 are indicated. The sequences of the primers and the ampli®cation conditions used are given in Billerey et al. (2001). The codons are numbered according to FGFR3c cDNA open reading frame. The FGFR3b isoform, which is expressed in epithelial cells, contains two more amino acids than the FGFR3c isoform, which is expressed in bone. Isoforms FGFR3b and FGFR3c result from a mutually exclusive splicing event in which the second half of the juxta-membrane Ig-like domain of FGFR3 is encoded by either the 151 nucleotides of exon 8 or the 145 nucleotides of exon 9 (Murgue et al., 1994). Therefore the G370C, Y373C, and K650E mutations in TD are equivalent to the G372C, Y375C and K652E mutations respectively in carcinomas Oncogene
FGFR3 mutations in carcinomas M Karoui et al
at codon 258 in exon 7 and an G?A substitution at nucleotide position 22 in intron 15. Germinal point mutations resulting in FGFR3 activation are responsible for dwar®sm syndromes. Surprisingly, similar FGFR3 mutations have also been implicated in tumorigenesis. Somatic FGFR3 mutations identical to those found in thanatophoric dysplasia (neonatal lethal dwar®sm syndrome) and a related genetic disease SADDAN were shown to be associated with rare cases of multiple myeloma (Chesi et al., 1997; Richelda et al., 1997) and more recently with bladder and cervical carcinomas (Cappellen et al., 1999). The frequency of FGFR3 mutations is high in bladder tumours (36%, Billerey et al., 2001; 41%, Sibley et al., 2001; 47%, van Rhijn et al., 2001), and lower in cervical tumours (3 of 12 in our ®rst study (Cappellen et al., 1999), 1 of 51 and 0 of 104 tumours in the studies of Wu et al. (2000) and Yee et al. (2000)). In this study, we examined the prevalence of thanatophoric and SADDAN dysplasia-type mutations in carcinomas of various types (upper aerodigestive tract, oesophagus, stomach, lung and skin) and found no mutations among the 116 primary tumours examined. These ®ndings suggest that these activating mutations are not common events in these carcinomas. A high proportion (25 ± 60%) of bladder cancer and some cervical carcinomas are related to tobaccosmoking (Cohen and Johansson, 1992). The absence of FGFR3 mutations in lung, upper aerodigestive tract
and oesophagus cancers indicates that FGFR3 mutations are not present in all smoking-related cancers. To date, FGFR3 mutations have been found to occur only in two genitourinary carcinomas, occurring with a particularly high frequency in bladder cancer. This is not the ®rst case of a particular oncogene restricted speci®cally to one or a few types of carcinoma. For example, somatic mutations of RET are found mostly in cases of thyroid cancer and the frequency of K-ras mutations is particularly high in lung, colon and pancreatic cancers. Urothelium is an epithelial structure with two functions: it helps to ensure smooth urinary ¯ow, and constitutes a barrier against aggression. The high frequency of FGFR3 mutations speci®c to bladder carcinomas suggests that FGFR3 ful®ls an important but as yet unknown function in this epithelium.
Acknowledgements We thank Prs FrancËois ReneÂ Pruvot, Jacques Belghiti, Olivier Farges, Jean BeÂnard, FrancËois Janot, Christian Brambilla and Philippe SaõÈ ag for their invaluable contribution to this work. This work was supported by grants from the ComiteÂ de Paris Ligue Nationale Contre le Cancer (UMR 144, laboratoire associeÂ), the CNRS, the Institut Curie, and the Groupement des Entreprises FrancËaises dans la Lutte contre le Cancer (HR).
References 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., 158, 1955 ± 1959. Cappellen D, De Oliveira C, Ricol D, Gil Diez de Medina S, Bourdin J, Sastre-Garau X, Chopin D, Thiery JP and Radvanyi F. (1999). Nat. Genet., 23, 18 ± 20. Chesi M, Nardini E, Brents LA, Schrock E, Ried T, Kuehl WM and Bergsagel PL. (1997). Nat. Genet., 16, 260 ± 264. Cohen SM and Johansson SL. (1992). Urol. Clin. North Am., 19, 421 ± 428. Delezoide AL, Benoist-Lasselin C, Legeai-Mallet L, Le Merrer M, Munnich A, Vekemans M and Bonaventure J. (1998). Mech. Dev., 77, 19 ± 30. Fracchiolla NS, Luminari S, Baldini L, Lombardi L, Maiolo AT and Neri A. (1998). Blood, 92, 2987 ± 2989. Johnson DE and Williams LT. (1993). Adv. Cancer Res., 60, 1 ± 41. Murgue B, Tsunekawa S, Rosenberg I, de Beaumont M and Podolsky DK. (1994). Cancer Res., 54, 5206 ± 5211. Naski MC, Wang Q, Xu J and Ornitz DM. (1996). Nat. Genet., 13, 233 ± 237.
Richelda R, Ronchetti D, Baldini L, Cro L, Viggiano L, Marzella R, Rocchi M, Otsuki T, Lombardi L, Maiolo AT and Neri A. (1997). Blood, 90, 4062 ± 4070. Robertson SC, Tynan JA and Donoghue DJ. (2000). Trends Genet., 16, 265 ± 271. Sibley K, Cuthbert-Heavens D and Knowles MA. (2001). Oncogene, 20, 686 ± 691. Tavormina PL, Bellus GA, Webster MK, Bamshad MJ, Fraley AE, McIntosh I, Szabo J, Jiang W, Jabs EW, Wilcox WR, Wasmuth JJ, Donoghue DJ, Thompson LM and Francomano CA. (1999). Am. J. Hum. Genet., 64, 722 ± 731. van Rhijn BW, Lurkin I, Radvanyi F, Kirkels WJ, van der Kwast TH and Zwartho EC. (2001). Cancer Res., 61, 1265 ± 1268. Webster MK and Donoghue DJ. (1997). Trends Genet., 13, 178 ± 182. Wu R, Connolly D, Ngelangel C, Bosch FX, Munoz N and Cho KR. (2000). Oncogene, 19, 5543 ± 5546. Yee CJ, Lin O and Boyd J. (2000). J Natl Cancer Inst, 92, 1848 ± 1849.