This is the author’s version of a work that was submitted/accepted for pub-lication in the following source:

Byron, Sara, Gartside, Michael, Wellens, Candice, Goodfellow, Paul, Bir-rer, Michael, Campbell, Ian, & Pollock, Pamela (2010) FGFR2 mutationsare rare across histologic subtypes of ovarian cancer. Gynecologic Oncol-ogy, 117 (1), pp. 125-129.

This file was downloaded from: http://eprints.qut.edu.au/45309/

c© Copyright 2010 Elsevier

NOTICE: this is the author’s version of a work that was accepted for pub-lication in Gynecologic Oncology. Changes resulting from the publishingprocess, such as peer review, editing, corrections, structural formatting,and other quality control mechanisms may not be reflected in this docu-ment. Changes may have been made to this work since it was submittedfor publication. A definitive version was subsequently published in Gyneco-logic Oncology, [VOL 117, ISSUE 1, (2010)] 10.1016/j.ygyno.2009.12.002

Notice: Changes introduced as a result of publishing processes such ascopy-editing and formatting may not be reflected in this document. For adefinitive version of this work, please refer to the published source:

http://dx.doi.org/10.1016/j.ygyno.2009.12.002

1

FGFR2 Mutations are Rare Across Histologic Subtypes of Ovarian Cancer 1

2

Sara A. Byron1, Michael G. Gartside1, Candice L. Wellens1, Paul J. Goodfellow2, 3

Michael J. Birrer3,4, Ian G. Campbell5,6, Pamela M. Pollock1* 4

5

1Cancer and Cell Biology Division 6

Translational Genomics Research Institute 7

Phoenix, Arizona 85004 8

2 Siteman Cancer Center 9

Washington University School of Medicine/Barnes-Jewish Hospital 10

St. Louis, Missouri 63110 11

3 Cell and Cancer Biology Department 12

Center for Cancer Research 13

National Cancer Institute 14

Bethesda, MD 20892 15

4 Present address: 16

Department of Medicine 17

Massachusetts General Hospital 18

Dana-Farber/Harvard Cancer Center 19

Boston, MA 02114 20

5 Victorian Breast Cancer Research Consortium 21

Cancer Genetics Laboratory 22

Peter MacCallum Cancer Centre 23

2

East Melbourne, Victoria, Australia 24

6 Department of Pathology 25

The University of Melbourne 26

Melbourne, Victoria, Australia 27

28

*Corresponding Author 29

445 North 5th Street 30

Phoenix, AZ 85004 31

Phone: 602-343-8815 32

Fax: 602-343-8840 33

e-mail: ppollock@tgen.org 34

35

Sources of Support: Department of Defense CDMRP Ovarian Cancer Research Program 36

(W81XWH-08-1-0244, PMP), National Health and Medical Research Council (Australia) (IGC), 37

the Victorian Breast Cancer Research Consortium (Australia) (IGC), American Cancer Society 38

(PF-07-215-01-TBE, SAB). 39

40

Running Title: FGFR2 Mutations in Ovarian Cancer 41

Key Words: Ovarian Carcinoma, FGFR2, Mutation 42

43

44

45

46

47

3

ABSTRACT 48

Objective: Ovarian cancer is the leading cause of death from gynecologic malignancies in the 49

Western world. Fibroblast growth factor receptor (FGFR) signaling has been implicated to play 50

a role in ovarian tumorigenesis. Mutational activation of one member of this receptor family, 51

FGFR2, is a frequent event in endometrioid endometrial cancer. Given the similarities in the 52

histologic and molecular genetics of ovarian and endometrial cancers, we hypothesized that 53

activating FGFR2 mutations may occur in a subset of endometrioid ovarian tumors, and possibly 54

other histotypes. 55

Methods: Six FGFR2 exons were sequenced in 120 primary ovarian tumors representing the 56

major histologic subtypes. 57

Results: FGFR2 mutation was detected at low frequency in endometrioid (1/46, 2.2%) and 58

serous (1/41, 2.4%) ovarian cancer. No mutations were detected in clear cell, mucinous, or 59

mixed histology tumors or in the ovarian cancer cell lines tested. Functional characterization of 60

the FGFR2 mutations confirmed that the mutations detected in ovarian cancer result in receptor 61

activation. 62

Conclusions: Despite the low incidence of FGFR2 mutations in ovarian cancer, the two FGFR2 63

mutations identified in ovarian tumors (S252W, Y376C) overlap with the oncogenic mutations 64

previously identified in endometrial tumors, suggesting activated FGFR2 may contribute to 65

ovarian cancer pathogenesis in a small subset of ovarian tumors. Despite 66

the low incidence of FGFR2 mutations in ovarian cancer, anti-FGFR therapies may prove useful 67

in the small subset of patients whose tumors have activating FGFR2 mutations. 68

69

INTRODUCTION 70

4

The late stage of diagnosis, resistance to chemotherapy, and biologic heterogeneity of 71

ovarian cancer make it a clinically challenging disease, with an overall 5-year survival rate of 72

only 46% [1]. The four most common histological subtypes of ovarian cancer are serous (80-73

85%), endometrioid (10%), clear cell (5%), and mucinous (3%) [2]. The histomorphologic 74

heterogeneity of ovarian cancers speaks to distinct etiologies and this is supported by the 75

presence of different underlying genetic alterations. For example, TP53 mutations are most 76

common in serous and high-grade endometrioid ovarian carcinomas [3], whereas PTEN, 77

PIK3CA, and CTNNB1(-catenin) mutations are more common in low-grade endometrioid 78

ovarian cancer [4-6]. Despite histomorphologic and underlying genetic differences, most 79

advanced ovarian carcinomas are treated similarly with a standard approach of surgical 80

cytoreduction followed by carboplatin and paclitaxel combination chemotherapy [7, 8]. 81

Fibroblast growth factor (FGF) signaling has been previously implicated in ovarian 82

tumorigenesis. The fibroblast growth factor family includes 18 ligands (FGF1-FGF10 and 83

FGF16-FGF23) which signal through four transmembrane receptor tyrosine kinases (FGFR1-84

FGFR4) and their alternatively spliced isoforms [9]. Alternative splicing of the exons that 85

encode the third immunoglobulin domain of the receptors is the primary determinant of 86

specificity in FGF/FGFR binding and signaling. These splicing events are highly tissue specific 87

and give rise to the “b” and “c” receptor isoforms for FGFR1-FGFR3, which possess distinct 88

ligand specificities [9, 10]. For FGFR2, cells of an epithelial lineage typically only express the 89

“b” isoform (FGFR2b encoded by exon 8) while mesenchymally derived cells only express the 90

“c” isoform (FGFR2c utilizing exon 9). 91

Ovarian carcinomas are thought to arise from cells of the ovarian surface epithelium 92

(OSE), a layer of poorly committed mesodermally derived epithelial cells surrounding the ovary. 93

5

During the process of malignant transformation, OSE cells become more committed to an 94

epithelial phenotype, acquiring expression of epithelial-specific markers such as E-cadherin and 95

CA125 [11]. Normal ovarian surface epithelial cells have been reported to lack the epithelial 96

FGFR2b isoform and instead express the mesenchymal FGFR2c isoform. The expression of the 97

FGFR2c isoform makes OSE unlike most other epithelial cell types and may reflect the 98

pluripotent nature of the OSE [12, 13]. Interestingly, most epithelial ovarian cancers express 99

FGFR2b [14], consistent with the increased epithelial differentiation of this tumor type during 100

malignant transformation. 101

In the NCI-60 cancer cell line panel [15], the ovarian cancer cell lines are unique with an 102

almost universal expression of all four FGFRs (FGFR1-FGFR4) and the highest incidence of 103

detectable expression of FGFR2 mRNA [16]. FGF-stimulated activation of FGFR in epithelial 104

ovarian cancer cell lines contributes to multiple aspects of the malignant phenotype, including 105

proliferation, motility, cell survival, and reorganization of the actin cytoskeleton [13]. In vivo, 106

mRNA and proteins levels of FGF1 (the universal FGF ligand) are associated with poorer overall 107

survival in patients with high-grade advanced stage serous ovarian tumors [17]. FGF9 is also 108

implicated as playing a key role in ovarian endometrioid adenocarcinomas carrying defects in the 109

Wnt/-catenin pathway [18]. 110

Activating mutations in FGFR2 have been reported in a variety of cancer types, including 111

ovarian carcinoma [19]. As part of the Cancer Genome Project, 26 ovarian tumors were screened 112

for mutations in FGFR1-4 [20]. A single mutation in FGFR2, G272V, was identified in a serous 113

ovarian tumor (1 out of 20 serous ovarian tumors screened for an estimated 5% mutation rate). 114

No mutations were identified in endometrioid, clear cell or mucinous ovarian tumors, although 115

6

only small numbers of these tumor subtypes were evaluated. No mutations were detected in 116

FGFR1, FGFR3, or FGFR4 in the 26 tumors investigated. 117

We and others recently identified activating mutations in FGFR2 in endometrial cancer, 118

primarily in the endometrioid histologic subtype [21, 22]. Interestingly, ovarian cancer and 119

endometrial cancer display similarities both histologically and at the molecular genetics level. 120

Endometrioid ovarian carcinoma bears close histological resemblance to endometrioid 121

carcinoma of the endometrium. Endometrioid cancers of the ovary and endometrium share 122

mutations in PTEN, PIK3CA, and CTNNB1 (-catenin) [4-6, 23-25]. 123

Previous reports implicating FGF signaling in ovarian tumorigenesis coupled with the 124

fact endometrioid ovarian cancer and endometrioid endometrial cancer share a number of cancer 125

associated molecular lesions led us to hypothesize that FGFR2 mutations similar to those seen in 126

endometrial cancers may occur in a subset of ovarian neoplasms. 127

128

MATERIALS AND METHODS 129

Clinical specimens and cell lines 130

120 fresh frozen ovarian tumor samples from multiple institutions were investigated, including 131

46 endometrioid, 41 serous, 14 mucinous, 12 clear cell, and 7 mixed histology ovarian tumors. 132

Ovarian tumor samples were obtained from Ian Campbell at the Peter MacCallum Cancer 133

Institute (26 endometrioid, 32 serous, 10 mucinous, 5 clear cell, and 3 mixed histology ovarian 134

tumors), Michael Birrer at the National Cancer Institute (5 endometrioid, 9 serous, 4 mucinous, 7 135

clear cell, and 4 mixed histology), and Paul Goodfellow at Washington University (15 136

endometrioid tumors). The majority of specimens used in this study contained >70% tumor 137

epithelial cells, as determined by H&E staining of multiple sections. For a subset of tumors 138

7

microdissection of frozen sections was required to enrich for neoplastic cells. All samples were 139

deidentified and the study approved by the Western Institutional Review Board (WIRB). 140

Ovarian tumor DNA provided by Dr Ian Campbell underwent whole genome amplification 141

(WGA) using the Repli-G Phi-mediated amplification system (Qiagen, Hilden, Germany), as 142

previously described [26]. Six ovarian cancer cell lines (CAOV3, ES-2, OVCAR-3, SKOV3, 143

SWS-26, TOV-21G) were also sequenced for FGFR2 mutations. 144

145

Detection of FGFR2 mutations 146

Ovarian tumor samples were screened for FGFR2 mutations in exons 7, 8, 9, 10, 13, and 15, as 147

previously described [21]. Primers pairs were designed to amplify from intronic sequences 148

flanking the target exon to include splice sites and the regions proximal to exon 9 previously 149

associated with disrupted FGFR2 splicing due to Alu insertion [27, 28]. All sequencing was 150

performed at the Translational Genomics Research Institute Sequencing Core. 151

152

BaF3 proliferation assays 153

The S252W and Y376C mutations were introduced to the pEF1.FGFR2b.IRES.neo plasmid 154

(NM_022970.3) using the Quikchange XL Site Directed Mutagenesis Kit (Stratagene) according 155

to manufacturer’s instructions. Mutagenesis primers were designed to introduce a missense 156

mutation to encode for the desired mutation in FGFR2 (indicated in bold) and a silent mutation 157

for diagnostic restriction digestion screening of clones (indicated by underline). Site-directed 158

mutagenesis primer sequences were as follows: FGFR2 S252W Forward: 5’-159

GTTGTGGAGCGCTGGCCTCACCGGCC-3’; FGFR2 S252W Reverse: 5’-160

GGCCGGTGAGGCCAGCGCTCCACAAC-3’; FGFR2 Y376C Forward: 5’–161

8

GATTACAGCTTCCCCAGACTGCCTCGAGATAGCCAT-3’; FGFR2 Y376C Reverse: 5’–162

ATGGCTATCTCGAGGCAGTCTGGGGAAGCTGTAATC-3’. After restriction enzyme 163

screening, the entire FGFR2 coding region was sequenced to confirm the presence of the 164

mutation of interest and to ensure that no other mutations were introduced during the 165

mutagenesis process. 166

The IL-3 dependent murine pro B BaF3 cell line was transduced with empty vector, 167

wildtype FGFR2b and mutant FGFR2b using Amaxa nucleofection, according to the 168

manufacturer’s instructions. Transduced cells were stably selected with 1.2 mg/mL Geneticin 169

in the presence of 5 ng/ml IL-3 for 14 days, and then maintained under selection in RPMI 170

supplemented with 10% FBS, 50 nM beta-mercaptoethanol, 100 U/mL penicillin, 100 g/mL 171

streptomycin sulfate, 1.2 mg/mL Geneticin and 5 ng/mL murine IL-3 (R&D Systems). For the 172

proliferation assays, cells were washed in PBS to remove IL-3, and plated at 1 x 104 cells per 173

well in triplicate in a 96 well plate in IL-3 free media containing 1 nM FGF7 and 5 g/mL 174

heparin. Cells had a 50% volume media change on day 3 to provide fresh FGF ligand. On day 175

five, bioluminescent measurement of ATP was assessed as an indicator of cell number using the 176

ViaLight Plus Cell Proliferation/Cytotoxity Kit (Lonza Rockland, Inc.), according to the 177

manufacturer’s instructions. Experiments were performed twice in each of two independent sets 178

of stable cell lines, with triplicate wells measured for each assay. 179

180

RESULTS AND DISCUSSION 181

To characterize the spectrum and frequency of FGFR2 mutations across the histological 182

subtypes of ovarian cancer, 120 ovarian tumors, including 46 endometrioid, 41 serous, 14 183

mucinous, 12 clear cell, and 7 mixed histology ovarian tumors, were screened for mutations in 184

9

FGFR2. Direct sequencing was undertaken for exons 7, 8, 9, 10, 13, and 15. These six exons 185

include >95% of mutations identified in endometrial cancers ([22] and our unpublished results). 186

Furthermore, the majority of germline mutations in the FGFR gene family associated with 187

skeletal and craniosynostosis syndromes occur within these exons. Alternative splicing of 188

FGFR2 exon 8 and exon 9 occurs in a tissue specific fashion, where cells of an epithelial lineage 189

usually only express the ‘b’ isoform encoded by exon 8 (FGFR2b) and mesenchymally derived 190

cells only express the ‘c’ isoform utilizing exon 9 (FGFR2c). The distinct ligand specificities of 191

these receptor isoforms, coupled with tissue-specific ligand expression, mediate normal paracrine 192

epithelial-mesenchymal signaling. Although no activating mutations in exon 8 have been 193

identified in endometrial cancer to date, a large number of activating mutations in exon 9 of 194

FGFR2 in craniosynostosis and skeletal dysplasia syndromes have been identified. Given that 195

normal ovarian surface epithelium express FGFR2c (utilizing exon 9) and a majority of ovarian 196

carcinomas express FGFR2b (utilizing exon 8) [11-13], we screened both exon 8 and exon 9 for 197

mutations. 198

Mutations in FGFR2 were identified in 1/46 (2.2%) endometrioid and 1/41 (2.4%) serous 199

ovarian tumors. Sequencing revealed the normal DNA was wildtype confirming the mutations 200

were somatic in origin. No mutations were seen in mucinous, clear cell, or mixed histology 201

ovarian tumors. In addition, no mutations were detected in FGFR2 in the six ovarian cancer cell 202

lines tested (CAOV3, ES-2, OVCAR-3, SKOV3, SWS-26, TOV-21G). 203

Both of the mutations identified in ovarian tumor samples have been previously reported 204

in endometrial cancer [21]. S252W, the most common mutation observed in endometrioid 205

endometrial cancer [21], was identified here in a grade 2, stage III, endometrioid ovarian tumor 206

and the Y376C mutation was identified in a grade 3, stage III, serous ovarian tumor. A novel 207

10

G272V mutation in FGFR2 has been previously reported in a single serous ovarian tumor [20]. 208

The functional consequence of this mutation is unknown. However, the absence of the G272V 209

mutation in our cohort suggests that this is not a hotspot mutation for FGFR2. Furthermore, in 210

contrast to the known causative role for S252W and Y376C mutations in FGFR2 in Apert and 211

Beare-Stevenson syndromes, respectively (reviewed in [29]), the absence of the G272V mutation 212

in the germline in FGFR2 mutation driven skeletal disorders (Apert, Crouzon, Pfeiffer, Beare-213

Stevenson, Jackson-Weiss syndromes) suggests it may be a passenger mutation, though 214

ultimately this needs to be functionally validated. Altered splicing of FGFR2 due to Alu 215

insertions has been described as an alternative mechanism of FGFR2 activation in a small subset 216

of patients with Apert syndrome [27, 28], however similar alterations were not identified in any 217

of the ovarian tumors analyzed. 218

S252W is located in the highly conserved linker region between the second and third 219

immunoglobulin-like domains in the extracellular domain of the receptor. Extensive structural 220

and biological studies evaluating the causative role of the S252W mutation in the 221

craniosynostosis and limb pathologies of Apert syndrome have shown that the mutation results in 222

activation of the FGFR2b and FGFR2c receptors by two mechanisms. First, the mutation 223

increases the binding affinity of the receptor isoforms for their cognate ligands. The S252W 224

mutant receptor also violates ligand binding specificity: e.g. the “c” isoform can now bind “b” 225

isoform specific ligands, resulting in autocrine receptor activation [30-32]. Normal ovarian 226

surface epithelial cells are unique in that they are the only epithelial tissue identified to date that 227

express FGFR2c and FGF7, members of the FGF signaling family that are typically expressed in 228

mesenchymally derived tissues [33]. Interestingly, the majority of ovarian carcinomas express 229

FGFR2b, and FGF7 has been shown to stimulate DNA synthesis in ovarian cancer cell lines 230

11

expressing FGFR2b, suggesting the establishment of an autocrine loop [14]. Whether these cells 231

have maintained expression of FGFR2c or have undergone an isoform switch from FGFR2c to 232

FGFR2b is currently unknown. Functional studies have shown that a neutralizing antibody to 233

FGF7 can partially inhibit DNA synthesis in the FGFR2b-expressing ovarian carcinoma cell line, 234

41 M [14], suggesting that this isoform switching may play a role in ovarian tumorigenesis rather 235

than just being a “passenger” event that accompanies the epithelial differentiation of this tumor 236

type during malignant transformation. Although it is unknown whether the ovarian tumor with 237

the S252W mutation in FGFR2 identified in this study predominantly expresses the “b” or the 238

“c” isoform of FGFR2, it is tempting to speculate that this tumor expresses the “c” isoform only. 239

The S252W mutation would then phenotypically mimic the previously identified isoform 240

switching, thereby allowing the autocrine activation of FGFR2c by FGF7 in OSE cells. The 241

identification of the pathogenic S252W mutation in this single tumor would in turn suggest that 242

the isoform switching previously observed in ovarian tumors plays a role in the pathogenesis of 243

these tumors. The low rate of activating mutations in FGFR2 identified to date may reflect that in 244

this tissue type FGFR2 is already activated in a ligand dependent manner by isoform switching, 245

and may not require additional mutational activation. 246

Y376C is located in the juxtamembrane domain of FGFR2. Studies investigating the 247

causative role of this mutation in Beare-Stevenson syndrome have proposed that the introduction 248

of an unpaired cysteine by this mutation results in receptor activation through formation of either 249

intermolecular bonds (resulting in covalent dimerization between mutant receptors) or 250

intramolecular bonds (resulting in interactions with two cysteines in the transmembrane domain) 251

[34]. Given that the S252W mutation results in ligand-dependent receptor activation, we 252

undertook studies to determine the ligand-dependency of the Y376C mutation. The homologous 253

12

Y372C mutation in FGFR1 results in ligand independent activation of an osteocalcin FGF 254

response element promoter-luciferase reporter in the osteogenic MC3T3 cell line [35], 255

presumably by the formation of intermolecular disulfide bonds [36]. The homologous Y373C 256

mutation in FGFR3c has been shown to result in receptor dimerization, but predominantly ligand 257

dependent activation of the MAPK pathway in HEK293 cells and L8 myoblasts [37]. We 258

assessed mitogen dependence using the BaF3 cell assay. BaF3 cells do not express endogenous 259

FGF ligands or FGFRs and are IL-3 dependent. Introduction and activation of FGFRs has been 260

shown to substitute for IL-3 to promote cell proliferation [38]. As shown in Figure 1A, the 261

Y376C mutation results in ligand independent receptor activation, as evidenced by BaF3 262

proliferation in the absence of FGF ligand. Stimulation with FGF7 resulted in increased 263

proliferation in the Y376C FGFR2b BaF3 cells compared to wildtype FGFR2b (Figure 1B), 264

demonstrating the mutant receptor can be further activated by the addition of ligand. This is in 265

contrast to the S252W mutation that does not lead to BaF3 proliferation in the absence of ligand, 266

similar to wildtype FGFR2b (Figure 1A). The S252W mutation does lead to increased BaF3 267

proliferation in response to its cognate ligand (Figure 1B), as previously reported [30]. We 268

should note that these functional studies were carried out using the “b” isoform of FGFR2 as the 269

effect of the Y376C mutation on receptor dimerization is independent of the “b” or “c” isoform 270

on which it arises. 271

In conclusion, we identified an S252W mutation in a single endometrioid ovarian 272

carcinoma and a Y376C mutation in a single serous ovarian tumor. The S252W mutation 273

violates the ligand binding specificity of FGFR2b and FGFR2c and phenotypically mimics the 274

occurrence of FGFR2 isoform switching previously observed in ovarian cancer whereas the 275

Y376C mutation results in constitutive receptor activation. In the era of personalized medicine, 276

13

we anticipate that hotspot FGFR2 mutations, including S252W and Y376C, will be incorporated 277

into a multi-gene panel to profile rare somatic mutations, allowing identification of FGFR2 278

mutant tumors even within cancer types with low mutation frequency. Additional studies are 279

needed to determine the pathogenic consequences and therapeutic potential of targeting aberrant 280

activation of FGFR2 due to mutation or isoform switching in ovarian cancer. However, the 281

identification of activating mutations in FGFR2 in ovarian tumors that overlap with the known 282

oncogenic mutations in FGFR2 in endometrial cancer, and the selective sensitivity of FGFR2 283

mutant endometrial cancer cell lines to FGFR inhibition [39], raises the possibility that Together 284

these data raise the possibility that inhibition of FGFR2 by either small molecule kinase 285

inhibitors or extracellular blocking antibodiesFGFR inhibitors may be beneficial in the small 286

subset of women whose ovarian tumors possesshave activating FGFR2 mutations. Furthermore, 287

given that at least one identified mutation functionally mimics FGFR2 isoform switching, there 288

is potential that , as well as inthis may extend to the yet undefined subset of women whose 289

tumorscases that have undergone FGFR2 isoform switching. 290

291

292

14

Conflict of Interest Statement 293

None of the authors have any conflicts of interest to disclose. 294

295

Acknowledgements/Grant Support 296

We would like to thank the TGen Sequencing Core for their excellent work and Dr. John Carpten 297

(Translational Genomics Research Insitute) for providing ovarian cancer cell line DNAs. We 298

gratefully acknowledge funding from the Department of Defense CDMRP Ovarian Cancer 299

Research Program (W81XWH-08-1-0244, PMP), the National Health and Medical Research 300

Council (Australia) (IGC), the Victorian Breast Cancer Research Consortium (Australia) (IGC), 301

and the American Cancer Society (PF-07-215-01-TBE, SAB). 302

303

References: 304

[1] Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA Cancer J 305

Clin 2009. 306

[2] Soslow RA. Histologic subtypes of ovarian carcinoma: an overview. Int J Gynecol Pathol 307

2008;27: 161-74. 308

[3] Kmet LM, Cook LS, Magliocco AM. A review of p53 expression and mutation in human 309

benign, low malignant potential, and invasive epithelial ovarian tumors. Cancer 2003;97: 389-310

404. 311

[4] Campbell IG, Russell SE, Choong DY, Montgomery KG, Ciavarella ML, Hooi CS, 312

Cristiano BE, Pearson RB, Phillips WA. Mutation of the PIK3CA gene in ovarian and breast 313

cancer. Cancer Res 2004;64: 7678-81. 314

[5] Obata K, Morland SJ, Watson RH, Hitchcock A, Chenevix-Trench G, Thomas EJ, 315

Campbell IG. Frequent PTEN/MMAC mutations in endometrioid but not serous or mucinous 316

epithelial ovarian tumors. Cancer Res 1998;58: 2095-7. 317

[6] Wright K, Wilson P, Morland S, Campbell I, Walsh M, Hurst T, Ward B, Cummings M, 318

Chenevix-Trench G. beta-catenin mutation and expression analysis in ovarian cancer: exon 3 319

mutations and nuclear translocation in 16% of endometrioid tumours. Int J Cancer 1999;82: 625-320

9. 321

[7] Muggia FM, Braly PS, Brady MF, Sutton G, Niemann TH, Lentz SL, Alvarez RD, 322

Kucera PR, Small JM. Phase III randomized study of cisplatin versus paclitaxel versus cisplatin 323

and paclitaxel in patients with suboptimal stage III or IV ovarian cancer: a gynecologic oncology 324

group study. J Clin Oncol 2000;18: 106-15. 325

15

[8] McGuire WP, Hoskins WJ, Brady MF, Kucera PR, Partridge EE, Look KY, Clarke-326

Pearson DL, Davidson M. Cyclophosphamide and cisplatin compared with paclitaxel and 327

cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 1996;334: 1-6. 328

[9] Ornitz DM, Itoh N. Fibroblast growth factors. Genome Biol 2001;2: REVIEWS3005. 329

[10] Mohammadi M, Olsen SK, Ibrahimi OA. Structural basis for fibroblast growth factor 330

receptor activation. Cytokine Growth Factor Rev 2005;16: 107-37. 331

[11] Auersperg N, Wong AS, Choi KC, Kang SK, Leung PC. Ovarian surface epithelium: 332

biology, endocrinology, and pathology. Endocr Rev 2001;22: 255-88. 333

[12] Valve E, Martikainen P, Seppanen J, Oksjoki S, Hinkka S, Anttila L, Grenman S, Klemi 334

P, Harkonen P. Expression of fibroblast growth factor (FGF)-8 isoforms and FGF receptors in 335

human ovarian tumors. Int J Cancer 2000;88: 718-25. 336

[13] Steele IA, Edmondson RJ, Leung HY, Davies BR. Ligands to FGF receptor 2-IIIb induce 337

proliferation, motility, protection from cell death and cytoskeletal rearrangements in epithelial 338

ovarian cancer cell lines. Growth Factors 2006;24: 45-53. 339

[14] Steele IA, Edmondson RJ, Bulmer JN, Bolger BS, Leung HY, Davies BR. Induction of 340

FGF receptor 2-IIIb expression and response to its ligands in epithelial ovarian cancer. Oncogene 341

2001;20: 5878-87. 342

[15] Monks A, Scudiero D, Skehan P, Shoemaker R, Paull K, Vistica D, Hose C, Langley J, 343

Cronise P, Vaigro-Wolff A, et al. Feasibility of a high-flux anticancer drug screen using a 344

diverse panel of cultured human tumor cell lines. J Natl Cancer Inst 1991;83: 757-66. 345

[16] Chandler LA, Sosnowski BA, Greenlees L, Aukerman SL, Baird A, Pierce GF. Prevalent 346

expression of fibroblast growth factor (FGF) receptors and FGF2 in human tumor cell lines. Int J 347

Cancer 1999;81: 451-8. 348

[17] Birrer MJ, Johnson ME, Hao K, Wong KK, Park DC, Bell A, Welch WR, Berkowitz RS, 349

Mok SC. Whole genome oligonucleotide-based array comparative genomic hybridization 350

analysis identified fibroblast growth factor 1 as a prognostic marker for advanced-stage serous 351

ovarian adenocarcinomas. J Clin Oncol 2007;25: 2281-7. 352

[18] Hendrix ND, Wu R, Kuick R, Schwartz DR, Fearon ER, Cho KR. Fibroblast growth 353

factor 9 has oncogenic activity and is a downstream target of Wnt signaling in ovarian 354

endometrioid adenocarcinomas. Cancer Res 2006;66: 1354-62. 355

[19] Katoh M. Cancer genomics and genetics of FGFR2 (Review). Int J Oncol 2008;33: 233-356

7. 357

[20] Greenman C, Stephens P, Smith R, Dalgliesh GL, Hunter C, Bignell G, Davies H, 358

Teague J, Butler A, Stevens C, Edkins S, O'Meara S, Vastrik I, Schmidt EE, Avis T, Barthorpe 359

S, Bhamra G, Buck G, Choudhury B, Clements J, Cole J, Dicks E, Forbes S, Gray K, Halliday K, 360

Harrison R, Hills K, Hinton J, Jenkinson A, Jones D, Menzies A, Mironenko T, Perry J, Raine K, 361

Richardson D, Shepherd R, Small A, Tofts C, Varian J, Webb T, West S, Widaa S, Yates A, 362

Cahill DP, Louis DN, Goldstraw P, Nicholson AG, Brasseur F, Looijenga L, Weber BL, Chiew 363

YE, DeFazio A, Greaves MF, Green AR, Campbell P, Birney E, Easton DF, Chenevix-Trench G, 364

Tan MH, Khoo SK, Teh BT, Yuen ST, Leung SY, Wooster R, Futreal PA, Stratton MR. Patterns 365

of somatic mutation in human cancer genomes. Nature 2007;446: 153-8. 366

[21] Pollock PM, Gartside MG, Dejeza LC, Powell MA, Mallon MA, Davies H, Mohammadi 367

M, Futreal PA, Stratton MR, Trent JM, Goodfellow PJ. Frequent activating FGFR2 mutations in 368

endometrial carcinomas parallel germline mutations associated with craniosynostosis and 369

skeletal dysplasia syndromes. Oncogene 2007;26: 7158-62. 370

16

[22] Dutt A, Salvesen HB, Chen TH, Ramos AH, Onofrio RC, Hatton C, Nicoletti R, 371

Winckler W, Grewal R, Hanna M, Wyhs N, Ziaugra L, Richter DJ, Trovik J, Engelsen IB, 372

Stefansson IM, Fennell T, Cibulskis K, Zody MC, Akslen LA, Gabriel S, Wong KK, Sellers 373

WR, Meyerson M, Greulich H. Drug-sensitive FGFR2 mutations in endometrial carcinoma. Proc 374

Natl Acad Sci U S A 2008;105: 8713-7. 375

[23] Risinger JI, Hayes AK, Berchuck A, Barrett JC. PTEN/MMAC1 mutations in 376

endometrial cancers. Cancer Res 1997;57: 4736-8. 377

[24] Oda K, Stokoe D, Taketani Y, McCormick F. High frequency of coexistent mutations of 378

PIK3CA and PTEN genes in endometrial carcinoma. Cancer Res 2005;65: 10669-73. 379

[25] Fukuchi T, Sakamoto M, Tsuda H, Maruyama K, Nozawa S, Hirohashi S. Beta-catenin 380

mutation in carcinoma of the uterine endometrium. Cancer Res 1998;58: 3526-8. 381

[26] Gorringe KL, Choong DY, Williams LH, Ramakrishna M, Sridhar A, Qiu W, Bearfoot 382

JL, Campbell IG. Mutation and methylation analysis of the chromodomain-helicase-DNA 383

binding 5 gene in ovarian cancer. Neoplasia 2008;10: 1253-8. 384

[27] Oldridge M, Zackai EH, McDonald-McGinn DM, Iseki S, Morriss-Kay GM, Twigg SR, 385

Johnson D, Wall SA, Jiang W, Theda C, Jabs EW, Wilkie AO. De novo alu-element insertions in 386

FGFR2 identify a distinct pathological basis for Apert syndrome. Am J Hum Genet 1999;64: 387

446-61. 388

[28] Bochukova EG, Roscioli T, Hedges DJ, Taylor IB, Johnson D, David DJ, Deininger PL, 389

Wilkie AO. Rare mutations of FGFR2 causing apert syndrome: identification of the first partial 390

gene deletion, and an Alu element insertion from a new subfamily. Hum Mutat 2009;30: 204-11. 391

[29] Webster MK, Donoghue DJ. FGFR activation in skeletal disorders: too much of a good 392

thing. Trends Genet 1997;13: 178-82. 393

[30] Yu K, Herr AB, Waksman G, Ornitz DM. Loss of fibroblast growth factor receptor 2 394

ligand-binding specificity in Apert syndrome. Proc Natl Acad Sci U S A 2000;97: 14536-41. 395

[31] Ibrahimi OA, Zhang F, Eliseenkova AV, Itoh N, Linhardt RJ, Mohammadi M. 396

Biochemical analysis of pathogenic ligand-dependent FGFR2 mutations suggests distinct 397

pathophysiological mechanisms for craniofacial and limb abnormalities. Hum Mol Genet 398

2004;13: 2313-24. 399

[32] Ibrahimi OA, Eliseenkova AV, Plotnikov AN, Yu K, Ornitz DM, Mohammadi M. 400

Structural basis for fibroblast growth factor receptor 2 activation in Apert syndrome. Proc Natl 401

Acad Sci U S A 2001;98: 7182-7. 402

[33] Parrott JA, Kim G, Mosher R, Skinner MK. Expression and action of keratinocyte growth 403

factor (KGF) in normal ovarian surface epithelium and ovarian cancer. Mol Cell Endocrinol 404

2000;167: 77-87. 405

[34] Przylepa KA, Paznekas W, Zhang M, Golabi M, Bias W, Bamshad MJ, Carey JC, Hall 406

BD, Stevenson R, Orlow S, Cohen MM, Jr., Jabs EW. Fibroblast growth factor receptor 2 407

mutations in Beare-Stevenson cutis gyrata syndrome. Nat Genet 1996;13: 492-4. 408

[35] White KE, Cabral JM, Davis SI, Fishburn T, Evans WE, Ichikawa S, Fields J, Yu X, 409

Shaw NJ, McLellan NJ, McKeown C, Fitzpatrick D, Yu K, Ornitz DM, Econs MJ. Mutations 410

that cause osteoglophonic dysplasia define novel roles for FGFR1 in bone elongation. Am J Hum 411

Genet 2005;76: 361-7. 412

[36] Robertson SC, Meyer AN, Hart KC, Galvin BD, Webster MK, Donoghue DJ. Activating 413

mutations in the extracellular domain of the fibroblast growth factor receptor 2 function by 414

disruption of the disulfide bond in the third immunoglobulin-like domain. Proc Natl Acad Sci U 415

S A 1998;95: 4567-72. 416

17

[37] Adar R, Monsonego-Ornan E, David P, Yayon A. Differential activation of cysteine-417

substitution mutants of fibroblast growth factor receptor 3 is determined by cysteine localization. 418

J Bone Miner Res 2002;17: 860-8. 419

[38] Naski MC, Wang Q, Xu J, Ornitz DM. Graded activation of fibroblast growth factor 420

receptor 3 by mutations causing achondroplasia and thanatophoric dysplasia. Nat Genet 1996;13: 421

233-7. 422

[39] Byron SA, Gartside MG, Wellens CL, Mallon MA, Keenan JB, Powell MA, Goodfellow 423

PJ, Pollock PM. Inhibition of activated fibroblast growth factor receptor 2 in endometrial cancer 424

cells induces cell death despite PTEN abrogation. Cancer Res 2008;68: 6902-7. 425

426

427

Figure Legends: 428

429

Figure 1. Effects of S252W and Y376C mutations on FGFR2b activity in BaF3 cells (A) 430

Proliferation of BaF3 cells expressing empty vector (EV), wildtype FGFR2b, or mutant (S252W, 431

Y376C) FGFR2b in the absence of FGF ligand. (B) Proliferation of BaF3 cells expressing 432

empty vector (EV), wildtype FGFR2b, or mutant (S252W, Y376C) FGFR2b in response to 433

FGF7, a cognate FGFR2b ligand. 434

435

436