| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of Medicine (S.E., L.Z.), Laboratory Medicine and Pathobiology (S.L.A.), University of Toronto; The Freeman Centre for Endocrine Oncology (S.E., L.Z., S.L.A.), Mount Sinai Hospital; and Ontario Cancer Institute (S.E., L.Z., S.L.A.), University Health Network, Toronto, Ontario, Canada M5G-1X5
Address all correspondence and requests for reprints to: Dr. S. Asa, 610 University Avenue, 4-302, Toronto, Ontario, Canada M5G-2M9. E-mail: sylvia.asa{at}uhn.on.ca.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
FGF signaling is critical in pituitary development. Deletion of FGF10 or its receptor, the FGFR2 IIIb isoform, leads to failure of primordial pituitary development (5). Midgestational expression of a soluble dominant-negative FGFR results in severe pituitary dysgenesis (6). FGF ligands are overexpressed in pituitary tumors. FGF-2, originally described in the bovine pituitary folliculostellate cells, regulates multiple pituitary hormones (7) and is overexpressed by pituitary adenomas (8). Estrogen administration in rats results in pituitary tumorigenesis accompanied by increased FGF-2 expression (9). Elevated circulating FGF-like immunoreactivity is found in patients with type 1 multiple endocrine neoplasia and associated pituitary tumors (10) and in patients with sporadic pituitary adenomas (8). The human endogenous FGF antisense gene (GFG) is expressed in the normal pituitary, where it inhibits pituitary cell proliferation, and is reduced in pituitary tumors (11).
We identified altered FGFR4 expression in pituitary tumors (12) due to expression of an N-terminally deleted isoform, pituitary tumor-derived FGFR4 (ptd-FGFR4) (13) generated by alternative transcription initiation using a cryptic promoter (14, 15). Wild-type FGFR4 is a 110-kDa membrane-anchored protein; ptd-FGFR4 is a 65-kDa cytoplasmic protein with transforming properties (13). The tumorigenic role of ptd-FGFR4, but not wild-type FGFR4, was demonstrated by targeted pituitary expression in transgenic mice (13). The basis for these functional differences between FGFR4 isoforms is unknown.
Cell adhesion is recognized as a determinant of organized growth and maintenance of architectural integrity. Reduced adhesiveness between cells and with extracellular matrix represents a hallmark of neoplastic growth (16). In neuroendocrine tissues, neural cell-adhesion molecule (NCAM) represents a prime candidate implicated in modulating cell growth, migration, and differentiation (17). We examined the ability of ptd-FGFR4 and FGFR4 to modulate cell adhesiveness and assemble an NCAM multiprotein complex that contains N-cadherin and its downstream target, ß-catenin. The data indicate that NCAM and N-cadherin represent important membrane targets for FGFR4-mediated cell adhesiveness and support a model of distinct proadhesive complex interactions by different FGFR4 isoforms. These data shed light on the fundamental basis of pituitary neoplasia that involves loss of normal stromal interactions as identified by reticulin staining and provide a rationale for the use of morphologic markers of stromal interaction in the critical distinction of hyperplasia from neoplasia in neuroendocrine lesions.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Control empty vector-transfected cells and cells transfected with FGFR4 formed tumor nodules that were well delineated and lacked invasive growth. In contrast, cells expressing ptd-FGFR4 formed larger tumors that invaded skin, nerves, and other local structures (Fig. 1B). These findings are reminiscent of the invasive behavior seen in pituitary tumors of mice transgenic for ptd-FGFR4 that invade brain (13).
|
). NIH 3T3 cells expressing ptd-FGFR4 adhered poorly to collagen IV with diminished adhesiveness (
50% reduction) compared with controls (Fig. 2A). In contrast, FGFR4-overexpressing cells demonstrated no significant loss of adhesiveness to the collagen IV extracellular matrix. Furthermore, stimulation of wild-type FGFR4-transfected cells with either the non-FGFR-selective FGF-1 or the FGFR4-selective ligand FGF-19 could not reproduce the effect of ptd-FGFR4 on cell adhesiveness (Fig. 2B). To further examine these FGFR4 isoform differences on cell adhesiveness, we preincubated cells with a neutralizing blocking antibody to activated ß1-integrin. This treatment resulted in marked inhibition of FGFR4-transfected cell adhesiveness but had no significant influence on the depressed adhesiveness of ptd-FGFR4-expressing cells (Fig. 2A). Taken together, these findings suggested that FGFR4 and ptd-FGFR4 recruit distinct adhesive complexes that cannot be explained on the basis of constitutive receptor activation alone.
|
). These differences were evident in NIH 3T3 as well as GH4 cells. Interestingly, overexpression of NCAM was not sufficient to enhance association of ptd-FGFR4 with NCAM or to rescue the loss of cell adhesiveness imparted by ptd-FGFR4 (data not shown).
|
). Inoculation of these stably transfected cells into nude mice revealed that N-cadherin expression in FGFR4-transfected cells was restricted predominantly to the cell membrane and was modestly enhanced compared with vector-transfected control cells (Fig. 4B). In contrast, ptd-FGFR4-transfected GH4 pituitary cells displayed marked reduction of N-cadherin staining, and the N-cadherin in these cells was aberrantly localized to the cytoplasmic compartment (Fig. 4B).
|
). In contrast, transgenic mice developed pituitary lactotroph tumors that exhibited a reduction in membrane N-cadherin reactivity; nontumorous cells in their pituitary glands retained a normal staining pattern (Fig. 4C).
N-Cadherin Alterations Induced by ptd-FGFR4 Impair ß-Catenin Stability
To investigate the molecular consequences of disruption of NCAM/FGFR4 interactions by ptd-FGFR4, we compared the ability of the two FGFR4 isoforms to influence the downstream target of N-cadherin, ß-catenin. GH4 cells stably expressing wild-type FGFR4 demonstrated levels of ß-catenin that were comparable to those in empty vector-transfected cells. In contrast, cells transfected with ptd-FGFR4 demonstrated a marked and consistent reduction in ß-catenin expression (Fig. 5A). By immunohistochemistry, control cells and cells transfected with FGFR4 showed clear and strong membrane staining for ß-catenin. Cells transfected with ptd-FGFR4 exhibited only scant cytoplasmic reactivity with focal nuclear translocation, whereas membrane reactivity was negligible (Fig. 5B). A similar pattern was seen in the pituitaries of transgenic mice expressing ptd-FGFR4 under the control of the prolactin promoter (PRL-ptd-FGFR4; Fig. 5C). In control mice, nontumorous cells expressed membrane-anchored ß-catenin throughout the pituitary, whereas in transgenic mice that pattern was seen only in nontumorous cells, and infiltrating adenomatous cells exhibited a marked reduction in ß-catenin reactivity that was not localized to the cell membrane.
|
). In marked contrast, N-cadherin/ß-catenin interaction was disrupted in cells transfected with ptd-FGFR4. Using immunohistochemistry and a phospho-ß-catenin-specific antibody, we identified scant cytoplasmic staining with no difference between control cells and cells overexpressing FGFR4 (Fig. 5E). Cells stably transfected with ptd-FGFR4 and injected in vivo demonstrated a marked reduction in cytoplasmic phospho-ß-catenin and focal nuclear translocation of phospho-ß-catenin. This difference was not as evident in PRL-ptd-FGFR4 transgenic mice (Fig. 5F) where nuclear phospho-ß-catenin was not seen; however, there was perinuclear accumulation of phospho-ß-catenin in tumor cells that was not evident in nontumorous cells or in the pituitaries of nontransgenic littermates. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Our findings represent one of the earliest reports implicating FGFR4 in cell adhesion. We demonstrate a remarkable difference in the ability of distinct FGFR4 isoforms to maintain normal cell adhesiveness. In contrast to wild-type FGFR4, which maintains affinity for collagen IV, ptd-FGFR4 resulted in marked loss of affinity to this extracellular matrix in vitro. These cell-adhesive differences were associated with corresponding differences in invasive properties in vivo. A role for FGFR4 in cell adhesion has been suggested by an FGFR4 polymorphism in which arginine is substituted for glycine in the transmembrane domain, resulting in increased cell motility and acceleration in breast and colorectal cancer progression (19). However, those changes are remarkable in view of the lack of appreciable change in receptor kinase activity in this polymorphism. The functional differences between FGFR4 isoforms is also consistent with a C-terminally truncated breast cancer cell-derived soluble FGFR4 that serves as a dominant-negative form abrogating wild-type FGFR4 signaling (20).
Our findings attribute dramatic differences between FGFR4 isoforms on cell adhesiveness to distinct interactions with NCAM, a well-described tumor suppressor gene that modulates neurite outgrowth and matrix adhesion by assembling an FGFR4 signaling complex. Inhibition of FGFR4 signaling repressed matrix adhesion induced by NCAM in pancreatic endocrine tumor cells (21). Moreover, neurite outgrowth could not be reconstituted in the absence of both NCAM and FGFR4 signaling (21). In contrast to wild-type FGFR4 that associates strongly with NCAM, ptd-FGFR4 has minimal direct interaction with NCAM.
The polysialated form of NCAM (PSA-NCAM) is highly expressed during brain and pituitary development, but not in the mature gland. PSA-NCAM is detected in pituitary tumors and has been suggested to be a prognostic marker for these neoplasms (22);expression of PSA-NCAM correlates with tumor growth and invasiveness (22). Polysialation has been proposed to involve steric inhibition of membrane-membrane apposition and cell adhesiveness, based on the biophysical properties of PSA (23). We propose that disruption of NCAM/FGFR4 proadhesive complexes, such as by ptd-FGFR4, represents an alternative mechanism for interruption of NCAM-mediated cell-adhesive functions.
N-cadherin is a member of the classical cadherin family of cell-cell adhesion molecules of particular importance in neuroendocrine cell function (24). It is the predominant member expressed in the nervous system and interacts directly with ß-catenin and, in turn, with 
-catenin as the intermediate link to the actin cytoskeleton. This cytoskeleton linkage is critical in stabilizing cadherin clusters at the cell surface and strengthening cell-cell adhesion. Estradiol, a recognized promoter of rodent pituitary proliferation, reduces membranous N-cadherin protein expression to barely detectable levels, whereas antiestrogen treatment reverses this effect (25). We demonstrate that FGFR4-expressing GH4 cells display noninvasive growth associated with intact N-cadherin membranous expression. In contrast, ptd-FGFR4-expression is associated with diminished and ectopic cytoplasmic expression of N-cadherin, and these changes manifest reduced cell adhesion in vitro and invasive growth in vivo. We demonstrate that ptd-FGFR4 disrupts N-cadherin/ß-catenin interaction, contributing to loss of an FGFR4/NCAM/N-cadherin proadhesive complex.
N-cadherin associates with dephosphorylated ß-catenin (26). Uncoupling of cadherins from the cytoskeleton may be an essential aspect of neuroendocrine cell migration; indeed, N-cadherin-mediated neurite outgrowth depends on inactivation of N-cadherin-mediated adhesions at pathway boundaries (27). Regulation of ß-catenin stability represents one means of effecting such changes. Several kinases and phosphatases have been implicated as regulators of ß-catenin phosphorylation. For example, ras has been shown to reduce the stability of the cadherin-catenin bond (28). At least two transmembrane tyrosine kinases, the epidermal growth factor receptor (29) and hepatocyte growth factor receptor (30), also target ß-catenin. Conversely, suppression of the association of the epidermal growth factor receptor with ß-catenin increases the association of ß-catenin with cadherin (31) and inhibits invasive gastric cancer cell growth, presumably through increased cadherin-mediated adhesions (32). Our findings are the first to demonstrate that a tumor-derived isoform of FGFR4 can potentially disrupt ß-catenin stability. Interestingly, ptd-FGFR4, but not FGFR4 in the absence or presence of FGF stimulation, was effective at destabilizing ß-catenin from N-cadherin with resultant diminished ß-catenin expression. Reduced ß-catenin expression is a feature of pituitary adenomas that also correlates with tumor invasiveness (33).
Integrins are heterodimeric cell surface receptors involved in adhesion to molecules in the extracellular matrix (34). Integrin receptors are expressed by many cell types, including anterior pituitary cells (35), where they mediate cell-matrix and cell-cell adhesion, migration, growth, and differentiation. Immunoneutralization of ß1-integrin repressed cell adhesiveness in FGFR4-transfected cells but did not alter the diminished adhesiveness conferred by ptd-FGFR4. Consistent with these data, ptd-FGFR4 failed to activate ß1-integrin and, when introduced in vivo, resulted in marked increase in invasive growth into skin, nerves, and other local structures. These findings are reminiscent of the invasive behavior of tumors in mice transgenic for ptd-FGFR4 (13) and provide suggestive evidence that this integrin represents an additional target for the FGFR4/NCAM/N-cadherin adhesive complex.
In addition to its role in maintaining adherens junctions, ß-catenin can also function as a transcription factor for target genes including cyclin D1 and Myc (36). Nuclear ß-catenin accumulation is a feature of rapidly proliferating and metastasizing solid neoplasms. Our current studies demonstrated loss of ß-catenin with only modest nuclear translocation, consistent with the more indolent nonmetastasizing but locally infiltrative nature of most pituitary adenomas (1).
Indeed, the altered stromal adhesiveness of cells expressing ptd-FGFR4 explains a key feature of pituitary neoplasia. Loss of the reticulin network represents the morphological hallmark of the transition from hyperplasia to adenoma and is a diagnostic marker for neoplasia (37). Neoplastic pituitary cells are characterized by their ability to form solid nests or trabecula in the absence of a stromal support or framework. Our data provide a pathobiological basis for this escape from collagen anchorage dependence by ptd-FGFR4 alteration of N-cadherin and ß-catenin stromal interactions. Disruption of distinct NCAM/N-cadherin proadhesive forces by ptd-FGFR4 provides a novel tumorigenic mechanism that explains the pathobiology of proliferative and infiltrative but nonmetastasizing pituitary neoplasms.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Plasmid Constructions and Transfections
Plasmids containing the coding regions of human FGFR4 or ptd-FGFR4 cDNA were prepared as previously described (13) using pcDNA 3.1 with or without a Topo/His/V5 tag. The NCAM expression vector in pcDNA 3.1 was kindly provided by Dr. U. Cavallaro (Basel, Switzerland).
FGF Stimulation
FGF stimulation was analyzed using the non-FGFR-selective FGF-1 (10 ng/ml; Sigma Chemical Co., St. Louis, MO) or the FGFR4-selective FGF-19 (10 ng/ml; Dr. A. L. Gurney, Genentech Inc., South San Francisco, CA) incubated with 10 U/ml heparin in serum-free media containing insulin (5 µg/ml) and transferrin (5 µg/ml) (38).
Western Blotting and Immunoprecipitations
Equal amounts of protein (50 µg) from whole-cell lysates were solubilized in 2x sodium dodecyl sulfate-sample buffer, separated on 8% polyacrylamide gels, and transferred to nitrocellulose. Blots were incubated with polyclonal affinity-purified rabbit antiserum against the carboxy terminus of FGFR4 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Immunoprecipitations and immunoblotting were performed using anti-FGFR4 (Santa Cruz) or monoclonal antibody to the V5-tag (Invitrogen, Burlington, Ontario, Canada). Antisera for N-cadherin (1:2500) and ß-catenin (1:1000) were purchased from Pharmingen (Missisauga, Ontario, Canada).
Adhesion Assay
Cells were cotransfected with the plasmid of interest and enhanced green fluorescent protein (EGFP) reporter (pEGFP-N2; CLONTECH, Mississauga, Ontario, Canada). In preliminary studies we demonstrated that approximately 95% of EGFP-positive cells also coexpressed the gene of interest by immunocytochemistry. GH4 or NIH 3T3 cells were cotransfected with EGFP and pcDNA or the different FGFR4 cDNAs using Lipofectamine-PLUS (Invitrogen) for 48–72 h. Transfected cells were plated at 1 x105 cells per well of Collagen IV- or fibronectin-coated culture plates in triplicate and incubated for 4 h, washed with PBS, and fixed with PBS containing 3% paraformaldehyde and 2% sucrose. Images of entire wells were obtained with a Leica fluorescence stereomicroscope (Leica Corp., Heidelberg, Germany) and were analyzed using the Image-Pro Plus Analysis Software (version 4.5; Media Cybernetics Inc., Silver Spring, MD).
Animals
The care of animals was approved by the Institutional Animal Care facilities at the Ontario Cancer Institute (OCI, Toronto, Ontario, Canada). Conditions were maintained at 70–74 F and 50–75% humidity. Exposure to light for alternating 12 h intervals was controlled automatically. Mice were fed an autoclaved formula diet and water ad libitum. Tissues were snap frozen in liquid nitrogen and stored at –70 C and/or fixed in formalin and embedded in paraffin for histological and immunohistochemical analysis.
In vivo Tumorigenesis
GH4 cells stably transfected with FGFR4, ptd-FGFR4, or vector alone were injected sc in Swiss nude mice (5 x 106 cells per injection). Tumor development at the site of injection was evaluated on a biweekly basis. Tumors were examined histologically and immunohistochemically to confirm GH4 origin with GH expression, and FGFR4 expression, and to exclude incidental neoplasms that occur spontaneously in nude mice.
Transgenic Mice
The generation and identification of transgenic mice lines expressing ptd-FGFR4 under the control of the pituitary prolactin promoter were described previously (13).
Immunocytochemical Localization
Tumors from nude mice inoculated with FGFR4-isoform stably transfected GH4 cells or pituitaries of transgenic mice were fixed in formalin and embedded in paraffin. For immunolocalization of FGFR4 isoforms, a polyclonal antiserum against human FGFR4 (Santa Cruz) was used at a dilution of 1:1500. A polyclonal antiserum to N-cadherin and a monoclonal antibody to ß-catenin (Pharmingen) were applied at dilutions of 1:200 and 1:500, respectively, and an antibody specific for phospho-ß-catenin (Cell Signaling Technology, Pickering, Ontario, Canada) was applied at 1:150. The specificity of all reactions was verified by replacing primary antiserum with normal rabbit serum, by examining negative control tissues, and by preabsorbing primary antibody with purified peptide.
Statistical Analysis
Data are presented as mean ± SE. Differences were assessed by Student’s paired t test. Significance level was assigned at P < 0.05.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: EGFP, Enhanced green fluorescent protein; FGF, Fibroblast growth factor; FGF4, FGF receptor 4; NCAM, neural cell adhesion molecule; PRL, prolactin; PSA-NCAM, polysialated form of NCAM; ptd-FGR4, pituitary tumor-derived FGF4.
Received for publication April 29, 2004. Accepted for publication June 22, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
recognizes a cryptic promoter in intron 4 of fibroblast growth factor receptor 4. J Biol Chem 278:19597–19602-, ß-, and 
-catenins, and p120 (cell adhesion molecules) in prolactinoma behavior. Mod Pathol 15:1357–1365[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  X. Zhu, S. L. Asa, and S. Ezzat Fibroblast Growth Factor 2 and Estrogen Control the Balance of Histone 3 Modifications Targeting MAGE-A3 in Pituitary Neoplasia Clin. Cancer Res., April 1, 2008; 14(7): 1984 - 1996. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Cristina, G. Diaz-Torga, A. Gongora, M. C. Guida, M. I. Perez-Millan, A. Baldi, and D. Becu-Villalobos Fibroblast growth factor-2 in hyperplastic pituitaries of D2R knockout female mice Am J Physiol Endocrinol Metab, November 1, 2007; 293(5): E1341 - E1351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Zhu, K. Lee, S. L. Asa, and S. Ezzat Epigenetic Silencing through DNA and Histone Methylation of Fibroblast Growth Factor Receptor 2 in Neoplastic Pituitary Cells Am. J. Pathol., May 1, 2007; 170(5): 1618 - 1628. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Boikos and C. A. Stratakis Molecular genetics of the cAMP-dependent protein kinase pathway and of sporadic pituitary tumorigenesis Hum. Mol. Genet., April 15, 2007; 16(R1): R80 - R87. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Shorts-Cary, M. Xu, J. Ertel, B. K. Kleinschmidt-Demasters, K. Lillehei, I. Matsuoka, S. Nielsen-Preiss, and M. E. Wierman Bone Morphogenetic Protein and Retinoic Acid-Inducible Neural Specific Protein-3 Is Expressed in Gonadotrope Cell Pituitary Adenomas and Induces Proliferation, Migration, and Invasion Endocrinology, March 1, 2007; 148(3): 967 - 975. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ezzat, L. Zheng, D. Winer, and S. L. Asa Targeting N-Cadherin through Fibroblast Growth Factor Receptor-4: Distinct Pathogenetic and Therapeutic Implications Mol. Endocrinol., November 1, 2006; 20(11): 2965 - 2975. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ezzat, X. Zhu, S. Loeper, S. Fischer, and S. L. Asa Tumor-Derived Ikaros 6 Acetylates the Bcl-XL Promoter to Up-Regulate a Survival Signal in Pituitary Cells Mol. Endocrinol., November 1, 2006; 20(11): 2976 - 2986. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Gillam, M. E. Molitch, G. Lombardi, and A. Colao Advances in the Treatment of Prolactinomas Endocr. Rev., August 1, 2006; 27(5): 485 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Ezzat and S L Asa The molecular pathogenetic role of cell adhesion in endocrine neoplasia J. Clin. Pathol., November 1, 2005; 58(11): 1121 - 1125. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ezzat, S. Yu, and S. L. Asa The Zinc Finger Ikaros Transcription Factor Regulates Pituitary Growth Hormone and Prolactin Gene Expression through Distinct Effects on Chromatin Accessibility Mol. Endocrinol., April 1, 2005; 19(4): 1004 - 1011. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ezzat, P. Huang, A. Dackiw, and S. L. Asa Dual Inhibition of RET and FGFR4 Restrains Medullary Thyroid Cancer Cell Growth Clin. Cancer Res., February 1, 2005; 11(3): 1336 - 1341. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
