Molecular Cancer Research Folkman AM No Date
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Moffa, A. B.
Right arrow Articles by Ethier, S. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Moffa, A. B.
Right arrow Articles by Ethier, S. P.
Molecular Cancer Research 2:643-652 (2004)
© 2004 American Association for Cancer Research

Signaling and Regulation

Transforming Potential of Alternatively Spliced Variants of Fibroblast Growth Factor Receptor 2 in Human Mammary Epithelial Cells1

Allison B. Moffa1,2, Stacey L. Tannheimer2 and Stephen P. Ethier1,2

1 Cellular and Molecular Biology Graduate Program and 2 Department of Radiation Oncology, Division of Radiation and Cancer Biology, University of Michigan Health System, Ann Arbor, Michigan

Requests for reprints: Stephen P. Ethier, Deputy Director, Karmanos Cancer Institute, 4100 John R. Street, Room 227 HWCRC, Detroit, MI 48201. Phone: 313-966-2110; Fax: 313-966-2659. E-mail: ethier{at}karmanos.org


   Abstract
Top
 Notes
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 References
 
A breast cancer cell line developed in our laboratory (SUM-52PE) has a 12-fold amplification and high-level overexpression of the oncogene fibroblast growth factor receptor 2 (FGFR2). Previously, nine different alternatively spliced FGFR2 variants were isolated from this cell line. Overexpression of two variants that differ only in their carboxyl termini (C1 and C3) has been successfully accomplished in the immortalized human mammary epithelial cell line H16N2. FGFR2 expression led to the activation of the mitogen-activated protein kinase and phosphatidylinositol 3-kinase signaling cascades. Phosphorylation of the adapter protein FGF receptor substrate 2 is much more robust in the cells expressing the C3 variant of FGFR2 compared with the C1 variant. H16N2 cells expressing the full-length FGFR2 with the C1 or C3 carboxyl terminus were tested for their ability to grow under epidermal growth factor (EGF)–independent conditions, in soft agar, and for their ability to invade naturally occurring basement membranes and compared with the parental SUM-52PE cell line. All three cell lines grew under EGF-independent conditions and all were inhibited by the FGFR family specific inhibitor PD173074. The full-length FGFR2-C1 and FGFR2-C3 variants grew robustly in soft agar similar to the parental cell line SUM-52PE. However, cells expressing the C3 variant formed large colonies in agar in both insulin-free and EGF-free medium, whereas the cells expressing the C1 variant required insulin for growth. Soft agar growth was also inhibited by PD173074. Because SUM-52PE was developed from a metastatic breast carcinoma, the FGFR2-overexpressing cell lines were assessed for their ability to invade sea urchin embryo cell membranes. H16N2 cells expressing the C1 carboxyl terminus failed to invade sea urchin embryo cell membranes. By contrast, FGFR2-C3-expressing cells were as invasive as the SUM-52 breast cancer cells and erbB-2-overexpressing H16N2 cells. These results indicate that FGFR2 is a transforming oncogene in human mammary epithelial cells when expressed to levels similar to that found in breast cancer cells with FGFR2 gene amplification. Furthermore, the results suggest that different splice variants have differing transforming activities and that signaling from variants expressing the C3 carboxyl terminus results in more autonomous signaling, cell growth, and invasion.


   Introduction
Top
 Notes
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 References
 
It is well known that ~25% of human breast cancers overexpress the erbB-2 oncoprotein as a result of gene amplification (1). Expression of high levels of p185-erbB-2 leads to its constitutive activation, which results in activation of many downstream signaling molecules. Studies done both in vitro and in vivo have shown conclusively the transforming potential of overexpressed erbB-2 in mammary epithelial cells (2-7). Thus, there is little doubt that erbB-2 gene amplification is a causally significant event in the molecular pathogenesis of human breast cancer.

The molecular mechanisms for the neoplastic progression of the 75% of human breast cancers in which the erbB-2 gene is not amplified remain poorly understood. To date, >50 genes have been shown to be amplified in human breast cancers, and many of them are overexpressed in specimens in which the gene is amplified (8, 9). Although there is some evidence for the causal involvement of genes such as c-myc and AIB1 in breast cancer progression (8, 10-12), there are no genes, besides erbB-2, for which a causal link has been unequivocally established.

The FGFR2 gene is amplified in ~5% of human breast cancers (13). The 10q26 locus, on which the gene resides, has also been found to be amplified in breast cancer in studies using comparative genomic hybridization (9). Furthermore, there is evidence for fibroblast growth factor receptor 2 (FGFR2) overexpression in breast cancers in which the gene is amplified. We have developed a breast cancer cell line (SUM-52) from a pleural effusion specimen obtained from a breast cancer patient (14). These cells have an amplification of the 10q26 locus as detected by comparative genomic hybridization and an amplification of the FGFR2 gene as determined by Southern blotting (15). SUM-52 cells also dramatically overexpress FGFR2 mRNA and protein. Recently, we isolated nine different splice variants of FGFR2 from SUM-52 cell mRNA. These cloning experiments revealed that SUM-52 cells exclusively express the IIIb isoform of FGFR2 that has been associated with epithelial cells. No mRNA for the stromal cell–associated IIIc variant was detected in these cells even by reverse transcription-PCR (RT-PCR). Despite the invariance with respect to the expression of the exon that codes for the third Ig loop in the ectodomain of the receptor, there was significant variability in the isoforms isolated with respect to other parts of the molecule. Of particular relevance to the present studies, we identified splice variants in SUM-52 cells that contain all three known carboxyl termini of FGFR2. The C1 and C2 carboxyl termini are alternatively spliced from the same exon, whereas the C3 terminus is expressed from a separate exon. All FGFR2 isoforms have a conserved tyrosine kinase domain followed downstream by a short stretch of carboxyl-terminal sequence. The C3 terminus is significantly shorter than the C1 terminus, containing only 1 amino acid residue beyond that conserved sequence, compared with 55 amino acids specific to the C1 terminus (16). FGFR2 molecules containing the C1 terminus have been found to be expressed in normal human mammary epithelial (HME) cells. SUM-52 cells, on the other hand, express FGFR2 molecules containing the C1, C2, and C3 termini (17). The C3 terminus is of particular interest because evidence obtained with NIH3T3 cells transformed with different FGFR2 isoforms suggests that C3-containing variants are more transforming than those containing either the C1 or the C2 terminus (16).

The present studies were undertaken to test two specific hypotheses. First, we wanted to determine if overexpression of FGFR2 in HME cells would induce transformed phenotypes expressed by the SUM-52 breast cancer cells and in that way behave similarly to overexpressed erbB-2 in HME cells. Second, we wanted to directly compare the transforming potential of FGFR2 isoforms containing either the C1 or the C3 carboxyl terminus in HME cells. We reported previously that one of the clones isolated from SUM-52 RNA, designated C1#38, was the full-length FGFR2 containing all three Ig loops (including exon IIIb) and the C1 carboxyl terminus. A different clone, designated C3#4, is identical to the C1#38 clone but contains the C3 terminus (17). Thus, in the experiments reported here, immortalized HME cells (H16N2 cells) were transduced with retroviral vectors containing either clone C1#38 or clone C3#4 and tested for their transformed growth phenotypes. The results indicate that, whereas both FGFR2 isoforms were able to transform cells to growth factor and anchorage independence, only the C3 isoform induced factor-independent growth in soft agar and induced an invasive phenotype. Our signal transduction data further corroborate this observation of C3 inducing greater transformation in a ligand-independent fashion. Thus, overexpression of FGFR2 in HME cells results in the acquisition of transformed phenotypes, and the C3 isoform is more transforming than an otherwise identical clone expressing the C1 carboxyl terminus.


   Results
Top
 Notes
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 References
 
Transduction and Expression of FGFR2 Variants in HME Cells
To study the influence of overexpression of specific FGFR2 splice variants in immortalized HME cells, two splice variants that differ only in their carboxyl termini were subcloned into bicistronic retroviral vectors. As we have shown previously, use of bicistronic retroviral vectors is important for the efficient transfer and expression of genes into HME cells (4, 5). The cells chosen for these experiments are H16N2 cells, which were immortalized by human papillomavirus (HPV)-16 (18). H16N2 cells are a relevant model for the SUM-52 cells, because both are luminal just like the majority of human breast cancers (14). Due to expression of the HPV-16 proteins E6 and E7, H16N2 cells have a disruption of the p53 and RB pathways. Most breast cancers, even if they do not have direct mutation of p53 or RB, have disruptions of other molecules epistatic to these cell cycle regulatory pathways (19–26). We have used H16N2 cells in previous studies to overexpress the erbB-2 oncoprotein and have found that these transformed cells recapitulate the transformed phenotypes of human breast cancer cells with an erbB-2 gene amplification. We also found in those studies that H16N2 cells are more permissive for complete transformation than are MCF-10A cells, likely because of the expression of the HPV-16 E6 and E7 proteins.

Figure 1 (left) shows the domain structure of the two FGFR2 isoforms used in these experiments, which differ only in their carboxyl termini. Figure 1 (right) shows the expression at the message and protein levels of the C1#38 and C3#4 clones after transduction into H16N2 cells. As shown in Fig. 1A and B, infected H16N2 cells effectively expressed FGFR2 protein at the molecular weight appropriate for each splice variant. As described previously, the C1#38 clone represents full-length FGFR2-IIIb isoform. Thus, the protein band most prominently detected in the Western blot corresponds in size to the 135-kDa protein also overexpressed in the SUM-52 donor breast cancer cell line. As can be seen in Fig. 1A, SUM-52 cells also expressed a prominent FGFR2 band at ~145 kDa, which represents the glycosylated form of the protein. The glycosylated form of the protein was much less abundant in the H16N2-C1 cells than in SUM-52 cells. Figure 1 also shows that H16N2-C1 cells express equivalent levels of FGFR2 protein regardless of the culture medium in which the cells are maintained. It is notable that protein expression is achieved in serum-free medium supplemented with insulin and hydrocortisone but lacking epidermal growth factor (EGF). This is the condition under which both transduced cell lines and the SUM-52 breast cancer cells are routinely cultured, indicatory of transformation, because untransduced H16N2 cells require EGF for growth. H16N2 cells transduced with the FLAG-tagged C3 vector expressed a single protein species as detected by anti-FLAG immunoblotting, and in this case, expression was maintained to a higher level when cells were cultured in the absence of EGF. Expression of FGFR2-C1 or C3 message, as detected by RT-PCR, is shown in Fig. 1C.



View larger version (48K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 1. Expression of the FGFR2-C1 and FGFR2-C3 isoforms in transduced H16N2 cells. Left, domain structure of the two isoforms used in these experiments, which differ only in their carboxyl termini. Right, expression of the two isoforms in transduced H16N2 cells. A. Western blot analysis of SUM-52 breast cancer cells and H16N2-C1 cells cultured under different conditions of growth factor supplementation. All are grown in serum-free medium supplemented with insulin (I), hydrocortisone (H), and/or EGF (E). Blots were probed with a C1-specific FGFR2 antibody. B. Western blot analysis of H16N2 cells expressing the FLAG-tagged C3#4 clone after growth under different culture conditions. Blots were probed with an anti-FLAG antibody. C. RT-PCR analysis of H16N2 cells transduced with a control vector, or a retroviral expression vector containing the C1#38 or C3#4 clones. -, experiments wherein the reverse transcriptase was omitted before the PCR reaction.

 
To quantify expression levels of each isoform, quantitative RT-PCR (Q-RT-PCR) was done. We designed sets of primers and a probe that recognize the Ig IIIb region of FGFR2 or either the C1 or the C3 carboxyl terminus. In comparison with H16N2 cells expressing only an empty vector (pNG), H16N2 cells overexpressing FGFR2-C1 or FGFR2-C3 expressed ~900- and 1,200-fold more total FGFR2 message, respectively. Furthermore, we found that H16N2 FGFR2-C1 and FGFR2-C3 cells express ~1.5- to 2-fold less total FGFR2 message than SUM-52 breast cancer cells. This finding is not unexpected because the value in SUM-52 cells represents amplified message levels of all nine alternatively spliced isoforms of FGFR2. We also designed carboxyl-terminal–specific primer sets to measure message levels of each isoform. The FGFR2-C1 level in transduced H16N2 cells was approximately equivalent to the amount of FGFR2-C1 in SUM-52 breast cancer cells. We cannot directly compare expression of FGFR2-C3 in SUM-52 cells to H16N2 cells overexpressing this isoform, because the overexpressed C3 isoform contains a carboxyl-terminal FLAG tag at the site where the primers would otherwise bind. As compared with expression of a housekeeping gene (GAPDH), our data suggest that SUM-52 cells express low levels of the FGFR2-C3 isoform relative to the total pool of FGFR2 mRNA and to the FGFR2-C1 isoform. Together, these data indicate that expression of the FGFR2-C1 isoform in transduced H16N2 cells is about equal to the expression of that isoform in SUM-52 cells, and all three cell lines express similar total FGFR2 message levels. These data are shown in Table 1. Importantly, these Q-RT-PCR data also show that FGFR2-C1 and FGFR2-C3 levels are comparable in the transduced H16N2 cells, thus excluding the possibility that the different behavior of these cells is due to an unequal expression of FGFR2 message (see details below).


View this table:
[in this window]
[in a new window]

  		Table 1. Q-RT-PCR Expression Levels of FGFR2

 
Growth Factor–Independent Proliferation of FGFR2-Overexpressing HME Cells
H16N2 cells, like other HME cell lines and strains, have a strict dependence on exogenous EGF for proliferation in serum-free medium. Figure 2 shows growth curves for H16N2 cells expressing either the C1 or the C3 isoform of FGFR2, and a growth curve for the SUM-52 cells cultured in the same EGF-free medium. We have shown previously that SUM-52 cells are EGF independent for growth in serum-free medium (14), and Fig. 2 shows that expression of either FGFR2 isoform induces a similar altered phenotype. Whereas vector control H16N2 cells require EGF for continuous growth, cells transduced with either expression vector have grown continuously in monolayer in the absence of EGF for many passages and are routinely cultured in this medium. Figure 2 also shows that growth of SUM-52 cells and the FGFR2 transduced H16N2 cells is completely blocked by the FGFR-specific kinase inhibitor PD173074. This inhibitor has no effect on the monolayer growth of other breast cancer cell lines that do not have FGFR2 activated (data not shown). Thus, these results show that overexpression of at least two FGFR2 isoforms can result in the acquisition of an EGF-independent phenotype.



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 2. Growth curves for FGFR2-overexpressing cells in EGF-free culture medium in the presence or absence 1 µmol/L concentrations of the FGFR inhibitor PD173074. Cells were seeded into 35-mm culture wells at 50,000 cells per well and grown for up to 11 days. All cells, including the SUM-52 cells, were cultured under serum-free conditions in the absence of EGF. At days 4, 7, and 11, nuclei were isolated from harvested cells and enumerated with a Coulter counter. Points, mean number of cells per dish from three dishes; bars, SD.

 
FGFR2 Activates Several Signal Transduction Pathways
We next investigated the effects of FGFR2 overexpression on activation of downstream signaling pathways. The adapter protein FGF receptor substrate 2 (FRS2, also known as SNT1) is involved in FGFR1 signaling (27). FRS2 binds directly to FGFR, and upon tyrosine phosphorylation, FRS2 recruits Grb2 and the Shp2 phosphatase (27, 28) to activate the mitogen-activated protein kinase pathway. Grb2 also has been shown to recruit the Gab1 adapter protein, which thus activates the phosphatidylinositol 3-kinase pathway as well (29).

To determine whether FRS2 plays a role in FGFR2 signaling in mammary epithelial cells, we studied expression levels and phosphorylation of this protein in SUM-52 and H16N2 cells overexpressing FGFR2-C1 and FGFR2-C3. Whereas the vector control H16N2 cells showed expression of FRS2 protein, it was not phosphorylated. However, we found strong constitutive tyrosine phosphorylation of FRS2 in SUM-52 and H16N2-C3 cells and weaker phosphorylation in FGFR2-C1-expressing H16N2 cells (Fig. 3A). This was our first of several observations suggesting that the FGFR2-C3 isoform is more transforming to epithelial cells than FGFR2-C1. Figure 3A shows that tyrosine phosphorylation of FRS2 in all the cell lines was completely abrogated by 24-hour exposure to 1 µmol/L of the FGFR kinase inhibitor PD173074, suggesting that FRS2 phosphorylation is dependent on FGFR2 activity in the transduced cell lines and SUM-52 breast cancer cells. Interestingly, exogenous addition of the ligand keratinocyte growth factor (KGF; also known as FGF7) to FGFR2-C1 cells led to an increase in phosphotyrosine intensity and an upward molecular weight shift of FRS2, indicative of additional phosphorylation sites being activated (Fig. 3B). KGF induced FGFR2-C1 cells to display a phosphorylated FRS2 level similar to the SUM-52 and FGFR2-C3 cells, suggesting that this full-length isoform is still dependent on ligand stimulation for strong receptor activation. Neither FGFR2-C3-expressing cells nor SUM-52 cells showed any response to exogenous ligand, indicating that these cell lines are ligand independent for signaling.



View larger version (86K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 3. FRS2 phosphorylation. FRS2 was affinity precipitated from SUM-52, H16N2-C1, C3, and pNG vector control cells, and subsequent Western blots were blotted with antibodies against phosphotyrosine and FRS2. Cells were lysed following exposure to (A) PD173074, (B) KGF growth factor, or no treatment, which is DMSO as a vehicle control (I, PD173074 inhibitor; -, no treatment; K, KGF with heparin).

 
We next examined the activation of the downstream mitogen-activated protein kinase and phosphatidylinositol 3-kinase pathways. Data in Fig. 4A show that phosphorylation of Erk-1 and 2 is completely blocked by PD173074 in SUM-52 cells. We also observe that the FGFR2-C3-expressing cells have a higher level of Erk-1 and Erk-2 phosphorylation than FGFR2-C1. The vector control (pNG) H16N2 cells do display phosphorylated Erk, because these cells are grown in the presence of EGF and express the EGF receptor. Erk phosphorylation levels in H16N2-C1 and H16N2-C3 cells are sensitive to the FGFR kinase inhibitor, whereas the vector control cells are not.



View larger version (68K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 4. A. Erk activation. SUM-52, H16N2-C1, C3, and pNG cells were treated for 24 hours in the presence or absence (DMSO) of PD173074 (1 µmol/L). Whole cell lysates were examined for expression of Erk and phospho-Erk. B. Akt activation. SUM-52, H16N2-C1, C3, and pNG cells were treated for 24 hours in the presence or absence of PD173074 (1 µmol/L). Whole cell lysates were examined for expression of Akt and phospho-Akt.

 
We next examined Akt phosphorylation as an indication of phosphatidylinositol 3-kinase pathway activity. Figure 4B shows that Akt is constitutively phosphorylated in SUM-52 breast cancer cells, and this phosphorylation is lost upon addition of inhibitor. In the H16N2 cells, expression of FGFR2-C1 or FGFR2-C3 leads to an increase in Akt phosphorylation, compared with the vector control cells, and this increase in phosphorylation is blocked by PD173074 treatment. These signaling results indicate that FGFR2 is capable of inducing phosphorylation of the adapter protein FRS2 in addition to Erk-1, Erk-2, and Akt. Furthermore, the FRS2 results suggest a mechanism by which the FGFR2-C3 isoform induces more strongly transformed phenotypes than C1.

FGFR2 Overexpression Induces Anchorage-Independent Growth of HME Cells
The previous results described in this article indicate that both FGFR2 isoforms can transform cells to an EGF-independent phenotype in monolayer culture, and in this regard, both splice variants seemed to be equally effective. Experiments were next done to examine the anchorage-independent growth of FGFR2-overexpressing cells and SUM-52 cells under different conditions of growth factor supplementation. Figure 5 shows that both SUM-52 and H16N2-C1 cells formed large colonies in soft agar. Furthermore, colony formation was completely inhibited by 1 µmol/L concentrations of the FGFR inhibitor PD173074 but was unaffected by exposure to the erbB-specific kinase inhibitor CI-1033. By contrast, SUM-149 cells that overexpress and have constitutively activated EGF receptor and H16N2-erbB-2 cells that overexpress erbB-2 also formed colonies in agar that were blocked by CI-1033 but not by the FGFR inhibitor PD173074. Next, experiments were done to directly compare the anchorage-independent growth potential of the H16N2-C1 cells to H16N2-C3 cells. Figure 6 shows that, whereas cells expressing both variants grew in soft agar in medium supplemented with insulin and EGF and in EGF-free medium, only the cells expressing the C3 variant were able to form colonies in insulin-free medium. This observation reinforces the hypothesis that expression of the C3 variant results in a more transformed phenotype than that resulting from overexpression of the C1 variant.



View larger version (112K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 5. Growth in soft agar of FGFR2-overexpressing cells compared with cells overexpressing either erbB-2 or EGF receptor (SUM-149 cells). Cells were seeded into 35-mm wells at 100,000 cells per well and grown in agar for 2 weeks. Cells were cultured either in their normal growth medium or in medium supplemented with the FGFR-specific inhibitor PD173074 or the pan-erbB kinase inhibitor CI-1033.

 


View larger version (69K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 6. Influence of growth factor supplementation on soft agar growth of C1-expressing or C3-expressing H16N2 cells in the presence or absence of PD173074. Cells were grown in agar as described, in culture medium supplemented with both insulin and EGF, or in insulin-free medium (HE) or EGF-free medium (IH). Hydrocortisone (H) was present under all culture conditions.

 
The data in Fig. 6 (right) also suggest that cells expressing the FGFR2-C3 isoform are more highly transformed than their C1-expressing counterparts. Whereas PD173074 (1 µmol/L) completely blocked the growth of H16N2 cells expressing FGFR2-C1 cells in soft agar, some colonies were still detectable in cultures of FGFR2-C3 cells after being cultured in the continuous presence of the kinase inhibitor. To quantify these observations, soft agar colonies that grew under the conditions described above were enumerated using an automated colony counter. The data in Fig. 7 show that, whereas growth of SUM-52 and H16N2-C1 cells was completely inhibited by PD173074 (1 µmol/L), growth of the cells expressing the C3 variant was only reduced by a factor of 2 at the same drug concentration. Indeed, ~3-fold higher concentrations of inhibitor were required to completely block the growth of the C3-expressing cells. At 3.5 µmol/L PD173074, the effect of the drug was still specific, because the EGF receptor–overexpressing SUM-149 breast cancer cells were not growth inhibited at this concentration and grew equally well in agar when grown in the presence of 10 µmol/L PD173074.



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 7. Growth of FGFR2-overexpressing H16N2, SUM-52, and SUM-149 cells in soft agar in the presence of increasing concentrations of PD173074. Cells were grown for 2 weeks in agar as described in Materials and Methods and then stained. Colonies were enumerated using an automated colony counter. Columns, mean number of colonies per well for triplicate wells.

 
Invasive Capacity of FGFR2-Overexpressing Cells
In recent studies, we have employed a novel assay developed by Livant et al. (30) to assess the invasive capacity of cancer cells under serum-free conditions. We have reported previously that SUM-52 and H16N2 cells overexpressing c-erbB-2 are potently invasive in this assay (5). To compare the transformed phenotypes of H16N2 cells expressing FGFR2 with the C1 or C3 carboxyl terminus, these cells were tested for their invasive capacity in the sea urchin embryo basement membrane assay. Figure 8 shows that, whereas H16N2-C1 cells were completely noninvasive in this assay, H16N2-C3 cells were invasive to levels equal to SUM-52 and other breast cancer cell lines (Fig. 8; data not shown). This result most dramatically shows the phenotypic difference between H16N2 cells expressing two different isoforms of FGFR2 and, when combined with the results described above, is consistent with the hypothesis that the FGFR2 variants expressing the C3 carboxyl terminus are more transforming than those expressing the C1.



View larger version (9K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 8. Invasive capacity of FGFR2-overexpressing cells. H16N2-C1 or H16N2-C3 cells were harvested and tested for their ability to invade naturally occurring sea urchin embryo cell membranes. Columns, percentage of cells attached to membranes that invaded into the lumens of the invasion substrates. *C1#38-overexpressing H16N2 cells were assayed twice, with 0% invasion each time. {dagger}C3#4-overexpressing cells were assayed four times.

 
In summary, the experiments reported here show that overexpression of FGFR2-IIIb isoforms can transform HME cells to express growth factor independence, anchorage independence, and invasive phenotypes. SUM-52 breast cancer cells are dependent on FGFR2 signaling to activate signal transduction pathways involving FRS2, mitogen-activated protein kinase, and phosphatidylinositol 3-kinase. H16N2 cells overexpressing the FGFR2-C1 and FGFR2-C3 isoforms also display an increase in signaling activity compared with vector-only H16N2 cells. These phenotypes are similar to those expressed by erbB-2-overexpressing HME cells and to a panel of breast cancer cell lines developed in our laboratory. Furthermore, our results are consistent with previous reports, which suggest that the expression of the shorter C3 carboxyl terminus results in cells that exhibit more transformed phenotypes than cells expressing FGFR2 with the C1 terminus (16). Thus, SUM-52 breast cancer cells, which highly overexpress the FGFR2 gene as a result of gene amplification, not only have high levels of the predominant FGFR2-IIIb isoforms expressed by normal cells but also express variants not expressed by normal cells, which may contribute directly to their aggressive growth potential.


   Discussion
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 References
 
Human breast cancer is a genetically heterogeneous disease. Given the level of genetic complexity in human breast cancer, novel approaches are needed to obtain evidence for primary and causal roles of specific alterations that occur in breast cancer cells. We have employed an experimental strategy that can help to shed new light on this important problem. This approach is based on drawing mechanistic connections between specific altered cellular phenotypes expressed by breast cancer cells and specific genetic alterations that occur in well-characterized model breast cancer cell lines. Normal HME cells are growth factor and anchorage dependent for growth in vitro. Furthermore, these cells are nonmotile and unable to invade naturally occurring basement membrane substrates in vitro. The H16N2 HME cell line used in these studies exhibits these normal cell phenotypes, but the cells are immortal. By contrast, breast cancer cell lines developed in our laboratory grow under defined conditions in vitro, exhibit specific factor independence for growth in monolayer and in soft agar, and are highly motile and invasive.

In the present work, we sought to examine the transforming ability of specific splice variants of FGFR2. We chose to study FGFR2 because one of the cell lines developed in our laboratory (SUM-52PE) has a FGFR2 gene amplification and dramatically overexpresses the gene at both the message and the protein levels. SUM-52 cells are highly growth factor independent in monolayer and in soft agar and have the ability to invade sea urchin embryo cell membranes. These transformed phenotypes and signaling pathway activation are completely blocked in the SUM-52 cells by administration of the PD173074 FGFR kinase inhibitor. The data reported here are consistent with a causal role for FGFR2 amplification in the development of the breast cancer from which the SUM-52 cells were isolated. We found that two different isoforms of FGFR2, identical full-length FGFR2 isoforms differing only in their carboxyl termini, were able to induce EGF-independent growth in immortalized HME cells, and both isoforms also induced robust growth potential under anchorage-independent conditions. These transformed phenotypes were reversed by the FGFR-specific inhibitor PD173074 in the FGFR2-transduced HME cells.

We chose to overexpress FGFR2 in the HPV-16 immortalized cell line H16N2 for several reasons. First, this cell line was derived from luminal mammary epithelial cells and they continue to express the luminal cytokeratins 18 and 19. This is significant because most human breast cancers, including the SUM-52 cell line, are of luminal origin. Second, we found previously that H16N2 cells were highly transformable by overexpression of erbB-2, which yielded phenotypes similar to those expressed by breast cancer cells with an erbB-2 gene amplification. However, it is important to keep in mind that, because these cells were artificially immortalized, they express oncoproteins not found in primary human breast cancer cells. The HPV-16 E7 and E6 proteins are well known to disrupt RB-mediated and p53-mediated checkpoints by inducing the degradation of their target proteins. Most breast cancers have some disruption of RB-regulated pathways, either as a result of p16 loss, cyclin D1 overexpression, or mutation or down-regulation of RB itself. Further, p53 is mutated in ~50% of breast cancers. Nevertheless, it is formally possible that the alterations induced in RB and p53 by HPV viral proteins differ in important ways from those that occur in primary breast cancer. In addition, it is possible that the HPV proteins could have effects independent of RB and p53, which influence the results of our studies. However, the changes induced in H16N2 cells by FGFR2 overexpression are similar to those expressed in the SUM-52 breast cancer cells both at the level of phenotype and in signal transduction and are reversible by FGFR-specific inhibitors.

The SUM-52 breast cancer cells express multiple FGFR2 isoforms (17), so we sought to explore the differences between variants with the C1 and C3 carboxyl termini. Interestingly, the FGFR2 isoform containing the C3 carboxyl terminus was significantly more transforming than the C1-containing counterpart, and this is consistent with observations made previously in NIH3T3 cells (16). Expression of the FGFR2-C3 variant message to levels equivalent to that of the C1 variant resulted in invasive capacity and greater growth factor independence in soft agar. The invasion data are particularly striking as the SUM-52 cells are highly invasive, but the FGFR2-C1 variant was unable to confer invasive capacity on HME cells. In addition, we observed a significant differential in the ability of the FGFR2-C1 and FGFR2-C3 isoforms to phosphorylate the FRS2 adapter protein. The FGFR2-C3 isoform phosphorylates FRS2 to a higher extent, similar to tyrosine-phosphorylated FRS2 levels observed in the SUM-52 breast cancer cells. Interestingly, addition of the KGF ligand increases FRS2 tyrosine phosphorylation levels in FGFR2-C1 cells to be similar to SUM-52 and FGFR2-C3 cells. These observations strongly suggest that the complexity of splice variants expressed in SUM-52 cells has direct relevance to their neoplastic potential and that the FGFR2-C3 variants, which are not expressed in normal HME cells, contribute directly to altered phenotypes expressed by SUM-52 cells.

Additionally, our Q-RT-PCR data suggest that the contribution of the FGFR2-C3 isoforms in SUM-52 cells is low compared with the C1-containing isoforms, suggesting that even disproportionately low levels of the C3 isoform can still induce transformed phenotypes in these breast cancer cells. However, because the only commercially available antibody against FGFR2 recognizes the C1 carboxyl terminus, we are unable to directly compare FGFR2-C1 and FGFR2-C3 protein levels. We hypothesize that the stronger transforming capability of the FGFR2-C3 variant might be due to increased protein expression and/or stability despite equal message levels as FGFR2-C1. In addition, the FGFR2-C1 carboxyl terminus is longer than the FGFR2-C3 terminus and thus contains multiple additional amino acid residues. Included among those is Tyr769, which on phosphorylation is a known binding site for phospholipase C{gamma} (31). Because only two of the nine splice variants originally isolated from SUM-52 cells have thus far been tested in our cell transformation assays, it remains to be determined if other cancer cell–associated isoforms have even greater transforming potential than the full-length, C3-containing variant described here. Further studies will also be conducted to gain insight into the ligand dependence of the various FGFR2 isoforms.

It has been reported that FGFR2 amplification and overexpression occurs in only 1% to 5% of human breast cancers (13). Thus, this genetic alteration is unlikely to be one of the major contributing events in breast cancer development. The low prevalence of this genetic event not withstanding, our results suggest that amplification and overexpression of FGFR2 can drive cellular phenotypes expressed by breast cancer cells and that small molecule inhibitors that block signaling from this receptor can specifically reverse the transformed phenotypes. Thus, FGFR2 is a reasonable target for clinical interventions aimed at breast and other cancer types that overexpress FGFR2 as a result of gene amplification.


   Materials and Methods
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
References
 
Cell Culture Conditions
H16N2 cells are an immortalized HME cell line that were immortalized by the HPV-16 genes E6, E7, and E5 (18). The SUM-52 and H16N2 cells are cultured in Ham's F-12 medium under serum-free conditions. The medium is supplemented with 0.1% bovine serum albumin, fungizone (0.5 µg/mL), gentamicin (5 µg/mL), ethanolamine (5 mmol/L), HEPES (10 mmol/L), transferrin (5 µg/mL), T3 (10 µmol/L), selenium (50 µmol/L), hydrocortisone (1 µg/mL), and insulin (5 µg/mL). Vector control H16N2 cells also grow in the presence of EGF (10 ng/mL). Transduced H16N2 cells grow in G418-supplemented medium to maintain expression of the transgene. All cell culture reagents were obtained from Sigma Chemical Co. (St. Louis, MO).

Development of FGFR2-Overexpressing Cell Lines
FGFR2 variants cloned from the SUM-52PE breast cancer cell line were previously isolated, sequenced, ligated into a bicistronic retroviral vector, and confirmed for protein synthesis by examination of whole cell lysates from transiently transfected 293 cells (17). Virus produced from the transiently transfected 293 cells was collected, sterile filtered using a 0.44 µm filter, and used to infect subconfluent H16N2 cells treated previously with polybrene. The medium was changed 24 hours after infection, and G418 selection began 48 hours after infection. Selected H16N2 cells were eventually divided into different growth factor–defined conditions (IHE, HE, and IH) and serially cultured in these conditions for all experiments conducted.

Q-RT-PCR Reactions
RNA was extracted from SUM-52, H16N2-C1, C3, and pNG cells using the Qiagen (Valencia, CA) RNeasy kit. RNA was converted into cDNA via a reverse transcription reaction using random hexamer primers. Primers and probes were ordered from Applied Biosystems (Foster City, CA) Assays-by-Design service. Primers were either specific for the Ig IIIb region of FGFR2 (pan-specific), the C1 carboxyl terminus, or the C3 terminus. The pan-specific FGFR2 primer sequence is forward: GGGCTGCCCTACCTCAAA and reverse: CAGCACTTCTGCATTGGAACTATTT, and the probe sequence is CCGAGTGCTTGAGAACC. The FGFR2-C1 primer sequence is forward: CAGTTGGTAGAAGACTTGGATCGA and reverse: ACTAGGTGAATACTGTTCGAGAGGTT, and the probe sequence is AACCAATGAGGAATACTTG. Finally, the FGFR2-C3 primers are forward: CTTGGATCGAATTCTCACTCTCACA and reverse: CCTGACCAACTTTTCCCAGTTTCT, and the probe is CCAATGAGATCTGAAAGTTT. GAPDH primer set was used as a control. RNA (5 µg) was used for the RT-PCR reaction, and the product was diluted 1:100 (1:10 for the C3 reactions and control). Q-RT-PCR was done in 25 µL reactions, in 96-well plates, using the Taqman Universal PCR Master Mix (Applied Biosystems). Reactions were done twice, in replicates of six and four, in the University of Michigan Cancer Center Microarray Core Facility.

Interpretation of relative expression data was calculated as described by Livak and Schmittgen (32). Briefly, average values were determined for number of cycles in each reaction to achieve a threshold of fluorescence. Then, from these values was subtracted the number of cycles necessary for the GAPDH reaction. To calculate fold differences between cell lines, the difference between values was calculated, for example, SUM52-C1 – pNG-C1 will yield the difference in number of cycles for the SUM-52 cells using the C1 reaction compared with that reaction in the H16N2 pNG vector control cells. The fold difference is then determined by raising 2 to the negative power of the calculated difference.

FGFR2 Western Blots
Cells were rinsed twice with ice cold HBSS (Life Technologies, Grand Island, NY) and then lysed on ice with a buffer consisting of Tris-HCl (50 mmol/L, pH 8.5), NaCl (150 mmol/L), 1% NP40 (ICN Biomedical, Inc., Aurora, OH), EDTA (5 mmol/L) supplemented with sodium orthovanadate (5 mmol/L), phenylmethsulfonyl fluoride (50 µg/mL), aprotinin (20 µg/mL), and leupeptin (10 µg/mL). Lysates were spun at 20,800 x g at 4°C for 10 minutes and then analyzed for protein using a modified Lowry method. For anti-FLAG immunoprecipitation, whole cell lysate (1 mg) was incubated with anti-FLAG agarose-conjugated beads (Sigma Chemical) for 1 hour at 4°C, washed thrice with lysis buffer and then resolved on a 7.5% polyacrylamide gels, transferred to polyvinylidene difluoride membrane (Millipore Corp., Bedford, MA), and probed with an anti-FLAG antibody (M2 antibody, Sigma Chemical). For detection of FGFR2-C1 variants, whole cell lysates (200 µg) were resolved on a 7.5% polyacrylamide gels, transferred to polyvinylidene difluoride membrane (Millipore), and probed with an anti-FGFR2 antibody (Santa Cruz Biotechnology, Santa Cruz, CA).

Assessment of Monolayer Growth
Cells were plated into six-well plates at 3.5 x 104 cells per well in SFIH plus 5% fetal bovine serum to allow attachment. The next day, plating medium was removed and cells were treated with SFIH ± DMSO or PD173074 (1 µmol/L) for 11 days, with fresh treatment added every other day. The number of cells was determined by counting isolated nuclei with a Coulter counter 4, 7, and 11 days after treatment (33). A plating efficiency was done 24 hours after plating to determine the number of attached cells per well. All experiments are done in triplicate, with mean ± SD. All experiments were done at least twice.

Growth in Soft Agar
A bottom layer of 1:1 Ham's F-12 serum-free medium to 1% agarose is poured and allowed to solidify. A 1-mL suspension of 1.0 x 105 cells in a 0.3% agarose solution was plated into six-well plates and fed every other day by adding 1-mL medium on top of the soft agar. For experiments using kinase inhibitors, control cells were treated with DMSO as a vehicle control and both control and kinase inhibitor were added fresh in the medium every other day. After 3 weeks, excess medium was carefully removed from the wells, and 1 mg/mL solution of p-iodonitrotetrazolium violet (Sigma Chemical) was added to the wells overnight to stain for viable cells. Pictures of these viable cell colonies were taken after staining.

Analysis of FGFR2 Transcript Overexpression by RT-PCR
Total RNA was isolated using Trizol according to their protocol (Life Technologies). RNA (1.5 µg) was used in the reverse transcriptase reaction along with DTT (10 mmol/L), antisense primer (100 mmol/L), deoxynucleotide triphosphates (0.5 mmol/L), and SuperScript II (100 units). Reverse transcription was carried out at 42°C for 50 minutes followed by incubation at 70°C for 15 minutes, with the addition of RNase H (1 unit, Life Technologies) at 37°C for 20 minutes. Product was then used for amplification of FGFR2 using gene-specific primers to the conserved transmembrane domain portion of the receptor (FGFR2 upstream: 5'-AGCAAGCGCCTGGAAGAGAAAA-3' and downstream: 5'-GGCTTATCCATTCTGTGTCCTTC-3'). The AccuTaq LA (Sigma Chemical) protocol was followed using 30 cycles of 98°C for 30 seconds, 60°C for 30 seconds, and 68°C for 2 minutes, with a final extension at 72°C for 7 minutes. RT-PCR product was then run on a 1% agarose gel for comparison. A no reverse transcriptase control was used for each cell line.

Invasion
Cells were suspended in 0.23% trypsin/EDTA (catalogue no. 15050-057, Life Technologies) and placed on sea urchin basement membranes with or without FCS according to established methods (33) for 4 hours at 37°C, the time required to observe maximal invasion percentages for normal and metastatic cells (ref. 33; data not shown). The percentages of spread and adherent cells were evaluated in each assay to check viability prior to fixation in 2% formaldehyde and scored at 400x magnification using phase-contrast optics. Viability ranged from 90% to 98% in all assays. Mean invasion percentages resulted from three independent determinations involving the scoring of all cells in contact with the invasion substrates.

Erk and Akt Western Blots
SUM-52, H16N2-C1, C3, and pNG cells were treated for 24 hours in the presence or absence of PD173074 (1 µmol/L; negative samples were treated with DMSO alone). Cells were lysed as described above, and whole cell lysate (100 µg) was run in a SDS-PAGE gel and blotted onto polyvinylidene difluoride membrane. Antibodies used were anti-Erk1 and phospho-Erk (Santa Cruz Biotechnology) and anti-Akt and phospho-Akt (Cell Signaling Technology, Beverly, MA). Blots were developed using the Amersham Biosciences (Piscataway, NJ) enhanced chemiluminescence Western blotting detection kit.

FRS2 Blots
Cells were treated for 24 hours in the presence or absence of PD173074 (1 µmol/L) or overnight in the presence or absence of KGF (25 ng/mL, Sigma Chemical) plus heparin (2 µg/mL). Cells were lysed as described above, and whole cell lysate (1 mg) was affinity precipitated with p13suc1 agarose-conjugated beads (Upstate, Lake Placid, NY). Samples were run in a SDS-PAGE gel and blotted onto polyvinylidene difluoride membrane. The FRS2 antibody was purchased from Santa Cruz Biotechnology, and the anti-phosphotyrosine antibody (4G10) is from Upstate. These blots were developed via enhanced chemiluminescence (Amersham Biosciences).


   Notes
Top
 Notes
Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 References
 
1 NIH grant 2RO1CA70354 and NIH through the University of Michigan's Cancer Center Support (grant 5 P30 CA46592).

Note: S.L. Tannheimer is currently in Cytokine Biology, ZymoGenetics, Seattle, Washington. Back

Received May 6, 2004; revised September 24, 2004; accepted October 7, 2004.


   References
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 References
 

  1. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire WL. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987;235:177–82.[Abstract/Free Full Text]
  2. Guy CT, Webster MA, Schaller M, Parsons TJ, Cardiff RD, Muller WJ. Expression of the neu protooncogene in the mammary epithelium of transgenic mice induces metastatic disease. Proc Natl Acad Sci U S A 1992;89:10578–82.[Abstract/Free Full Text]
  3. Pianetti S, Arsura M, Romieu-Mourez R, Coffey RJ, Sonenshein GE. Her-2/neu overexpression induces NF-{kappa}B via a PI3-kinase/Akt pathway involving calpain-mediated degradation of I{kappa}B-{alpha} that can be inhibited by the tumor suppressor PTEN. Oncogene 2001;20:1287–99.[CrossRef][Medline]
  4. Woods Ignatoski KM, LaPointe AJ, Radany EH, Ethier SP. erbB-2 overexpression in human mammary epithelial cells confers growth factor independence. Endocrinology 1999;140:3615–22.[Abstract/Free Full Text]
  5. Woods Ignatoski KM, Maehama T, Markwart SM, Dixon JE, Livant DL, Ethier SP. ERBB-2 overexpression confers PI 3' kinase-dependent invasion capacity on human mammary epithelial cells. Br J Cancer 2000;82:666–74.[CrossRef][Medline]
  6. Xie YM, Hung MC. p66Shc isoform down-regulated and not required for HER-2/neu signaling pathway in human breast cancer cell lines with HER-2/neu overexpression. Biochem Biophys Res Commun 1996;221:140–5.[CrossRef][Medline]
  7. Zrihan-Licht S, Lim J, Keydar I, Sliwkowski MX, Groopman JE, Avraham H. Association of csk-homologous kinase (CHK) (formerly MATK) with HER-2/ErbB-2 in breast cancer cells. J Biol Chem 1997;272:1856–63.[Abstract/Free Full Text]
  8. Fejzo MS, Godfrey T, Chen C, Waldman F, Gray JW. Molecular cytogenetic analysis of consistent abnormalities at 8q12-q22 in breast cancer. Genes Chromosomes Cancer 1998;22:105–13.[CrossRef][Medline]
  9. Forozan F, Mahlamaki EH, Monni O, et al. Comparative genomic hybridization analysis of 38 breast cancer cell lines: a basis for interpreting complementary DNA microarray data. Cancer Res 2000;60:4519–25.[Abstract/Free Full Text]
  10. Anzick SL, Kononen J, Walker RL, et al. AIB1, a steroid receptor coactivator amplified in breast and ovarian cancer. Science 1997;277:965–8.[Abstract/Free Full Text]
  11. Roux-Dosseto M, Romain S, Dussault N, et al. c-myc gene amplification in selected node-negative breast cancer patients correlates with high rate of early relapse. Eur J Cancer 1992;28A:1600–4.
  12. Watson PH, Safneck JR, Le K, Dubik D, Shiu RPC. Relationship of c-myc amplification to progression of breast cancer from in situ to invasive tumor and lymph node metastasis. J Natl Cancer Inst 1993;85:902–7.[Abstract/Free Full Text]
  13. Adnane J, Gaudray P, Dionne CA, et al. BEK and FLG, two receptors to members of the FGF family, are amplified in subsets of human breast cancers. Oncogene 1991;6:659–63.[Medline]
  14. Ethier SP, Kokeny KE, Ridings JE, Dilts CA. erbB family receptor expression and growth regulation in a newly isolated human breast cancer cell line. Cancer Res 1996;56:899–907.[Abstract/Free Full Text]
  15. Forozan F, Veldman R, Ammerman CA, et al. Molecular cytogenetic analysis of 11 new breast cancer cell lines. Br J Cancer 1999;81:1328–34.[CrossRef][Medline]
  16. Itoh H, Hattori Y, Sakamoto H, et al. Preferential alternative splicing in cancer generates a K-sam messenger RNA with higher transforming activity. Cancer Res 1994;54:3237–41.[Abstract/Free Full Text]
  17. Tannheimer SL, Rehemetulla A, Ethier SP. Characterization of fibroblast growth factor receptor 2 overexpression in the human breast cancer cell line SUM-52PE. Breast Cancer Res 2000;2:311–20.[CrossRef][Medline]
  18. Band V, Zajchowski D, Swisshelm K, et al. Tumor progression in four mammary epithelial cell lines derived from the same patient. Cancer Res 1990;50:7351–7.[Abstract/Free Full Text]
  19. Ozbun MA, Butel JS. Tumor suppressor p53 mutations and breast cancer: a critical analysis. Adv Cancer Res 1995;66:71–141.[Medline]
  20. Phillips HA. The role of the p53 tumour suppressor gene in human breast cancer. Clin Oncol (R Coll Radiol) 1999;11:148–55.
  21. Barnes DM, Camplejohn RS. P53, apoptosis, and breast cancer. J Mammary Gland Biol Neoplasia 1996;1:163–75.[CrossRef][Medline]
  22. Gasco M, Shami S, Crook T. The p53 pathway in breast cancer. Breast Cancer Res 2002;4:70–6.[Medline]
  23. Varley JM, Armour J, Swallow JE, et al. The retinoblastoma gene is frequently altered leading to loss of expression in primary breast tumours. Oncogene 1989;4:725–9.[Medline]
  24. Borg A, Zhang QX, Alm P, Olsson H, Sellberg G. The retinoblastoma gene in breast cancer: allele loss is not correlated with loss of gene protein expression. Cancer Res 1992;52:2991–4.[Abstract/Free Full Text]
  25. Fung YK, T'Ang A. The role of the retinoblastoma gene in breast cancer development. Cancer Treat Res 1992;61:59–68.[Medline]
  26. Nielsen NH, Emdin SO, Cajander J, Landberg G. Deregulation of cyclin E and D1 in breast cancer is associated with inactivation of the retinoblastoma protein. Oncogene 1997;14:295–304.[CrossRef][Medline]
  27. Kouhara H, Hadari YR, Spivak-Kroizman T, et al. A lipid-anchored Grb2-binding protein that links FGF-receptor activation to the Ras/MAPK signaling pathway. Cell 1997;89:693–702.[CrossRef][Medline]
  28. Hadari YR, Kouhara H, Lax I, Schlessinger J. Binding of Shp2 tyrosine phosphatase to FRS2 is essential for fibroblast growth factor-induced PC12 cell differentiation. Mol Cell Biol 1998;18:3966–73.[Abstract/Free Full Text]
  29. Ong SH, Hadari YR, Gotoh N, Guy GR, Schlessinger J, Lax I. Stimulation of phosphatidylinositol 3-kinase by fibroblast growth factor receptors is mediated by coordinated recruitment of multiple docking proteins. Proc Natl Acad Sci U S A 2001;98:6074–9.[Abstract/Free Full Text]
  30. Livant DL, Linn S, Markwart S, Shuster J. Invasion of selectively permeable sea urchin embryo basement membranes by metastatic tumor cells, but not by their normal counterparts. Cancer Res 1995;55:5085–93.[Abstract/Free Full Text]
  31. Shi E, Kan M, Xu J, Wang F, Hou J, McKeehan WL. Control of fibroblast growth factor receptor kinase signal transduction by heterodimerization of combinatorial splice variants. Mol Cell Biol 1996;13:3907–18.
  32. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 2001;25:402–8.[CrossRef][Medline]
  33. Ethier SP, Kudla A, Cundiff KC. Influence of hormone and growth factor interactions on the proliferative potential of normal rat mammary epithelial cells in vitro. J Cell Physiol 1987;132:161–7.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
L. Raskin, M. Pinchev, C. Arad, F. Lejbkowicz, A. Tamir, H. S. Rennert, G. Rennert, and S. B. Gruber
FGFR2 Is a Breast Cancer Susceptibility Gene in Jewish and Arab Israeli Populations
Cancer Epidemiol. Biomarkers Prev., May 1, 2008; 17(5): 1060 - 1065.
[Abstract] [Full Text] [PDF]

Home page
 Mol Cancer ResHome page
J. Y. Cha, Q. T. Lambert, G. W. Reuther, and C. J. Der
Involvement of Fibroblast Growth Factor Receptor 2 Isoform Switching in Mammary Oncogenesis
Mol. Cancer Res., March 1, 2008; 6(3): 435 - 445.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
E. Lee, C. A. Haiman, H. Ma, D. Van Den Berg, L. Bernstein, and G. Ursin
The Role of Established Breast Cancer Susceptibility Loci in Mammographic Density in Young Women
Cancer Epidemiol. Biomarkers Prev., January 1, 2008; 17(1): 258 - 260.
[Full Text] [PDF]

Home page
 Cancer Res.Home page
T. Kondo, L. Zheng, W. Liu, J. Kurebayashi, S. L. Asa, and S. Ezzat
Epigenetically Controlled Fibroblast Growth Factor Receptor 2 Signaling Imposes on the RAS/BRAF/Mitogen-Activated Protein Kinase Pathway to Modulate Thyroid Cancer Progression
Cancer Res., June 1, 2007; 67(11): 5461 - 5470.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
M. Takeda, T. Arao, H. Yokote, T. Komatsu, K. Yanagihara, H. Sasaki, Y. Yamada, T. Tamura, K. Fukuoka, H. Kimura, et al.
AZD2171 Shows Potent Antitumor Activity Against Gastric Cancer Over-Expressing Fibroblast Growth Factor Receptor 2/Keratinocyte Growth Factor Receptor
Clin. Cancer Res., May 15, 2007; 13(10): 3051 - 3057.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
Z. Q. Yang, K. L. Streicher, M. E. Ray, J. Abrams, and S. P. Ethier
Multiple Interacting Oncogenes on the 8p11-p12 Amplicon in Human Breast Cancer
Cancer Res., December 15, 2006; 66(24): 11632 - 11643.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Moffa, A. B.
Right arrow Articles by Ethier, S. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Moffa, A. B.
Right arrow Articles by Ethier, S. P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal