| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Departments of Immunology [K. W. F., R. D. G., D. M. S.], Cell and Molecular Biology [B. E. W., B. A. F., J. M. R., N. M. G.], and Pathology [M. O., M. I.], Baylor College of Medicine, Houston, Texas 77030
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To resolve the role of FGFR1 and FGFR2 in CaP tumor growth and progression, we have used CID to temporally control FGFR signaling in transplant tumors. In CID, a lipid-permeable, synthetic dimerizing drug, such as AP20187, is used to noncovalently bind and oligomerize modified CID-binding domains, such as FK506 binding protein 12 KD (FKBP12), that are genetically fused to the target protein domains. In turn, CID removal terminates signaling. Hence, CID allows the conditional and reversible activation of targets proteins when target proteins are naturally activated via oligomerization (11, 12, 13, 14) . Using CID, we could directly determine the quantitative or qualitative signaling differences between FGFR1 and FGFR2 in CaP-derived cells.
Here we describe the application of iFGFRs to temporal studies of the role of the FGFR axis in tumor progression using prostate tumor cell (15) lines derived from the TRAMP model (16, 17, 18) . We demonstrate that iFGFR1 affects tumor "establishment" and early growth of transplanted prostate tumors in syngeneic hosts by promoting proliferation. However, at later stages of tumor development, iFGFR1 no longer promotes tumor growth. In contrast, iFGFR2 appears to interfere with tumor establishment and growth. Additionally, FGFR1 activation leads to higher Erk phosphorylation than FGFR2 and up-regulation of the metastasis-associated protein OPN, which may be responsible for some of the selective tumor-promoting effects of iFGFR1.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell Lines.
TRAMP-C2N and TRAMP-C1A are subclones of TRAMP-C2 and TRAMP-C1, respectively (15)
, derived from a primary tumor of the TRAMP model (17
, 21)
. To establish C2N and C1A subclones expressing iFGFRs, cells were transfected with linearized pSH1/M-FGFR1-FvFvls-E, pSH1/M-FGFR2-FvFvls-E, or pSH1/MFvFvls-E plus vector pBJ5-Neo at a 10:1 ratio using FuGENE-6 (Roche Applied Sciences, Indianapolis, IN) following the manufacturer’s instructions. All TRAMP-derived cell lines were cultured as described previously (15)
. Clonal iFGFR cell lines were derived by limiting dilution. For all in vitro experiments, cells were grown to confluence in 100-mm tissue culture plates, washed twice with DMEM, and serum starved for 48 h in DMEM unless otherwise stated.
Western Blot Analysis and Immunoprecipitation.
After starving cells, a time course was done with addition of AP20187 (100 nM) or serum (10%)-containing media. Immunoblotting was carried out as described previously (14)
. Membranes were incubated with primary antibodies against phospho-Erk, total Erk (Cell Signaling Technology, Beverly, MA), or 
-tubulin (Santa Cruz Biotechnology, Santa Cruz, CA) or anti-OPN (Santa Cruz Biotechnology) diluted 1:500 in blocking buffer overnight or anti-HA (HA.11; Covance, Richmond, CA) at 1:1500 for 1 h, followed by a secondary antibody of either HRP-linked polyclonal goat antimouse antibody (1:2000; Santa Cruz Biotechnology) or HRP-linked polyclonal goat antirabbit antibody (1:1000; Santa Cruz Biotechnology) for 30 min. Immunoprecipitation was carried out using the IMMUNOcatcher kit (Cytosignal Research Products, Irvine, CA) following the manufacturer’s protocol using 2 µg anti-phosphotyrosine antibody 4G10 (Upstate Biotechnology, Lake Placid, NY) for immunoprecipitation and HA.11 immunoblotting as described above. Phosphatase inhibitors I and II (Sigma, St. Louis, MO) were added to lysate.
RNase Protection Assay.
RNase protection assays were performed following the manufacturer’s instruction for the RNase Protection Kit (Roche Applied Sciences). Briefly, murine OPN cDNA was PfuI amplified using forward primer 5'-ACAGATTGGCAGTGATTTGCTTTTGCC-3' and reverse primer 5'-ACAGCTTGTCCTTGTGGCTGTGAAAC-3' and cloned into pCR-Blunt (Invitrogen, Carlsbad, CA) to make pCR-osteo. PvuII-linearized pCR-osteo and MunI-linearized plasmid pA544 containing ß-actin (Roche Applied Sciences) were used to synthesize [32P]CTP-labeled antisense RNA probes of 150 and 255 bp in size, respectively, using the SP6/T7 Transcription Kit (Roche Applied Sciences). Template DNA was digested with RNase-free DNase I, and probes were gel purified before hybridization with 1 µg of total RNA. Total RNA was isolated using Trizol reagent (Life Technologies, Inc., Gaithersburg, MD) from cells after starvation and incubation with 100 nM AP20187 for various times. Hybridization was carried out for 4 h at 45°C. "Nonprotected" RNA molecules and excess probe were digested with RNase A and T1. "Protected" RNA molecules were purified and resolved on a denaturing polyacrylamide gel, dried, and imaged by X-ray film.
Apoptosis Assay and Proliferation Assays.
Cells were grown to confluence in 12-well plates, washed twice with DMEM, and plated in either 10% serum media or with DMEM in triplicate wells ± 100 nM AP20187. After 22 h of culture, media were removed, and cells received 3 J of UVC using a UV Stratalinker 1800 (Stratagene, La Jolla, CA). Media were replaced, and cells were harvested with a cell scraper 2 h later. Cells were centrifuged and resuspended in 50 µl of DMEM. An equal volume of cells was mixed with 0.2% trypan blue for live/dead counting. To measure effects on proliferation, 5000 cells were plated in DMEM with log dilutions of AP20187 in duplicate in 96-well plates. After 24 h, a CyQUANT Cell Proliferation Kit (Molecular Probes, Eugene, OR) was used following the manufacturer’s protocol. A Titertek Fluoresckan II (ICN Pharmaceuticals, Costa Mesa, CA) was used to measure CyQUANT fluorescence. Probabilities were determined by t test.
Immunohistochemistry.
Immunohistochemistry was carried out as described previously (18)
. Slides were incubated with anti-Ki-67 antibody (Novacastra, Newcastle upon Tyne, United Kingdom) at 1:1000 in Power Block for 1 h, followed by incubation with biotin-conjugated goat antirabbit antibody at 1:2000 (Vector Laboratories, Burlingame, CA) for 1 h. Probability was determined by t test.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
|
Initially, to determine whether differences in autophosphorylation between C2N-iFGFR1 and C2N-iFGFR2 could explain the disparate effects of distinct FGFRs seen in previous studies (5
, 6)
, we examined AP20187-induced tyrosine phosphorylation of iFGFRs (Fig. 1B)
. No phosphorylation over background was observed for either iFGFR1 or iFGFR2 after addition of serum, confirming their insulation from activation by serum FGFs or endogenous serum-activated receptors (Fig. 1B)
. Furthermore, 5 min after drug addition, we observed an equivalent increase in phosphorylation of both iFGFR1 and iFGFR2, suggesting that the opposing effects of each receptor on prostate tumors was likely mediated by downstream signaling pathways.
Based on a previous observation in PC12 cells between FGFR1 and FGFR3, in which FGFR1 signaling led to greater Erk activation (24)
, and the established importance of Erk in growth factor-induced proliferation, we reasoned that FGFR1 and FGFR2 may also differ with regard to Erk phosphorylation (Fig. 1C)
. Whereas both iFGFRs receptors were able to phosphorylate Erk within 5–15 min of CID administration, iFGFR1 signaling consistently led to an approximately 3-fold higher phosphorylation of Erk than iFGFR2 (Fig. 1D)
. Additionally, Erk phosphorylation after treatment with serum for 15 min was 2.5-fold higher in C2N-iFGFR1 cells than in C2N-iFGFR2 cells. When compared with parental C2N cells, iFGFR1 and iFGFR2 showed no increase in background Erk phosphorylation (Fig. 1E)
. Thus, differential signaling to Erk by iFGFR1 and iFGFR2 may mirror the opposing effects attributed to these receptors in CaP.
Effects of iFGFR1 and iFGFR2 on Cell Death and Proliferation in Vitro.
Numerous cellular effects have been attributed to the activation of FGFRs, including proliferation and survival. To determine the effects of iFGFR1 and iFGFR2 signaling on preventing cell death in vitro, we examined the ability of either iFGFR1 or iFGFR2 to protect cells from UV irradiation after serum starvation. The addition of either serum or 100 nM AP20187 led to a 2-fold increase in viable C2N-iFGFR1 cells after UV irradiation over untreated cells (Fig. 2A)
. In contrast, only serum could provide protection from UV for C2N-iFGFR2 cells. Thus, observed differences between iFGFR1 and iFGFR2 in their ability to protect cells from death after UV irradiation may underlie differences in tumor growth rates seen in vivo.
|
. Thus, no significant difference between iFGFR1 and iFGFR2 signaling on proliferation could be detected under in vitro growth conditions.
Influence of iFGFR1 on Tumor Initiation and Maintenance.
Due to the consistently observed ectopic expression of FGFR1 in human CaP and rodent CaP models, it was surprising that no major effects of FGFR1 signaling on proliferation and survival were observed in vitro. However, these in vitro studies may not adequately explore the contributions of the FGFR axis to CaP progression in vivo. To address this, we injected 2 x 106 C2N-iFGFR1.1 cells s.c. into the backs of syngeneic C57BL/6 mice. Two days later, we commenced biweekly i.p. injections of 50 µg of AP20187, unless otherwise stated. Tumor volumes were estimated regularly with Vernier calipers. To distinguish between signaling leading to tumor establishment (or "take") versus tumor maintenance, we used three distinct drug regimens. In the first treatment group, AP20187 was administered continuously throughout the experiment to show the cumulative effects of iFGFR1 on tumor take and progression. In the second group, AP20187 was administered for the first 24 days. In the third group, AP20187 treatment was initiated 17 days after cell injections, and when well-established tumors (
200 mm3 at 42 days) appeared, CID administration was stopped. Also, to assess the normal baseline growth characteristics of clonal line C2N.iFGFR1.1 in mice, a fourth CID-free control growth curve was established.
In three independent experiments, differences in tumor growth were consistently observed between the control group and those mice that received continuous AP20187 (Fig. 3A)
. After 50 days of continuous CID treatment, tumors reached an average size of 1600 mm3, whereas untreated tumors were 90 mm3 at the same time point. Moreover, early and continuous CID treatment always (31 of 31 mice) led to tumor establishment compared with only 60% (14 of 24 mice) in nontreated controls and 50% (5 of 10 mice) for parental C2N cells (Table 1
; data not shown), demonstrating that FGFR1 signaling not only promoted growth but also contributed to tumor establishment or the ability of the cells to progress to a palpable size. Delaying drug administration by 17 days did not alter the ultimate proportion of animals with tumors (100%), implying that the increased number of tumors observed after CID treatment was not due to greater early tumor survival but rather that some tumors required removal of a barrier to growth, possibly neoangiogenesis. Furthermore, when drug administration was delayed, growth was also consistently delayed in comparison with tumors continuously stimulated with AP20187.
|
|
. After this short latency period, the inability to resume rapid growth occurred only in the smallest tumors (data not shown), demonstrating that exogenous iFGFR1 signaling was not necessary for most TRAMP.C2N-derived tumors to grow.
Various explanations could account for escape from CID-dependent growth. Besides becoming completely independent of iFGFR1 signaling, tumors could theoretically escape AP20187 dependence by transgene amplification, leading to CID-independent iFGFR1 signaling. However, transgene immunoblotting revealed no consistent differences in average transgene expression between CID-treated and mock-treated tumors (Fig. 3B)
. Furthermore, when comparing mice that received drug continuously after tumor challenge with mice that received drug only at early time points, we observed only a slight delay in growth rates, reaching 1000 mm3 by days 48 and 50, respectively (Fig. 3A)
. This implies that iFGFR1 signaling is most critical for tumor take and early outgrowth but has diminished, if any, contribution to growth as tumors become relatively large and established.
To ascertain whether iFGFR1 signaling would be sufficient to drive tumor initiation and growth in nontumorigenic C1A cells, we injected C1A-iFGFR1.1 cells into the backs of mice and administered AP20187 starting 4 or 17 days later. In both CID-treated and untreated C1A-iFGFR1.1-injected mice, no tumors were detected up to 2 months after injection (Table 1)
. Thus, in the nontumorigenic TRAMP.C1A line, iFGFR1 signaling was not sufficient for tumorigenesis or tumor take, demonstrating that iFGFR1 signaling is not sufficient to establish tumor growth in otherwise nontumorigenic cells. However, iFGFR1 can greatly augment tumor growth in already tumorigenic cell lines.
Influence of iFGFR2 on Tumor Initiation.
In normal prostate epithelium, the FGFR2IIIb isoform of FGFR2 is widely expressed and responds to stromal-derived FGF-7 and FGF-10, leading to survival and differentiation of prostate epithelial cells. In previous studies, overexpression of FGFR2 kinase in Dunning rat-derived cell lines inhibited prostate tumor growth; however, these experiments could not rule out whether heterodimerization with endogenous FGFR1 (or other FGFRs) had occurred (5
, 6)
. Additionally, in the Dunning model, prostate adenocarcinoma cells are mixed with stromal cells that likely provide additional factors that contribute to differentiation or inhibit proliferation of the epithelial cells. Although C2N cells are tumorigenic, injection of 2 x 106 C2N-iFGFR2 cells did not lead to tumor growth under any drug treatment regimen (Table 1)
. Even when 4 x 106 cells were injected, no drug-dependent (or independent) growth occurred.
To rule out the formal possibility that a clone-specific loss of tumorigenicity was responsible for these observations, we repeated the C2N.iFGFR2 experiment with both a second clone and also a "pooled" heterogeneous line of C2N-iFGFR2 cells. Again, consistent with previous reports, in all three experiments, no measurable iFGFR2-expressing C2N tumors developed (Table 1)
. It is clear that low-level basal signaling due to iFGFR2 expression is sufficient to inhibit tumor establishment; however, due to the absence of tumor growth, this model system cannot be used to ascertain whether FGFR2 signaling can inhibit or reverse the growth of established tumors.
FGFR1 Signaling in Vivo Augments Proliferation.
Finally, we wanted to determine the underlying mechanism for the differential growth rates of iFGFR1-stimulated tumors. In the above-mentioned in vivo studies, we observed that once tumors achieved a certain size (500 mm3; Fig. 3A
), growth rates were indistinguishable regardless of iFGFR1 signaling. This was consistent with our immunohistochemistry of large (>1000 mm3) tumors excised at euthanasia, in which we saw no significant differences in proliferation (by Ki-67 staining), apoptosis (by terminal deoxynucleotidyltransferase-mediated nick end labeling assay), or vascularity (by CD31 signaling; data not shown). This contrasts with the result when CID-dependent iFGFR1 stimulation was terminated in relatively small (200 mm3) tumors (Fig. 3A)
, in which growth rates declined with AP20187 withdrawal. Therefore, to attribute these growth rate differences to differential proliferation or apoptosis rates, we repeated a simplified version of the above C2N-iFGFR1 tumor experiment. After injecting 2 x 106 C2N-iFGFR1.1 cells into 14 mice, we injected all mice with CID until tumors reached approximately 200 mm3 in size. Thereafter, half of the tumors were harvested. For the remaining animals, administration of AP20187 was terminated, and 6 days later, the remaining tumors were harvested. Unlike our results based on large, primarily CID-independent, tumors, loss of iFGFR1 signaling in small tumors led to a 2-fold reduction in proliferation rates based on Ki-67 staining (Fig. 4)
, but no changes in apoptosis (by terminal deoxynucleotidyltransferase-mediated nick end labeling assay) or vascularization (by CD31 staining) were detected (data not shown). Thus, whereas iFGFR signaling drives proliferation of moderately sized tumors (200 mm3), it does not appear to influence apoptosis and angiogenesis in those tumors.
|
when normalized to -tubulin levels (Fig. 5B)
. We also measure OPN protein levels by immunoblotting secreted OPN (Fig. 5C)
. At the protein level, we observed (by densitometry) a 4-fold increase in secreted OPN after iFGFR1 activation, whereas, in contrast, C2N-iFGFR2 cells had both lower basal levels of OPN, similar to parental cell lines (data not shown), and only very slight induction of OPN. Lending support to this finding, in a recent study, drug inhibition of FGFR1 and the Erk pathway blocked OPN expression in smooth muscle cells (28)
. Thus, in addition to demonstrating a dramatic difference in the signaling potential of FGFR1 versus FGFR2 in vivo at the cellular level, the CID system led to the identification of the tumor-associated OPN gene as a likely candidate for FGFR1-induced genes that can promote tumor growth.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Using CID technology, we elucidated a quantitative signaling difference between the two receptors. Although we observed comparable activation of iFGFR1 and iFGFR2 by autophosphorylation, activation of Erk was 3-fold higher by iFGFR1 when compared with iFGFR2. However, in comparison with C2N.iFGFR1 cells, we also observed 2-fold lower Erk phosphorylation in C2N.iFGFR2 cells after addition of serum, implying that ectopic expression of iFGFR2 impedes Erk signaling in C2N cells independently of CID. This may help explain the loss of tumorigenicity of multiple clones of C2N.iFGFR2 cells we observed in our studies. In independent studies of human salivary adenocarcinoma-derived cells, expression of full-length FGFR2 blocked tumor growth and basally inhibited endogenous FGFR phosphorylation of FRS2, the major upstream substrate for FGFR activation of the Erk cascade (29) . Although no mechanism was provided, an active suppression of FRS2 signaling due to FGFR2 expression is implied. Because FRS2 associates with FGFRs independently of receptor phosphorylation, overexpression of iFGFR2 may sequester FRS2 from endogenous FGFR1, thereby preventing Erk signaling.
The enhanced proliferation after iFGFR1 signaling in C2N.iFGFR1 cells in vivo versus in vitro suggests that FGFR1-induced genes may support mechanisms that can overcome barriers to tumor growth in vivo, such as neoangiogenesis, stromal independence, and anchorage independence. OPN, which we have shown to be differentially up-regulated by iFGFR1, has previously been shown to enhance anchorage independence (30) and facilitate tumor growth through increased proliferation (31) , whereas OPN-specific antisense transcripts can retard tumor growth (32) . Possibly OPN and/or other FGFR1-induced factors support the proliferation we observed in vivo.
Beyond 0.5 mm in size, tumors become hypoxic without additional vasculature (33)
. iFGFR1 signaling may help overcome this barrier to tumor establishment through proangiogenic signals, such as VEGF up-regulation or thrombospondin (TSP-1) inhibition (34)
, acting at a very early stage in tumor growth. This mechanism would not be expected to contribute to growth in culture where oxygen is abundant, consistent with our inability to observe changes in VEGF expression in C2N cells after iFGFR1 signaling in vitro (data not shown). Furthermore, this mechanism would be expected to be less important in already established well-vascularized tumors, consistent with our observations that iFGFR1 signaling leads to an increase in proliferation, but not to an increase in vasculature density in tumors 200 mm3 in size. Extrapolating from the effects of iFGFR1 signaling on tumors 200 mm3 in size and our additional observations based on transgenic mice expressing iFGFR1 in the prostatic epithelium, where vascularization occurs after a long period of hyperplasia,5
we conclude that neovascularization is secondary to proliferation in our model. A likely scenario is that iFGFR1 signaling stimulates proliferation, either via the Erk pathway or indirectly via OPN or other proteins, driving tumors growth beyond vascular capacity. This, in turn, would lead to hypoxia and the stabilization of hypoxia-inducible factor 1, a transcription factor that induces the expression of proangiogenic factors, such as VEGF. With sufficient paracrine signaling from hypoxic cells, rapidly dividing tumors would promote neoangiogenesis, allowing further growth (33)
. Although this would also occur within unmodified C2N-derived tumors, iFGFR1 signaling would augment this process, increasing both the rate of growth and the proportion of established tumors.
Although these experiments were conducted in the presence of endogenous FGFR1 signaling, additional ectopic iFGFR1 signaling also influenced tumor growth rates until the tumors were quite large (at least 200 mm3), vascularized, and well established. By blocking endogenous FGFR1 signals, an even greater role for FGFR1 signaling on tumor growth could be revealed. Additionally, the C2N cell line produces poorly differentiated tumors, with high genetic heterogeneity, potentially leading to more genetic resources for circumventing the loss of iFGFR1 signaling. Nevertheless, some tumors (20%) were still unable to overcome the loss of iFGFR signals, suggesting that therapies aimed at selectively blocking FGFR1 signaling might have dramatic consequences for a subset of prostate tumors.
As demonstrated here, CID technology can be a powerful tool in elucidating the roles of distinct signaling molecules in disease progression and for modeling CaP development. In a previous paper, we used iFGFR1 to create a mouse model with progressive transformation of the mammary gland (19) . As a second step, creation of transgenic mice expressing inducible signaling molecules in the prostate epithelium and breeding these conditional models with established CaP models (e.g., TRAMP) or to other cancer susceptibility models (e.g., p53, Rb deficient) will allow for a more accurate determination of the temporal, additive, and synergistic roles of many signaling factors for tumor development, progression, and dissemination. This should permit the creation of "susceptibility maps," where tumor grade is correlated with susceptibility to loss of an inducible signaling pathway, thereby providing information into when therapeutics for that pathway would be most useful.
These studies show that FGFR signaling is important for the earliest stages of tumor establishment; however, FGFR1 involvement in other stages of cancer progression cannot be excluded. Additionally, FGFR1 promotes in vivo proliferation and transcriptionally up-regulates the metastasis-associated protein OPN. Also, using iFGFRs, we have shown that quantitative differences exist in signaling between FGFR1 and FGFR2. We confirmed previous results by showing a positive contribution by FGFR1 signaling to prostate tumor growth and a negative effect by FGFR2 with the intracellular signaling domains of the receptors largely responsible for the opposing effects of FGFR1 and FGFR2 on tumor growth (5) . Finally, we have used these CID-inducible receptors to better understand FGFR signaling in tumor establishment and maintenance, while illuminating the importance of FGFR1 signaling as an excellent target for CaP intervention and ectopic overexpression of FGFR2 as a potential gene therapy.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by NIH Grants R01 CA87569 and UO1 CA84296 and NIH Training Grant T32-AI07495 (to K. W. F.). 
2 Present address: Department of Anatomy, University of California San Francisco, San Francisco, CA 94143.
3 To whom requests for reprints should be addressed, at Baylor College of Medicine, One Baylor Plaza/M929, Houston, TX 77030. E-mail: dspencer{at}bcm.tmc.edu
4 The abbreviations used are: CaP, prostate cancer; CID, chemically induced dimerization; TRAMP, transgenic adenocarcinoma of the murine prostate; FGF, fibroblast growth factor; FGFR, fibroblast growth factor receptor; iFGFR, inducible fibroblast growth factor receptor; OPN, osteopontin; VEGF, vascular endothelial growth factor; Erk, extracellular signal-regulated kinase; HRP, horseradish peroxidase; HA, hemagglutinin.
5 Kevin W. Freeman and David M. Spencer, unpublished results.
Received 11/ 6/02. Revised 6/23/03. Accepted 7/22/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
vß3 in prostate cancer progression. Neoplasia, 4: 191-194, 2002.[Medline]
B-mediated induction of membrane type 1 matrix metalloproteinase in murine melanoma cells. J. Biol. Chem., 276: 44926-44935, 2001.This article has been cited by other articles:
|
|
 |
|
  F. Yasuhara, G. R. O. Gomes, E. R. Siu, C. I. Suenaga, E. Marostica, C. S. Porto, and M. F. M. Lazari Effects of the Antiestrogen Fulvestrant (ICI 182,780) on Gene Expression of the Rat Efferent Ductules Biol Reprod, September 1, 2008; 79(3): 432 - 441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Deng, J. Yang, M. McCarty, and B. Su MEKK3 is required for endothelium function but is not essential for tumor growth and angiogenesis Am J Physiol Cell Physiol, October 1, 2007; 293(4): C1404 - C1411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Xian, K. L. Schwertfeger, and J. M. Rosen Distinct Roles of Fibroblast Growth Factor Receptor 1 and 2 in Regulating Cell Survival and Epithelial-Mesenchymal Transition Mol. Endocrinol., April 1, 2007; 21(4): 987 - 1000. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
 |
|
 L. Duchesne, B. Tissot, T. R. Rudd, A. Dell, and D. G. Fernig N-Glycosylation of Fibroblast Growth Factor Receptor 1 Regulates Ligand and Heparan Sulfate Co-receptor Binding J. Biol. Chem., September 15, 2006; 281(37): 27178 - 27189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Schwertfeger, W. Xian, A. M. Kaplan, S. H. Burnett, D. A. Cohen, and J. M. Rosen A Critical Role for the Inflammatory Response in a Mouse Model of Preneoplastic Progression Cancer Res., June 1, 2006; 66(11): 5676 - 5685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Zhong, G. Saribekyan, C.-P. Liao, M. B. Cohen, and P. Roy-Burman Cooperation between FGF8b Overexpression and PTEN Deficiency in Prostate Tumorigenesis Cancer Res., February 15, 2006; 66(4): 2188 - 2194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Cheng, L. Yu, D. Zhang, Q. Huang, D. Spencer, and B. Su Dimerization through the Catalytic Domain Is Essential for MEKK2 Activation J. Biol. Chem., April 8, 2005; 280(14): 13477 - 13482. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Cario, M. Stanulla, B. M. Fine, O. Teuffel, N. v. Neuhoff, A. Schrauder, T. Flohr, B. W. Schafer, C. R. Bartram, K. Welte, et al. Distinct gene expression profiles determine molecular treatment response in childhood acute lymphoblastic leukemia Blood, January 15, 2005; 105(2): 821 - 826. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B Kwabi-Addo, M Ozen, and M Ittmann The role of fibroblast growth factors and their receptors in prostate cancer Endocr. Relat. Cancer, December 1, 2004; 11(4): 709 - 724. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Shain Exogenous Fibroblast Growth Factors Maintain Viability, Promote Proliferation, and Suppress GADD45{alpha} and GAS6 Transcript Content of Prostate Cancer Cells Genetically Modified to Lack Endogenous FGF-2 Mol. Cancer Res., November 1, 2004; 2(11): 653 - 661. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Billottet, N. Elkhatib, J.-P. Thiery, and J. Jouanneau Targets of Fibroblast Growth Factor 1 (FGF-1) and FGF-2 Signaling Involved in the Invasive and Tumorigenic Behavior of Carcinoma Cells Mol. Biol. Cell, October 1, 2004; 15(10): 4725 - 4734. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. W. Freeman, B. E. Welm, R. D. Gangula, J. M. Rosen, M. Ittmann, N. M. Greenberg, and D. M. Spencer Inducible Prostate Intraepithelial Neoplasia with Reversible Hyperplasia in Conditional FGFR1-Expressing Mice Cancer Res., December 1, 2003; 63(23): 8256 - 8263. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
