| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Advances in Brief |
1 Lombardi Cancer Center, Georgetown University, Washington, DC, and
2 Department of Human Oncology, University of Wisconsin Medical School, Madison, Wisconsin
 |
ABSTRACT |
|---|
 Top ABSTRACT IntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
|
Introduction |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
We found that FGF-BP is expressed at high levels in the murine gut during embryonic development, down-regulated in the adult (4 , 5) , but expressed at high levels in some colon cancer tissues and cell lines (1) . To evaluate regulation of FGF-BP during colon carcinogenesis we initiated a series of studies with normal and pathological colon biopsies to determine at what stage of transformation the gene is up-regulated. Here we report that FGF-BP expression is highly up-regulated in dysplastic lesions, i.e., early on during colon carcinogenesis. These early lesions are associated with mutations in ß-catenin, and/or a loss of function of the APCtumor suppressor gene has been identified in >80% of sporadic colon carcinomas (6) . To assess the possible contribution of the loss of APC to FGF-BP up-regulation, we used a well-defined murine model, the B6 ApcMin/+ mouse, which carries one mutant APC allele and develops polyps on loss of the residual wild-type APC allele (7 , 8) . In this model we found that FGF-BP and ß-catenin expression was induced in polyps, as well as in a rare ACF, the earliest discernible stages of transformation (9) . Furthermore, cell culture studies show that increases in endogenous ß-catenin by treatment with LiCl result in a significant increase in FGF-BP mRNA levels, and cotransfection assays demonstrate transcriptional activation of the FGF-BP gene promoter by ß-catenin through T-cell factor (TCF) sites. We conclude that FGF-BP is a novel target gene of the Wnt/ß-catenin pathway.
|
Materials and Methods |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
Cell Culture, Transfections, and Reporter Assays.
The cell lines CaCo-2 (colon cancer), SKBR3, and MDA-MB468 (breast cancer) were from the American Type Culture Collection (Springfield, VA). The HCT-116 cell line with somatic cell knockout of the activated ß-catenin allele was provided by Dr. Todd Waldman (Georgetown University; Ref. 14
). The cells were maintained in DMEM with 10% FBS at 37°C and 5% CO2. Twenty-four h before transfection, cells were seeded in 12-well plates at a density of 1 x 105 cells/well in DMEM +10% FBS. With CaCo-2 cells, for each well 0.75 µg of DNA constructs and 10 µl of LipofectAMINE reagent were combined with 200 µl of OPTI-MEM1 (Life Technologies, Inc.) and incubated for 30 min at room temperature. Appropriate amounts of OPTI-MEM1 were added to the solution to bring the volume to 1 ml, and the mixture was placed on cells for 3 h at 37°C. The cells were then washed twice with Iscove’s Modified Medium (IMEM, Life Technologies, Inc.) and incubated in DMEM +10% FBS for 18 h. For SKBR3 cells, DNA constructs were mixed in a 1:2 ratio with FuGENE (Roche) reagent in serum-free IMEM and incubated at room temperature for 30 min. The FuGENE/DNA solution was then added to the cells with DMEM +10% FBS medium and incubated at 37°C for 18 h. Transfection efficiency was determined by cotransfection with 4.0 ng of the Renilla luciferase reporter vector pRL-CMV (Promega). After the 18-h incubation, cells were lysed in 100 µl of Passive Lysis buffer (Promega). Ten µl of the cell extract was assayed for firefly and Renilla luciferase activity with the Dual-Luciferase Reporter assay system (Promega). FGF-BP promoter constructs (-1060/+62, -118/+62, -93/+62, and -77/+62) were cloned into the pGL3 vector as described previously (15)
. Two consensus TCF sites (5'-A/T A/T CAAAG-3') located at -1030 and at -545 were deleted by PCR-based mutagenesis using the following primers: forward primers 1030-del-F: 5'-CAA ATG TCT GTT TAT ACA ACT TAA GAC CC-3' and 545-del-F: 5'-CAG TCACCC ATT CAT TTA TTG AGA GTG G-3'; reverse primers 1030-del-R: 5'-GGG TCT TAA GTT GTA TAA ACA GAC ATT TG-3' and 545-del-R 5-CCA CTC TCA ATA AAT GAA TGG GTG ACT G-3'. All of the constructs were sequenced to confirm mutations. SKBR3 cells were transfected with 100 ng of luciferase cDNA, 0.1 ng Renilla, and 300 ng of pCDNA3 or 300 ng pCDNA3-ß-catenin using Fugene 6 transfection reagent (Roche). Experiments were typically performed in duplicate and repeated as indicated in the legends to the figures. ß-Catenin, E-cadherin, and Topflash expression constructs were described previously (16)
. The pcDNA3 cloning vector was purchased from Invitrogen.
Northern Analysis.
MDA-MB468 cells were plated in 10-cm dishes and grown to 70% confluence in IMEM +10% FBS 24 h before treatment. The cells were treated with LiCl (30 mM) + inositol or with NaCl (30 mM) + inositol dissolved in 10 ml of IMEM +10% FBS. Sixteen h after initiation of treatment total RNA was isolated using the RNA STAT-60 protocol (RNA STAT-60; Tel-Test, Friendswood, TX). Thirty µg of total RNA were run on a 1.2% formaldehyde-agarose gel. Blotting and hybridization with the human FGF-BP probe were performed as described previously (15)
.
Data Analysis.
The Prism/GraphPad program was used for data analysis. Ps < 0.05 were considered significant.
|
Results |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
showed that FGF-BP expression was detectable in 12 of 76 histologically normal appearing samples. In contrast with this, a high portion of the samples with moderate to severe dysplasia expressed FGF-BP (62 of 85; P < 0.0001 normal versus dysplastic mucosa; 
2 test). Interestingly, even individual dysplastic crypts (closed arrow in Fig. 1A
) located within otherwise normal mucosa (open arrow in Fig. 1
) show expression of FGF-BP. It is of note that the increased expression of FGF-BP in dysplastic lesions coincides with a significant increase in blood vessel density from 80 ± 7 to 154 ± 9 vessels/field as measured by CD31 staining in the lamina propria (normal mucosa versus severe dysplasia; P < 0.001, t test).
|
2). Because inflammatory bowel disease does not induce FGF-BP expression, we hypothesized that the up-regulation of FGF-BP at the onset of colon epithelial malignant transformation may be due to an early genetic event, such as loss of the APC tumor suppressor function associated with the initiation of dysplasia. To address this question, we used the APC heterozygous B6 ApcMin/+ mouse (17)
as an animal model system.
FGF-BP Expression in B6 ApcMin/+ Mouse Adenoma Coincides with Cellular Relocation and Increase in ß-Catenin Protein.
As a first step, we compared FGF-BP expression in the intestines of wild-type C57BL/6J mice relative to that in B6 ApcMin/+ mice. We found no differences in baseline expression of FGF-BP (data not shown). This suggests that the loss of function of one allele in the B6 ApcMin/+ mice is not sufficient to alter the signal toward FGF-BP expression. In the normal epithelium ß-catenin is sequestered at the membrane and is rarely found in the cytoplasm or nucleus (Ref. 18
; see Fig. 2J
). When APC becomes defective in intestinal crypt cells of B6 ApcMin/+ mice, regulation of ß-catenin is lost (17)
, and the epithelium progresses to early stages of malignancy. We used the accumulation of cytoplasmic and nuclear ß-catenin in microscopic sections as a read-out for the loss of APC function, and probed serial sections of normal and adenoma tissues for both FGF-BP and ß-catenin expression.
|
, darker H&E staining) showed an elevation of the ß-catenin protein (Fig. 2, B and F
, brown stain) and of FGF-BP mRNA (Fig. 2, C and G
, dark blue stain). Fig. 2, D and H
, show negative controls for FGF-BP detection. In a survey of tissues we found that 21 of 27 adenomas were strongly positive for FGF-BP (>30% of the adenoma surface area; Fig. 2, C and G
). Adjacent normal intestine was also examined for FGF-BP expression, and only 5 of 19 samples showed any expression of FGF-BP (P < 0.001, normal versus adenomas, 2; Fig. 2K
). On careful inspection of serial sections, we found a striking coincidence of expression of FGF-BP mRNA (Fig. 2, C and G)
and of ß-catenin protein (Fig. 2, B and F)
in the same tissue areas. We also searched tissues for the earliest microscopically discernible stage of dysplasia, ACF, that is observed at a rate of 0.14 per mouse (19)
. Although we observed only a single ACF, there was coincidence of expression of the two genes in cells that formed this ACF in the midst of normal mucosal areas (Fig. 2, I and J)
. In addition to the tissues derived from the ApcMin/+ model, we also examined sections from intestinal polyps in a dextran sulfate-induced model of inflammatory colon disease (20) . No increase in FGF-BP expression was observed in this model (data not shown). This finding corroborates the lack of expression of FGF-BP in human clinical inflammatory bowel disease of different stages (see above). We concluded from these studies that FGF-BP expression is induced during the initiation of malignancy, and we hypothesized that this could occur is a result of the activation of the Wnt/ß-catenin pathway.
Lithium Induces Endogenous FGF-BP mRNA Expression.
To determine whether ß-catenin is directly involved in the regulation of the FGF-BP gene, we examined whether lithium-induced ß-catenin stabilization affects the levels of endogenous FGF-BP mRNA. Lithium inhibits glycogen synthase kinase-3ß, a negative regulator of ß-catenin (21)
. For the experiments we used MDA-MB468 breast cancer cells because they express detectable FGF-BP and show intact ß-catenin regulation. The MDA-MB468 cells were treated for 16 h with LiCl and inositol, which prevents inositol 1,4,5-triphosphate depletion by LiCl (21)
. Cells treated with LiCl and inositol increased ß-catenin protein levels and showed a 3-fold induction of FGF-BP mRNA as compared with control treatment (NaCl+inositol; Fig. 3
). The NaCl+inositol control showed no significant effect on basal FGF-BP mRNA expression (data not shown). Thus, increasing the level of free ß-catenin coincides with induction of endogenous FGF-BP mRNA, lending additional support to the notion of ß-catenin as a regulator of FGF-BP expression.
|
. This increase in promoter activity in CaCo-2 cells is comparable with ß-catenin induction of the TopFlash promoter, a known ß-catenin sensitive promoter containing multimerized TCF sites (data not shown).
|
. To determine whether this induction is a ß-catenin-specific effect, we cotransfected E-cadherin, an adhesion molecule that binds to the internal armadillo repeats of ß-catenin and functions as a dominant-negative regulator of ß-catenin by preventing its translocation to the nucleus. Cotransfection with E-cadherin reversed ß-catenin induction of the FGF-BP promoter (Fig. 4B)
. We also found that FGF-BP promoter activity of the full-length promoter -1060/+62 was reduced in isogenic HCT-116 colon cancer cells with somatic cell knockout of the activated ß-catenin allele (14)
as compared with promoter activity in wild-type HCT-116 cells (Fig. 4D)
. Taken together with the results of the ß-catenin effects on endogenous levels of FGF-BP mRNA, these data indicate that ß-catenin is a transcriptional regulator of the FGF-BP gene.
ß-Catenin Regulatory Region in the FGF-BP Promoter.
To identify the regions necessary for regulation of the FGF-BP promoter by ß-catenin, we transfected SKBR3 cells with 5' deletion constructs of the FGF-BP promoter/reporter constructs (15)
. ß-Catenin had a minimal background effect (<2-fold) on luciferase activity of the pGL3-basic empty vector (Fig. 4B)
, similar to nonspecific background effects that we observed previously with this vector (15)
. Deletion from -1060 to -118 reduced the ß-catenin induced promoter activity by >70%. An additional deletion to -93 had no effect on the induction of the promoter by ß-catenin, but deletion to -77 negated all of the ß-catenin induction of the promoter to background levels of the pGL3 vector (Fig. 4B)
. The experiments with the FGF-BP promoter/reporter constructs in the HCT-116 knockout cells, which have their activated ß-catenin allele deleted (14)
, showed a significant reduction of constitutive promoter activity of the full-length construct. Constitutive activity of the -118/+62 construct, however, was not altered by the deletion of the activated ß-catenin allele (Fig. 4D)
. This finding complements the different inducibility of the activity of these constructs by transfection of exogenous ß-catenin in the SKBR3 cells (see Fig. 4B
).
We found that the FGF-BP promoter contains two TCF consensus binding sites at -545 and -1030, and deleted these sites by PCR. Interestingly, only deletion of the distal TCF site at -1030 reduced ß-catenin induction of promoter activity, whereas deletion of the proximal site at -545 had no significant effect (Fig. 4C)
. We conclude from these results that ß-catenin induction of FGF-BP promoter activity involves regulatory regions in the distal promoter.
|
Discussion |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
Our previous analysis of FGF-BP expression in the developing mouse gut had shown that epithelial cells positioned at the bottom of the crypts express FGF-BP and that this expression is lost in cells maturing along the crypt/villus axis (4)
. More recently, positioning of epithelial cells along the crypt/villous axis and imposition of a crypt precursor phenotype was found associated with a gradient of ß-catenin/TCF activity that shows its maximum at the bottom of the crypts and is reduced as cells differentiate during their migration up the crypt (22
, 23) . It is likely that the FGF-BP expression that we observed in histologically normal tissues represents staining of sections of the lower third of crypts and, hence, the region with high ß-catenin activity. Also, FGF-BP expression may indicate a very early stage in the transition to dysplasia that is not yet manifest from the H&E staining. Fig. 1, A
and B, shows such an example of dysplastic lesions with surrounding normal mucosa. Interestingly, the histologically normal crypts that do not express FGF-BP show some staining for the protein in sections that transverse the bottom of the crypt as indicated by the narrow opening of the crypt (compare the two crypts indicated by open arrows in Fig. 1A
). With respect to distinct pathological alterations, the lack of FGF-BP expression in inflammatory human bowel disease and in the rodent animal model equivalent (20)
suggest that inflammatory pathways in the colon do not lead to an up-regulation of FGF-BP either on their own or through cross-talk with the ß-catenin signaling.
FGF-BP is an activator of growth factors in the FGF-family, and our studies lend support to the idea that the FGF family plays a role in the development of the early angiogenic phenotype in colon cancer. The induction of the angiogenic phenotype in colon cancer is a multifaceted process requiring the cooperation of numerous factors during the different steps of malignant progression. In a study of levels of FGF-2 and VEGF in serum samples from colon cancer patients, it was suggested that FGF-2 may act as an early inducer of the angiogenic phenotype (24) . FGF-BP up-regulation in intestinal adenomas may indeed trigger this by providing the chaperone that can release the immobilized FGF-2. In support of this notion, we found increased angiogenesis coincident with FGF-BP expression in human colon dysplastic lesions. Other factors, such as VEGF, probably cooperate with FGF-2 to maintain the process of angiogenesis throughout the stages of tumor formation. VEGF expression is found in adenomas of the colon; however, unlike our findings with FGF-BP, increased VEGF expression levels are correlated with later stages of the disease, and VEGF expression is increased in carcinomas as compared with adenomas as well as in metastatic versus nonmetastatic colon cancer (25) .
Not only does FGF-BP appear to be a novel proangiogenic target of ß-catenin that is up-regulated at an early stage of premalignant lesions, it seems that its regulation is through areas of the FGF-BP promoter that are not required for basal growth factor or TPA regulation of this gene in either squamous cell carcinoma or breast cancer cells (e.g., see Refs. 15
, 26 ). In fact, the regulation by EGF and TPA in these cell types involves an activator protein, a CAAT/enhancer binding protein ß site, and an E-box repressor that are all situated downstream of -118 and that are activated predominantly through the p38/mitogen-activated protein kinase pathway. Therefore, an unexpected result was the involvement of the region between -1060 and -118 in the ß-catenin regulation of FGF-BP. This also indicates that ß-catenin is not activating the promoter via indirect activation of the mitogen-activated protein kinase signaling pathways. Examination of the 1-kb promoter region between -1060 and -118 for possible transcription factor recognition sites using Transfac analysis revealed two potential TCF/lymphoid enhancer factor (LEF) sites, which are known to be involved in ß-catenin regulated gene transcription. Deletion of these sites showed that only the -1030 site contributes to ß-catenin-induced promoter activation, whereas the site at -545 does not. Interestingly, there is also a perfect consensus site for REL/nuclear factor 
B present in this region, which is of interest because ß-catenin can interact directly with nuclear factor B and might, thus, contribute in addition to regulation of the gene promoter (27)
. The second surprise of the FGF-BP gene promoter analysis was that the region between -93 and -77 was also required for full ß-catenin induction of the promoter. We have demonstrated previously that this region harbors an SP1 binding site and can bind SP1 specifically. However, this site is not required for growth factor or TPA-regulated gene transcription (15)
. It remains to be determined whether this site acts cooperatively with upstream regulatory factors in the ß-catenin induction of FGF-BP gene transcription. The Sp1 and Krueppel-like factors that bind to GC boxes are known to play a role in cell growth and tumor progression. However, their role in early events in colon carcinogenesis has not been defined.
In conclusion, ß-catenin, one the most significant oncogenic proteins in colon cancer, has been implicated in several key steps of the path to malignancy. The cell cycle regulatory genes, c-Myc and cyclin-D1, have both been identified as targets of ß-catenin. These two proteins were also found overexpressed in intestinal adenomas of the B6 ApcMin/+ mouse. Furthermore, ß-catenin activates matrix metalloproteinase 7, an enzyme that plays a role in invasion and metastasis. Additionally, APC and E-cadherin, two proteins that are closely tied to ß-catenin function, are important for induction of apoptosis and cell-cell adhesion, respectively. Our identification of FGF-BP as a direct target of ß-catenin transcriptional activation suggests that ß-catenin can also play a role in promoting the switch to the angiogenic phenotype observed early in the malignant progression of colon cancer.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: Anton Wellstein, Lombardi Cancer Center, Georgetown University, Washington, DC 20057. Phone: (202) 687-3672; Fax: (202) 687-4821; E-mail: wellstea{at}georgetown.edu
3 The abbreviations used are: FGF-BP, fibroblast growth factor binding protein; APC, adenomatous polyposis coli; FBS, fetal bovine serum; ACF, aberrant crypt focus; VEGF, vascular endothelial growth factor; TPA, 12-O-tetradecanoylphorbol-13-acetate. 
Received 6/29/03. Revised 10/16/03. Accepted 10/21/03.
|
REFERENCES |
|---|
|
TopABSTRACTIntroductionMaterials and MethodsResultsDiscussionREFERENCES |
|---|
B activity and intestinal inflammation in dextran sulphate sodium (DSS)-induced colitis in mice is suppressed by gliotoxin. Clin. Exp. Immunol., 120: 59-65, 2000.[Medline]
B in human colon and breast cancer. Cancer Cell, 2: 323-334, 2002.[Medline]This article has been cited by other articles:
|
|
 |
|
  K. N. Nejak-Bowen, G. Zeng, X. Tan, B. Cieply, and S. P. Monga {beta}-Catenin Regulates Vitamin C Biosynthesis and Cell Survival in Murine Liver J. Biol. Chem., October 9, 2009; 284(41): 28115 - 28127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Gibby, K. McDonnell, M. O. Schmidt, and A. Wellstein A distinct role for secreted fibroblast growth factor-binding proteins in development PNAS, May 26, 2009; 106(21): 8585 - 8590. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Zhang, Y. Chen, M. R. Swift, E. Tassi, D. C. Stylianou, K. A. Gibby, A. T. Riegel, and A. Wellstein Effect of FGF-binding Protein 3 on Vascular Permeability J. Biol. Chem., October 17, 2008; 283(42): 28329 - 28337. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Sonvilla, S. Allerstorfer, S. Stattner, J. Karner, M. Klimpfinger, H. Fischer, B. Grasl-Kraupp, K. Holzmann, W. Berger, F. Wrba, et al. FGF18 in colorectal tumour cells: autocrine and paracrine effects Carcinogenesis, January 1, 2008; 29(1): 15 - 24. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Newman, E. Walsh, K. B. C. Apparao, and P. L. Sannes Fibroblast growth factor-binding protein and N-deacetylase/N-sulfotransferase-1 expression in type II cells is modulated by heparin and extracellular matrix Am J Physiol Lung Cell Mol Physiol, November 1, 2007; 293(5): L1314 - L1320. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Tassi, S. Walter, A. Aigner, R. H. Cabal-Manzano, R. Ray, P. J. Reier, and A. Wellstein Effects on neurite outgrowth and cell survival of a secreted fibroblast growth factor binding protein upregulated during spinal cord injury Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2007; 293(2): R775 - R783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. D. Hendrix, R. Wu, R. Kuick, D. R. Schwartz, E. R. Fearon, and K. R. Cho Fibroblast Growth Factor 9 Has Oncogenic Activity and Is a Downstream Target of Wnt Signaling in Ovarian Endometrioid Adenocarcinomas Cancer Res., February 1, 2006; 66(3): 1354 - 1362. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Tassi, R. T. Henke, E. T. Bowden, M. R. Swift, D. P. Kodack, A. H. Kuo, A. Maitra, and A. Wellstein Expression of a Fibroblast Growth Factor-Binding Protein during the Development of Adenocarcinoma of the Pancreas and Colon Cancer Res., January 15, 2006; 66(2): 1191 - 1198. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Xie, E. Tassi, M. R. Swift, K. McDonnell, E. T. Bowden, S. Wang, Y. Ueda, Y. Tomita, A. T. Riegel, and A. Wellstein Identification of the Fibroblast Growth Factor (FGF)-interacting Domain in a Secreted FGF-binding Protein by Phage Display J. Biol. Chem., January 13, 2006; 281(2): 1137 - 1144. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Dihlmann, M. Kloor, C. Fallsehr, and M. von Knebel Doeberitz Regulation of AKT1 expression by beta-catenin/Tcf/Lef signaling in colorectal cancer cells Carcinogenesis, September 1, 2005; 26(9): 1503 - 1512. [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 |
-Tubulin is a loading control. Treatment was for 16 h. Duplicate samples were loaded. The experiment was repeated three times. B, levels of FGF-BP mRNA were quantified by phosphoimaging and corrected for loading by glyceraldehyde-3-phosphate dehydrogenase mRNA levels. **, P < 0.01 relative to NaCl treatment.