| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cell Growth & Differentiation |
-Catenin Mutations, but not E-Cadherin Inactivation, Underlie T-Cell Factor/Lymphoid Enhancer Factor Transcriptional Deregulation in Gastric and Pancreatic Cancer1
Departments of Internal Medicine [K. C., F. T. K., X. J., M. H., J-m. Q., E. R. F.], Surgery [A. Y.], Human Genetics [E. R. F.], and Pathology [E. R. F.], The Cancer Center and the Division of Molecular Medicine and Genetics, University of Michigan Medical School, Ann Arbor, Michigan 48109; and Department of Pathology, Yale University School of Medicine, New Haven, Connecticut 06510 [D. L. R., J. C.]
 |
Abstract |
|---|
 TOP Abstract IntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
-cat, a potential phosphorylation site in this ß-cat-related protein. Although E-cad is an important binding partner for ß-cat and -cat, E-cad inactivation did not result in CTTA. The ß-cat and -cat mutant proteins identified in our studies strongly activated Tcf transcription in vitro, whereas ß-cat mutant proteins with large NH2-terminal deletions had only modest effects on Tcf. Our results suggest a role for Tcf deregulation in gastric and pancreatic cancer, resulting from ß-cat and -cat mutations in some cases and, in others, from yet to be defined defects. Furthermore, these data imply that the consequences of APC and ß-cat mutations are distinct from the effects of E-cad inactivation. |
Introduction |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
70% of colorectal adenomas and carcinomas but are rare in other cancers (reviewed in Ref. 1
). The APC protein has been found to bind to a number of proteins, including ß-cat (2
, 3)
, -cat (also known as plakoglobin; Refs. 4
and 5
), GSK3ß (6)
, EB1 (7)
, hDLG (8)
, microtubules (9
, 10)
, and the related proteins axin and conductin (11, 12, 13, 14)
. With the exception of ß-cat, GSK3ß, and the conductin and axin proteins, the significance and role of APC’s interactions with its various binding partners is not well understood.
ß-cat, initially identified as a coprecipitating protein with E-cad, was subsequently shown to link E-cad to 
-cat, a vinculin-like protein, which, in turn, links the E-cad/cat cell-cell adhesion complex to the cortical cytoskeleton (15, 16, 17, 18, 19)
. Molecular cloning revealed that ß-cat is a member of the armadillo (arm) family. The prototype molecule Arm functions in the wingless signaling pathway in Drosophila, with ß-cat functioning in the conserved wnt pathway in vertebrates (12
, 15) . The -cat protein is closely related to ß-cat and appears to have similar functions in cell-cell adhesion and the wnt pathway (16, 17, 18, 19)
.
The identification of Tcf/Lef transcriptional factor proteins as downstream targets of ß-cat (20, 21, 22) , together with the observation of CTTA in colorectal cancer cell lines with APC defects (23) , supported the hypothesis that APC had a critical role in regulating ß-cat (24) . Further support was provided by data, indicating that a subset of the colorectal cancers lacking APC mutations had CTTA as a result of mutations altering potential GSK3ß phosphorylation sites in the ß-cat NH2 terminus (25) . Other studies have reported mutations of the NH2-terminal phosphorylation sites of ß-cat in melanoma cell lines (26) and medulloblastoma (27) as well as carcinomas of the ovary (28) , endometrium (29) , liver (30) , and prostate (31) . The mutant ß-cat proteins that are expressed are presumed to be oncogenic because they cannot be appropriately regulated by the protein complex containing APC, GSK3ß, and axin or conductin (11, 12, 13, 14, 15) . Mutant ß-cat accumulates in the cell and deregulates expression of downstream Tcf-/Lef-regulated target genes, including perhaps c-myc (32) , via ß-cat’s role as a coactivator of Tcf/Lef transcription.
As noted above, in addition to their role in Wnt signaling, ß-cat and -cat are key components in E-cad cell-cell adhesion, a function that is critical in development, cell differentiation, and maintenance of tissue architecture (15, 16, 17, 18, 19)
. In many epithelial cancer types, loss of E-cad expression has been frequently observed (18)
. The mechanisms underlying loss of E-cad are rather poorly understood in the majority of cases. However, in some cancers, including diffuse-type gastric carcinoma, somatic mutations in the E-cad gene (CDH1) underlie loss of its expression (18
, 33
, 34)
. CDH1 germ-line mutations have recently been found in several families with inherited predisposition to gastric cancer (35
, 36)
, further highlighting the significance of E-cad alterations in GI cancer. In contrast to the extensive body of data in colorectal cancer, there have been only limited reports of APC mutations and no reports of CTNNB1 or CTNNG1 mutations in tumors of the upper GI tract or pancreas (37, 38, 39, 40, 41)
. Decreased expression of E-cad and/or ß-cat, however, has been seen in a significant fraction of esophageal, gastric, and pancreatic cancers, and altered expression is correlated with high grade or advanced stage (42, 43, 44, 45)
.
To explore defects in the APC/ß-cat/Tcf and E-cad/cat pathways in cancers of the GI tract further, we examined gastric and pancreatic cancer cell lines for CTTA, using Tcf reporter gene constructs. Our studies indicate the presence of Tcf deregulation in gastric and pancreatic cancer, resulting from CTNNB1 or CTNNG1 mutations in some cases and from yet to be defined defects in other cases. Our findings also imply that the consequences of APC or catenin mutations in cancer development are distinct from those that are attributable to E-cad inactivation.
|
Results |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
. We then assessed Tcf transcriptional activity in four gastric and eight pancreatic cancer lines. The majority of the lines displayed no evidence of CTTA, with mean relative Tcf activities between 0.25 and 2.0. However, CTTA was found in two gastric (NCI-N87 and AGS) and two pancreatic (ASPC and HS766T) lines, with mean relative Tcf activities ranging from 5 in the NCI-N87 line to 50 in the ASPC line (Fig. 1)
. The specific Tcf/Lef factor(s) responsible for the observed Tcf reporter gene activity are not known, although RT-PCR studies indicated that Tcf-4 was expressed in all cell lines studied (data not shown).
|
-cat Mutations in Gastric and Pancreatic Cancers.
. Although CTNNG1 mutations have not been previously reported in cancer, given the very high sequence conservation between the -cat and ß-cat proteins, we analyzed the sequences encoding the NH2-terminal region of -cat in all 12 gastric and pancreatic cell lines. In NCI-N87, we found a missense mutation at CTNNG1 codon 28, resulting in a serine-to-leucine substitution and alteration of a potential phosphorylation site (Fig. 2)
. No CTNNB1 or CTNNG1 mutations were found in ASPC or the eight lines that failed to manifest CTTA. Sequencing studies of exons 2–4 of CTNNB1 and the corresponding region of CTNNG1 in 36 primary gastric cancers, using genomic DNA microdissected from regions enriched in neoplastic cells, failed to identify any CTNNB1 or CTNNG1 mutations, suggesting that mutations in ß-cat and -cat are likely to be present in only a minority of gastric cancers.
|
. Tcf activity was unaffected by APC in HCT-116 colorectal cancer cells containing a mutant CTNNB1 allele (Fig. 3)
. Although Tcf transcriptional activity was only subtly elevated in ASPC, the fact that Tcf activity could not be suppressed by overexpression of wild-type APC (Fig. 3)
implies that CTTA in ASPC is not attributable to APC inactivation. Thus, defects besides those in the APC, CTNNB1, or CTNNG1 genes appear to be associated with CTTA in some cancers.
|
), we found that overexpression of APC decreased Tcf activity in a cell line with mutant -cat (i.e., NCI-N87; Fig. 3
). The ability of APC to suppress Tcf activity in a cell line expressing a mutant -cat protein is consistent with the proposal that mutant -cat may affect Tcf activity indirectly. Perhaps mutant -cat protein interferes with the degradation machinery regulating ß-cat protein levels, causing ß-cat protein levels to rise, with resultant Tcf activation. Support for this potential mechanism of Tcf activation by mutant -cat is provided by the studies of Miller and Moon (46)
, who demonstrated that expression of ectopic (aberrantly localized) ß-cat and -cat proteins mimicked Wnt pathway activation in Xenopus.
Mutant ß-cat and -cat Proteins Activate Tcf Transcriptional Activity.
Mutant ß-cat proteins with missense mutations or in-frame deletions of single amino acids in the NH2-terminal phosphorylation sites have been shown previously to strongly activate Tcf transcriptional activity when expressed in heterologous cell types, such as the 293 kidney cell line (21)
. We sought to confirm that the mutant CTNNB1 and CTNNG1 alleles that we had identified did, in fact, encode proteins that activated Tcf transcription. Consistent with prior results, a mutant CTNNB1 allele encoding a ß-cat protein with the codon 33 serine-to-tyrosine substitution found in the SW48 colorectal cancer line strongly activated Tcf-mediated transcription in 293 cells, whereas wild-type ß-cat had readily detectable but more modest effects (Fig. 4A)
. We found an in-frame deletion of serine codon 45 (
S45) in the HS766T pancreatic line; this mutation had previously been found in the HCT-116 colorectal line. The S45 mutant CTNNB1 allele strongly activated Tcf transcription, as did the G34E mutant allele from the AGS gastric cancer line (Fig. 4A)
.
|
-cat protein was 2-fold more potent than wild-type -cat (Fig. 4)
. The differing strength of the mutant ß-cat and -cat proteins in our assay may reflect true differences in their in vivo function, or it may be attributable, at least in part, to the heterologous 293 cell line used for the assay. Nevertheless, the data demonstrating the S28L mutant CTNNG1 allele had less activity than the CTNNB1 alleles with point mutations and small deletions (i.e., S33Y, S45, and G34E) are consistent with the observation that the NCI-N87 line (i.e., the line from which the S28L CTNNG1 allele was derived) had less CTTA relative to lines with mutant ß-cat (e.g., AGS and HS766T; see Fig. 1
).
In addition to missense mutations and single amino acid deletions in the NH2-terminal phosphorylation sites of ß-cat, larger in-frame deletions of ß-cat NH2-terminal sequences have been reported in a few cancer cell lines and primary tumors (22
, 26
, 47)
. The deletions are presumed to promote cancer growth through the same mechanisms as do mutant ß-cat alleles with localized mutations. We found, however, that mutant ß-cat proteins with substantial NH2-terminal deletions, such as deletion of ß-cat amino acids 1–90 (N90) or deletion of amino acids 5–80 (ex3), had activities in the Tcf assay that were essentially indistinguishable from those of wild-type ß-cat (Fig. 4A)
. We confirmed expression of Flag epitope-tagged wild-type ß-cat and selected mutant ß-cat proteins in 293 cells. Like the S33Y ß-cat protein, the N90 ß-cat protein accumulated to higher levels than did wild type ß-cat (Fig. 4B)
. Despite being expressed from the same vector as the ß-cat proteins, wild-type and mutant -cat proteins were rather poorly expressed in 293 cells (Fig. 4B)
.
On the basis of our data, at least two explanations for the reduced transcriptional activation potential of NH2-terminal truncated forms of ß-cat can be considered. (a) It is entirely possible that the in vitro assay of Tcf transcriptional activity in 293 cells does not accurately reflect the in vivo activity of all mutant ß-cat proteins. (b) A second, albeit less likely, explanation is that the assay does, in fact, reflect the ability of mutant ß-cat proteins to activate Tcf transcription, but ß-cat proteins with larger NH2-terminal deletions contribute to cancer through mechanisms that are independent of Tcf activation. Our rationale for favoring the former explanation is that mutant ß-cat proteins with substantial deletions of the NH2 terminus were previously shown to activate Tcf transcription in certain cell types (48) .
Loss of Endogenous E-cad Expression Does Not Lead to CTTA.
In normal cells, ß-cat is a highly abundant protein, complexed primarily with E-cad at the plasma membrane. As reviewed above, E-cad expression is often very reduced or absent in many epithelial cancers, including gastric cancer. To examine the possibility that E-cad inactivation might lead to increased levels of ß-cat and CTTA, we first examined the expression of E-cad and then that of -catenin, ß-cat, -cat, and GSK3ß in the four gastric and eight pancreatic cancer cell lines. Three of the four gastric cancer lines and one pancreatic cancer line had very reduced or absent E-cad expression (Fig. 5)
. Comparison of the pattern of E-cad expression in the lines (Fig. 5)
to their Tcf transcriptional activity (Fig. 1)
revealed no clear association. Specifically, of the three gastric cancer lines that had loss endogenous E-cad expression, CTTA was found only in the AGS line, in which a mutant ß-cat protein was also present. No evidence of CTTA was found in the SNU-5 or RF-1 lines. Similarly, despite the fact that the MIA PaCa-2 pancreatic line lacks endogenous E-cad expression, CTTA was not seen. Comparison of the levels of ß-cat and -cat in the four gastric cancer lines revealed modest elevations of ß-cat in AGS and -cat in NCI-N87 (AGS has a ß-cat mutation and NCI-N87 has a -cat mutation; Fig. 5
). However, no clear change in ß-cat levels was seen in the pancreatic line with a ß-cat mutation (HS766T; Fig. 5
). Our failure to find a strong correlation between CTNNB1 and CTNNG1 mutations and the relative abundance of the respective protein products may be largely attributable to the fact that we assessed total levels of ß-cat and -cat in the cells, rather than "free" cytosolic pools. The cell lines expressed roughly equivalent levels of GSK3ß, with only modest changes in abundance observed (Fig. 5)
.
|
-cat proteins were intact in breast cancer, we assessed the ability of wild-type and mutated forms of ß-cat and -cat to activate Tcf transcription in BT549 breast cancer cells, a line lacking E-cad expression. As shown in Fig. 6
, mutated ß-cat and -cat proteins were able to stimulate Tcf transcriptional activation more potently than their wild-type counterparts. The stronger activation of Tcf by the S28L mutant -cat in BT549 cells (Fig. 6)
versus 293 cells (Fig. 4A)
may reflect increased stability of -cat in BT549 cells. Overall, our results indicate that, despite the fact that Tcf/Lef family members are expressed in breast cancer cells, deregulation of Tcf transcription is a rare event in breast cancer. Furthermore, the data establish that E-cad inactivation does not lead to Tcf transcriptional deregulation. Although E-cad levels may modulate ß-cat levels, loss of endogenous E-cad expression is not sufficient to activate Tcf transcription constitutively.
|
|
Discussion |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
The studies presented here were undertaken with the intent of further defining the role of defects in the APC/ß-cat/Tcf and E-cad/cat pathways in cancer. Our findings indicate a role for Tcf deregulation in gastric and pancreatic cancers, resulting from mutations in ß-cat and -cat in some cases and from yet to be defined defects in others. Mutations in ß-cat were found in one gastric and one pancreatic cancer cell line, and we confirmed that the mutant ß-cat proteins activate Tcf transcription in vitro. Like most other previously described CTNNB1 mutations in colorectal and other cancers, the mutations affect presumptive GSK3ß phosphorylation sites of ß-cat. Our studies also identified a missense mutation of a potential GSK3ß phosphorylation site in the NH2-terminal region of -cat, and we found that this mutant protein had clearly detectable but more modest effects on Tcf transcription than comparable mutant ß-cat proteins. To the best of our knowledge, our studies are the first to identify activating mutations in the CTNNG1 gene in cancer, and the findings imply that missense mutations and small in-frame deletions of the NH2-terminal phosphorylation sites of -cat are likely to have effects in cancer cells that are analogous to those of ß-cat mutations. We failed to find CTNNB1 or CTNNG1 mutations in studies of a panel of 36 primary gastric cancers. Hence, similar to colorectal cancer, in which ß-cat mutations are only found in 2–5% of primary tumors (47
, 49
, 50)
, mutations in ß-cat and -cat appear to be rare in primary gastric cancers. Nevertheless, our data imply that Tcf transcriptional deregulation may be important in the pathogenesis of a subset of pancreatic and gastric cancers. In some cases, such as the ASPC pancreatic cancer line, it may result from defects other than those in the APC, CTNNB1, or CTNNG1 genes.
We had hoped to reconcile the large body of data demonstrating frequent inactivation of E-cad by mutational or epigenetic mechanisms in various cancers with the recent findings implicating ß-cat deregulation and Tcf transcriptional activation in colon and other cancers. E-cad is a major binding protein for ß-cat in cells, and ß-cat has an important role in E-cad’s cell adhesion function, via its ability to link E-cad to -cat and the cytoskeleton. We initially predicted that loss of endogenous E-cad might lead to a significant increase in free cytosolic ß-cat and Tcf deregulation. However, this prediction was not borne out by our studies. Loss of endogenous E-cad expression in gastric, pancreatic, and breast cancers was not associated with obvious increases in ß-cat levels, and no evidence of CTTA was found in any of the cancer lines lacking endogenous E-cad expression. A likely explanation for the observations is that the proteins and pathways responsible for regulating ß-cat cytosolic levels, such as GSK3ß, APC, and axin, remain fully functional in cancer cells with E-cad defects.
Sadot et al. (51)
have recently reported that overexpression of full-length N-cadherin or the cytoplasmic (ß-cat binding) region of N-cadherin effectively inhibits the ability of ß-cat to activate Tcf transcription in Chinese hamster ovary cells and SW480 colorectal cancer cells. Although the findings stand in apparent contradiction to our results, it is difficult to make direct comparisons between studies of the effects on Tcf activation when various forms of N-cadherin are overexpressed and studies of Tcf activation in cancer cell lines lacking endogenous E-cad expression. We believe that our studies better reflect the situation found in primary cancers in which endogenous E-cad expression is lost. As such, our data imply that there is no direct connection between the E-cad/cat and APC/cat/Tcf pathways in cancer cells. Moreover, our findings indicate that the functional consequences of inactivating mutations in APC or activating mutations in ß-cat or -cat are distinct from those attributable to E-cad inactivation. Future studies will provide novel and definitive insights into the means by which ß-cat functions in both E-cad cell-cell adhesion and Tcf transcriptional activation as well as the specific means by which E-cad inactivation contributes to altered cell growth control and cancer development and progression. Perhaps then, the apparently differing consequences of catenin and E-cad defects on the cancer cell phenotype can be reconciled.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
DNA Transfections and Reporter Assays.
Transfections to assess Tcf transcriptional activation were performed on cells growing at 70% confluence in six-well plates, using 1 ml of Opti-MEM reduced serum medium (Life Technologies, Inc., Grand Island, NY), 6 µl of Lipofectin reagent (Life Technologies, Inc.), 0.75 µg of pcH110, and 0.75 µg of pTopflash (optimal motif) or pFopflash (mutant motif) in each well (21)
. To assess the effect of wild-type and mutated forms of ß-cat and -cat on Tcf transcriptional activity, we undertook cotransfections of 293 cells as described above, except that 0.5 µg of pcH110, 0.5 µg of pTopflash or pFopflash, and 0.5 µg of the ß-cat or -cat expression vector were used. To assess the effect of APC on Tcf activity, we cotransfected cells with 0.5 µg of pcH110, 0.5 µg of pTopflash or pFopflash, and 0.25 or 0.75 µg of the wild-type APC expression construct. Varying amounts of an empty control vector (pcDNA3) were used to normalize the DNA mass to 2 µg in each transfection. Cell extracts were prepared 24–30 h posttransfection using reporter lysis buffer (Promega). Luciferase and ß-galactosidase assays were carried out as recommended by the manufacturer (Promega). Luciferase activities were measured in a luminometer (model TD-20E; Turner Corp., Mountain View, CA). All transfections were repeated three or more times. For each cell line, luciferase activities in a given series of transfection experiments were first normalized for ß-galactosidase activity. Then, the corrected mean luciferase activity of the pFopflash reporter construct was assigned a value of 1, and the luciferase activity with pTopflash was reported relative to the pFopflash activity.
ß-cat and -cat Mutation Analyses.
For the four gastric and eight pancreatic cancer cell lines, first strand cDNA was prepared from total RNA, using avian myeloblastosis virus reverse transcriptase (Promega) and random hexamers. The entire open reading frames of CTNNB1 and CTNNG1 were then amplified from cDNA by PCR with Pfu (Stratagene, La Jolla, CA) and primers ßCATFL-S and ßCATFL-A and PLAKFL-S and PLAKFL-A, respectively (see Table 1
for primer sequences). In 36 primary gastric cancer specimens, DNA was isolated by microdissection of regions of carcinoma from thick, unstained sections of formalin-fixed and paraffin-embedded tumors. To amplify a CTNNB1 genomic DNA fragment containing exons 2–4 (codons 1–86), we carried out PCR with Pfu, using oligonucleotide primers ßCATEX2-S and ßCATEX4-A. Similarly, PCR with Pfu was carried out to amplify a genomic DNA fragment containing codons 7–70 of CTNNG1, using primers PLAKG-S and PLAKG-A. PCR products were purified following agarose gel electrophoresis and sequenced directly with the following primers: ßCATEX2-S and ßCATEX4-A for ß-cat RT-PCR products; ßCATGS-S and ßCATGS-A for CTNNB1 genomic DNA products; and PLAKG-S and PLAKG-A for CTNNG1 RT-PCR and genomic DNA products. Thermo-Sequenase and 33P-labeled ddNTPs were used according to the manufacturer’s instructions (Amersham, Arlington Heights, IL). Mutations were verified in independent PCR and sequencing studies.
|
-cat expression constructs. Full-length wild-type CTNNB1 and CTNNG1 cDNAs were amplified by PCR from a normal colon cDNA library (Clontech, Palo Alto, CA), using primers ßCATFL-S and ßCATFL-A for CTNNB1 and PLAKFL-S and PLAKFL-A for CTNNG1 (Table 1)
. The oligonucleotides were modified to yield BamHI (5' end) and XbaI (3' end) recognition sites at the ends of the CTNNB1 cDNA and KpnI (5' end) and EcoRI (3' end) sites for the CTNNG1 cDNA. Mutant CTNNB1 and CTNNG1 cDNAs were amplified by PCR from cDNA of selected gastric and pancreatic cancer cell lines with the respective primer pairs. To generate the ß-cat N90 and exon 3 deletion constructs, the forward primers ßCATN90-S and ßCATEX3-S were used together with the ßCATFL-A reverse primer. PCR amplification of CTNNB1 and CTNNG1 cDNAs was carried out with Pfu polymerase, and the sequences of all PCR products was verified by manual or automated DNA sequencing. The pCH110 eukaryotic expression vector (Pharmacia, Piscataway, NJ), containing a functional LacZ gene cloned downstream of a cytomegalovirus early region promoter/enhancer element was used to control for transfection efficiency in the luciferase assays.
Western Blot Analysis.
Approximately 40 µg of each protein lysate in radioimmunoprecipitation lysis buffer (Tris-buffered saline, 0.5% deoxycholate, 0.1% SDS, and 1% NP40) buffer was separated by electrophoresis on SDS/polyacrylamide gels and transferred to Immobilon P membranes (Millipore, Bedford, MA) by semidry electroblotting (Transblot, Bio-Rad, Hercules, CA). ECL Western blot analysis was carried out with the mouse monoclonal antibodies against -cat, ß-cat, -cat, GSK-3ß (all from Transduction Laboratories, Lexington, KY), and E-cad (Zymed Laboratories, Inc., San Francisco, CA). A polyclonal antibody against Na+/K+ ATPase (Research Diagnostics, Inc, Flanders, NJ) was used to verify equal loading of Western blots. For studies of the expression of Flag epitope-tagged wild-type and mutated forms of ß-cat and -cat, anti-Flag antibody (Sigma Chemical Co., St. Louis, MO) was used at a 1:5000 dilution, and antiactin (Sigma) antibody was used at 1:1000. The secondary antibodies used for ECL-Western studies were horseradish peroxidase-conjugated goat-antimouse IgG or goat-antirabbit immunoglobulin antibodies (Pierce Biochemicals, Rockford, IL). Antibody complexes were detected with the ECL-Western blot kit (Amersham) and exposure to X-OMAT film (Kodak, Rochester, NY).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 This work was supported by grants from the William and Catherine Weldon Donaghue Foundation for Medical Research, NIH Grants GM57604 and CA70097, United States Army Grant DAMD17-94-J4366, and Deutsche Forschungsgemeinschaft Grants CA207/1 and KO1826/1. 
2 The first two authors contributed equally to this work.
3 Present address: Department of Internal Medicine, University of Leipzig, Leipzig 04103, Germany.
4 Present address: GI Division, Department of Internal Medicine, Peking Union Medical College Hospital, Beijing 100730, People’s Republic of China.
5 To whom requests for reprints should be addressed, at Division of Molecular Medicine and Genetics, University of Michigan Medical Center 4301 MSRB III, Box 0638, 1150 West Medical Center Drive, Ann Arbor, MI 48109-0638. Phone: (734) 764-1549; Fax: (734) 647-7979; efearon{at}mmg.im.med.umich.edu
6 The abbreviations used are: APC, adenomatous polyposis coli; ß-cat, ß-catenin; -cat, -catenin; GSK3ß, glycogen synthase kinase 3ß; E-cad, E-cadherin; Tcf, T-cell factor; Lef, lymphoid enhancer factor; CTTA, constitutive Tcf transcriptional activity; GI, gastrointestinal; RT-PCR, reverse transcriptase-PCR ECL, enhanced chemiluminescence.
Received for publication 1/19/99. Revision received 3/30/99. Accepted for publication 4/ 2/99.
|
References |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
-catenin, ß-catenin, and plakoglobin. J. Biol. Chem., 270: 5549-5555, 1995.- and ß-catenin genes in human gastric and breast carcinomas. Cancer Res., 56: 49-52, 1996.-catenin expression in human esophageal cancer. Cancer Res., 54: 291-296, 1994.-catenin, ß-catenin, and plakoglobin in esophageal carcinomas and its prognostic significance: immunohistochemical analysis of 96 lesions. Oncology, 54: 158-165, 1997.[Medline]
This article has been cited by other articles:
 |
|
 |
|
  K. Ganesan, T. Ivanova, Y. Wu, V. Rajasegaran, J. Wu, M. H. Lee, K. Yu, S. Y. Rha, H. C. Chung, B. Ylstra, et al. Inhibition of Gastric Cancer Invasion and Metastasis by PLA2G2A, a Novel {beta}-Catenin/TCF Target Gene Cancer Res., June 1, 2008; 68(11): 4277 - 4286. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Salim and L. Ratner Modulation of {beta}-Catenin and E-Cadherin Interaction by Vpu Increases Human Immunodeficiency Virus Type 1 Particle Release J. Virol., April 15, 2008; 82(8): 3932 - 3938. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Winter, A. H. Ting, F. Vilardell, E. Gallmeier, S. B. Baylin, R. H. Hruban, S. E. Kern, and C. A. Iacobuzio-Donahue Absence of E-Cadherin Expression Distinguishes Noncohesive from Cohesive Pancreatic Cancer Clin. Cancer Res., January 15, 2008; 14(2): 412 - 418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Dvory-Sobol, E. Sagiv, D. Kazanov, A. Ben-Ze'ev, and N. Arber Targeting the active {beta}-catenin pathway to treat cancer cells. Mol. Cancer Ther., November 1, 2006; 5(11): 2861 - 2871. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Andersen, J. Mejlvang, S. Mahmood, I. Gromova, P. Gromov, E. Lukanidin, M. Kriajevska, J. K. Mellon, and E. Tulchinsky Immediate and Delayed Effects of E-Cadherin Inhibition on Gene Regulation and Cell Motility in Human Epidermoid Carcinoma Cells Mol. Cell. Biol., October 15, 2005; 25(20): 9138 - 9150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K S Nair, R Naidoo, and R Chetty Expression of cell adhesion molecules in oesophageal carcinoma and its prognostic value J. Clin. Pathol., April 1, 2005; 58(4): 343 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Shiina, J. E. Breault, W. W. Basset, H. Enokida, S. Urakami, L.-C. Li, S. T. Okino, M. Deguchi, M. Kaneuchi, M. Terashima, et al. Functional Loss of the {gamma}-Catenin Gene through Epigenetic and Genetic Pathways in Human Prostate Cancer Cancer Res., March 15, 2005; 65(6): 2130 - 2138. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Gottardi and B. M. Gumbiner Distinct molecular forms of {beta}-catenin are targeted to adhesive or transcriptional complexes J. Cell Biol., October 25, 2004; 167(2): 339 - 349. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Al-Aynati, N. Radulovich, R. H. Riddell, and M.-S. Tsao Epithelial-Cadherin and {beta}-Catenin Expression Changes in Pancreatic Intraepithelial Neoplasia Clin. Cancer Res., February 15, 2004; 10(4): 1235 - 1240. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Lowy, C. Fenoglio-Preiser, O. J. Kim, J. Kordich, A. Gomez, J. Knight, L. James, and J. Groden Dysregulation of {beta}-Catenin Expression Correlates With Tumor Differentiation in Pancreatic Duct Adenocarcinoma Ann. Surg. Oncol., April 1, 2003; 10(3): 284 - 290. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Shtutman, J. Zhurinsky, M. Oren, E. Levina, and A. Ben-Ze'ev PML Is a Target Gene of {beta}-Catenin and Plakoglobin, and Coactivates {beta}-Catenin-mediated Transcription Cancer Res., October 15, 2002; 62(20): 5947 - 5954. [Abstract] [Full Text] [PDF]
|
|
