| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
Developmental Genetics of Melanocytes, UMR 146 CNRS-Institut Curie, 91405 Orsay Cedex, France [S. J. G., L. L.]; Human Genetics Program [A. B., J. U.] and Department of Pathology [A. J. K-S.], Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111; Department for Molecular Biomedical Research, VIB-Ghent University, B-9000 Ghent, Belgium [F. v. R.]; Hopkins Center for Epithelial Disorders, Johns Hopkins School of Medicine, Baltimore, Maryland 21205 [W. L-K., M. D.]; and Molecular Oncology Research Institute, Tufts-New England Medical Center, Boston, Massachusetts 02111 [P. N. T.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The oncogenic serine/threonine kinase AKT1 (also known PKB
) is a downstream effector of the PI3K5
and is frequently activated in human cancer (8)
. The gene for the closely related AKT2 is amplified and overexpressed in ovarian, pancreatic, breast, and follicular thyroid carcinomas, and AKT2 kinase activity is high in ovarian cancer (9, 10, 11, 12, 13)
. Furthermore, total AKT kinase activity is activated in non-small cell lung cancer, squamous cell carcinomas of the oral cavity, and breast and prostate carcinomas (14, 15, 16)
. In ovarian cancer, AKT2 amplification and overexpression are associated with undifferentiated histology and aggressive clinical behavior, suggesting that AKT contributes to tumor progression (10)
.
AKT activation contributes to the neoplastic phenotype. AKT stabilizes the cell cycle inhibitors p21Cip1 and p27Kip1 and inhibits the transport of both proteins into the nucleus. AKT also enhances the translation of mRNAs for cyclins D1 and D3. These changes lead to increased cyclin-dependent kinase and E2F activity and promote cell cycle progression (reviewed in Ref. 8
). AKT also promotes cell survival. The antiapoptotic function of AKT has been linked to inhibition of cytochrome c release from mitochondria, stimulation of glucose uptake and utilization, phosphorylation and inactivation of Bad and (pro)caspase 9, activation of nuclear factor 
B, overexpression of Bcl-2 or Bcl-xL, and phosphorylation and nuclear exclusion of FKHRL (8
, 17) . AKT activation is also associated with enhanced tumor cell invasion. AKT enhances invasiveness of pancreatic carcinoma cells via up-regulation of IGF-I (18)
and increases secretion of matrix metalloproteinases 2 and 9 from immortalized mammary epithelial cells and ovarian carcinomas (19
, 20)
.
In this article, we investigate the role of AKT in the biology of human squamous cell carcinoma lines and show that AKT activation causes EMT characterized by down-regulation of numerous epithelial cell-specific proteins, including E-cadherin and ß-catenin, and up-regulation of the mesenchymal cell-specific protein vimentin. Interestingly, EMT was accompanied by increased in vivo cell motility on fibronectin-coated surfaces and increased invasiveness in animals. These findings expand the spectrum of biological activities of AKT and suggest that therapeutic inhibition of AKT may be a useful strategy to control tumor cell invasion and metastasis.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibodies, Immunoblotting, and Immunostaining.
The mouse monoclonal antibodies used were anti-p130cas, anti-ß-tubulin, and anti-desmoplakin I/II (gifts from Drs. E. Golemis and M. Pasdar); anti-vimentin (Biodesign); anti-pan-cadherin (Sigma); and antihuman E-cadherin and anti-ß-catenin (Transduction Laboratories). The rabbit polyclonal antibodies used were anti-HEF1 and anti--catenin (gifts from Drs. E. Golemis and R. Kemler), anti-Akt (19992), anti-phospho-AKT threonine 308 (Cell Signaling), and anti-
-catenin (Transduction Laboratories).
Immunoblotting was performed as described previously (22)
. Anti-AKT and anti-HEF1 antibodies were diluted 1:500 before use, anti-p130cas antibody was diluted 1:750 before use, and anti-ß-tubulin and anti--catenin antibodies were diluted 1:2000 before use. Anti--catenin, anti-ß-catenin, anti-E- cadherin, anti-desmoplakin, and anti-pan-cadherin antibodies were used at final concentrations of 2, 0.2, 0.2, 0.5, and 15 µg/ml, respectively. Enhanced chemiluminescence detection was used (ECL; Amersham).
Immunostaining was performed as described previously (22) . Antibody concentrations were 2.5 µg/ml for anti-desmoplakin, 0.3 µg/ml for anti-vimentin, 10 µg/ml for anti-E-cadherin, and 2.5 µg/ml for anti-ß-catenin. Slides were examined under a Leica DM IRB light microscope equipped for epifluorescence or a laser scanning confocal microscope driven by Scanware software (Leica). Images were collected using the same settings and processed with Adobe Photoshop 4.0 at identical thresholds to allow semiquantitative comparisons.
Adenoviral Infection.
Recombinant adenoviruses encoding ß-galactosidase, Akt T308A, or myristylated Akt (myr-Akt) were amplified in HEK293 cells and purified according to standard procedures (23)
. SCC15 cells were infected with 50 plaque-forming units/cell in serum-free medium. After 3 h, infection was stopped by adding medium containing 20% fetal bovine serum. The cells were infected again 2 days later by the same procedure. Phase-contrast photographs were taken 2 days later. The proportion of infected cells was estimated from the number of cells producing ß-galactosidase, as revealed by X-gal staining.
Retroviral Infection.
Inserts harboring c-Akt, v-Akt, or myr-Akt were cloned into the retroviral vectors pLXSN (24)
or MSV-SR (25)
. Infectious viral supernatants were generated by transfection of the amphotropic packaging cell line PA-137 for pLXSN-based constructs and by transient cotransfection of COS cells with the amphotropic packaging plasmid pSV-A-MLV (a gift of Dr. Landau) for the MSV-SR-based constructs. Retroviral infections involved treating subconfluent cultures of SCC13 and SCC15 with DEAE-dextran (40 µg/ml) for 1 h and then with viral supernatants overnight. G418 (400 µg/ml) was used for selection 48 h after infection, and resistant colonies were pooled. Several mass cultures from independent infections were generated.
Northern Blot Analysis.
Total and poly(A)+ RNA were isolated with Roti-Quick and Quick Prep kits, respectively. Samples (20 µg of total RNA or 4 µg of poly(A)+ RNA) were subjected to electrophoresis in 1% agarose formaldehyde gels, transferred onto Hybond N+ membranes, and hybridized with 32P-labeled probes, as indicated. Signals were quantified with Storm 820 using Image Quant 5.2.
Promoter Reporter Assays.
SCC15 and SCC15 v-Akt C cells were transiently transfected with Exgen 500 (Euromedex) in 6-well plates. Each well contained serum-free medium containing 1.5 µg of pPGKßgeopA and 2 µg of either pGL3 basic vector or pGL3 hE-cad prom. Both constructs contain the luciferase gene under no promoter (pGL3) or under the wild-type human E-cadherinpromoter (pGL3 hE-cad prom) (6)
. Luciferase was assayed by standard procedures, and transfection efficiency was normalized for ß-galactosidase activity.
Cell Migration Assay.
Freshly trypsinized cells were plated at 105 cells on 35-mm Falcon 3004 dishes, coated with 10 µg/ml fibronectin, 25 µg/ml collagen, or 50 µg/ml laminin. The average speed (µm/h) of locomotion in complete medium, for optimal migration (26)
, was calculated for 20–30 cells for each experimental condition.
Detachment Assay.
Cells were seeded at 1.5 x 105 cells/25-cm2 flask. After 48 h, cells were washed with 5 ml of warm PBS and then trypsinized with 1 ml of fresh 0.25% trypsin (Life Technologies, Inc.) at 20°C with gentle agitation. The number of detached cells was determined at various times, and the total number of cells/flask was determined after complete trypsinization. One flask was used for each time point, and each experiment was performed at least five times independently.
Growth Curve.
Cells were seeded at 1 x 105 cells/25-cm2 flask, fed every other day, and counted every day. Growth curves were constructed, and doubling times were estimated.
Assays for Tumorigenesis and Invasiveness.
To determine tumorigenicity, SCC13 cells (5 x 106) were transferred s.c. into athymic nude mice. Eight weeks after injection, mice were killed and examined for gross evidence of tumors. Tissues were collected from the injection area, stained with H&E, and analyzed.
A tracheal invasion assay was performed as described previously (21 , 27 , 28) . Cells (5 x 105) were injected into the lumen of de-epithelialized rat tracheas that were closed with metal clips at both ends and then transplanted s.c. into nude mice. After 4 or 8 weeks, transplanted tracheas were removed and processed for histopathology. Invasiveness was estimated as the extent of penetration into the tracheal wall from the center of the cell mass to the most distal invasive point (28) .
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. ß-Galactosidase- or Akt T308A-infected cells displayed no morphological changes (Fig. 1, A, B, D, E, G, and H)
. Cells infected with myr-Akt lost their epithelial cell morphology; they were dispersed and assumed a fibroblast-like appearance (Fig. 1, C, F, and I)
. Similar results were obtained with NBT-II, a rat carcinoma cell line (data not shown).
|
. Uninfected SCC15 produced the Mr 60,000 endogenous AKT. SCC15 c-Akt cells contained at least three times more Akt than SCC15 cells. Various amounts of a protein larger than Mr 85,000 corresponding to v-Akt were detected in SCC15 v-Akt A, C, and D cells. AKT activation status was examined by Western blotting with anti-phospho-AKT threonine 308 antibody, a marker of AKT activation. Total AKT activity was very low in parental SCC15 cells and high in SCC15 cells expressing exogenous Akt (Fig. 1M)
. Expression of constitutively active Akt (v-Akt) was associated with the transition from an epithelial to a fibroblast-like morphology (Fig. 1, J–L)
.
Constitutively Active Akt Promotes a Shift in Expression from an Epithelial to a Mesenchymal Repertoire.
The morphological effect of v-Akt expression on squamous cell carcinoma lines suggested that active Akt promotes an EMT. The morphological changes characteristic of cells undergoing EMT are accompanied by a shift in gene expression from an epithelial to a mesenchymal repertoire. To determine whether Akt promotes such a shift, we used immunofluorescence to examine the expression and subcellular distribution of desmoplakin and vimentin, two markers of EMT. In parental SCC15 cells, desmoplakin was mostly in spots at the sites of cell-cell contact (Fig. 2A)
. All v-Akt-infected cells contained desmoplakin, but only in cytoplasmic granules (Fig. 2, B and C)
, and desmoplakin was less abundant in v-Akt-infected cells than in SCC15 cells, as assessed by Western blot analysis (data not shown). The parental SCC15 cell line did not produce vimentin, a mesenchymal marker (Fig. 2D)
, whereas all v-Akt cell lines produced vimentin microfilaments (Fig. 2, E and F)
. The heights of SCC15, SCC15 v-Akt C, and SCC15 v-Akt D cells were determined by confocal microscopy using E-cadherin as marker (data not shown). The mean height of SCC15 cells was 6.0 µm, and the mean heights of SCC15 v-Akt C and SCC15 v-Akt D cells were similar and equal to about 3.8 µm. We conclude that constitutively active Akt indeed induces EMT.
|
. E-cadherin, -catenin, ß-catenin, and p130cas were all dramatically down-regulated in v-Akt-infected cells, but not in c-Akt- or vector-infected cells. The level of plakoglobin was slightly down-regulated, and the level of the p130cas substrate HEF1 was not down-regulated. N-cadherin was slightly up-regulated (data not shown). Similar experiments with SCC13, another human squamous cell carcinoma line, and MSV-SR retroviruses encoding myr-Akt revealed changes in expression indistinguishable from those in SCC15 v-Akt (data not shown).
|
. Parental SCC15 cells produced both E-cadherin and ß-catenin, and both proteins were at contact sites between cells (Fig. 3, B and E)
. v-Akt-expressing-SCC15 cells contained less ß-catenin and E-cadherin: most of the E-cadherin was in cytoplasmic granules, with only a small amount at the plasma membrane (Fig. 3, C and D)
; and ß-catenin was diffuse in the cytoplasm, although some remained at the membrane (Fig. 3, F and G)
. Slight quantitative differences in ß-catenin distribution were observed between SCC15 v-Akt C and SCC15 v-Akt D cells; the amount of ß-catenin at the membrane was slightly higher in SCC15 v-Akt C cells. In the cytoplasm of Akt-expressing cells, E-cadherin and ß-catenin did not colocalize (Fig. 3, F and G
; data not shown).
Constitutively Active Akt Down-Regulates the Expression of the E-Cadherin Gene Transcriptionally by Inducing SNAIL.
During EMT there is a massive shift of gene expression from a pattern characteristic of epithelial cells to that of mesenchymal cells. There may be genes that are the primary targets of the EMT-inducing signal(s) and others whose expression is affected secondarily. E-cadherin and ß-catenin are potential candidate genes whose down-regulation by Akt may be direct, so we examined their mRNAs in v-Akt-expressing and parental SCC15 cells. Akt did not significantly affect the level of ß-catenin mRNA but substantially reduced the level of E-cadherin mRNA (Fig. 4A)
. Parental SCC15 and v-Akt-expressing SCC15 cells were transiently transfected with pGL3 or pGL3 hE-cadherin promoter luciferase constructs to determine whether Akt modulates the activity of the E-cadherinpromoter. Luciferase assays of extracts from the transfected cells revealed that the activity of the E-cadherinpromoter was repressed by Akt (Fig. 4B)
. Similar results were obtained for the murine E-cadherinpromoter (data not shown). Thus, Akt appears to down-regulate E-cadheringene transcription.
|
.
Constitutively Active Akt Alters Cell Motility on Different Substrates and Cell Attachment to Plastic.
Motility of parental and Akt-expressing SCC15 cells (SCC15, SCC15 c-Akt A, SCC15 v-Akt C, and SCC15 v-Akt D cells) was examined on uncoated plastic and plastic that had been coated with fibronectin, collagen, or laminin. On fibronectin, SCC15 v-Akt C and v-Akt D cells migrated more rapidly than parental SCC15 cells, whereas SCC15 c-Akt A cells exhibited the same motility as the parental SCC15 cells (Fig. 5A)
. On laminin, SCC15 v-Akt C cells migrated more slowly than SCC15 cells, whereas on collagen, the speed of migration of all cell lines was similar (data not shown).
|
. Thus, constitutively active Akt attenuates the attachment of SCC15 cells to plastic.
Constitutively Active Akt Promotes Proliferation, Tumorigenicity, and Invasiveness of Squamous Cell Carcinoma Lines.
E-cadherin suppresses tumor invasiveness (3
, 33)
, so we examined whether Akt promotes the tumorigenic potential and invasiveness of epithelial cells (Fig. 6)
. Growth curves were constructed for SCC15, SCC15 v-Akt B, and SCC15 v-Akt D cells (Fig. 6A)
. The doubling time of the parental cells was 23 h, and those of SCC15 v-Akt B and v-Akt D were only 13 and 12 h, respectively. Similar experiments were performed with SCC13 cell lines transfected with empty vector (Sr), v-Akt, and myr-Akt. The respective doubling times were estimated to be 28, 13, and 17 h for SCC13-Sr, SCC13-v Akt, and SCC13-myr Akt, respectively (data not shown).
|
. Microscopic examination of these tumors revealed evidence of invasion of the surrounding muscular tissue. Unlike parental SCC13 cells, parental SCC15 cells are fully tumorigenic in nude mice (21)
and therefore were not used to score for the effect of constitutive Akt activation on tumorigenesis.
We next measured the ability of parental and Akt-expressing SCC15 cells to pass through the pars membranacea of tracheal walls to assess invasiveness (Fig. 6, C–F)
. The cells expressing ectopic v-Akt migrated more efficiently through tissues than did the parental SCC15 cells. The test of invasiveness was also performed with the various SCC13 cell lines yielding virtually identical results: SCC13 cell lines expressing v-Akt or myr-Akt, but not SCC13-Sr cells, were able to pass through the pars membranacea (data not shown). Therefore, constitutively active Akt is a potent promoter of tumorigenicity and invasion.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Here we show that constitutively active Akt induces EMT and stimulates proliferation and motility of squamous cell carcinoma lines plated on fibronectin-coated surfaces (Fig. 5)
. Also, active Akt promotes invasiveness (Fig. 6)
. Cells expressing a constitutively active mutant of Akt (v-Akt or myristylated Akt) displayed several features typical of EMT: reduction in cell-cell adhesion; and flattening and spreading or scattering (Fig. 1)
. We detected Akt-induced EMT in cells stably infected with retroviral vectors and in those transiently infected with adenoviral vectors. EMT occurred a minimum of 72–96 h after transient infection, which may reflect the time required for the reprogramming of gene expression and/or structural reorganization associated with EMT (34)
.
At least two separate correlates of EMT have been identified, namely, cell-cell dissociation and cell movement (34)
. Akt activation appears to mediate both processes. In particular, expression of active mutants of Akt increases cell migration on fibronectin-coated plates but reduces migration on laminin-coated plates. This pattern is compatible with induction of 4ß1 integrin that interacts specifically with fibronectin (35)
. Indeed, integrin activation often follows EMT (36
, 37)
.
In several cell culture models, EMT is induced by transforming growth factor ß or by peptide growth factors via receptor tyrosine kinase signaling (22 , 38, 39, 40, 41) . In both cases, PI3K is a critical mediator of EMT. Oncogenic SRC and RAS, both inducers of EMT, also activate PI3K (42) . Our data suggest that AKT kinases are major effectors of EMT signals downstream of PI3K. Potential targets of the P13K/AKT pathway include Rac and Rho, two small G proteins involved in cytoskeletal reorganization, cell migration, and invasiveness (34) .
Akt-induced EMT involves a large down-regulation of E-cadherin and ß-catenin protein levels (Fig. 3)
. Down-regulation is specific for E-cadherin because the closely related N-cadherin is not affected. E-cadherin and ß-catenin are also relocalized to separate compartments, an indication that their interaction is disrupted. E-cadherin is internalized and displays a punctate cytoplasmic staining pattern, compatible with a vesicular localization. Localization of E-cadherin in vesicles has been described during IGF-I-induced EMT and may point to alterations in protein trafficking, possibly induced by activation of Rab5-mediated endocytosis (22
, 43, 44, 45)
.
Akt also down-regulated ß-catenin, a result that appeared paradoxical. In the Wnt pathway, GSK3 phosphorylates axin and ß-catenin, causing degradation of the latter. Upon Wnt-induced inhibition of GSK3, ß-catenin accumulates in the cytoplasm and translocates into the nucleus, acting as a cofactor for the transcription factor LEF/TCF (lymphoid enhancer factor/T cell factor), affecting the transcription of genes that promote cell survival and proliferation (46, 47, 48, 49, 50) . AKT, activated by peptide growth factor signals, phosphorylates and inhibits GSK3 (51) , so we expected stabilization and nuclear translocation of ß-catenin in squamous cell carcinoma lines expressing active Akt. The apparent discrepancy can be resolved by recent crystallographic and biochemical studies elucidating the mechanisms of GSK3 regulation (52 , 53) : Wnt and insulin/AKT signaling pathways affect two distinct pools of GSK3 that in turn target different substrates, thereby giving selective responses and differential substrate phosphorylation (54) . In the presence of sustained Wnt signaling, phosphorylation of GSK3 by AKT potentiates the Wnt pathway leading to ß-catenin stabilization; however, AKT signaling alone cannot initiate the Wnt signaling process (55, 56, 57) . Consequently, in our system, activation of Akt alone, i.e., in the absence of Wnt signaling, may not be able to phosphorylate and inhibit the GSK3 pool involved in ß-catenin up-regulation.
This does not, however, explain the down-regulation of ß-catenin by active Akt. This appears to be a posttranscriptional effect. Possibly, ß-catenin is destabilized and even degraded as a secondary consequence of the down-regulation of E-cadherin. After Akt activation, ß-catenin was displaced and did not colocalize with E-cadherin (Fig. 3)
, and in both Drosophila and mouse development, binding to E-cadherin stabilizes ß-catenin (58
, 59)
.
In addition to internalization of E-cadherin, Akt activation represses E-cadheringene transcription (Fig. 4)
. In the presence of active Akt, the E-cadherinpromoter is less active, and this repression appears to be the consequence of up-regulation of the transcription repressor SNAIL. Indeed, SNAIL induces EMT by repressing E-cadherintranscription (4
, 5)
. Other potential modulators of E-cadherin transcription are SNAIL-related repressors, such as Slug, Smuc, and SIP1 that bind to E-boxes in the E-cadherinpromoter (6
, 32
, 60)
. The mechanisms by which Akt activates transcription of SNAIL remain unclear.
Akt-induced EMT endows squamous cell carcinoma lines with an invasive phenotype as demonstrated by an in vivo assay of invasion (Fig. 6)
. The identical assay has demonstrated that AKT2 antisense RNA can inhibit invasiveness in cancer cells that amplify/overexpress the AKT2 gene (11)
. Although the invasiveness may be due in part to the ability of Akt to stimulate cell cycle progression (Fig. 6)
, it is more likely to result from loss of cell-cell adhesion (Figs. 1
2
3
), increased motility (Fig. 5)
, and tissue degradation. Akt activation can lead to increased production of matrix metalloproteinases (19
, 20)
, and, in turn, low levels of E-cadherin are associated with stromelysin 1 activation (61)
. Overexpression of IGF-I receptor in the pancreas in vivo leads to transformation and invasion associated with down-regulation of E-cadherin (62)
. An intriguing possibility is that this effect of IGF-I receptor is mediated by AKT.
In summary, we show that activation of the AKT pathway in cancer cells leads to EMT and invasion in vivo. Thus, an important consequence of the AKT activation often detected in human carcinomas is the acquisition of an invasive phenotype. Therapy based on AKT inhibition may therefore complement conventional treatments to control tumor cell invasion and metastasis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by Grants CA78412 (to A. B.), CA75028 (to A. J. K-S.), CORE Grant CA06927, and CA57436 (to P. N. T.) from the NIH; by an appropriation to the Fox Chase Cancer Center from the Commonwealth of Pennsylvania; and grants from La Ligue Nationale contre le Cancer and Association sur la Recherche sur le Cancer [L. L.], Ministère de la Recherche, France [S. G.], Fund for Scientific Research, Flanders [F. W. O.], and Fortis Bank-Verzekeringen, Brussels (F. v. R.). 
2 These authors contributed equally to this work.
3 To whom requests for reprints may be addressed, at Fox Chase Cancer Center, Division of Population Science, 7701 Burholme Avenue, Philadelphia, PA 19111. Phone: (215) 728-4012; Fax: (215) 214-1623; E-mail: a_bellacosa{at}fccc.edu
4 To whom requests for reprints may be addressed, at Developmental Genetics of Melanocytes, UMR 146 CNRS-Institut Curie, Bat. 110, 91405 Orsay Cedex, France. Phone: 33-1-69-86-71-07; Fax: 33-1-69-86-71-09; E-mail: lionel.larue{at}curie.fr
5 The abbreviations used are: PI3K, phosphatidylinositol 3'-kinase; EMT, epithelial-mesenchymal transition; IGF, insulin-like growth factor; X-gal, 5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside; HEF1, human enhancer of filamentation 1; GSK3, glycogen synthase kinase 3.
Received 8/28/02. Accepted 3/ 5/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B, IB kinase , and Akt kinase in squamous cell carcinoma of the oral cavity. Cancer (Phila.), 92: 3037-3044, 2001.[Medline]
kinase is frequently elevated in human cancers and its constitutive activation is required for oncogenic transformation in NIH3T3 cells. Am. J. Pathol., 159: 431-437, 2001.4ß1 integrin through the ß1 subunit induces recognition of the RGDS sequence in fibronectin. J. Cell Biol., 126: 271-279, 1994. induces a motile fibroblast-like phenotype with extracellular matrix-degrading potential in a rat bladder carcinoma cell line. Cell Regul., 1: 1003-1014, 1990.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. K. Nath, R. M. Brown, M. Michaud, M. R. Sierra-Honigmann, M. Snyder, and J. A. Madri Leptin affects endocardial cushion formation by modulating EMT and migration via Akt signaling cascades J. Cell Biol., April 21, 2008; 181(2): 367 - 380. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Escriva, S. Peiro, N. Herranz, P. Villagrasa, N. Dave, B. Montserrat-Sentis, S. A. Murray, C. Franci, T. Gridley, I. Virtanen, et al. Repression of PTEN Phosphatase by Snail1 Transcriptional Factor during Gamma Radiation-Induced Apoptosis Mol. Cell. Biol., March 1, 2008; 28(5): 1528 - 1540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ye, D. Hu, L. Tu, X. Zhou, F. Lu, B. Wen, W. Wu, Y. Lin, Z. Zhou, and J. Qu Involvement of PI3K/Akt Signaling Pathway in Hepatocyte Growth Factor-Induced Migration of Uveal Melanoma Cells Invest. Ophthalmol. Vis. Sci., February 1, 2008; 49(2): 497 - 504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Zeng, Y. Yao, M. Han, X. Zhao, X.-C. Liu, J. Wei, Y. Luo, J. Zhang, J. Zhou, S. Wang, et al. Biliverdin Reductase Mediates Hypoxia-Induced EMT via PI3-Kinase and Akt J. Am. Soc. Nephrol., February 1, 2008; 19(2): 380 - 387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Guan, X.-r. Wang, X.-f. Zhu, X.-f. Huang, J. Xu, L.-h. Wang, X.-b. Wan, Z.-j. Long, J.-n. Liu, G.-k. Feng, et al. Aurora-A, a Negative Prognostic Marker, Increases Migration and Decreases Radiosensitivity in Cancer Cells Cancer Res., November 1, 2007; 67(21): 10436 - 10444. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. S. Tan, G. Icre, A. Montagner, B. B.-t. Heggeler, W. Wahli, and L. Michalik The Nuclear Hormone Receptor Peroxisome Proliferator-Activated Receptor {beta}/{delta} Potentiates Cell Chemotactism, Polarization, and Migration Mol. Cell. Biol., October 15, 2007; 27(20): 7161 - 7175. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. F.L. O'Shaughnessy, B. Akgul, A. Storey, H. Pfister, C. A. Harwood, and C. Byrne Cutaneous Human Papillomaviruses Down-regulate AKT1, whereas AKT2 Up-regulation and Activation Associates with Tumors Cancer Res., September 1, 2007; 67(17): 8207 - 8215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. C. Willis and Z. Borok TGF-beta-induced EMT: mechanisms and implications for fibrotic lung disease Am J Physiol Lung Cell Mol Physiol, September 1, 2007; 293(3): L525 - L534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Yoon, Z. Liang, X. Zhang, M. Choe, A. Zhu, H. T. Cho, D. M. Shin, M. M. Goodman, Z. Chen, and H. Shim CXC Chemokine Receptor-4 Antagonist Blocks Both Growth of Primary Tumor and Metastasis of Head and Neck Cancer in Xenograft Mouse Models Cancer Res., August 1, 2007; 67(15): 7518 - 7524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kaimori, J. Potter, J.-y. Kaimori, C. Wang, E. Mezey, and A. Koteish Transforming Growth Factor-beta1 Induces an Epithelial-to-Mesenchymal Transition State in Mouse Hepatocytes in Vitro J. Biol. Chem., July 27, 2007; 282(30): 22089 - 22101. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
, SCC15 cells; 
, SCC15 v-Akt C cells; 
, SCC15 v-Akt D cells.
), 13 h for SCC15 v-Akt B cells (