QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 15 February 2005
doi: 10.1242/jcs.01634

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Maeda, M. Articles by Wheelock, M. J. Search for Related Content PubMed PubMed Citation Articles by Maeda, M. Articles by Wheelock, M. J.

Cadherin switching: essential for behavioral but not morphological changes during an epithelium-to-mesenchyme transition

¹ Department of Oral Biology (College of Dentistry), University of Nebraska Medical Center, Omaha, NE 68198-7696, USA
² Department of Biochemistry and Molecular Biology, University of Nebraska Medical Center, Omaha, NE 68198-7696, USA
³ Department of Genetics, Cell Biology and Anatomy, University of Nebraska Medical Center, Omaha, NE 68198-7696, USA
⁴ Department of Pathology and Microbiology (College of Medicine), University of Nebraska Medical Center, Omaha, NE 68198-7696, USA
⁵ Eppley Institute for Research in Cancer and Allied Diseases, University of Nebraska Medical Center, Omaha, NE 68198-7696, USA
⁶ Eppley Cancer Center, University of Nebraska Medical Center, Omaha, NE 68198-7696, USA

View larger version (88K): [in a new window]   Fig. 1. Subcloning NMuMG cells. (A) Parental NMuMG cells (a-c), clone NMuMG/E9 (d,e) and clone NMuMG/E11 (f,g) were stained for N-cadherin (a,d,f) and E-cadherin (b,e,g). (c) A merged picture of a,b. (a-c) Photographs were taken using a 10x objective; scale bar, 50 µm. (d-g) Photographs were taken using a 40x dry objective; scale bar, 10 µm. (B) TNE extracts of NMuMG/E9 (lane 1) and NMuMG/E11 (lane 2) were analysed for N-cadherin (top) and E-cadherin (bottom) by immunoblots. (C) Both clones underwent morphological changes (elongation of cell shape, b,d) in response to TGF-ß1 [Tß1; 5 ng ml–1 for 1 day (d)]. Photographs were taken of living cells using a 10x objective; scale bar, 50 µm.

View larger version (76K): [in a new window]   Fig. 2. TGF-ß1-mediated EMT in NMuMG/E9 cells. (A) NMuMG/E9 cells were stained for E-cadherin (a,b), ZO-1 (c,d), paxillin (e,f) and F-actin (phalloidin; g,h) in the absence of TGF-ß1 (a,c,e,g) or in the presence of TGF-ß1 (5 ng ml–1 for 1 day; b,d,f,h). In the presence of TGF-ß1, the signals for E-cadherin and ZO-1 became weak and punctate, and focal adhesions (f) and stress fibers (h) were induced. Photographs were taken using a 63x oil objective; scale bar, 10 µm. (B) NMuMG/E9 cells were treated with TGF-ß1 and cultured without passage for the indicated times. TNE extracts were immunoblotted for N-cadherin, E-cadherin, ß-catenin and GAPDH at the indicated times from t=0 hour (h) to t=9 days (d). Immunoblots were quantified and normalized to GAPDH as shown on the graph. ZO-1 (an epithelium marker) and fibronectin (a mesenchyme marker) were examined in the same time course.

View larger version (42K): [in a new window]   Fig. 3. E-Cadherin continued to function after morphological EMT was induced. (A) Permeabilized (0.2% Triton X-100, Tx100; a) and unpermeabilized (b) NMuMG/E9 cells that had been treated with 5 ng ml–1 TGF-ß1 for 1 day were stained with antibodies against the extracellular portion of E-cadherin (ECCD2). E-Cadherin was localized to the cell surface (b). Photographs were taken using a 40x dry objective; scale bar, 10 µm. (B) Extracts of untreated NMuMG/E9 cells (lanes 3, 4) or NMuMG/E9 cells treated with 5 ng ml–1 TGF-ß1 for 1 day (lanes 5, 6) were immunoprecipitated (IP) with anti-E-cadherin (EC; lanes 4, 6) or control antibodies (Ctr; lanes 3, 5) and immunoblotted for E-cadherin, ß-catenin and -catenin. Comparable amounts of ß-catenin and -catenin were coimmunoprecipitated with E-cadherin in treated (lane 6) and untreated (lane 4) NMuMG/E9 cells, indicating that E-cadherin junctional complexes were functional. Lanes 1, 2 are immunoblots of treated and untreated cells to show N-cadherin and fibronectin were induced in this experiment, as expected, and to show the input levels of E-cadherin, ß-catenin and -catenin in the immunoprecipitations.

View larger version (47K): [in a new window]   Fig. 4. Snail and SIP1 were induced during TGF-ß1-mediated EMT in NMuMG/E9 cells. (A) RT-PCR using total RNA extracted from untreated (lane 1) and TGF-ß1-treated (5 ng ml–1 for 2 days) NMuMG/E9 cells (lane 2) was done for snail, SIP1, slug, E12/E47 and ß-actin (as a control). The identity of the slower migrating band in the E12/E47 lane is not known. (B) RT-PCR using total RNA extracted from NMuMG/E9 cells over the same time course as Fig. 2B. E-Cadherin, snail, SIP1 and GAPDH were analysed. (C) The snail and SIP1 mRNA levels were analysed by quantitative real-time RT-PCR using the same total RNA as in B.

View larger version (92K): [in a new window]   Fig. 5. Cadherin switching was reversible in TGF-ß1-treated NMuMG/E9 cells. (A) Untreated NMuMG/E9 cells (a,d,g), NMuMG/E9 cells treated with 1 ng ml–1 TGF-ß1 and 20 ng ml–1 EGF for 30 days (b,e,h), and NMuMG/E9 cells treated with TGF-ß1 and EGF for 23 days and subsequently cultured upon removal of both reagents for an additional 7 days (c,f,i) were stained for E-cadherin (a-c), ZO-1 (d-f) or F-actin (phalloidin) (g-i). Photographs were taken using a 40x dry objective; scale bar, 10 µm. (B) TNE extracts from untreated NMuMG/E9 cells (lane 1), TGF-ß1- and EGF-treated NMuMG/E9 cells for 60 days (lane 2) or reversibly treated NMuMG/E9 cells (lane 3) were examined for N-cadherin, E-cadherin, fibronectin and GAPDH by immunoblot. (C) RT-PCR was performed using total RNA extracted from untreated NMuMG/E9 cells (lane 1), TGF-ß1- and EGF-treated NMuMG/E9 cells (for 1 day, lane 2, or for 30 days, lane 3) or reversibly treated NMuMG/E9 cells (lane 4) to examine the levels of mRNAs encoding E-cadherin, N-cadherin, snail, SIP1 and GAPDH.

View larger version (84K): [in a new window]   Fig. 6. Manipulation of N-cadherin expression in NMuMG/E9 cells did not interfere with TGF-ß1-induced morphological change. (A) N-Cadherin-knockdown NMuMG/E9 cells (NMuMG/E9Ncad cells) showed similar morphology to parental NMuMG/E9 cells in phase-contrast microscopy (a,b). Photographs were taken of living cells using a 10x phase objective; scale bar, 50 µm. Parental NMuMG/E9 cells and NMuMG/E9Ncad cells were stained with N-cadherin mAb in the absence of TGF-ß1 (c,d). Cell-cell border staining of N-cadherin was significantly reduced. Photographs were taken using a 40x dry objective; scale bar, 10 µm. (B) TNE extracts of NMuMG/E9 cells and NMuMG/E9Ncad cells in the absence and presence of TGF-ß1 (5 ng ml–1 for 1 day) were immunoblotted for N-cadherin, E-cadherin, GAPDH and fibronectin. N-Cadherin and E-cadherin levels were quantified and normalized to GAPDH as shown on the bar graph. (C) NMuMG/E9Ncad cells in the absence (a,c) or presence of TGF-ß1 (5 ng ml–1 for 1 day; b,d) were stained for paxillin (a,b) and F-actin (c,d). Photographs were taken using a 40x dry objective; scale bar, 10 µm. (D) Transwell motility assays were done using control knockdown NMuMG/E9 cells (ctr KD) and NMuMG/E9Ncad cells in the absence of TGF-ß1 showed that NMuMG/E9Ncad cells were significantly less motile than control cells (P<0.05). (E) NMuMG/E9 cells expressing 2x-Myc-tagged human N-cadherin (NMuMG/E9 Ncad cells) showed similar morphology to parental NMuMG/E9 cells in phase-contrast microscopy (a,b). Photographs were taken using a 40x dry objective; scale bar, 10 µm. (F) TNE extracts of parental NMuMG/E9 cells and NMuMG/E9 Ncad cells were immunoblotted for N-cadherin (top) and E-cadherin (bottom).

View larger version (90K): [in a new window]   Fig. 7. E-Cadherin and N-cadherin expression in MCF10A cells was dependent on cell confluence. (A) Phase-contrast micrographs of MCF10A cells in sparse culture (a) or confluent culture (b). Photographs were taken of living cells using a 10x objective; scale bar, 50 µm. (B) MCF10A cells in sparse culture (a,c) or confluent culture (b,d) were stained for E-cadherin (a,b) or N-cadherin (c,d). Photographs were taken using a 40x dry objective; scale bar, 10 µm. (C) TNE extracts of MCF10A cells in increasing confluence were immunoblotted for N-cadherin (top), E-cadherin (middle) and ß-catenin (bottom).

View larger version (55K): [in a new window]   Fig. 8. TGF-ß1-mediated EMT in MCF10A cells. (A) MCF10A cells in the absence (a) or presence (b) of TGF-ß1 [2 ng ml–1 for 1 day (d)] were stained for F-actin. Photographs were taken using a 63x oil objective; scale bar, 10 µm. (B) Phase-contrast micrographs of TGF-ß1-treated MCF10A cells (t=0-3 days). Photographs were taken of living cells using a 10x objective; scale bar, 50 µm. (C) TNE extracts of TGF-ß1-treated parental MCF10A cells (t=0-3 days; lanes 1-4) control knockdown cells (ctr KD cells) (t=0-3 days; lanes 5-8), N-cadherin knockdown cells (Ncad cells) (t=0-3 days; lanes 9-12) and untreated N-cadherin overexpressing cells (Ncad cells) (lane 13) were immunoblotted for N-cadherin (regular and longer exposure), E-cadherin, GAPDH and fibronectin. The slower migrating bands in the N-cadherin immunoblot are pro-region-containing precursor forms of N-cadherin. (D) Quantification of N- and E-cadherin levels using normalization with GAPDH. (E) MCF10ANcad in the absence (a,c) or presence (b,d) of TGF-ß1 (2 ng ml–1 for 2 days) were co-stained for paxillin (a,b) and F-actin (c,d). Photographs were taken using a 63x oil objective; scale bar, 10 µm.

View larger version (84K): [in a new window]   Fig. 9. Knockdown of N-cadherin in MCF10A cells decreased cell motility. (A) Wound-healing assays. Phase-contrast micrographs of living cultures of parental MCF10A cells (a-f), MCF10A/ctr KD cells (g-l) and MCF10ANcad cells (m-r) in a wound-healing assays performed in the absence (a-c,g-i,m-o) or presence (d-f,j-l,p-r) of TGF-ß1 (2 ng ml–1). TGF-ß1 treatment began 2 days before initiation of the assays. Photographs were taken just after the incision (0 hour) (a,d,g,j,m,p) and at 5 hours (b,e,h,k,n,q) and 13 hours (c,f,i,l,o,r) after incision using a 10x objective; scale bar, 50 µm. (B) Motility was quantified by measuring the decrease in the denuded area at 5 hours and 13 hours, and presented as the average decrease in the number of pixels with standard deviation in three independent experiments. TGF-ß1 treatment increased motility in all cell lines tested. In the presence of TGF-ß1, MCF10ANcad cells were significantly slower than parental or control knockdown cells at 5 hours (h) and 13 hours. (C) Transwell motility assays of parental MCF10A cells, MCF10A/ctr KD cells and MCF10ANcad cells in the absence (–) or presence (+) of TGF-ß1 (2 ng ml–1). TGF-ß1 treatment began 2 days before initiation of the assays. Data are presented as the average number of migrating cells in nine random fields of view in three independent experiments. In the presence of TGF-ß1, MCF10ANcad cells were significantly less motile than parental or control knockdown cells.

View larger version (90K): [in a new window]   Fig. 10. N-Cadherin overexpression promoted motility in MCF10A cells. (A) Phase-contrast micrographs of living cells (a,b; 10x objective; scale bar, 50 µm) and N-cadherin immunostaining micrographs (c,d; 40x objective; scale bar, 10 µm) of parental MCF10A cells (a,c) or MCF10A cells overexpressing 2x-Myc-tagged human N-cadherin (MCF10A/Ncad; b,d). (B) Phase-contrast micrographs of living cultures of parental MCF10A cells (a-c), MCF10A/ctr KD cells (d-f) or MCF10A/Ncad cells (g-i) in a wound-healing assay performed in the absence of TGF-ß1. Photographs were taken immediately after incision (0 hour; a,d,g) and at 13 hours (b,e,h) or 27 hours (c,f,i) after incision using a 10x objective; scale bar, 50 µm. Some of these panels are the same as those shown in Fig. 9A, for easier comparison to MCF10A/Ncad cells. (C) Wound-healing assays were quantified as described in Fig. 9. MCF10A/Ncad cells were significantly more motile than either control cell line at 13 hours and at 27 hours. (D) Transwell motility assay comparing parental MCF10A cells, MCF10A/ctr KD cells and MCF10A/Ncad cells in the absence of TGF-ß1. Data are presented as the average number of migrating cells in nine random fields of view in three independent experiments.