spacer gif spacer gif spacer gif spacer gif spacer gif
 QUICK SEARCH:   [advanced]


spacer gif
     Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents    

First published online 12 December 2006
doi: 10.1242/jcs.03311


Journal of Cell Science 120, 86-100 (2007)
Published by The Company of Biologists 2007
This Article
Right arrow Summary Freely available
Right arrow Full Text
Right arrow Full Text (PDF)
Right arrow Supplementary Material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yamazaki, D.
Right arrow Articles by Takenawa, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yamazaki, D.
Right arrow Articles by Takenawa, T.
Social Bookmarking
 Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?

Rac-WAVE-mediated actin reorganization is required for organization and maintenance of cell-cell adhesion

Daisuke Yamazaki, Tsukasa Oikawa and Tadaomi Takenawa*

Department of Biochemistry, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan


Figure 1
View larger version (78K):
[in this window]
[in a new window]

 
Fig. 1. Effects of cytochalasin D on cell-cell adhesion. MDCK cell layers were treated with DMSO or 10 µM cytochalasin D for the indicated times and then cytochalasin D was washed out, and cells were incubated in fresh medium for the indicated times. (A) Cell layers were fixed and stained for WAVE2, ß-catenin and actin filaments (F-actin). Bar, 10 µm. (B) Vertical sections. Bar, 5 µm. (C) The signal intensities for ß-catenin and WAVE2 of cross sections as shown in A were quantified with ImageJ software. All data are mean ± s.d. *P<0.05; **P<0.0005. Statistical significance was examined using the Student's t-test.

 

Figure 2
View larger version (58K):
[in this window]
[in a new window]

 
Fig. 2. Expression and localization of WAVE2 in epithelial MDCK cells. (A) Comparison of WAVE1 and WAVE2 expression in MDCK cells and mouse embryonic fibroblasts (MEFs). Cell lysates were immunoblotted with anti-WAVE1 and anti-WAVE2 antibodies. (B) Localization of endogenous WAVE2 in MDCK cells. Before (No chelate) or after (Chelate) treatment with EGTA, MDCK cell layers were fixed and then stained with anti-E-cadherin and anti-WAVE2 antibody. The regions from which vertical sections were taken are indicated by white lines. Arrows indicate cell-cell contacts. Bar, 10 µm. (C) Detergent solubility of WAVE2. MDCK cells were treated with 0.5% Triton X-100 containing buffer before fixation. Fixed cell layers were stained with anti-E-cadherin and anti-WAVE2 antibody. Scale bar, 15 µm. (D) Interaction of WAVE2 with adhesion complex proteins. Protein complexes were immunoprecipitated from lysates of DMSO- or cytochalasin D (CytoD)-treated MDCK cells with anti-WAVE2 antisera. Immunoprecipitates were immunoblotted for E-cadherin, ß-catenin, afadin and WAVE2. Negative controls for immunoprecipitations were pre-immune antisera (Pre-Immune IP). (E) DECMA-1 (anti-E-cadherin antibody) inhibits recruitment of WAVE2 to cell-cell contacts. Trypsinized cells were plated on coverslips and treated with DECMA-1. After incubation for 12 hours, cells were fixed and stained for E-cadherin and WAVE2. The lower panels are magnified images of the areas indicated by white squares in each upper panel. Broken lines indicate localization of WAVE2 at the lamellipodial edge. Bars, 15 µm (upper panels); 5 µm (lower panels).

 

Figure 3
View larger version (73K):
[in this window]
[in a new window]

 
Fig. 3. WAVE1 and WAVE2 have redundant functions at cell-cell adhesions. (A) Cells treated with the indicated siRNA were fixed and stained for E-cadherin and actin filaments. Cross and vertical sections were shown. The regions from which vertical sections were taken are indicated by white lines. Bar, 15 µm (cross sections); 5 µm (vertical sections). (B) Signal intensities of E-cadherin and actin filaments in cross sections were quantified with ImageJ software. All data are mean ± s.d. *P<0.05; **P<0.005; ***P<0.0001. Statistical significance was examined using the Student's t-test. (C) Cells treated with siRNAs against CTRL, WAVE1, WAVE2, WAVE1 + WAVE2 and Arp2 were lysed, and WAVE was detected by immunoblotting. (D) Cells treated with siRNAs against CTRL, Abi1 (440 and 764) and HSPC300 (110 and 180) siRNAs were lysed, and WAVE was detected by immunoblotting. The results of two independent experiments are shown. (E) Signal intensities of WAVE, Abi1 and HSPC300 in D were quantified with ImageJ software. All data are mean ± s.d.

 

Figure 4
View larger version (53K):
[in this window]
[in a new window]

 
Fig. 4. Effects of WAVE2 RNAi in MDCK cells. (A) Depletion of WAVE2 in MDCK cells inhibits lamellipodium formation on collagen. CTRL or WAVE2 siRNA-treated cells were trypsinized and then plated on collagen. After 4 hours, cells were fixed and stained for WAVE2 and actin filaments (F-actin). Insets in each panel show the magnified images of the areas indicated by white squares. Bar, 10 µm. The lower graph shows the number of cells with lamellipodia. The percentage of cells with lamellipodia covering 50-100% of the cell perimeter was presented as >50%. 20-50%, 20-50% of the perimeter; <20%, 0-20% of the perimeter. Data are mean ± s.d. Three independent experiments were performed. (B) Morphologies of colonies of CTRL or WAVE2 siRNA-treated cells. Bar, 100 µm. (C,D) Ca2+-switch assay with subconfluent MDCK cell layers. Subconfluent CTRL siRNA (C) or WAVE2 siRNA (D) treated cell layers were treated with 5 mM EGTA for 30 minutes. Cell layers were then treated with Ca2+-containing medium for the indicated times. Cell layers were fixed and then stained for WAVE2 and E-cadherin. Merged images of WAVE2 (red) and E-cadherin (green) are shown in insets. Insets in each panel show the magnified images of the areas indicated by white squares. Scale bar, 15 µm. (E) Quantification of cell-cell contacts with E-cadherin staining. The number of cell-cell contacts with E-cadherin staining was counted during the Ca2+-switch assay in C and D. Data are mean ± s.d. *P<0.05. Statistical significance was examined with the Student's t-test. Three independent experiments were performed.

 

Figure 5
View larger version (80K):
[in this window]
[in a new window]

 
Fig. 5. Recruitment of actin filaments to new cell-cell contact sites. CTRL siRNA (A-F,M-P) or WAVE2 siRNA (G-L,Q-T) treated cells were plated on collagen-coated coverslips and then incubated for 2 or 3 hours as indicated. Cells were fixed and stained for E-cadherin and actin filaments (F-actin). (C,F,I,L) Cells in B,E,H,K photographed at the apical level. (A'-L') Magnified images of the areas in the white squares in A-L, respectively. (B'',E'',H'',K'') Magnified images of B,E,H,K, respectively. Broken lines indicate the overlapping areas in H'' and K''. (M-T) vertical sections. The regions from which vertical sections were taken are indicated by white lines. Arrows indicate the locations of cell-cell contacts. Broken lines indicate the locations of coverslips. Brackets in Q and R indicate the abnormal localization of E-cadherin. Bars, 15 µm (A-L); 5 µm (A'-L'); 7.5 µm (B'',E'',H'',K''); 2.5 µm (M-T).

 

Figure 6
View larger version (72K):
[in this window]
[in a new window]

 
Fig. 6. Maturation of cell-cell adhesions. (A,B) Ca2+-switch assay was performed with CTRL siRNA (A) or WAVE2 siRNA (B) treated confluent cell layers. Confluent cell layers were treated with 4 mM EGTA and then incubated with Ca2+-containing medium for the indicated times. Cell layers were fixed and stained for ß-catenin and actin filaments (F-actin). Arrows indicate disorganized cell-cell adhesions. Bar, 10 µm. (C) Cell-cell contacts were classified into three categories on the basis of the ß-catenin staining pattern. Categories were thick, strong signals (thick); thin, weak signals (thin); and broad or no signal (disorganized). All data are mean ± s.d. Three independent experiments were performed. (D,E) Vertical sections of CTRL siRNAi (D) or WAVE2 siRNA (E) treated cells in the Ca2+-switch assays. Brackets indicate the width of cell-cell adhesions. Bar, 7.5 µm.

 

Figure 7
View larger version (65K):
[in this window]
[in a new window]

 
Fig. 7. WAVE2 functions downstream of Rac cell-cell contact sites. (A) DECMA-1 inhibits the RacCA-induced development of cell-cell adhesions. Cells expressing Myc-tagged RacCA were trypsinized and then re-plated in media containing DECMA-1. Cells were fixed and stained for Myc, WAVE2, and actin filaments. Arrows indicate the areas magnified the right-hand panels. Bars, 15 µm. (B) Effects of WAVE2 RNAi on RacCA-induced development of cell-cell adhesion. Myc-tagged RacCA was transiently expressed in CTRL siRNA and WAVE2 siRNA-treated cells, and its effects on cell-cell adhesions were examined. Cells were fixed and stained for ß-catenin and actin filaments. Arrowheads and arrows indicate the cell-cell contacts in Myc-tagged RacCA expressing or non-expressing cells, respectively. Bar, 7.5 µm. (C) Signal intensities of ß-catenin and actin filaments were quantified in vertical sections with ImageJ software. All data are mean ± s.d. *P<0.05; **P<0.005; ***P<0.0001. Statistical significance was examined using the Student's t-test.

 

Figure 8
View larger version (65K):
[in this window]
[in a new window]

 
Fig. 8. Rac signaling is important for WAVE2 localization at cell-cell contacts. (A) Myc-tagged constitutively active (CA) and dominant-negative (DN) Rac were transiently expressed, and the localization of WAVE2 was examined by immunostaining. The regions from which sections were taken are indicated by white lines. Arrows indicate the positions of cell-cell contacts. Bar, 7.5 µm. (B) In MDCK cells transiently expressing GFP-Rac, the Rac-GFP formed clusters at sites other than cell-cell contacts (arrows). These clusters contained E-cadherin, actin filaments, and WAVE2. Bar, 15 µm. (C) CTRL or Tiam1 siRNA-treated cells were fixed and stained for WAVE2, E-cadherin, and actin filaments (F-actin). The regions from which vertical sections were taken are indicated by white lines. Arrows indicate localization of WAVE2. Bar, 10 µm.

 

Figure 9
View larger version (44K):
[in this window]
[in a new window]

 
Fig. 9. WAVE2 is recruited to cell-cell contacts through its N-terminus. (A) GFP-tagged WAVE2 constructs were expressed and their localization was examined and quantified in MDCK cells. All data are mean ± s.d. Three independent experiments were performed. WHD, WAVE-homology domain; B, basic domain; PR, proline-rich domain; VCA, verprolin-homology, cofilin-homology and acidic domain. Bar, 15 µm. (B) Pull-down assay. (Left panels) MDCK cell lysates were incubated with GST-tagged truncated WAVE2 proteins and immunoblotted with anti-Abi1 and anti-HSPC300 antibodies. (Right panels) MDCK cells in which GFP-tagged Abi1 or HA-tagged HSPC300 were expressed were lysed and then incubated with GST-tagged full-length WAVE2 proteins and immunoblotted with anti-GFP and anti-HA antibodies, respectively. (C) WHD-GFP was transiently expressed in CTRL, Abi1 or HSPC300 siRNA-treated cells. Expression levels of WHD-GFP were similar in these cells (data not shown). Cells were fixed and the localization of WHD-GFP was examined and quantified. These data indicate that localization of WHD was dependent on Abi1 and HSPC300. Bar, 30 µm. All data are mean ± s.d. of three independent experiments.

 

Figure 10
View larger version (21K):
[in this window]
[in a new window]

 
Fig. 10. A model for the function of WAVE2 during cell-cell adhesion formation in MDCK cells. (1) Non-adherent cells form WAVE2-dependent lamellipodial protrusions at the cell periphery. Such membrane protrusions facilitate efficient formation of new cell-cell contacts. (2) Clusters of cadherin appear at the sites of cell-cell contact. After cells contact each other, lamellipodial activity continues to expand the cell-cell contact and fill any gaps between the cells. (3) As cadherin-dependent cell-cell adhesions develop, WAVE2 is recruited and makes new actin structures along the adhesion, thereby stabilizing the adhesions. (4) In subconfluent cell layers, lamellipodial protrusions are often observed at sites of cell-cell contact and increase the area of cell-cell contact. (5) In confluent cell layers, membrane protrusions are not observed by light microscopy and WAVE2 is concentrated at cell-cell adhesions. At the basal part of the lateral membrane, WAVE2-mediated actin assembly pushes the membrane and promotes formation of new cell-cell contacts.

 

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati   Add to Twitter Twitter    What's this?




© The Company of Biologists Ltd 2007