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

First published online 9 August 2005
doi: 10.1242/jcs.02508

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 Related articles in JCS 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 Kim, Y. J. Articles by Knudsen, K. A. Search for Related Content PubMed PubMed Citation Articles by Kim, Y. J. Articles by Knudsen, K. A.

Modulating the strength of cadherin adhesion: evidence for a novel adhesion complex

Young J. Kim¹, Christa Sauer², Karla Testa^3,*, James K. Wahl¹, Robert A. Svoboda¹, Keith R. Johnson¹, Margaret J. Wheelock¹ and Karen A. Knudsen^2,

¹ University of Nebraska Medical Center, College of Dentistry, 769605 Nebraska Medical Center, Omaha, NE 68583, USA
² Lankenau Institute for Medical Research, 100 Lancaster Avenue, Wynnewood, PA 19096, USA
³ Biology Department, Saint Joseph's University, 5600 City Avenue, Philadelphia, PA 19131, USA

View larger version (66K): [in a new window]   Fig. 1. N-cadherin/ER-LBD construct and its expression in L cells. (A) Full-length human N-cadherin (Ncad) cDNA fused at the 3' end to a modified estrogen receptor ligand-binding domain (ER-LBD) that binds 4-hydroxytamoxifen (4OHT) but not 17ß-estradiol. Pre and pro regions of the N-cadherin protein are cleaved off in cells to generate the mature N-cadherin. tm, transmembrane domain; cyto, cytoplasmic domain. (B) Western immunoblot analysis of N-cadherin, NcadER, -catenin, ß-catenin and p120ctn with GAPDH as a control. L cells stably expressing the N-cadherin/ER-LBD construct (LNER) were treated with 100 nM 4OHT (+) or vehicle (-) for 5 hours. Non-transfected L cells (L) and L cells expressing wild-type N-cadherin (LN), in the absence of 4OHT (-), served as negative and positive controls, respectively. Cells were extracted with 0.5% NP40 and soluble proteins resolved by SDS-PAGE, transblotted electrophoretically to nitrocellulose and immunoblotted with antibodies to N-cadherin, the three catenins, and GAPDH. The parent L cells lacked endogenous N-cadherin, and - and ß-catenin were not detected in these cells as the proteins are degraded in the absence of a cadherin. Similar levels of NcadER (molecular weight 150 kDa) were detected in LNER cells with or without 4OHT treatment. - and ß-catenin were detected in LN cells and at similar levels with or without 4OHT in LNER cells, suggesting formation of a cadherin/catenin complex under either condition. p120ctn, which is stable in L cells lacking a cadherin, was present at similar levels in L, LN and LNER cells with or without 4OHT. Positions of molecular weight markers are indicated on the right. (C) Immunofluorescent light microscopic localization of NcadER in LNER cells with or without 4OHT. LNER monolayers plus or minus 100 nM 4OHT overnight were fixed and stained with anti-N-cadherin. The NcadER signal in cells with 4OHT appeared more organized and was particularly strong at sites of cell-cell contact, compared to cells without 4OHT. (D) Localization of N-cadherin in L cells and LN cells. As expected, L cells lacked N-cadherin, whereas LN cells exhibited strong N-cadherin staining, particularly at cell-cell borders. Bar, 50 µm.

View larger version (116K): [in a new window]   Fig. 2. Adhesive activity of NcadER activated or not by 4OHT. (A) Aggregation of L cells, LN cells, and LNER cells with or without 4OHT treatment. Trypsin harvested cells were pipetted into a single cell suspension and identical numbers placed in hanging drop cultures, with LNER cells treated with or without 100 nM 4OHT. After 24 hours the cells were collected, triturated and photographed under phase-contrast microscopy. Non-transfected parent L cells (L) failed to aggregate, whereas LN cells formed tightly compacted aggregates. LNER cells without 4OHT (LNER - 4OHT) formed large aggregates only partly dispersed by pipetting and where individual cells were clearly discernable. LNER cells with 4OHT (LNER + 4OHT) formed tightly compacted aggregates similar to those formed by LN cells. Compacted aggregates were highly resistant to dispersion by trituration, individual cells were difficult to discern, and a phase-dense ring was observed around the aggregate (arrows). (B) Behavior of LNER monolayers with or without 4OHT. 50,000 cells were plated in 100 µl medium in wells of six-well culture dishes. After the cells attached (5 hours), the dishes were gently flooded with 2 ml medium, with or without 100 nM 4OHT, and the cells were observed for 3 days. The cultures were photographed within (top panels), and outside of (bottom panels) the original plating area. The line defines the edge of the plating area, which is left of the line. LNER cells without 4OHT piled up on one another in the plating area, detached, and eventually seeded new colonies outside the plating area. In the presence of 4OHT the LNER cells stayed predominantly as a monolayer within the plating area, and seeded far fewer colonies outside this area, compared to cells without 4OHT.

View larger version (43K): [in a new window]   Fig. 3. Proteins co-immunoprecipitating with NcadER activated or not by 40HT. (A) Immunoblot analysis of N-cadherin immunoprecipitates from LN cells and LNER cells with or without 4OHT treatment. LNER cells without (-) or with (+) 100 nM 4OHT overnight were extracted with 0.5% NP40 and the NcadER immunoprecipitated with anti-N-cadherin using identical conditions. N-cadherin was immunoprecipitated from LN cells as a positive control. The immunoprecipitates were resolved by SDS-PAGE, transblotted to nitrocellulose and immunoblotted with antibodies to N-cadherin, ß-catenin, -catenin, actin and p120ctn. (B) Quantification of N-cadherin co-immunoprecipitating proteins normalized to N-cadherin. Means of two separate experiments are shown. (C) N-cadherin immunoprecipitation of 35S-radiolabeled LNER cells, treated or not with 4OHT, analyzed by SDS-PAGE and autoradiography. The control lane was processed using hybridoma supernatant without

View larger version (64K): [in a new window]   Fig. 4. Immunofluorescence light microscopic detection of NcadER following low concentration NP40 treatment of living LNER cells treated or not with 4OHT. Monolayer cultures of LNER cells previously treated or not with 100 nM 4OHT for 24 hours were exposed to 0.05% NP40 for 1 minute at room temperature prior to fixing and staining the cells with anti-N-cadherin. In the absence of 4OHT treatment much of the NcadER was removed by NP40 treatment of live cells (A). In contrast, the NcadER signal in LNER cells treated with 4OHT was significantly greater and was apparent at cell-cell borders (B). Bar, 50 µm.

View larger version (130K): [in a new window]   Fig. 5. Effect of cytochalasin B on aggregation and compaction of LNER cells treated or not treated with 4OHT. LNER cells were trypsinized and placed in suspension culture in the presence of 1 µM cytochalasin B, with (+) or without (-) 4OHT, for 24 hours. Cytochalasin B disrupted F-actin as evidenced by phalloidin staining (right panels). In the absence of 4OHT cytochalasin B had a negative impact on LNER aggregation, when compared to cells without either agent. In the presence of 4OHT the negative effect of cytochalasin B on LNER cell aggregation was attenuated. Moreover, the line of phase density (arrows) seen in compacted aggregates of 4OHT-treated LNER cells without cytochalasin B were observed in the smaller aggregates formed by LNER cells treated with both 4OHT and cytochalasin B.

View larger version (65K): [in a new window]   Fig. 6. Immunofluorescence light microscopic analysis of the effect of cytochalasin B on ß-catenin in LNER cells with or without 4OHT. LNER cells with or without 4OHT treatment were cultured in the presence of 1 µM cytochalasin B (Cyto B) to disrupt the actin cytoskeleton (see Fig. 5). The cells were fixed and stained with anti-ß-catenin to detect the NcadER/catenin complex. In LNER cells treated with Cyto B, but not 4OHT, the ß-catenin staining was weak and not particularly distinct at cell-cell borders (A). However, in cells treated with 4OHT and Cyto B, the ß-catenin staining was strong and distinct at cell-cell borders (B).

View larger version (141K): [in a new window]   Fig. 7. Immunofluorescent light microscopic localization of F-actin, tubulin and vimentin in LNER cells treated or not with 4OHT. Monolayer cultures of LNER cells were treated with (+) or without (-) 100 nM 4OHT for 24 hours prior to being fixed and stained for F-actin, tubulin and vimentin as markers of the actin, microtubule, and intermediate filament systems, respectively. Rhodamine-phalloidin was used to detect F-actin, whereas antibodies were used to detect tubulin and vimentin. (A) Cells at low magnification. (B) Cells at higher magnification using confocal microscopy. Although no differences were apparent in F-actin or tubulin staining in LNER cells treated or not treated with 4OHT, the vimentin appeared better organized in the 4OHT-treated cells, compared to cells without 4OHT. Bar, 50 µm.

View larger version (52K): [in a new window]   Fig. 8. Evidence for vimentin association with the cadherin/catenin complex. (A) Confocal microscopic localization of vimentin in L, LN and LNER cells treated or not treated with 4OHT. Monolayer cultures of parent L cells, LN cells expressing N-cadherin and LNER cells without (-) or with (+) 4OHT were fixed and stained with anti-vimentin. The vimentin in L cells, which lack a cadherin and do not aggregate, was largely perinuclear. In LN cells, which aggregate and compact, the vimentin was organized into filaments that extended throughout the cell body. In LNER cells lacking 4OHT treatment, which aggregate but do not compact, the vimentin pattern more closely resembled that in the L cells. In contrast, in LNER cells treated with 4OHT, which aggregate and compact, the vimentin pattern was closer to that seen in the LN cells. (B) Immunoblot analysis of vimentin in LNER cells with or without 4OHT. GAPDH was included as a sample loading control. (C) Immunoblot analysis of N-cadherin (N-cad) and vimentin (Vim) immunoprecipitates from LNER cells with or without 4OHT treatment. Control (Cont) was monoclonal anti-myc. Bar, 50 µm.

View larger version (75K): [in a new window]   Fig. 9. siRNA knock down of vimentin and its effect on the aggregation and compaction of LNER cells treated or not with 4OHT. (A) Immunoblot analysis of vimentin expression in LNER cells treated or not with 4OHT and transfected with control or vimentin siRNAs. GAPDH served as a sample loading control. (B) Quantification of the vimentin signal (means of two experiments), showing a similar knockdown of vimentin (50%) by vimentin siRNA, compared to control siRNA, in LNER cells treated or not with 4OHT. (C) Aggregation/compaction assay of LNER cells with or without 4OHT treatment and transfected with control or vimentin siRNAs. Control siRNA had no effect on the aggregation of LNER cells lacking 4OHT treatment, or on the aggregation or compaction (arrows) of LNER cells treated with 4OHT. Nor did vimentin siRNAs affect the aggregation of LNER cells lacking 4OHT treatment. In contrast, vimentin siRNAs inhibited the compaction of aggregates of 4OHT-induced LNER cells.

View larger version (94K): [in a new window]   Fig. 10. Effect of 4OHT on p120ctn localization in LNER cells and association of p120ctn and vimentin in L cells. (A) Immunofluorescent light microscopic localization of p120ctn in LNER cells treated (+) or not (-) with 4OHT. (B) Immunoblot analysis of p120ctn in immunoprecipitates from non-transfected parental L cells immunoprecipitated with anti-vimentin (Vm) or control (Cont) anti-myc antibody. Immunoblot analysis of L cell extract (Extract) indicates the position of p120ctn. (C) Non-transfected parental L cells were processed for immunofluorescence using rabbit anti-p120ctn (p120) and mouse anti-vimentin (Vm) followed by Cy3-labeled anti-rabbit IgG and FITC-labeled anti-mouse IgG. Arrows indicate regions of the cell where both p120ctn and vimentin are abundant and colocalize. Bar, 50 µm.