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

First published online 3 January 2006
doi: 10.1242/jcs.02738

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 Zhang, W. Articles by Garlick, J. A. Search for Related Content PubMed PubMed Citation Articles by Zhang, W. Articles by Garlick, J. A.

E-cadherin loss promotes the initiation of squamous cell carcinoma invasion through modulation of integrin-mediated adhesion

Weitian Zhang^1,*, Addy Alt-Holland^1,*, Alexander Margulis¹, Yulia Shamis¹, Norbert E. Fusenig², Ulrich Rodeck³ and Jonathan A. Garlick^1,

¹ Division of Cancer Biology and Tissue Engineering, Department of Oral and Maxillofacial Pathology, School of Dental Medicine, Tufts University, 55 Kneeland Street, Boston, MA 02111, USA
² German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany
³ Department of Dermatology and Cutaneous Biology, Thomas Jefferson University, 233 South 10th Street, Philadelphia, PA 19107, USA

View larger version (95K): [in a new window]   Fig. 1. Invasive tumor formation after in vivo transplantation was dependent upon loss of E-cadherin. (A-C) Clinical appearance of 3D surface transplants (left panel): 3D tissues were engineered in vitro as 4:1 mixtures of HEK and either pBabe-, H-2kd-EcadC25- or H-2kd-Ecad-II-4-expressing cells and transplanted to the dorsum of nude mice for 4 weeks. Grafts harboring E-cadherin-deficient, H-2kd-Ecad-expressing II-4 cells generated nodular tumors with focal areas of erythema (C, arrow). By contrast, grafts containing II-4-pBabe (A) or H-2kd-EcadC25-expressing cells (B) generated normal grafts without any evidence of tumor formation. (D-F) Histological appearance of 3D surface transplants. Tumors generated in transplants of 4:1 mixtures of HEK and H2kd-Ecad-expressing II-4 cells demonstrated invasion of individual cells (F, inset) and small clusters of poorly differentiated tumor cells (F). By contrast, grafted mixtures of HEK with either pBabe- (D) or H-2kd-EcadC25-expressing II-4 cells (E) generated normal epithelia without any evidence of residual tumor cells. Bars, 20 µm.

View larger version (104K): [in a new window]   Fig. 2. Tumor cell invasion was linked to the cytoplasmic redistribution of ß-catenin and increased expression of 2 integrin subunit. (A-F) Four weeks after in vivo transplantation of 4:1 mixtures, excised tissues were stained by double immunofluorescence for ß-gal (green) and ß-catenin (red). In transplanted mixtures comprised of HEK and control pBabe- (A) or EcadC25-expressing II-4 cells (B), ß-gal-positive II-4 cells were not present and ß-catenin was localized at cell-cell borders of the HEK cells. By contrast, as can be seen by the same immunostaining pattern, mixtures of HEK and H2kd-Ecad-expressing II-4 cells demonstrated islands of invasive tumor cells that showed cytoplasmic co-localization of ß-gal (C, green) and ß-catenin (D, red), indicating that II-4 cell invasion was associated with loss of ß-catenin from cell junctions and abrogation of cell-cell adhesion. The invading clusters of tumor cells showed increased expression of 2 integrin subunit (E, green), whereas in control grafts comprised of HEK and control pBabe-II-4 cells, 2 integrin expression was limited to the basal HEK at the epithelial-stromal interface (F, green). Bars, 10 µm.

View larger version (71K): [in a new window]   Fig. 3. Invasion of cell-cell adhesion-deficient II-4 cells was preceded by their attachment and localization at the epithelial-stromal interface in 3D in vitro tissues. In vitro 3D cultures were constructed as 4:1 mixtures of HEK with either pBabe-, H-2kd-EcadC25- or H-2kd-Ecad-II-4 cells and were grown at an air-liquid interface for 7 days. Tissues were stained by double immunofluorescence for the basement membrane component Type IV collagen (red) and ß-gal (green). Constructs harboring mixtures of HEK with either pBabe- (A, arrows) or H-2kd-EcadC25-expressing II-4 cells (B, arrows) demonstrated tumor cells that were limited to the suprabasal layers of the epithelium. By contrast, tissues harboring H-2kd-Ecad-II-4 cells demonstrated ß-gal-positive cells in the basal layer of the epithelium, adjacent to the Type IV collagen present at the epithelial-stromal interface (C, arrows). Bars, 10 µm.

View larger version (19K): [in a new window]   Fig. 4. Loss of E-cadherin increases II-4 cell attachment to extracellular matrix proteins. 2D cultures of pBabe-, H-2kd-EcadC25- or H-2kd-Ecad-II-4 cells were trypsinized briefly and replated onto plates coated with either Type I collagen (A), Type IV collagen (B), laminin-1 (C) or fibronectin (D) for 20 minutes. Attached cells were quantified by optical density determined at 590 nm. Results are calculated as the mean ± s.d. of four replicates and experiments were repeated four times. P<0.01 for H-2kd-Ecad-II-4 cells vs control cells in adherence to Types I and IV collagen substrates. P<0.05 for H-2kd-Ecad-II-4 cells vs control cells in adherence to laminin-1 and fibronectin substrates.

View larger version (28K): [in a new window]   Fig. 5. Loss of E-cadherin in II-4 cells is linked to increased expression of 2, 3 and ß1 intregrin subunits. (A) Expression levels of integrin subunits on the cell surface of pBabe-, H-2kd-EcadC25- or H-2kd-Ecad-II-4 in 2D cultures. The three cell types were extracted and 5 µg samples of membrane proteins were immunoblotted in non-reducing conditions with anti-2, anti-3 or anti-ß1 integrin antibodies and analyzed by ECL. Expression levels of these specific integrin subunits were markedly increased in E-cadherin deficient H-2kd-Ecad-II-4 cells, when compared with their levels in the control pBabe- and H-2kd-EcadC25-II-4 cells (left panel). Scanning densitometry of the relative intensity of the presented immunoblots is shown on the right. (B) FACS analysis of pBabe-, H-2kd-EcadC25- or H-2kd-Ecad-II-4 cells immunoreacted with antibodies against specific integrin subunits. Elevated levels of cell surface 2, 3 and ß1 subunits in H-2kd-Ecad-II-4 cells (black line in right-hand panels) in comparison to control pBabe-II-4 cells (gray line) were identified by increased fluorescence intensity seen in E-cadherin deficient cells. The expression levels of 2, 3 and ß1 subunits were similar in H-2kd-EcadC25-II-4 (black line in left-hand panels) and the control pBabe-II-4 cells (gray line) as seen by the superimposition of these lines. (C) Functional blocking of ß1 integrin in H-2kd-Ecad-II-4 cells. Two-dimensional cultures of Ecad-deficient H-2kd-Ecad-II-4 cells were trypsinized briefly and replated for 15 minutes onto Type I and IV collagens or laminin-1 substrates, in the presence or absence of ß1-integrin-blocking antibodies. Attached cells were quantified by optical density determined at 590 nm. Results are calculated as the mean ± s.d. of four replicates and experiments were repeated three times. P<0.001 for cell adherence to Type I collagen, Type IV collagen and laminin-1 substrates compared with levels in blocked controls.