Carcinoma cells induce lumen filling and EMT in epithelial cells through soluble E-cadherin-mediated activation of EGFR

Abstract

In epithelial cancers, carcinoma cells coexist with normal cells. Although it is known that the tumor microenvironment (TME) plays a pivotal role in cancer progression, it is not completely understood how the tumor influences adjacent normal epithelial cells. In this study, a three-dimensional co-culture system comprising non-transformed epithelial cells (MDCK) and transformed carcinoma cells (MSV-MDCK) was used to demonstrate that carcinoma cells sequentially induce preneoplastic lumen filling and epithelial-mesenchymal transition (EMT) in epithelial cysts. MMP-9 secreted by carcinoma cells cleaves cellular E-cadherin (encoded by CDH1) from epithelial cells to generate soluble E-cadherin (sE-cad), a pro-oncogenic protein. We show that sE-cad induces EGFR activation, resulting in lumen filling in MDCK cysts. Long-term sE-cad treatment induced EMT. sE-cad caused lumen filling by induction of the ERK signaling pathway and triggered EMT through the sustained activation of the AKT pathway. Although it is known that sE-cad induces MMP-9 release and consequent EGFR activation in tumor cells, our results, for the first time, demonstrate that carcinoma cells can induce sE-cad shedding in adjacent epithelial cells, which leads to EGFR activation and the eventual transdifferentiation of the normal epithelial cells.

Keywords: EGFR; MMP-9; Soluble E-cadherin; Transdifferentiation.

Fig. 1.

Co-culture of carcinoma cells with MDCK cysts disrupt luminal architecture. MDCK–RFP cysts (red) were formed by culturing the cells for 72 h in Matrigel™ (A). MSV-MDCK–GFP cells (green) were then added in co-cultures and imaged after 4 h (B), 8 h (C) and 24 h (D). The arrow in D points to a filopodium from an MSV-MDCK cell contacting a cyst. Control cysts were cultured alone for an additional 24 h (E). MDCK–RFP cysts were treated with conditioned medium (CM) from MSV-MDCK cells for 24 h (F). Lumen-filled cysts in three independent experiments were counted and compared to controls after 24 h in the co-culture (G) or after exposure to the conditioned medium from tumor cells for 24 h (H). Results are mean±s.e.m. **P<0.005 (Student's t-test).

Fig. 2.

Conditioned medium from co-culture contains active MMP-9, which is crucial for lumen filling and mediates sE-cad shedding in MDCK cysts. (A) Representative immunoblot from three independent experiments showing MMP-9 levels in the supernatant from MDCK 3D cultures. Quantification data are shown underneath the blot and are expressed as mean±s.e.m. from three independent experiments. (B) Representative zymogram from three independent experiments showing MMP-9 activity. Lane 1, MDCK control; lane 2, co-culture; lane 3, co-culture+10 µM MMP-9 inhibitor. The intensity of MMP-9 bands normalized to control is shown. (C) Confocal images showing co-culture in presence and absence of 10 µM MMP-9 inhibitor after 24 h treatment. Note the absence of lumen filling with MMP-9 inhibition (right panel). (D) Quantification of lumen filling in MDCK cysts in the presence and absence of MMP-9 inhibitor from an average of three independent experiments. Results are mean±s.e.m. ***P<0.0001 (Student's t-test). (E) Representative immunoblot from two independent experiments showing E-cadherin levels in MDCK and MSV-MDCK cells. β-actin is used as a loading control. (F) Diagrammatic representation of the Transwell co-culture system. (G) Top panel, zymogram showing MMP-9 activity in the conditioned medium from the bottom chamber of the Transwell. A dose-dependent increase in MMP-9 levels with increasing number of MSV-MDCK cells in the bottom chamber was observed. 10 µM of MMP-9 inhibitor was used to block MMP-9 activity in the bottom chamber. Bottom panel, immunoblot showing sE-cad levels in the conditioned medium (CM) in the bottom chamber of the Transwell. (H) Immunoblot showing increased sE-cad levels in the co-culture conditioned medium. Note that in the presence of MMP-9 inhibitor, sE-cad levels were reduced. (I) Immunoblot showing sE-cad levels in the bottom chamber of a Transwell assay consisting of MCF10A and MDA-MB435S cell lines. Quantification data are shown as mean±s.e.m. from three experiments underneath the blots for A,G,H and I.

Fig. 3.

sE-cad is necessary for induction of lumen filling in MDCK cysts. (A) Representative confocal images showing lumen filling and disrupted epithelial architecture induced by 10 μg/ml purified sE-cad over 24 h. (B) Quantification of lumen filling from three independent experiments. Results are mean±s.e.m. ***P<0.0001 (Student's t-test). (C) Immunoblot showing sE-cad levels in the conditioned medium before and after immunodepletion of sE-cad. Quantification data represent mean±s.e.m. from three independent experiments. Note that the immunodepleted sE-cad is bound to Protein G beads. (D) Representative confocal images of control cysts, conditioned medium (CM)-treated cysts and cysts treated with sE-cad-depleted conditioned medium. Cysts were stained for actin (red) and TOPRO-3 (nuclear marker, blue). (E) Quantification of lumen-filled cysts in presence of conditioned medium and sE-cad-depleted conditioned medium. Results are mean±s.e.m. for B,E and the blot shown in C. ***P<0.001 (Student's t-test).

Fig. 4.

MSV-MDCK cells induce lumen filling in MDCK cells by sE-cad-mediated activation of EGFR and downstream AKT and ERK1/2 pathways in MDCK cysts. (A) Representative immunofluorescence images showing increased pEGFR expression in MDCK cysts treated with MSV-MDCK cells, conditioned medium (CM), sE-cad, EGF with or without CL-387,785 (CL). Immunofluorescence shows actin (red), E-cadherin (blue) and pEGFR (green). (B) Representative immunoblot from three independent experiments showing pEGFR, total EGFR, phosphoryalted AKT (pAKT), phosphorylated ERK1/2 (pERK1/2) and total AKT and ERK1/2 levels in MDCK 3D cyst lysates co-cultured with MSV-MDCK cells, conditioned medium, sE-cad and EGF for 4 h. 1 µM CL-387,785, an EGFR kinase inhibitor was used to block EGFR activation for 4 h where indicated. EGF treatment was for 15 min. (C) Co-immunoprecipitation of pEGFR (Tyr1068) with Grb2 in MDCK 3D cysts treated with conditioned medium or sE-cad at 2 h and 4 h. (D) Graphs represent quantification data as mean±s.e.m. from three independent experiments. *P<0.05, **P<0.005 (Student's t-test).

Fig. 5.

Lumen filling is a consequence of reduced apoptosis and increased proliferation. (A) Immunofluorescence images showing increased Ki67 and Bcl2 expression in MDCK cysts treated with conditioned medium (CM) and sE-cad for 48 h. Images were obtained from staining cysts with anti-Bcl2 antibody (green), phalloidin–Alexa-Fluor-546 (actin, red) and Ki67 (blue). (B) Representative immunoblot showing Ki67, Bcl2 and cyclin D1 expression in MDCK cysts treated with sE-cad and conditioned medium. 1 µM CL-387,785 (CL) EGFR inhibitor was used where indicated. Quantification data represent mean±s.d. from two independent experiments.

Fig. 6.

sE-cad and conditioned medium induces an EMT-like phenotype in MDCK cysts. (A) Phase-contrast images of control MDCK cysts and cysts treated with conditioned medium (CM) and sE-cad for 96 h. (B) Immunofluorescence staining with different EMT markers. Representative merged confocal images showing actin (red), fibronectin (green), N-cadherin (green) and MMP-9 (blue).

Fig. 7.

AKT is involved in the induction EMT in MDCK cysts. (A,B) Immunofluorescence showing EMT expression in cysts treated with sE-cad (A) and conditioned medium (CM; B) in presence of inhibitors at 96 h. U0126 and LY294002 were used at 1 µM each. Representative confocal images obtained from staining cysts with anti-fibronectin antibody (green), anti-N-cadherin antibody (blue), phalloidin–Alexa-Fluor-546 (for actin, red) are shown. (C,D) Quantification of cysts displaying an EMT phenotype after 96 h with sE-cad treatment (C) or conditioned medium treatment (D) in presence of inhibitors. *P<0.05, ***P<0.001 (Student's t-test). (E,F) Representative immunoblots from two independent experiments showing fibronectin and N-cadherin levels in sE-cad-treated cysts (E) and conditioned-medium-treated cysts (F) in the presence of inhibitors. U0126 and LY294002 were used at 1 µM each. β-actin was used as a loading control. Quantification data represent mean±s.d. from two independent experiments.

Fig. 8.

Proposed model for sequential lumen filling and EMT induced by carcinoma cells in MDCK cysts. See text for details.