p120-catenin modulates airway epithelial cell migration induced by cigarette smoke

Abstract

Cigarette smoking has been linked to almost all major types of cancer. Emerging evidence suggests that smoking initiates transformed cell growth and migration by disrupting cell-cell interactions in the polarized mucosal epithelium. Together with other adherens junction proteins, p120-catenin (p120ctn) maintains cell-cell adhesion through its direct interaction with E-cadherin (E-cad). Mislocalization and/or loss of p120ctn have been reported in all lung cancer subtypes and are related to poor prognosis. Here, we showed that p120ctn modulates smoke-induced cell migration via the EGFR/Src-P pathway. Chemical blockade of EGFR/Src signaling inhibited smoke-induced activation of cofilin (an actin severing protein) and promoted cell migration in the presence of p120ctn but had little effect on blocking migration in the absence of p120ctn. These data suggested that smoke-induced cell migration was mediated via an EGFR/Src-dependent signaling pathway in cells that expressed p120ctn, but upon loss of p120ctn, migration continued to occur via an alternative, EGFR/Src-independent pathway. Thus, gradual loss of membrane p120ctn with lung cancer progression may contribute to reduced effectiveness of conventional chemotherapies, such as those directed against EGFR.

Fig. 1

siRNA knockdown of p120ctn in primary HBE cells disrupts adherens junctions. (A) Immunofluorescent staining of adherens junction markers p120-catenin (p120ctn) in untreated cells (Ctrl) and cells treated with smoke medium at 24 mg/m³ TSP (Smk) for 24 h. (B) Immunofluorescent staining of p120ctn, β-ctn, and E-cad in scrambled siRNA-transfected cells (p120WT) and p120ctn knockdown cells (p120KD) at 48 h posttransfection. Bar represents 50 μm. (C) Cell lysates from p120WT and p120KD cells (transfected with p120siRNA pool 1 or p120siRNA pool 2) were analyzed by Western blots probed with p120ctn, β-ctn and E-cad antibodies at 48 h posttransfection. GAPDH was used as a loading control. (D) Quantitation of p120ctn, β-ctn and E-cad levels in p120WT and p120KD cells at 48 h posttransfection by densitometry. The levels of proteins in p120KD cells (black bars) were normalized to those in p120WT cells (white bars, designated 1-fold). Results shown are the average of three independent experiments. **p < 0.01, p120KD versus p120WT cells.

Fig. 2

Smoke-mediated signaling events in HBE cells. (A–D) Smoke causes p120ctn-independent phosphorylation of EGFR/Src and dephosphorylation of Limk1/Cofilin (Cof) in HBE cells after 3 h of exposure. (A) Cell lysates obtained from p120WT and p120KD cells after a 3 h exposure to smoke-free (Ctrl) or smoke medium at 48 mg/m³ TSP (Smk) were analyzed by Western blot using antibodies directed against Tyr-phosphorylated EGFR (EGFR-P), Tyr⁴¹⁶-phosphorylated Src (Src-P), Thr⁵⁰⁸-phosphorylated Limk1 (Limk1-P) and Ser³-phosphorylated Cofilin (Cof-P). Equal loading of lysates was confirmed using antibodies directed against total EGFR, Src and GAPDH. Blots are representative of three independent experiments. (B) Relative fold change in EGFR-P, Src-P, Limk1-P and Cof-P following smoke exposure in p120WT (grey bar) and 120KD cells (black bar) was normalized relative to Ctrl (white bars, designated 1-fold). Results shown are the mean ± SEM of three independent experiments. **p < 0.01, p120KD or p120WT versus Ctrl cells. (C) HBE cell lysates exposed to smoke-free (Ctrl) or smoke medium at 12, 24 and 48 mg/m³ TSP (Smk) for 3 h were analyzed by Western blot using antibodies directed against Src-P, Limk1/2-P and Cof-P. Levels of total Src and GAPDH were used as loading controls. (D) Relative fold change in Src, Limk1 and Cof phosphorylation induced by smoke was normalized to Ctrl (designated 1-fold) at 3 h and plotted as dose-response curves. Results shown are the mean ± SEM of three independent experiments. **p < 0.01, Smk-exposed versus Ctrl cells.

Fig. 3

EGFR and Src inhibitors fail to block decreased Cof-P and increased cell migration in response to smoke when cells lack p120ctn. (A–D) p120WT and p120KD cells were treated with smoke-free (Ctrl) or smoke medium at 48 mg/m³ TSP (Smk) for 3 h in the presence of 1 μM AG1478 (A and B) or 12.5 μM PP2 (C and D) to inhibit EGFR or Src, respectively. Western blot (A and C) and densitometric analysis (B and D) of Cof-P in Smk-treated cells were normalized to Ctrl (designated 1-fold). Data are expressed as the mean ± SEM of three independent experiments. (E) P120WT and p120KD cells were loaded on Boyden chambers and incubated with Ctrl (white bar) or smoke medium at 48 mg/m³ TSP (black bar) in the presence of DMSO (vehicle control) or 12.5 μM PP2 for 4 h. Migrating cells were quantified by measuring the OD at 595 nm and plotted as the mean ± SEM of three independent chambers. **p < 0.01, smoke-treated p120WT or p120KD cells versus respective control. **p < 0.01, smoke-treated p120KD cells versus smoke-treated p120WT cells.

Fig. 4

Smoke mediates EGFR/Src-dependent and independent cell migration. In the normal human airway, adherens junctions (p120ctn, β-ctn and E-cad) are intact on the epithelial cell membrane. Smoke induces rapid activation of membrane Src through EGFR. Src activation promotes cell migration through suppression of Cof-P. When airway epithelial cells express p120ctn, EGFR and Src specific inhibitors (as those used in chemotherapy) help to stabilize p120ctn on the membrane where it functions as a tumor suppressor. In the advanced stages of lung cancer, p120ctn is translocated into the cytoplasm by smoke where it is degraded or modified. Membrane loss of p120ctn results in disruption of adherens junctions and decreased cell–cell adhesion. Src continues to be activated through EGFR but EGFR and Src inhibitors fail to block cell migration because loss of tumor-suppressing membrane p120ctn, activates an alternative EGFR/Src-independent pathway (denoted with a dotted-line). Cell migration continues via this alternative pathway in the absence of p120ctn, bypassing traditional EGFR/Src signaling and thus providing a potential mechanism of chemotherapeutic resistance.