Inhibition of mTORC1 induces loss of E-cadherin through AKT/GSK-3β signaling-mediated upregulation of E-cadherin repressor complexes in non-small cell lung cancer cells

Abstract

Background: mTOR, which can form mTOR Complex 1 (mTORC1) or mTOR Complex 2 (mTORC2) depending on its binding partners, is frequently deregulated in the pulmonary neoplastic conditions and interstitial lung diseases of the patients treated with rapalogs. In this study, we investigated the relationship between mTOR signaling and epithelial mesenchymal transition (EMT) by dissecting mTOR pathways.

Methods: Components of mTOR signaling pathway were silenced by shRNA in a panel of non-small cell lung cancer cell lines and protein expression of epithelial and mesenchymal markers were evaluated by immunoblotting and immunocytochemistry. mRNA level of the E-cadherin repressor complexes were evaluated by qRT-PCR.

Results: IGF-1 treatment decreased expression of the E-cadherin and rapamycin increased its expression, suggesting hyperactivation of mTOR signaling relates to the loss of E-cadherin. Genetic ablation of rapamycin-insensitive companion of mTOR (Rictor), a component of mTORC2, did not influence E-cadherin expression, whereas genetic ablation of regulatory-associated protein of mTOR (Raptor), a component of mTORC1, led to a decrease in E-cadherin expression at the mRNA level. Increased phosphorylation of AKT at Ser473 and GSK-3β at Ser9 were observed in the Raptor-silenced NSCLC cells. Of the E-cadherin repressor complexes tested, Snail, Zeb2, and Twist1 mRNAs were elevated in raptor-silenced A549 cells, and Zeb2 and Twist1 mRNAs were elevated in Raptor-silenced H2009 cells. These findings were recapitulated by treatment with the GSK-3β inhibitor, LiCl. Raptor knockdown A549 cells showed increased expression of N-cadherin and vimentin with mesenchymal phenotypic changes.

Conclusions: In conclusion, selective inhibition of mTORC1 leads to hyperactivation of the AKT/GSK-3β pathway, inducing E-cadherin repressor complexes and EMT. These findings imply the existence of a feedback inhibition loop of mTORC1 onto mTORC2 that plays a role in the homeostasis of E-cadherin expression and EMT, requiring caution in the clinical use of rapalog and selective mTORC1 inhibitors.

Figure 1

Involvement of the mTOR pathway in the expression of E-cadherin and components of the adherens junctional complex. (A) Representative chest computed tomographic imaging of 67-years old female patient, who had taken rapalog for 3 months under the diagnosis of metastatic recurred breast cancer. Compared with imaging prior to the rapalog treatment (left), size of multiple metastatic nodules (arrows) were decreased but ground glass opacity with interstitial thickenings were newly developed throughout the lung fields (right). (B) A549 cells were treated with 100 ng/mL IGF-1 for the indicated time, and expression of E-cadherin was evaluated by immunoblotting. A549 cells were treated with the indicated dose of IGF-1 for 16 hr and expression of E-cadherin was evaluated by immunoblotting (C) and qRT-PCR (D). (E) A549 cells were treated with the indicated dose of rapamycin for 16 hr and expression of E-cadherin was evaluated by immunoblotting. (F) A549, H596, and H2009 NSCLC cells were treated with 100 nM rapamycin for 16 hr and E-cadherin mRNA was evaluated by qRT-PCR. (G) A549 and H596 NSCLC cells were treated with 100 nM of rapamycin for 16 hr and expression of the components of the adherens junction complex were evaluated by immunoblotting. Data was analyzed by one-way ANOVA followed by Tukey’s multiple comparison test (D) or an independent sample t-test (F). β-actin was used as a loading control. RQ: relative quantitation, Rapa: rapamycin, veh: vehicle. Error Bars, SD of 3 independent experiments; *, P < 0.05; †, P < 0.01.

Figure 2

Disruption of mTORC1 suppresses E-cadherin expression. (A) A549 cells were stably transduced with lentiviral shRNA vectors targeting Raptor, Rictor, or TSC2. Expression of E-cadherin was analyzed by Western blotting. (B) A549 and H2009 cells were transduced with 2 different pLKO.1-Raptor-shRNA constructs and expression of E-cadherin was analyzed using Western blotting. (C) E-cadherin mRNA was analyzed by qRT-PCR in Raptor-silenced A549 and H2009 cells. Data was analyzed by one-way ANOVA followed by Turkey’s multiple comparison test. β-actin was used as a loading control. Error Bars, SD of 3 independent experiments; RQ, relative quantitation; *, P < 0.05; †, P < 0.01.

Figure 3

Aberrant Akt-GSK3β signaling in Raptor-silenced cells. A549, H2009, H460, and H1299 NSCLC cells were transduced with scrambled or Raptor shRNA and pAKT-Ser473 and pGSK-Ser9 levels were evaluated by Western blotting. β-actin was used as a loading control.

Figure 4

E-cadherin repressor complexes are elevated in Raptor-silenced cells. (A) A549 and H2009 cells were transduced with scrambled or Raptor shRNA and E-cadherin repressor complex mRNAs were evaluated by qRT-PCR. Data were analyzed by the t-test. (B) A549 cells were treated with 40 mM LiCl for the indicated times and E-cadherin repressor complex mRNAs were evaluated by qRT-PCR. Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparison test. (C) A549 cells were treated with MG132 (10 μM) and LiCl (40 mM), and/or IGF (10 ng/mL) for 6 hr and expression of Snail was evaluated by immunoblotting. β-actin was used as a loading control. Error Bars, SD of 3 independent experiments; RQ, relative quantitation; *, P < 0.05; †, P < 0.01.

Figure 5

EMT in Raptor-silenced cells. (A) Phase contrast and immunocytochemical image of scrambled and Raptor shRNA-transduced A5549 and H2009 cells (X1,000). E-cadherin: Red, Vimentin: Green. (B) Expression of EMT markers α-smooth muscle actin (SMA), Vimentin, and N-cadherin were evaluated in scrambled and Raptor shRNA-transduced A549 and H2009 cells by Western blotting. β-actin was used as a loading control. (C) Illustration of the role of mTORC1 in the regulation of mTORC2-mediated EMT.