Curcumin inhibited HGF-induced EMT and angiogenesis through regulating c-Met dependent PI3K/Akt/mTOR signaling pathways in lung cancer

Abstract

The epithelial-mesenchymal transition (EMT) and angiogenesis have emerged as two pivotal events in cancer progression. Curcumin has been extensively studied in preclinical models and clinical trials of cancer prevention due to its favorable toxicity profile. However, the possible involvement of curcumin in the EMT and angiogenesis in lung cancer remains unclear. This study found that curcumin inhibited hepatocyte growth factor (HGF)-induced migration and EMT-related morphological changes in A549 and PC-9 cells. Moreover, pretreatment with curcumin blocked HGF-induced c-Met phosphorylation and downstream activation of Akt, mTOR, and S6. These effects mimicked that of c-Met inhibitor SU11274 or PI3 kinase inhibitor LY294002 or mTOR inhibitor rapamycin treatment. c-Met gene overexpression analysis further demonstrated that curcumin suppressed lung cancer cell EMT by inhibiting c-Met/Akt/mTOR signaling pathways. In human umbilical vein endothelial cells (HUVECs), we found that curcumin also significantly inhibited PI3K/Akt/mTOR signaling and induced apoptosis and reduced migration and tube formation of HGF-treated HUVEC. Finally, in the experimental mouse model, we showed that curcumin inhibited HGF-stimulated tumor growth and induced an increase in E-cadherin expression and a decrease in vimentin, CD34, and vascular endothelial growth factor (VEGF) expression. Collectively, these findings indicated that curcumin could inhibit HGF-promoted EMT and angiogenesis by targeting c-Met and blocking PI3K/Akt/mTOR pathways.

Figure 1

The effects of curcumin on hepatocyte growth factor (HGF)-induced cell proliferation, migration, invasion and epithelial-mesenchymal transition in lung cancer A549 and PC-9 cells. (a) A549 cells and PC-9 cells were starved for 12 hours and then stimulated by 40 ng/ml of HGF in the presence of 2% of fetal bovine serum for 24, 48, 72, and 96 hours. Cell proliferation was detected at indicted times. When curcumin was used, it was added 4 hours before HGF stimulation. **P < 0.01 compared with HGF group. (b) A549 cells and PC-9 cells were starved for 12 hours then both the cells were stimulated with 40 ng/ml of HGF in the presence of 2% of fetal bovine serum. Cell migration capability of A549 cells and PC-9 cells were determined by wound healing assay. When curcumin was used, it was added 4 hours before HGF stimulation. Data are means of three separated experiments ± SD, *P < 0.05, **P < 0.01 compared with HGF group. (c) The cells were added to the upper chamber in 2% fetal bovine serum (FBS) media and invaded toward 2% FBS and 40 ng/ml HGF containing growth media in the lower chamber. Invasion capability of A549 cells and PC-9 cells were determined by transwell assay. When curcumin was used, it was added to the upper chamber. Data are means of three separated experiments ± SD, *P < 0.05, **P < 0.01 compared with HGF group. (d,e) A549 cells and PC-9 cells were starved for 12 hours, then treated with 40 ng/ml of HGF (with 0.5% FBS) for 48 hours. The cells morphology (d) was observed by contrast microscopy (original magnification, ×200). The expression of E-cadherin and vimentin (e) were detected by western blotting analysis. When curcumin was used, curcumin was added 4 hours before HGF stimulation. Quantitative results are also illustrated. The data presents the average of three independent experiments; CUR, curcumin.

Figure 2

c-Met/PI3K/Akt/mTOR signaling pathway is involved in inhibition effect of curcumin on hepatocyte growth factor (HGF)-induced lung cancer cell epithelial-mesenchymal transition. A549 cells (a) and PC-9 cells (b) were starved for 12 hours and then stimulated with 40 ng/ml of HGF for 15 minutes with or without pretreated with different concentrations of curcumin (10–30 μmol/l) for 4 hours, protein expression of c-Met, p-c-Met, Akt, p-Akt, mTOR, p-mTOR, S6, and p-S6 were detected by western blotting analysis. Quantitative results are also illustrated. The data presents the average of three independent experiments. (c,d) A549 cells and PC-9 cells were starved for 12 hours, then treated with 40 ng/ml of HGF (with 0.5% fetal bovine serum) for 48 hours. MET inhibitors SU11274 (5 μmol/l), or PI3K inhibitor LY294002 (25 μmol/l), or mTOR inhibitor rapamycin (200 nmol/l) was used 4 hours before HGF stimulation. The cell morphology was observed by contrast microscopy (Original magnification, ×100) (c). The expression of E-cadherin, vimentin was detected by western blotting (d). Quantitative results are also illustrated. The data presents the average of three independent experiments. CUR, curcumin; SU: SU11274; LY: LY294002; Ra, Rapamycin.

Figure 3

SU11274, LY294002, rapamycin inhibited hepatocyte growth factor (HGF)-induced lung cancer cell migration, invasion and c-Met/PI3K/Akt/mTOR signaling activation. (a,b) A549 cells were starved for 12 hours, then treated with 40 ng/ml of HGF (with 2% fetal bovine serum). MET inhibitors SU11274 (5 μmol/l), or PI3K inhibitor LY294002 (25 μmol/l), or mTOR inhibitor rapamycin (200 nmol/l) was used 1 hour before HGF stimulation. Cell migration capability was determined by wound healing assay (a). Invasion capability was determined by transwell assay (b). Data are means of three separated experiments ± SD, *P < 0.05, **P < 0.01 compared with HGF group. (c) A549 cells and PC-9 cells were starved for 12 hours, then treated with 40 ng/ml of HGF for 15 minutes. MET inhibitors SU11274 (5 μmol/l), or PI3K inhibitor LY294002 (25 μmol/l), or mTOR inhibitor rapamycin (200 nmol/l) was used 4 hours before HGF stimulation. Protein expression of c-Met, p-c-Met, Akt, p-Akt, mTOR, p- mTOR, S6, and p-S6 were detected by western blotting analysis. Quantitative results are also illustrated. The data presents the average of three independent experiments. SU, SU11274; LY: LY294002; Ra, Rapamycin.

Figure 4

Curcumin could inhibit c-Met/PI3K/Akt/mTOR signaling activation, epithelial-mesenchymal transition, migration and invasion induced by c-MET overexpression in lung cancer A549 cells and PC-9 cells. A549 cells and PC-9 cells were transfected with c-Met overexpression plasmid (ex-MET) and then treated with 20 μmol/l of curcumin or 5 μmol/l of SU11274 for 24 hours, the expression of c-Met, p-c-Met, Akt, p-Akt, mTOR, p-mTOR, S6 p-S6, E-cadherin, and vimentin protein were detected by western blotting analysis (a,b). The cell morphology was observed by contrast microscopy (Original magnification, ×200) (c). Cell migration capability of A549 cells (d) and PC-9 cells (e) were determined by wound healing assay. Invasion capability of A549 cells (f) and PC-9 cells (g) were determined by transwell assay. Data are means of three separated experiments ± SD, **P < 0.01 compared with ex-Met group. NC, negative control; CUR, curcumin; SU, SU11274.

Figure 5

Curcumin suppresses the migration and tube formation of HUVEC cells by c-Met/Akt/mTOR pathway. (a,b) Representation (a) and quantification (b) of tube formation assay showing the angiogenic capability of HUVECs stimulated with hepatocyte growth factor (HGF) for 6 hours, or cotreated with different concentrations of curcumin (10–30 μmol/l) or different inhibitors (SU11274: 5 μmol/l, LY294002:25 μmol/l; rapamycin: 200 nmol/l). Original magnification: ×100. (c) Quantification of scratch migration assay showing the migration of HUVEC cells stimulated with HGF or combined with curcumin or different inhibitors. Data are means of three separated experiments ± SD, *P < 0.05, **P < 0.01 compared with HGF group. (d) HUVECs cells were starved for 12 hours, and stimulated with 40 ng/ml of HGF for 15 minutes. Different concentrations of curcumin or different inhibitors were used 4 hours before HGF stimulation. The cells were then harvested and lysed for the detection of p-c-Met, c-Met, p-Akt, Akt, p-mTOR, mTOR, S6, and p-S6. (e) HUVECs cells were treated with different concentrations of HGF for 24 hours, or pretreated with curcumin or different inhibitors for 4 hours. The vascular endothelial growth factor expression was examined by western blotting. Quantitative results are also illustrated. The data presents the average of three independent experiments. SU, SU11274; LY, LY294002; Ra, Rapamycin.

Figure 6

Curcumin effectively inhibited hepatocyte growth factor-induced tumor growth, epithelial-mesenchymal transition, vascular endothelial growth factor (VEGF) expression and angiogenesis in the xenograft tumor model of lung cancer. (a) Image of tumor size and (b) wet tumor weight at the time of dissection. (c) Immunohistochemistry (IHC) of the expression pattern of E-cadherin, vimentin, VEGF and CD34 using tumor tissues derived from xenograft tumor model. Representative images of each group are shown (Magnifications: 100×). Data are means ± SD; *P < 0.05. ** P<0.01.