Atorvastatin reduces vascular endothelial growth factor (VEGF) expression in human non-small cell lung carcinomas (NSCLCs) via inhibition of reactive oxygen species (ROS) production

Abstract

The high metastatic potential of non-small cell lung cancers (NSCLCs) is closely correlated with the elevated expression of vascular endothelial growth factor (VEGF) and resultant tumor angiogenesis. However, no effective strategies against VEGF expression have been available in NSCLCs therapy. This study demonstrated that elevated reactive oxygen species (ROS) levels derived from both mitochondria and NADPH oxidase were required for VEGF expression in NSCLC cells. Atorvastatin administration could significantly inhibit VEGF expression both in vitro and in vivo via inhibition of ROS production. Atorvastatin inhibited ROS generation partly through suppression of Rac1/NADPH oxidase activity. Specifically, atorvastatin could upregulate the activity of glutathione peroxidase (GPx) and catalase, which are responsible for elimination of hydrogen peroxide (H(2)O(2)) in the mitochondria and peroxisomes, respectively. Thus, inhibition of ROS production by concomitant suppression of Rac1/NADPH oxidase activity and upregulation of the activity of GPx and catalase contributes critically to atorvastatin-reduced VEGF expression in NSCLCs. Atorvastatin may be a potential alternative against VEGF expression and angiogenesis in NSCLCs therapy.

Figure 1

Atorvastatin (ATV) inhibits VEGF expression independent of its pro‐apoptotic effect in A549 cells. (A) A549 cells were treated with indicated doses of atorvastatin for 24 h. Then, total protein lysates were extracted and the VEGF expression was subjected to ELISA. Bars are mean ± SD from four independent experiments, *P < 0.05, ***P < 0.001. (B) The apoptotic rate of A549 cells treated with atorvastatin (5 μM) in the presence or absence of caspase inhibitor, z‐VAD‐fmk (25 μM) was determined by flow cytometry after 24‐h incubation. Bars are mean ± SD from five independent experiments, ***P < 0.001. (C) The expression of VEGF treated with atorvastatin (5 μM) in the presence or absence of z‐VAD‐fmk (25 μM). Bars are mean ± SD from four‐five independent experiments, n.s, no significant difference, **P < 0.01.

Figure 2

The dependence of reduced VEGF expression by atorvastatin (ATV) on inhibition of ROS production in A549 cells. (A) A549 cells were treated with indicated dose of atorvastatin for 24 h and then treated with 5 μM CM2‐DCFH‐DA for 15 min at 37 °C. The fluorescent intensity was analyzed by Flow cytometry. The below graph represented the mean ± SD from five independent experiments, ***P < 0.001. (B) The effect of widely accepted ROS scavengers, NAC (25 μM), DPI (10 μM) or rotenone (10 μM) and atorvastatin (5 μM) on VEGF expression. Bars are mean ± SD from six independent experiments. (C–D) A549 cells were exposed to H2O2 (100 μM) in presence or absence of catalase (500 U/ml) (C) or atorvastatin (5 μM) (D) for 24 h. Then, the expression of VEGF was examined by ELISA. Bars are mean ± SD from five‐seven independent experiments, n.s, no significant difference, **P < 0.01,***P < 0.001.

Figure 3

The involvement of NADPH oxidase, GPx and catalase in atorvastatin (ATV)‐induced suppression of VEGF expression in A549 cells. (A) A549 cells were transfected with 0.1 μM p47phox siRNA using Lipofectamine 2000 transfecting. Total protein lysates were extracted and subjected to immunoblotting using antibodies against p47phox and β‐tubulin. Bar graphs are derived from densitometric scanning of the blots. Bars are mean ± SD from three independent experiments, ***P < 0.001. (B) The effect of p47phox siRNA on VEGF expression assayed by ELISA. Bars are mean ± SD from four independent experiments, **P < 0.01. (C) The effect of atorvastatin (5 μM) on NADPH oxidase activity after 24‐h incubation in A549 cells determined by lucigenin‐enhanced chemiluminescene assay. Bars are mean ± SD from three independent experiments, **P < 0.01. (D) A549 cells were differentially treated atorvastatin (5 μM), p47phox siRNA, or their combination for 24 h. Then the expression of VEGF was analyzed by ELISA. Bars are mean ± SD from three‐five independent experiments, *P < 0.05. (E–F) The effect of atorvastatin (5 μM) on the activity of GPx(E) and catalase (F) in A549 cells. Bars are mean ± SD from three independent experiments, *P < 0.05. (G) The effect of catalase (500 U/ml) treatment on the expression of VEGF after 24‐h incubation analyzed by ELISA. Bars are mean ± SD from four independent experiments, *P < 0.05.

Figure 4

Inhibition of Rac1‐GTPase activity and Rac1 membrane translocation by atorvastatin (ATV) in A549 cells. (A) The effect of atorvastatin (5 μM) on Rac1‐GTPase activity in A549 cells. Bars are mean ± SD from four independent experiments, ***P < 0.001. (B) Immunoblots of Rac1 protein in member (M) and cytosolic (C) in A549 cells exposed to atorvastatin (5 μM). Calsequestrin (csq) was used as control for equal protein loading.

Figure 5

The dependence of atorvastatin (ATV)‐reduced VEGF expression on ROS production in another NSCLC cell line (Calu3 cells). (A) Calu3 cells were differentially treated with NAC (25 μM), DPI (10 μM) or rotenone (10 μM) and atorvastatin (5 μM) for 24 h. Then, total protein lysates were extracted and the expression of VEGF was subjected to ELISA. Bars are mean ± SD from five‐seven independent experiments. (B) Calu3 cells were treated with atorvastatin (5 μM), or exposed to H2O2 (100 μM) in presence or absence of atorvastatin (5 μM). The expression of VEGF was subjected to ELISA. Bars are mean ± SD from five independent experiments, n.s, no significant difference, **P < 0.01. (C–F) Calu3 cells were exposed to atorvastatin (5 μM) for 24 h. The effect of atorvastatin (5 μM) on the activity of Rac1‐GTPase (C), NADPH oxidase (D), GPx (E) and catalase (F) were differentially analyzed by respective method according to above experiments. Bars are mean ± SD from three‐five independent experiments, *P < 0.05, **P < 0.01.

Figure 6

Atorvastatin (ATV) and ROS inhibitors decrease VEGF expression in A549‐beared tumors. 1 × 107 A549 cells were injected subcutaneously into nude mice to develop tumor‐harbored mice. When tumors grew to 80–100 mm3, the mice were differently treated (A) saline, (B) DPI (1 mg/kg/day, by gavage), (C) NAC (7 mg/ml) given in the drinking water for the length of the experiment or (D) atorvastatin (10 mg/kg/day, by gavage). Tumor sections were analyzed by immunohistochemistry analysis using antibody against VEGF.