Role of TGF-β signaling in curcumin-mediated inhibition of tumorigenicity of human lung cancer cells

Abstract

Purpose: Curcumin has been shown to have potent anticancer activities like inhibition of cell proliferation, induction of apoptosis, and suppression of angiogenesis. Transforming growth factor-β (TGF-β) signaling plays a complex role in tumor suppression and promotion depending on the tumor type and stage. However, the effect of curcumin on TGF-β signaling in cancer cells and the role of TGF-β signaling in curcumin-induced anticancer activities have not been determined. Here, we investigate the role of curcumin on TGF-β signaling, and whether TGF-β signaling is involved in the antitumor activities of curcumin.

Methods: Human non-small cell lung cancer (NSCLC) cell lines, ACC-LC-176 (without TGF-β signaling), H358, and A549 (with TGF-β signaling) were treated with curcumin to determine cell growth, apoptosis, and tumorigenicity. Antitumor activities of curcumin were determined using these cell lines and an in vivo mouse model. We also tested the effect of curcumin on TGF-β/Smad signaling by western blotting and by luciferase assays.

Results: Curcumin inhibited cell growth and induced apoptosis of all three NSCLC cell lines in vitro and in vivo. It significantly reduced subcutaneous tumor growth by these three cell lines irrespective of TGF-β signaling status. Curcumin inhibited TGF-β-induced Smad2/3 phosphorylation and transcription in H358 and A549 cells, but not in ACC-LC-176 cells.

Conclusions: Curcumin reduces tumorigenicity of human lung cancer cells in vitro and in vivo by inhibiting cell proliferation and promoting apoptosis. These results suggest that TGF-β signaling is not directly involved in curcumin-mediated growth inhibition, induction of apoptosis, and inhibition of tumorigenicity.

Fig. 1

Curcumin inhibits cell growth and induces apoptosis of NSCLC cells. NSCLC cell lines ACC-LC-176 (a), H358 (b), and A549 (c) were treated with curcumin with indicated concentrations for 5 days. Cell counting was performed every 24 h using cell counting hemocytometer after treatment (a–c). The percentage of apoptotic cells was analyzed by flow cytometry (d–f) after 24-h treatment. Data were presented as the mean ± SD of triplicate determinations. g NSCLC cells ACC-LC-176, H358, and A549 were treated with curcumin as indicated for 24 h (left panel), or with 10 μM curcumin for different time points (6, 12, 24, 48 h; right panel). Protein lysates from treated cells were used for analyzing the level of PARP and cleaved PARP by western blot analyses

Fig. 2

Curcumin inhibits tumorigenicity of NSCLC cells in vitro. a, b ACC-LC-176, H358, and A549 cells growing in semisolid 0.4 % sea plaque agarose were treated with curcumin with indicated concentrations for 10 days. Colonies grown on the soft agarose were counted and analyzed with GEL COUNT™ software. Representative pictures of colonies treated with curcumin for 10 days are shown. Pictures of colonies were taken at ×100 magnification. Data are presented as the mean ± SD from three individual plates

Fig. 3

Curcumin inhibits tumorigenicity of NSCLC cells in vivo. NSCLC cells were inoculated at the right flank of athymic nude mice. Mice bearing subcutaneous tumor from NSCLC cell line ACC-LC-176 (a, n = 5 in each group), H358 (b, n = 5 in each group), and A549 (c, n = 5 in each group) were treated with curcumin as soon as the tumors had reached a minimum accurately measurable size. Curcumin (50 mg/kg) or DMSO was administrated i.p. every 3 days. Mice were euthanized at the end of the experiment, and pictures of the tumor mass are presented. Tumor volume was measured every 3 days with a slide calipers from the beginning of drug treatment, and the growth curve was plotted. Tumors were weighed at the end of the experiment. Each plot presents the mean volume ± SD of 5 mice in each group. *P < 0.001

Fig. 4

No obvious toxicities are present in mice with curcumin treatment. a Mice with or without curcumin treatment were weighed every 3 days from the beginning after inoculation of tumor cells. Each plot represents the mean weight ± SD value of 5 mice in each group at different time point. b Heart and liver were removed at the end of the experiment. Formalin-fixed paraffin-embedded tissue samples were analyzed by performing H&E staining. Pictures are presented at ×400 magnification

Fig. 5

Curcumin inhibits cell proliferation and induces apoptosis in tumor xenografts. a–c Tumor xenografts described in Fig. 3 were used for immunohistochemical staining for H&E, Ki67, and caspase 3. A representative picture from each group is shown. Pictures are shown at ×400 magnification

Fig. 6

Curcumin inhibits TGF-β/Smad signaling in H358 and A549 cells. Western blot analyses were performed using lysates from ACC-LC-176 (a), H358 (b), and A549 (c) cells treated with TGF-β (5 ng/ml) and 10 or 20 μM curcumin for 22 h. Luciferase reporter assay. ACC-LC-176 (d), H358 (e), and A549 (f) cells were transiently transfected with p3TP-Lux, (GAGA)₉ MLP-Luc, and CMV-βgal reporter plasmids, and then pretreated with curcumin of indicated concentrations for 2 h, and 5 ng/ml TGF-β was added thereafter for 22 h. Cell lysates were used to measure luciferase and β-gal activities, and normalized luciferase activities were presented