Amurensin G Sensitized Cholangiocarcinoma to the Anti-Cancer Effect of Gemcitabine via the Downregulation of Cancer Stem-like Properties

Abstract

Cholangiocarcinoma (CCA) is a malignant biliary tract tumor with a high mortality rate and refractoriness to chemotherapy. Gemcitabine is an anti-cancer chemotherapeutic agent used for CCA, but the efficacy of gemcitabine in CCA treatment is limited, due to the acquisition of chemoresistance. The present study evaluated the chemosensitizing effects of Amurensin G (AMG), a natural sirtuin-1 inhibitor derived from Vitis amurensis, in the SNU-478 CCA cells. Treatment with AMG decreased the SNU-478 cell viability and the colony formation ability. Annexin V/ Propidium iodide staining showed that the AMG increased apoptotic death. In addition, AMG downregulated anti-apoptotic Bcl-2 expression, while upregulating pro-apoptotic cleaved caspase-3 expression. Treatment with AMG decreased the migratory ability of the cells in a wound healing assay and transwell migration assay. It was observed that AMG decreased the gemcitabine-induced increase in CD44^highCD24^highCD133^high cell populations, and the expression of the Sox-2 protein was decreased by AMG treatment. Co-treatment of AMG with gemcitabine significantly enhanced the production of reactive oxygen species, as observed through mitochondrial superoxide staining, which might be associated with the downregulation of the Sirt1/Nrf2 pathway by AMG. These results indicate that AMG enhances the chemotherapeutic ability of gemcitabine by downregulating cancer stem-like properties in CCA cells. Hence, a combination therapy of AMG with gemcitabine may be an attractive therapeutic strategy for cholangiocarcinoma.

Keywords: amurensin G; cancer stem cell; chemoresistance; cholangiocarcinoma; gemcitabine.

Figure 1

Effects of co-treatment of AMG and gemcitabine on cell viability and proliferation of SNU-478 cells. (A) The representative graph shows the changes in cell viability by AMG, for 72 h. (B) The graph shows the change in cell viability by AMG and/or gemcitabine, for 72 h. (C) The representative images show colony formation in SNU-478 cells. (D) Quantification of colony-forming area. Data are presented as means ± S.D. from at least three independent experiments. ** p < 0.01 vs. Con; ^@ p < 0.05, ^@@ p < 0.01 vs. GEM; ^## p < 0.01 vs. AMG (Dunnet’s test). Con, control; AMG, amurensin G; GEM, gemcitabine.

Figure 2

Effects of co-treatment of AMG and gemcitabine on types of cell death in SNU-478 cells. (A) Representative flow cytometry plots with Annexin V/PI staining for apoptosis. (B) Percentage of apoptotic cells. (C) The protein expression of Bcl-2 and c-cas3 was detected by Western blotting after the treatment for 48 h. (D) The protein expression of Bcl-2 was normalized with β-actin. (E) The protein expression of c-cas3 was normalized with GAPDH. Data are presented as means ± S.D. from at least three independent experiments. * p < 0.05, ** p < 0.01 vs. Con; ^@ p < 0.05, ^@@ p < 0.01 vs. GEM; ^## p < 0.01 vs. AMG (Dunnet’s test). Con, control; AMG, amurensin G; GEM, gemcitabine.

Figure 3

Effect of co-treatment of AMG and gemcitabine on cell migratory ability in SNU-478 cells. (A) The representative images of a wound healing assay in SNU-478 cells. Bar = 500 μm. (B) Quantification of wound-healing scratch area. (C) The representative images of a transwell migration assay in SNU-478 cells. Bar = 200 μm. (D) Quantification of transwell migrated cells. Data are presented as means ± S.D. from at least three independent experiments. * p < 0.05, ** p < 0.01 vs. Con; ^@@ p < 0.01 vs. GEM (Dunnet’s test). Con, control; AMG, amurensin G; GEM, gemcitabine.

Figure 4

Effect of co-treatment of AMG and gemcitabine on cancer stem-like cell properties in SNU-478 cells. (A) Representative histogram of CD44. (B) The graph shows the percentage of CD44^high populations. (C) Representative plot of CD24 and CD133 among CD44^high cells. (D) The graph shows the percentage of CD44^highCD24^highCD133^high populations. (E) The protein expression of Sox-2 was detected using Western blotting. (F) The Sox-2 protein expression was normalized with β-actin. Data are presented as means ± S.D. from at least three independent experiments. * p < 0.05, ** p < 0.01 vs. Con; ^@@ p < 0.01 vs. GEM; ^## p < 0.01 vs. AMG (Dunnet’s test). Con, control; AMG, amurensin G; GEM, gemcitabine.

Figure 5

Effects of co-treatment of AMG and gemcitabine on mitochondrial ROS production. (A) The representative images show ROS formation using MitoSOX^TM and Hoechst 33342 in SNU-478 cells. Bar = 200 μm. (B) Quantification of MitoSOX fluorescence intensity. (C) The protein expression of Sirt1, Nrf2, and Keap1 was detected using Western blotting. (D) The Sirt1 protein expression was normalized with GAPDH. (E) The Keap1/Nrf2 protein expression ratio was assessed. Data are presented as means ± S.D. from at least three independent experiments. * p < 0.05, ** p < 0.01 vs. Con; ^@ p < 0.05 vs. GEM; ^## p < 0.01 vs. AMG (Dunnet’s test). Con, control; AMG, amurensin G; GEM, gemcitabine.

Figure 6

Schematic model of the mechanism of action upon treatment with AMG in SNU-478 cells (created with BioRender.com). Treatment of AMG reduces colony formation and cell migration, by inhibiting cancer stem-like properties. Inhibition of Sirt1 reduces the expression of Nrf2, leading to apoptosis through increased ROS formation. Additionally, AMG enhances cell death by sensitizing CCA to gemcitabine.