MTA1-Dependent Anticancer Activity of Gnetin C in Prostate Cancer

Abstract

The overexpression of metastasis-associated protein 1 (MTA1) in prostate cancer (PCa) contributes to tumor aggressiveness and metastasis. We have reported the inhibition of MTA1 by resveratrol and its potent analog pterostilbene in vitro and in vivo. We have demonstrated that pterostilbene treatment blocks the progression of prostatic intraepithelial neoplasia and adenocarcinoma in mouse models by inhibiting MTA1 expression and signaling. In the current study, we investigated the MTA1 targeted anticancer effects of Gnetin C, a resveratrol dimer, in comparison with resveratrol and pterostilbene. Using DU145 and PC3M PCa cells, we found that Gnetin C downregulates MTA1 more potently than resveratrol and pterostilbene. Further, Gnetin C demonstrated significant MTA1-mediated inhibitory effect on cell viability, colony formation, and migration, while showing a more potent induction of cell death than resveratrol or pterostilbene. In addition, we identified Gnetin C-induced substantial ETS2 (erythroblastosis E26 transformation-specific 2) downregulation, which is not only MTA1-dependent, but is also independent of MTA1 as a possible mechanism for the superior anticancer efficacy of Gnetin C in PCa. Together, these findings underscore the importance of novel potent resveratrol dimer, Gnetin C, as a clinically promising agent for the future development of chemopreventive and possibly combinatorial therapeutic approaches in PCa.

Keywords: ETS2; Gnetin C; MTA1; prostate cancer; pterostilbene; resveratrol.

Figure 1

Gnetin C inhibits metastasis-associated protein 1 (MTA1) protein and mRNA expression in prostate cancer (PCa) cells more potently than resveratrol and pterostilbene. (A) Chemical structures of resveratrol, pterostilbene (Pter), and dimer-resveratrol (Gnetin C). (B,C) Dose-dependent inhibition of MTA1 protein expression by Gnetin C in DU145 and PC3M cells, respectively. Bottom, quantitation of MTA1 levels under Gnetin C treatment. (D,E) Comparison of MTA1 inhibition by Res (50 μM), Pter (50 μM), and Gnetin C (25 μM, 50 μM) in DU145 and PC3M cells, respectively. Bottom, quantitation of MTA1 levels under Res, Pter, and Gnetin C treatment. (F,G) MTA1 mRNA levels were analyzed by qRT-PCR after treatment with Res (50 μM), Pter (50 μM), and Gnetin C (25 μM, 50 μM) in DU145 and PC3M cells, respectively. Quantifications represent the mean ± SEM of three independent experiments. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 (one-way ANOVA).

Figure 2

Gnetin C induces MTA1-dependent cytotoxicity in PCa cells more potently than resveratrol and pterostilbene. (A,B) Cell viability analysis of DU145 and PC3M cells treated with various concentrations (5–100 μM) of Res, Pter, and Gnetin C. Bottom, IC₅₀ quantifications from the cell viability analysis. (C,D) Cell cycle analysis of DU145 and PC3M cells showing percentage of cell death after treatment with Res (50 μM), Pter (50 μM), and Gnetin C (25, 50 μM). Data represent the mean ± SEM of three independent experiments. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 (one-way ANOVA).

Figure 3

Gnetin C reduces MTA1-mediated clonogenic survival in PCa cells more potently than resveratrol and pterostilbene. (A,B) Representative images showing the colony formation ability of DU145 and PC3M MTA1 expressing (NS) and MTA1 knockdown (shMTA1) cells after treatment with 5 μM of Res, Pter, and Gnetin C. Right, Quantifications of the number of colonies using ImageQuant software are shown for each cell line. Data represent the mean ± SEM of three independent experiments with duplicate wells. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 (one-way ANOVA).

Figure 4

Gnetin C reduces the MTA1-mediated motility of PCa cells more potently than resveratrol and pterostilbene. (A,B) Representative images showing the migration ability of DU145 and PC3M MTA1 expressing (NS) and MTA1 knockdown (shMTA1) cells after treatment with 1 μM of Res, Pter, and Gnetin C for 48 h. Right, Quantifications of wound widths as a percentage of wound area using ImageJ software is shown for each cell line. Data represent the mean ± SEM of six separate wounds and three independent experiments. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 (one-way ANOVA).

Figure 5

Erythroblastosis E26 transformation-specific 2 (ETS2) expression correlates with MTA1 in PCa. (A) ETS2 expression positively correlates with MTA1 expression in human PCa cell lines. (B) MTA1 and ETS2 mRNA (top) and protein (bottom) levels were analyzed in PC3M (NS) and silenced for MTA1 (shMTA1) cells by qRT-PCR and Western blot, respectively. Quantifications represent the mean ±SEM of three independent experiments. * p < 0.05; ** p < 0.01; **** p < 0.0001 (two-way ANOVA). (C) Quantitation of MTA1 and ETS2 mRNA (top) and protein (bottom) levels in prostate tissues (n = 3) of 13-week-old Pb-Cre⁺, R26^MTA1/+ mice, and Cre-negative normal prostate control. Data represent the mean ±SEM of three prostate tissues. * p < 0.05; ** p < 0.01; **** p < 0.0001 (two-way ANOVA). (D) MTA1 and ETS2 protein expression was analyzed by immunofluorescence. Images were pseudo-colored in green to represent MTA1 and in red to represent ETS2. Images show the co-localization of MTA1 and ETS2 (merge, yellow) (magnification x 200). (E) Subcellular localization and predominantly nuclear expression of MTA1 and ETS2. Lamin A and Erk1/2 levels were used as a loading control for nuclear and cytoplasmic fractions, respectively. (F) MTA1 interacts with ETS2 in an ectopic system. A549 cells were co-transfected with Myc-MTA1, Myc-ETS2, and Myc-MTA1 + Myc-ETS2, and cell lysates were immunoprecipitated with antibodies to ETS2, followed by Western blotting with Myc antibodies.

Figure 6

MTA1-mediated mobility of PCa cells is dependent on ETS2. (A,B) ETS2 overexpression in MTA1 knockdown DU145 and PC3M cells: left, mRNA levels were assessed by qRT-PCR; right, protein levels were assessed by Western blot. β-actin is a loading control. Data represent the mean ± SEM of three independent experiments. * p < 0.05; **p < 0.01; *** p < 0.001; **** p < 0.0001 (two-way ANOVA). (C,D) Representative images showing migration ability of DU145 and PC3M MTA1 expressing (NS) and MTA1 knockdown (shMTA1) cells overexpressing ETS2. Bottom, quantitation of wound widths as percentage of wound area using ImageJ software is shown for each cell line. Data represent the mean ± SEM of six separate wounds and three independent experiments. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 (one-way ANOVA).

Figure 7

Gnetin C is more potent than resveratrol and pterostilbene in inhibiting MTA1-dependent ETS2 in PCa cells. (A,B) Comparison of ETS2 inhibition by Res (50 μM), Pter (50 μM), and Gnetin C (25 μM, 50 μM) in DU145 and PC3M cells, respectively. Bottom, quantitation of ETS2 levels under Res, Pter, and Gnetin C treatment. (C,D) ETS2 mRNA levels were analyzed by qRT-PCR after treatment with Res (50 μM), Pter (50 μM), and Gnetin C (25, 50 μM) in DU145 and PC3M cells, respectively. Data represent the mean ± SEM of three independent experiments. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 (one-way ANOVA). (E,F) MTA1 and ETS2 protein levels were determined in DU145 and PC3M cells expressing MTA1 (NS) and MTA1 knockdowns (shMTA1) treated with Res (50 μM), Pter (50 μM), and Gnetin C (25,50 μM). Bottom, quantifications of MTA1 and ETS2 levels under Res, Pter, and Gnetin C treatment in DU145 and PC3M cells. Data represent the mean ± SEM of three independent experiments. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 (two-way ANOVA).