Resveratrol inhibits uveal melanoma tumor growth via early mitochondrial dysfunction

Abstract

Purpose: To test the efficacy of resveratrol, a nontoxic plant product, in the treatment of uveal melanoma.

Methods: The effect of oral administration and peritumor injection of resveratrol was tested on tumor growth in two animal models of uveal melanoma. The mechanism of resveratrol action on uveal melanoma cells was studied in vitro in a cell-viability assay: with JC-1 dye, to measure mitochondrial membrane potential; by Western blot analysis, to analyze the cellular redistribution of cytochrome c and Smac/diablo; and in a fluorescence assay with specific substrates, to measure activation of different caspases.

Results: Resveratrol treatment inhibited tumor growth in animal models of uveal melanoma. Since oral administration resulted in relatively low bioavailability of resveratrol, the effect of increased local levels was tested by peritumor injection of the drug. This method resulted in tumor cell death and tumor regression. In vitro experiments with multiple uveal melanoma cell lines demonstrate that resveratrol causes a decrease in cell viability, resulting at least in part from an increase in apoptosis through a mitochondrial pathway. An early event in drug action is the direct targeting of mitochondria by resveratrol, which leads to a decrease in mitochondrial membrane potential and the eventual activation of caspase-3.

Conclusion: These data suggest that resveratrol can inhibit tumor growth and can induce apoptosis via the intrinsic mitochondrial pathway and that by further increasing bioavailability of resveratrol the potency of the drug can be increased, leading to tumor regression. The nontoxic nature of the drug at levels needed for therapy make resveratrol an attractive candidate for the treatment of uveal melanoma.

Figure 1

Inhibition of uveal melanoma tumor growth. Tumor growth kinetics during resveratrol treatment of C918 (A) and Mum2b (B) xenograft models by oral gavage. Data show the average tumor growth in resveratrol- and vehicle-treated groups of mice. *P < 0.05, **P < 0.0005.

Figure 2

Tumor regression after peritumor injection of resveratrol in the C918 xenograft model. (A) Tumor size before and after three injections of vehicle or 20 mg resveratrol. H&E-stained sections of vehicle-treated tumor (B), 20 mg resveratrol-treated tumor (C), and adjacent normal skin tissue in the resveratrol-treated condition (D). Insets: detail of area around black arrows. (B, inset) Mitotic body; (C, inset) apoptotic cells. (C, D, *) Necrotic area. (C, white arrow) Calcified cell; (D, white arrow) inflammatory cells.

Figure 3

Resveratrol decreased the cell viability of uveal melanoma cell lines by inducing apoptosis. (A) The cell viability of M619, C918, Mum2b, and 92.1 cells was measured as a function of resveratrol concentration and time after drug addition and plotted as relative fluorescence units (rfu). (B) Viability of dense (>90% confluent) 92.1 cells was measured similarly. Each experiment was performed in triplicate. (C) Apoptotic nuclei of M619, C918, Mum2b, and 92.1 cells were visualized in resveratrol time courses with Hoechst 33528 dye and plotted as the percentage of apoptotic cells relative to the total number of cells. The number of cells was counted in five different viewing areas by fluorescence microscope and averaged. Experiments were performed in duplicate.

Figure 4

Resveratrol caused loss of mitochondrial membrane potential in uveal melanoma cells and in isolated mitochondria. Mitochondrial membrane potential in M619, C918, Mum2b, and 92.1 cells (A) and 92.1 mitochondria (B) was measured as a function of resveratrol concentration 15 minutes after drug addition, by using JC-1 dye. As a positive control, the cells were treated with FCCP or sodium azide. Data are plotted as the ratio of red to green fluorescence intensity at 530 and 590 nm, respectively, in cells and as the red fluorescence intensity in isolated mitochondria. Each experiment was performed in duplicate. UD, undetectable.

Figure 5

Resveratrol causes loss of cytochrome c and Smac/Diablo from the mitochondria and induces activation of caspase-9 and -3. (A) Cytochrome c and Smac/Diablo were measured by Western analysis in the cytoplasmic fractions of M619, C918, and Mum2b cells in a resveratrol time course. Each experiment was performed in duplicate. UNT, untreated cells. Caspase-9 (B) and -3 (C) activation was measured in M619, C918, and Mum2b cells with the substrates Ac-LEHD-AFC and Ac-DEVD-AMC, respectively, in a resveratrol-concentration course and plotted as relative fluorescence units (Rfu) after 48 hours of drug treatment. Each experiment was performed in triplicate. (D) Caspase-8 activation induced by resveratrol in M619, C918, and Mum2b cells was measured in whole cell lysates by Western analysis. The GAPDH level was measured as the loading control. Experiments were performed in duplicate.