Cisplatin combined with zidovudine enhances cytotoxicity and oxidative stress in human head and neck cancer cells via a thiol-dependent mechanism

Abstract

Oxidative stress and mitochondrial dysfunction in cancer cells represent features that may be exploited therapeutically. We determined whether agents that induce mitochondrial dysfunction, such as zidovudine (AZT) and cisplatin (CIS), could enhance killing of human head and neck cancer cells via oxidative stress. AZT- and/or CIS-induced cytotoxicity was determined using clonogenic survival, mitochondrial membrane potential was analyzed to investigate mitochondrial function, and glutathione was measured to determine thiol metabolism perturbations. AZT+CIS significantly increased toxicity and reduced mitochondrial membrane potential in FaDu, Cal-27, and SQ20B head and neck cancer cells while increasing the percentage of glutathione disulfide (%GSSG). Treatment with the thiol antioxidant N-acetylcysteine (NAC) reversed the loss of mitochondrial membrane potential and the increase in %GSSG and partially protected FaDu and Cal-27 cells from AZT+CIS. Finally, an inhibitor of glutathione synthesis, l-buthionine-[S,R]-sulfoximine, sensitized the cells to AZT+CIS-induced cytotoxicity, which was partially reversed by NAC. These results suggest that exposure of cancer cells to agents that induce mitochondrial dysfunction, such as AZT, causes significant sensitization to CIS-induced toxicity via disruptions in thiol metabolism and oxidative stress. These findings provide a biochemical rationale for evaluating agents that induce mitochondrial dysfunction in combination with chemotherapy and inhibitors of glutathione metabolism in head and neck cancer.

Figure 1

Effect of zidovudine (AZT) on survival of FaDu (black bars), Cal-27 (hatched bars) and SQ20B (white bars) cells. The cells were treated with 10 – 1000 μmol/L AZT for 24 h then plated for clonogenic survival. Clonogenic cell survival data were normalized to control (CON). Error bars represent ± standard error of the mean of N=3 experiments performed on different days with at least 4 cloning dishes taken from 1 treatment dish. *, p<0.05 versus respective control.

Figure 2

Effect of zidovudine (AZT) and cisplatin (CIS) on cytotoxicity (A), mitochondrial membrane potential (B), total glutathione (GSH) levels (C) and percentage oxidized glutathione (% GSSG) levels (D) in FaDu (black bars), Cal-27 (hatched bars) and SQ20B (white bars) cells. A: Cells were treated with 300 μM AZT and/or 0.2 μM CIS for 24 h. Clonogenic cell survival data were normalized to control (CON). B: Cells were treated as stated above, labeled with JC-1 dye for 15 min and analyzed by flow cytometry. C,D: Cells were treated as stated above and harvested for glutathione analysis using the spectrophotometric recycling assay. Error bars represent ± standard error of the mean of 2-3 experiments. *, p< 0.001 versus control; ¥, p<0.001 versus AZT or CIS alone.

Figure 3

Effect of zidovudine (AZT) and cisplatin (CIS) and N-acetyl-cysteine (NAC) on cytotoxicity (A), mitochondrial membrane potential (B), total glutathione (GSH) levels (C) and percentage oxidized glutathione (% GSSG) levels (D) in FaDu (black bars) and Cal-27 (hatched bars) cells. A: Cells were treated with 300 μM AZT and 0.2 μM CIS for 24 h with or without treatment with 20 mM NAC for 1 h before and during AZT and CIS exposure. Clonogenic cell survival data were normalized to control (CON). B: Cells were treated as stated above, labeled with JC-1 dye for 15 min and analyzed by flow cytometry. C,D: Cells were treated as stated above and harvested for glutathione analysis using the spectrophotometric recycling assay. Error bars represent ± standard error of the mean of N=3-7 experiments. *, p< 0.001 versus control; ¥, p<0.001 versus respective treatment without NAC.

Figure 4

Effect of zidovudine (AZT) and cisplatin (CIS) and L-buthionine-[S,R]-sulfoximine (BSO) on cytotoxicity in FaDu (black bars), Cal-27 (hatched bars) and SQ20B (white bars) cells. Cells were treated with 300 μM AZT and 0.2 μM CIS for 24 h with or without treatment with 1 mM BSO for 1 h before and during AZT and CIS exposure. Clonogenic cell survival data were normalized to control (CON). Cells were harvested for glutathione analysis using the spectrophotometric recycling assay. Error bars represent ± standard error of the mean of N=3 experiments performed on different days with at least 4 cloning dishes taken from 1 treatment dish. *, p<0.001 versus control; ¥, p< 0.05 versus respective treatment without BSO.

Figure 5

Effect of N-acetyl-cysteine (NAC) on zidovudine (AZT), cisplatin (CIS) and L-buthionine-[S,R]-sulfoximine (BSO)-induced cytotoxicity in FaDu cells. Cells were treated with 1 μM BSO, 300 μM AZT and 0.2 μM CIS for 24 h with or without treatment with 20 mM NAC for 1 h before and during AZT, CIS and BSO exposure. Clonogenic cell survival data were normalized to control (CON). Error bars represent ± standard error of the mean of N=3 experiments performed on different days with at least 4 cloning dishes taken from 1 treatment dish. *, p<0.001 versus control; ¥, p< 0.05 versus respective treatment without NAC.