Disruption of thioredoxin metabolism enhances the toxicity of transforming growth factor β-activated kinase 1 (TAK1) inhibition in KRAS-mutated colon cancer cells

Abstract

Transforming growth factor β-activated kinase 1 (TAK1) is critical for survival of many KRAS mutated colorectal cancer cells, and TAK1 inhibition with 5Z-7-oxozeaenol has been associated with oxidative stress leading to tumor cell killing. When SW 620 and HCT 116 human colon cancer cells were treated with 5µM 5Z-7-oxozeaenol, cell viability, growth, and clonogenic survival were significantly decreased. Consistent with TAK1 inhibition being causally related to thiol-mediated oxidative stress, 10mM N-acetylcysteine (NAC) partially reversed the growth inhibitory effects of 5Z-7-oxozeaenol. In addition, 5Z-7-oxozeaenol also increased steady-state levels of H2DCFDA oxidation as well as increased levels of total glutathione (GSH) and glutathione disulfide (GSSG). Interestingly, depletion of GSH using buthionine sulfoximine did not significantly potentiate 5Z-7-oxozeaenol toxicity in either cell line. In contrast, pre-treatment of cells with auranofin (Au) to inhibit thioredoxin reductase activity significantly increased levels of oxidized thioredoxin as well as sensitized cells to 5Z-7-oxozeaenol-induced growth inhibition and clonogenic cell killing. These results were confirmed in SW 620 murine xenografts, where treatment with 5Z-7-oxozeaenol or with Au plus 5Z-7-oxozeaenol significantly inhibited growth, with Au plus 5Z-7-oxozeaenol trending toward greater growth inhibition compared to 5Z-7-oxozeaenol alone. These results support the hypothesis that thiol-mediated oxidative stress is causally related to TAK1-induced colon cancer cell killing. In addition, these results support the hypothesis that thioredoxin metabolism is a critical target for enhancing colon cancer cell killing via TAK1 inhibition and could represent an effective therapeutic strategy in patients with these highly resistant tumors.

Keywords: Colorectal cancer; KRAS; TAK1; Thioredoxin; Treatment resistance.

Graphical abstract

Fig. 1

TAK1 inhibition decreased cell viability, growth, and clonogenic survival in HCT 116 and SW 620 cells; N-acetylcysteine, a thiol antioxidant, reversed these changes. Cells were treated with 5 µM 5Z-7-oxozeaenol and then 10 mM N-acetylcysteine. Vehicle=0.1% DMSO. A. Cell viability after 24 h of treatment was measured by MTT. (B) Cells were plated at time 0 and treatment applied at 24 h; cell counts were obtained at 48, 72, and 96 h after plating to determine cell growth rate. (C) After 24 h of treatment, cells were detached using TrypLE Express and plated at low density. 14 days later, colonies were fixed, stained, and counted to determine the rate of clonogenic survival. Data are mean±SEM of at least 3 separate experiments. *P <0.01, **P <0.05.

Fig. 2

Oxidative stress is increased with TAK1 inhibition, and cells respond to TAK1 inhibition with increased intracellular glutathione. Cells were treated with 5 µM 5Z-7-oxozeaenol and then 10 mM N-acetylcysteine. Vehicle=0.1% DMSO. (A) Flow cytometry was used to analyze intracellular reactive oxygen species generation after 24 h of treatment using H₂DCFDA (redox-sensitive probe) and DCFDA (redox-insensitive probe). Data normalized to vehicle control and presented as fold change of mean fluorescence intensity for H₂DCFDA-stained cells/DCFDA-stained cells. (B, C) After 24 h treatment with TAK1 inhibitor, cells were scrape-harvested and reduced glutathione (GSH), (B) and oxidized glutathione (GSSG), (C) levels were assayed. Normalized to vehicle control. Errors represent ±1 SEM of at least 3 separate experiments. *P <0.01, **P <0.05.

Fig. 3

Depleting intracellular glutathione had little effect on cell growth and clonogenic survival when coupled with TAK1 inhibition. Cells were pretreated with 100 µM BSO for 8 h to deplete intracellular glutathione, then treated with 5 µM 5Z-7-oxozeaenol. (A) After 24 h of treatment, cells were scrape harvested and total intracellular glutathione levels, shown as normalized to vehicle control, were assayed. (B) Cells were plated at time 0 and treatment applied at 24 h; cell counts were obtained at 48, 72, and 96 h after plating to determine cell growth rate. (C) After 24 h of treatment, cells were detached using TrypLE Express and plated at low density. 14 days later, colonies were fixed, stained, and counted to determine the rate of clonogenic survival. Values represent mean±1 SEM of at least 3 separate experiments. #P-value<0.01 versus control. *P-value<0.01. **P-value<0.05.

Fig. 4

Auranofin inhibits thioredoxin reductase (TrxR) activity and, when combined with the TAK1 inhibitor, increases the ratio of oxidized to total thioredoxin in SW 620 cells. SW 620 cells were pretreated with 500 nM auranofin, then with 5 µM 5Z-7-oxozeaenol. (A) Cells were scrape harvested and thioredoxin reductase activity was assayed and normalized to vehicle control. (B) SW 620 cells were scrape harvested. Thioredoxin-1 western blot was performed showing oxidized [(S-S) trx] and total thioredoxin (trx). (C) Semiquantitative analysis of Western blots for oxidized to total thioredoxin. Values represent the mean±1 SEM from 3 separate experiments. #P-value<0.01 versus control. **P-value<0.05.

Fig. 5

Auranofin inhibits thioredoxin reductase (TrxR) activity and enhances toxicity of TAK1 inhibition in both HCT 116 and SW 620 cells. SW 620 cells were pretreated with 500 nM auranofin, then with 5 µM 5Z-7-oxozeaenol. (A) Cells were plated at time 0 and treatment applied at 24 h; cell counts were obtained at 48 and 72 h after plating to determine cell growth rate. (B) After 24 h of treatment, cells were detached using TrypLE Express and plated at low density. 14 days later, colonies were fixed, stained, and counted to determine the rate of clonogenic survival. Errors represent ±1 SEM of at least 3 separate experiments. #P-value<0.01 versus control. ##P-value<0.05 versus control. *P-value<0.01. **P-value<0.05.

Fig. 6

The combination of auranofin plus TAK1 inhibitor yields the slowest growth of hind-limb human colon cancer xenografts. 2×10⁶ SW 620 cells were injected into the hind limbs of nude mice. After 11 days of tumor growth, mice were randomly assigned to treatment with intraperitoneal injection of vehicle control, TAK1 inhibitor (15 mg/kg), auranofin (1.6 mg/kg), or TAK1 inhibitor plus auranofin. Tumor size was measured with calipers and tumor volume calculated. N=7 per group; #P-value<0.01 versus control; ##P-value<0.05 versus control.