Epoxyeicosatrienoic acids attenuate reactive oxygen species level, mitochondrial dysfunction, caspase activation, and apoptosis in carcinoma cells treated with arsenic trioxide

Abstract

Epoxyeicosatrienoic acids (EETs) and the cytochrome P450 epoxygenase CYP2J2 promote tumorogenesis in vivo and in vitro via direct stimulation of tumor cell growth and inhibition of tumor cell apoptosis. Herein, we describe a novel mechanism of inhibition of tumor cell apoptosis by EETs. In Tca-8113 cancer cells, the antileukemia drug arsenic trioxide (ATO) led to the generation of reactive oxygen species (ROS), impaired mitochondrial function, and induced apoptosis. 11,12-EET pretreatment increased expression of the antioxidant enzymes superoxide dismutase and catalase and inhibited ATO-induced apoptosis. 11,12-EET also prevented the ATO-induced activation of p38 mitogen-activated protein kinase, c-Jun NH(2)-terminal kinase, caspase-3, and caspase-9. Therefore, 11,12-EET-pretreatment attenuated the ROS generation, loss of mitochondrial function, and caspase activation observed after ATO treatment. Moreover, the CYP2J2-specific inhibitor compound 26 enhanced arsenic cytotoxicity to a clinically relevant concentration of ATO (1-2 μM). Both the thiol-containing antioxidant, N-acetyl-cysteine, and 11,12-EET reversed the synergistic effect of the two agents. Taken together, these data indicate that 11,12-EET inhibits apoptosis induced by ATO through a mechanism that involves induction of antioxidant proteins and attenuation of ROS-mediated mitochondrial dysfunction.

Fig. 1.

Effect of EETs on ROS production in ATO-treated Tca-8113 cells. Cells were preincubated with 100 nM EETs or vehicle for 24 h, then exposed to 10 μM ATO for 2 h. DCFH-DA was then incubated with cells at 37°C for 15 min to detect ROS. A, quantification of mean ROS levels in Tca-8113 cells after ATO and EETs treatment. B, representative flow cytometric histogram of DCF fluorescent levels. C, quantification of mean ROS levels in Tca-8113 cells after ATO and 11,12-EET treatment (n = 3). D and E, production of hydrogen peroxide measured by Amplex Red (n = 6). Results shown are mean ± S.E.M. *, p < 0.05 versus control; **, p < 0.05 versus ATO treatment alone.

Fig. 2.

Effect of 11,12-EET on antioxidant enzyme expression and activity. A, antioxidant enzyme expression after 11,12-EET treatment for the indicated times and doses. B, total SOD and catalase activity. Tca-8113 cells were preincubated with 100 nM 11,12-EET for 12 h, and enzymatic activities were then measured. Results are shown as percentage of control ± S.E.M. *, p < 0.05 versus control.

Fig. 3.

11,12-EET inhibition of ATO-induced mitochondrial impairment. A, flow cytometric analysis of Tca-8113 cells treated with ATO (10 μM) and 11,12-EET (100 nM) for 24 h by using Annexin V-FITC and PI staining. The lower left quadrants represent nonapoptotic cells; the upper left quadrants represent cells that have lost their cell membranes and are dead, either end-stage apoptotic or necrotic (Annexin V-negative, PI-positive); the lower right quadrants represent early apoptotic cells (Annexin V-positive, PI-negative); and the upper right quadrants represent late apoptotic or necrotic cells (Annexin V- and PI-positive). B, graph represents the mean percentage of Annexin V-positive Tca-8113 cells expressed as the proportion of positive cells in each group (n = 3). C, detection of the mitochondrial transmembrane potential collapse by JC-1 staining. In cells with normal mitochondrial function, membrane potential-driven accumulation of these dyes results in the formation of red fluorescent J-aggregates as shown in the control panel. In cells treated with ATO, the green mitochondria stained with JC-1 dye indicate depressed ΔΨm. D, caspase-3 and caspase-9 activities in Tca-8113 cells. Cells were treated with 10 μM ATO and 100 nM of 11,12-EET for 24 h, lysed, and analyzed spectrophotometrically for caspase activity. E, cytochrome c release in Tca-8113 cells. Cells were treated with 10 μM ATO and 100 nM of 11,12-EET for 24 h and lysed, and cytosolic protein were analyzed by Western blot. Data are reported as mean absorption relative to control ± S.E.M. (n = 5). *, p < 0.05 versus control; **, p < 0.05 versus ATO treatment alone.

Fig. 4.

11,12-EET suppression of ATO-induced p38 and JNK activation. Cells were starved overnight, preincubated with 11,12-EET or p38/JNK inhibitor for 12 h, and incubated with ATO or H₂O₂ for 15 min. Data are representative of three independent experiments. Results shown are mean ± S.E.M. (n = 3). A and B, p38 and JNK activation in Tca-8113 cells treated with ATO (10 μM), 11,12-EET (100 nM), and H₂O₂ (200 μM) as indicated. C and D, flow cytometric analysis of Tca-8113 cells treated with p38 inhibitor (20 μM) as indicated for 24 h using Annexin V-FITC and PI staining. E and F, flow cytometric analysis of Tca-8113 cells treated with JNK inhibitor (50 μM) as indicated for 24 h using Annexin V-FITC and PI staining. *, p < 0.05 versus control; **, p < 0.05 versus ATO; #, p < 0.05 versus ATO+EET or ATO+p38/JNK Inhibitor.

Fig. 4.

11,12-EET suppression of ATO-induced p38 and JNK activation. Cells were starved overnight, preincubated with 11,12-EET or p38/JNK inhibitor for 12 h, and incubated with ATO or H₂O₂ for 15 min. Data are representative of three independent experiments. Results shown are mean ± S.E.M. (n = 3). A and B, p38 and JNK activation in Tca-8113 cells treated with ATO (10 μM), 11,12-EET (100 nM), and H₂O₂ (200 μM) as indicated. C and D, flow cytometric analysis of Tca-8113 cells treated with p38 inhibitor (20 μM) as indicated for 24 h using Annexin V-FITC and PI staining. E and F, flow cytometric analysis of Tca-8113 cells treated with JNK inhibitor (50 μM) as indicated for 24 h using Annexin V-FITC and PI staining. *, p < 0.05 versus control; **, p < 0.05 versus ATO; #, p < 0.05 versus ATO+EET or ATO+p38/JNK Inhibitor.

Fig. 5.

Effect of CYP2J2 inhibitor on cell proliferation. Cells were treated with the indicated drugs for 24 h, and cell number was measured by the SRB assay. A, effect of C26 on number of tumor cells. B, effect of ATO on number of tumor cells. C, effect of the combination of C26, ATO, 11,12-EET, and NAC on tumor cell number. Cells were incubated with 2 μM ATO, 1 μM C26, 100 of nM 11,12-EET, and/or 2 mM of NAC, as indicated for 24 h. Data are expressed as percentage of untreated controls ± S.E.M. (n = 5). *, p < 0.05 versus control; **, p < 0.05 versus ATO+C26.

Fig. 6.

Effects of CYP2J2 inhibition on sensitization of tumor cells to ATO-induced ROS production and apoptosis. Tca-8113 cells were incubated with 2 μM ATO, 1 μM C26, 100 nM 11,12-EET, and/or 2 mM NAC. A, Tca-8113 cells were preincubated with C26 and NAC for 1 h, followed by 24-h incubation with 11,12-EET, and 2 h before detection, ATO was added to induce ROS. Production of hydrogen peroxide was measured by Amplex Red. Results shown are mean ± S.E.M. (n = 6). B, Tca-8113 cells were preincubated with C26 and NAC for 1 h, and then EET and ATO were added for 24 h. Density plots of Annexin V/PI staining were measured by flow cytometry. C, graph represents the mean number of Annexin V-positive Tca-8113 cells expressed as percentage of control untreated cells ± S.E.M. (n = 3). Each sample was run in duplicate, and the data are representative of three independent assays. D, Tca-8113 cells were treated with CYP2J2-specific siRNA (100 nM) for 24 h. CYP2J2 expression was determined by Western blot (n = 3). E, Tca-8113 cells were treated with CYP2J2-specific siRNA (100 nM), EET, and ATO for 24 h. Density plots of Annexin V/PI staining for apoptosis were measured by flow cytometry. F, graph represents the mean number of Annexin V-positive Tca-8113 cells expressed as percentage of control untreated cells ± S.E.M. Each sample was run in duplicate, and the data are representative of three independent assays. *, p < 0.05 versus control; **, p < 0.05 versus ATO+C26.

Fig. 6.

Effects of CYP2J2 inhibition on sensitization of tumor cells to ATO-induced ROS production and apoptosis. Tca-8113 cells were incubated with 2 μM ATO, 1 μM C26, 100 nM 11,12-EET, and/or 2 mM NAC. A, Tca-8113 cells were preincubated with C26 and NAC for 1 h, followed by 24-h incubation with 11,12-EET, and 2 h before detection, ATO was added to induce ROS. Production of hydrogen peroxide was measured by Amplex Red. Results shown are mean ± S.E.M. (n = 6). B, Tca-8113 cells were preincubated with C26 and NAC for 1 h, and then EET and ATO were added for 24 h. Density plots of Annexin V/PI staining were measured by flow cytometry. C, graph represents the mean number of Annexin V-positive Tca-8113 cells expressed as percentage of control untreated cells ± S.E.M. (n = 3). Each sample was run in duplicate, and the data are representative of three independent assays. D, Tca-8113 cells were treated with CYP2J2-specific siRNA (100 nM) for 24 h. CYP2J2 expression was determined by Western blot (n = 3). E, Tca-8113 cells were treated with CYP2J2-specific siRNA (100 nM), EET, and ATO for 24 h. Density plots of Annexin V/PI staining for apoptosis were measured by flow cytometry. F, graph represents the mean number of Annexin V-positive Tca-8113 cells expressed as percentage of control untreated cells ± S.E.M. Each sample was run in duplicate, and the data are representative of three independent assays. *, p < 0.05 versus control; **, p < 0.05 versus ATO+C26.

Scheme 1.

The hypothesis of the functions of EETs in tumor cell apoptosis induced by ATO.