Article Figures & Data
Figures
- Figure 1.
Catecholamines inhibit chemotherapy-induced apoptosis in ovarian cancer cells. HeyA8 (A and B) and SKOV3ip1 (C and D) cells were treated with the chemotherapeutic agents paclitaxel (A and C) or cisplatin (B and D), alone (IC50) or in combination with the catecholamines norepinephrine (NE; 10 μmol/L) or isoproterenol (ISO; 10 μmol/L). A β-blocker (propranolol; 1 μmol/L) was administered 30 minutes prior to catecholamine exposure. Apoptosis assays were performed using annexin V phycoerythrin/7AAD, followed by flow cytometry analysis. Results shown represent mean ± SEM, indicated by the error bar. *, P < 0.05; **, P < 0.01, compared with paclitaxel or cisplatin alone.
- Figure 2.
Catecholamines increase DUSP1 production through ADRB2. A, DUSP1 gene expression in HeyA8 and SKOV3ip1 ovarian cancer cells treated with norepinephrine (NE) compared with untreated cells (*, P < 0.01). B, DUSP1 mRNA levels, determined by real-time RT-PCR in HeyA8 and SKOV3ip1 ovarian cancer cells after exposure to different concentrations (0, 0.1, 1, or 10 μmol/L) of norepinephrine or isoproterenol (ISO) for different time periods (1, 3, or 6 hours). The mean fold change in DUSP1 mRNA expression compared with control is shown. Error bars, SEM. *, P < 0.01, compared with vehicle-treated control condition. C, Western blots analysis of DUSP1 protein expression. HeyA8 cells were stimulated with 10 μmol/L norepinephrine for 3 hours, and protein was obtained from the cell lysate for Western blot analysis using a DUSP1 antibody. The quantification of band intensity relative to β-actin intensity is shown at the bottom. *, P < 0.01, compared with the control. Adrenergic signaling plays a role in DUSP1 production. HeyA8 cells were pretreated with receptor-specific agonists or inhibitors and stimulated with norepinephrine for 3 hours; DUSP1 mRNA expression levels were then examined using real-time RT-PCR. Data, percentage of the control (medium only). The relative DUSP1 mRNA level is graphed as the mean fold change in DUSP1 production relative to control. Error bars, SEM. D, ADRA1 antagonist prazosin and ADRA2 antagonist yohimbine. E, nonspecific β-adrenergic antagonist propranolol. *, P < 0.01, compared with the norepinephrine-treated only. F, ADRB1 antagonist atenolol, ADRB2 antagonist ICI118,551, and ADRB3 antagonist SR59230A. *, P < 0.01, compared with the norepinephrine-treated only. G, ADRB2 and ADRB3 siRNA. *, P < 0.01, compared with control siRNA.
- Figure 3.
Norepinephrine (NE) plays a role in transcriptional control of DUSP1 promoter. A, DUSP1 promoter activity was determined by expression of a luciferase reporter gene in HeyA8 ovarian cancer cells after 3 hours of exposure to norepinephrine (1 or 10 μmol/L) or an equivalent volume of vehicle control solution. The role of specific β-adrenergic receptors in mediating norepinephrine effects on DUSP1 promoter activity was assessed by pretreating cells for 2 hours with 1 μmol/L concentrations of the α-adrenergic antagonist phentolamine, the ADRB1-selective antagonist metoprolol, the ADRB2-selective antagonist ICI118,551, or the ADRB3-selective antagonist SR59230A. To determine whether β-adrenergic signaling alone was sufficient to activate the DUSP1 promoter, cells were treated with 1 μmol/L concentrations of the nonselective β-agonist isoproterenol, the ADRB1-selective agonist dobutamine, the ADRB2-selective agonist terbutaline, or the ADRB3-selective agonist BRL37344. Values, mean ± standard error of 5 independent experiments. *, P < 0.05, compared with vehicle-treated control condition. B–E, to identify the specific transcription factor and promoter response element mediating norepinephrine induction of the DUSP1 promoter, we conducted systematic mutagenesis of a luciferase reporter construct under control of the 914 bases upstream of the human DUSP1 transcription start site. B and C, *, P < 0.05, compared with vehicle-treated control condition; E, *, P < 0.05, compared with the full-length promoter construct in the magnitude of norepinephrine-induced luciferase activity. F, transcription factors activated by norepinephrine. *, P < 0.05, compared with vehicle-treated control condition.
- Figure 4.
Downstream signaling of ADRB2 is involved in norepinephrine(NE)-induced increases in DUSP1 expression. A, cAMP agonist forskolin. B, PLC inhibitor U73122 and PKC inhibitor staurosporine. C, PKC, CREB1, and SP1 siRNA. D, PI3K inhibitor LY294002 and AKT inhibitor AKT1/2. E, Epac inhibitor brefeldin A and Epac agonist 8CPT-2Me-cAMP. F, PKA inhibitor KT5720 and H-89. G, MEK inhibitor U0126. H, p38 inhibitor SB203580. Error bars, SEM. DUSP1 siRNA inhibits norepinephrine-induced dephosphorylation of JNK and c-Jun. I, Western blot analysis of p-JNK (p-JNK1 and p-JNK2) and p-c-Jun, evaluated using the appropriate phospho-specific antibodies. Total JNK and total c-Jun are shown for comparison. These results represent 3 independent experiments. J, Western blot analysis of JNK/c-Jun phosphorylation after 24 hours of treatment with norepinephrine and paclitaxel in DUSP1 siRNA-expressing cells.
- Figure 5.
Effects of chronic restraint stress on ovarian cancer chemosensitivity. One week after being intraperitoneally injected with HeyA8 (A) or SKOV3ip1 (B) cells, nude mice were subjected to 2 hours of daily restraint stress each morning until the end of the experiment. Mice were randomly assigned to 8 groups (10 mice each, 4 without stress and 4 with stress treated with control, paclitaxel alone, propranolol alone, or paclitaxel with propranolol). Treatment was initiated at 3 to 4 weeks after injection. Paclitaxel at a dose of 2 mg/kg (SKOV3ip1) or 2.5 mg/kg (HeyA8) was given intraperitoneally weekly; propranolol at a dose of 2 mg/kg was given intraperitoneally every day. At the end of the study, mice were killed and their tumors were harvested. Tumor weights (A and B, left) and tumor nodules (A and B, right) were quantified in the HeyA8 and SKOV3ip1 models. Immunohistochemical staining of tumor samples from the SKOV3ip1 model showing the effects of chronic restraint stress on cell apoptosis (C), proliferation (D), DUSP1 (E), and JNK phosphorylation (F). All photographs were taken at 200× magnification. The bars in the graphs correspond sequentially to the labeled columns of the images at left. Error bars, SEM. *, P < 0.05; **, P < 0.01, compared with the control.
- Figure 6.
DUSP1 plays a role in mediating stress-induced chemoresistance. A, analysis of apoptosis in SKOV3ip1 cells overexpressing DUSP1 in no-stress conditions. *, P < 0.05, compared with the SKOV3ip1-EV group (cells transfected with pCMV6-entry vector). Error bars, SEM. B, analysis of apoptosis in HeyA8 cells expressing DUSP1 siRNA, treated with norepinephrine (NE) and paclitaxel. *, P < 0.05, compared with the control siRNA, norepinephrine, and paclitaxel group. Error bars, SEM. C and D, effects of DUSP1 silencing on stress-mediated tumor growth. Nude mice were subjected to 2 hours of daily restraint stress each morning for 3 to 4 weeks using HeyA8 ovarian cancer models. Mice (n = 10 per group) were injected subcutaneously with HeyA8 cells and randomly assigned to 1 of 4 groups: (i) control siRNA twice weekly; (ii) control siRNA twice weekly and intraperitoneal paclitaxel weekly; (iii) DUSP1 siRNA twice weekly; or (iv) DUSP1 siRNA twice weekly and intraperitoneal paclitaxel weekly. After 3 to 4 weeks of treatment, mice were killed and tumor weights and number of nodules were recorded. Mean tumor weight (C) and mean number of nodules (D) are shown in the graphs. *, P < 0.05, compared with control siRNA and paclitaxel. Error bars, SEM. E, immunohistochemical analysis of cleaved caspase-3 staining in a sample of tumor tissue from each group. All photographs were taken at 200× magnification. **, P < 0.01, compared with control siRNA and paclitaxel. Error bars, SEM. Overall survival (F) and progression-free survival (G): Kaplan–Meier curves for ovarian cancer patients with high and low DUSP1 expression is shown. Median survival durations for each group are shown. Data are compiled from the TCGA project. H, working model of stress-induced chemoresistance. After being exposed to a stressor, catecholamines are released and bind to the β-adrenergic receptors on the tumor cell surface to initiate downstream signaling through cAMP–PLC–PKC and CREB, which initiates transcription of DUSP1 gene expression. DUSP1 protein is produced and JNK and c-Jun are dephosphorylated, protecting the cells from apoptosis.
Additional Files
Supplementary Data
- Supplementary data - Supplementary Figure 1 Functional genomic analysis of HeyA8 and SKOV3ip ovarian cancer cells 4 h after exposure to norepinephrine and compared to control cells. Supplementary Figure 2 A, Western blot analysis of ADRB1, ADRB2, ADRB3, and DUSP1 expression in 10 epithelial ovarian cancer cell lines and one breast cancer cell line (MDA-231), using non-transformed ovarian epithelium cells (HIO-180) as control. B, DUSP1 mRNA levels of cells exposed to norepinephrine 10 μM for 4 h. Supplementary Figure 3 A, DUSP1 silencing. Two commercially available siRNAs that directly target the DUSP1 gene were transfected to norepinephrine-treated HeyA8 cells. B, DUSP1 overexpression. SKOV3ip1 cells were transiently transfected with a DUSP1-myc/DDK-tagged expression vector (Origene). Supplementary Figure 4 SKOV3ip1 cells were treated with paclitaxel alone (10nM) or in combination with norepinephrine (NE; 10μM) or vascular endothelial growth factor (VEGF; 10ng/ml or 50ng/ml). NE or VEGF was administered 30 minutes prior to paclitaxel exposure.
Volume 22, Issue 7
