Glucocorticoid receptor antagonism as a novel therapy for triple-negative breast cancer

Abstract

Purpose: Triple-negative breast cancer (TNBC) accounts for 10% to 20% of newly diagnosed invasive breast cancer. Finding effective targets for chemotherapy-resistant TNBC has proven difficult in part because of TNBC's molecular heterogeneity. We have previously reported that likely because of the antiapoptotic activity of glucocorticoid receptor (GR) in estrogen receptor (ER)-negative breast epithelial and cancer cells, high GR expression/activity in early-stage TNBC significantly correlates with chemotherapy resistance and increased recurrence. We hypothesized that pretreatment with mifepristone, a GR antagonist, would potentiate the efficacy of chemotherapy in GR+ TNBCs by inhibiting the antiapoptotic signaling pathways of GR and increasing the cytotoxic efficiency of chemotherapy.

Experimental design: TNBC cell apoptosis was examined in the context of physiologic glucocorticoid concentrations, chemotherapy, and/or pharmacologic concentrations of mifepristone. We used high-throughput live microscopy with continuous recording to measure apoptotic cells stained with a fluorescent dye and Western blot analysis to detect caspase-3 and PARP cleavage. The effect of mifepristone on GR-mediated gene expression was also measured. TNBC xenograft studies were performed in female severe combined immunodeficient (SCID) mice and tumors were measured following treatment with vehicle, paclitaxel, or mifepristone/paclitaxel.

Results: We found that although mifepristone treatment alone had no significant effect on TNBC cell viability or clonogenicity in the absence of chemotherapy, the addition of mifepristone to dexamethasone/paclitaxel treatment significantly increased cytotoxicity and caspase-3/PARP cleavage. Mifepristone also antagonized GR-induced SGK1 and MKP1/DUSP1 gene expression while significantly augmenting paclitaxel-induced GR+ MDA-MB-231 xenograft tumor shrinkage in vivo.

Conclusions: These results suggest that mifepristone pretreatment could be a useful strategy for increasing tumor cell apoptosis in chemotherapy-resistant GR+ TNBC.

Figure 1. In vitro analysis of paclitaxel-induced cell death +/− mifepristone in MBA-MB-231, BT-20, and MDA-MB-468 cells

A, TNBC cell lines were treated with vehicle (EtOH 0.01%), mif (100 nM), paclitaxel (100 nM), dex (100 nM)/paclitaxel (100 nM), or dex (100 nM)/mif (100 nM)/paclitaxel (100 nM). Error bars represent +/−SEM. *Denotes the two treatments compared to derive the p-value shown. B, Representative phase-contrast and cytotoxic fluorescent dye images of three TNBC cell lines using automated IncuCyte imaging. C, Total GR (NR3C1) mRNA expression was first normalized to β-Actin mRNA expression and then compared (fold difference) to one technical replicate associated with MDA-MB-231 NR3C1 levels (error bars reflect +/− SEM of three independent experiments). D, Western blot of glucocorticoid receptor (GR) and β-Actin. GRα translational isoforms are labeled.

Figure 2. Caspase-3 and PARP cleavage following treatment with mifepristone

MDA-MB-231 cells were treated with vehicle (EtOH 0.01%), dex (100 nM), mif (100 nM), paclitaxel (100 nM), dex/paclitaxel, or dex/mif/paclitaxel for 24 and 48 hours. The addition of mifepristone resulted in increased cleaved caspase-3 and PARP compared to dex/pac treatment alone.

Figure 3. SGK1 and MKP1/DUSP1 steady-state mRNA analysis following 4 hrs of dex vs. dex/mif treatment

MDA-MD-231, MDA-MB-468, or BT-20 cells were treated with vehicle (EtOH 0.01%), dex (100 nM) or dex/mif (100 nM) for 4 hours. SGK1 and MKP1/DUSP1 mRNA expression was normalized to β-Actin mRNA levels. Normalized SGK1 and MKP1/DUSP1 fold-change relative to vehicle treatment is shown (n=3 experiments +/− SEM). A, In MDA-MB-231 cells, dex-induced SGK1 (P=0.02) and MKP1/DUSP1 (P=0.008) mRNA expression was significantly reversed following concomitant mifepristone treatment. B, BT-20 cells showed significantly less induction of both SGK1 (P=0.004) and MKP1/DUSP1 (P=0.005) mRNA expression with the addition of mifepristone. C, In MDA-MB-468 cells, dex-induced SGK1 (P=0.001) steady-state mRNA levels were significantly higher compared with dex/mif treated cells, although the difference in MKP1/DUSP1 mRNA levels was not significant (NS). (* P<0.05, ** P<0.01, *** P<0.001).

Figure 4. Paclitaxel +/− mifepristone treatment of MDA-MB-231 tumor xenografts

A, Mice bearing MDA-MB-231 human TNBC xenograft tumors in the thoracic mammary fat pad were treated (arrows) with vehicle (closed squares, n=5), paclitaxel (open circles, n=7) mifepristone (closed triangles, n=6) via IP injection for five consecutive days. Tumors were measured three times a week for 35 days. Daily pre-treatment with mifepristone significantly improved tumor response compared to treatment with paclitaxel alone (P=0.04). To determine the significance of treatment effects on tumor growth, repeated measures ANOVA was performed. Error bars represent +/− SEM. B, Tumors were removed at the endpoint of the experiment, fixed, paraffin-embedded and sectioned followed by fluorescent immunohistochemistry. A higher percentage of GR+ cells remained in tumors following treatment with paclitaxel alone (n=6 animals) compared to tumors from SCID mice treated with mif/paclitaxel (n=5 mice) (P=0.11). C, MDA-MB-231 cells were treated with vehicle (EtOH 0.01%), mif (100 nM), paclitaxel (100 nM), dex (100 nM)/paclitaxel (100 nM), or dex (100 nM)/mif (100 nM)/paclitaxel (100 nM) and colonies were counted after 96 hours of treatment. Error bars represent +/−SEM. *Denotes the two treatments compared to derive the p-value shown.