Article Figures & Data
Figures
- Figure 1.
AR inhibition decreases ER+/AR+ breast cancer growth. A, proliferation of MCF7 cells treated with increasing concentrations of enzalutamide (Enza) was monitored by IncuCyte. B, MCF7 cells were grown in soft agar with enzalutamide or tamoxifen and colony size was measured by ImageJ. C, immunoblotting for AR, ER, and Tubulin in MCF7 cells expressing a nontargeting (shNeg) or AR-targeting (shAR15 and shAR17) shRNA constructs (top). Proliferation was monitored by IncuCyte (bottom). D, MCF7 cells were grown in media with CSS for 72 hours then treated with vehicle (Veh), E2, or E2 + enzalutamide or MJC13 and cell number was measured by crystal violet. E, MCF7 cells were grown in media with CSS for 72 hours then treated with Veh, E2, or E2 + enzalutamide for 24 hours followed by cell-cycle analysis. F, MCF7 cells expressing shNeg, shAR15, or shAR17 were cultured in media with CSS for 72 hours then treated with vehicle or E2 and growth was measured by crystal violet. Error bars, SEM. *, P < 0.05; ****, P < 0.0001 by ANOVA with Dunnett multiple comparison test.
- Figure 2.
AR inhibitors diminish ER genomic binding. ChIP-seq for ER in MCF7 cells grown in CSS for 3 days then treated with E2 ± enzalutamide or MJC-13. A, heatmap of ER binding. The heatmap is shown with a horizontal window of ± 2 kb. B, the number of binding sites identified by MACS2, using vehicle treatment as the control. C, the ER ChIP-seq signal at individual sites with E2 alone (x-axis) versus E2 + enzalutamide (blue) or MJC13 (red; y-axis). D and E, ChIP-qPCR (D) and ChIP-seq read depth (E) at well-characterized ER-binding sites. Error bars, SEM. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by ANOVA with Dunnett multiple comparison test.
- Figure 3.
AR and ER colocalize in the nucleus in response to E2. A, MCF7 cells were grown in media with CSS for 72 hours then pretreated with vehicle (veh), 10 μmol/L enzalutamide, or 1 μmol/L bicalutamide (bic). Following pretreatment, cells were treated with vehicle or 10 nmol/L E2 ± enzalutamide or bicalutamide as shown for an additional 3 hours. Cells were then fixed and ICC was performed for AR (green) and ER (red). B and C, MCF7 cells were grown in media with CSS for 72 hours then treated with the indicated treatment for 3 hours, and nuclear extracts were immunoblotted for AR and TOPO1. D, ER+/AR+ ZR-75-1 or (E) ER−/AR+ MDA-453 cells were grown in media with CSS for 72 hours, then pretreated for 3 hours with enzalutamide or vehicle control. Following pretreatment, cells were treated with vehicle, 10 nmol/L DHT, or 10 nmol/L E2 ± enzalutamide as shown for 3 additional hours. Nuclear extracts were then obtained and subjected to Western blotting for AR and TopoI. F, MCF7 cells were grown in media with CSS for 72 hours then treated with E2 ± enzalutamide for 1 hour followed by fixation and PLA staining for AR and ER (red). Nuclei were stained with DAPI (blue). G, fluorescent intensity per nuclei was measured by CellProfiler. Error bars, SEM. ****, P < 0.0001 by ANOVA with Dunnett multiple comparison test.
- Figure 4.
E2 induces AR genome binding that overlaps with ER binding. ChIP-seq for AR in MCF7 cells grown in CSS for 3 days then treated with E2 for 1 hour or DHT for 4 hours. A, heatmap of binding showing a horizontal window of ± 2 kb and enriched motifs from each category. B, the number of binding sites identified by MACS2, using vehicle treatment as the control. C, the number of AR-binding sites that are unique to DHT (red), unique to E2 (blue), or shared (overlap) are shown. D and E, ChIP-qPCR (D) and ChIP-seq read depth (E) results show AR binding at well-characterized ER-binding sites following E2 treatment. F, the percentage of AR-binding sites in response to DHT (left) or E2 (right) that were also identified as ER-binding sites (blue) is shown.
- Figure 5.
Enzalutamide (Enza) synergizes with tamoxifen (tam) and fulvestrant in vitro. A, T47D cells were grown in media with complete serum and enzalutamide and/or tamoxifen, and cell number was monitored by IncuCyte. Percent inhibition was compared to vehicle after 5 days, and synergy was calculated using CalcuSyn software. A combination index (CI) value < 1 is indicative of synergy. B and C, BCK4 or PT12 cells were grown in phenol red–free media with CSS for 1 day then treated with E2 with enzalutamide and/or fulvestrant and cell number was monitored by IncuCyte.
- Figure 6.
Enzalutamide (Enza) inhibits tamoxifen-resistant tumor growth in vitro and in vivo, and AR is expressed in recurrent breast cancers. A, growth of MCF7-TamR cells treated with vehicle, tamoxifen (Tam), enzalutamide, or MJC13 for 7 days. B, MCF7-TamR cells were plated in soft agar and the number of colonies was counted after 14 days. C and D, MCF7-TamR cells were implanted into the mammary glands of nude mice with estrogen pellets and were matched into groups to receive either control chow (CTRL), tamoxifen pellets, enzalutamide-containing chow, or both (tam+enza). C, tumor growth was measured over time by luminescence. D, final tumor weights of mice from each group. E, IHC for AR and ER in clinical samples of patient-matched primary tumor and recurrence 110 months later. F, IHC for AR and ER in clinical samples of patient-matched primary tumor and metastasis 167 months later (400×). *, P < 0.05; ***, P < 0.001; ****, P < 0.0001 by ANOVA with Dunnett multiple comparison test.
- Figure 7.
Enzalutamide (Enza) decreases hormone-driven growth of PT12 primary tumors and metastases. A–D, 1 × 106 GFP-luciferase expressing PT12 cells were injected orthotopically into the mammary fat pad of NOD-SCID-IL2Rgc−/− mice followed by implantation of either an E2 or DHT pellet. When tumors reached an average of 39 mm3, mice were matched into the following groups: E2 with control chow (CTRL; n = 10) or enzalutamide chow (n = 10), or DHT with control chow (n = 5) or enzalutamide chow (n = 5). Tumor viability was measured by IVIS for mice with E2 pellets (A) or DHT pellets (B). C, BrdUrd staining of tumors from mice with E2 pellets. D, Cleaved caspase staining of tumors from mice with DHT pellets. E, PT12 GFP-luciferase cells were injected intracardially in NOD-SCID-IL2Rgc−/− mice. Metastatic burden of mice treated with E2 or E2 + enzalutamide was monitored using IVIS (photons/second) over 12 weeks (total signal supine+prone). F, IVIS signal from mice in supine position at 2 weeks versus 12 weeks in E2 versus E2 + enzalutamide mice. G, IVIS image of mice in the supine position in the E2 (n = 11) or E2 + enzalutamide (n = 12) groups after 12 weeks, red denotes high IVIS signal. Error bars, SEM. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001 by repeated-measures mixed model approach for (A and B), ANOVA with Bonferroni's multiple comparison test for (C and D), and Wilcoxon rank sum test for E.
Additional Files
Supplementary Data
- Supplementary Methods - Supplementary Methods
- Supplementary Figure Legends - Supplementary Figure Legends
- Supplementary Tables 1-3 - S1. E2 induces AR binding at sites that overlap with ER binding. S2. Enza alters expression of E2-regulated genes in PT12 xenograft tumors. S3. Pathway analysis of Enza-regulated genes in PT12 xenograft tumors.
- Supplementary Figures 1-6 - S1. Enza or AR knockdown decrease baseline and E2-induced proliferation of breast cancer cells in vitro. S2. E2 induces and Enza inhibits AR nuclear localization. S3. AR and ER binding in MCF7 cells. S4. Enza synergizes with anti-estrogens. S5. The combination of Enza plus tam reduces ER. S6. PT12 cells are AR+ and Enza alters expression of E2-regulated genes in PT12 xenograft tumors.
Volume 14, Issue 11
