Androgen receptor splice variants circumvent AR blockade by microtubule-targeting agents

Abstract

Docetaxel-based chemotherapy is established as a first-line treatment and standard of care for patients with metastatic castration-resistant prostate cancer. However, half of the patients do not respond to treatment and those do respond eventually become refractory. A better understanding of the resistance mechanisms to taxane chemotherapy is both urgent and clinical significant, as taxanes (docetaxel and cabazitaxel) are being used in various clinical settings. Sustained signaling through the androgen receptor (AR) has been established as a hallmark of CRPC. Recently, splicing variants of AR (AR-Vs) that lack the ligand-binding domain (LBD) have been identified. These variants are constitutively active and drive prostate cancer growth in a castration-resistant manner. In taxane-resistant cell lines, we found the expression of a major variant, AR-V7, was upregulated. Furthermore, ectopic expression of two clinically relevant AR-Vs (AR-V7 and ARV567es), but not the full-length AR (AR-FL), reduced the sensitivities to taxanes in LNCaP cells. Treatment with taxanes inhibited the transcriptional activity of AR-FL, but not those of AR-Vs. This could be explained, at least in part, due to the inability of taxanes to block the nuclear translocation of AR-Vs. Through a series of deletion constructs, the microtubule-binding activity was mapped to the LBD of AR. Finally, taxane-induced cytoplasm sequestration of AR-FL was alleviated when AR-Vs were present. These findings provide evidence that constitutively active AR-Vs maintain the AR signaling axis by evading the inhibitory effects of microtubule-targeting agents, suggesting that these AR-Vs play a role in resistance to taxane chemotherapy.

Keywords: androgen receptor; microtubule; prostate cancer; splice variants; taxane chemotherapy.

Figure 1. Upregulation of AR-V7 in taxane-resistant prostate cancer cells

A. and B. 22Rv1 with acquired resistance to taxanes were established by culturing in escalating doses of docetaxel (DTX) or paclitaxel (PTX). MTT assays were performed in passage-matched 22Rv1 or 22Rv1 resistant cells to determine the responses to taxanes. C. The response of DTX-resistant LNCaP95 to docetaxel treatment. D. and E. Western blotting using an anti-N terminal antibody or an AR-V7-specific antibody in 22Rv1 (D) or LNCaP95 (E) resistant cells. Rv1/LN95, passage-matched parental line; DR, docetaxel-resistant; PR, paclitaxel-resistant. The P values were determined by the Student's t-tests, ** denotes P < 0.01. The results presented are mean ± SEM from three experiments.

Figure 2. Expression of constitutively active AR-Vs negatively impact the cytotoxicities of taxanes

A. LNCaP cells were transfected with vector, AR-FL, AR-V7, or AR^v567es, and cell viability was determined by the MTT assay after 48 h of treatment with docetaxel. Western analysis was performed with an antibody recognizes the N-terminus of AR. The P values were determined by the Student's t-tests. *P < 0.05; **P < 0.01 vs vector. B. LNCaP95 cells were cultured in an androgen-depleted condition, and transfected with a control or an AR-V7-specific shRNA. **P < 0.01. CTX, cabazitaxel. The results presented are mean ± SEM.

Figure 3. Transcriptional activities of constitutively active AR-Vs are refractory to taxane treatment

COS-7 cells were transfected with the ARR3-luc reporter plasmid along with a plasmid encoding AR-FL, AR-V7, or AR^v567es. The luciferase reporter assay was performed after 24 h treatment. The P values were determined by the Student's t-tests. **P < 0.01 vs untreated. Doses: DTX, 1 and 2.5 nM; PTX, 2.5 and 5 nM. The results presented are mean ± SEM from three experiments.

Figure 4. Nuclear imports of constitutively active AR-Vs are microtubule-independent

FRAP assays were performed in COS-7 cells expressing different fluorescence-tagged AR proteins. Cells transfected with EGFP-AR-FL were cultured in the presence of androgen. Cells were treated with 20 nM docetaxel for 2 h before photobleaching. A. Confocal images taken at different intervals after photobleaching of the nuclei. White and yellow arrows indicate the nucleus and the cytoplasm, respectively. B. Recovery plot of the nuclear:cytoplamic fluorescence ratio (Fn/c) over time in cells treated with different microtubule inhibitors. Fn/c ratios are expressed as fractions of the pre-photobleach Fn/c. Nocodazole (NCZ) was used at 5 μg/ml and KX-01 was at 100 nM. FRAP images for NCZ and KX-01 are in Supplementary Figure S4.

Figure 5. The full-length AR associates with the microtubules

COS-7 cells were transfected with an expression vector for AR-FL and in vivo microtubule binding assay was performed with a commercial kit (Cytoskeleton, BK038). Nocodazole (NCZ), CaCl₂, and low temperature (cold) were used to disrupt microtubule integrity. Assembled microtubules were precipitated by ultracentrifugation and the pellet was resuspended and analyzed by Western blot (Top). Importin β and p53 were used as negative and positive controls, respectively, and histone H3 was used to detect nuclear contamination. P, pellet; W, wash; S, supernatant. Bottom, the microtubule-binding activities for AR and p53 were quantitated by the P/S ratios. The results presented are mean ± SEM from three experiments.

Figure 6. Microtubule-binding activity is mapped to the ligand-binding domain of AR

Left panel, a series of deletion constructs encompassing different domains of AR were generated and expressed in COS-7 cells. Right panel, the microtubule-binding activities of these constructs were analyzed by the in vivo microtubule binding assay and the Western blots (Supplementary Figure S5) were quantitated to calculate the P/S ratios. The results presented are mean ± SEM from three experiments. MT, microtubule.

Figure 7. Poor microtubule-binding activities of the AR-Vs

COS-7 cells were transfected with an expression vector for AR-FL, AR-V7, and AR^v567es and cultured in an androgen-deprived condition. A. In vivo MT-binding assays. B. quantitation of the results in A. The results presented are mean ± SEM from three experiments. C. Western blot showing that the proteins were expressed at similar levels after transfection.

Figure 8. Cytoplasmic sequestration of AR-FL by docetaxel is attenuated by AR-V7 and AR^v567es

A. Confocal fluorescence microscopy of EGFP-AR-FL subcellular localization when it was expressed with TurboFP or with a TurboFP-tagged AR-V in COS-7 cells. B. Based on distribution of the green fluorescence signal, cells were categorized into cytoplasmic (N < C), or nuclear and equally nuclear and cytoplasmic (N ≥ C).% of cells in each category were quantified. DRAQ5 was used to stain the nuclei. Cells cultured in an androgen-deprived condition were pre-treated with 10 nM docetaxel for 6 hr, followed by treatment with 1 nM R1881 for 4 hr. ** and ## P < 0.01. C. In vivo MT-binding assay in COS-7 cells expressing AR-FL alone, or with AR-V7 or AR^v567es.

Figure 8. Cytoplasmic sequestration of AR-FL by docetaxel is attenuated by AR-V7 and AR^v567es

A. Confocal fluorescence microscopy of EGFP-AR-FL subcellular localization when it was expressed with TurboFP or with a TurboFP-tagged AR-V in COS-7 cells. B. Based on distribution of the green fluorescence signal, cells were categorized into cytoplasmic (N < C), or nuclear and equally nuclear and cytoplasmic (N ≥ C).% of cells in each category were quantified. DRAQ5 was used to stain the nuclei. Cells cultured in an androgen-deprived condition were pre-treated with 10 nM docetaxel for 6 hr, followed by treatment with 1 nM R1881 for 4 hr. ** and ## P < 0.01. C. In vivo MT-binding assay in COS-7 cells expressing AR-FL alone, or with AR-V7 or AR^v567es.

Figure 9. Nuclear translocation of AR-Vs is importin β-dependent

A. FRAP assays were performed in COS-7 cells expressing EGFP-tagged AR-V7. Cells were treated with DMSO or 50 μM importazole (IPZ) for 2 h before photobleaching. Confocal images taken at different intervals after photobleaching of the nuclei. Red and yellow arrows indicate nucleus and cytoplasm, respectively. B. Fn/c recovery plot for EGFP-AR-V7. C. COS-7 cells transfected with pEGFP-AR-V7 were treated with DMSO or 10 μM importazole for 48 h. DAPI was used for staining the nuclei. D. & E. confocal images (D) and Fn/c recovery plot (E) of FRAP assays in COS-7 cells expressing TurboFP635-tagged AR^v567es and treated with IPZ. D. & E. confocal images (D) and Fn/c recovery plot (E) of FRAP assays in COS-7 cells expressing TurboFP635-tagged AR^v567es and treated with IPZ.