SEARCH

SEARCH BY CITATION

Keywords:

  • androgen-ablation therapy;
  • autophagy;
  • chemotherapy

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

What's known on the subject? and What does the study add?

  • Androgen-ablation therapy (AAT) and chemotherapy are commonly used to treat incurable prostate cancer. To improve outcome, there is major on-going research to develop more effective treatments with less toxicity. Autophagy has been suggested from previous studies to play a potential role in cell survival and may be associated with resistance to chemotherapy.
  • Autophagy is known to be upregulated by nutrient starvation or AAT in prostate cancer. However, its functional impact is not fully known. The present study describes the potential synergism between the blockade of autophagy and AAT alone or AAT combined with taxane chemotherapy. Hence, future combined treatment options are warranted to further investigate the clinical impact of autophagy suppression as a treatment strategy.

Objective

  • To study the cellular effects of the anti-androgen bicalutamide on autophagy and its potential impact on response to androgen-ablation therapy (AAT) alone or combined with docetaxel chemotherapy in human prostate cancer LNCaP cells.

Materials and Methods

  • LNCaP cells were treated with bicalutamide ± docetaxel, and cellular effects were assayed: lipidated LC3 (a microtubule-associated protein) for autophagy and its trafficking to fuse with lysosome; flow cytometry using propidium iodide or caspase 3 for cell death; and sulforhodamine B assay for cell growth.

Results

  • Bicalutamide treatment enhanced autophagy in LNCaP cells with increased level of autophagosome coupled with an altered cellular morphology reminiscent of neuroendocrine differentiation.
  • Consistent with the literature on the interaction between androgen receptor activation and taxane chemotherapy, bicalutamide diminished docetaxel mediated cytotoxicity.
  • Significantly, pharmacological inhibition of autophagy with 3-methyladenine significantly enhanced the efficacy cell kill mediated by AAT ± docetaxel.

Conclusion

  • Autophagy associated with bicalutamide treatment in LNCaP cells may have a pro-survival effect and strategy to modulate autophagy may have a potential therapeutic value.

Abbreviations
AAT

androgen-ablation therapy

AR

androgen receptor

DAPI

4′,6-diamidino-2-phenylindole

FBS

fetal bovine serum

FITC

fluorescein isothiocyanate

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

GFP

green fluorescent protein

3-MA

3-methyladenine

mRFP

monomeric red fluorescent protein

mTOR

mammalian target of rapamycin

siRNA

small interfering RNA

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

Androgen receptor (AR) signalling critically contributes to prostate carcinogenesis and is an important target for therapy [1]. Androgen-ablation therapy (AAT) with non-steroidal AR antagonists, e.g. bicalutamide, and/or GnRH agonists often results in the development of castrate-resistant disease. Bicalutamide (or Casodex) binds to the ligand-binding domain of the AR and competes against natural AR ligands to disrupt the recruitment ofco-factors [2], suppressing proliferation and inducing apoptosis [3, 4].

Docetaxel is the first chemotherapy found to extend survival in the patients with castrate resistant prostate cancer (CRPC) [5, 6]. However, there is a significant toxicity profile associated with taxane-based chemotherapy while a proportion of prostate cancer is intrinsically resistant to or quickly develops relapsed disease. Docetaxel is a microtubule targeting agent that acts by disrupting the microtubular dynamics, causing aberrant cell division and cell death [7]. Microtubules have also been shown to have a role in the activity of nuclear hormone receptors. Recently, oestrogen receptor has been shown to interact with the β-tubulin subunit, and microtubule stabilisation has been found to inhibit nuclear activity of the oestrogen receptor by preventing nuclear accumulation whilst enhancing the cytoplasmic signalling through the PI3-K pathway [8].

A recently emerging theme from studies investigating the mechanisms of therapeutic resistance is autophagy [9]. Autophagy is a homeostatic process whereby cellular components are engulfed into vesicles known as autophagosomes, which fuse with lysosomes and are consequently subjected to proteolytic degradation. This process is upregulated in response to cellular stress, e.g. starvation, hypoxia or exposure to noxious agents, as a means to generate nutrients and energy from pre-existing molecules or to clear damaged intracellular structures [10, 11]. In many cases, this process is induced as a survival mechanism, allowing tumour cells to evade damaging effects of therapy and to promote chemo- or radio-resistance [9, 12, 13]. We have recently reported that ligand activation of AR delays the induction of autophagy in response to starvation in prostate cancer cells [14]. This finding led us to ask whether, conversely, suppressed AR activation is a sufficient stimulus to induce autophagy. In addition, we also studied the significance of autophagy after AAT alone or AAT combined with chemotherapy.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

Cell Culture Assays

Cells were routinely maintained in RPMI 1640-based medium containing 10% fetal calf serum (FCS). Where media depleted of androgens was required, dextran coated charcoal was used to remove endogenous steroids present in the FCS as previously described [15]. LNCaP-AI cells, an androgen independent subclone of LNCaP cells, were serially maintained in RPMI 1640 medium supplemented with 10% charcoal stripped FCS as previously described [16]. Synthetic androgen R1881 (Perkin-Elmer) was added where indicated at a concentration of 10 nm and where appropriate cells were pre-treated with 10 μm bicalutamide (Casodex, AstraZeneca).

For green fluorescent protein (GFP)-LC3 expression, subconfluent LNCaP cells were infected with an adenoviral vector encoding GFP-LC3 fusion, a kind gift of Aviva Tolkovsky (University of Cambridge, UK). At 24-h after infection, cells were washed with fresh medium before treatment. Stable LNCaP cell lines expressing pWZL-monomeric red fluorescent protein (mRFP)-GFP-LC3 vector were established as previously described [14]. Propidium iodide was purchased from Sigma (St Louis, MO, USA). 3-methyladenine (3-MA) was purchased from Calbiochem (Merck, Darmstadt, Germany).

Cellular proliferation and growth inhibition in response to docetaxel was determined by Sulphorhodamine B (SRB) assay based on a protocol by Skehan et al. [17]. Cells were seeded into 96-well plates at an appropriate density and allowed to adhere for 24 h, then placed into various concentrations of docetaxel in quadruplicate. Cells were then incubated for three doubling times (or 72 h) before being fixed with 10% cold trichloroacetic acid (TCA) for 1 hat 4 °C. Cells were then washed five times with water, dried at 37 °C for 1 h and stained with 100 μL of SRB dye (0.4% SRB in 1% acetic acid) for 30 min. After staining, cells were washed in 1% acetic acid and air dried before the solubilisation of the dye in 10 mm Tris (pH 10.5) for 30 min with agitation. Optical density was read at 570 nm using a BioRad 680 Microplate Reader.

Flow Cytometry

Cell death was assayed either by an Activated Caspase-3 fluorescein isothiocyanate (FITC) MAb Apoptosis Kit (BD PharmingenTM) or propidium iodide staining. Cells were exposed to varying concentration of docetaxel for 48 h before being harvested, fixed and stained with an anti-active Caspase-3-FITC labelled antibody, according to the manufacturers' protocol, and subsequently analysed by flow cytometry. Alternatively, adherent and floating cells were collected, pelleted and re-suspended in 0.1 mg/mL propidium iodide. Cells were analysed according to fluorescence intensity using FACSCaliber flow cytometer and Cell Quest software, Version 5.2.1 (BD Bioscience, San Jose, CA, USA).

Cell cycle distribution was also analysed by flow cytometry. Cells were plated in 6-well plates at a density of 80 000 cells per well. At the end of the experimental period, adherent and non-adherent cells were harvested and re-suspended in 2% FCS. Permeabilisation of cells was achieved by treatment with 5% Triton X-100, then stained with 2.5 mg/mL propidium iodide. To ensure exclusive DNA staining 100 μg/mL RNase was added to the cell suspension.

After respective fluorochrome staining, cells were injected into a Becton Dickinson FACScan. At least 10 000 cells were evaluated for each sample. Isotype controls were used as negative controls where appropriate. CellQuest software was used for data acquisition and data analysis performed on the WINMDI and Cychlred software programs. Cells were gated on Forward Scatter vs Side Scatter to exclude cell debris and doublet discrimination was carried out on the FL2-Width vs FL2-Area dot-blot to ensure only single cells were examined for cell cycle.

Western Blotting

Western blot analysis was performed as described previously [14]. Primary antibodies used in this study were purchased from Sigma (actin, clone AC-40; glyceraldehyde 3-phosphate dehydrogenase (GAPDH)-Peroxidase, clone GAPDH-71.1); Cell Signalling Technology, Danvers, MA, USA (phospho AKT (S473); phospho p70 S6 Kinase (T421/S424); phospho S6 (S235/6), clone 2F9; AKT; p70 S6 Kinase; mammalian target of rapamycin [mTOR]; S6); Novus, Littleton, CO, USA (LC3); BD Bioscience, San Jose, CA, USA (Cathepsin L; LAMP-2) and Santa Cruz Biotechnology, Santa Cruz, CA, USA (AR; PSA, clone C-19, Cathepsin D). Secondary antibodies used were anti-mouse, anti-rabbit and anti-sheep IgG-horseradish peroxidase conjugate from Cell Signalling Technology.

Small Interfering RNA (siRNA)

siRNA oliogonucleotides used in this study were purchased from Ambion/Applied Biosystems (Austin, TX, USA). For siRNA uptake, cells were nucleofected using Nucleofector Kit R (Lonza, Basel, Switzerland) according to the manufacturer's instructions. Knock down of AR expression was achieved using 2 μL of a 100 μm solution per 100 μL of cells in nucleofector solution to give a final concentration of 200 pmol. At 24 h after nucleofection, medium was changed to treatment conditions for the indicated time.

Immunostaining and Confocal Microscopy

To visualise punctae in GFP-LC3 or mRFP-GFP-LC3-expressing cells, cultures were fixed in 4% paraformaldahyde, washed in PBS and mounted in Vectashield with 4′,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Burlingame, CA, USA). For Cathepsin D immunostaining, fixed cells were blocked in PBS containing 10% fetal bovine serum (FBS) and 0.5% BSA, followed by incubation with Cathepsin D antibody (Santa Cruz) and anti-sheep AlexaFluor555 secondary antibody (Invitrogen) and mounted as described.

Cells were visualised using an Olympus FV-1000 confocal microscope under 63X objective. The size, number and spatial distribution of punctae and lysosomes were analysed using Image J software (Version 1.40 g).

Statistical Analysis

Statistical analysis was performed using the unpaired Student's t-test, with differences of P < 0.05 considered to indicate statistical significance.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

The Anti-Androgen Bicalutamide Modulates Autophagy in LNCaP Cells

LC3 is a microtubule-associated protein. The lipidated form of LC3 is recruited to the membrane of autophagosomes during autophagy and is commonly used as a marker for the process [10]. To examine the relationship between autophagy and pharmacological blockade of AR, we treated LNCaP cells transiently expressing GFP-tagged LC3 with 10 μm bicalutamide for 24 h. Bicalutamide increased the number of cells positive for GFP-LC3 punctae from 30% to >60% (Fig. 1A). This suggests that inhibition of AR may be a sufficient stimulus to modulate autophagy in LNCaP cells. Furthermore, bicalutamide treatment increased the level of lipidated LC3 (LC3-II) [18] while AR and PSA protein levels decreased as expected (Fig. 1B). In addition, AR has been linked with activation of the mTOR/Akt pathway [19, 20], the suppression of which has been shown in many systems to be critical for autophagy induction [21]. Attenuation of the mTOR pathway by bicalutamide reduced phosphorylation of p-Akt, p-S6K and p-S6 kinases (Fig. 1B).

figure

Figure 1. Bicalutamide (bical) treatment induces autophagy in an AR ligand-binding dependent manner. (A) Top panel: representative confocal images of cells transiently expressing GFP-LC3 maintained for 24 h in full medium (FBS) with ethanol vehicle or 10 μm bicalutamide. Scale bar, 20 nm. Green, GFP-LC3; blue, DAPI. Bottom panel: quantification of GFP-LC3 puncta-positive cells depicted in A. A positive cell contained five or more punctae. 50 cells per condition per experiment were analysed. (B) Cell lysates maintained as described in A were Western blotted with the indicated antibodies. The lipid-conjugated (LC3-II) and unconjugated (LC3-I) forms are indicated by arrows. Blots are representative of three independent experiments. (C) Control, scrambled siRNA- and AR siRNA-treated cells were maintained for 24 h in FBS with vehicle or 10 μm bicalutamide before lysis. Lysates were analysed by Western blot for AR and PSA expression, with actin as loading control. Densitometric analysis of AR and PSA levels normalised to actin control are indicated below, relative to FBS control. (D) Representative confocal images of LNCaP cells transiently expressing GFP-LC3. Scale bar, 20 nm. Green, GFP-LC3; blue, DAPI. (E) Quantification of GFP-LC3 puncta-positive cells depicted in (D). A positive cell contained five or more punctae. 50 cells per condition per experiment were analysed, n = 3, error bars indicate sem, **P < 0.01.

Download figure to PowerPoint

AR itself is an androgen-regulated gene [22] and bicalutamide treatment suppressed AR expression (Fig. 1B). Hence, bicalutamide could induce autophagy via either reduction in AR protein level or blockade of ligand mediated AR activation. We addressed this by treating cells with siRNA against AR, in the presence or absence of bicalutamide. Compared with control and scrambled siRNA-treated cells, AR siRNA markedly reduced AR protein as well as PSA levels, confirming AR knockdown is sufficient in suppressing its transcriptional function (Fig. 1C). Similarly, bicalutamide led to decreased AR and PSA expression in control and scrambled siRNA-treated cells, an effect further pronounced with the addition of AR siRNA (Fig. 1C).

Confocal microscopy was used to quantify GFP-LC3 positive cells as an indicator of autophagy (Fig. 1D). In AR knockdown cells maintained with serum supplemented, a condition where the low residual levels of AR are exposed to androgens present in serum, autophagy did not exceed basal levels despite lower AR expression than that observed with bicalutamide (Fig. 1D,E). However, in bicalutamide treated cells, where (residual) AR protein is subjected to androgen blockade, autophagy was significantly induced in all treatment groups (Fig. 1E). There was no additional increase in the proportion of GFP-LC3 punctae positivity in bicalutamide-treated cells with AR siRNA compared with scrambled siRNA. Taken together, bicalutamide induced autophagy depends on inhibition of the AR ligand-mediated activation, rather than on a net reduction of AR protein expression.

The Rate of Autophagy Induction and Turnover is Enhanced by Bicalutamide

The accumulation of GFP-LC3 positive punctae (Fig. 1A) and upregulated LC3 lipidation (LC3-II, Fig. 1B) can indicate either an induction of autophagy and/or an inhibition of autophagosomal–lysosomal turnover. To study the dynamic of autophagosomal flux, LNCaP cells stably expressing mRFP-GFP-LC3 were treated with bicalutamide. The pH-sensitive GFP signal is lost upon fusion with the acidic lysosomal compartment, while mRFP fluorescence persists. Therefore, the fate of autophagosomes can be tracked by merging images of GFP and mRFP fluorescence [14].

Bicalutamide treated mRFP-GFP-LC3-expressing LNCaP cells showed an increase in the numbers of co-localised mRFP+/GFP+ (autophagosomal) and single mRFP+ (autolysosomal) punctae (Fig. 2A–C) after 24 h. While both PSA and AR expression continued to be suppressed at 96 h (Fig. S1), there was a further increase in the number of mRFP+/GFP+ and mRFP+ punctae (Fig. 2B,C). While the sizes of mRFP+/GFP+ punctae remained fairly static over the study period, the sizes of mRFP+ punctae significantly increased (Fig. 2D). This is consistent with highly dynamic effects of bicalutamide on autophagosome physiology.

figure

Figure 2. Trafficking of autophagosomes induced by long-term treatment with bicalutamide. (A) Representative confocal images of cells stably expressing mRFP-GFP-LC3. Merged images depict overlay of GFP (green) and mRFP (red) images. For clarity, GFP and mRFP in isolation are shown in white; their colours are shown in the merged image – co-localisation is represented by a yellow colour. Scale bar, 20 nm. (BD) Quantification of mRFP+/GFP+ (B) and mRFP+ (C) punctae in cells shown in (A). (D) Average size in pixels of mRFP+/GFP+ and mRFP+ punctae in cells maintained as in A. For B, C and D, 15–25 cells per condition per experiment were analysed, n = 3, error bars indicate sem, **P < 0.01 vs corresponding FBS control. In B, #P < 0.01 96 h vs 24 h and 48 h.

Download figure to PowerPoint

Bicalutamide Alters Cellular Morphology and Re-Distributes Autolysosomes

After sustained (96 h) bicalutamide treatment, LNCaP cells exhibited an elongated morphology with dendritic-like processes (Fig. 3A, top panels), consistent with neuroendocrine-like differentiation associated with castrate resistance [23]. In mRFP-GFP-LC3-expressing LNCaP cells, we observed that many punctae were concentrated in these peripheral processes (Fig. 3A, bottom panels). As lysosomal positioning is increasingly recognised to coordinate cellular nutrient responses including autophagy and mTORC1 signalling [24], we examined the dynamic of LC3-labelled structures by measuring the distance of both mRFP+/GFP and mRFP puncta types from the nucleus after bicalutamide treatment and subsequently performing a frequency distribution analysis. There was a dramatic re-localisation of mRFP+ punctae from peri-nuclear to the periphery (Fig. 3B,C), with only minor differences seen in the distribution of mRFP+/GFP punctae (Fig. 3B,D). These data are consistent with (i) bicalutamide induced autophagosomes being formed centrally, and (ii) as these vesicles mature, they fuse with lysosomes (signified by mRFP positivity) and redistribute to the periphery of bicalutamide treated morphologically altered LNCaP cells. The significance of such vesicular traffic requires further investigations.

figure

Figure 3. Long-term bicalutamide (bical) treatment promotes peripheral re-localisation of autophagosomes, autolysosomes and lysosomes. (A) Phase-contrast images of parental LNCaP cells (top panels) and merged confocal images of DAPI-stained mRFP-GFP-LC3 stable cells (middle panels) maintained for 96 h in FBS with vehicle or 10 μm bicalutamide. The bottom panels show boxed area in high magnification. Scale bars, 100 nm (top panels) or 20 nm (bottom panels). (B) Average distance from the nucleus in pixels of mRFP+/GFP+ and mRFP+ punctae. Cells were maintained in FBS with vehicle or 10 μ sem bicalutamide for the indicated time points. In all, 15–25 cells per condition per experiment were analysed, n = 3, error bars indicate sem, **P < 0.01 vs mRFP + FBS control or as indicated, #P < 0.05 vs mRFP+/GFP+ FBS control. (C and D) Representative frequency distribution of mRFP+ (C) and mRFP+/GFP+ (D) punctae in pixels. Measurements were taken from 15–25 cells per condition as described in A. (E) Cells were maintained in FBS with vehicle control or 10 μm bicalutamide for 96 h, lysed and Western blotted with the indicated antibodies, with GAPDH as a loading control. (F, G) The average distance from nucleus (F) and average number (G) of Cathepsin D+ lysosomes in cells maintained in FBS with vehicle or bicalutamide for 96 h. In all, 15–25 cells per condition per experiment were analysed, n = 3, error bars indicate sem, **P < 0.01.

Download figure to PowerPoint

As bicalutamide treatment led to the anterograde translocation of autolysosomes, we considered whether this was a general effect of AR inhibition upon lysosomal trafficking. In keeping with enhanced lysosome formation, expression of the mature cleaved form of lysosomal aspartate and cysteine proteases, Cathepsins D and L, were markedly enhanced upon bicalutamide treatment. In contrast, the lysosomal membrane protein LAMP2 was only weakly upregulated (Fig. 3E). Similar to the changes observed for mRFP+ autolysosome after bicalutamide treatment, Cathepsin D+ lysosomes increased in number and re-distributed to a peripheral location (Fig. 3F,G; Figs S2,S3) but, unlike mRFP+ punctae, did not change in their sizes (Fig. S4). Hence, the redistribution of autolysosomes may reflect a global trafficking effect of bicalutamide on an overlapping lysosomal population.

Autophagy after AAT ± Docetaxel Treatment may have a Pro-Survival Effect

We further tested the functional significance of autophagy in LNCaP cells after AAT with or without docetaxel chemotherapy. First, docetaxel treatment is confirmed to be highly cytotoxic to all prostate cancer cells tested (Supplementary Table S1). LNCaP cells chronically maintained in an androgen-depleted condition (LNCaP-AI cells) were ≈2.5-fold more resistant to docetaxel than the parental LNCaP cells, with GI50 values (the concentration required to achieve 50% growth inhibition) at 3.7 and 1.4 nm, respectively. Consistent with this finding, LNCaP cells showed a higher level of docetaxel induced cell death than LNCaP-AI cells (Fig. 4A,B; P < 0.05). Like LNCaP-AI cells, acute AAT in LNCaP cells reduced docetaxel induced cell death two-fold (P < 0.01; Fig. 4C). Furthermore, restoration of androgen to LNCaP-AI cells significantly enhanced docetaxel induced cell death (P < 0.04; Fig. 4D), and this effect was abolished by bicalutamide (P < 0.01; Fig. 4D). Hence, combined AAT and taxane chemotherapy induced sub-optimal cell kill when compared with taxane treatment alone.

figure

Figure 4. Assessment of growth inhibitory effects of docetaxel in prostate cancer cell lines: (A) Representative data showing the increased caspase 3-positive population in response to docetaxel in both LNCaP and LNCaP-AI cell lines. Apoptotic cells are those shifted upwards in the Y axis, which is a measure of fluorescence. (B) Fold change in cell death in parental LNCaP and LNCaP-AI cells (maintained in their normal culture conditions) following increasing doses of docetaxel (up to 1 nm). (C) Fold change in cell death induced by docetaxel in LNCaP cells (acutely androgen starved) in presence and absence of synthetic androgen analogue R1881 (10 nm). (D) Fold change in cell death in LNCaP-AI cells upon treatment with increasing doses of docetaxel alone (left panel), and androgen supplement with R1881 (10 nm; middle panel) and androgen supplement with anti-androgen bicalutamide (1 μm; right panel). Error bars represent sd from three independent experiments (*P < 0.01). (E and F) Bar chart showing cell death analysis in LNCaP cells comparing treatment with bicalutamide (Bic) and combined treatment with docetaxel (Doc) and bicalutamide (Bic). Data were normalised to the respective vehicle controls and showed the comparison between Bic vs Bic + Doc with or without 3-MA treatment (n = 3, error bars indicate sem, *P < 0.01).

Download figure to PowerPoint

Autophagy is reported to have both pro-survival as well as cell death promoting effects. Based on our data, we hypothesised that autophagy may have a pro-survival role after treatment with bicalutamide ± taxane chemotherapy. LNCaP cells were treated with bicalutamide in the presence or absence of the autophagy inhibitor 3-MA for 24 h. Analysing normalised data to their respective controls, the addition of 3-MA to bicalutamide treatment significantly enhanced cell death by 1.5-fold (P < 0.05; Fig. 4E). Furthermore, suppression of autophagy with 3-MA along with bicalutamide and docetaxel treatment resulted in a two-fold increase in cytotoxic effect (P = 0.01; Fig. 4F). When docetaxel was used simultaneously with bicalutamide, the mean (sd) percentage of cell death was at 9.4 (2.1)%, while upon the addition of 3-MA to bicalutamide and docetaxel treatment, the cell death rate increased to 19.4 (2.4)%, signifying a two-fold increase.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

AAT is intrinsically complicated by the eventual development of castrate-resistant disease. The present data in LNCaP cells treated with bicalutamide and the pro-survival effects of autophagy warrant further investigation with additional cell line models and in vivo studies. Future studies on additional AR-positive cell lines, e.g. CWR and LNCaP-AI cells, will be informative. Together, LNCaP and these additional cell models will be useful tools for mechanistic studies. Autophagy imparts cells with a degree of self-sustainability, allowing for their survival in unfavourable conditions [10, 11]. To our knowledge, this report is the first to attribute the upregulation of autophagy due to suppression of AR function by bicalutamide, in the absence of other stressors. Although bicalutamide is a well-documented promoter of apoptosis and growth-arrest, its ability to induce autophagy may in fact counteract its anti-tumorigenic effects [3]. Even in the face of widespread apoptosis, bicalutamide-induced autophagy may prime the survival of tumour cells in its microenvironment, which is often hypoxic, acidic and nutrient poor [14]. In addition, autophagy can also have a pro-survival effect when AAT is combined with taxane chemotherapy.

Prostate cancer cells subjected to long-term bicalutamide treatment also exhibit caspase-dependent death [4]. The simultaneous activation of caspases and autophagy is not uncommon; however, whether autophagy opposes or contributes to cell death is context specific [14, 25]. In the present study, we show that acute induction of autophagy by bicalutamide acts as a pro-survival response, akin to that observed in tamoxifen-resistant breast cancer cells. Future in vitro and in vivo studies into the impact of autophagy after AAT are warranted. The mechanism by which AR inhibition by bicalutamide activates autophagy is unclear; however, the present results suggest that reduction of mTOR signalling, potentially via abrogated Akt activation, may underlie this response. Numerous reports have implicated a cross-talk between AR and PI3-K/Akt pathways in prostate cells, showing both positive and negative effects upon non-genomic signalling and AR transactivation [20, 26]. However, whether reduced Akt signalling underlies the present reduction in mTOR activation is not known, given that androgen can also activate S6K, S6 and 4E-binding protein 1 without Akt phosphorylation [19].

Docetaxel is a taxane-based chemotherapeutic agent acting as a microtubule stabiliser to disrupt microtubule dynamics. Defects of the mitotic spindle may result in aberrantly divided cells with G1 arrest and subsequent apoptosis [27-30]. However, it is increasingly appreciated that induction of apoptosis may occur regardless of the cell cycle state of a cell [31]. The use of docetaxel has improved clinical outcome in patients with castrate-resistant prostate cancer, but the survival benefit remains modest, and much research is on-going to further optimise its use and to identify novel tubulin-binding agents, either alone or in combination [32, 33]. Hence, improved insight in autophagy in association with conventional therapy may facilitate the development of a new treatment strategy.

The clinical significance of interaction between the microtubule and AR in prostate cancer remains to be determined. Previously, AR activity has been reported to enhance cellular sensitivity to paclitaxel [34]. To our knowledge, the present report has shown for the first time the ability of the anti-androgen bicalutamide in inducing autophagy as a pro-survival effect after AAT alone or combined with docetaxel chemotherapy. The present data suggests that simultaneous AAT and taxane-based therapy may reduce the overall anti-tumour efficacy, which may have an implication to the scheduling of microtubule-targeting agents and AAT. Hence, ‘decoupling’ of the timing of chemotherapy and AAT may be beneficial [35-37]. Future studies with either additional anti-androgens and/or AR knocked out prostate epithelial cells will be of use to further define the molecular basis of anti-androgen in autophagy. In addition, the present data on the use of bicalutamide is consistent with our recent data suggesting that ligand activation of AR delays the induction of autophagy in response to starvation in prostate cancer cells [14].

The cytoskeleton has long been reported to have a role in nuclear hormone receptor trafficking, in particular the movement of activated receptor complexes to the nucleus. This was first reported in relation to the movement of the glucocorticoid receptor, and its interaction with microtubules [38-40]. Hsp-90, a chaperone protein of the glucocorticoid receptor (and AR), was hypothesised to be a potential link between the receptor and the cytoskeleton [41]. The role of the microtubule network has been investigated in several nuclear hormone receptors. Previous work has found other cytoskeletal components to be important in the movement of the AR [42]. A recent study by Gan et al. [43] also showed that taxol may play a role in inhibiting the AR, with both paclitaxel and docetaxel treatment decreasing PSA expression.

The effect of bicalutamide upon lysosomal localisation could reflect cytoskeletal reorganisation, a possibility reinforced by the observed morphological change. Lysosomes are dynamic intracellular organelles intimately involved both in the activation of mTOR complex 1 signalling and in degrading autophagic substrates [44, 45]. Recently, lysosomal positioning has been implicated to coordinate anabolic and catabolic responses with nutrient availability by orchestrating early plasma–membrane signalling events, mTORC1 signalling and autophagy [24]. Autophagosomes are generated throughout the cytoplasm and are relocated along microtubules towards the centrosome, where fusion with the clustered lysosomes takes place [46]. Similarly, microtubule-dependent trafficking and peripheral lysosomal redistribution are closely linked in several conditions [47, 48]. Furthermore, androgen signalling promotes the accumulation of γ-tubulin at the centrosome to regulate microtubule nucleation [49].

It is unclear if the lysosomal redistribution and cathepsin upregulation in response to bicalutamide are related to the observation on survival effects, and future investigations may clarify the interaction between bicalutamide, lysosome/autophagosome physiology and cell survival. Cathepsins have been implicated in promoting invasion and metastasis [47, 50]. Peripheral redistribution of Cathepsin D+ lysosomes raises the possibility that protease exosecretion could be an outcome of AR inhibition in vivo, perpetuating an invasive and treatment resistant phenotype [51]. However, cathepsins have also been implicated in contributing to cell death either through their role in autophagy or caspases mediated death [52]. Thus, the precise role of cathepsin upregulation in response to bicalutamide treatment warrants further investigation [12].

These data support AAT-induced autophagy as a pro-survival process, as suppression of autophagy enhanced cytotoxic effect in LNCaP cells after bicalutamide treatment ± docetaxel chemotherapy. As taxane chemotherapy is clinically indicated for castrate resistant disease (similar to our in vitro model of AAT), the combination of taxane chemotherapy and autophagy targeted therapy may enhance the overall anti-tumour effects. In summary, the present study shows AAT mediated autophagy and its potential pro-survival impact in LNCaP cells after AAT ± chemotherapy. Future studies using additional cell lines in both in vitro and in vivo studies are warranted. To our knowledge, the present data has shown for the first time that inhibition of AR with bicalutamide induces autophagy as a pro-survival process.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

We would like to thank Tanatsu Yoshimori for mRFP-GFP-LC3 fragment, Aviva Tolkovsky for adenovirus-expressing GFP-LC3 and Liz Hall for assistance in the preparation of the manuscript. This study is supported by Cancer Research UK.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
bju11409-sup-0001-si.pdf224K

Fig. S1. Western blot analysis for AR and PSA expression in LNCaP cells after 10 μm bicalutamide (bical) treatment.

Fig. S2. Representative frequency distribution of cathepsin D+ lysosomes in pixels, measured from cells maintained in 10 μm bicalutamide for the indicated time points.

Fig. S3. Representative confocal images of cells maintained in bicalutamide as described in supplementary Figure 2 and stained with cathepsin D antibody and DAPI. Scale bar, 20 nm.

Fig. S4. The average size of cathepsin D+ lysosomes in cells maintained in FBS with vehicle or bicalutamide (bical) for 96 h. In all, 15–25 cells per condition per experiment were analysed, n = 3, error bars indicate sem, **P < 0.01.

Table S1 GI50 values for docetaxel treatment.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.