The small molecule WNT/β-catenin inhibitor CWP232291 blocks the growth of castration-resistant prostate cancer by activating the endoplasmic reticulum stress pathway

CWP232291 was obtained from JW Pharmaceutical Corporation (Seoul, Korea). The compound z-valine-alanine-aspartate-fluoromethylketone (ZVAD-FMK) was purchased from R&D Systems (Minneapolis, MN). The following primary antibodies were used: cleaved caspase-3, poly (ADP-ribose) polymerase (PARP), phospho-eIF2a serine 51 [peIF2a (serine-51)], eIF2a, inositol-requiring kinase 1 (IRE1), C/EBP-homologous protein (CHOP) (Cell Signaling Technology, Danvers, MA), AR, survivin, GAPDH, β-actin, bcl-2 (Santa Cruz Biotechnology, Dallas, TX), β-catenin (Merck Millipore, Burlington, MA) and AR-V7 (Precision Antibody, Columbia, MD).

The human prostate cancer cell lines PC3, DU145, VCaP, LNCaP, and 22Rv1 were obtained from the American Type Culture Collection (ATCC; Manassas, VA) and maintained in RPMI 1640 (Invitrogen, Waltham, MA) with 5–10% heat-inactivated fetal bovine serum (FBS), 100 units/mL penicillin, and 100 μg/mL streptomycin in a 5% CO₂ atmosphere at 37 °C. After informed consent, prostate cancer tissues were obtained from patients who underwent palliative transurethral resection of the prostate to relieve bladder outlet obstruction and were subsequently diagnosed with prostate cancer at Asan Medical Center. The study protocol was approved by the Institutional Review Board of Asan Medical Center. Tumor specimens were minced with scissors and digested by incubation in RPMI containing 1 mg/mL type I collagenase (Sigma Aldrich, St. Louis, MO) for 1 h at 37 °C. Cells were washed with medium containing 10% FBS to inactivate collagenase and then with PBS to remove FBS. Next, cells were plated and maintained in Human Prostate Epithelial Cell Growth Medium (Lonza, Portsmouth, NH) in a 5% CO₂ atmosphere at 37 °C. To ensure authentication and consistency throughout the study, only low-passage cells (< passage 8–15) were used in the experiments. Mycoplasma testing was performed using a PCR-based e-mycoplasma test kit (iNtRON Biotechnology, Seongnam, Korea) for all cells.

Cell viability was measured using the CellTiter Glo® cell viability assay (Promega, Madison, WI). Briefly, 3 × 10³ cells were seeded per well in 96-well plates and incubated overnight. After exposure to CWP232291, cells were incubated with 20 μL of CellTiter Glo reagent for 10 min, after which luminescence intensity was measured on a MicroLumatPlus LB luminometer (EG&G Berthold, Bad Wildbad, Germany). All plates had blank wells containing cell-free medium to measure background luminescence. Data represent the percentage of control untreated cells [(treatment value − blank) / (vehicle value − blank)] expressed as the mean ± SD of at least three repetitions. Results were analyzed using the GraphPad Prism® version 6 software. Drug concentrations that reduced response to 50% of control (IC₅₀) were determined for each cell line.

The cell cytotoxicity assay was performed using the Cytotoxicity Detection Kit (Sigma Aldrich, St. Louis, MO) according to the manufacturer’s protocol. Briefly, 3 × 103 cells were seeded per well into 96-well plates and incubated overnight. After exposure to CWP232291, 50 μL of sample medium (from control and concentration dependent treatments) were transferred to a 96-well plate in triplicate wells. Reaction mixture (50 μL) was then added to each sample and incubated for 30 min in the dark at room temperature. The reaction was stopped by adding 50 μL of stop solution and mixing by gentle tapping. The absorbance was measured at 492 nm and 680 nm. The 680n m absorbance value (background) was subtracted from the 492 nm absorbance values before calculation of % cytotoxicity. The maximum lactate dehydrogenase (LDH) activity was determined after treating cell with lysis buffer as described in the manufacturer’s protocol. Cytotoxicity (%) was calculated using the following formula:

Cytotoxicity (%) = (CWP-treated LDH activity-Low LDH activity) / (High LDH activity-Low LDH activity) X 100.

Whole cell lysates were prepared in lysis buffer [150 mmol/L NaCl, 1% Nonidet P-40, 50 mmol/L Tris–HCl (pH 7.4), 50 mmol/L NaF, 5 mmol/L EDTA, 0.1 mmol/L Na₃VO₄, and 0.1% SDS] containing protease inhibitor cocktail (Sigma Aldrich). The cell lysates were microcentrifuged at 13,000×g for 10 min, and the supernatants were stored at − 80 °C. Protein concentration was measured using the Bradford protein assay (Bio-Rad, Hercules, CA). Proteins were separated by electrophoresis and transferred to PVDF membrane. After blocking with 5% bovine serum albumin (BSA) for 1 h at room temperature, the membranes were incubated with primary antibody, then with secondary antibody conjugated with peroxidase. Protein bands were detected using the chemiluminescence detection system (Millipore Corp., Billerica, MA).

Total RNA was extracted from untreated or CWP232291-treated cell lines using Trizol® (Invitrogen) according to the manufacturer’s instructions. The DNase-treated RNA was reverse-transcribed using a cDNA synthesis kit (Toyobo, Osaka, Japan). Quantitative PCR was conducted using the SYBR method (Toyobo). The PCR thermal cycling conditions were as follows: 95 °C for 20 s, followed by 40 cycles of 95 °C for 3 s, and 60 °C for 30 s. The melting curve stage proceeded at 95 °C for 15 s, melting from 60 °C for 1 min to 95 °C for 15 s with a ramp rate of 1%, and 60 °C for 15 s. Melting curve analysis was performed to ensure the specificity of the PCR products. GAPDH was selected for internal reference and loading control. The following primers were used: CHOP forward, 5′-AGAACCAGGAAACGGAAACAGA-3′, reverse, 5′-TCTCCTTCATGCGCTGCTTT-3′; AR forward, 5′-CAGTGGATGGGCTGAAAAAT-3′, reverse 5′- AAGCGTCTTGAGCAGGATGT-3′; PSA forward, 5′-CATCAGGAACAAAAGCGTGA-3′, reverse, 5′-ATATCGTAGAGCGGGTGTGG-3′; UBE2C forward, 5′-AGTGGCTACCCTTACAATGCG-3′, reverse, 5′-TTACCCTGGGTGTCCACGTT-3′; UGT2B17 forward, 5′- ACCAGCCAAACCCTTGCCTAAG-3′, reverse, 5′-GGCTGATGCAATCATGTTGGCAC-3′; TMPSS2 forward, 5′-CAGGAGTGTACGGGAATGTGATGGT-3′, reverse, 5′-GATTAGCCGTCTGCCCTCATTTGT-3′; c-myc forward, 5′- GCTGCTTAGACGCTGGATTT-3′, reverse, 5′- GGCATTCGACTCATCTCAGC-3′; cyclin D1 forward, 5′- ATGTTCGTGGCCTCTAAGATGA-3′, reverse, 5′- CCAGTGGTTACCAGCAGCTC-3′; MMP-7 forward, 5′- TGAGCTACAGTGGGAACAGG-3′, reverse, 5′- ACCACCCCAAAGAAAATTCC-3′; Annexin-2 forward, 5′-ACGCTGGAGTGAAGAGGAAA-3′, reverse, 5′- AAGGCACTGAGACTCCCTCA-3′; Axin-2 forward, 5′- AGTGTGAGGTCCACGGAAAC-3′, reverse, 5′- TGGCTGGTGCAAAGACATAG-3′; and GAPDH forward, 5′-CAATGACCCCTTCATTGACC-3′, reverse, 5′-GACAAGCTTCCCGTTCTCAG-3′.

Cells were fixed with 4% paraformaldehyde in PBS for 20 min and permeabilized with 0.1% Triton X-100 in PBS for 30 min. Non-specific binding sites were blocked with 5% BSA in PBS for 1 h, then cells were incubated overnight at 4 °C with primary antibody against AR followed by secondary antibody conjugated with fluorescent dye (Molecular Probes, Eugene, OR). The samples were mounted in Vectashield medium containing 4′,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories, Burlingame, CA). Confocal images were obtained using a confocal laser microscopy system (Leica Geosystems, Heerbrugg, Switzerland).

The TCF/LEF reporter construct is a mixture of an inducible β-catenin-responsive luciferase construct and a constitutively expressed Renilla element. The β-catenin-responsive luciferase construct encodes the firefly luciferase reporter gene under the control of a minimal (m) CMV promoter and tandem repeats of the TCF/LEF transcriptional response element. For transfection, cells were seeded at 3 × 10⁵ cells/well (PC3, DU145) and 6 × 10⁵ cells/well (LNCaP, 22Rv1) in 6-well plates. After 18–24 h, cells were transfected with 1 μg of TCF/LEF reporter construct (Qiagen GmbH, Hilden, Germany) using 2 μL of Lipofectamine 2000™ (Invitrogen) in 50 μL of OptiMEM® reduced-serum medium, achieving a DNA/Lipofectamine 2000™ ratio of 1:2. After transfection, cells were treated with or without 100 ng/mL WNT3A (R&D Systems) and CWP232291 for 24 h; control cells were untreated. Luciferase assays were performed using the Dual-Glo luciferase assay system (Promega). Each assay was performed in triplicates, and the reporter activity was expressed as mean ± SD.

Cells were seeded in 6-well plates in RPMI 1640 medium containing 10% FBS for 18–24 h. The cells were then exposed to CWP232291 for 72 h, after which apoptosis and necrosis were assessed by flow cytometry using the annexin V-FITC apoptosis detection assay (BD Biosciences, Bedford, MA) in accordance with the manufacturer’s instructions. In this assay, cells positive for annexin V (bottom right quadrant) and those positive for both annexin V and propidium iodide (PI) (top right quadrant) represent the early and late apoptotic populations, respectively, whereas cells positive for PI only (top left quadrant) represent the necrotic population. The apoptosis cell population analysis was carried out using CellQuestPro software (BD Biosciences).

The experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Asan Medical Center (2015–14-182). 22Rv1 cells (5 × 10⁶) were injected subcutaneously into the right dorsal flanks of 6-week-old male BALB/C nude mice (OrientBio, Seoul, Korea). When the tumors reached an average volume of 70–100 mm³, the mice were randomly divided into control and treatment groups. Mice were treated with CWP232291 in 3% dimethyl sulfoxide in phosphate buffered saline, and control groups received an equal volume of the corresponding diluent alone. 22Rv1-bearing mice received intravenous (tail vein) CWP232291 (50 or 100 mg/kg/day) or vehicle alone once per week for 4 weeks (n = 6 per group). The tumor volume was measured three times a week and calculated using the following formula: 0.52 × length × (width)² (length: longest diameter across the tumor; width: corresponding perpendicular diameter). The concentration-time profile of the active metabolite (CWP232204) after a single intravenous injection of CWP232291 into nude mice is shown in Additional file 1: Figure S1, and CWP232291 pharmacokinetic variables are shown in Additional file 1: Table S1.

First, we investigated whether CWP232291 induces apoptosis in prostate cancer cells. We selected four prostate cancer cells: PC3 and DU145 (AR-negative and androgen-independent), LNCaP (AR-expressing and androgen-dependent), and 22Rv1 (AR-expressing and androgen-independent). Flow cytometry analysis of prostate cancer cells doubly labeled with annexin V and PI showed that CWP232291 induced apoptosis (Fig. 1a). In addition, exposure of cells for 72 h to a concentration of CWP232291 equivalent to the IC₅₀ (PC3, 200 nM; DU145, 400 nM; LNCaP, 60 nM; 22Rv1, 70 nM) significantly increased apoptosis compared with the control in PC3, DU145, LNCaP, and 22Rv1 cells. At the same time point, cleaved caspase-3 and cleaved PARP were significantly increased by CWP232291 compared with the control (Fig. 1b). mRNA levels of Axin-2, a negative regulator of canonical WNT signaling, was increased in prostate cancer cells after treatment with IC₅₀ doses of CWP232291 (Fig. 1c). Inhibition of caspase activity with ZVAD-FMK, a cell-permeable pan-caspase inhibitor, abrogated CWP232291-induced apoptosis, demonstrating that CWP232291 kills cells through the caspase pathway (Fig. 1d).

Next, we investigated whether CWP232291 induces endoplasmic reticulum (ER) stress and the unfolded protein response (UPR). ER stress triggers the UPR, which is mediated by three sensors: protein kinase RNA-like ER kinase (PERK), inositol-requiring enzyme 1 (IRE1), and activating transcription factor 6 (ATF6) [22]. PERK activation attenuates global translation initiation via phosphorylation of eukaryotic translation initiation factor 2α (eIF2α), resulting in increased expression of the proapoptotic CHOP [23]. Cells were exposed to CWP232291 for 24 h at the IC₅₀ (PC3, 200 nM; DU145, 400 nM; LNCaP, 60 nM; 22Rv1, 70 nM), after which ER stress-related markers were assessed by western blotting and real-time PCR (Fig. 2). A concentration-dependent increase in eIF2α protein phosphorylation, IRE1 protein, and CHOP protein and mRNA was observed, suggesting that CWP232291 induces ER stress. Taken together, these data indicate that CWP232291 induces ER stress, resulting in PERK activation, upregulation of CHOP, and caspase-3-dependent apoptosis.

The targeting of WNT/β-catenin signaling by CWP232291 was investigated. In the canonical WNT pathway, LEF/TCF family members are the key mediators of β-catenin-dependent transcription [7, 24]. PC3, DU145, LNCaP, and 22Rv1 cells were transfected with a LEF/TCF luciferase reporter construct and exposed to CWP232291 (PC3, 200 nM; DU145, 400 nM; LNCaP, 60 nM; 22Rv1, 70 nM) in the presence or absence of WNT3a (100 ng/mL) (Fig. 3a). CWP232291 significantly decreased luciferase activity compared with the control, demonstrating inhibition of β-catenin and WNT target gene survivin. Indeed, western blotting revealed that CWP232291 markedly suppressed the expression of the β-catenin and survivin in prostate cancer cells in a concentration-dependent manner (Fig. 3b). In addition, by fluorescence microscopy, a decrease in the amount of β-catenin staining was observed in the nucleus of prostate cancer cells exposed to CWP232291 compared with the control. In the absence of CWP232291, β-catenin staining was diffuse and distributed throughout the cytoplasm. Treatment of prostate cancer cells with WNT3a increased the nuclear localization and expression of β-catenin (Additional file 1: Figure S2). β-catenin staining was lower in the nucleus of prostate cancer cells treated with CWP232291 than in the untreated control (Fig. 3c). Next, we investigated whether CWP232291 treatment reduces WNT/β-catenin target gene expression. mRNA levels of c-myc, cyclin D1, MMP-7, and annexin-2 were lower in LNCaP and 22Rv1 cells after treatment with IC₅₀ doses of CWP232291 for 24 h (LNCaP, 60 nM; 22Rv1, 70 nM) than in the untreated control (Fig. 3d).

Given that β-catenin interacts with the AR, it is conceivable that CWP232291 may also affect AR activity. Prostate cancer cell lines were classified according to AR status: PC3 and DU145 (AR-negative), LNCaP (wild-type AR), 22Rv1 (AR splice variant), VCaP (AR overexpression). To investigate the effects of CWP232291 on AR, cell viability was assessed in AR-expressing (LNCaP, 22Rv1, and VCaP) and AR-negative (PC3 and DU145) prostate cancer cells (Fig. 4a). After prostate cancer cells were exposed to 0–10 μM CWP232291 for 72 h, cell viability was measured. CWP232291 reduced the proliferation of all prostate cancer cells. The IC₅₀ values were lower in AR-expressing cells than in AR-negative cells (Fig. 4b; IC₅₀ of 0.097 μM in LNCaP, 0.060 μM in 22Rv1, 0.070 μM in VCaP vs. 0.188 μM in PC3 and 0.418 μM in DU145 cells). The LDH cytotoxic assay was performed to investigate whether CWP232291 has cytotoxic effects in prostate cancer cells (Additional file 1: Figure S3). Released LDH, an indicator of cytotoxicity, increased at concentrations of CWP232291 higher than 0.01 μM. Western blotting showed that CWP232291 decreased the expression of AR and its splice variants in both LNCaP and 22Rv1 cells (Fig. 4c). Similarly, immunofluorescence staining and confocal microscopy revealed that CWP232291 markedly decreased AR expression in both LNCaP and 22Rv1 cells (Fig. 4)d.

Transcriptional regulation of AR by β-catenin through TCF/LEF-binding sites on the AR promoter has been well documented [10]. To investigate whether CWP232291 directly targets AR transcription, AR and its target gene mRNA levels were analyzed in LNCaP and 22Rv1 cells, which were androgen-deprived for 48 h and then treated with vehicle or 1 nM DHT with or without IC₅₀ doses of CWP (LNCaP 100 nM; 22Rv1 60 nM) for 24 h. As shown in Fig. 4e, CWP232291 treatment decreased AR, PSA, UBE2C, and TMPRSS2 mRNA levels, suggesting that CWP232291 may directly target AR transcription. Additionally, we assessed the activity of androgen-response elements using a reporter assay. CWP232291 significantly suppressed androgen-response element luciferase activity in prostate cancer cells treated with DHT (Additional file 1: Figure S4). We performed the chromatin immunoprecipitation assay to confirm the transcriptional regulation of AR by β-catenin. The results revealed that CWP232291 inhibits β-catenin occupancy of the TCF-binding site within the AR promoter (Additional file 1: Figure S5).

The effect of CWP232291 was compared with that of the standard of care and tested in the context of castration resistance. 22Rv1 and VCaP cells, which are AR-positive and androgen-independent, were exposed to CWP232291 or docetaxel. CWP232291 inhibited the growth of both 22Rv1 and VCaP cells in a similar manner to docetaxel (Fig. 5a). When tested in docetaxel-resistant prostate cancer cells, CWP232291 showed similar growth inhibition in docetaxel-resistant DU145 cells as in parental DU145 cells (Fig. 5b). Next, CWP232291 was tested in primary prostate cancer cells derived from four CRPC patients (0–10 μM, 72 h) (Fig. 5c). Three patients (P1, P2, and P3) progressed after docetaxel chemotherapy, and one patient (P4) progressed after failure of enzalutamide. As shown in Fig. 5c, CWP232291 showed antitumor activity against all primary prostate cancer cells. Interestingly, CWP232291 suppressed the growth of primary prostate cancer cells from a patient (P4) who progressed after enzalutamide, whereas docetaxel was not effective in these cells.

The in vivo antitumor activity of CWP232291 was investigated in the 22Rv1 xenograft mouse model (Fig. 6a-e). Mice bearing 22Rv1 tumors were treated with CWP232291 (27 days, 50 mg/kg/day or 100 mg/kg/day). CWP232291 inhibited tumor growth compared with vehicle (Fig. 6a). Tumor growth inhibition reached 52.0% (50 mg/kg/day) and 73.7% (100 mg/kg/day) after 27 days of CWP232291 treatment, while body weight did not change significantly (Fig. 6b). CWP232291 significantly reduced tumor weight (50 mg/kg/day: 245 mg; 100 mg/kg/day: 160 mg; control: 592 mg) after 27 days compared with vehicle (Fig. 6c). Western blotting showed that CWP232291 decreased expression of β-catenin and increased that of cleaved caspase-3 compared with the control, indicating potent growth-inhibiting and proapoptotic effects in vivo in a time- and dose-dependent manner (Fig. 6e).