Role of WNT7B-Induced Noncanonical Pathway in Advanced Prostate Cancer

Published OnlineFirst February 5, 2013; DOI: 10.1158/1541-7786.MCR-12-0520

Molecular Cancer Chromatin, Gene, and RNA Regulation Research

Role of WNT7B-induced Noncanonical Pathway in Advanced Prostate Cancer

Dali Zheng1,3, Keith F. Decker1,3, Tianhua Zhou1,3, Jianquan Chen3,4, Zongtai Qi1,3, Kathryn Jacobs1,3, Katherine N. Weilbaecher2,3, Eva Corey5, Fanxin Long3,4, and Li Jia1,3

Abstract Advanced prostate cancer is characterized by incurable castration-resistant progression and osteoblastic bone metastasis. While androgen deprivation therapy remains the primary treatment for advanced prostate cancer, resistance inevitably develops. Importantly, mounting evidence indicates that androgen receptor (AR) signaling continues to play a critical role in the growth of advanced prostate cancer despite androgen deprivation. While the mechanisms of aberrant AR activation in advanced prostate cancer have been extensively studied, the downstream AR target genes involved in the progression of castration resistance are largely unknown. Here, we identify WNT7B as a direct AR target gene highly expressed in castration-resistant prostate cancer (CRPC) cells. Our results show that expression of WNT7B is necessary for the growth of prostate cancer cells and that this effect is enhanced under androgen-deprived conditions. Further analyses reveal that WNT7B promotes androgen-independent growth of CRPC cells likely through the activation of protein kinase C isozymes. Our results also show that prostate cancer- produced WNT7B induces osteoblast differentiation in vitro through a direct cell–cell interaction, and that WNT7B is upregulated in human prostate cancer xenografts that cause an osteoblastic reaction when grown in bone. Taken together, these results suggest that AR-regulated WNT7B signaling is critical for the growth of CRPC and development of the osteoblastic bone response characteristic of advanced prostate cancer. Mol Cancer Res; 11(5); 482–93. 2013 AACR.

Introduction effectiveness against osteoblastic bone disease in prostate Prostate cancer is dependent on androgens for growth. cancer. Antiosteoblastic approaches still lack therapeutic While androgen deprivation therapy for advanced prostate targets and remain restricted to palliative radiotherapy and cancer effectively reduces tumor burden, the disease typically systemic chemotherapy (3). Increased understanding of the recurs as incurable castration-resistant prostate cancer mechanisms underlying castration resistance and osteoblas- (CRPC). The progression of CRPC is often associated with tic bone metastases will provide opportunities to develop metastases that occur primarily in bone (1). While both more effective therapies for advanced prostate cancer. osteolytic (bone lysing) and osteoblastic (bone forming) WNT signaling plays a central role in both developmental and oncogenic processes. There are 19 closely related WNT processes occur during prostate cancer bone metastasis, fi prostate cancer predominantly yields osteoblastic lesions, genes identi ed in humans. Secreted WNT proteins bind to in contrast to the mostly osteolytic lesions observed in other the Frizzled receptors and other coreceptors at the plasma cancers (lung, kidney, and breast). Studies suggest that membrane and initiate canonical or noncanonical intracellular signaling cascades (4). Canonical WNT signaling tumor-induced osteoblastic and osteolytic activity may play b different roles in promoting tumor growth (2). While anti- stabilizes intracellular -catenin by preventing its phosphor- osteolytic therapies have been developed, they show limited ylation-dependent degradation, resulting in transcriptional activation of WNT target genes. Noncanonical WNT sig- Jun 1 2 naling activates intracellular kinases such as c- -NH2 Authors' Affiliations: Center for Pharmacogenomics, Division of Oncol- 2þ ogy, 3Department of Medicine, 4Department of Developmental Biology, kinase (JNK), Ca /calmodulin-dependent protein kinase Washington University School of Medicine, St. Louis, Missouri; and II (CaMKII), and protein kinase C (PKC). Aberrant WNT 5Department of Urology, University of Washington School of Medicine, Seattle, Washington signaling has been linked to a number of cancers. In prostate cancer, nuclear b-catenin was frequently detected in Note: Supplementary data for this article are available at Molecular Cancer b Research Online (http://mcr.aacrjournals.org/). advanced prostate cancer despite the infrequency of -catenin mutations (5%) in tumors (5, 6). Gene expression Corresponding Author: Li Jia, Center for Pharmacogenomics, Depart- fi ment of Medicine, Washington University School of Medicine, 660 S Euclid pro les of prostate cancer often show overexpression of Avenue, Campus Box 8220, St. Louis, MO 63110. Phone: 314-362-6968; WNT ligands or underexpression of secreted WNT inhi- Fax: 314-362-8844; E-mail: [email protected] bitors (7–10). These changes are likely to be involved in doi: 10.1158/1541-7786.MCR-12-0520 enhancement of nuclear b-catenin and prostate cancer 2013 American Association for Cancer Research. development and progression.

482 Mol Cancer Res; 11(5) May 2013

WNT7B in Advanced Prostate Cancer

It is well known that AR signaling is essential during all Materials and Methods phases of prostate cancer development and progression. b Cell lines and materials While -catenin is a coactivator for TCF/LEF transcription LNCaP, C4-2B, 22RV1, HCT116, ST2, and C3H10T1/2 factors in the canonical WNT pathway, it also interacts with cells were maintained in cell culture medium with 10% AR and enhances AR-mediated transcription in the presence – – FBS as previously described (23 25). LAPC4 cells (from of androgen (11 13). Although cooperative gene regulation Dr. Charles Sawyers, University of California, Los Angeles; by AR and TCF/LEF transcription factors has been ref. 26) were maintained in RPMI-1640 with 20% FBS. All reported, these 2 transcription factors often compete for b cell lines were frozen within 2 passages after receipt from -catenin, with both signaling pathways promoting cell the cell bank and used within less than 6 months after resusc- growth (14, 15). The relative importance of AR and itation. All prostate cancer cell lines were authenticated on the TCF/LEF factors in prostate cancer growth depends on the basis of growth, morphology, and RNA-sequencing results. expression levels of each component at any given time, as Recombinant human WNT3A (R&D Systems) and 5a- well as the local androgen levels. In the presence of androgen, dihydrotestosterone (DHT, Sigma-Aldrich) were purchased. activation of WNT/b-catenin signaling in AR-positive prostate cancer cells often acts through AR-dependent mechan- Stable prostate cancer cell lines with WNT7B isms rather than classical TCF/LEF-dependent mecha- knockdown or overexpression nisms (16). For stable WNT7B knockdown, C4-2B cells were infected Emerging evidence suggests that WNT signaling plays an with lentiviruses encoding short hairpin RNA (shRNA) against important role during progression to CRPC. WNT signal- human WNT7B or LacZ control for 24 hours, and selected ing supports the survival and growth of prostate epithelial with puromycin (1.5 mg/mL). Pooled populations were main- cells after androgen deprivation (17, 18). Enhancement of tained in RPMI-1640 containing 10% FBS and puromycin. the WNT/b-catenin pathway modulates AR signaling and WNT7B knockdown efficiency was examined by quantitative augments AR transcriptional activity under androgen- reverse transcription PCR (qRT-PCR). The shRNA lentiviral deprived conditions (19). Interaction between AR and vectors were obtained from the RNAi Core at Washington WNT signaling pathways in CRPC is also significantly University (St. Louis, MO). The shRNA sequences are listed increased in vivo (20). Recent exome sequencing further in Supplementary Table S1. The pLKO.1 vector system was revealed a significant enrichment of mutations in WNT used to produce lentiviral particles. For stable WNT7B over- signaling components in castration-resistant compared with expression, viruses expressing mouse WNT7B and coexpres- paired castration-sensitive tumors (21). While these results sing GFP via an internal ribosome entry site were generated as support the importance of WNT signaling in prostate cancer previously described (25, 27). LNCaP, C4-2B, and LAPC4 progression, the WNT members critical for CRPC growth cells were infected with viruses expressing WNT7B or empty and the mechanisms of WNT regulation and signaling in vector, and more than 90% infection efficiency was achieved as CRPC remain unclear. determined by GFP detection. Given the critical role of WNT signaling in the differentiation and activity of osteoblasts (22), WNT signaling is Western blot analysis likely to contribute to the development of prostate cancer Whole-cell lysates were prepared from prostate cancer cells as osteoblastic bone metastasis. This is supported by evidence indicated, proteins were separated on 4%–15% SDS-PAGE that the WNT antagonist DKK1 strongly inhibits bone gels (Bio-Rad), and transferred onto polyvinylidene difluoride formation in osteoblastic lesions (9). It is unclear, however, membranes. Blotted proteins were detected by the following which WNT ligands induce osteoblastic activity in prostate antibodies: anti-WNT7B (AF3460; 1:1,000, R&D Systems), cancer bone metastasis and why osteoblastic lesions are often anti-AR (ab74272; 1:1,000, Abcam), anti-b-tubulin (sc-80011; observed in prostate cancer given that WNT signaling is 1:4,000, Santa Cruz Technology), anti-phospho-b-catenin activated in other cancers as well. (T41/S45; 9565, 1:1,000, Cell Signaling), anti-b-catenin In the present study, we have established a new connection (9582; 1:1,000, Cell Signaling), anti-phospho-MARCKS between AR and WNT signaling with important implica- (2741; 1:250, Cell Signaling), anti-MARCKS (5607; tions for prostate cancer progression and bone metastasis. 1:1,000, Cell Signaling), anti-lamin B1 (ab16048; 1:1,000, Our data show that WNT7B expression is regulated by AR, Abcam), anti-PKCa (sc-208; 1:2,000, Santa Cruz Technolo- WNT7B remains at a high level in CRPC cells after gy), anti-PKCd (sc-937; 1:1,000, Santa Cruz Technology). androgen deprivation, and that WNT7B activates a nonca- Nuclear proteins were extracted using NE-PER Nuclear and nonical WNT signaling pathway promoting prostate cancer Cytoplasmic Extraction Reagent Kit (Thermo Scientific). growth and survival after androgen deprivation. Further- Membrane proteins were extracted using Mem-PER Eukaryotic more, we show that prostate cancer-produced WNT7B is Membrane Protein Extraction Reagent Kit (Thermo Scientific). associated with osteoblastic responses in the bone. Given that AR is expressed in almost all prostate cancer cells and Chromatin immunoprecipitation remains active even under castrated conditions, AR-regu- Chromatin immunoprecipitation (ChIP) analyses were lated WNT7B signaling is likely one of the critical mechan- conducted as described previously (23). The primers for isms underlying osteoblastic bone metastasis in advanced quantitative real-time PCR (qPCR) are listed in Supplemen- prostate cancer. tary Table S1.

www.aacrjournals.org Mol Cancer Res; 11(5) May 2013 483

Zheng et al.

qRT-PCR R&D systems) for 24 hours. Fireﬂy and Renilla luciferase qRT-PCR was conducted as described previously (23). activities were measured using Dual-Glo Luciferase Assay Values are means SDs of triplicate wells. The primers are System (Promega). The results are represented as Fireﬂy/ listed in Supplementary Table S1. Renilla ratio.

RNA interference Osteoblast differentiation assay A predesigned pool of 4 siRNA duplexes (Thermo Sci- ST2 and C3H10T1/2 cells (1 105 cells/well) trans- entific) or single siRNA duplexes (Qiagen or IDT) against duced to overexpress WNT7B or vector control were seeded WNT7B, AR, b-catenin, PKCa, PKCd, and MARCKS in a 12-well plate in a-minimum essential medium were purchased. A nonspecific pool of 4 siRNA duplexes (a-MEM) and basal medium eagle (BME), respectively, (siNS1) and 2 different nonspecific single siRNA duplexes with 10% FBS in the presence of 50 mg/mL ascorbic acid and (siNS2 and siNS3) were used as controls. All siRNA 50 mmol/L b-glycerophosphate. After 1 week, the expres- sequences are listed in Supplementary Table S1. Cells were sion levels of osteoblast differentiation markers, alkaline transfected with siRNA as indicated at a final concentration phosphatase (ALP), and bone sialoprotein (BSP) were mea- of 15 nmol/L using Lipofectamine RNAiMAX Transfection sured by qRT-PCR. ALP enzymatic activity was also mea- Reagent (Life Technologies) according to the manufacturer's sured as previously described (25, 27). After 3 weeks of instruction. incubation, mineralization was examined using von Kossa staining as previously described (25, 27). Cell proliferation In coculture experiments, ST2 or C3H10T1/2 cells (1 4 3 LNCaP (1 10 cells/well), C4-2B (6 10 cells/well), 105 cells/well) were seeded in a 12-well plate together with 4 or 22RV1 (1 10 cells/well) cells were plated in 96-well prostate cancer cells at 1:1 ratio. Prostate cancer cells were plates and transfected with gene-specific siRNA. Cells were stably overexpressing WNT7B or transfected with WNT7B maintained in phenol red-free RPMI-1640 containing 5% siRNA 1 day before coculture. The cells were grown in charcoal-stripped serum (CSS) with ethanol or 10 nmol/L a-MEM or BME medium with 10% FBS in the presence of DHT for 3 or 5 days. The number of viable cells was 50 mg/mL ascorbic acid and 50 mmol/L b-glyceropho- analyzed using the CCK-8 kit (Dojindo Molecular sphate. ST2 or C3H10T1/2 osteoblast differentiation was Technologies). examined as described above.

Soft agar colony formation assay Microarray data analysis C4-2B cells with stable WNT7B knockdown or over- For gene expression from prostate cancer xenograft expression were plated in 24-well plate (2,000 cells/well). tumors, RNA was extracted from 24 LuCaP prostate cancer Cells were suspended in RPMI-1640 containing 10% FBS xenografts as described previously (28) and amplified and and 0.5% agar, and overlaid onto a solid layer of the same hybridized to Agilent 44K whole human genome expression medium containing 1.0% agar. After 2-week incubation, oligonucleotide microarrays (Agilent Technologies). Probe colonies were counted and photographed at 40 labeling and hybridization were conducted following the magnification. Agilent suggested protocols, and fluorescent array images were collected using the Agilent DNA microarray scanner Apoptosis assay G2565BA. Agilent Feature Extraction Software was used to LNCaP, C4-2B, or 22RV1 cells were grown in phenol grid, extract, and normalize data. Differences in gene expres- red-free RPMI-1640 containing 5% CSS with ethanol or 10 – fi sion were analyzed via Welch test with the Benjamini nmol/L DHT for 3 days after WNT7B or nonspeci c Hochberg correction for multiple testing. WNT7B expres- siRNA transfection. Cell apoptosis was determined by sion data from clinical prostate cancer tumors were obtained caspase-3 and -7 activities using Caspase-Glo 3/7 assay kit from the NCBI GEO database GSE2443 (29) and (Promega) or by DNA fragmentation using terminal deox- GSE3325 (30), and analyzed using limma (31). ynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) assay kit (Promega) according to the manufacturer's instruction. Results WNT7B expression is regulated by the AR in CRPC cells Luciferase assay Using RNA-seq, we examined WNT gene expression in C4-2B (3 104 cells/well) and HCT116 cells (1 104 androgen-dependent LNCaP and LNCaP-derived castra- cells/well) were plated in 48-well plates and grown in RPMI- tion-resistant C4-2B cells in the presence and absence of 1640 containing 5% FBS for 24 hours. Cells were trans- 10 nmol/L DHT (23). Of 19 WNT genes, WNT7B was the fected with TOPFlash or FOPFlash luciferase reporter only WNT member found to be upregulated after DHT plasmids (200 ng/well, Addgene) using Lipofectamine LTX stimulation in both LNCaP and C4-2B cells. RNA-seq Reagent (Life Technologies). pRL-TK Renilla luciferase results for all WNT genes are presented in Supplementary reporter (10 ng/well, Promega) was cotransfected as an Table S2. WNT7B expression results were confirmed by internal control. After plasmid transfection, cells were trea- qRT-PCR and Western blot analysis (Fig. 1A and B). ted with or without recombinant WNT3A (50 ng/mL, Notably, the high level of WNT7B expression in C4-2B

484 Mol Cancer Res; 11(5) May 2013 Molecular Cancer Research

WNT7B in Advanced Prostate Cancer

Figure 1. WNT7B is an AR target gene in prostate cancer cells. A, qRT-PCR results showing expression of WNT7B in LNCaP and C4-2B cells. Cells were grown in RPMI-1640 medium with 5% CSS for 3 days followed by the treatment with ethanol or 10 nmol/L DHT for 16 hours. B, Western blot analysis showing WNT7B protein levels under the same conditions as in (A). C, ChIP-seq results showing 2 AR-binding sites at the WNT7B locus in LNCaP and C4-2B cells. D, ChIP- qPCR analyses were conducted to confirm the ChIP-seq results. The prostate specific antigen (PSA) enhancer site was used as a positive control. AR occupancy at the negative control region was defined as 1. E, C4-2B cells were grown in RPMI-1640 medium with 5% CSS for 2 days followed by AR siRNA transfection. qRT-PCR analyses of WNT7B mRNA levels were conducted 3 days after AR siRNA transfection. AR protein levels were examined by Western blot analysis in parallel (inset). , P < 0.05; , P < 0.01, determined by a 2-tailed Student t test. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. cells in the absence of androgen is comparable with the deprived conditions. We further examined AR-mediated DHT-induced WNT7B expression levels in LNCaP cells, WNT7B expression in an additional CRPC 22RV1 cell implying that WNT7B may play a role in androgen-inde- line. Notably, the AR was constitutively bound to the pendent growth of C4-2B cells. WNT7B locus (Site 2) in 22RV1 cells (5-fold enrichment ChIP-seq analyses of LNCaP and C4-2B cells revealed one over the control, Supplementary Fig. S1A), which may be strong AR binding site approximately 1.7 kb downstream of due to the predominant expression of an AR splice variant the transcription start site (within the first intron) and one lacking the ligand-binding domain in this cell line (33). weak AR binding site approximately 700 bp downstream of WNT7B expression levels in the absence of androgen were the 30 end of the WNT7B gene (Fig. 1C). Both binding sites similar to those in the presence of androgen (Supplementary were validated by an independent site-specific ChIP-qPCR Fig. S1B) but decreased about 40% after RNA interference analysis (Fig. 1D). Interestingly, we observed AR occupancy against AR (Supplementary Fig. S1C and S1D). Taken (5-fold enrichment over the control) at the weak AR-binding together, these results show that WNT7B is an AR target site (Site 1) in the absence of androgen in C4-2B cells but not gene in both androgen-dependent and CRPC cells, and that in LNCaP. This is consistent with the fact that AR protein is WNT7B remains at a high level in CRPC cells in the absence more active, stable, and constitutively nuclear in CRPC cells of androgen due to distinct AR-binding events. (32) and may explain the higher basal expression of WNT7B in C4-2B cells under androgen-deprived conditions. Fur- WNT7B is required for the growth of both androgen- thermore, the basal expression level of WNT7B in C4-2B dependent and castration-resistant prostate cancer cells cells was decreased more than 70% after knockdown of AR To investigate the potential oncogenic role of WNT7B in using 2 different AR siRNA (Fig. 1E), indicating that prostate cancer cells, we examined LNCaP and C4-2B cell WNT7B expression depends on AR even in androgen- proliferation after knockdown of WNT7B. A high

www.aacrjournals.org Mol Cancer Res; 11(5) May 2013 485

Zheng et al.

knockdown efficiency of WNT7B after RNA interference cells in the absence of androgen resulted in increased Thr41/ was confirmed by Western blot analysis (Fig. 2A). Knock- Ser45 phosphorylation of b-catenin and decreased total down of WNT7B significantly inhibited LNCaP and C4-2B protein levels of b-catenin (Fig. 3A). In contrast, overexpres- growth (Fig. 2B). Notably, WNT7B RNA interference sion of WNT7B slightly decreased Thr41/Ser45 phosphor- abolished prostate cancer cell growth and resulted in cell ylation of b-catenin, yet had little effect on total b-catenin death in the absence of androgen. Further analyses revealed protein levels. Importantly, while WNT7B had a modest that prostate cancer cell apoptosis substantially increased 3 effect on total b-catenin levels, no differences in nuclear days after WNT7B siRNA transfection as determined by b-catenin protein levels were detected. Confocal microscopy caspase-3 and -7 activity assays (Fig. 2C) as well as TUNEL confirmed these results; b-catenin staining was present assays (Fig. 2D; Supplementary Fig. S2). In line with the cell predominately at the cell membrane and diffused in the proliferation results, the most prominent apoptosis was cytoplasm, and weak nuclear b-catenin staining was unal- observed in androgen-deprived conditions. To test whether tered after knockdown or overexpression of WNT7B (Sup- overexpression of WNT7B promotes C4-2B proliferation, plementary Fig. S6). We were unable to detect significant we established WNT7B-overexpressing LNCaP and C4-2B TCF/LEF activity in C4-2B cells using TOPFlash reporter stable cell lines (Fig. 2E). Overexpression of WNT7B had assays (Fig. 3B), and knockdown of b-catenin did not inhibit little effect on LNCaP and C4-2B cell proliferation (Fig. 2F). C4-2B cell growth in the absence of androgen (Fig. 3C). To further assess the role of WNT7B in prostate cancer These results indicate that b-catenin/TCF transcriptional progression, we conducted anchorage-independent colony signaling is unlikely to be involved in WNT7B-induced formation assays on soft agar using C4-2B cells. Stable androgen-independent growth of C4-2B cells. knockdown of WNT7B through lentiviral shRNA trans- We have previously reported that WNT7B induces oste- duction (Supplementary Fig. S3) significantly inhibited C4- oblast differentiation through the PKCd-mediated nonca- 2B cell anchorage-independent growth as indicated by both nonical WNT signaling pathway (25). In this study, we colony number and size (Fig. 2G), whereas overexpression of tested whether WNT7B stimulates androgen-independent WNT7B significantly increased C4-2B cell anchorage-inde- C4-2B growth through PKC signaling, as shown in osteo- pendent growth (Fig. 2H), suggestive of a potential role of blasts. We observed that phosphorylation of MARCKS, a WNT7B in prostate cancer progression. In addition, we prototypic substrate of PKC, was significantly reduced after found that WNT7B is also required for CRPC 22RV1 cell knockdown of WNT7B and enhanced after overexpression growth in both the presence and absence of androgen of WNT7B in C4-2B cells in the absence of androgen (Fig. (Supplementary Fig. S4). Taken together, these results 3A). The total protein levels of MARCKS were slightly support the notion that WNT7B is required for the growth decreased after WNT7B knockdown but not increased after of both androgen-dependent prostate cancer cells and CRPC WNT7B overexpression, indicating that alterations of phos- cells under androgen-deprived conditions. phorylated MARCKS were unlikely to be attributable to Previous studies have shown that WNT7B is expressed in changes in total protein levels. In fact, the change in total a subset of high-grade prostate cancer and bone metastasis MARCKS levels may be secondary to altered phosphoryla- specimens but not in normal prostate tissues (34). To further tion, as phosphorylated MARCKS is more protected from investigate whether WNT7B is critical for prostate cancer proteolytic cleavage than the dephosphorylated form (35). progression, we analyzed 2 publicly available prostate cancer Previous studies have shown that PKCa and PKCd are microarray datasets (29, 30). Comparison between 10 abundantly expressed in prostate cancer cells and exhibit untreated androgen-dependent primary tumors versus 10 distinct functional properties (36). Phosphorylation of androgen-independent primary tumors revealed that MARCKS was effectively reduced by knockdown of either WNT7B is significantly upregulated at the primary site of PKCa or PKCd (Fig. 4A and 4B), implicating both PKC patients with metastatic CRPC (Supplementary Fig. S5A). isozymes in regulation of MARCKS phosphorylation. Fur- WNT7B is also expressed at a higher level in hormone- thermore, knockdown of PKCa or PKCd inhibited C4-2B refractory metastatic tumors compared with localized pros- cell growth in the absence of androgen (Fig. 4C and 4D), tate tumors (Supplementary Fig. S5B). These results are suggesting that both PKC isozymes are required for andro- consistent with a role for WNT7B in promoting CRPC gen-independent C4-2B cell proliferation. Interestingly, our tumor growth. RNA-seq and ChIP-seq results suggest that PKCa and PKCd are also direct AR target genes in C4-2B cells WNT7B promotes CRPC growth after androgen (Supplementary Fig. S7 and Supplementary Table S2). As deprivation through a noncanonical PKC pathway a result, AR may bypass WNT7B and upregulate PKCa and We next sought to investigate mechanisms of the PKCd directly in the presence of androgen, leading to WNT7B-induced intracellular cascades responsible for functional compensation. This mechanism may explain why C4-2B cell growth under androgen-deprived conditions. knockdown of WNT7B has less inhibitory effects on andro- b-catenin is an important cofactor in the regulation of gen-dependent prostate cancer growth (Fig. 2B). Finally, canonical WNT signaling as well as AR signaling. Therefore, knockdown of MARCKS also substantially affected C4-2B we first investigated whether WNT7B alters b-catenin levels, cell growth (Fig. 4E), suggesting an important function of phosphorylation, and nuclear translocation. Our results this gene in cell proliferation. Taken together, our results show that knockdown of endogenous WNT7B in C4-2B have suggested a previously undescribed PKC-mediated

486 Mol Cancer Res; 11(5) May 2013 Molecular Cancer Research

WNT7B in Advanced Prostate Cancer

Figure 2. WNT7B is necessary for androgen-dependent and CRPC cell growth. A, WNT7B protein levels were examined by Western blot analysis 3 days after WNT7B siRNA transfection. B, LNCaP and C4-2B cell proliferation was examined using CCK8 assays 3 and 5 days after WNT7B siRNA transfection in the presence or absence of 10 nmol/L DHT. C, Apoptosis assay with LNCaP and C4-2B cells showing caspase-3 and -7 activity 3 days after WNT7B siRNA transfection in the presence or absence of 10 nmol/L DHT. The results are presented as mean SD of 2 independent experiments. D, TUNEL assays were conducted under the same conditions as in (C). A total of 10 fields were photographed. Total number of cells and the number of TUNEL-positive cells in each field were counted and percentage of apoptotic cells was calculated. E, WNT7B-overexpressing stable prostate cancer cell lines were confirmed by Western blot analysis. F, LNCaP and C4-2B cells overexpressing WNT7B or vector control were grown in the presence or absence of 10 nmol/L DHT for 5 days before CCK8 assays. G, C4-2B cells with stable WNT7B knockdown (shWNT7B-1 and -2) or control knockdown (shLacZ) were examined in soft agar colony formation assays. The assays were conducted in RPMI-1640 medium containing 10% FBS. The colony number is presented as mean SD of 3 independent wells. The experiment was conducted for at least 3 independent times. H, WNT7B-overexpressing or vector control C4-2B cells were used in soft agar colony formation assays as in (G). , P < 0.05; , P < 0.01; , P < 0.001; NS, not significant, determined by a 2-tailed Student t test.

www.aacrjournals.org Mol Cancer Res; 11(5) May 2013 487

Zheng et al.

Figure 3. WNT7B promotes androgen-independent C4-2B cell growth in a b-catenin–independent manner. A, Western blot analysis results showing b-catenin, phosphorylated b-catenin (p-b-catenin), MARCKS and phosphorylated MARCKS (p-MARCKS) protein levels in C4-2B cells after knockdown (siNS1 vs. siWNT7B) or overexpression (Vector vs. WNT7B) of WNT7B in the absence of androgen. B, luciferase assays in HCT116 and C4-2B (with or without WNT7B overexpression) cells after transient transfection of TOPFlash and FOPFalsh reporters. Cells were treated with or without recombinant WNT3A. C, C4-2B cell proliferation was examined using CCK8 assays 3 and 5 days after b-catenin siRNA (sib-catenin) transfection in the absence of DHT. Two different sib-catenin were used and b-catenin protein levels were examined by Western blot analysis 3 days after sib-catenin transfection in C4-2B cells (inset).

noncanonical pathway in CRPC by which WNT7B pro- cell differentiation despite overexpression of WNT7B by motes C4-2B cell growth after androgen deprivation. these cells (Supplementary Fig. S9C), suggesting that the lipid-modified WNT7B is likely linked to plasma mem- WNT7B promotes the development of osteoblastic bone brane. We then cocultured C4-2B with ST2 cells in oste- reactions ogenic medium and observed moderate increases in ALP The importance of WNT signaling in bone development activity and mineralization of ST2 cells. Both effects were has been well established (27). Consistent with our previous diminished after WNT7B knockdown in C4-2B cells (Fig. results (25), we observed WNT7B-induced osteoblast dif- 5A). The mRNA levels of osteoblast differentiation markers, ferentiation in mouse bone marrow–derived stromal ST2 ALP and BSP, were decreased more than 50% as determined cells and mouse pluripotent mesenchymal C3H10T1/2 cells by qRT-PCR. Because we used primers specific for mouse (Supplementary Fig. S8A and S8B). Increased osteoblastic genes, the mRNA levels of ALP and BSP reflected the activity after WNT7B overexpression was shown by expression of these 2 genes in ST2 cells. The expression of increases in ALP enzymatic activity and mineralization. It ALP and BSP were barely detectable in C4-2B cells using should be noted that overexpression of WNT7B in C4-2B these primers (Supplementary Fig. S10). In addition, cocul- cells did not increase ALP activity and mineralization of C4- ture of C4-2B cells overexpressing WNT7B with ST2 cells 2B cells (Supplementary Fig. S8C). markedly enhanced ALP activity, mineralization, and ALP We next asked whether prostate cancer-produced and BSP mRNA levels in ST2 cells (Fig. 5B). Similar results WNT7B promotes osteoblast differentiation. Studies have were obtained with 2 additional prostate cancer cell lines. suggested that WNT7B is highly insoluble after posttrans- Overexpression of WNT7B in LNCaP or LAPC4 cells lational modification and that WNT7B activation is likely markedly promoted ST2 osteoblast differentiation in the mediated through a direct cell–cell interaction (37). We coculture system (Supplementary Fig. S11). To examine observed enriched WNT7B protein on C4-2B cell mem- whether these effects are specific for ST2 cells, we conducted brane using immunofluorescence and Western blot analyses similar assays on C3H10T1/2 cells. Coculture of WNT7B- (Supplementary Fig. S9A and S9B). Conditioned medium overexpressing C4-2B cells with C3H10T1/2 cells dramat- from LNCaP and C4-2B cell culture had little effect on ST2 ically enhanced ALP activity, mineralization, and ALP and

488 Mol Cancer Res; 11(5) May 2013 Molecular Cancer Research

WNT7B in Advanced Prostate Cancer

Figure 4. WNT7B promotes androgen-independent C4-2B cell growth through the PKC signaling pathway. A and B, PKCa, PKCd, phosphorylated MARCKS (p-MARCKS), and MARCKS protein levels were examined by Western blot analysis 3 days after PKCa (A) or PKCd (B) RNA interference in C4-2B cells in the absence of androgen. C–E, C4-2B cell proliferation was examined using CCK8 assays 3 and 5 days after transfection with siPKCa (C), siPKCd (D), or siMARCKS (E) in the absence of androgen. MARCKS protein levels were examined by Western blot analysis 3 days after MARCKS RNA interference in C4-2B cells (inset). , P < 0.001, between nonspeciﬁc siRNA and gene-speciﬁc siRNA as determined by a 2-tailed Student t test.

BSP mRNA levels in C3H10T1/2 cells (Fig. 5C). Similar These 4 samples were excluded from further analyses. The effects were observed when C3H10T1/2 cells were cocul- xenograft tumors were grouped on the basis of whether they tured with WNT7B-overexpressing LAPC4 cells (Supple- did or did not elicit an osteoblastic reaction when growing in mentary Fig. S12). Thus, WNT7B produced from andro- the tibia of intact mice. Xenografts that elicit a large volume gen-dependent and -independent prostate cancer cell lines of new bone formation were designated as osteoblastic can induce osteoblast differentiation. tumors, whereas the remaining tumors were classiﬁed as nonosteoblastic. Comparison of the mRNA expression levels WNT7B is upregulated in human prostate cancer tumors of 8 WNT ligands in osteoblastic (n ¼ 5) and nonosteo- that cause osteoblastic lesions in bone metastasis blastic (osteolytic, mixed, and no reaction, n ¼ 15) xeno- To further investigate the involvement of WNT7B in the grafts showed that only WNT7B was overexpressed in osteoblastic reaction elicited by human prostate cancer prostate cancer xenografts that caused an osteoblastic reac- tumor growth in the bone, we queried the levels of WNT7B tion (Fig. 6). No differences were detected for WNTs3, 4, in LuCaP prostate cancer xenografts (Supplementary Table 5A, 5B, 6, 10B, and 11. Notably, the tumor with the highest S3). LuCaP xenografts were established from primary and WNT7B expression level (LuCaP 23.1) induces a robust metastatic human prostate tumors. Twenty-one androgen- bone reaction in the bone [see Fig. 2 in ref. 38]. A LuCaP dependent LuCaP xenografts were grown in intact mice, 23.1-derived CRPC xenograft tumor (LuCaP 23.1V) con- whereas 3 castration-resistant variants (LuCaP 23.1V, tinued to express WNT7B at a very high level and elicit LuCaP 35V, and LuCaP 96V) were grown in castrated osteoblastic lesions (Supplementary Table S3). These results mice. Of 24 xenograft tumors, 4 were AR-negative neuro- indicate that WNT7B may play an important role in the endocrine tumors and expressed very low levels of WNT7B, osteoblastic reaction resulting from growth of prostate can- consistent with the fact that WNT7B is an AR target gene. cer in the bone.

www.aacrjournals.org Mol Cancer Res; 11(5) May 2013 489

Zheng et al.

Figure 5. C4-2B-produced WNT7B promotes osteoblast differentiation. A, C4-2B cells were transfected with WNT7B siRNA (siWNT7B) or nonspeciﬁc siRNA (siNS1). After 24 hours, C4-2B cells were cocultured with ST2 cells in osteogenic medium. To determine osteoblastic activity of ST2 cells, alkaline phosphatase staining was conducted on day 7 and von Kossa staining for calcium deposits was conducted on day 21 (left). Mouse ALP and BSP mRNA levels were examined by qRT-PCR on day 7 (right). B, WNT7B-overexpressing C4-2B cells or vector control cells were cocultured with ST2 cells in osteogenic medium. The same assays were conducted as in (A). C, WNT7B-overexpressing C4-2B cells or vector control cells were cocultured with C3H10T1/2 cells in osteogenic medium. The same assays were conducted as in (A). , P < 0.05; , P < 0.01, determined by a 2-tailed Student t test.

Discussion WNT-induced signaling pathways are cell- and context- Despite androgen deprivation, AR signaling persists in dependent. We have shown that WNT7B promotes andro- CRPC through a variety of mechanisms. While it is believed gen-independent CRPC cell proliferation likely through a that a subset of AR target genes remain transcriptionally PKC-mediated noncanonical WNT pathway. A similar active, major efforts have been made to understand which pathway was observed in WNT7B-induced osteoblast dif- AR target genes are responsible for the growth of CRPC. ferentiation (25). PKC proteins are a family of serine/ Here, we report that WNT7B is an AR-regulated gene that is threonine kinases involved in the transduction of signals persistently upregulated in CRPC cells and promotes CRPC for cell proliferation, differentiation, and apoptosis. These cell proliferation under androgen-deprived conditions. Our isozymes may exhibit overlapping or opposing functions. data also show that WNT7B produced by prostate cancer While PKCa and PKCd generally function as proapoptotic cells induces osteoblastic bone lesions, and that the WNT7B proteins, they can also act as antiapoptotic proteins (39, 40). paracrine signal is mediated through a direct cell–cell inter- Whether pro- or antiapoptotic functions are exhibited action (Fig. 7). depends on cell type, stimulus, and stage of cancer

490 Mol Cancer Res; 11(5) May 2013 Molecular Cancer Research

WNT7B in Advanced Prostate Cancer

Figure 6. WNT7B is overexpressed in prostate cancer xenograft tissues causing osteoblastic lesions. WNT gene expression levels from microarray studies were analyzed between prostate cancer LuCaP xenograft tissues with osteoblastic lesions (n ¼ 5) versus nonosteoblastic lesions (n ¼ 15).

development and progression. The specific role of each PKC adhesion, spreading, motility, and mitogenesis through actin isozyme in different types of cancers is not completely cytoskeletal regulation (42, 43). Upon phosphorylation by understood (36). It is well documented that PKCa and PKC, MARCKS translocates from the cell membrane to the PKCd mediate phorbol ester-induced apoptosis in prostate cytosol, leading to actin rearrangement and enhanced cell cancer cells (41). Our data, however, suggest a novel onco- migration and invasion. While our data show that MARCKS genic function of these 2 PKC isozymes through their role in is required for the growth of CRPC cells, it is conceivable WNT7B-induced proliferation of CRPC cells. MARCKS is that WNT7B-induced phosphorylation of MARCKS may a ubiquitously expressed substrate of PKC involved in cell increase metastatic activity as observed in other cancer cells (44, 45). Prostate cancer cells secrete a number of osteogenic factors, such as endothlin-1 (ET-1), bone morphogenetic proteins (BMP), and WNTs. These factors promote differentiation of osteoblasts that, in turn, release growth factors that facilitate prostate cancer growth in bone. As a result, prostate cancer cells, capable of producing factors that induce bone formation, tend to survive and proliferate in bone. Several lines of evidence indicate that WNT signaling is likely the primary determinant of tumor-induced bone remodeling. It is well known that increasing DKK1 expression may convert prostate cancer cells from an osteoblastic to an osteolytic type. ET-1 may also increase bone formation by suppression of DKK1 (46). WNTs promote BMP expression and modulate osteoblastic activity in prostate cancer bone metastasis (47). The present study further supports the notion that AR-mediated upregulation of WNT signaling might play a major role in osteoblastic bone metastasis in advanced prostate cancer. Interestingly, WNT7B has been identified as an AR target gene in ER-negative breast cancer cells (48), and approximately 10%–15% of breast cancer bone metastases display predominantly osteoblastic lesions (49). Whether these bone lesions are associated with high AR activity and WNT7B expression is unknown. Current antiandrogen therapies for advanced prostate cancer have focused on blocking AR ligand binding and androgen synthesis. While these novel therapies have been a major breakthrough, increases in survival have been measured in months. Our findings suggest a unique opportunity Figure 7. Schematic model of WNT7B signaling in CRPC cells and for development of more effective therapies by targeting the osteoblasts. AR/WNT7B/PKC signaling cascade, which would inhibit

www.aacrjournals.org Mol Cancer Res; 11(5) May 2013 491

Zheng et al.

prostate cancer proliferation and prevent the formation of a Acknowledgments metastatic reservoir in the bone. The authors thank Drs. David R. Piwnica-Worms, John R. Edwards, Scot Matkovich (Washington University in St. Louis), and Robert L. Vessella (University of Washington, Seattle) for valuable discussions. Disclosure of Potential Conﬂicts of Interest No potential conﬂicts of interest were disclosed. Grant Support Authors' Contributions This work was supported by the Concern Foundation, the Siteman Cancer Center Conception and design: D. Zheng, T. Zhou, L. Jia Developmental Research Award in Prostate Cancer Research, and the BRIGHT Development of methodology: D. Zheng, K.N. Weilbaecher, L. Jia Institute at Washington University School of Medicine/P50 Molecular Imaging Acquisition of data (provided animals, acquired and managed patients, provided Center Pilot Research Fund. facilities, etc.): D. Zheng, J. Chen, Z. Qi, E. Corey, L. Jia The costs of publication of this article were defrayed in part by the payment Analysis and interpretation of data (e.g., statistical analysis, biostatistics, compu- of page charges. This article must therefore be hereby marked advertisement tational analysis): D. Zheng, K.F. Decker, T. Zhou, E. Corey, L. Jia in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Writing, review, and/or revision of the manuscript: D. Zheng, K.F. Decker, T. Zhou, Z. Qi, K.N. Weilbaecher, E. Corey, F. Long, L. Jia Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): T. Zhou, J. Chen, L. Jia Received August 31, 2012; revised January 16, 2013; accepted January 16, 2013; Study supervision: L. Jia published OnlineFirst February 5, 2013.

References 1. Logothetis CJ, Lin SH. Osteoblasts in prostate cancer metastasis to androgen receptor in prostate cancer cell lines. Int J Oncol 2005; bone. Nat Rev Cancer 2005;5:21–8. 26:1033–40. 2. Ortiz A, Lin SH. Osteolytic and osteoblastic bone metastases: two 17. Placencio VR, Sharif-Afshar AR, Li X, Huang H, Uwamariya C, Neilson extremes of the same spectrum? Recent Results Cancer Res 2012; EG, et al. Stromal transforming growth factor-beta signaling mediates 192:225–33. prostatic response to androgen ablation by paracrine Wnt activity. 3. Sturge J, Caley MP, Waxman J. Bone metastasis in prostate cancer: Cancer Res 2008;68:4709–18. emerging therapeutic strategies. Nat Rev Clin Oncol 2011;8:357–68. 18. Yu X, Wang Y, Jiang M, Bierie B, Roy-Burman P, Shen MM, et al. 4. Angers S, Moon RT. Proximal events in Wnt signal transduction. Nat Activation of beta-Catenin in mouse prostate causes HGPIN and Rev Mol Cell Biol 2009;10:468–77. continuous prostate growth after castration. Prostate 2009;69: 5. Voeller HJ, Truica CI, Gelmann EP. Beta-catenin mutations in human 249–62. prostate cancer. Cancer Res 1998;58:2520–3. 19. Verras M, Brown J, Li X, Nusse R, Sun Z. Wnt3a growth factor induces 6. Saha B, Arase A, Imam SS, Tsao-Wei D, Naritoku WY, Groshen S, et al. androgen receptor-mediated transcription and enhances cell growth Overexpression of E-cadherin and beta-catenin proteins in metastatic in human prostate cancer cells. Cancer Res 2004;64:8860–6. prostate cancer cells in bone. Prostate 2008;68:78–84. 20. Wang G, Wang J, Sadar MD. Crosstalk between the androgen receptor 7. Thiele S, Rauner M, Goettsch C, Rachner TD, Benad P, Fuessel S, et al. and beta-catenin in castrate-resistant prostate cancer. Cancer Res Expression profile of WNT molecules in prostate cancer and its 2008;68:9918–27. regulation by aminobisphosphonates. J Cell Biochem 2011;112: 21. Kumar A, White TA, MacKenzie AP, Clegg N, Lee C, Dumpit RF, et al. 1593–600. Exome sequencing identifies a spectrum of mutation frequencies in 8. Wang Q, Symes AJ, Kane CA, Freeman A, Nariculam J, Munson P, advanced and lethal prostate cancers. Proc Natl Acad Sci U S A et al. A novel role for Wnt/Ca2 þsignaling in actin cytoskeleton 2011;108:17087–92. remodeling and cell motility in prostate cancer. PLoS One 2010;5: 22. Long F. Building strong bones: molecular regulation of the osteoblast e10456. lineage. Nat Rev Mol Cell Biol 2012;13:27–38. 9. Hall CL, Bafico A, Dai J, Aaronson SA, Keller ET. Prostate cancer cells 23. Decker KF, Zheng D, He Y, Bowman T, Edwards JR, Jia L. Persistent promote osteoblastic bone metastases through Wnts. Cancer Res androgen receptor-mediated transcription in castration-resistant 2005;65:7554–60. prostate cancer under androgen-deprived conditions. Nucleic Acids 10. Wissmann C, Wild PJ, Kaiser S, Roepcke S, Stoehr R, Woenckhaus M, Res 2012;40:10765–79. et al. WIF1, a component of the Wnt pathway, is down-regulated in 24. Jia L, Landan G, Pomerantz M, Jaschek R, Herman P, Reich D, et al. prostate, breast, lung, and bladder cancer. J Pathol 2003;201:204–12. Functional enhancers at the gene-poor 8q24 cancer-linked locus. 11. Masiello D, Chen SY, Xu Y, Verhoeven MC, Choi E, Hollenberg AN, PLoS Genet 2009;5:e1000597. et al. Recruitment of beta-catenin by wild-type or mutant androgen 25. Tu X, Joeng KS, Nakayama KI, Nakayama K, Rajagopal J, Carroll TJ, receptors correlates with ligand-stimulated growth of prostate cancer et al. Noncanonical Wnt signaling through G protein-linked PKCdelta cells. Mol Endocrinol 2004;18:2388–401. activation promotes bone formation. Dev Cell 2007;12:113–27. 12. Mulholland DJ, Cheng H, Reid K, Rennie PS, Nelson CC. The androgen 26. Klein KA, Reiter RE, Redula J, Moradi H, Zhu XL, Brothman AR, et al. receptor can promote beta-catenin nuclear translocation indepen- Progression of metastatic human prostate cancer to androgen inde- dently of adenomatous polyposis coli. J Biol Chem 2002;277: pendence in immunodeficient SCID mice. Nat Med 1997;3:402–8. 17933–43. 27. Hu H, Hilton MJ, Tu X, Yu K, Ornitz DM, Long F. Sequential roles of 13. Yang F, Li X, Sharma M, Sasaki CY, Longo DL, Lim B, et al. Linking Hedgehog and Wnt signaling in osteoblast development. Develop- beta-catenin to androgen-signaling pathway. J Biol Chem 2002;277: ment 2005;132:49–60. 11336–44. 28. Sun S, Sprenger CC, Vessella RL, Haugk K, Soriano K, Mostaghel EA, 14. Mulholland DJ, Read JT, Rennie PS, Cox ME, Nelson CC. Functional et al. Castration resistance in human prostate cancer is conferred by a localization and competition between the androgen receptor and T-cell frequently occurring androgen receptor splice variant. J Clin Invest factor for nuclear beta-catenin: a means for inhibition of the Tcf 2010;120:2715–30. signaling axis. Oncogene 2003;22:5602–13. 29. Best CJ, Gillespie JW, Yi Y, Chandramouli GV, Perlmutter MA, Gath- 15. Amir AL, Barua M, McKnight NC, Cheng S, Yuan X, Balk SP. A direct right Y, et al. Molecular alterations in primary prostate cancer after beta-catenin-independent interaction between androgen receptor and androgen ablation therapy. Clin Cancer Res 2005;11:6823–34. T cell factor 4. J Biol Chem 2003;278:30828–34. 30. Varambally S, Yu J, Laxman B, Rhodes DR, Mehra R, Tomlins SA, et al. 16. Cronauer MV, Schulz WA, Ackermann R, Burchardt M. Effects of WNT/ Integrative genomic and proteomic analysis of prostate cancer reveals beta-catenin pathway activation on signaling through T-cell factor and signatures of metastatic progression. Cancer Cell 2005;8:393–406.

492 Mol Cancer Res; 11(5) May 2013 Molecular Cancer Research

WNT7B in Advanced Prostate Cancer

31. Smyth GK. Linear models and empirical bayes methods for assessing 40. Lahn M, Sundell K, Gleave M, Ladan F, Su C, Li S, et al. Protein kinase differential expression in microarray experiments. Stat Appl Genet Mol C-alpha in prostate cancer. BJU Int 2004;93:1076–81. Biol 2004;3:Article3. 41. Gavrielides MV, Gonzalez-Guerrico AM, Riobo NA, Kazanietz MG. 32. Gregory CW, Johnson RT Jr, Mohler JL, French FS, Wilson EM. Androgens regulate protein kinase Cdelta transcription and modulate Androgen receptor stabilization in recurrent prostate cancer is asso- its apoptotic function in prostate cancer cells. Cancer Res 2006; ciated with hypersensitivity to low androgen. Cancer Res 2001;61: 66:11792–801. 2892–8. 42. Disatnik MH, Boutet SC, Pacio W, Chan AY, Ross LB, Lee CH, et al. The 33. Chan SC, Li Y, Dehm SM. Androgen receptor splice variants activate bi-directional translocation of MARCKS between membrane and androgen receptor target genes and support aberrant prostate cancer cytosol regulates integrin-mediated muscle cell spreading. J Cell Sci cell growth independent of canonical androgen receptor nuclear 2004;117:4469–79. localization signal. J Biol Chem 2012;287:19736–49. 43. Rombouts K, Mello T, Liotta F, Galli A, Caligiuri A, Annunziato F, et al. 34. Li ZG, Yang J, Vazquez ES, Rose D, Vakar-Lopez F, Mathew P, et al. MARCKS actin binding capacity mediates actin ﬁlament assembly Low-density lipoprotein receptor-related protein 5 (LRP5) mediates during mitosis in human hepatic stellate cells. Am J Physiol Cell Physiol the prostate cancer-induced formation of new bone. Oncogene 2012;303:C357–67. 2008;27:596–603. 44. Chen X, Rotenberg SA. PhosphoMARCKS drives motility of mouse 35. Spizz G, Blackshear PJ. Protein kinase C-mediated phosphorylation of melanoma cells. Cell Signal 2010;22:1097–103. the myristoylated alanine-rich C-kinase substrate protects it from 45. Techasen A, Loilome W, Namwat N, Takahashi E, Sugihara E, Puapairoj speciﬁc proteolytic cleavage. J Biol Chem 1996;271:553–62. A, et al. Myristoylated alanine-rich C kinase substrate phosphorylation 36. Caino MC, von Burstin VA, Lopez-Haber C, Kazanietz MG. Differential promotes cholangiocarcinoma cell migration and metastasis via the regulation of gene expression by protein kinase C isozymes as deter- protein kinase C-dependent pathway. Cancer Sci 2010;101:658–65. mined by genome-wide expression analysis. J Biol Chem 2011;286: 46. Clines GA, Mohammad KS, Bao Y, Stephens OW, Suva LJ, Shaugh- 11254–64. nessy JD Jr, et al. Dickkopf homolog 1 mediates endothelin-1-stim- 37. Lobov IB, Rao S, Carroll TJ, Vallance JE, Ito M, Ondr JK, et al. WNT7b ulated new bone formation. Mol Endocrinol 2007;21:486–98. mediates macrophage-induced programmed cell death in patterning 47. Dai J, Hall CL, Escara-Wilke J, Mizokami A, Keller JM, Keller ET. of the vasculature. Nature 2005;437:417–21. Prostate cancer induces bone metastasis through Wnt-induced bone 38. Corey E, Quinn JE, Bladou F, Brown LG, Roudier MP, Brown JM, et al. morphogenetic protein-dependent and independent mechanisms. Establishment and characterization of osseous prostate cancer mod- Cancer Res 2008;68:5785–94. els: intra-tibial injection of human prostate cancer cells. Prostate 48. Ni M, Chen Y, Lim E, Wimberly H, Bailey ST, Imai Y, et al. Targeting 2002;52:20–33. androgen receptor in estrogen receptor-negative breast cancer. Can- 39. Basu A, Pal D. Two faces of protein kinase Cdelta: the contrasting cer Cell 2011;20:119–31. roles of PKCdelta in cell survival and cell death. Sci World J 2010;10: 49. Roodman GD. Mechanisms of bone metastasis. N Engl J Med 2272–84. 2004;350:1655–64.

www.aacrjournals.org Mol Cancer Res; 11(5) May 2013 493

Role of WNT7B-induced Noncanonical Pathway in Advanced Prostate Cancer

Dali Zheng, Keith F. Decker, Tianhua Zhou, et al.

Mol Cancer Res 2013;11:482-493. Published OnlineFirst February 5, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-12-0520

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2013/02/05/1541-7786.MCR-12-0520.DC1

Cited articles This article cites 48 articles, 18 of which you can access for free at: http://mcr.aacrjournals.org/content/11/5/482.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/11/5/482.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/11/5/482. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.