CCN3 regulates motility of prostate cancer cells through v 3 integrin, ILK, Akt and NF- B pathway

Po-Chun Chen1, Hsu-Chen Cheng1 * and Chih-Hsin Tang2,3 *

1 2

Department of Life Sciences, National Chung Hsing University, Taichung, Taiwan Department of Pharmacology, School of Medicine, China Medical University,

Taichung, Taiwan
3

Graduate Institute of Basic Medical Science, China Medical University, Taichung,

Taiwan

*Corresponding authors Chih-Hsin Tang, PhD Department of Pharmacology, School of Medicine, China Medical University No. 91, Hsueh-Shih Road, Taichung, Taiwan Tel: 886-4-22052121-7726; Fax: 886-4-22053764 E-mail: chtang@mail.cmu.edu.tw Or Hsu-Chen Cheng, PhD Department of Life Sciences, National Chung Hsing University 250 Kuo Kuang Rd., Taichung, Taiwan E-mail: hcheng@dragon.nchu.edu.tw Tel: (886) 4-22840416, Ext. 505; Fax: (886) 4-22874740

Abstract Nephroblastoma overexpressed (NOV or CCN3) is a secreted, matrix-associated protein that belongs to the CCN gene family and is involved in many cellular functions, including growth, differentiation, and adhesion. The effect of CCN3 on human prostate cancer cells, however, is unknown. Here, we have shown that CCN3 and intercellular adhesion molecule 1 (ICAM-1) affected the mobility of prostate cancer cells. In addition, expression of CCN3 was positively correlated with both cell migration and ICAM-1 expression in human prostate cancer cells. CCN3 activated a signal transduction pathway that included v 3 integrin, integrin-linked kinase (ILK), Akt, and NF- B. Reagents that inhibit specific components of this pathway each diminished the ability of CCN3 to effect cell migration and ICAM-1 expression. Moreover, CCN3 increased binding of p65 to an NF- B binding element in the ICAM-1 promoter. Finally, knockdown of CCN3 expression markedly inhibited cell migration, tumor growth in bone, and bone metastasis. Taken together, our results indicate that CCN3 enhances the migration of prostate cancer cells by increasing ICAM-1 expression through a signal transduction pathway that involves v 3 integrin, ILK, Akt, and NF- B. CCN3 thus represents a promising new target for treating prostate cancer.

Running title: CCN3 induces prostate cancer metastasis Keyword: CCN3; ICAM-1; prostate cancer; migration; bone

Introduction Prostate cancer is the most commonly diagnosed malignancy in the U.S. and other western countries (1). In early stages of prostate cancer, surgery is the most frequent therapeutic intervention. In advanced disease states, however, more systemic interventions are required to inhibit the growth and spread of secondary metastases. Bone metastasis is a common complication associated with advanced prostate cancer, often causing acute pain and bone fractures. Bone metastasis has prognostic value in prostate cancer, because the extent of disease in the bone significantly affects survival (2). Cancer metastasis is a critical step in tumor progression and the major cause of mortality for patients with cancer. This process is composed of several steps by which cells detach from the primary tumor and form a secondary tumor at a distant site (3). Environmental signals serve as cellular attractors, thereby promoting cancer cell motility. To invade nearby tissues, tumor cells must change their cell-cell adhesion properties, rearrange the extracellular matrix of their environment, suppress anoikis, and reorganize their cytoskeleton (4). Cell adhesion molecules such as integrin, cadherin, and immunoglobulin superfamilies have been studied extensively in the context of tumor progression and metastasis (5). Intercellular adhesion molecule 1 (ICAM-1, also called CD54) is an inducible surface glycoprotein that belongs to the immunoglobulin supergene family and mediates cell-cell adhesion (6-8). The surface domain of ICAM-1 is crucial for transendothelial migration of leukocytes from the capillary bed into the surrounding tissue (9), but ICAM-1 may facilitate migration of other cell types as well (10-12). ICAM-1 plays a key role in breast and lung cancer cell invasion (13,14), and knockdown of ICAM-1 expression reduces the invasiveness of breast cancer cells (13). Give the critical role that ICAM-1 plays in tumorigenesis, disruption of ICAM-1 may prove useful for preventing cancer cell metastasis. Nephroblastoma overexpressed (NOV, also known as CCN3) belongs to the CCN gene family, which includes CYR61 (cysteine-rich, angiogenic inducer, 61), CGTF (connective tissue growth factor), and NOV. This gene family encodes secreted
3

proteins that interact with the extracellular matrix and thereby regulate many cellular functions, including cell division, chemotaxis, apoptosis, adhesion, motility, and ion transport (15-17). In recent years, CCN3 has been shown to promote mesenchymal cell differentiation, angiogenesis, and cell adhesion (18-20). Elevated CCN3 expression is also associated with osteosarcoma and poor prognosis in this context (20,21), and with higher rates of metastasis in melanoma (22). Upregulation of CCN3 in prostate cancer cells suggests that it has a role in prostate tumorigenesis (23). Previous studies have shown that CCN3 modulates the migration and invasiveness of human cancer cells (24). The molecular mechanism by which CCN3 affects human prostate cancer cells, however, is essentially unknown. Here, we provide the first evidence that CCN3 increases both migration and ICAM-1 expression through v 3 integrin in cultured human prostate cancer cells.

Furthermore, knockdown of CCN3 decreased cell migration in cultured human prostate cancer cells and bone metastasis in vivo. In addition, a signal transduction pathway including v 3 integrin, integrin-linked kinase (ILK), Akt, and NF- B

mediated the response to CCN3 stimulation in prostate cancer cells. The elucidation of this CCN3-responsive signaling pathway provides valuable clues toward understanding the mechanisms of human prostate cancer metastasis and presents an opportunity to develop more effective clinical therapies in the future.

Materials and methods Materials Protein A/G beads, rabbit polyclonal antibodies specific for NOV were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Mouse monoclonal antibody specific for v 3 integrin was purchased from Chemicon (Temecula, CA). KP-392, Akt inhibitor, PDTC, TPCK were purchased from Calbiochem (San Diego, CA, USA). The recombinant human CCN3 was purchased from R&D Systems (Minneapolis, MN, USA). The Akt (Akt K179A) dominant-negative mutant was gift from Dr. W.M. Fu (National Taiwan University, Taipei, Taiwan),the IKK (KM) and IKK (KM) mutants
4

were

gifts

from

Dr.

H.

Nakano

(Juntendo

University,

Tokyo,

Japan).

pSV- -galactosidase vector and luciferase assay kit were purchased from Promega (Madison, MA). All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA).

Cell culture Human prostate cancer cell lines (PC-3, DU145, and LNCaP) were obtained from the American Type Culture Collection. Cells were maintained at 37 C, 5% CO2, in RPMI-1640 medium supplemented with 20 mM HEPES, 10% heat-inactivated fetal calf serum, 2 mM glutamine, 100 U/ml penicillin, and 100 Qg/ml streptomycin. CCN3 shRNAs plasmids were purchased from National RNAi Core Facility Platform (Taipei, Taiwan). CCN3 shRNAs plasmids were transfected with Lipofectamine 2000 (Invitrogen, CA) and CCN3 shRNA-expressing cells were puromycin selected. Surviving cells were picked and expanded to make clonal cell populations. For monolayer growth curves, 104 cells were plated in 6 well plates and grown for 1 to 6 days. Cells were trypsinized, and cell numbers was counted.

Migration assay The migration assay was performed using Transwell inserts (Costar, NY; 8-mm pore size) in 24-well dishes. Before performing the migration assay, cells were pretreated for 30 min with different concentrations of inhibitors (KP-392, Akti, PDTC or TPCK) or vehicle control (0.1% DMSO). Approximately 1 104 cells in 200 Ql of serum-free medium were placed in the upper chamber, and 300 Ql of the same medium containing different concentrations of CCN3 were placed in the lower chamber. The cells were incubated for 24 h at 37 C in 5% CO2, then fixed in 3.7% formaldehyde for 5 min, and stained with 0.05% crystal violet in PBS for 15 min. Cells on the upper side of the filters were removed with cotton-tipped swabs and the filters were washed with PBS. Cells on the underside of the filters were examined and counted under a microscope. Each experiment was performed in triplicate and repeated at least three
5

times. The number of invading cells in each experiment was corrected for proliferation effects using a cell viability assay (corrected invading cell number = counted invading cell number/percentage of viable cells) (25).

siRNA transfection Small-interfering RNAs (siRNAs) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Cells were transfected with siRNAs (0.4 nmol) using Lipofectamine 2000 (Invitrogen, CA).

Flow cytometric analysis Human prostate cells were grown in 6-well dishes and then washed with PBS and detached using trypsin (Gibco, CA) at 37 C. Cells were fixed for 10 min in PBS containing 3.7% paraformaldehyde, rinsed in PBS, and incubated with mouse anti-human-ICAM-1 (1:100) (BD biosciences, CA) for 1 h at 4 C. Cells were then washed in PBS and incubated with fluorescein isothiocyanate-conjugated goat anti-mouse secondary IgG (1:100; Leinco Technologies, St Louis, MO) for 45 min at 4 rC. After a final rinse, cells were analyzed using a FACSCalibur flow cytometer and CellQuest software (BD Biosciences, CA).

Western blot analysis Cellular lysates were prepared and proteins were resolved by SDS-PAGE. Proteins were transferred to Immobilon polyvinylidene fluoride membranes. The blots were blocked with 4% bovine serum albumin for 1 h at room temperature and probed with rabbit anti-human antibodies against p-GSK3 , GSK3 ILK, p-Akt, Akt, p-IKK, IKK, p-I B , I B , p-p65, p65, ICAM-1, CCN3 or -actin (1:1000) for 1 h at room

temperature (Santa Cruz, CA). After three washes, the blots were incubated with peroxidase-conjugated donkey anti-rabbit secondary antibody (1:1000) for 1 h at room temperature. The blots were visualized with enhanced chemiluminescence using X-OMAT LS film (Eastman Kodak, Rochester, NY).
6

ILK kinase activity assay ILK enzymatic activity was assayed in PC-3 cells lysed in Nonidet P-40 buffer (0.5% sodium deoxycholate, 1% Nonidet P-40, 50 mM HEPES, pH 7.4, 150 mM NaCl) as reported (26). Briefly, ILK was immunoprecipitated from 250 Qg of lysate using an ILK antibody overnight at 4 C. After immunoprecipitation, beads were resuspended in 30 Ql kinase buffer containing 1 Qg recombinant substrate (a GSK3 fusion protein) and 200 QM ATP. The reaction was allowed to proceed for 30 min at 30 C. Phosphorylated substrate was visualized by western blot analysis using an antibody against phospho-GSK3 . Total GSK3 was also detected using the appropriate

antibody. Anti-ILK was used as a loading control.

Quantitative real-time PCR Total RNA was extracted from prostate cancer cells using a TRIzol kit (MDBio, Taipei, Taiwan). Reverse transcription was performed using 1 g total RNA and an oligo(dT) primer. Quantitative real-time PCR (qPCR) was carried out using a TaqMan One-step PCR Master Mix (Applied Biosystems, CA, USA). Total cDNA (100 ng) was added to each 25 l reaction with sequence-specific primers and TaqMan probes. All target gene primers and probes were purchased commercially, including those for (GAPDH as an internal control (Applied Biosystems). qPCR was carried out in triplicate with a StepOnePlus (Applied Biosystems) sequence detection system. The cycling conditions were 10 min at 95 C, followed by 40 cycles of 95 C for 15 s and 60 C for 60 s. To calculate the cycle number at which the transcript was detected (CT), the threshold was set above the non-template control background and within the linear phase of target gene amplification.

Reporter gene assay Human prostate cancer cells were co-transfected with 0.8 Qg B-luciferase plasmid and 0.4 Qg -galactosidase expression vector. PC-3 cells were grown to 80%
7

confluent in 12-well plates and were transfected the following day by Lipofectamine 2000 (LF2000; Invitrogen). DNA and LF2000 were premixed for 20 min and then applied to the cells. After a 24 h transfection, the cells were then incubated with the indicated agents. After further 24 h incubation, the media were removed, and cells were washed once with cold PBS. To prepare lysates, 100 Ql reporter lysis buffer (Promega) was added to each well, and cells were scraped from dishes. The supernatant was collected after centrifugation at 13,000 rpm for 2 min. Aliquots of cell lysates (20 Ql) containing equal amounts of protein (2030 Qg) were placed into wells of an opaque black 96-well microplate. An equal volume of luciferase substrate was added to all samples, and luminescence was measured in a microplate luminometer. The value of luciferase activity was normalized to transfection efficiency monitored by the co-transfected -galactosidase expression vector.

Chromatin immunoprecipitation Chromatin immunoprecipitation (ChIP) was performed as described (27). DNA was immunoprecipitated using an anti-p65 antibody, then extracted, purified, and resuspended in H2 O. Immunoprecipitated DNA was amplified by PCR using the following primers: 5d-AGACCTTAGCGCGGTGTAGA-3d and

5d-GCGACTCGAGGAGACGATGA-3d (28). PCR products were resolved by 1.5% agarose gel electrophoresis and visualized by UV light.

Electrophoretic mobility shift assay Electrophoretic mobility shift assay was performed using an EMSA gel shift kit (Panomics, Redwood City, CA) and an oligonucleotide corresponding to the NF- B binding sequence (5d-AGCTTGGAAATTCCGGA-3d,(29)). Nuclear extracts of PC-3 cells (3 Qg) were incubated with poly [d(I-C)] at room temperature for 5 min. The nuclear extract was incubated with biotin-labeled probes at room temperature for 30 min. After electrophoresis on a 6% polyacrylamide gel, the samples were transferred onto a presoaked Immobilon-Nyt membrane (Millipore, Billerica, MA).
8

The membrane was baked at 80C for 1 h, cross-linked in an oven for 3 min, incubated with blocking buffer followed by streptavidin-horseradish peroxidase conjugate, and then subjected to western blot analysis.

Intratibial injection of PC-3 cells in nude mice CCN3 shRNA-expressing PC-3 cells were cultured with fresh culture medium for 24 h before intratibial injection. Controls included untransfected PC-3 cells and cells transfected with vector control plasmid. Cells were harvested with trypsin-EDTA (Invitrogen, Carlsbad, CA), resuspended in PBS, and kept at 4 C before injection. Male CB17-SCID mice (4 weeks old) were deeply anesthetized using

trichloroacetaldehyde monohydrate (0.4 mg/g body weight; KANTO Chemical Co., Tokyo, Japan). Intratibial injection of cells was performed using polyethylene tubing (Recorder No. 427401, Becton Dickinson) fit around a 30-G needle. This design was used to ensure the depth (1.5 mm) of the needle as it was inserted into the proximal tibia, thereby preventing the cells from spilling out of the injection site. The cell suspension (150 Ql containing 2 105 cells) was slowly injected into the bone marrow cavity of the tibia. A tumor mass was visualized around the proximal tibia 28 days after injection. To make sure of bone osteolysis, radiographs were taken using a soft X-ray generating unit. Animals were deeply anesthetized with trichloroacetaldehyde monohydrate, placed in a prone position on a Kodak Scientific Imaging film (13 18 cm), and exposed to X-rays (45 kV) for 5 s.

Statistical analysis Data are presented as mean standard error of the mean (SEM). Statistical analysis

between two samples was performed using the Students t test. Statistical comparisons of more than two groups were performed using one-way analysis of variance with Bonferronis post-hoc test. In all cases, P < 0.05 was considered significant.

Results CCN3 expression is positively correlated with prostate cancer cell motility, and this effect requires v 3 integrin receptor CCN3 has been shown to increase cell migration and metastasis in a variety of human cancers (20,30). Here, we examined whether CCN3 affects cell migration and metastasis in prostate cancer cells. As a first step, we examined the levels of CCN3 protein and mRNA expression in three human prostate cancer cell lines: PC-3, DU145, and LNCaP. CCN3 protein and mRNA expression were significantly elevated in the highly metastatic PC-3 cell line compared to both DU145 and LNCaP cell lines (Fig. 1A&B). To determine whether CCN3 affects migration of prostate cancer cells, a Transwell assay was used. PC-3 cells had a higher rate of migration compared to the other cell lines (Fig. 1C). Moreover, treatment with CCN3 (10100 ng/ml) dramatically increased migration in all three cell lines (Fig. 1D). Therefore, the expression of CCN3 is positively correlated with motility of prostate cancer cells. Previous studies have shown that CYR61, another member of the CCN family, affects cell migration by binding to integrin receptors on the cell surface (17,31). We hypothesized, therefore, that v 3 integrin signaling pathway may be involved in

CCN3-induced prostate cancer cell migration. Pretreatment of cells with an anti- v 3 integrin monoclonal antibody (5 Qg/ml) for 30 min markedly inhibited CCN3-induced cancer cell migration (Fig. 1E). Similarly, CCN3-induced migration was essentially abolished by pretreatment with 100 nM cyclic RGD peptide (which is known to block v 3 function (32)), but not by pretreatment with cyclic RAD peptide (a negative control). To confirm the positive correlation between integrin v 3 and

CCN3-induced cell migration, we measured integrin v and 3 mRNA expression by qPCR following CCN3 treatment. As expected, CCN3 up-regulated both v and 3 integrin expression in a dose-dependent manner (Fig. 1F). Collectively, these data indicated that CCN3 expression was positively correlated with a higher metastatic potential (i.e., rate of migration) in prostate cancer cells, and that this effect may have been mediated by integrin v 3 receptor.
10

ICAM-1 is involved in CCN3-induced cell migration Recently, it has been reported that ICAM-1 plays a key role in cancer cell migration and tissue invasion (10,33). We examined, whether ICAM-1 was involved in CCN3-induced migration of prostate cancer cells. Western blot and qPCR analyses indicated that ICAM-1 level was elevated in highly metastatic PC-3 cells (Fig. 2A&B). We next asked whether CCN3-induced cell migration correlated with ICAM-1 expression. Results showed that CCN3 increased ICAM-1 protein and mRNA expression in a dose-dependent manner (Fig. 2C&D). Finally, we asked whether loss of ICAM-1 affects CCN3-induced cell migration. PC-3 cells were transfected with ICAM-1 siRNA (100 nM) to knock down ICAM-1 levels. After 24 h, CCN3 was added and cell migration was analyzed using the Transwell assay. Transfection of cells with ICAM-1 siRNA markedly inhibited CCN3-induced cell migration (Fig. 2E). To improve integrin v 3 signaling pathway was involved in ICAM-1 expression. We used anti- v 3 integrin monoclonal antibody (5 Qg/ml) and cyclic RGD peptide (100 nM) to abolish integrin v 3 signaling pathway. In the result, ICAM-1 mRNA expression was inhibited by anti- v 3 integrin antibody and RGD peptide but not RAD peptide (negative control) (Fig. 2F). These data clearly show that CCN3-induced cancer cell migration may occur via upregulation of ICAM-1 and through v 3 integrin receptor.

CCN3 promotes cell migration and ICAM-1 expression via an ILK- and Akt-dependent pathway Integrin-linked kinase (ILK) is a downstream regulator of integrin signaling (34). To determine whether ILK participates in CCN3-dependent effects in prostate cancer cells, we first treated PC-3 cells with CCN3 (30 ng/ml) and measured ILK activity. As shown in Fig. 3A, GSK3 was used as substrate to measure ILK activity. Following CCN3 stimulation, ILK activity increased in a time-dependent manner, reaching a maximum level between 15 and 60 min. Treatment with an ILK-specific inhibitor
11

(KP-392, 1 QM) or transfection with an ILK siRNA (100 nM) decreased CCN3-induced migration (Fig. 3B&C). In addition, CCN3-dependent increases in ICAM-1 expression were reduced by KP-392 treatment or ILK siRNA transfection (Fig. 3D&E). Therefore, ILK is involved in CCN3-mediated migration and ICAM-1 expression. It is well established that Akt is regulated by ILK signaling (26,35). We asked, therefore, whether CCN3 stimulation could activate the Akt signaling pathway. CCN3 treatment increased the level of phosphorylated Akt (Fig. 4A). CCN3-induced migration of prostate cancer cells was greatly reduced by addition of Akt inhibitor (Akti) (Fig. 4B) or expression of a dominant-negative mutant of Akt (Fig. 4C). Akt

inhibition similarly reduced ICAM-1 expression in CCN3-treated cells (Fig. 4D&E). Taken together, our results indicated that the Akt pathway was involved in CCN3-induced cell migration and ICAM-1 upregulation in human prostate cancer cells.

NF- B is involved in CCN3-induced cell migration and ICAM-1 expression NF- B is a primarily cytosolic transcription factor whose activity is often correlated with cancer cell migration and invasion (25,36). Therefore, we used western blot analysis to investigate whether components of the NF- B signaling pathway are involved in the CCN3-dependent effects on prostate cancer cells. Levels of phosphorylated IKK / , I B  and NF- B subunit p65 were higher after 15 min of CCN3 treatment and diminished after ~120 min (Fig. 5A). Moreover, cell migration in response to CCN3 was decreased by pretreatment with the NF- B inhibitor PDTC (10 QM) or the I B protease inhibitor TPCK (1 QM; Fig. 5B). Furthermore, expression of dominant-negative mutant of IKK or IKK markedly inhibited

CCN3-induced cell migration (Fig. 5B). Similarly, each of these treatments reduced ICAM-1 expression (Fig. 5C). It is clear, therefore, that CCN3-mediated effects on cell migration and ICAM-1 expression are mediated by the NF- B signaling pathway.

12

CCN3 increases binding of NF- B to the NF- B element on the ICAM-1 promoter The promoter region of human ICAM-1 contains an NF- B binding site (28); therefore, we asked whether CCN3 stimulation caused NF- B to bind to the ICAM-1 promoter in prostate cancer cells. To this purpose, the electrophoretic mobility shift assay was used. As shown in Fig. 5D, the p65 subunit of NF- B bound to the NF- B element was increased by CCN3 treatment (0-100 ng/ml). We confirmed that NF- B activity was increased by B-luciferase activity assay (Fig. 5E). In addition,

CCN3-mediated binding of p65 to the NF- B element was repressed by KP-392, Akti, PDTC, and TPCK (Fig. 5F). Based on these results, activation of the ILK, Akt, and NF- B signaling pathways was required for the CCN3-induced increase in ICAM-1 expression in human prostate cancer cells.

Knockdown of CCN3 expression inhibits cell migration and osteolytic metastasis in a mouse model of prostate cancer To confirm the role of CCN3 in prostate tumor metastasis, we took advantage of PC-3 cells that stably express CCN3 shRNAs. Following puromycin (1 Qg/ml) selection, individual stable clones (CCN3 sh1, CCN3 sh2 and CCN3 sh3) were collected for analysis. Empty vector plasmid was used as a negative control. Levels of both CCN3 and ICAM-1 were decreased in all three CCN3 shRNA stable clone, and CCN3 sh3 showed most decreased (Fig. 6A). CCN3 knockdown did not alter the rate of cell proliferation (Fig. 6B), but significantly reduced migration (Fig. 6C). The PC-3 cell line is an androgen-independent human prostate cancer line that frequently metastasizes to bone (37,38). To examine the effect of CCN3 on prostate tumor growth in vivo, PC-3 cells (2 105) were locally injected into the tibial bone marrow cavity of 4-week-old severe combined immunodeficiency (SCID) mice. In each mouse, one tibia was injected with negative PC-3 cells and the contralateral tibia was injected with CCN3 sh3 stable clone. Mice were sacrificed after 28 days later. To analyze bone osteolysis in these mice, radiographs were taken using a soft X-ray generating unit. Radiographs which were taken 28 days post-injection revealed
13

osteolytic lesions in tibia injected with negative PC-3 cells. In contrast, osteolytic lesions were dramatically reduced in mice injected with CCN3 sh3 stable clone (Fig. 6D). Tumor weight was recorded after the animals were sacrificed. Quantitative assessment of tumor weight and osteolytic lesions confirmed that knockdown of CCN3 inhibited both tumor growth and formation of osteolytic lesions (Fig. 6E&F).

Discussion In recent years, correlations between CCN3 and tumorigenesis have been identified in a variety of cancers, but the mechanism of CCN3 activity in these contexts remains unclear (20). In both osteosarcoma and Ewing sarcoma, high levels of CCN3 expression indicate a poor prognosis for patients (21,39). In contrast, reduced levels of CCN3 expression are thought to promote cancer progression in childhood adrenocortical tumors and melanomas (22,40). A correlation between prostate cancer and CCN3 has been reported previously (23), but the mechanism by which CCN3 affects prostate cancer remains unclear. In this study, we have demonstrated for the first time that CCN3 (an extracellular matrix-associated protein) increased migration and metastasis of prostate cancer cells by transcriptionally upregulating ICAM-1 expression. CCN3 increased ICAM-1 expression by activating the v 3 integrin, ILK, Akt, and NF- B signaling pathways. Prostate cancer is prevalent in developed countries worldwide. Most prostate tumors remain confined to the prostate gland and adjacent soft tissue and cause little to no harm. However, nearly one in eight cases leads to metastasis, typically to bone (41). The role of androgenic hormones in prostate cancer progression and survival has been reported previously, and is supported by the ability of androgen ablation therapy to cause regression of both primary and metastatic disease (42,43). Here, we found a correlation between CCN3 expression and the rate of prostate cancer cell migration. The highest level of CCN3 expression and the greatest ability to migrate were seen in the most malignant prostate cancer cell line (PC-3, Fig. 1A-C). However, CCN3 was found to support the chemo-migration of both androgen-dependent (LNCaP) and
14

androgen-independent (PC-3 and DU145) prostate cancer cells (Fig. 1D). Therefore, CCN3 induced cancer migration in androgen-dependent and independent prostate cancer cells. Cell adhesion molecules have been implicated in tumor progression and metastasis (44). In solid tumors such as prostate cancer, the level of ICAM-1 expression has been shown to dictate metastatic potential, which in turn determines cancer lethality (45). Here, we found that ICAM-1 expression was upregulated by CCN3 treatment and was directly responsible for the increase in prostate cancer cell migration by CCN3 (Fig. 2). This result is similar to previous studies involving prostate cancer (46). In addition, ICAM-1 levels were reduced following knockdown of CCN3. Therefore, ICAM-1 appeared to mediate CCN3-induced cell migration. Integrins, which link the extracellular matrix to intracellular signaling molecules, regulate a number of cellular processes, including adhesion, signaling, motility, survival, gene expression, growth, and differentiation. Integrins are known to serve as receptors for CCN3 (17). CCN3 has been shown to bind v 3 integrin and increase cell migration (20,47). In this report, v and 3 integrin mRNA expression was

elevated following CCN3 treatment. Moreover, CCN3-induced cell migration was inhibited by pretreatment with a neutralizing antibody against integrin v 3 or with an RGD peptide. This indicates that integrin v 3 receptor played an important role in CCN3-induced migration and ICAM-1 expression in prostate cancer cells. ILK is activated by integrins, growth factors, and chemokines (34). In bone, ILK promotes chondrocyte proliferation, adhesion, and spreading (48), but its role in prostate cancer differentiation is still unclear. In this study, we presented the first evidence to indicate that ILK plays a critical role in prostate cancer progression. Treatment of prostate cancer cells with CCN3 increased the activity of ILK, and CCN3-induced ICAM-1 expression and cell migration were both inhibited by the ILK inhibitor KP-392 and by ILK-specific siRNA. These data suggest that ILK activation is an obligatory step in CCN3-induced prostate cancer cell migration. ILK may regulate these cellular functions by promoting phosphorylation of Akt on Ser473,
15

thereby activating this important downstream pathway (49). Akt has been shown to regulate cancer progression (50,51). We found that CCN3 treatment increased the level of Akt phosphorylation in a time-dependent manner, and that inhibition of Akt reduced the effects of CCN3 on cell migration and ICAM-1 expression (Fig. 4). Taken together, our results provide evidence that CCN3 upregulated ICAM-1 expression and promoted cell migration via the integrin/ILK/Akt signaling pathway in prostate cancer cells. Previous reports have indicated that NF- B is involved in tumor metastasis (52). Here, we demonstrated that the NF- B inhibitors PDTC and TPCK reduced CCN3-induced cell migration, indicating that NF- B acts downstream of CCN3. In its inactivated state, NF- B is held in the cytoplasm by the inhibitory protein I B. Upon stimulation (for example, by tumor necrosis factor , I B proteins become

phosphorylated by the multisubunit IKK complex. This targets I B for ubiquitination and subsequent degradation by the 26S proteasome. Free NF- B can then translocate to the nucleus, where it regulates the transcription of target genes (53). The NF- B pathway has also been linked to CCN3 signaling (54). Treatment of PC-3 cells with CCN3 led to increased levels of phosphorylated IKK, I B , and p65 (Fig. 5A). Using transient transfection of a B-luciferase reporter construct that indicates NF- B

activity, we found that CCN3 increased NF- B activity. Moreover, chromatin immunoprecipitation and electrophoretic mobility shift assays clearly demonstrated that p65 binds to the NF- B element in the ICAM-1 promoter in response to CCN3 stimulation, and that p65 binding to the NF- B element was attenuated by KP-392, Akti, PDTC, and TPCK (Fig. 5D-F). Taken together, these results indicate that CCN3 acted through the v 3 integrin, ILK, Akt, and NF- B pathways to induce ICAM-1 expression in human prostate cancer cells. Finally, to directly determine the effect of CCN3 on prostate cancer progression, we knocked down CCN3 expression in PC-3 cells using shRNA (Fig. 6). CCN3 knockdown significantly reduced ICAM-1 expression and inhibited migration in PC-3 cells. We then injected negative PC-3 cells and PC-3 cells expressing CCN3 shRNA
16

into the contralateral tibia of SCID mice to investigate the in vivo effects of CCN3 on tumor growth and bone resorption (i.e., osteolysis). Tumor growth in the bone was clearly decreased in PC-3 cells that lacked CCN3, as was bone metastasis. These data indicated that CCN3 play an import role in prostate cancer metastasis to bone in vivo. The rate-limiting step in metastasis, and a critical stage in cancer progression, is the acquisition of motility by a tumor cell. Bone metastasis is the major cause of mortality for patients with prostate cancer. Here, we have revealed critical new insights into CCN3 function and its role in prostate cancer progression. CCN3 expression is upregulated in prostate cancer cells, promotes cell migration in vitro, and promotes tumor growth in vivo. Although the mechanisms involved in CCN3-induced bone metastasis are not yet complete, we have provided evidence that CCN3 works via the integrin v 3, ILK, Akt, and NF- B signaling pathways. Our data demonstrate the importance of CCN3 in the growth and metastasis of prostate cancer, and identify CCN3 as a novel therapeutic target for the clinical treatment of prostate cancer.

17

References 1. 2. Bryant, R.J. and Hamdy, F.C. (2008) Screening for prostate cancer: an update. Eur Urol, 53, 37-44. Ernst, D.S., Hanson, J. and Venner, P.M. (1991) Analysis of prognostic factors in men with metastatic prostate cancer. Uro-Oncology Group of Northern Alberta. J Urol, 146, 372-6. Van't Veer, L.J. and Weigelt, B. (2003) Road map to metastasis. Nat Med, 9, 999-1000. Gupta, G.P. and Massague, J. (2006) Cancer metastasis: building a framework. Cell, 127, 679-95. Makrilia, N., Kollias, A., Manolopoulos, L. and Syrigos, K. (2009) Cell adhesion molecules: role and clinical significance in cancer. Cancer Invest, 27, 1023-37. 6. Zimmerman, T. and Blanco, F.J. (2008) Inhibitors targeting the LFA-1/ICAM-1 cell-adhesion interaction: design and mechanism of action. Curr Pharm Des, 14, 2128-39. Lawson, C. and Wolf, S. (2009) ICAM-1 signaling in endothelial cells. Pharmacol Rep, 61, 22-32. van de Stolpe, A. and van der Saag, P.T. (1996) Intercellular adhesion molecule-1. J Mol Med (Berl), 74, 13-33. Duperray, A., Languino, L.R., Plescia, J., McDowall, A., Hogg, N., Craig, A.G., Berendt, A.R. and Altieri, D.C. (1997) Molecular identification of a novel fibrinogen binding site on the first domain of ICAM-1 regulating 10. leukocyte-endothelium bridging. J Biol Chem, 272, 435-41. Yang, S.F., Chen, M.K., Hsieh, Y.S., Chung, T.T., Hsieh, Y.H., Lin, C.W., Su, J.L., Tsai, M.H. and Tang, C.H. (2010) Prostaglandin E2/EP1 signaling pathway enhances intercellular adhesion molecule 1 (ICAM-1) expression and 11. cell motility in oral cancer cells. J Biol Chem, 285, 29808-16. Rauch, B.H., Muschenborn, B., Braun, M., Weber, A.A. and Schror, K. (2007) ICAM-1 and p38 MAPK mediate fibrinogen-induced migration of human vascular smooth muscle cells. Eur J Pharmacol, 577, 54-7. 12. Roche, Y., Pasquier, D., Rambeaud, J.J., Seigneurin, D. and Duperray, A. (2003) Fibrinogen mediates bladder cancer cell migration in an ICAM-1-dependent pathway. Thromb Haemost, 89, 1089-97. Rosette, C., Roth, R.B., Oeth, P., Braun, A., Kammerer, S., Ekblom, J. and Denissenko, M.F. (2005) Role of ICAM1 in invasion of human breast cancer cells. Carcinogenesis, 26, 943-50.
18

3. 4. 5.

7. 8. 9.

13.

14.

Grothey, A., Heistermann, P., Philippou, S. and Voigtmann, R. (1998) Serum levels of soluble intercellular adhesion molecule-1 (ICAM-1, CD54) in patients with non-small-cell lung cancer: correlation with histological

15. 16. 17. 18.

expression of ICAM-1 and tumour stage. Br J Cancer, 77, 801-7. Perbal, B. (2004) CCN proteins: multifunctional signalling regulators. Lancet, 363, 62-4. Planque, N. and Perbal, B. (2003) A structural approach to the role of CCN (CYR61/CTGF/NOV) proteins in tumourigenesis. Cancer Cell Int, 3, 15. Chen, C.C. and Lau, L.F. (2009) Functions and mechanisms of action of CCN matricellular proteins. Int J Biochem Cell Biol, 41, 771-83. Lin, C.G., Leu, S.J., Chen, N., Tebeau, C.M., Lin, S.X., Yeung, C.Y. and Lau, L.F. (2003) CCN3 (NOV) is a novel angiogenic regulator of the CCN protein family. J Biol Chem, 278, 24200-8.

19.

Lin, C.G., Chen, C.C., Leu, S.J., Grzeszkiewicz, T.M. and Lau, L.F. (2005) Integrin-dependent functions of the angiogenic inducer NOV (CCN3): implication in wound healing. J Biol Chem, 280, 8229-37. McCallum, L. and Irvine, A.E. (2009) CCN3--a key regulator of the hematopoietic compartment. Blood Rev, 23, 79-85. Perbal, B., Zuntini, M., Zambelli, D., Serra, M., Sciandra, M., Cantiani, L., Lucarelli, E., Picci, P. and Scotlandi, K. (2008) Prognostic value of CCN3 in osteosarcoma. Clin Cancer Res, 14, 701-9.

20. 21.

22.

Fukunaga-Kalabis, M., Martinez, G., Telson, S.M., Liu, Z.J., Balint, K., Juhasz, I., Elder, D.E., Perbal, B. and Herlyn, M. (2008) Downregulation of CCN3 expression as a potential mechanism for melanoma progression. Oncogene, 27, 2552-60.

23.

Maillard, M., Cadot, B., Ball, R.Y., Sethia, K., Edwards, D.R., Perbal, B. and Tatoud, R. (2001) Differential expression of the ccn3 (nov) proto-oncogene in human prostate cell lines and tissues. Mol Pathol, 54, 275-80. Benini, S., Perbal, B., Zambelli, D., Colombo, M.P., Manara, M.C., Serra, M., Parenza, M., Martinez, V., Picci, P. and Scotlandi, K. (2005) In Ewing's sarcoma CCN3(NOV) inhibits proliferation while promoting migration and

24.

25.

invasion of the same cell type. Oncogene, 24, 4349-61. Tang, C.H., Tan, T.W., Fu, W.M. and Yang, R.S. (2008) Involvement of matrix metalloproteinase-9 in stromal cell-derived factor-1/CXCR4 pathway of lung cancer metastasis. Carcinogenesis, 29, 35-43.

26.

Tseng, W.P., Yang, S.N., Lai, C.H. and Tang, C.H. (2010) Hypoxia induces BMP-2 expression via ILK, Akt, mTOR, and HIF-1 pathways in osteoblasts. J Cell Physiol, 223, 810-8.
19

27.

Chiu, Y.C., Yang, R.S., Hsieh, K.H., Fong, Y.C., Way, T.D., Lee, T.S., Wu, H.C., Fu, W.M. and Tang, C.H. (2007) Stromal cell-derived factor-1 induces matrix metalloprotease-13 expression in human chondrocytes. Mol Pharmacol,

28.

72, 695-703. Lin, C.C., Lee, C.W., Chu, T.H., Cheng, C.Y., Luo, S.F., Hsiao, L.D. and Yang, C.M. (2007) Transactivation of Src, PDGF receptor, and Akt is involved in IL-1beta-induced ICAM-1 expression in A549 cells. J Cell Physiol, 211,

29.

771-80. Lin, Y.C., Shun, C.T., Wu, M.S. and Chen, C.C. (2006) A novel anticancer effect of thalidomide: inhibition of intercellular adhesion molecule-1-mediated cell invasion and metastasis through suppression of nuclear factor-kappaB.

30. 31.

Clin Cancer Res, 12, 7165-73. Perbal, B. (2006) The CCN3 protein and cancer. Adv Exp Med Biol, 587, 23-40. Su, J.L., Chiou, J., Tang, C.H., Zhao, M., Tsai, C.H., Chen, P.S., Chang, Y.W., Chien, M.H., Peng, C.Y., Hsiao, M., Kuo, M.L. and Yen, M.L. (2010) CYR61 regulates BMP-2-dependent osteoblast differentiation through the {alpha}v{beta}3 integrin/integrin-linked kinase/ERK pathway. J Biol Chem, 285, 31325-36.

32.

Pfaff, M., Tangemann, K., Muller, B., Gurrath, M., Muller, G., Kessler, H., Timpl, R. and Engel, J. (1994) Selective recognition of cyclic RGD peptides of NMR defined conformation by alpha IIb beta 3, alpha V beta 3, and alpha 5 beta 1 integrins. J Biol Chem, 269, 20233-8.

33.

Sun, J.J., Zhou, X.D., Liu, Y.K., Tang, Z.Y., Feng, J.X., Zhou, G., Xue, Q. and Chen, J. (1999) Invasion and metastasis of liver cancer: expression of intercellular adhesion molecule 1. J Cancer Res Clin Oncol, 125, 28-34. Dedhar, S., Williams, B. and Hannigan, G. (1999) Integrin-linked kinase (ILK): a regulator of integrin and growth-factor signalling. Trends Cell Biol, 9, 319-23.

34.

35.

Peroukides, S., Bravou, V., Varakis, J., Alexopoulos, A., Kalofonos, H. and Papadaki, H. (2008) ILK overexpression in human hepatocellular carcinoma and liver cirrhosis correlates with activation of Akt. Oncol Rep, 20, 1337-44. Huang, S., Pettaway, C.A., Uehara, H., Bucana, C.D. and Fidler, I.J. (2001) Blockade of NF-kappaB activity in human prostate cancer cells is associated with suppression of angiogenesis, invasion, and metastasis. Oncogene, 20,

36.

37.

4188-97. Feeley, B.T., Krenek, L., Liu, N., Hsu, W.K., Gamradt, S.C., Schwarz, E.M., Huard, J. and Lieberman, J.R. (2006) Overexpression of noggin inhibits
20

BMP-mediated growth of osteolytic prostate cancer lesions. Bone, 38, 154-66. 38. Guise, T.A., Mohammad, K.S., Clines, G., Stebbins, E.G., Wong, D.H., Higgins, L.S., Vessella, R., Corey, E., Padalecki, S., Suva, L. and Chirgwin, J.M. (2006) Basic mechanisms responsible for osteolytic and osteoblastic bone metastases. Clin Cancer Res, 12, 6213s-6216s. 39. Perbal, B., Lazar, N., Zambelli, D., Lopez-Guerrero, J.A., Llombart-Bosch, A., Scotlandi, K. and Picci, P. (2009) Prognostic relevance of CCN3 in Ewing sarcoma. Hum Pathol, 40, 1479-86. Doghman, M., Arhatte, M., Thibout, H., Rodrigues, G., De Moura, J., Grosso, S., West, A.N., Laurent, M., Mas, J.C., Bongain, A., Zambetti, G.P., Figueiredo, B.C., Auberger, P., Martinerie, C. and Lalli, E. (2007) Nephroblastoma overexpressed/cysteine-rich protein 61/connective tissue growth factor/nephroblastoma overexpressed gene-3 (NOV/CCN3), a selective adrenocortical cell proapoptotic factor, is down-regulated in childhood adrenocortical tumors. J Clin Endocrinol Metab, 92, 3253-60. 41. 42. 43. 44. 45. 46. Mundy, G.R. (2002) Metastasis to bone: causes, consequences and therapeutic opportunities. Nat Rev Cancer, 2, 584-93. Heinlein, C.A. and Chang, C. (2004) Androgen receptor in prostate cancer. Endocr Rev, 25, 276-308. Feldman, B.J. and Feldman, D. (2001) The development androgen-independent prostate cancer. Nat Rev Cancer, 1, 34-45. of

40.

Kobayashi, H., Boelte, K.C. and Lin, P.C. (2007) Endothelial cell adhesion molecules and cancer progression. Curr Med Chem, 14, 377-86. Roland, C.L., Harken, A.H., Sarr, M.G. and Barnett, C.C., Jr. (2007) ICAM-1 expression determines malignant potential of cancer. Surgery, 141, 705-7. Conrad, F., Zhu, X., Zhang, X., Chalkley, R.J., Burlingame, A.L., Marks, J.D. and Liu, B. (2009) Human antibodies targeting cell surface antigens overexpressed by the hormone refractory metastatic prostate cancer cells: ICAM-1 is a tumor antigen that mediates prostate cancer cell invasion. J Mol

47.

Med (Berl), 87, 507-14. Tzeng, H.E., Chen, J.C., Tsai, C.H., Kuo, C.C., Hsu, H.C., Hwang, W.L., Fong, Y.C. and Tang, C.H. (2011) CCN3 increases cell motility and MMP-13 expression in human chondrosarcoma through integrin dependent pathway. J

48.

Cell Physiol. Terpstra, L., Prud'homme, J., Arabian, A., Takeda, S., Karsenty, G., Dedhar, S. and St-Arnaud, R. (2003) Reduced chondrocyte proliferation and chondrodysplasia in mice lacking the integrin-linked kinase in chondrocytes. J Cell Biol, 162, 139-48.
21

49.

Persad, S., Attwell, S., Gray, V., Mawji, N., Deng, J.T., Leung, D., Yan, J., Sanghera, J., Walsh, M.P. and Dedhar, S. (2001) Regulation of protein kinase B/Akt-serine 473 phosphorylation by integrin-linked kinase: critical roles for kinase activity and amino acids arginine 211 and serine 343. J Biol Chem, 276, 27462-9.

50.

Huang, C.Y., Fong, Y.C., Lee, C.Y., Chen, M.Y., Tsai, H.C., Hsu, H.C. and Tang, C.H. (2009) CCL5 increases lung cancer migration via PI3K, Akt and NF-kappaB pathways. Biochem Pharmacol, 77, 794-803. Chang, F., Lee, J.T., Navolanic, P.M., Steelman, L.S., Shelton, J.G., Blalock, W.L., Franklin, R.A. and McCubrey, J.A. (2003) Involvement of PI3K/Akt pathway in cell cycle progression, apoptosis, and neoplastic transformation: a

51.

52. 53.

target for cancer chemotherapy. Leukemia, 17, 590-603. Inoue, J., Gohda, J., Akiyama, T. and Semba, K. (2007) NF-kappaB activation in development and progression of cancer. Cancer Sci, 98, 268-74. Chang, M.S., Lee, W.S., Chen, B.C., Sheu, J.R. and Lin, C.H. (2004) YC-1-induced cyclooxygenase-2 expression is mediated by cGMP-dependent activations of Ras, phosphoinositide-3-OH-kinase, Akt, and nuclear factor-kappaB in human pulmonary epithelial cells. Mol Pharmacol, 66, 561-71.

54.

Le Dreau, G., Kular, L., Nicot, A.B., Calmel, C., Melik-Parsadaniantz, S., Kitabgi, P., Laurent, M. and Martinerie, C. (2010) NOV/CCN3 upregulates CCL2 and CXCL1 expression in astrocytes through beta1 and beta5 integrins. Glia, 58, 1510-21.

22

Figure legends Fig. 1 CCN3-induced migration of human prostate cancer cells. (A) Total protein was extracted from PC-3, DU145, and LNCaP cells and subjected to western blot analysis using anti-CCN3. Actin was used as a loading control. (B) Total mRNA was collected from PC-3, DU145, and LNCaP cells, and CCN3 expression was determined by qPCR. (C) In vitro migration activity of PC-3, DU145, and LNCaP cells was measured using the Transwell assay. (D) PC-3, DU145, and LNCaP cells were incubated with CCN3 (0, 10, 30, or 100 ng/ml) for 24 h, and in vitro migration was measured using the Transwell assay. (E) PC-3 cells were pretreated with anti- v 3 mAb (5 Qg/ml), cyclic RGD (100 nM), or cyclic RAD (100 nM) for 30 min followed by stimulation with CCN3 (30 ng/ml) for 24 h (Control = no CCN3 treatment). In vitro migration was measured after 24 h using the Transwell assay. (F) PC-3 cells were incubated with CCN3 for 24 h, and v- and 3-integrin mRNA levels were determined by qPCR. Results are expressed as mean SEM of triplicate samples. In

(B, C), * P < 0.05 compared with PC-3. In (D-F), * P < 0.05 compared with control; # P < 0.05 compared with CCN3 treatment.

Fig. 2 CCN3-directed migration of human prostate cancer cells involves upregulation of ICAM-1. (A) Total protein was extracted from PC-3, DU145, and LNCaP cells, and western blot analysis was used to detect ICAM-1. Actin was used as a loading control. (B) Total mRNA was collected from PC-3, DU145, and LNCaP cells, and ICAM-1 mRNA expression was determined by qPCR. (C) PC-3 cells were incubated with CCN3 (0, 10, 30, or 100 ng/ml) for 24 h and the cell lysates were collected. The level of ICAM-1 in cell lysates was determined by western blot analysis. Anti-actin was used as a loading control. (D) PC-3, DU145, and LNCaP cells were incubated with CCN3 (0, 10, 30, or 100 ng/ml) for 24 h, and ICAM-1 mRNA level was measured by qPCR. (E) PC-3 cells were transfected with ICAM-1 or negative control siRNA (Neg) for 24 h, followed by treatment with CCN3 for 24 h, and ICAM-1 protein level was examined by western blot analysis. Inset: PC-3 cells were
23

transfected with ICAM-1 or control siRNA for 24 h. After 24 h, CCN3 was added and cell migration was analyzed using the Transwell assay. Actin was used as a loading control. (F) PC-3 cells were pretreated with anti- v 3 mAb (5 Qg/ml), cyclic RGD (100 nM), or cyclic RAD (100 nM) for 30 min followed by stimulation with CCN3 (30 ng/ml) for 24 h (N = no CCN3 treatment). Results are expressed as mean SEM

of triplicate samples. In (B, D), * P < 0.05 compared with PC-3. In (E, F), * P < 0.05 compared with control; # P < 0.05 compared with CCN3 treatment.

Fig. 3 ILK signaling affects the CCN3 response in human prostate cancer cells. (A) PC-3 cells were incubated with CCN3 (30 ng/ml) for the indicated times. Cell lysates were prepared and immunoprecipitated with anti-ILK. Immunoprecipitated proteins were subjected to western blot analysis using anti-pGSK3 , anti-GSK3 , and anti-ILK. (B) PC-3 cells were pretreated with the ILK inhibitor KP-392 (3 QM) or vehicle (Control) for 30 min, followed by stimulation with CCN3 (30 ng/ml) for 24 h. In vitro migration was measured using the Transwell assay. (C) PC-3 cells were transfected with ILK siRNA or negative control siRNA (Neg) and stimulated with CCN3 (30 ng/ml). In vitro migration was measured using the Transwell assay after 24 h. (D, E) PC-3 cells were treated as in (B, C) and ICAM-1 mRNA levels were determined by qPCR. In (B-E), results are expressed as mean SEM of triplicate

samples. * P < 0.05 compared to control; # P < 0.05 compared to CCN3 treatment.

Fig. 4 Akt affects the response of human prostate cancer cells to CCN3. (A) PC-3 cells were incubated with CCN3 (30 ng/ml) for the indicated times and p-Akt or Akt levels were determined by western blot analysis. (B) PC-3 cells were pretreated for 30 min with the Akt inhibitor Akti (1 M) or vehicle (Control), followed by stimulation with CCN3 (100 ng/ml) for 24 h. In vitro migration was measured using the Transwell assay. (C) PC-3 cells were transfected with a dominant-negative (DN) Akt mutant for 24 h followed by stimulation for with CCN3 (30 ng/ml) for 24 h. Vector represents a control transfection. In vitro migration was measured using the
24

Transwell assay. (D, E) PC-3 cells were treated as in (B, C) and ICAM-1 mRNA levels were determined by qPCR. Results in (B-E) are expressed as mean SEM of

triplicate wells. * P < 0.05 compared to control; # P < 0.05 compared to CCN3 treatment.

Fig. 5 NF- B mediates the response of human prostate cancer cells to CCN3 stimulation. (A) PC-3 cells were incubated with CCN3 (30 ng/ml) for the indicated times, and phosphorylation of IKK /  I B  and p65 was determined by western blot analysis. (B) PC-3 cells were pretreated for 30 min with PDTC (10 QM), TPCK (1 QM), or vehicle (Control), followed by stimulation with CCN3 (30 ng/ml) for 24 h. PC-3 cells were transfected with dominant-negative (DN) mutant IKK or IKK for 24 h, followed by stimulation with CCN3 (30 ng/ml) for 24 h. Vector represents a control transfection. In vitro migration was measured using the Transwell assay. (C) PC-3 cells were treated as described in (B) and ICAM-1 mRNA levels were determined by qPCR. (D) Cells were stimulated with CCN3 (0-100 ng/ml) for 120 min and an electrophoretic mobility shift assay was performed. (E) PC-3 cells were transfected with an NF- B promoter reporter plasmid for 24 h. Cells were then incubated with CCN3 (0-100 ng/ml) for 24 h and luciferase activity was measured. (F) Cells were pretreated with KP-392 (1 QM), Akti (3 QM), PDTC (10 QM), or TPCK (1 QM) for 30 min, followed by stimulation with CCN3 (30 ng/ml) for 120 min. Chromatin immunoprecipitation was performed with anti-p65. One percent of immunoprecipitated chromatin was assayed to verify equal loading (input). In (B, C, E), results are expressed as mean SEM of triplicate samples. * P < 0.05 compared to

control; # P < 0.05 compared to CCN3 treatment.

Fig. 6 CCN3 knockdown decreases cell migration and osteolytic metastasis in a mouse model. (A) Total protein was collected from untransfected PC-3 cells and PC-3 cells stably expressing shRNAs directed against CCN3. A vector-only control is also shown (Neg). Western blot analysis was used to detect CCN3 and ICAM-1. Actin was
25

used as the loading control. (B) PC-3 cells stably expressing shRNA constructs were seeded as monolayers and counted daily. Cells (104) were reseeded after each count, and the cell numbers were plotted. (C) In vitro migration of PC-3 cells stably expressing shRNA constructs was measured using the Transwell assay. (D) SCID mice were treated by intratibial injection of CCN3 sh3 stable clone in one leg and negative cells in the other leg, and photographed after 28 days. Radiographs taken at 28 days show an osteolytic lesion on the side injected with negative PC-3 cells (left, n=9) but not on the side injected with CCN3- knockdown cells (right, n=9). (E) Quantification of tumor weights following intratibial injection. (F) Quantification of the bone resorption area following intratibial injection. In (C, F), results are expressed as mean SEM of triplicate samples. * P < 0.05 compared to control.

26