Skip Navigation


Carcinogenesis Advance Access originally published online on July 24, 2006
Carcinogenesis 2007 28(2):310-320; doi:10.1093/carcin/bgl134
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/2/310    most recent
bgl134v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Lin, S.-L.
Right arrow Articles by Ying, S.-Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lin, S.-L.
Right arrow Articles by Ying, S.-Y.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Hyaluronan stimulates transformation of androgen-independent prostate cancer

Shi-Lung Lin*, Donald Chang and Shao-Yao Ying*

Department of Cell and Neurobiology, Keck School of Medicine BMT-403, University of Southern California, 1333 San Pablo Street, Los Angeles, CA 90033, USA

*To whom correspondence should be addressed. Tel: +1 323 442 1856; Fax: +1 323 442 3466; Email: lins{at}usc.edu or sying{at}usc.edu


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Summary
 References
 
Interaction between extracellular matrices and cancer cell receptors frequently alters signal transduction pathways, leading to malignant transformation and metastasis. Hyaluronan (HA) is a tumor promoter and enhancer in transformation of androgen-independent (AI) prostate cancer (CaP); however, the signal transduction pathway involved in this mechanism remains unclear. We report here that HA-mediated CD168, a receptor for HA-mediated motility, and its downstream signal molecules, including ROK1, Gab-1, PI3K•p110{alpha} and eIF4E3, accelerate the progression of AI rather than androgen-dependent CaP and enhance AI cell invasion and metastasis in human bone marrow endothelial layers. MicroRNA-based small hairpin RNA-mediated suppression of ROK1 can reverse the malignant role of CD168 signaling in human AI CaP PC3 and DU145 cells. This differential activation of ROK–PI3K signaling in AI CaP cells may provide clues to shed light on some mechanisms of cancer relapse after androgen ablation. These findings reveal a novel signal transduction mechanism for matrix-mediated cancer transformation and metastasis in hormone-refractory CaP.

Abbreviations: AD, androgen-dependent; AI, androgen-independent; AR, androgen receptor; CaP, prostate cancer; ECM, extracellular matrix; EGFR, epidermal growth factor receptor; eIF4E, eukaryotic initiation factor 4E; HA, hyaluronan; HA-R, hyaluronan receptors; hBMEC, human bone marrow endothelial cell; HRCaP, hormone-refractory prostate cancer; IHC, immunohistochemical; LCM, laser-capture machine; miR, microRNA; MLC, myosin light chain; PBS, phosphate-buffered saline; PI3K, phosphatidylinositol-(3,4,5)P3 kinase; RNAi, RNA interference; RNA–PCR, RNA–polymerase cycling reaction; ROK, RhoA-activated protein kinase; shRNA, small hairpin RNA


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Summary
 References
 
Hormone-refractory prostate cancer (HRCaP) is a leading cause of cancer-related death in men. Normal prostatic cell growth is controlled by androgens, and prostate cancer (CaP) occurs when this control is perturbed in aging. The mainstay treatment of androgen-dependent (AD) CaP is to remove androgen stimulation by surgery or hormonal therapy, which is temporarily effective, but eventually results in deadly androgen-independent (AI) CaP transformation (1).

The extracellular matrix (ECM) plays an important role in transformation of AI CaP; however, the signal transduction pathway involved in this mechanism is unclear. Previous studies have revealed that CD168, the receptor for hyaluronan (HA) mediated motility, was highly stimulated in HRCaP cells (2,3). HRCaP cells often lose androgen regulation, leading to neoplastic transformation into AI CaP (3,4). Using microarray analysis of gene profiles in prostatic epithelium obtained from 50 CaP patients, we further observed that major components of the CD168 signaling pathway were consistently overexpressed in metastatic HRCaP cells (5), suggesting the in vivo significance of HA-mediated CD168 signaling in AI CaP transformation.

HA, an ECM polymer composed of the repeating disaccharide unit 2-deoxy, 2-acetamido-D-glucopyranosyl-ß(1,4)-D-glucuronopyranosyl-ß(1,3), is frequently localized in the stroma of solid tumors, facilitating cell migration, tumor invasion and metastasis (6–8). HA is synthesized by stromal fibroblasts in response to paracrine factors produced by tumor cells (9), and such a tumor–stromal interaction is crucial to the development and progression of HRCaP (10). HA binds to hyaluronan receptors (HA-R), for example, CD168 and CD44, and stimulates the RhoA-activated protein kinase (ROK) signal transduction pathway in various cancers (11–13). Three tumorigenic mechanisms have been proposed in CaP (Figure 1). The cytoplasmic domain of HA-R isoforms coupled with RhoA, a Rho•GTPase of the Ras superfamily, forms a complex to activate ROK. Then, the active ROK can either increase myosin light chain (MLC) phosphorylation and actin/myosin-coupled contraction to enhance cancer migration and metastasis as shown in Figure 1, route 1 (12,13), or phosphorylate its linker molecule, Gab-1, and promote the membrane localization of both Gab-1 and HA-R to activate the phosphorylation of phosphatidylinositol-(3,4,5)P3 kinases (PI3K) and to convert phosphatidylinositol (PtdIns)(4,5)P2 to PtdIns(3,4,5)P3 (IP3), subsequently leading to the activation of Akt-TOR-eIF4E signal transduction as proposed in this study (Figure 1, route 2 and right panel). Activation of Akt/TOR/eIF4E signaling causes malignant transformation and drug resistance in advanced CaP (14,15). Alternatively, the PI3K–IP3 signaling cascade promotes M-CSF production, resulting in osteolytic metastasis as shown in Figure 1, route 3 (14).


Figure 1
View larger version (33K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 In vivo significance of HA/CD168-mediated, PI3K-dependent tumorigenecity in CaP. Three mechanisms for HA–CD168 signaling in advanced CaP are proposed, including direct MLC phosphorylation to increase cell migration and invasion (route 1), activation of the PI3K-mediated Akt/TOR/eIF4E signaling pathway that supports cancer cell proliferation and anti-apoptosis (route 2) and augmentation of M-CSF cytokine production to facilitate osteolytic metastasis (route 3). In addition, TGF{alpha}–EGFR interaction may function through PI3K, resulting in a parallel and/or redundant relationship to HA/CD168-mediated cancer progression and metastasis. The PI3K-mediated Akt/TOR/eIF4E signaling pathway is an important regulator in the translation of malignancy-related gene transcripts, such as proto-oncoproteins [c-myc, CDC6, cyclin D1, ornithine decarboxylase (ODC)], angiogenic factors (IGF2, FGF2 and VEGF), and degradative enzymes (MMP-9) (right panel).

 
CD44 and CD168 interact with Ras and RhoA-ROK, respectively, in activation of PI3K–IP3 signaling in many epithelial neoplasms including CaP. Nevertheless, expression of CD44 and its isoforms is not correlated with HA level, Gleason grade, pathological T classification, PSA recurrence, clinical invasion and infiltration of prostatic cancer cells (16,17). Furthermore, CD44v(3,8–10), the major variants that interact with RhoA-ROK, are not expressed in either AI CaP cell lines or in vivo CaP at all stages (18,19). In contrast, CD168 is present in HRCaP and consists of two isoforms; intracellular CD168 mainly exists in adherent cells, whereas membrane CD168 presumably promotes motility of Ras-transformed fibroblasts and acts as an oncogene to cause metastatic transformation of immortalized cells (14,20). Given that previous studies demonstrated that CD168 rather than CD44 was overexpressed in metastatic HRCaP cells (2,3,5) and that CD168 could compensate for CD44 in CD44-knockout mice (21), these findings suggest that HA–CD168 interaction is a predominant pathway for activating ROK and subsequent PI3K–IP3 signaling in HRCaP. Thus, this differential activation of PI3K by HA–CD168 signaling may provide clues for solving the mystery of recurring AI CaP.

We present here a novel signal transduction model for HA/CD168-mediated activation of ROK–PI3K signaling in AI transformation of HRCaP (Figure 1, route 2 and right panel). Differing from the HA–CD44 signaling pathway, in which PI3K signaling is activated via its membrane localization of p110•p85 complex and binding to the active conformation of Ras, Ras•GAP (14), the HA–CD168 complex first activates ROK and then phosphorylates Gab-1. The ROK-mediated Gab-1 phosphorylation and Gab-1•PI3K membrane localization are accompanied by a robust stimulation of PI3K activity, via the {alpha}-form, but not the {delta}-form of the p110 catalytic subunit (12,13). The PI3K activation leading to overexpression of cap-dependent eukaryotic initiation factor 4E (eIF4E) has been observed in HRCaP (4,5) and is related to malignant transformation and anti-apoptosis in many advanced human cancers (15). In this study, we further provide evidence that HA–CD168 interaction triggers the ROK–PI3K–eIF4E signaling cascade, which occurs mainly in AI CaP but not in AD CaP cells, supporting the functional role of HA–CD168 signaling in AI CaP transformation.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Summary
 References
 
Cell culture and reagents
Human CaP cell lines, LNCaP, PC3 and DU145 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and grown as described (5). For HA stimulation, HA (50–400 µg/ml) was added to the cultured cells at ~40% confluency. To produce ROK inhibition, Y-27632 (10–120 µg/ml) was applied to the cell culture at the same time with the HA treatment (200 µg/ml). One day before the treatments, the culture medium was replaced with medium containing 2% fetal bovine serum for 30 h.

Single-cell-type mRNA amplification
Single-cell-type (homologous) mRNA libraries were amplified from laser-capture machine (LCM) dissected prostatic epithelial cells using the RNA–polymerase cycling reaction (RNA–PCR) as described (5,22,23). Human CaP tissue array sections were dewaxed in xylene, prefixed by 100% ethanol and dissected into homogenous cell populations using a microscopic LCM (Leica, Germany). About 50 dissected cells were lysed in 20 µl of ice-cold Cell Lysis II buffer (Ambion, Austin, TX) and incubated at 75°C for 10 min. Next, the crude cell lysate was treated with DNase I (0.04 U/µl) at 37°C for 5 min and then 75°C for 5 min. A quarter of the crude cell lysate was then used in a two-cycle RNA–PCR for mRNA amplification (5,23).

Microarray and northern blot analyses in vivo
Microarray and northern blot analyses of HRCaP-specific genes in vivo were performed as described (5,23). RNA samples amplified from LCM-dissected prostatic epithelium were biotin-labeled and hybridized with U133A,B human genome arrays (Affymetrix, Santa Clara, CA), respectively. The microarrays were read at 7.5 µm with a confocal scanner (Molecular Dynamics) and analyzed with the Affymetrix Microarray Suite version 5.0 and Genetrix (Epicenter, CA) softwares. The samples were normalized using the total average difference between perfectly matched probes and mismatched probes. Differential signals >3-fold were collected for northern blot analysis.

Immunohistochemical (IHC) staining of tissue arrays
Human CaP tissue arrays and IHC staining kits were obtained from Imgenex (San Diego, CA) and used according to the manufacturer's suggestions. For antigen retrieval, the arrays were merged in the target retrieval solution (DakoCytomation, CA) and heated by microwave at the high and low setting for 3 min each. Then the specimens were rinsed in phosphate-buffered saline (PBS) three times and incubated with a primary antibody (diluted in Zeller's solution) overnight in a humidified chamber at 4°C. We detected signal molecules related to the CD168–ROK pathway, using primary antibodies against CD168, ROK1, ROK2, Gab-1, PI3K•p110{alpha}, PI3K•p110{delta}, EGFR, CD44, Ras, Akt1/2, eIF4E and the androgen receptor (AR) (Santa Cruz Biotechnology, Santa Cruz, CA; Upstate LLC, Charlottesville, VA; EMD Biosciences, San Diego, CA), and their phosphorylated forms (EMD Biosciences; Imgenex), respectively, following the manufacturers' suggestions. Biotinylated goat anti-rabbit or horse anti-mouse antibody was used as the secondary antibody (Chemicon, Temecula, CA) and streptavidin-HRP was added as the tertiary antibody. Then the bound antibody was detected with DAB substrates. Positive results were observed under a 100x microscope with whole field scanning and recorded at 100x and 400x magnification (Nikon 80i microscopic quantification system).

Immunoprecipitation and western blot analysis
One to two days after treatment, cells were washed with ice-cold PBS twice and lysed in the CelLytic-M lysis/extraction reagent (Sigma Chemicals). Protein-containing cell lysates were then collected and tested for immunoprecipitation and western blot analyses as described (5,24), except using primary antibodies against signal molecules of the CD168–ROK pathway. ROK1, ROK2, Gab-1, PI3K•p110{delta}, EGFR, CD44, Ras, Akt1/2, eIF4E and AR could be directly detected in western blotting, whereas CD168 and PI3K•p110{alpha} required purification by immunoprecipitation before western blotting.

Flow cytometry
Cells were trypsinized, pelleted and fixed by resuspending in 1 ml of pre-chilled 70% methanol in PBS for 1 h at –20°C. The cells were pelleted and washed once with 1 ml of PBS. The cells were pelleted again and resuspended in 1 ml of 1 mg/ml propidium iodide, 0.5 mg/ml RNase in PBS for 30 min at 37°C. Approximately 15 000 cells were then analyzed on a BD FACSCalibur (San Jose, CA). Cell doublets were excluded by plotting pulse width versus pulse area and gating on the single cells. The collected data were analyzed using the software package Flowjo using the ‘Watson Pragmatic’ algorithm (25).

Invasion assay
Chamber inserts (12 µm pore size, Chemicon) were coated with 200 µg/ml of matrigel alone or supplemented with 200 µg/ml of HA and/or 50 µg/ml Y-27632 (EMD) in phenol red-free-DMEM with 1% L-glutamine and dried overnight under sterile conditions. Cells were harvested, washed and resuspended in phenol red-free-DMEM to give a final cell density of 1 x 106 cells/ml for PC3 and 1 x 105 cells/ml for DU145 cells, respectively. Five hundred microliters of the resulting cell suspension was then dispensed into the top chamber, whereas DMEM-conditioned medium (1.5 ml) was added to the bottom chamber to create a chemotactic gradient. Invasion was measured after overnight incubation at 37°C for 16 h. Top chambers were wiped with cotton wool, and invading cells on the underside of the membrane were fixed in 100% methanol for 10 min, air-dried, stained in cresyl violet for 20 min and gently rinsed in water. When dry, the cresyl violet stain on membranes was eluted using a 100% ethanol/0.2 M NaCitrate (1:1) wash for 20 min and absorbance was read at 570 nm using a Precision Microplate Reader (Molecular Dynamics). The percentage of invading cells was calculated by comparison of absorbance in test samples against absorbance determined on membrane inserts that were not wiped (total cells).

Adhesion assay
Cells were trypsinized, washed in adhesion media [RPMI 1640/0.1% BSA/20 mM HEPES (pH 7.4)] and sterile saline once, and resuspended at 1 x 106 cells/ml in PBS with 10 µM fura-4 acetoxymethyl ester (fluorescent probe, Sigma) for 1 h at 37°C in the dark. Cells were then pelleted, washed in serum-free medium containing 1% (v/v) of probenecid (100 mM) and incubated for 20 min in adhesion media at 37°C in the dark to activate the intracellular fluorescent probe. Cells (3 x 105 cells/ml) were resuspended in adhesion medium and protected from the light until experimentation. Human bone marrow endothelial cells (hBMECs) were seeded at a density of 1 x 105 cells/ml in 96-well plates and washed with adhesion media before assay. The hBMEC cell line was isolated and grown as described previously (26). Cancer cells were added (300 µl cell suspension/well) to confluent endothelial monolayers and incubated for specified times at 37°C with or without treatments of HA (200 µg/ml) and/or Y-27632 (50 µg/ml), respectively. Non-adherent cells were removed using 2 x 250 µl washes of adhesion medium. Finally, plates were read in a fluorescent plate reader (Molecular Dynamics) at 37°C using an excitation wavelength of 485 nm and an emission wavelength of 530 nm.

RNA interference (RNAi) mediated ROK depletion
ROK-depleted PC3 and LNCaP cell lines were generated using cytomegalovirus (CMV) promoter-driven microRNA (miR) based small hairpin RNA (shRNA) expression vectors as described (24). The designed miR-based shRNA molecule was directed against the ROK1 gene nucleotide 1044–1138 region (accession number AF077033 [GenBank] ), which contains a partial Rho-binding domain required for ROK kinase activity. For in-cell transfection, the designed vectors were liposomally encapsulated in FuGENE reagent (Roche, Indianapolis, IA) and applied to cell cultures at 40% confluency. After a 24 h incubation, positively transfected cells were isolated using flow cytometry with anti-rGFP monoclonal antibody (Clontech) as described (24).

Statistical analysis
Results were presented as mean ± SE. Statistical analysis of data was performed by one-way ANOVA. When main effects were significant, the Dunnett's post hoc test was used to identify the groups that differed significantly from the controls. For pairwise comparison between two treatment groups, the two-tailed student t-test was used. For experiments involving more than two treatment groups, ANOVA was performed followed by a post hoc multiple range test. Probability values of P < 0.05 were considered significant. All P-values were determined from two-tailed tests.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Summary
 References
 
Identification of differential gene markers in HRCaP
Prostatic carcinoma often shows a heterogeneous and multifocal incidence with diverse clinical and morphological manifestations. Knowledge of the molecular basis for such heterogeneity is however limited. To accurately distinguish CaP progression, we divided patients' tissue samples into four distinct stages, including (i) normal and benign prostatic hyperplasia (BPH) (stage 1); (ii) localized CaP, in which cancer is confined in situ and shows no signs of elevated PSA and metastasis (stage 2); (iii) metastatic disease, in which cancer is growing outside the prostate and surrounding tissues but still responds well to male hormone therapy (stage 3); and (iv) hormone-refractory disease, in which CaP continues to grow despite androgen/testosterone treatments (stage 4). In addition to these progressive stages, we selected cancerous tissues with corresponding Gleason grades, which were assigned by two clinical pathologists to indicate cancer cell malignancy. For example, the stages 2, 3 and 4 CaP contained cancerous epithelial cells with Gleason scores 5–7, 7–9 and 9–10, respectively. Under this stringent category system, we synchronized both CaP stages and grades for identifying cancer-stage-specific genes in vivo.

Our previous microarray approach has provided an accurate and consistent strategy for analyzing the heterogeneity of CaP in vivo (5,22,23). On the basis of the same combined technology of LCM, RNA–PCR and microarray analysis, we have shown that an average of 93–98% similarity was well maintained between the microarray profiles of RNAs amplified from 10 pg source RNAs and those of 10 µg unamplified source RNAs (Figure 2A; P < 0.05, n = 3). To examine differential gene profiles in HRCaP, we then compared microarray results from stage-3 CaP with those from stage-4 CaP epithelial cells (both Gleason score 9). As shown in Figure 2C, we first subtracted the gene profile of AD CaP cells (red circle) from that of AI CaP cells (blue circle) in each section to rule out individual variation. Subsequently, the subtracted gene profile of the stage-3 CaP cells was further subtracted from that of the stage-4 CaP cells, providing a unique differential gene expression pattern in HRCaP. These microarray results were further confirmed by northern blot analysis using RNA samples amplified from AD and AI CaP cells of the HRCaP patients' sections (Figure 2B).


Figure 2
View larger version (109K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Microarray and tissue array analyses of the CD168–ROK signaling cascade in human CaP tissue sections in vivo. (A) Comparison between source RNAs (10 µg) and RNA–PCR–amplified RNAs from one millionth of the source RNAs using a combined LCM, RNA–PCR and microarray analysis method. A total RNA library isolated from 106 LNCaP cells by RNeasy spin columns (Qiagen, CA) and a purchased reference RNA library (Stratagene, CA) were used as source RNAs, respectively. The correlation coefficiency between these sources exceeded 0.93, indicating the high sensitivity and reliability of this approach. (B) Confirmation of overexpression of CD168/ROK-associated signal genes (red square) by both microarray and northern blot analyses, showing 3- to 10-fold increases in HRCaP (Gleason score 9). No difference was detected in house-keeping GAPDH and ß-actin gene expression (bottom). (C) Samples from LCM-dissected AD (red circle) and AI (blue circle) prostatic epithelial cells in vivo, respectively. (D) In vivo correlation of the CD168–ROK signaling cascade (route 2 of Figure 1) in human CaP tissue arrays, including IHC staining of phosphorylated forms of ROK1 (p-ROK), Gab-1 (p-Gab1), PI3K (PI3K•p110{alpha}) and eIF4E (p-eIF4E). Ratios (upper right corner) showed the positive immunostaining samples versus the total patient numbers for each CaP stage. Over 85% of these signal molecules were concurrently and consistently activated in HRCaP epithelium (P < 0.001; n = 4), suggesting the in vivo significance of HA-mediated CD168–ROK signaling in AI CaP transformation.

 
As shown in Figure 2B, we determined that the expression of CD168 and its downstream signal genes (ROK1, eIF4E3, CDC6 and MLC) were concurrently increased in AI CaP. The mean increases were 4.7 ± 1.1 fold for CD168, 9.1 ± 2.3 fold for ROK1, 3.4 ± 0.7 fold for eIF4E3, 5.1 ± 1.7 fold for CDC6 and 2.9 ± 1.3 fold for MLC (P < 0.05; n = 4). Such concurrent and consistent overexpression of the HA–CD168 signal genes in HRCaP demonstrated their possible functions in AI CaP transformation. Thus, we proposed the involvement of the HA–CD168–ROK signaling pathway in AI CaP as indicated in the route 2 of Figure 1.

Components of the CD168–ROK signaling cascade in human CaP tissue arrays in vivo
Our microarray data have established that the CD168/ROK1-associated signal genes were concurrently overexpressed in HRCaP cells in vivo (2,5). To determine the significance of this signal transduction pathway in AI CaP transformation, we conducted IHC staining for CD168/ROK-associated signal proteins in human CaP tissue arrays, which contained a total of 50 patients' samples staged as described above, according to Gleason scores 5–10. We observed highly concurrent and consistent activation of proteins of the CD168–ROK signaling pathway in response to CaP progression in agreement with the microarray results at the RNA level (Figure 2D).

Normal prostatic epithelium showed very limited CD168–ROK signal transduction, whereas the immunoreactivity of phosphorylated (active) forms of the CD168–ROK signal gene products, including CD168, ROK1, Gab-1, PI3K•p110{alpha} and eIF4E, was proportionally increased in concert with both clinical CaP stages and pathological Gleason scores. As shown in Figure 2D, IHC analysis barely detected CD168 expression and ROK1 activation in all normal and BPH prostatic epithelium regions (non-cancer). The IHC results of stage-2 CaP (mostly Gleason score 6 and a few Gleason score 7) showed weak expression and activation of the entire CD168–ROK1 signaling cascade in ~50% of the stage-2 patients, whereas those of stage-3 (mostly Gleason score 8) and stage-4 (Gleason scores 9–10) CaP displayed much stronger CD168 expression and CD168–ROK1 signaling activation in the prostatic epithelium. In particular, overexpression and activation of the entire CD168–ROK signaling cascade were concurrently co-localized in over 85% of metastatic HRCaP (stage-4, Gleason scores 9–10). These results clearly demonstrated a positive correlation between the activation of HA–CD168 signaling in prostatic epithelium and the progress of AI CaP in vivo.

HA-stimulated CD168–ROK1 signaling in AI rather than AD CaP cells
To test the relevance of the HA–CD168 interaction, we utilized three well-established human CaP cell lines; LNCaP, PC3 and DU145 because CD168 is expressed abundantly in the AI PC3 and DU145 cells but relatively less in the AD LNCaP cells, while the standard CD44 form (CD44s) is not present in AD LNCaP and is barely expressed in AI PC3 and DU145 cells (18,27,28). In addition, LNCaP and DU145 cells do not retain any cell surface HA and are relatively insensitive to HA-mediated adhesion, whereas PC3 cells express abundant HA and respond strongly to HA stimulation (29). For these reasons, studies of HA–CD168 interaction were carried out primarily in PC3 cells. The PC3 cell line was isolated from a bone metastasis of HRCaP, which may have preserved more HA-responding features than the other two cell lines derived from metastatic brain and lymph nodes, respectively (30).

Northern blot analysis showed that ROK1 mRNA was overexpressed in AI CaP PC3 and DU145 cells, but not in AD LNCaP cells (Figure 3A). Figure 3B illustrates that HA treatment (200 µg/ml) significantly increased ROK1 activation in AI PC3 and DU145 cells, but not in AD LNCaP cells as determined by western blot analysis. In PC3 cells, HA treatments (50–400 µg/ml) concurrently increased phosphorylated CD168/ROK-associated signal molecules, including ROK1, Gab-1, PI3K•p110{alpha} and eIF4E, which were completely consistent with the findings of human microarray and tissue array analyses (Figures 3C, 2B and D). All these signal molecules showed a dose-dependent response to HA concentration, suggesting a sequential signal cascade from ROK1 to Gab-1, PI3K•p110{alpha} and then eIF4E in the transformation of AI CaP cells. No significant difference was observed in ROK2, PI3K•p110ß, PI3K•p110{delta}, Ras and house-keeping ß-actin and GAPDH genes. Because epidermal growth factor receptor (EGFR) signaling predominantly functions through PI3K•p110{delta} and Ras is essential for CD44 signaling, these results also suggest that neither EGFR nor CD44 is involved in the HA-mediated ROK–PI3K signaling in AI PC3 cells.


Figure 3
View larger version (69K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 HA-mediated activation of the CD168–ROK signaling cascade in human CaP cell lines. (A) Northern blotting (left) and western blotting (right) analyses of CD168 and ROK1 expression levels, both showing high levels in AI PC3 and DU145 cells but low levels in AD LNCaP cells. (B) Western blot analysis of HA (200 µg/ml) stimulated ROK activation in AI PC3 and DU145 cells, but lack of activation in AD LNCaP cells (P < 0.001; n = 4). (C) Western blot analysis of HA-stimulated CD168–ROK signaling in AI PC3 cells. The CD168–ROK signaling cascade, including ROK1, Gab-1, PI3K•p110{alpha} and eIF4E, was concurrently activated (phosphorylated) in response to HA stimulation, whereas other HA-R pathways such as CD44, Ras, ROK2, PI3K•p110ß and PI3K•p110{delta} were not affected by the same HA treatment. Activation of the entire CD168–ROK signaling cascade was proportionally dose-dependent on HA concentration (50–400 µg/ml). (D) Inhibition of HA-mediated CD168–ROK signaling by Y-27632. Co-treatment with Y-27632 (50 µg/ml) and HA markedly blocked the HA-mediated PI3K activation in AI CaP but not AD CaP cells (P < 0.001; n = 6), demonstrating that HA stimulates AI CaP transformation via the CD168–ROK signaling cascade rather than the CD44–Ras pathway.

 
Blockade of ROK signaling by Y-27632 (50 µg/ml), a ROK-specific small molecule inhibitor, completely inhibited the HA-mediated activation of CD168–ROK signaling in AI CaP cells (Figure 3D), suggesting that HA predominantly functions through the CD168–ROK signal transduction pathway. HA activates the CD168–ROK signaling pathway and thus leads to PI3K/eIF4E-dependent tumorigenecity in AI CaP cells. Recent studies have demonstrated that PI3K-dependent Akt/TOR/eIF4E signaling is an important mediator of multiple cancer cell functions, including cell proliferation, anti-apoptosis (drug resistance), migration, metastasis and angiogenesis (20,31).

HA/CD168-mediated cell proliferation in AI CaP cells
Normal HA concentration in bone may be sufficient to activate CD168–ROK signaling, consequently leading to AI CaP cell proliferation and metastasis. In the following studies, we used 200 µg/ml of HA as an estimate of HA concentration in bone marrow (32,33), and 50 µg/ml of Y-27632 because it produced effective ROK inhibition (decrease of 78% ROK phosphorylation) in both AI CaP PC3 and DU145 cells without significant pharmacological toxicity.

Figure 4A shows a cell cycle analysis as determined by flow cytometry of cellular DNA density, indicating a significant increase of HA-mediated cell proliferation in the mitotic (M) phase of AI PC3 and DU145, but not in AD LNCaP cells. The mitotic cell population was increased from 31.5 ± 1.1 to 45.4 ± 2.3% (44% enhancement) in AI PC3, from 12.1 ± 0.7 to 20.7 ± 1.3% (71% increase) in AI DU145 cells, whereas no significant increase was observed in the HA-treated AD LNCaP cells, suggesting a supportive role of HA–CD168 signaling in AI CaP cell growth. Accordingly, the cell population at the G0/G1 phase of the cell cycle was reduced from 50.5 ± 3.4 to 40.4 ± 4.5% in PC3 cells and from 51.4 ± 1.9 to 33.2 ± 0.6% in DU145 cells. Blockade of CD168–ROK signaling by Y-27632 completely suppressed the stimulatory effect of HA on AI PC3 and DU145 cell proliferation, but not that of AD LNCaP. Y-27632 not only reduced the HA-stimulated mitotic cell population, but also restored the HA-decreased G0/G1 phase cell population back to the normal level. No apoptosis was detected after Y-27632 treatment as determined by the fragmented DNA content of the sub-G0/G1 phase cell population (the left-hand area of the first peak). We thus conclude that Y-27632 effectively prevented the CD168–ROK signaling cascade and suppressed the HA-stimulated AI cell proliferation at the G0/G1 phase of the cell cycle. Given that ROK1 was barely expressed in AD LNCaP cells (Figure 3A) and the cell cycle of LNCaP was not affected by HA and Y-27632, these findings suggest that HA-mediated CD168–ROK signaling plays an important role in AI CaP cell proliferation and is a potential mechanism of AI cell outgrowth in HRCaP.


Figure 4
View larger version (98K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Characterization of HA-stimulated cell proliferation, tumor invasion and metastasis in AD LNCaP, AI PC3 and DU145 cells. (A) Flow cytometry analysis of HA-stimulated CaP cell proliferation (P < 0.05; n = 4), showing assignment of cell populations with different DNA contents (y-axis) to different cell cycle stages (x-axis); from left to right, C (G0/G1 phase), D (S phase) and E (G2/M phase). The first (left) and second (right) peaks represent the levels of G0/G1 and M phase cell populations in the entire cell population, respectively. Bar charts (right) indicate the ratios of different cell populations (x-axis) versus different treatments (y-axis), including untreated control cells (Ctl), cells treated with HA (Ctl + HA) and cells treated with HA and Y-27632 (Ctl + HA + Y-27632). The white bar refers to the cell population resting in the G0/G1 non-dividing phase, whereas the black bar represents the mitotic (M) cell population. (B) IHC staining of human CaP tissue array demonstrated that CD44 expression was not correlated with AR activation [p-AR(ser213/210)] in prostatic epithelium. (C) Functional analysis of HA-stimulated tumor invasion in matrigel chambers (P < 0.05; n = 4). A significant increase in tumor invasion was detected in AI PC3 (black bar, labeled by an asterisk), whereas no significant increase was observed in AI DU145 (white bar) and AD LNCaP (gray bar) cells. Blockade of CD168–ROK signaling by Y-27632 markedly suppressed the HA-stimulated AI PC3 cell invasion. (D) Comparison of cell adhesion with the hBMEC monolayer between AI PC3 and DU145 cells (P < 0.05; n = 6). About 28% of the PC3 cell population (y-axis) quickly adhered to the hBMEC layer by 50 min after HA treatment (filled square) as compared with a 14% adhesion rate in PC3 cells without HA treatment (filled circle). DU145 cells responded poorly to HA treatment and barely showed any adhesion to hBMEC. Blockade of CD168–ROK signaling by co-treatment with Y-27632 and HA (filled triangle) markedly reduced the AI PC3 cell adhesion rate (from 28 to 5%).

 
HA/CD168-mediated cell invasion and metastasis in AI CaP
To ascertain the malignant role of HA–CD168 interaction in CaP invasion and metastasis, we tested matrigel cell invasion and AI CaP cell adhesion to hBMEC. Given that both CD44s and Ras are barely expressed in AI PC3 and DU145 cells (Figure 3C), the HA-mediated CaP cell invasion and metastasis should primarily reflect the effects of CD168–ROK signaling on AI CaP transformation. Again, CD44s was moderately present in normal epithelium and prostatic carcinoma in all stages, showing no relationship to the activation of ROK and the AR (Figure 4B), whereas ROK1 signaling is inversely related to activation of the AR in CaP epithelium in all stages of the tested patient samples (Figure 2D). Because these samples have been pre-selected to present a consistent correlation among cancer stage, Gleason score and AR insensitivity, the differential activation pattern between ROK1 and AR in these samples suggests that CD168–ROK1 signaling is probably an alternative route, other than the CD44–Ras pathway, to modulate HA-mediated transformation of AI CaP. This result however did not rule in the role of CD168–ROK1 signaling in patients with inconsistent CaP stage, Gleason score or AR activation.

Matrigel cell invasion assays showed that HA elevated the invasive population of AI PC3 cells from 7.6 to 13.9% (83% enhancement), whereas a much smaller effect was observed in DU145 and LNCaP cells (31 and 32% enhancement, respectively) (Figure 4C). Co-treatment with Y-27632 completely prevented the HA-promoted PC3 cell invasion (5.9%). A similar, but less dramatic, effect of Y-27632 was also observed in AI DU145 cells, but not in AD LNCaP cells, suggesting that HA increases the proliferation and invasiveness of bone-metastatic AI PC3 cells via the CD168–ROK signaling pathway. Furthermore, as a result of AI CaP cell adhesion to hBMEC, there was a significant increase in PC3 cell adhesion to hBMEC after HA treatment (Figure 4D). Within 50 min, 28% of PC3 cells adhered to the hBMEC monolayer in comparison with 14% in blank controls. In contrast, DU145 cells responded weakly to HA stimulation and showed limited adhesion to hBMEC. Blockade of ROK signaling by Y-27632 markedly reduced PC3 cell adhesion, indicating that this increase in metastatic cell adhesion was mediated through the HA-stimulated CD168–ROK signaling pathway as well. Given that PC3 and DU145 cells are derived from AI CaP cells metastasizing to bone and brain, respectively, HA-stimulated CD168–ROK signaling may preferentially promote the invasion and adhesion of AI CaP cells to bone endothelium, accounting for the high preference of HRCaP metastases for bone marrow under clinical circumstances.

Confirmation of HA-stimulated CD168–ROK tumorigenecity in AI CaP cells using an miR-based RNAi approach
To avoid possible cytotoxic effects of Y-27632 in CaP cells, we adopted a vector-based RNAi technology to deplete cellular ROK expression (24). After vector transfection, the CMV-mediated miR-based shRNA expression successfully suppressed >95% of ROK1 expression in AD LNCaP and AI PC3 cells (Figure 5A). The ROK-depleted PC3 cells displayed a slower proliferation rate (19% decrease) and a complete abolition of HA-stimulated cell proliferation, tumor invasion and adhesion to the hBMEC layer, reminiscent of the Y-27632-treated cells. On the basis of cell cycle analysis using flow cytometry (Figure 5B), administration of HA (200 µg/ml) increased the mitotic cell population in wild-type PC3 cultures by ~47%, but had no effect on ROK-depleted PC3 cultures. Similarly, matrigel cell invasion assays and hBMEC adhesion tests indicated much lower rates of tumor invasion and adhesion in ROK-depleted PC3 cells than in wild-type PC3 cells after HA stimulation (Figure 5C and D). Presumably because of the low expression rate of ROK in LNCaP cells, RNAi-mediated ROK depletion did not affect the properties of these cells. In fact, wild-type LNCaP cells are very similar to ROK-depleted AI PC3 cells in the measures reported here. Taken together, both ROK depletion experiments using RNAi and Y-27632, respectively, confirmed that CD168–ROK signaling is essential for HA-mediated AI CaP transformation.


Figure 5
View larger version (103K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 RNAi-mediated ROK gene silencing effects on AD LNCaP and AI PC3 cells. (A) Western blot analysis of ROK1 depletion in AI PC3 cells. The CMV-mediated miR-based shRNA (miR–ROK1) method successfully depleted >95% of ROK expression, while off-target integrin ß1 (ITGb1) and GAPDH gene products were not affected (P < 0.001; n = 3). (B) Significant suppression of HA-stimulated AI PC3 but not AD LNCaP cell proliferation after ROK depletion. Photographs were taken at 0 and 24 h after HA treatment (200 µg/ml). Bar charts (bottom) indicate the ratios of different cell populations (x-axis) versus different treatments (y-axis), including untreated CaP cell controls (Ctl), cells treated with HA (Ctl + HA), cells transfected with anti-eGFP shRNA and then treated with HA (Ctl + HA + miR-GFP) and cells transfected with anti-ROK1 shRNA and then treated with HA (Ctl + HA + miR-ROK). The white bar refers to the cell population resting in the G0/G1 non-dividing phase, whereas the black bar indicates the mitotic (M) cell population. (C) Functional analysis of HA-stimulated tumor invasion in matrigel chambers after ROK depletion (P < 0.001; n = 3). After 24 h of HA stimulation, tumor invasion was significantly increased in wild-type and miR-GFP-transfected AI PC3 cells (labeled by an asterisk), whereas no change was observed in ROK-depleted AI PC3 cells (black bar) and AD LNCaP cells (white bar) in all treatments. (D) Comparison of cell adhesion to hBMEC between AI PC3 and AD LNCaP cells after ROK depletion (P < 0.001; n = 3). Within 50 min of HA treatment (x-axis), ~29% of wild-type (filled square) and 27% of miR-GFP-transfected (filled diamond) PC3 cells quickly adhered to hBMEC, while only an 8% adhesion rate was observed in ROK-depleted PC3 cells (filled triangle), which is even lower than the 18% adhesion rate in wild-type PC3 cells without HA treatment (filled circle). Both HA treatment and RNAi-mediated ROK depletion did not affect LNCaP cells, which exhibited a limited adhesion to hBMEC.

 
Possible correlation between EGFR and CD168 signaling pathways in advanced CaP
Many signal transduction pathways can activate the PI3K-dependent Akt/TOR/eIF4E signaling cascade in cancers. In CaP, EGFR-stimulated PI3K•p110 consists primarily of the {delta}- and ß-isoforms (34–36), whereas the CD168/ROK1-mediated PI3K targets the p110{alpha} form (Figure 3D). As shown in Figure 6, IHC-stained epithelial tissues of normal and BPH prostate (non-cancer) showed strong EGFR activation but no ROK1 expression, stage-2 CaP (mostly Gleason score 6) showed weak EGFR and ROK1 activation in the patients' prostatic epithelium, while stage-3 (mostly Gleason score 8) and stage-4 (Gleason scores 9–10) CaP showed strong activation of both EGFR and ROK1 signaling in advanced cancer cells. The overall CD168–ROK1 signaling activation in the prostatic epithelium is proportional to the clinical stages and pathological Gleason scores of CaP progression, and also correlated with the overexpression and activation levels of EGFR in ~65% of advanced CaP patients with Gleason scores 7–10. Therefore, signal transduction in EGFR and CD168–ROK1 pathways showed a highly consistent pattern of synergistic activation in the cancerous epithelium of metastatic and hormone-refractory CaP.


Figure 6
View larger version (110K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Comparison of the EGFR and CD168–ROK1 signaling pathways in human CaP. Immunostaining of phosphorylated ROK1 (p-ROK), EGFR (p-EGFR) and their responding PI3K molecules was performed in human CaP tissue arrays (P < 0.001; n = 4). Ratios (upper right corner) showed the positive immunostaining samples versus the total patient numbers for each CaP stage.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Summary
 References
 
These findings point to a novel ECM-dependent signaling pathway responsible for cancer transformation in HRCaP, involved in AI cell proliferation, invasion and metastasis. The identification of this tumorigenic mechanism is supported by our recent observations that (i) HA receptor CD168 and its interacting kinase ROK1 were specifically overexpressed and activated in HRCaP cells, as determined by subtractive hybridization, microarray, northern blot and tissue array analyses (2,5); (ii) activation of the CD168–ROK1–PI3K•p110{alpha}–eIF4E signaling cascade was increased in proportion to the progression of CaP stages and Gleason scores; (iii) CD168 and ROK1 were highly expressed and activated in HA-stimulated AI PC3 and DU145 but not AD LNCaP cells; (iv) HA sequentially activated the CD168–ROK1–PI3K•p110{alpha}–eIF4E signaling cascade in a dose-dependent manner in AI CaP cells, which could be prevented by ROK depletion using RNAi or Y27632; and (v) activation of HA-mediated CD168–ROK signaling was manifested as enhanced AI CaP cell invasion and metastasis in human bone marrow endothelial layers. Thus, the HA-stimulated activation of CD168–ROK–PI3K signaling may be one of the causes of metastatic AI CaP transformation.

In this study, we further found that EGFR and CD168–ROK1 signaling pathways were co-activated in ~65% of advanced CaP patients with Gleason scores 7–10 (Figure 6). In HRCaP, such co-activation could be observed in >70% of the patients. Because both EGFR and CD168–ROK1 signal pathways function through PI3K, this dual activation of PI3K-dependent tumorigenecity may explain the drug resistance of recurring AI CaP. EGFR has been found to interact with ARs, in the absence of androgen, activating mitogen-activated protein kinase and PI3K, and therefore causing HRCaP (36). On the basis of this proposed model, clinical trials were conducted in HRCaP patients, testing agents that selectively interfered with EGFR signaling, including the anti-ErbB monoclonal antibodies trastuzumab, erlonitib and gefitinib (ZD1839, Iressa), and small molecule inhibitors directed against tyrosine kinases. To date, only 10–15% of the patients responded positively to these anti-EGFR drugs (37,38). If these anti-EGFR drugs succeeded in blocking EGFR-mediated PI3K activation, the HA–CD168 interaction might still be able to activate PI3K and trigger AI CaP transformation. Conceivably, HA-stimulated CD168–ROK signaling is probably involved in the mechanism underlying such high drug resistance in 85% of the trial patients. Because our tissue array data have revealed that CD168/ROK-mediated PI3K activation accounts for 85% of metastatic HRCaP in patients (Figure 2D), these results implicate a very risky mechanism by which AI CaP cells adopt an alternative pathway to activate PI3K and escape from anti-EGFR therapy. If indeed CD168–ROK signaling is a mediator for drug resistance and relapse in HRCaP, our findings signal an urgent need to elucidate the role of the HA–CD168 interaction in advanced CaP and the therapeutic value of blockade of this pathway in preventing cancer transformation and metastasis.

HA-mediated CD168–ROK signaling in HRCaP reflects a strong stromal–epithelial interaction in AI CaP transformation. Malignant cancer transformation is frequently characterized by active interactions between extracellular matrices and tumor cells, resulting in alterations of the signal transduction pathways essential for stromal–epithelial regulation. The present study provides evidence for this mechanism and also for the molecular correlation between tissue rigidity and tumor behavior in HRCaP. High HA deposits in the tissues surrounding the tumor often stimulate solid tumor formation, with high activities in cell invasion and metastasis (39). Our data show that bone-derived, metastatic PC3 cells opt for HA-mediated tumorigenecity more than brain-derived, metastatic DU145 cells, probably due to a more solid tissue environment in bone than in brain. The elevated deposit of HA in prostatic carcinoma promotes AI CaP transformation via the CD168–ROK–PI3K signaling and thus induces eIF4E–associated tumor invasion and metastasis. In tissues with high HA levels, the CD168/ROK1-dependent AI CaP cells such as PC3 precursors are further stimulated by HA and increase their invasion of ECM, leading to preferential metastasis into bone tissues. We therefore postulate that HA binds to CD168, activates RhoA-ROK and augments the PI3K-Akt/TOR/eIF4E signaling pathway, which mediates CD168/ROK1-dependent cancer cell proliferation, invasion and metastasis in AI CaP. This postulated chain of reactions is supported by the present study and other recent observations, showing that tissue rigidity reflects matrix stiffness, which in turn regulates tumor dissemination via growth factor signaling and RhoA-ROK activation, promoting tumor transformation through Rho•GTPase-dependent cytoskeletal tension and cell contractility (39).

In sum, the HA–CD168 signal transduction pathway constitutes a novel and important part of AI CaP transformation, accounting for resistance in androgen-ablation therapies. An elevated HA concentration in HRCaP triggers the CD168–ROK signaling cascade and raises PI3K-associated tumorigenecity in AI CaP cells. Normal prostatic HA deposition is restricted to stroma, whereas HA expression gradually becomes dysregulated in both stromal and epithelial regions in CaP tissues (40). In advanced CaP, the increased deposit of HA in the prostatic epithelium stimulates CD168–ROK signaling in AI CaP cells, which in turn activates PI3K/eIF4E-associated tumorigenecity and promotes AI cell outgrowth. As a result, cancer cells invade the ECM and preferentially metastasize in the bone marrow endothelium. In this study, the HA-mediated cancer cell transformation was only observed in AI CaP rather than AD CaP cells and HA concentration and AI-associated ROK activation were closely related. These findings are consistent with previous observations in biopsy samples of HRCaP, in which differential deposit of HA adjacent to prostatic epithelium was an effective and negative predictor for patient survival (40,41). Thus, further investigation of the HA–CD168 signaling cascade in AI CaP transformation may be essential for preventing the relapse and metastasis of HRCaP.


   Summary
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Summary
References
 
Our data indicate that HA-mediated CD168 signaling is part of the stimulus for transformation of AI CaP via the ROK–PI3K–eIF4E signaling cascade. Activation of this signal transduction pathway was directly related to the clinical staging, cell proliferation, cell invasion and metastasis of advanced CaP. These data also show that CD168 substitutes in the role of CD44 in hormone-refractory CaP and may shed light on the low effective rate of several currently used anti-EGFR drugs in clinical trials.


   Acknowledgments
 
This study was supported by NIH/NCI Grant CA-85722, American Cancer Society Grant IRG-58-007-45 and Wright Foundation Award 23-5104-4291.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Summary
 References
 

  1. Zhou J., Scholes J., Hsieh J.T. (2001) Signal transduction targets in androgen-independent prostate cancer. Cancer Metastasis Rev. 20:351–362.[CrossRef][Web of Science][Medline]
  2. Lin S.L. and Ying S.Y. (1999) Differentially expressed genes in activin-induced apoptotic LNCaP cells. Biochem. Biophys. Res. Commun. 257:187–192.[CrossRef][Web of Science][Medline]
  3. Shah R.B., Mehra R., Chinnaiyan A.M., et al. (2004) Androgen-independent prostate cancer is a heterogeneous group of diseases: lessons from a rapid autopsy program. Cancer Res. 64:9209–9216.[Abstract/Free Full Text]
  4. Cheng Y.G., Hannah M.L., Lin S.L., Lee J.M., Ying S.Y. (2002) Regulation of cell proliferation, apoptosis, and carcinogenesis by activin. Exp. Biol. Med. 227:75–87.[Abstract/Free Full Text]
  5. Ying S.Y. and Lin S.L. (2004) Gene expression in precursor cells of prostate cancer associated with activin by combination of subtractive hybridization and microarray technologies. Biochem. Biophys. Res. Commun. 313:104–109.[CrossRef][Web of Science][Medline]
  6. Laurent T.C. and Fraser J.R. (1992) Hyaluronan. FASEB J. 6:2397–2404.[Abstract]
  7. Turley E.A. (1992) Hyaluronan and cell locomotion. Cancer Metastasis Rev. 11:21–30.[CrossRef][Web of Science][Medline]
  8. Toole B.P., Zoltan-Jones A., Misra S., Ghatak S. (2005) Hyaluronan: a critical component of epithelial-mesenchymal and epithelial-carcinoma transitions. Cells Tissues Organs 179:66–72.[CrossRef][Web of Science][Medline]
  9. Csoka T.B., Frost G.I., Stern R. (1997) Hyaluronidases in tissue invasion. Invasion Metastasis 17:297–311.[Medline]
  10. Delpech B., Girard N., Olivier A., Maingonnat C., van Driessche G., van Beeumen J., Bertrand P., Duval C., Delpech A., Bourguignon J. (1997) The origin of hyaluronectin in human tumors. Int J Cancer 72:942–948.[CrossRef][Web of Science][Medline]
  11. Itoh K., Yoshioka K., Akedo H., Uehata M., Ishizaki T., Narumiya S. (1999) An essential part for Rho-associated kinase in the transcellular invasion of tumor cells. Nat. Med. 5:221–225.[CrossRef][Web of Science][Medline]
  12. Somlyo A.V., Phelps C., Dipierro C., Eto M., Read P., Barrett M., Gibson J.J, Burnitz M.C., Myers C., Somlyo A.P. (2003) Rho kinase and matrix metalloproteinase inhibitors cooperate to inhibit angiogenesis and growth of human prostate cancer xenotransplants. FASEB J. 17:223–234.[Abstract/Free Full Text]
  13. Bourguignon L.Y.W., Singleton P.A., Zho H., Diedrich F. (2003) Hyaluronan-mediated CD44 interaction with RhoGEF and Rho kinase promotes Grb2-associated binder-1 phosphorylation and phosphatidylinositol 3-kinase signaling leading to cytokine (macrophage-colony stimulating factor) production and breast tumor progression. J. Biol. Chem. 278:29420–29434.[Abstract/Free Full Text]
  14. Rodriguez-Viciana P., Warne P.H., Vanhaesbroeck B., Waterfield M.D., Downward J. (1996) Activation of phosphoinositide 3-kinase by interaction with Ras and by point mutation. EMBO J. 15:2442–2451.[Web of Science][Medline]
  15. De Benedetti A. and Graff J.R. (2004) eIF-4E expression and its role in malignancies and metastases. Oncogene 23:3189–3199.[CrossRef][Web of Science][Medline]
  16. Aaltomaa S., Lipponen P., Tammi R., Tammi M., Viitanen J., Kankkunen J.P., Kosma V.M. (2002) Strong stromal hyaluronan expression is associated with PSA recurrence in local prostate cancer. Urol. Int. 69:266–272.[CrossRef][Web of Science][Medline]
  17. Liu A.Y., Roudier M.P., True L.D. (2004) Heterogeneity in primary and metastatic prostate cancer as defined by cell surface CD profile. Am. J. Pathol. 165:1543–1556.[Abstract/Free Full Text]
  18. Dhir R., Gau J.T., Krill D., Bastacky S., Bahnson R.R., Cooper D.L., Becich M.J. (1997) CD44 expression in benign and neoplastic human prostates. Mol. Diagn. 2:197–204.[CrossRef][Web of Science][Medline]
  19. Pilarski L.M., Masellis-Smith A., Belch A.R., Yang B., Savani R.C., Turley E.A. (1994) RHAMM, a receptor for hyaluronan-mediated motility, on normal human lymphocytes, thymocytes and malignant B cells: a mediator in B cell malignancy? Leuk. Lymphoma 14:363–374.[Web of Science][Medline]
  20. Fruman D.A., Meyers R., Cantley L.C. (1998) Phosphoinositide kinases. Annu. Rev. Biochem. 67:481–507.[CrossRef][Web of Science][Medline]
  21. Nedvetzki S., Gonen E., Assayag N., et al. (2004) RHAMM, a receptor for hyaluronan-mediated motility, compensates for CD44 in inflamed CD44-knockout mice: a different interpretation of redundancy. Proc. Natl Acad. Sci. USA 101:18081–18086 [Erratum (2005). Proc. Natl Acad. Sci. USA, 102, 1263.].[Abstract/Free Full Text]
  22. Lin S.L., Chuong C.M., Widelitz R.B., Ying S.Y. (1999) In vivo analysis of cancerous gene expression by RNA-polymerase chain reaction. Nucleic Acid Res. 27:4585–4589.[Abstract/Free Full Text]
  23. Lin S.L. and Ying S.Y. (2003) mRNA/cDNA library construction using RNA-polymerase cycling reaction. Methods Mol. Biol. 221:129–143.[Medline]
  24. Lin S.L., Chang D., Wu D.Y., Ying S.Y. (2003) A novel RNA splicing-mediated gene silencing mechanism potential for genome evolution. Biochem. Biophys. Res. Commun. 310:754–760.[CrossRef][Web of Science][Medline]
  25. Watson J.V., Chambers S.H., Smith P.J. (1987) A pragmatic approach to the analysis of DNA histograms with a definable G1 peak. Cytometry 8:1–8.[Medline]
  26. Rafii S., Shapiro F., Rimarachin J., Nachman R.L., Ferris B., Weksler B., Moore M.A., Asch A.S. (1994) Isolation and characterization of human bone marrow microvascular endothelial cells: hematopoietic progenitor cell adhesion. Blood 84:10–19.[Abstract/Free Full Text]
  27. Miyake H., Hara I., Okamoto I., Gohji K., Yamanaka K., Arakawa S., Saya H., Kamidono S. (1998) Interaction between CD44 and hyaluronic acid regulates human prostate cancer development. J. Urol. 160:1562–1566.[CrossRef][Web of Science][Medline]
  28. Liu A.Y., LaTray L., van Den Engh G. (2000) Changes in cell surface molecules associated with in vitro culture of prostatic stromal cells. Prostate 44:303–312.[CrossRef][Web of Science][Medline]
  29. Simpson M.A., Reiland J., Burger S.R., Furcht L.T., Spicer A.P., Oegema T.R. Jr, McCarthy J.B. (2001) Hyaluronan synthase elevation in metastatic prostate carcinoma cells correlates with hyaluronan surface retention, a prerequisite for rapid adhesion to bone marrow endothelial cells. J. Biol. Chem. 276:17949–17957.[Abstract/Free Full Text]
  30. Webber M.M., Bello D., Kleinman H.K., Wartinger D.D., Williams D.E., Rhim J.S. (1996) Prostate specific antigen and androgen receptor induction and characterization of an immortalized adult human prostatic epithelial cell line. Carcinogenesis 17:1641–1646.[Abstract/Free Full Text]
  31. Bjornsti M.A. and Houghton PJ. (2004) Lost in translation: dysregulation of cap-dependent translation and cancer. Cancer Cell 5:519–523.[CrossRef][Web of Science][Medline]
  32. Vincent T., Molina L., Espert L., Mechti N. (2003) Hyaluronan, a major non-protein glycosaminoglycan component of the extracellular matrix in human bone marrow, mediates dexamethasone resistance in multiple myeloma. Br. J. Haematol. 121:259–269.[CrossRef][Web of Science][Medline]
  33. Courel M.N., Maingonnat C., Bertrand P., Chauzy C., Smadja-Joffe F., Delpech B. (2004) Biodistribution of injected tritiated hyaluronic acid in mice: a comparison between macromolecules and hyaluronic acid-derived oligosaccharides. In Vivo 18:181–187.[Abstract/Free Full Text]
  34. El-Rayes B.F. and LoRusso P.M. (2004) Targeting the epidermal growth factor receptor. Br. J. Cancer 91:418–424.[CrossRef][Web of Science][Medline]
  35. Sherwood E.R. and Lee C. (1995) Epidermal growth factor-related peptides and the epidermal growth factor receptor in normal and malignant prostate. World J. Urol. 13:290–296.[Web of Science][Medline]
  36. Sawyer C., Sturge J., Bennett D.C., O'Hare M.J., Allen W.E., Bain J., Jones G.E., Vanhaesebroeck B. (2003) Regulation of breast cancer cell chemotaxis by the phosphoinositide 3-kinase p110delta. Cancer Res. 63:1667–1675.[Abstract/Free Full Text]
  37. Lorusso P.M. (2003) Phase I studies of ZD1839 in patients with common solid tumors. Semin Oncol. 30:Suppl. 1, 21–29.[Web of Science][Medline]
  38. Arribas J. and Baselga J. (2005) Who will benefit from treatment against EGFR? Lancet Oncol. 6:257–258.[CrossRef][Web of Science][Medline]
  39. Akiri G., Sabo E., Dafni H., et al. (2003) Lysyl oxidase-related protein-1 promotes tumor fibrosis and tumor progression in vivo. Cancer Res. 63:1657–1666.[Abstract/Free Full Text]
  40. Lokeshwar V.B., Rubinowicz D., Schroeder G.L., Forgacs E., Minna J.D., Block N.L., Nadji M., Lokeshwar B.L. (2001) Stromal and epithelial expression of tumor markers hyaluronic acid and HYAL1 hyaluronidase in prostate cancer. J. Biol. Chem. 276:11922–11932.[Abstract/Free Full Text]
  41. Lipponen P., Aaltomaa S., Tammi R., Tammi M., Agren U., Kosma V.M. (2001) High stromal hyaluronan level is associated with poor differentiation and metastasis in prostate cancer. Eur. J. Cancer. 37:849–856.[CrossRef][Web of Science][Medline]
Received April 27, 2006; revised July 10, 2006; accepted July 12, 2006.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 RNAHome page
S.-L. Lin, A. Chiang, D. Chang, and S.-Y. Ying
Loss of mir-146a function in hormone-refractory prostate cancer
RNA, March 1, 2008; 14(3): 417 - 424.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
S.-L. Lin, D. Chang, A. Chiang, and S.-Y. Ying
Androgen receptor regulates CD168 expression and signaling in prostate cancer
Carcinogenesis, February 1, 2008; 29(2): 282 - 290.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/2/310    most recent
bgl134v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Lin, S.-L.
Right arrow Articles by Ying, S.-Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lin, S.-L.
Right arrow Articles by Ying, S.-Y.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?