Tel.: +1-604-875-5006, Fax: +1-604-875-5654
Cancer Cell Biology
The importance of integrin-linked kinase in the regulation of bladder cancer invasion
Article first published online: 28 MAY 2011
Copyright © 2011 UICC
International Journal of Cancer
Volume 130, Issue 3, pages 521–531, 1 February 2012
How to Cite
Matsui, Y., Assi, K., Ogawa, O., Raven, P. A., Dedhar, S., Gleave, M. E., Salh, B. and So, A. I. (2012), The importance of integrin-linked kinase in the regulation of bladder cancer invasion. Int. J. Cancer, 130: 521–531. doi: 10.1002/ijc.26008
- Issue published online: 24 NOV 2011
- Article first published online: 28 MAY 2011
- Accepted manuscript online: 23 FEB 2011 11:10AM EST
- Manuscript Accepted: 10 JAN 2011
- Manuscript Received: 27 AUG 2010
- integrin-linked kinase;
- bladder cancer;
It is important to understand the molecular mechanisms of bladder cancer progression not only to prevent cancer progression but also to detect new therapeutic targets against advanced bladder cancer. The integrin-linked kinase (ILK) is a major signaling integrator in mammalian cells and plays an important role in epithelial–mesenchymal transition (EMT) of human cancers, but its mechanisms are not completely understood. In this study, we investigated the importance and mechanisms of ILK in bladder cancer progression. When the expression of ILK in bladder cancer cell lines and N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN)-induced murine bladder cancer was evaluated, ILK has a tendency to be overexpressed in invasive cell lines and invasive BBN-induced murine bladder cancer. Overexpression of ILK in 253J bladder cancer cells suppressed E-cadherin expression, resulting in the promotion of cell invasion. Conversely, ILK knockdown by siRNA suppresses cell invasion in invasive bladder cancer cells through the regulation of E-cadherin or matrix metalloprotease 9 (MMP-9). To regulate E-cadherin expression, our results showed that the glycogen synthase kinase 3β (GSK3β)-Zeb1 pathway may play an important role downstream of ILK. Finally, the results of a human bladder tissue microarray (TMA) showed that ILK expression correlates with the invasiveness of human bladder cancer. Our study suggests that ILK is overexpressed in invasive bladder cancer and plays an important role in the EMT of bladder cancer via the control of E-cadherin and MMP-9 expression. ILK may be a new molecular target to suppress tumor progression in advanced and high-risk bladder cancer patients.
Metastatic or invasive bladder cancer is a life-threatening disease despite radical surgery or chemotherapy. Although approximately 70% of bladder cancers are non-muscle invasive at initial diagnosis,1 unfortunately, transurethral resection or adjuvant treatment options, including intravesical therapy, are limited in preventing tumor recurrence or progression, and 20–30% of those tumors progress into more aggressive and potentially lethal tumors. Therefore, it is important to understand the molecular mechanisms of bladder cancer progression to prevent cancer progression from superficial to invasive cancer or to detect new therapeutic targets against advanced bladder cancer.
Epithelial–mesenchymal transition (EMT) has been implicated as having a role in tumor invasion/migration and metastasis.2 Loss of E-cadherin expression is a hallmark of the EMT process, which is probably required for epithelial cells to undergo changes in cell morphology and motility and to adopt mesenchymal characteristics. A reduction or loss in expression of E-cadherin has been recognized as an important primary event in bladder tumorigenesis often linked to a poor prognosis. Expression of E-cadherin is controlled by several transcriptional repressors, including Twist, Snail1, Snail2/Slug, E47, Zeb1/TCF8 and Zeb2/SIP1, which bind to E-boxes in the E-cadherin promoter, and the importance of these transcriptional repressors has been recently investigated in various malignancies.3
The integrin-linked kinase (ILK) is a major signaling integrator in mammalian cells. ILK plays critical roles in development, cell motility, adhesion-dependent signaling, cytoskeleton reorganization and tumor invasion.4–6 Upon activation by different stimuli, ILK phosphorylates the serine/threonine protein kinase (Akt) and glycogen synthase kinase 3β (GSK3β) and participates in the related signaling cascades. Akt phosphorylation at Ser473 by ILK contributes to cell survival in cancer cells. With phosphorylation of GSK3β at serine 9 (ser9), ILK induces the activation of activator protein transcription factor, cAMP-responsive element binding protein or β-catenin/TCF transcription factor and the expression of cyclin D1 and matrix metalloprotease-9 (MMP-9).7–9 Furthermore, ILK was reported to regulate Snail expression transcriptionally by controlling poly(ADP-ribose) polymerase-1 (PARP-1) binding to the Snail promoter ILK responsive element (SIRE) or post-translationally via the phosphorylation by GSK3β, resulting in the suppression of E-cadherin expression.10, 11
Numerous cancers have now been described to undergo changes in levels of expression of ILK consequent upon acquisition of increasingly more malignant properties.4 These have been reported for colon, pancreas, prostate and gastric cancers, as well as melanoma.12–16 However, until now, the role of ILK expression and activity in bladder cancer has not been elucidated. In this study, we examine the importance of ILK in bladder cancer progression. We investigated the expression of ILK in a panel of bladder cancer cell lines, N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN)-induced murine bladder cancer and human bladder cancer specimens. We also studied the mechanisms of ILK-induced EMT in bladder cancer cells.
Material and Methods
Antibodies and reagents
Antibodies were obtained as follows: anti-pGSK3β (ser9), anti-Snail, anti-Slug, anti-MMP2 and anti-Zeb1 from Cell Signaling Technology (Danvers, MA), anti-Vinculin from Sigma (St. Louis, MO), anti-GSK3β and anti-MMP9 from Santa Cruz (Santa Cruz, CA), anti-ILK, anti-β catenin, anti-N-cadherin and anti-E-cadherin from BD Biosciences (Franklin Lakes, NJ). Lithium chloride (LiCl) was obtained from Sigma.
Eleven bladder tumor-derived cell lines and one SV40-transformed urothelial cell line (SV-HUC1) were used for screening ILK expression. Among the 11 bladder cancer cell lines, 253J, TCCsup, KK47, J82, RT112, UMUC3 and KU7 were kindly provided by Dr. O. Ogawa (Kyoto University, Japan), MGH-U3 cells were kindly provided by Dr. Yves Fradet, Laval University, and the remaining cell lines were purchased from the American Type Culture Collection (ATCC). Six cancer cell lines (253J, 5637, TCCsup, KK47, J82 and T24) were maintained in RPMI1640 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS). RT4, RT112, MGH-U3, UMUC3 and KU7 were maintained in McCoy's 5A supplemented with 10% FBS, RPMI1640 supplemented with 10% FBS, 2 mML-glutamine and 10 mM HEPES, minimum essential medium (MEM) supplemented with 10% FBS and 2 mML-glutamine, MEM-Eagle supplemented with 10% FBS and DMEM supplemented with 5% FBS, respectively. SV-HUC1 cells were purchased from ATCC and maintained in F12K medium supplemented with 10% FBS. Cells were incubated at 37°C in a humidified atmosphere containing 5% CO2.
Plasmid and transfection
253J cells were transiently transfected with either pcDNA 3.1 His-V5-tagged kinase constitutively active ILK (ILK S343D),17 expressing constitutively kinase-active mutant ILK, or pcDNA 3.1 His-V5 was used as a control (mock). Cells grown in 6-cm dishes were transfected for 4 h with 3 μg DNA using a 1:3 ratio of Lipofectamine 2000 transfection reagent (Invitrogen) according to the manufacturer's recommendations. Assays were carried out 24–72 h after transfection. Transfection efficiency was approximately 80–90% (data not shown).
Small interfering RNA treatment
A 21-bp double-stranded small interfering RNA (siRNA) molecule specifically targeting the integrin-binding domain of ILK and a control non-silencing sequence was synthesized (Dharmacon). The sequence of the DNA target of ILK-A is 5-GACGCTCAGCAGACATGTGGA-3.18 Cells at 40% confluency were transfected with 25 nmol/L of siRNA in serum-free OPTI-MEM (Invitrogen) using OligofectamineTM transfection reagent (Invitrogen). Four hours later, medium containing 15–30% FBS was added, resulting in a final FBS concentration of 5–10% and 72 h later, cells were harvested for protein analysis.
Cell viability assay
Cell viability was determined by MTS assays as an indicator of potential changes to cell proliferation. Briefly, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulphophenyl)-2H-tetrazolium (MTS, Sigma–Aldrich) was combined with phenazine methosulphate in a ratio of 20:1. This mixture was added to the culture media in a 1:5 ratio and incubated for 4 h at 37°C in a humidified 5% CO2 atmosphere. Absorbance was read at 490 nm using a microplate reader (Becton Dickinson Labware, Lincoln Park, NJ).
Cell migration and invasion assay
Cell migration was determined by a wound healing assay. After cells reached confluency, a wound was made in the monolayer by pressing a pipette tip down on the plate. The debris was removed by washing the monolayer twice with serum-free medium, and the cells were cultured for an additional 18–24 h. Cell migration was recorded in six different microscopic fields. The percentage of wound healing was calculated by the equation: (percent wound healing) = average of ([gap length: 0 h] − [gap length: 18–24 h])/[gap length: 0 h]).
Cell invasion was administrated using Matrigel invasion chambers (BD Biosciences). Cells (5 × 104) suspended in 500 μl of serum-free medium were applied on the upper compartment, and the lower compartment was filled with 750 μl of RPMI containing 10% fetal bovine serum. After 22–48 h of incubation, noninvaded cells on the upper surface of the filter were removed carefully with a cotton swab and cells were fixed with 100% methanol for 2 min. Invading cells on the lower side of the filter were stained with 0.5% crystal violet for 2 h, and the total number of invading cells was counted under a light microscope.
Western blot analysis
After drug treatments, cells were washed with PBS and lysed in an appropriate volume of ice-cold RIPA buffer composed of 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.5% sodium deoxycholate, 1% Nonidet P-40 and 0.1% sodium dodecyl sulfate (SDS) containing 1 mM Na3VO4, 1 mM NaF, 1 mM phenylmethylsulfonyl fluoride and protease inhibitor cocktail tablets (Complete, Roche Diagnostics GmbH, Mannheim, Germany). Cellular lysates were clarified by centrifugation at 13,000g for 20 min and protein concentrations of the lysates were determined by a BCA protein assay kit (PIERCE, Rockford, IL). Ten to thirty micrograms of the lysates were boiled for 5 min in SDS sample buffer and separated by SDS-PAGE on a 10–15% Tris-HCl minigel and transferred onto a polyvinylidene difluoride (PVDF) membrane following standard methods. Membranes were probed with appropriate dilutions of primary antibodies followed by incubation with horseradish peroxidase-conjugated secondary antibodies. After extensive washing, proteins were visualized by a chemiluminescent detection system (GE Healthcare, Buckinghamshire, UK).
Cells were incubated in serum-free media for 24 h, and the proteins in the conditioned medium were concentrated with Amicon Ultra-4 (Millipore) at 12,000 rpm for 30 min at 4°C. Proteins (20 μg) were loaded in nonreducing conditions on a 10% polyacrylamide gel containing 0.1% gelatin (Sigma). After electrophoresis, gels were incubated in Triton X-100 exchange buffer [20 mmol/L Tris-HCl (pH 8.0), 150 mmol/L NaCl, 5 mmol/L CaCl2 and 2.5% Triton X-100] for 30 min followed by a 10-min wash thrice with the incubation buffer (same buffer without Triton X-100). Gels were incubated in the incubation buffer overnight at 37°C, stained with 0.5% Coomassie blue R250 (Sigma) for 2 h and destained overnight with 30% methanol and 10% glacial acetic acid. Gelatinolytic activity was shown as clear areas in the gel.
Mouse bladder cancer specimen and tissue microarray immunohistochemistry
Six- to eight-weeks old C57BL/6 mice were given drinking water with 0.025% BBN (Tokyo Kasei Kogyo, Japan). The BBN solution was prepared freshly twice a week. Histopathological analysis of bladders was performed from mice fed BBN for 8 weeks for carcinoma in situ (Cis) or 20 weeks for invasive tumors. Negative control mice were given only water and were sacrificed at 16–20 weeks. In total two normal bladders, three Cis and superficial tumors and four invasive bladder tumors were immunohistologically stained for ILK expression. Statistical analysis could not be performed due to the small sample number.
For tissue microarray (TMA) immunohistochemistry, a BL803 TMA was purchased from US Biomax, Inc. (Rockville, MD). This TMA contains 30 cases of transitional cell carcinoma (TCC), five each of tumor-adjacent tissue and normal tissue with duplicate cores per case. Each core measures 1.5 mm in diameter and cores are arrayed in a rectangular fashion with a 10 by 8 layout. Each core from each cancer tissue represents a single specimen that was selected and pathologically confirmed. In this study, we used 29 cases of TCC with histologic grading according to the World Health Organization system19 and seven normal tissues to examine ILK, E-cadherin and MMP-9 expression in the bladder. Immunohistochemical staining was conducted on a Discover XT™ Ventana autostainer (Ventana Medical System, Inc. Tuscan, AZ). TMA slides were scanned on a BLISS System from Bacus Laboratories (Olympus, Center Valley, PA). The staining intensity in each section was graded on a four-point visual scoring scheme with concern to the percentage of the stained area. All comparisons of staining intensity and percentages were made by a pathologist (L.F.) at 200× magnification.
All results are expressed as means ± SE. The mean values and corresponding SEs were calculated using Prism 4.03 (GraphPad Software). Data were analyzed by chi-square, Kruskal-Wallis and Mann-Whitney U tests using InStat 3.06 (GraphPad). Comparison of multiple groups was performed by ANOVA in Sigmaplot 11.0 (Systat Software Inc.). P values of ≤0.05 were considered statistically significant.
Invasive bladder cancer cells tend to have high expression of ILK
ILK protein has been reported to relate closely to tumor progression and invasion in various cancers through the regulation of E-cadherin or MMP9 expression. In this study, 11 bladder cancer cell lines and one SV40-transformed urothelial cell line (SV-HUC1) were screened for the expression of ILK, E-cadherin, MMPs and other EMT markers. Among 12 cell lines, six showed high expression of ILK protein (Fig. 1a). They showed low expression of E-cadherin and had a tendency to show mesenchymal morphology defined by poorly adherent carcinoma cells displaying a stellate morphology. Conversely, six different cell lines with epithelial morphology, defined by tightly adherent cuboidal cells growing as discrete colonies, showed low expression of ILK protein and high expression of E-cadherin. Phosphorylation of GSK3β at serine 9 (ser9), which is a representative downstream target of ILK, was increased in cell lines with high ILK expression. All cell lines with high ILK expression expressed Zeb1 protein, whereas only SV-HUC1 showed low expression of Zeb1 among those with low or no ILK expression. No clear correlation with ILK expression was found among other EMT markers such as N-cadherin, Vimentin, Snail, Slug, MMP2 and MMP-9, although some mesenchymal cell lines showed high expression of such proteins.
BBN-induced mouse bladder cancer showed a progressive change in ILK expression during carcinogenesis (Fig. 1b). Normal urothelium expressed minimal amounts of ILK homogeneously, while carcinoma in situ (Cis) showed a low to moderate amount of intracellular expression of ILK. Invasive bladder cancers showed strong ILK expression, and this expression was strongest at the invasion front [Fig 1b (D)].
From these results, ILK expression becomes more pronounced with bladder cancer progression and may play some role in the EMT of bladder cancer through E-cadherin regulation.
Overexpression of ILK suppresses E-cadherin expression and promotes cell invasion in 253J bladder cancer cells
To examine the influence of an activation of the ILK pathway on bladder cancer cells with epithelial morphology, 253J cells were transiently transfected with either a constitutively active ILK expressing vector or a control vector; 253J cells were chosen because of their transfection efficiency and epithelial characterization. Immuocytoflourescence has shown that 253J cells highly express membranous E-cadherin at cell junctions (data not shown), and 253J has previously been characterized as having an epithelial morphology.20, 21 Western blot analysis demonstrated that kinase-active ILK overexpression induced GSK3β phosphorylation at ser9, which was associated with suppression of E-cadherin (Fig. 2a). At the same time, Zeb1 expression was induced by active ILK overexpression, although there was no change in Snail or β-catenin expression.
Regarding phenotypic change, while 253J cell viability did not change (Fig. 2b), kinase-active ILK overexpression promoted cell migration and cell invasion significantly compared with control vector-transfected cells (Figs. 2c and 2d, p < 0.0001 in both the cell migration assay and the cell invasion assay).
To investigate whether induction of phosphorylated GSK3β (pGSK3β) by active ILK is important to suppress E-cadherin expression, phosphorylation of GSK3β was induced by LiCl and the change in E-cadherin expression was evaluated. LiCl induced phosphorylation of GSK3β and suppressed E-cadherin expression in a dose-dependent manner (Fig. 3a). This was associated with Zeb1 induction with a decrease in Snail levels. No change was seen in cell viability (Fig. 3b), migration (Fig. 3c) or invasion (Fig. 3d) as a result of LiCl addition. This is likely due to the decrease in Snail protein expression with LiCl addition which would reduce invasion and counteract any effects resulting from the increased expression of Zeb1. Furthermore, LiCl may reduce proliferation in general, obscuring phenotypic change from induced Zeb1. Taken together, these results suggest that activation of ILK in bladder cancer may promote cell invasion through a suppression of E-cadherin via the control of the GSK3β-Zeb1 pathway.
ILK knockdown suppresses cell invasion in invasive bladder cancer cells through the regulation of E-cadherin and MMP-9
To further confirm the importance of ILK in invasive bladder cancer, we knocked down ILK expression using an ILK-siRNA (ILK-A) transfection in two invasive bladder cancer cell lines. In both the cell lines examined, the 25 nM ILK-A transfection almost completely knocked down ILK expression after 72 h (Fig. 4b). When ILK was knocked down, cell viability did not change significantly; however, cell migration and invasion were significantly inhibited in both cell lines (Fig. 4a).
Western blot analysis demonstrated that ILK knockdown suppressed GSK3β phosphorylation at ser9 in both the cell lines, but it resulted in E-cadherin re-expression only in TCCsup cells (Fig. 4b). In TCCsup cells, the suppression of ILK led to Zeb1 downregulation, although there was no change in Snail expression. In UMUC3 cells, the suppression of ILK led to Snail downregulation but not Zeb1 suppression, and E-cadherin expression was downregulated. From these results, the ILK-GSK3β-Zeb1 pathway may be important in regulating the EMT of bladder cancer via the control of E-cadherin expression but other EMT pathways under ILK regulation are also likely involved.
When alternative pathways regulating cell migration and invasion were examined in UMUC3 cells, Western blot analysis and zymography showed that MMP-9 expression and activity was suppressed in UMUC3 cells by ILK knockdown, whereas it was not changed in TCCsup cells (Fig. 4c). No change in MMP-2 protein levels or activity was observed in TCCsup and UMUC3 cells either. These results suggest that ILK may regulate EMT through multiple pathways, including suppression of E-cadherin expression and also through alteration of MMP-9 regulation in bladder cancer.
ILK expression correlates with the invasiveness of human bladder cancer
Using bladder cancer TMAs, we evaluated the relationship between the pathological stage of bladder cancer and ILK, E-cadherin and MMP-9 expression. Using four-point visual scoring, the expression level was divided into two categories (low or high expression). The results showed that ILK and MMP-9 expression is related to tumor stage; ILK and MMP-9 expression were significantly upregulated in invasive bladder cancer compared with both normal urothelium and non-muscle invasive bladder cancer (Fig 5). Conversely, E-cadherin expression was suppressed in invasive bladder cancer and was inversely related to pathological stage.
However, sample numbers are limited and, therefore, although bladder tumors with high ILK expression tended to show lower E-cadherin (76.5%) or high MMP-9 expression levels (66.7%), associations did not reach statistical significance. However, 72.7% (8/11 cases) showed high E-cadherin expression or low MMP-9 expression among bladder cancers with low ILK expression, whereas 94.4% (17/18 cases) showed low E-cadherin expression or high MMP-9 expression in those with high ILK expression (p = 0.0004). Collectively from these results combined with data from our in vitro and in vivo studies, we propose that ILK regulates the EMT of human bladder cancer through the E-cadherin and MMP-9 pathways.
The prognostic significance of various EMT markers has been studied in bladder cancer, a malignancy that may be aggressive; advanced and metastatic bladder disease may be treated with cisplatin-based chemotherapy but is incurable, highlighting the importance of the development of novel therapeutics.22, 23 In this study, we analyzed ILK expression in bladder cancer cell lines and cancer specimens and found that there was a significant correlation between EMT and ILK expression. ILK is a serine/threonine kinase that interacts through its COOH terminus with beta-1 and beta-3 integrins, which are receptors for extracellular matrix (ECM) proteins such as collagen and laminin.24 ILK, via its COOH terminus, also interacts with actin cytoskeleton proteins through association with CH-ILKBP-actopaxin, paxillin and affixin, which are responsible for focal adhesion formation.4, 25–28 In addition, ILK interacts with growth factor receptors through the association of its four ankyrin repeats at the NH2 terminus with PINCH and Nck-2 adaptor proteins.29 ILK also contains a pleckstrin homology-like domain-binding phosphatidylinositol triphosphate24–26 which regulates its kinase activity. Furthermore, ILK binds to Akt and GSK3β and controls their activities via phosphorylation.30 ILK phosphorylates and activates Akt at Ser473, which regulates genes essential for survival.31–33 ILK also phosphorylates GSK3β at ser9 and inactivates it, leading to the activation of activator protein transcription factor, cAMP-responsive element binding protein, or beta-catenin/TCF transcription factor and the expression of cyclin D1 and MMP-9.7–9 All together, ILK is upstream of various EMT markers and thus may be important to regulate and integrate those pathways, making it an attractive target for cancer therapeutics.
We demonstrated that through ILK overexpression and knockdown experiments that ILK may control E-cadherin expression. As E-cadherin expression is strongly correlated with cancer invasion, ILK overexpression promoted cell migration and invasion, and ILK knockdown inhibited those activities. While the transcriptional downregulation of E-cadherin expression has been reported to involve ILK-mediated activation of the E-cadherin repressors Snail34, 35 and Zeb1,35 we did not observe E-cadherin induction after the suppression of Snail expression with ILK knockdown in UMUC3 cells. Conversely, E-cadherin was induced by the suppression of Zeb1 with ILK knockdown in TCCsup cells. Further, when active ILK is overexpressed in 253J cells, Zeb1 is induced but there is no change in Snail protein level, resulting in the suppression of E-cadherin expression. These results suggest that Zeb1 may be more important than Snail in the regulation of E-cadherin expression in some bladder cancers.
Phosphorylation of GSK3β at ser9 may play an important role in the mechanism of ILK-mediated activation of Zeb1. McPhee et al. have shown that the ILK/Akt pathway promotes PARP-1 to bind SIRE and modulate Snail expression.10 Adam et al. reported that the microRNA-200 family regulates transcription of Zeb1 and Zeb2 in UMUC3 bladder cancer cells.36 However, our results showed that the inhibitory phosphorylation of GSK3β at ser9 by LiCl induced not Snail expression but Zeb1 expression, leading to E-cadherin suppression. Interestingly, Zhou et al. have reported that GSK3β can post-transcriptionally regulate the Snail protein level, and the phosphorylation of GSK3β at ser9 is accompanied by the increased expression of Snail protein.11 Although the opposite was seen for Snail protein in 253J cells, Zeb1 may be regulated by GSK3β in a similar manner.
Separately from E-cadherin regulation, ILK also regulates MMP-9 protein, which is involved in the degradation of the extracellular matrix necessary for cell migration and invasion. Activation of AP-1 binding sites in the MMP-9 promoter through the phosphorylation of GSK3β at ser9 appears to be responsible for the ILK-induced expression of MMP-9.37 In our results, ILK knockdown in UMUC3 cells dephosphorylated GSK3β at ser9 and suppressed MMP-9 protein, leading to the suppression of cell migration and invasion, although there is no change in the E-cadherin level. No change in MMP-9 level or activity in TCCsup cells was detected, despite the dephosphorylation of GSK3β at ser9 by ILK knockdown. These results indicate that ILK regulates the EMT of bladder cancer but the mechanism depends on cell types. It is difficult to find a common pathway controlling EMT under ILK regulation, suggesting that various EMT pathways may be involved. Recently, QLT0267, a selective ILK inhibitor, has been developed and its effect was demonstrated in various cancer cells in vitro and in vivo.38, 39 Bladder cancer may be a good model for such an ILK inhibitor as well as other cancers such as melanoma, thyroid cancer and glioblastoma.
In our TMA, ILK expression correlated with the invasiveness of bladder cancer, although the correlation with the prognosis could not be examined. Because the sample number is very limited, the correlation of ILK expression with E-cadherin or MMP-9 expression did not reach significance, but tumors with high ILK expression showed lower E-cadherin and higher MMP-9 expression. Although it is still controversial whether MMP-9 or E-cadherin correlates with overall survival of the patients with bladder cancer,22, 23, 40–43 the ratio of MMP-9 and E-cadherin was reported to be strongly prognostic for disease specific survival.44 From our results indicating that ILK expression correlates with lower E-cadherin or higher MMP-9 expression, ILK can be considered as a promising prognostic factor of bladder cancer.
In conclusion, our study indicates that ILK is important in the EMT of bladder cancer, which regulates E-cadherin and MMP-9 expression. We propose that ILK may be a new target to suppress tumor progression; collectively, the results of this study provide the preclinical proof of principle to further evaluate the benefits of ILK inhibition in bladder cancer, either systemically through novel molecules or intravesically using nucleotide-based agents against high-risk non-muscle invasive bladder cancers.
- 3The cadherin superfamily. WormBook 2005: 1–9..