This article is mentioned in:

Introduction

Urothelial cancer of the bladder is the 4th most commonly diagnosed cancer in men worldwide (1). Patients with metastatic disease are often treated with a combination chemotherapy containing gemcitabine and cisplatin as a standard of care (2,3). However, the treatment success is limited, resulting in a median survival of 12–15 months. Treatment failure is commonly caused by acquired resistance after primary response (2,3). Therefore, efficient second line chemotherapies are urgently needed.

Integrins have been identified to play an important role in the development of resistance to chemotherapy in bladder cancer (4). These molecules are transmembrane receptors with two different chains, an α (alpha) and a β (beta) subunit. Integrins are bridges for cell-cell and cell-extracellular matrix (ECM) interactions. Cell-matrix contact plays a fundamental role in the metastatic potential of tumors (5). Alterations of integrin expression may result in an enhanced adhesive behavior in bladder cancer (6). Moreover, the expression patterns of integrin subtypes are known to be important mediators of tumor cell de-differentiation and tumor proliferation (7). Furthermore, it was shown that integrins were involved in the development of metastasis and recurrence of urothelial cancer (6,8,9).

Drug-adapted cancer cell lines have been successfully used to study cancer cell resistance mechanisms (10,11). To reflect the heterogeneity of individual bladder cancer patients and to enable a systematic evaluation of the role of integrins concerning resistance acquisition, we used a panel of 12 urothelial cancer cell lines consisting of 4 parental chemosensitive cell lines and their sublines with acquired resistance to gemcitabine or cisplatin (12,13).

Materials and methods

Drugs

Cisplatin was purchased from Gry-Pharma (Kirchzarten, Germany), gemcitabine from Lilly (Bad Homburg, Germany).

Cell lines

The cell lines 5637, T24, HT1376, and TCC-SUP were obtained from the American Type Culture Collection (Manassas, VA, USA). The following drug-resistant sublines were established by continuous exposure to increasing drug concentrations as described previously (12,14) and are part of the Resistant Cancer Cell Line (RCCL) collection (http://www.kent.ac.uk/stms/cmp/RCCL/RCCLabout.html): 5637rCDDP1000 (cisplatin-resistant, 1,000 ng cisplatin/ml), 5637rGEMCI20 (gemcitabine-resistant, 20 ng gemcitabine/ml), T24rCDDP1000, T24rGEMCI20, HT1376rCDDP1000, HT1376rGEMCI20, TCC-SUPrCDDP1000, and TCC-SUPrGEMCI20. All cell lines were grown in Iscove's modified Dulbecco's medium supplemented with 10% fetal calf serum (FCS; Gibco, Karlsruhe, Germany), 100 IU/ml penicillin, and 100 µg/ml streptomycin at 37°C. Cell line authentication was performed by short tandem repeat profiling.

Cell adhesion to extracellular matrix components

24-well plates were coated with extracellular matrix components (Matrigel; Corning, Amsterdam, The Netherlands) overnight. Plates were washed with 1% bovine serum albumin (BSA) in phosphate buffered saline (PBS) to block nonspecific cell adhesion. Thereafter, 0.5×106 tumor cells were added to each well for 60 min. Subsequently, non-adherent tumor cells were washed off. The adherent cells were fixed with 1% glutaraldehyde and counted in five different fields using a microscope (20× objective) to calculate the mean cellular adhesion rate.

Chemotaxis

Serum induced cell migration was examined using 6-well transwell chambers (Greiner, Frickenhausen, Germany) with 8 µm pores. To evaluate cell migration, cells were placed in the upper chamber for 20 h in serum-free medium. The lower chamber contained 10% serum. After incubation, the upper surface of the transwell membrane was wiped gently with a cotton swab to remove non-migrating cells. Cells migrating to the lower surface of the membrane were stained using hematoxylin and counted. Cells migrating into the lower chamber were counted separately under the microscope.

Blocking study

Cells were preincubated for 60 min with a function-blocking anti-integrin β1 monoclonal antibody (20 mg/l) (MAB 2253Z; MerckMillipore, Darmstadt, Germany). Controls remained untreated. Adhesion and chemotaxis was tested as indicated above.

Flow cytometry

Cells were washed in blocking solution (PBS, 0.5% BSA) and then incubated for 60 min at 4°C with phycoerythrin (PE)-conjugated monoclonal antibodies directed against the following integrin subtypes: Anti-α3 (IgG1; clone C3II.1), anti-α5 (IgG1; clone IIA1), anti-α6 (IgG2b, clone MP 4F10), anti-β1 (IgG1; clone MAR4), anti-β3 (IgG1; clone VI-PL2) or anti-β4 (IgG2a; clone 439-9B; all: BD Biosciences, Heidelberg, Germany). Integrin expression was measured by flow cytometry (FACSCalibur; BD Biosciences, Heidelberg, Germany). Mouse IgG1-PE (MOPC-21) or mouse IgG2a-PE (G155-178; all: BD Biosciences) antibodies were used as isotype control.

Immunohistochemistry

33 cases of invasive and non-invasive bladder cancers as well es corresponding normal urothelium were taken from the archive of the Dr. Senckenberg Institute of Pathology in Frankfurt. Tissue sections were stained for Integrin β1, (D2E5) Rabbit mAb, Cell Signaling Technology (Waltham, MA, USA), dilution 1:100. In brief, 4 µm sections were cut and pretreated with Trilogy™, Cell Marcque (Rocklin, CA, USA), incubated with the antibody, antigen retrieval was performed at pH 6 in a microwave oven using the Peroxidase-FLEX EnVision kit (Dako, Jena, Germany).

A pathologist, who was blinded to clinical history and therapeutic response, scored the immunohistochemical staining using a five-stage staining score: 0=negative; 1=weak; 2=moderate; 3=strong; 4=very strong.

Images were acquired using a digital slide scanner (ScanScope XT; Aperio, Vista, CA, USA).

Statistical analysis

Results are expressed as mean ± SD of at least three independent experiments. For statistical analysis student's t-test, analysis of variance (ANOVA), and Student-Newman-Keuls-Test were performed whenever applicable. P<0.05 was considered to indicate a statistically significant difference.

Results

Influence of acquired resistance on adhesion to extracellular matrix components

In untreated cells, adhesion to extracellular matrix components was decreased in 2 of 4 gemcitabine-resistant sublines (HT1376rGEMCI20 and TCC-SUPrGEMCI20) and upregulated in 2 of 4 cell lines (T24rGEMCI20 and 5637rGEMCI20) compared to parental cells. In cisplatin-resistant sublines, adhesion was decreased in 1 of 4 cisplatin-resistant sublines (TCC-SUPrCDDP1000) and enhanced in 2 cell lines (HT1376rCDDP1000 and 5637rCDDP1000) (Fig. 1).

Figure 1. Evaluation of cell adhesion. Comparison of adhesive behavior of parental bladder cancer cell lines as well as of their gemcitabine- and cisplatin-resistant sublines. The plates were coated with extracellular matrix components overnight. Plates were washed with 1% BSA in PBS to block nonspecific cell adhesion. Thereafter, 0.5×106 tumor cells were added to each well for 60 min. Subsequently, non-adherent tumor cells were washed off. The adherent cells were fixed with 1% glutaraldehyde and counted in five different fields using a microscope (20× objective) to calculate the mean cellular adhesion rate. Adhesion of parental cell lines was set as 100%. *P≤0.05 vs. controls.

Influence of acquired chemoresistance on chemotaxis

Chemotaxis was enhanced in 3 of 4 gemcitabine-resistant urothelial cancer cell lines (gemcitabine-resistant sublines of TCC-SUP, HT1376 and T24). In cisplatin-resistant sublines, chemotaxis was enhanced in HT1376rCDDP1000 and T24rCDDP1000 compared to parental cell lines (Fig. 2).

Figure 2. Evaluation of chemotaxis. Serum-induced chemotactic movement was examined using six-well transwell chambers with 8-µm pores. A total of 0.5×106 tumor cells per ml were placed in the upper chamber in serum-free medium. The lower chamber contained 10% serum. After 20 h incubation, the upper surface of the transwell membrane was gently wiped with a cotton swab to remove nonmigrating cells. Cells which had moved to the lower surface of the membrane were stained using haematoxylin and counted with a microscope (20× objective). The mean chemotaxis rate was calculated from five different fields. Chemotaxis of parental cell lines was set as 100%. *P≤0.05 vs. controls.

Differential expression of cell surface integrins

Expression of integrins on the cell surface was analyzed by flow cytometry (Fig. 3). In gemcitabine-resistant sublines, the expression of integrin α3 was enhanced in 3 sublines (gemcitabine-resistant sublines of T24, 5637, and TCC-SUP) and diminished in HT1376rGEMCI20 compared to parental cell lines. Integrin β1 expression was upregulated in all gemcitabine-resistant sublines compared to parental cells. Integrin β4 expression was enhanced in TCC-SUPrGEMCI20 and diminished in T24rGEMCI20 and in 5637rGEMCI20.

Figure 3. Flow cytometry analysis of integrin surface expression. Cells were washed in blocking solution and stained with specific monoclonal antibodies as listed in materials and methods. Mouse IgG1-PE or mouse IgG2a-PE antibodies were used as isotype controls. Fluorescence was analyzed using a FACScan flow cytometer. The relative fluorescence unit values are given in percentage difference to the parental cell lines. Parental cell lines were set as 100%. One of three independent experiments is shown here. *P≤0.05 vs. controls.

Comparing cisplatin-resistant cell lines, Integrin α3 expression was upregulated in 3 of 4 sublines (cisplatin-resistant sublines of T24, 5637, and TCC-SUP) and downregulated in HT1376rCDDP1000. Integrin α5 was upregulated in 3 of 4 sublines (cisplatin-resistant sublines of 5637, TCC-SUP, and HT1376). Integrin β1 expression was upregulated in 3 of 4 sublines (cisplatin-resistant sublines of HT1376, T24, and 5637) and downregulated in TCC-SUPrCDDP1000. Integrin β4 was upregulated in HT1376rCDDP1000 and downregulated in all other tested sublines (cisplatin-resistant sublines of T24, 5637, and TCC-SUP).

Immunohistochemical staining of integrin β1

In non-malignant tissue samples, integrin β1 was only visible in the basal layer of the urothelium. The medium staining score for normal urothelium was between ‘negative’ to ‘weak’ (0.88±0.33). In bladder cancer samples, the mean integrin β1 staining score was between ‘moderate’ and ‘strong’ (2.48±1.42). There was no significant difference of integrin β1 staining between low grade tumors or high grade tumors and no significant difference between non-muscle invasive tumors and muscle invasive tumors (Fig. 4).

Figure 4. Expression of integrin β1 in non-malignant urothelium and urothelial cancer. (A) Normal urothelium showed a negative or a weak expression of intergrin β1 only in the basal cell layer. A higher integrin β1 expression was visible in samples of (B) non invasive low grade urothelial cancer, (C) non invasive high grade cancer, and (D) muscle invasive high grade cancer. Immunostaining of 4 representative tissue samples.

Influence of blocking integrin β1 on adhesion and chemotaxis

Functional blocking of integrin β1 resulted in a reduced adhesion in 2 of 4 parental urothelial cancer cell lines (HT1376 and T24). In gemcitabine-resistant cells, adhesion was downregulated in 3 of 4 cell lines (gemcitabine-resistant sublines of HT1376, T24 and TCC-SUP). In cisplatin-resistant cells, adhesion was downregulated in all 4 tested cell lines (Fig. 5). We could not detect an influence on chemotaxis after blocking integrin β1 (Fig. 6).

Figure 5. Influence on adhesion after blocking integrin β1. Bladder cancer cell lines were preincubated for 60 min with a monoclonal anti-integrin β1 blocking antibody (20 mg/l). Controls remained untreated. Cells were then subjected to the adhesion assay as indicated. Untreated parental cells were set as 100%. *P≤0.05 vs. controls.

Figure 6. Influence on chemotaxis after blocking integrin β1. Bladder cancer cell lines were preincubated for 60 min with a monoclonal anti-integrin β1 blocking antibody (20 mg/l). Controls remained untreated. Cells were then subjected to the chemotaxis assay as indicated. Untreated parental cells were set as 100%. *P≤0.05 vs. controls.

Discussion

In the present study, we used a well-established panel of urothelial cancer cell lines with acquired resistance to gemcitabine or cisplatin, the standard therapeutics for metastatic urothelial cancer of the bladder (2,3). Cell line panels seem to be necessary to reflect the heterogeneity of different patient-derived cancer cell lines. Although the complex scenario of metastatic colonization is not fully understood, there is strong evidence that alterations of tumor-matrix contact are necessary to allow motile crawling into the surrounding tissue (15).

In several cancer cells, resistance to gemcitabine seems to be connected with integrins and associated proteins (16–18). In addition, resistance to cytotoxic drugs and proliferation regulation was shown to be dependent on extracellular matrix proteins (19). In this study, acquisition of resistance to gemcitabine or cisplatin showed a changed adhesive behavior with some resistant sublines showing an enhanced adhesive behavior and other resistant sublines being less adhesive (Fig. 1). In contrast, the influence on chemotaxis was more uniform, with 5 of 8 sublines showing an enhanced chemotaxis and no subline with a significantly diminished chemotaxis after acquisition of resistance (Fig. 2). This is in line with Ploenes et al who reported about an enhanced chemotaxis in lung cancer cell lines with an increased chemoresistance (20).

Since integrins seem to be involved in the development of resistance to chemotherapy in bladder cancer (4) and alterations of integrin expression change adhesive and invasive behavior of bladder cancer cells (6), we aimed to elucidate the role of integrins in this context.

Integrin α3 might be involved in resistance acquisition, since it was upregulated in 3 of 4 gemcitabine-resistant and also in 3 of 4 cisplatin-resistant sublines in this study (Fig. 3). Litynska et al (21) tried to analyze the role of integrin α3 in bladder cancer by blocking its function. They described that adhesion was up- or downregulated after blocking integrin α3 depending on the tested cell line. The cell line specific effects that can be triggered after acquisition of resistance show the heterogeneity between independent cell lines and underline the importance of using a panel of cell lines for a better interpretation.

It was reported that integrin α5 contributes to a more malignant phenotype in urothelial bladder cancer (22). In our study, this integrin subunit was overexpressed in most of the tested chemoresistant sublines what might underline the more malignant phenotype of the chemoresistant sublines (Fig. 3).

We observed that most chemoresistant sublines showed a diminished expression of integrin β4 compared to their parental counterparts (Fig. 3). Therefore, a downregulation of integrin β4 could be connected with a more malignant behavior. This is in line with reports that an overexpression of integrin β4 inhibits growth and migration in bladder cancer cell lines and plays an anti-tumoral role (23,24).

In all gemcitabine-resistant and in 3 of 4 cisplatin-resistant sublines, surface expressed integrin β1 was upregulated compared to parental cell lines (Fig. 3). Since chemotaxis was frequently enhanced after acquisition of resistance, these results might support the conclusions of Chakraborty et al (25) who postulated that blockade of β1-integrin with a specific antibody could result in alteration of multiple signaling pathways related to adhesion and migration. Interestingly, Zhang and coworkers showed that they could reverse chemoresistance to mitomycin c by blocking integrin β1 (4). Integrin β1 was overexpressed in most of the tested sublines and it was reported to contribute to a more malignant phenotype in urothelial bladder cancer (4,25). We could confirm in this study that overexpression of integrin β1 is associated with a malignant phenotype since we detected a stronger expression in malignant tissue samples compared to normal urothelium. Nevertheless, there was no different expression comparing low grade with high grade tumors or between non-muscle invasive bladder cancer and muscle invasive bladder cancer (Fig. 4).

To further analyze the role of integrin β1, we suppressed the function of integrin β1 and meassured adhesion and chemotaxis afterwards. There was an influence on adhesion after blocking integrin β1 with a reduced adhesion in 2 of 4 parental and 3 of 4 gemcitabine-resistant sublines. In cisplatin-resistant cells, adhesion was even downregulated in all 4 tested cell lines (Fig. 5).

We could not show an influence on chemotaxis after blocking integrin β1 (Fig. 6). If there is no influence on chemotaxis or if the used transwell migration assay is not able to reflect the impact on chemotaxis is not clear. An explanation for the latter might be that effects that influence invasion after blocking integrin β1 are delayed and therefore not detectable with the used transwell migration assay.

Discussing the role of different integrins and the influence on adhesive and invasive behavior, we should be aware that differentially guided adhesive behavior of different tumor cell lines has been previously observed (21). Blocking integrin β1 inhibited cell-matrix interactions in HCV29 and BC3726 cell lines, whereas binding of the bladder cancer cell lines T24 and Hu456 was enhanced (21). Each cell line therefore may possess a characteristic receptor set and long-term treatment with chemotherapy may influence integrin subfamilies differently. Therefore systematic analysis of cell line panels is fundamental (6).

We did not investigate the relevance of each integrin member in detail that was used here. To provide a complete picture of the role of integrin subtypes in gemcitabine- and cisplatin-resistant bladder cancer ongoing studies are necessary. Particularly, blocking experiments using integrins α3, α5, and β4 would be of interest. In addition, these findings are limited to bladder cancer cell lines and bladder tissue. The role of integrins in chemoresistant bladder cancer should be further evaluated in vivo in an animal model.

Overall, evidence is presented here that acquired resistance to gemcitabine or cisplatin frequently enhances chemotaxis, what might be a surrogate for an increased invasive behavior in chemoresistant bladder cancer cell lines. Since overexpression of integrin β1 seems to be frequently upregulated in chemoresistant urothelial bladder cancer cell lines, further in vivo studies should evaluate downregulation of integrin β1 as a potential therapeutic target especially in chemotherapy refractory cases.

Acknowledgements

The work was supported by the charity Hilfe für krebskranke Kinder Frankfurt e.V., its trust Frankfurter Stiftung für krebskranke Kinder, the Patenschaftsmodell program of the University Hospital Frankfurt and the Kent Cancer Trust.

References