Signal transducer and activator of transcription 3 activation is associated with bladder cancer cell growth and survival

Tissue microarray immunohistochemistry indicated that Stat3 phosphorylation was elevated in bladder cancer tissues. Three representative bladder cancer tissues with p-Stat3 positive immunostaining (scale 2–3) are shown (Figure 1B–D), whereas normal bladder tissues were negative or very weak (scale 0–1) with immunostaining (Figure 1A). The elevated p-Stat3 in the bladder cancer tissues was scored and summarized according to immunostaining intensities. Two of the bladder cancer tissues were not included for immunostaining scoring because no staging information is available. Nineteen out of 100 bladder cancer tissues were positive for p-Stat3 immunostaining (scale 2–3).

The clinicopathological data for 102 bladder cancer tissues are classified in Table 1. The samples represented 76 male and 26 female patients, with the majority of patients (95.1%, 97 out of 102) between 41–90 years old. Forty-five tissues and 57 tissues were staged and graded, respectively. Forty of 45 staged tissues (89%) had no signs of regional lymph nodes nor distant organ metastases. According to histological features, the bladder cancer tissues were identified as urothelial carcinoma (85%), squamous cell carcinoma (3%), adenocarcinoma (2%) and mixed carcinomas (10%).

Western blot analysis showed that elevated p-Stat3 was also found in bladder cancer cell lines, UMUC-3, 253J and WH (Figure 2). Very low or no p-Stat3 was detected in the same amount of T24 and TCC cell lysates as revealed by the internal controls of GAPDH. All bladder cancer cell lines examined have very similar total Stat3 expression levels.

One way to investigate the functions of activation of Stat3 in bladder cancer is to interrupt Stat3 signaling pathway in bladder cancer cell lines with elevated p-Stat3. To that end, WH and UMUC-3 were transduced with rAd/dnStat3. The two cell lines were infected with rAd/dnStat3 (moi = 100, 250, and 500). The expression of dnStat3 in WH cells was observed and shown at 48 hours post infection (Figure 3). FLAG-tagged dnStat3 expressions in 10 μg of bladder cancer cell lysates were detected by western blots using an anti-FLAG antibody. Total Stat3 expressions correspondingly reflected the dose-dependent increase of dnStat3 expressions in these bladder cancer cells.

Bladder cancer cell growth was significantly suppressed in the presence of dnStat3 and STA-21. UMUC-3 and WH cells were transduced with either rAd/dnStat3 or rAd/eGFP (moi = 100 and 500). Cell growth is presented in cell densities normalized to untransduced controls at day 2 and day 4 post-infection (Figure 4A and 4B). The growth rates of untransduced cells were set at 100%. There was 20–30% reduction in UMUC-3 cell growth at day 4 post-infection with rAd/eGFP; more notably, the UMUC-3 cell line transduced with rAd/dnStat3 (moi = 100 and 500) was reduced down to 11.9 ± 0.3% (P < 0.05) and 1.3 ± 0.3% (P < 0.005) of cell growth at day 4 compared to untransduced controls (Figure 4A). The discrepancies between cell growth of bladder cancer cells transduced with rAd/eGFP and rAd/dnStat3 apparently reflects dnStat3-specific inhibitory effects. The dnStat3-mediated inhibition of cell growth was even more dramatic as observed in WH cells (Figure 4B). WH cells transduced with rAd/dnStat3 (moi = 100 and 500) had only 50.2 ± 13.5% (P < 0.05) and 16.1 ± 2.2% (P < 0.005) of cell growth compared to untransduced controls at day 2 post-infection; evenmore, at day 4 post-infection, cell growth of rAd/dnSTat3-transduced WH cells was decreased to only 1.1% and 0.5% of untransduced controls. After 4 day of infection with a higher dose of rAd/Stat3 (moi = 1000), a cell viability assay (MTT) also revealed that WH and UMUC with higher content of p-Stat3 only maintained 34.8 ± 1.4% and 51.9 ± 8.5% cell viability of untreated controls (Figure 4C). On the contrary, TCC and T24 with much lower or undetectable p-Stat3 contents had much higher cell viability with 90.5 ± 3.6% and 73.6 ± 7.5% of untreated controls under the same experimental conditions. The overall cell viability of cells treated with rAd/dnStat3 was decreased along the time, while that of cells untransduced or treated with rAd/eGFP was increased (data not shown).

As MTT assay showed, similar inhibition on cell growth and viability was observed in bladder cancer cells treated with 30 μM STA-21 (Figure 4C). Viability of bladder cancer cells and BdSMC treated with DMSO was about the same as untreated controls. However, exposure to STA-21 greatly reduced cell viability of 253J, UMUC3 and WH (38.1 ± 0.74%, 11.4 ± 1.5%, and 29.0 ± 6.7%). Interestingly, STA-21 had very minimal effects on BdSMC cell viability (91.3 ± 4.4%). This indicated that STA-21 inhibition is specific to bladder cancer cells that have constitutive activation of Stat3. In addition, the decreased overall viability of cells treated with STA-21 was consistent with that observed in cells treated with rAd/dnStat3. These together suggested that bladder cancer cell death associated with inhibition of Stat3 pathway might have occurred.

Transduction of dnStat3 in bladder cancer cells induced activation of apoptotic caspases 3, 8, and 9 in those cells transduced with rAd/dnStat3. UMUC-3 and WH were fixed at day 2 and day 1, respectively, post-transduction of rAd/eGFP or rAd/dnStat3 and then subject to immuno-fluorescent staining using antibodies that recognize cleaved caspases 3, 8 and 9 for apoptosis evaluations. Only rare sporadic cells were found stained by anti-cleaved caspases 3, 8, and 9 antibodies in negative UMUC-3 and WH controls (untransduced or transduced with rAd/eGFP) (Figure 5A &5B). However, many dnStat3-transduced cells were positive in anti-cleaved caspases 3 (44.2 and 53%), 8 (53.5 and 74.4%), and 9 (74.2 and 42.4%) immunostaining (Figure 5B).

Targeting Stat3 pathway using STA-21 also led to increased cleavage of caspase 3 after 72 hours of treatment in 253J, UMUC-3 and WH (Figure 5C and data not shown). About 6.4–13% cells were cleaved caspase 3 positive in these three STA-21-treated bladder cancer cell lines, compared to less than 0.5% in untreated or DMSO-treated cells. STAT-21 treatment did not seem to induce caspase 3 cleavage in BdSMC cells with less than 0.1% appearing positive with anti cleaved caspase 3 immunoreactivity. The increased apoptotic caspase activation implicated that apoptosis could be one of mechanisms underlying the decreased cell viability in bladder cancer cells in that Stat3 pathway was compromised by treatment of rAd/Stat3 or STA-21.

We then would like to explore possible mechanisms for dnStat3-induced cell growth inhibition and activation of apoptotic caspases in bladder cancer cells. It is likely that inhibition of Stat3 pathway by dnStat3 or STA-21 down regulated Bcl-2, Bcl-xL and survivin as well as cell cycle regulating gene, cyclin D1. Reduced expressions of these genes may contribute to bladder cancer cell growth inhibition and apoptotic caspase activation. UMUC-3 and WH cells were transduced with either rAd/eGFP or rAd/dnStat3 (moi = 1000 and 500) for 27 hours and 48 hours, respectively. Expression of survivin, Mcl-1, Bcl-2, Bcl-xL and cyclin D1 proteins in these cells were evaluated using western blot analysis and densitometric quantification. In UMUC-3 cells, Bcl-2 (47% untreated control), Bcl-xL (55.4%) and survivin (9.7%) were down regulated by the transduction of dnStat3 as compared to expressions in untransduced cells or cells transduced with rAd/eGFP (Figure 6A), whereas Mcl-1 (100%) remained intact. Expression of the three genes in rAd/eGFP-transduced cells (moi = 1000) was only slightly changed as compared to the untransduced cells. survivin(67.1% untreated control), Bcl-xL (61.1%) and cyclin -D1 (62.7%) expression was detected to be decreased when the same cell line treated with 30 μM STA-21 for 4 days (Figure 6C).

We also examined the expression of anti-apoptotic genes and cyclin D1 in WH bladder cancer cells when Stat3 pathway was targeted by rAd/dnStat3 or STA-21. Mcl-1 (80% untreated control) was apparently not affected as compared with the control treated with rAd/eGFP but Bcl-2 (16.6%) and Bcl-xL (46.9%) protein expressions were down regulated by the transduction of dnStat3 (Figure 6B). In addition, cyclin D1 expression (1.9% untreated control) was almost completely inhibited by the interference of Stat3 signaling pathway. Expression reduction in two anti-apoptosis genes (Bcl-2 and Bcl-xL) and cyclin D1 were consistent with apoptosis and cell growth inhibition in WH cells. Four days of treatment of STA-21 showed similar but less reduction in survivin (19.6% untreated control), Bcl-2 (64%) and Bcl-xL (79.2%) expressions.

Bladder cancer cell lines were purchased from American Type Culture Collection (ATCC). Cell lines were maintained in 1× DMEM supplemented with 10% fetal bovine serum and 100 U/ml penicilin/streptomycin/amphotericin B (Mediatech, Herndon, VA) at 37°C, aired with 5% CO₂. Bladder smooth muscle cells (BdSMC) were purchased from Cambrex Bio Science and maintained in SmGM^®-2-Smooth muscle medium (Cambrex, Chicago, IL) supplemented with 5% FBS.

Stat3 phosphorylation status in bladder cancer tissues were examined using immunohistochemistry with a p-Stat3 (Y705)-specific monoclonal antibody (Cell Signaling Tech., Danvers, MA). We stained bladder cancer tissue samples (n = 102) on tissue microarray slides from two different providers (US Biomax, Inc., Rockville, MD and ISU ABXIS Co., Seoul, Korea). The immunohistochemistry and scoring of p-Stat3 expression were described previously [49]. Most of the p-Stat3 positive cancer tissues showed staining in greater than 50% of each sample.

Western blots were carried out according to protocols described previously [49]. 10 to 100 μg of cellular proteins were resolved on 10% or 14% SDS-PAGE gels before transfer, immunoblotting, and visualization of specific protein bands. Antibodies were purchased separately and used to recognize FLAG (Sigma, St. Louis, MO) GAPDH (Chemicon International, Temecula, CA). Stat3, p-Stat3 (Y705) (Cell Signaling Tech., Danvers, MA), Bcl-2, Bcl-x_L (Biosciences, Inc. Franklin Lakes, NJ,), Mcl-1, cyclin D1 (Lab Vision Corp., Fremont, CA) and survivin (UpState, Charlotteville, VG)

For expression comparison, each protein expression was presented in a percentage of its corresponding untreated control after densitometric quantification and normalization to the GAPDH expression. A representative one from duplicated experiments was presented.

The construction of recombinant Adenovirus/CMV-dnStat3 Y705F (rAd/dnStat3) is described previously [23]. DnStat3 was generated from Stat3 by changing the tyrosine at position 705 into phenylalanine. The dnStat3 protein product is tagged with a 6-repeat FLAG sequence for detection and cannot be activated through tyrosine phosphorylation. About 2 × 10⁵ WH and UMUC-3 cells were transduced with rAd/dnStat3 or rAd/CMV-eGFP (rAd/eGFP) (Applied Viromics, Fremont, CA) at variant multiplicities of infection (moi) based on TCID50 assay using 293T cells. For cell growth experiments, cell numbers were enumerated at day 2 or 4 post-infection. Cell counts in 5 random fields of view (magnification 100×) were obtained for each treatment and control. Cell growth rates were presented in percentages of cell density of untreated controls and averaged from triplicate experiments.

Approximately 5000 cells were grown in 100 μl 10% FBS-supplemented DMEM medium in 96-well flat-bottomed plates overnight. Cells were exposed to STA-21 (30 μM) that was dissolved in dimethyl sulfoxide (DMSO) before added to the medium. Cell viability was analyzed by the MTT [3-(4, 5-dimethyl-2-thiazolyl)-2, 5-diphenyl-2H-tetrazolium bromide] (Sigma) assay in three replicates. At the time of assay end-point, cells were treated with MTT (1 mg/ml) for 3–4 hours. Colormetric quntitation was determined by an EL808 Ultra Microplate Reader (Bio-Tek Intruments, Inc) after formazan was dissolved in 25% N, N-dimethylformamide and 10% SDS in a light-proof condition overnight.

Approximately 1 or 2 × 10⁵ cells (UMUC-3, WH, 253J, and BdSMC) were seeded on sterile coverslips in a 6-well plate overnight, either transduced by either rAd/eGFP or rAd/dnStat3 (moi = 500) for 48 or 28 hours respectively, for UMUC-3 and WH cells. For the small molecule inhibition, cells were treated with 30 μM STA-21 for 72 hours. Cleaved caspase immunostaining and documentation were described previously [50]. The primary rabbit antibodies were diluted with 1:100, 1:50, and 1:100 dilutions, respectively, for detecting cleaved-caspase-3 (Asp175), cleaved-caspase-8 (Asp374), or cleaved-caspase-9 (Asp330) (Cell Signaling Tech)