The following items were purchased: anti-STAT5a (sc1081) rabbit polyclonal antibodies, anti-STAT5b (sc1656) mouse monoclonal antibodies, and anti-phosphotyrosine (sc7020) mouse monoclonal antibodies from Santa Cruz Biotechnology (Santa Cruz, CA, USA); Alexa Fluor 488 goat anti-mouse IgG (A-11001), Hoechst 33258, pentahydrate (bisbenzimide), and anti-V5 Tag (#46-0705) mouse monoclonal antibodies from Invitrogen (Life Technologies; Carlsbad, CA, USA); anti-lamin B1 (ab16048) rabbit polyclonal antibodies from Abcam plc (Cambridge, UK); anti-PARP (#9542) rabbit polyclonal antibodies, anti-cleaved caspase-3 (#9664) rabbit monoclonal antibodies, and anti-Bcl-xL (#2746) rabbit monoclonal antibodies from Cell Signaling Technology (Danvers, MA, USA); and anti-β-actin (A1978) mouse monoclonal antibodies from Sigma-Aldrich (St. Louis, MO, USA).

The following items were purchased: fetal bovine serum (FBS), DMEM, RPMI, MEM and trypsin solution from Gibco (Life Technologies); RNeasy Mini kit from Qiagen (Hilden, Germany); High Capacity RNA-to-cDNA kit from Applied Biosystems (Darmstadt, Germany); NE-PER Nuclear and Cytoplasmic Extraction reagents and SuperSignal West Pico and Femto chemiluminescent substrates from Thermo Fisher Scientific (Rockford, IL, USA); FuGENE HD Transfection reagent from Promega (Madison, WI, USA); G418 (Geneticin) from Roche Diagnostics Deutschland GmbH (Mannheim, Germany); laminin, collagen from human placenta type IV [(collagen IV) from Sigma-Aldrich]; Cell Counting kit-8 from Dojindo Laboratories (Kumamoto, Japan); BioCoat Matrigel Invasion Chamber (8-µm pore size) and fibronectin from BD Biosciences (Franklin Lakes, NJ, USA); gemcitabine hydrochloride from Wako Pure Chemical Industries (Osaka, Japan); Fluorescent Mounting Medium from Dako Japan (Tokyo, Japan); Histofine Simple Stain Max PO (M) or (R) kit from Nichirei Biosciences, Inc., (Tokyo, Japan); and recombinant human epidermal growth factor (EGF, 236-EG) and recombinant human platelet-derived growth factor (PDGF, 220-BB) from R&D Systems (Minneapolis, MN, USA).

AsPC-1, BxPC3, Capan-1, HPAF-2, MIA PaCa-2, PANC-1 and SW1990 PDAC cell lines were purchased from the American Type Culture Collection and the PK-45H cell line was obtained from the Cell Resource Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University.

PANC-1, MIA PaCa-2 and SW1990 cells were grown in DMEM, AsPC-1, Capan-1, BxPC3 and PK-45H cells were grown in RPMI, and HPAF-2 cells were grown in MEM. All of the cultures were supplemented with 10% FBS, 100 units/ml penicillin, 100 µg/ml streptomycin and 250 ng/ml amphotericin B. Cells were maintained at 37°C in a humid atmosphere with 5% CO₂.

To evaluate the STAT5b expression levels in the pancreatic cancer cell lines, TaqMan quantitative RT-PCR was performed. Total RNA was isolated with RNeasy Mini kit according to the manufacturer's protocol. For cDNA synthesis, the High Capacity RNA-to-cDNA kit was used according to the manufacturer's protocol. Quantitative RT-PCR was performed using the 7500 Fast Real-Time PCR System (Applied Biosystems, Darmstadt, Germany) and for evaluation, commercial TaqMan^® Gene Expression- assays (Applied Biosystems, AoD, Assay-ID: Hs00273500_m1 for the studied genes STAT5b and Hs02758991_g1 for GAPDH as the reference gene) were used with optimized primer and probe concentrations.

To extract protein from cell lines, cells were washed twice with cold PBS and reacted with lysis buffer Triton X-100 at 4°C for 30 min. After the reaction, cells were harvested by cell scraping and centrifuged at 4°C and 15,000 rpm for 30 min. For western blot analysis, 20 µl of each protein was separated by gel electrophoresis on a polyacrylamide gel and transferred to nitrocellulose membranes. The membranes were blocked with Tris-buffered saline containing 0.1% Tween-20 (TBST) and 5% skim milk, then incubated overnight at 4°C with the indicated primary antibodies and then for 60 min with the corresponding horseradish-conjugated secondary antibodies. The membranes were washed twice with TBST and incubated with goat anti-mouse or anti-rabbit IgG-horseradish peroxidase conjugated secondary antibodies for 1 h at 4°C. The blots were washed three times with TBST and an enhanced chemiluminescence system (ImageQuant LAS 4000 mini; GE Healthcare Japan, Tokyo, Japan) was used to detect the bands. The band densities of PARP and Bcl-xL were measured with ImageJ (National Institutes of Health, Bethesda, MD, USA), and densitometry analysis was performed as previously described (24).

Cells were plated onto chamber slides and allowed to adhere overnight. Slides were fixed for 10 min in 2% paraformaldehyde and free aldehydes were quenched with 50 mM/l NH₄Cl in PBS for 10 min. Then, the cells were permeabilized in 0.1% Triton X-100 in PBS-2% BSA for 15 min. Slides were then incubated at 23°C for 1 h with anti-STAT5b antibody (in 1:50 dilution). After washing with PBS, slides were incubated for 30 min with Alexa Fluor 488-conjugated goat anti-mouse secondary antibody (in 1:500 dilution), and the nuclei were counterstained with 0.2 µg/ml Hoechst 33258 dye for 1 min at 23°C, as per a previously reported method (25). Slides were next washed 3 times in PBS and then mounted in a fluorescent mounting medium. Immunofluorescence scans were taken with a Nikon TE2000-E inverted fluorescence microscope system equipped with a Nikon Digital Eclipse C1 laser scanning confocal microscope (Nikon, Tokyo, Japan).

Cell fractionation was carried out utilizing NE-PER Nuclear and Cytoplasmic Extraction reagents according to the manufacturer's protocol. Preparation of cell lysates and western blotting were performed as described above with primary antibodies against STAT5b (in 1:500 dilution) and against lamin B1 (1:10,000 dilution).

For immunoprecipitation, a total of 1 mg of cell lysate was incubated overnight at 4°C against STAT5b antibody followed by 2-h incubation with protein A/G plus agarose, as per a previously reported method (23). The beads were then washed three times with TBST, then boiled and subjected to western blotting as described above with primary antibodies against phosphotyrosine (1:500 dilution) and STAT5b (1:500 dilution).

To construct expression vectors for human STAT5b short hairpin (sh) RNA containing the sense target sequence for STAT5b (5′-GGA CAC AGA GAA TGA GTT A-3′), the antisense target sequence was synthesized and inserted into a pBAsi-hU6 Neo DNA vector. Likewise, the scrambled sequence (5′-TCT TAA TCG CGT ATA AGG C-3′) was used to construct the sham vectors that served as negative controls. PANC-1 cells (2×10⁶ cells/well) were plated on a 10-cm plate and grown in DMEN medium at 37°C. Transfection of STAT5b shRNA expression vectors and sham vectors was performed using FuGENE HD Transfection reagent according to the manufacturer's instructions, whereby 2×10⁶ cells/well were transfected with 19 µg of DNA using FuGENE HD and cells were passaged and cultured with 500 µg/ml of G418. Cell lysates were collected and protein levels were measured by western blotting in the same manner as described above. In order to construct the STAT5b overexpression clones, the STAT5b full-length cDNA was subcloned into pcDNA3.1/V5-His TOPO vector, which expresses a V5 tag fusion protein. Authenticity was confirmed by sequencing. AsPC-1 cells (2×10⁶ cells/well) were transfected with FuGENE HD Transfection reagent and stably transfected clones were selected with 800 µg/ml of G418 in RPMI. Individual clones were isolated, and expression of the V5-tagged STAT5b protein was determined by western blotting with anti-STAT5b and anti-V5 Tag antibodies as described above.

To assess cell proliferation, cells were seeded in a 96-well plate at a density of 8.0×10³ cells/well in DMEN and incubated at 37°C for 24, 48 or 72 h. After the indicated time, the medium was replaced with fresh DMEN and 10 µl of Cell Counting kit-8 was added to the wells. The plate was incubated for 2 h at 37°C followed by an absorbance measurement of the wells at 450 nm using a microtiter plate reader. Analysis was performed in triplicate.

To assess cell proliferation after treatment with the antineoplastic agent, 8.0×10³ cells were seeded in a 96-well plate in DMEN and incubated at 37°C for 24 h similarly to the cell proliferation assay. After incubation, the medium was replaced with 100 µl of 0.1% BSA containing 0, 10, 100 or 1,000 µM of gemcitabine reagent and incubated at 37°C for 48 or 72 h. After treatment with gemcitabine for the indicated time, the medium was changed to 10 µl of Cell Counting kit-8. Absorbance measurements at 450 nm were taken after 2 h of incubation with a microtiter plate reader. Analysis was performed in triplicate.

PARP and cleavage of caspase-3 were monitored to assess apoptosis. After treatment with absence or presence of 1,000 µM gemcitabine for 72 h at 37°C, protein extraction and western blotting (as described above) were performed with antibodies against PARP (in 1:1,000 dilution), cleaved caspase-3 (in 1:1,000 dilution) and β-actin (in 1:10,000 dilution).

Adhesion assay was performed as per a previously reported method with some modification (26). Cells were suspended in serum-free medium containing 0.1% BSA and were seeded at a density of 3×10⁴ cells/plate on non-adhesive NUNC 96-well culture plates (Thermo Fisher Scientific) coated with fibronectin, laminin or collagen IV. The cells were then incubated for 3 h at 37°C and washed with PBS. Adherent cells were fixed in 50 µl of 96% ethanol for 10 min, stained with 50 µl of 0.1% crystal violet, rinsed with water and dried at room temperature for 30 min. Stained cells were solubilized with 50 µl of 0.2% Triton X-100 and absorbance was measured at 595 nm with a microtiter plate reader. Adhesion assays were performed in triplicate.

The invasiveness of stable transfected PANC-1 cells was measured as per a previously reported method with some modification (27). Cells (2.5×10⁴ in quantity) were suspended in 500 µl of serum-free medium [0.1% bovine serum albumin (BSA)] and placed onto the upper compartment of Matrigel-coated Transwell chambers (8-µm pore size, BioCoat Matrigel Invasion Chambers; Becton-Dickinson Labware). The lower compartment was filled with 750 µl of medium containing 1% FBS, 1 nM EGF or 1 nM PDGF. After 20 h, cells on the upper surface of the filter were carefully removed with a cotton swab and membranes were fixed in methanol and stained with crystal violet. The cells that had migrated through the membrane to the lower surface of the filter were counted using a microscope. Each assay was performed in duplicate and experiments were repeated three times.

Tissues from 44 patients with invasive PDAC were obtained for the present study. The patients received treatment at Nippon Medical School Hospital (Tokyo, Japan) from 2009 to 2013. All patients underwent either a pancreatoduodenectomy or a distal pancreatectomy, and received adjuvant or neoadjuvant chemotherapy with gemcitabine alone or gemcitabine combination regimen. Patients were 27 males and 17 females, and median age was 69 years (range, 37–88 years). Clinicopathological stage was determined according to the TNM classification system of the International Union Against Cancer (UICC) and additionally characterized according to the Japan Pancreas Society classification (Table I). The median follow-up period was 21 months. Paraffin-embedded samples were prepared for immunohistochemical analysis as previously described (28). This study was carried out in accordance with the principles embodied in the Declaration of Helsinki, 2008, and informed consent for the use of pancreatic tissues was obtained from each patient.

Paraffin-embedded sections (3 µm) were immunostained using a Histofine Simple Stain MAX PO (R) or (M) kit. After deparaffinization, the tissue sections were preheated in 10 mM citrate buffer solution (pH 6.0) for 5 min at 121°C. Then, endogenous peroxidase activity was blocked by incubation for 30 min with 0.3% hydrogen peroxide in methanol. The tissue sections were then incubated with the anti-STAT5b antibody (1:50 in dilution) in phosphate-buffered saline (PBS) containing 1% bovine serum albumin (BSA) overnight at 4°C. Bound antibodies were detected with the Simple Stain MAX PO (R) or (M) reagent, using diaminobenzidine tetrahydrochloride as the substrate. The sections were then counterstained with Mayer's hematoxylin. Negative control tissue sections were prepared by omitting the primary antibody. As for the evaluation of immunostaining, both intensity and proportion of positively stained cancer cells were analyzed at magnification ×200.

Student's t-test was used for statistical analysis of the cell proliferation assay, sensitivity of antineoplastic agents assay, adhesion assay, invasion assay and densitometry analysis. χ² test and Fisher's exact test were used to analyze the correlation between STAT5b expression and clinicopathological characteristics. Cumulative survival rates were calculated by the Kaplan-Meier method. Significant difference was set at P<0.05. All statistical tests were determined using the Software Package SPSS for Windows (version 12.0; SPSS, Inc., Chicago, IL, USA).

We performed qRT-PCR analysis to investigate the relative expression levels of STAT5b mRNA in pancreatic cancer cell lines, including AsPC-1, BxPC3, Capan-1, HPAF-2 MIA PaCa-2, PANC-1, PK-45H and SW1990. STAT5b mRNA was detected in all pancreatic cancer cell lines, and the relative levels were highest in PANC-1 cells, second highest in Capan-1 cells and lowest in PK-45H cells. In PANC-1 cells, STAT5b mRNA levels were 13.5-fold higher than in PK-45H cells (Fig. 1A). Western blot analysis was also performed to examine the STAT5b protein levels with the same eight pancreatic cancer cell lines. Consistent with our qRT-PCR results, STAT5b protein was detected in all cell lines, with the highest protein levels in PANC-1 cells and second highest levels in Capan-1 cells (Fig. 1B).

Confocal immunofluorescence microscopy analysis was next carried out to determine the subcellular localization of STAT5b in PANC-1 and MIA PaCa-2 cells. This analysis revealed that STAT5b was distributed in the nuclei and cytoplasm in both cell types (Fig. 2A). Similarly, western blotting using nuclear and cytoplasmic fractions detected STAT5b protein in both the nuclei and the cytoplasm (Fig. 2B). To investigate the activation of STAT5b in the two cell types, the cell lysates were immunoprecipitated with anti-STAT5b antibody followed by western blotting with anti-phosphotyrosine or anti-STAT5b antibodies. This analysis revealed the phosphorylation of STAT5b in the same two pancreatic cancer cell types (Fig. 2C).

PANC-1 cells are ideal for assessing the role of endogenous STAT5b in pancreatic cancer cells because they express relatively high levels of STAT5b protein. Therefore, PANC-1 cells were stably transfected with a plasmid vector encoding shRNAs targeting the STAT5b transcripts. Clones transfected with the STAT5b shRNAs showed inhibition of STAT5b, and uninhibited STAT5a protein expression by western blotting (Fig. 3A). To assess the consequence of reduced STAT5b expression on proliferation ability, cell growth was investigated by cell proliferation assay. There was no significant difference between the sham-transfected cells and the STAT5b shRNA clones (Fig. 3B).

The chemotherapeutic agent gemcitabine has been the standard treatment in patients with PDAC for more than a decade since Burris et al (4). We therefore, compared the effects of gemcitabine on growth and apoptosis in sham-transfected and STAT5b shRNA clones by cell proliferation assay as described above to investigate the effect of gemcitabine treatment on proliferation ability with STAT5b suppression. Treatment with gemcitabine for 48 or 72 h resulted in a dose-dependent reduction in cell growth. In the sham-transfected cells, a maximum decrease of 19.4% occurred at a concentration of 1,000 µM gemcitabine and 72-h incubation. By contrast, both STAT5b shRNA clones exhibited a significantly greater growth inhibitory effect of 32.6 and 52%, respectively with the same concentration of gemcitabine and incubation time (Fig. 4).

We next investigated the effects of gemcitabine on growth in PANC-1, AsPC-1 and BxPC cells by cell proliferation assay as described above, in order to compare the effect of gemcitabine treatment on proliferation of pancreatic cancer cells which express high and low levels of STAT5b. Treatment with gemcitabine for 72 h resulted in a dose-dependent reduction in cell growth. In PANC-1 cells which expressed relatively high levels of STAT5b, a maximum decrease of 22.1% occurred at a concentration of 1,000 µM gemcitabine and 72-h incubation. Both AsPC-1 and BxPC3 cells, which express relatively low levels of STAT5b, exhibited a significant decrease in growth by treatment with gemcitabine compared to PANC-1 cells for 72 h (Fig. 5).

AsPC-1 cells, which express relatively low levels of STAT5b protein, were stably transfected with a plasmid vector encoding STAT5b full-length cDNA, in order to investigate and compare the role of STAT5b in pancreatic cancer cells. Clones transfected with the STAT5b cDNA exhibited overexpression of STAT5b, verified by western blotting (Fig. 6A). To assess the effect of overexpressed STAT5b expression on proliferation ability, cell growth was measured by cell proliferation assay. No significant difference between the sham-transfected cells and the STAT5b overexpressing clones was observed (Fig. 6B). We performed cell proliferation assay with gemcitabine treatment, as described above, to determine the effect of gemcitabine treatment on proliferation ability with STAT5b overexpression clones. Treatment with gemcitabine for 72 h resulted in a significant increase in growth of overexpressed clones compared to sham-transfected cells (Fig. 6C).

The deregulation of apoptosis (29) is an indicator of carcinogenesis and the induction of apoptosis is a standard strategy for anticancer therapies (30). Therefore, we examined the expression of known apoptosis-related proteins, cleaved caspase-3 and PARP (31), by western blotting to investigate whether gemcitabine treatment for STAT5b shRNA clones induced greater apoptotic actions than the sham-transfected cells. After treatment with gemcitabine for 72 h, both STAT5b shRNA clones exhibited a stronger expression of cleaved caspase-3 and PARP than the sham-transfected cells (Fig. 7A). STAT5 is known to upregulate Bcl-xL expression and STAT5 has been implicated in the regulation of apoptosis (32). Therefore, we next sought to determine whether the levels of these transcription factors were altered in our cells. Gemcitabine markedly reduced Bcl-xL levels in the STAT5b shRNA cells compared with the sham-transfected cells (Fig. 7B). Relative values of PARP and Bcl-xL proteins were calculated by quantitative densitometry. Band densities of PARP protein significantly increased in STAT5b shRNA clones compared to sham-transfected cells (Fig. 7C). Band densities of Bcl-xL protein significantly decreased in STAT5b shRNA clones compared to sham-transfected cells (Fig. 7D).

Cancer cell adhesion is a process that involves cancer cells interacting with adjacent cancer cells as well as with extracellular matrix components (33). For tumors to metastasize and grow, neoplastic and endothelial cells must invade into surrounding tissues. The ability to block the invasive capacity of tumor cells therefore offers a new approach to treating patients with malignant disease (34). Hence, we next compared the effects of STAT5b suppression on cell adhesion and invasion. In comparison to sham-transfected cells, the adhesion assay for both STAT5b shRNA clones exhibited a significantly reduced adhesion to fibronectin (P<0.001), laminin (P<0.001), and collagen IV (P<0.002), which are known as major types of extracellular matrices (ECM) (Fig. 8A). Similarly, compared to sham-transfected cells, both clones exhibited significantly reduced ability to invade across a Matrigel membrane treated with FBS (P<0.02), EGF (P<0.01) and PDGF (P<0.02) in the invasion assay (Fig. 8B).

Immunohistochemical results for STAT5b were evaluated as the expression index; 'staining percentage of the tumor cell' multiplied by 'staining intensity (0–2)'. We set 0–20 index as weak (29 cases) (Fig. 9A), and 30–120 index as strong (15 cases) (Fig. 9B). However, there was a trend toward reduced overall survival of patients in the STAT5b-strong group, but a significant difference in the overall survival rate between the STAT5b-weak group and STAT5b-strong group was not observed (P=0.35) (Fig. 9C). Clinicopathologically, a significant correlation between STAT5b expression in the cancer cells and main pancreatic duct invasion (P=0.014). (Table I) was observed.