In the present study, a total of 96 patients with pancreatic cancer were enrolled from the First Affiliated Hospital of Wenzhou Medical University (Wenzhou, China) between March 2015 and August 2016. All pancreatic cancer cases, which were confirmed by pathological examination, enrolled in the current study had not received any radiation or chemotherapy before surgery. The demographic and clinical characteristics of the subjects are shown in Table I, including age, sex, smoking, tumor stage, TNM classification, lymph node metastasis, and vascular infiltration. Based on the TNM classification system promulgated by the American Joint Committee on Cancer (AJCC), the pathologic stage was divided into localized and aggressive cancer, as determined by transrectal ultrasound, magnetic resonance imaging (MRI) and emission computed tomography (ECT) (25).

All subjects were informed about the contents of the study and provided their informed consent. This study was approved by the Ethics Committee of Wenzhou Medical University.

Sections, 4-µm-thick, from pancreatic cancer tissues in patients were dewaxed with xylene and hydrated using sequential ethanol (100, 95, 85, and 75%) and distilled water washes. The endogenous peroxidase was blocked with 3% hydrogen peroxide. Antigen retrieval was performed by heating the sections in 0.1% sodium citrate buffer (pH 6.0). The expression of IL-6 (CAS No: ab6672, 1:400; Abcam, Cambridge, MA, USA) was determined using the immunochemical streptavidin-peroxidase method. All samples were semi-quantitatively or quantitatively assessed by two independent investigators in a blinded manner. Slides were examined and images were captured using a DM4000 B LED microscope system (Leica Microsystems GmbH, Hannheim, Germany) and a DFC 420 C 5M digital microscope camera (Leica Microsystems GmbH). For H-score assessment, 10 fields were chosen at random at an ×400 magnification and the staining intensity in the malignant cell nuclei was scored as 0, 1, 2, or 3 corresponding to the presence of negative, weak, intermediate, and strong brown staining, respectively. The total number of cells in each field and the number of cells stained at each intensity were counted. The average percentage of positive cells was calculated and the following formula was applied:

IHA H-score=(% of cells stained at intensity category 1×1)+(% of cells stained at intensity category 2×2)+(% of cells stained at intensity category 3×3) (26).

Human PCCs (PANC-1) were obtained from the American Type Culture Collection (ATCC^® CRL-1469™; Manassas, VA, USA). Trichostatin A (TSA; CAS No: 58880-19-6; Selleckchem, Houston, TX, USA)-induced resistance of PCCs (PANC-1-TSA) were generated in our laboratory. PANC-1 and PANC-1-TSA cells were maintained in DMEM (Gibco; Invitrogen; Thermo Fisher Scientific, Inc., Grand Island, NY, USA) and supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 µg/ml streptomycin (all from Invitrogen; Thermo Fisher Scientific, Inc.) were treated with 5% tyrphostin B42 (CAS No: 133550-30-8; Selleckchem) or recombinant IL-6 (CAS No: 200-06; PeproTech, Inc., Rocky Hill, NJ, USA) for 24 h. Tyrphostin B42a and IL-6 were dissolved in 100% dimethyl sulfoxide (DMSO; Sigma-Aldrich, St. Louis, MO, USA) and diluted with the culture medium for experiments. The final concentration of DMSO for all treatments was maintained at 0.1%.

PANC-1-TSA was obtained by culture of PCCs (PANC-1) in vitro with intermittent increase of the concentration of TSA in the culture for 30 weeks. After cultivating PANC-1 cells with different concentrations of TSA for 1 week, we ascertained the cell death conditions and chose the concentration (0.05 µmol/l) of median lethal dose (LD₈₀) as the initial concentration to cultivate the resistant cell line. When cells grew stably and entered the logarithmic growth phase, the concentration gradually increased at 0.01 µmol/l every 2 weeks. After 15 concentration gradients and 30 weeks of cultivation, the final concentration was stable at 0.2 µmol/l.

Cell suspension (100 µl; 5.0×10³ cells/well) was dispensed in a 96-well plate with the outer wells left empty for the addition of water. The plate was pre-incubated for 24 h in humidified tyrphostin B42 and DMSO was used as a control. A CCK-8 solution (10 µl) was added (Dojindo Molecular Technologies, Inc., Tokyo, Japan) to each well of the plate. The plate was then incubated for 4 h. The absorbance at 450 nm was measured using a microplate reader (Thermo Fisher Scientific, Inc.). Six parallel wells were set up in each experiment, and all of the measurements were repeated at least three times.

Total RNA was isolated using TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.). The concentration and purity of RNA were determined using a spectrophotometer. cDNA was synthesized with reverse transcriptase (Thermo Fisher Scientific, Inc.). qRT-PCR assays were carried out using FastStart Universal SYBR Green Master Mix (Roche Diagnostics GmbH, Mannheim, Germany). PCR parameters were as follows: 95°C for 10 min, then 95°C for 15 sec, and 60°C for 30 sec for 40 cycles. The PCR assays were performed in triplicate and were conducted using a Real-Time PCR Detection System (ABI 7500; Applied Biosystems; Thermo Fisher Scientific, Inc., Carlsbad, CA, USA). The expression of mRNA was calculated using the 2^−ΔΔCt method as previously described (27). Three replicate reactions per sample and endogenous control were used to ensure statistical significance. All the primers that were used in this study are listed in Table II.

Whole-cell protein extracts and nuclear protein extracts from PCCs were prepared with RIPA Lysis Buffer [50 mM Tris-HCl, pH 8.0, with 150 mM sodium chloride, 1.0% Igepal CA-630 [(NP-40]), 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate]) and a Nuclear Extract kit (Active Motif, Carlsbad, CA, USA), respectively, according to the manufacturer's instructions. Protein concentrations were determined using an assay kit (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Lysates containing 100 µg of protein were mixed with loading buffer containing 5% β-mercaptoethanol and heated for 5 min at 100°C. The samples were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes by blotting. The membranes were probed with antibodies JAK2 (D2E12) (1:1,000; rabbit monoclonal; cat. no. 3230), STAT3 (124H6) (1:1,000; mouse monoclonal; cat. no. 9139), and p-STAT3 (D3A7) (1:1,000; rabbit monoclonal; cat. no. 9145; all from Cell Signaling Technolgy, Inc., Danvers, MA, USA) using an enhanced chemiluminescence kit (cat. no. 34095; Thermo Fisher Scientific), and GAPDH (1:5,000; cat. no. AP0063; Bioworld Technology, St. Louis Park, MN, USA) was used as a control.

All in vitro experiments were repeated at least three times to confirm the results. All the statistical calculations were performed using Prism 6.0 (GraphPad Software, Inc., San Diego, CA, USA) and the data were expressed as the mean ± SEM. Statistical significance was determined with the Student's t-test (two-tailed) when comparing two groups of data set. P-values <0.05 were considered to indicate statistical significance. Asterisks shown in the figures indicate significant differences of the experimental groups in comparison with the corresponding control groups.

We first evaluated the activity of IL-6, an upstream inducer of JAK2/STAT3 signaling, in patients with pancreatic cancer. As shown in Fig. 1A and Table I, the expression levels of IL-6 in pancreatic cancer tissues were increased, and positively associated with lymph node metastasis (P=0.027), tumor differentiation (P=0.003) and vascular infiltration (P=0.049) of pancreatic cancer. In pancreatic cancer patients, enhanced IL-6 expression had a high risk of lymph node metastasis [Odds ratio (OR)=4.321; 95% CI: 1.082–17.252; P=0.027; metastasis vs. non-metastasis), poorly differentiated (OR=6.120; 95% CI: 1.999–18.734; P=0.003; (poorly + moderately) vs. mildly)], and perineural vascular infiltration (OR=4.653, 95% CI: 1.001–21.633; P=0.035; perineural vs. adipose), suggesting that IL-6 levels and downstream JAK2/STAT3 signaling may be essential for the development of pancreatic cancer.

We next investigated the activity of IL-6/JAK2/STAT3 signaling in different PCC cell lines. We found overexpression of JAK2 and STAT3, but not the phosphorylation of STAT3, in PCC cell lines including ASPC-1, SW1990, Puta 8988, and PANC-1 (Fig. 1B and C). Among them, PANC-1 cells exhibited the highest expression of the STAT3 protein. In addition, in PANC-1 cells, IL-6 activated JAK2/STAT3 signaling by inducing the expression and phosphorylation of STAT3 (Fig. 1D-F). Similarly, in PANC-1-TSA cells, IL-6 also induced the phosphorylation of STAT3, although the changes of STAT3 expression did not exhibit significance (Fig. 1D-F). These results identified again the crucial role of IL-6/JAK2/STAT3 signaling in pancreatic cancer. Notably, the activity of IL-6/JAK2/STAT3 signaling in PANC-1-TSA cells was higher when compared with those in PANC-1 cells. Moreover, IL-6 treatment exacerbated this difference, as indicated by the phosphorylation, but not the expression, of IL-6/JAK2/STAT3 signaling in PANC-1-TSA cells. Thus, over-activation of IL-6/JAK2/STAT3 signaling in PCC cells may be one explanation for TSA-induced resistance.

Given that TSA triggers the resistance in PCCs by inducing the activation of IL-6/JAK2/STAT3 signaling, we investigated whether TSA exerted similar effects on STAT3-downstream target genes. Thus, the mRNA expression levels of c-Myc, c-Src, HIF-α, and CCND1 in PANC-1-TSA cells were quantified by the qRT-PCR. Enhanced levels of c-Myc and c-Src are required for the proliferation of PCCs, which are regulated by cyclin D1 (encoded by the CCND1 gene), a key molecule involved in regulating cell cycle progression (28–30). As shown in Fig. 2A, the mRNA expression levels of c-Myc, c-Src and CCND1 were markedly increased in PANC-1-TSA cells compared with those in PANC-1 cells (P<0.05), indicating that PANC-1-TSA cells may have a more proliferative activity than PANC-1 cells. In addition, overexpression of HIF-α mRNA was also observed in PANC-1-TSA cells (Fig. 2B). In response to a hypoxic tumor microenvironment, TSA-induced HIF-α expression plays an important role in differentiation, tumor growth and angiogenesis in pancreatic cancer (31).

In addition to the induction of proliferation, TSA also exerted an inhibitory effect on the apoptosis of PCCs. As shown in Fig. 2C, the expression of Bcl-2 mRNA was upregulated in PANC-1-TSA cells whereas the expression of Bax was downregulated compared with that in PANC-1 cells (P<0.05), suggesting that TSA inhibited the apoptosis of PCCs through the mitochondrial pathway.

As aforementioned, PANC-1-TSA cells may possess more proliferative activity than PANC-1 cells. In this case whether tyrphostin B42 improves TSA-mediated resistance in PCCs through the inhibition of proliferation remains unknown. As shown in Fig. 3A, we determined that tyrphostin B42 (30 µM) significantly decreased the number of proliferated PCCs, especially in PANC-1-TSA cells in a time-dependent manner. The results from the CCK-8 assay indicated that the viabilities of PANC-1 and PANC-1-TSA cells were decreased with the treatment of tyrphostin B42 (Fig. 3B, P<0.05). In addition, high-doses of tyrphostin B42 (≥30 µM) had a stronger inhibitory effect on the proliferation of PANC-1-TSA cells compared with PANC-1 cells (Fig. 3B, P<0.05). These results revealed that tyrphostin B42 inhibited the over-proliferative activity of PCCs and improved TSA-mediated resistance.

After TSA treatment, IL-6/JAK2/STAT3 signaling was activated and then induced the over-proliferation of PCCs. In the present experiment, in PANC-1-TSA cells, tyrphostin B42 reduced the expression of STAT3 and inhibited the phosphorylation of p-STAT3 (Fig. 4). Moreover, high-doses of tyrphostin B42 (≥30 µM) exerted stronger inhibitory effects on IL-6/JAK2/STAT3 signaling. Consequently, tyrphostin B42-mediated inhibition of PCC proliferation may be induced through IL-6/JAK2/STAT3 signaling.

In PANC-1 cells, tyrphostin B42 (0–60 µM) did not downregulate the expression of downstream target genes of IL-6/JAK2/STAT3 signaling, including c-Myc, c-Src and CCND1 (Fig. 5A). Similarly, in PANC-1-TSA cells, a low-dose (<30 µM) of tyrphostin B42 did not inhibit but promote the expression of these target genes. However, high-doses (≥30 µM) of tyrphostin B42 downregulated the mRNA expression of c-Myc, c-Src and CCND1. In addition, high-doses of tyrphostin B42 also reduced low-dose-mediated overexpression of HIF-α mRNA in PANC-1-TSA cells (Fig. 5B). Furthermore, decreased expression of Bcl-2 and increased expression of Bax in PANC-1-TSA cells were inhibited with tyrphostin B42 treatment at low doses (<30 µM) (Fig. 5C). These results indicated that tyrphostin B42 at higher doses (≥30 µM) exerted an inhibitory effect on the expression of downstream target genes of IL-6/JAK2/STAT3 signaling, leading to the inhibition of proliferation, the induction of apoptosis, and the reduction of TSA-mediated resistance in PCCs.