Anti-β2-AR (cat. no. ab182136) was purchased from Abcam (Cambridge, UK), anti-interleukin-6 (IL-6; cat. no. 12153), anti-E-cadherin (cat. no. 9101s), anti-vimentin (cat. no. 9102s), anti-Snail (cat. no. 3879), anti-phospho-signal transducer and activator of transcription 3 (Stat3; cat. no. 12640) and anti-GAPDH (cat. no. 2118s) antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). Human IL-6 antibody (cat. no. AF-206-NA) were purchased from R&D Systems, Inc. (Minneapolis, MN, USA). Non-selective β-AR blocker, propranolol hydrochloride (cat. no. 318-98-9), non-specific β-AR agonist, isoproterenol hydrochloride (cat. no. 51-30-9), the β1-AR antagonist, metoprolol tartrate (cat. no. 56392-17-7) and the β2-AR antagonist ICI-118,551 hydrochloride (cat. no. 72795-19-8), were purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Anti-rabbit IgG, HRP-linked secondary antibody (cat. no. 7074s) was purchased from Cell Signaling Technology.

All tissue samples were collected between January 2007 and December 2009, obtained from the First Affiliated Hospital, Sun Yat-Sen University (Guangzhou, China). Patients with TSCC (n=75) were enrolled in this study. Adjacent non-cancerous tissue samples (n=20) and TSCC specimens (n=75) were used for immunohistochemical assessment. All tissue samples used in this study were used following the guidelines set by the Institution Review Board of Sun Yat-Sen University. All patients received radical surgery and none received any form of adjuvant therapy prior to surgery. The tumor extent was classified based on the TNM system by Union for International Cancer Control, and the tumor grade was classified following the World Health Organization classification of histological differentiation (6). Survival was calculated according to the date of surgery and the date of the last follow-up (or death). Informed consent was signed by all patients for the use of their tissue and clinical data in clinical analysis and research studies prior to treatment.

Immunohistochemical staining was performed according to standard protocols (7). Samples were fixed in 4% paraformaldehyde for 24 h at room temperature and embedded in paraffin. Briefly, 4 µm thick tissue sections were dewaxed and hydrated routinely. Following antigen retrieval, tissue sections were incubated with primary antibodies at 4°C overnight. Samples were washed for five times with phosphate-buffered saline, then incubated with secondary antibody (1:200) at room temperature for 30 min. Tissues were treated with 3,3′-diaminobenzidine (1:200) for 1 min at room temperature and counterstained with hematoxylin for 10 sec at room temperature. After dehydration, slides were observed and analyzed using an Axioskop 40 (Carl Zeiss AG, Oberkochen, Germany) microscope. Five fields were randomly selected for each specimen under a light microscope at a magnification of ×400, and three experienced observers analyzed the images. The scoring criteria were according to a standard protocol described previously (7). A final score >4 was classified as high β2-AR expression and a score ≤4 was classified as low β2-AR expression.

Cal27 and SCC15 TSCC cell lines were obtained from American Type Culture Collection (Manassas, VA, USA). Cells were cultured in Dulbecco's modified Eagle's medium-F12 (Gibco: Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (HyClone; GE Healthcare Life Sciences, Logan, UT, USA), 1% penicillin and streptomycin. All cells were incubated at 37°C in a humidified atmosphere containing 5% CO₂ with the culture media changed every 2 days. Cultured cells were treated at 70 to 90% confluence and were serum-starved for 8 h before treatment. Briefly, cells were treated for 48 h according the groups as follows: i) Culture medium (control); ii) 1 µM propranolol (Prop); iii) 1 µM propranolol + 1 µM isoproterenol (Prop + ISO); iv) 1 µM metoprolol + 1 µM isoproterenol (Meto + ISO); v) 1 µM ICI-118,551 + 1 µM isoproterenol (ICI + ISO); and vi) 1 µM ISO. Propranolol, metoprolol and ICI-118,551 were added in the culture media 30 min before adding ISO.

Cancer cells were cultured with medium containing 500 ng/ml IL-6 antibody (R&D Systems, Inc.) to neutralize and block IL-6 signaling in TSCC cells.

Samples were blocked with goat serum (Boster Biological Technology, Wuhan, China) and incubated with the primary antibody [E-cadherin (1:250), vimentin (1:250)] diluted in 5% bovine serum albumin (BSA; Boster Biological Technology) at 4°C overnight. Secondary antibody conjugated to fluorescein isothiocyanate was used to bind to the primary antibody (1:200) diluted in PBS at 37°C for 30 min. Nuclei were stained with DAPI (1:250; Sigma-Aldrich; Merck KGaA) diluted in PBS at 37°C for 5 min. Samples were placed on glass coverslip. Cells were observed using a confocal laser scanning microscope (Zeiss LSM 510 Meta; Carl Zeiss AG).

Cell migration assay was evaluated by using wound healing and Transwell membrane chambers (8 µm pore size; Corning Inc., New York, NY, USA). The wound healing assay was performed as previously described (16). Briefly, the cells were seeded in 12-well plates and treated with ISO at 90% confluence and normal culture medium was added to the control cells. For Transwell membrane chambers, cells were seeded into the upper chambers and cultured with or without ISO for 36 h. For invasion assay, top chambers coated with Matrigel (BD Biosciences, Franklin Lakes, NJ, USA) at a concentration of 1.37 mg/ml. Upper chambers were fixed in 4% paraformaldehyde at room temperature for 30 min, stained using 0.1% crystal violet at room temperature for 5 min, then counted using a Axioskop 40 (Carl Zeiss AG) microscope.

Cells were lysed on ice in 120 µl radioimmunoprecipitation lysis buffer containing phosphatase inhibitors (Sigma-Aldrich; Merck KGaA) for 30 min. Total proteins (30 µg; determined by bicinchoninic acid assay) were loaded and separated by 10% SDS-PAGE, and then transferred into nitro-cellulose membrane (EMD Millipore, Billerica, MA, USA). Membranes were blocked in 5% evaporated skimmed milk at room temperature for 1 h and then incubated with primary antibody at 4°C overnight. Primary antibodies (1:1,000) were diluted in 5% evaporated skimmed milk at 4°C overnight. Then incubated the membrane with horseradish peroxidase-conjugated secondary antibody (1:2,000; diluted in 5% evaporated skimmed milk at room temperature for 1 h; Cell Signaling Technology, Inc.) and observed using chemiluminescence (EMD Millipore).

Data was presented as mean ± SD. All experiments were performed at least three times and all statistical analyses were performed using SPSS 13.0 software for Windows (SPSS, Inc., Chicago, IL, USA). The association between β2-AR expression and the clinical pathological parameters was determined by χ² test. The survival analysis was performed by the Kaplan-Meier method. Student's t-test was used for statistical analysis. P<0.05 was considered to indicate a statistically significant difference. A Cox regression was used to analyze the association between the survival time and the clinicopathological variables.

The expression of p-β2-AR was detected by immunohistochemistry in 75 cases with TSCC. The β2-AR was detected in at low levels in adjacent non-cancerous epithelium (ANCE). In primary tumor samples, predominant cytoplasm and membrane staining of β2-AR was detected in cancer cells (Fig. 1A). The semi-quantitative analysis revealed that β2-AR was overexpressed in TSCC compared with ANCE (Fig. 1B). Enhanced expression of β2-AR was also observed in TSCC with lymph node metastasis compared with TSCC samples without lymph node metastasis (Fig. 1C). β2-AR expression was increased SCC tissue with poor differentiation compared with well differentiated TSCC tissue (Fig. 1D). Among 75 cases examined in the study, 36 (48%) cases exhibited high β2-AR staining and 40 (52%) cases displayed low β2-AR expression.

Associations were tested between β2-AR expression and clinical and pathological features in this TSCC cohort (Table I). χ² analysis demonstrated that β2-AR expression was associated with tumor differentiation (P=0.004) and N stage (P=0.013). There were no associations between β2-AR expression and age, gender, T stage and clinical stage.

Kaplan-Meier analysis demonstrated a statistically significant difference in 5-year survival rate between patients with high β2-AR expression and low β2-AR expression (P=0.003; Fig. 2). As demonstrated in Table II, univariate analysis indicated that tumor size, grade, pathological T stage, lymph node metastasis, clinical stage, and β2-AR were all significant prognostic factors for survival of patients with TSCC. Multivariate analysis showed that lymph node metastasis, clinical stage, and β2-AR were all independent prognostic factors for patient survival.

TSCC cell lines Cal27 and SCC15 were treated with ISO (β-AR agonist) to activate β2-AR signaling. The cell morphology was dramatically changed from an epithelial phenotype to mesenchymal phenotype with decreased E-cadherin expression and increased vimentin expression (Fig. 3A and B). A wound healing assay indicated that these changes were accompanied by enhanced cell motility (Fig. 4A and B). Similar results were also observed in Transwell assay coated with or without Matrigel (Fig. 5A and B). Western blot analyses further confirmed that E-cadherin expression was downregulated and vimentin expression was upregulated by treatment with ISO (Fig. 5C and D). By contrast, western blot analysis indicated that ISO-mediated activation of β2-AR (p-β2-AR), increased expression of vimentin and Snail1, and reduced E-cadherin expression were inhibited by co-treatment with ICI-118,551, but not metoprolol (Fig. 6A and B). The expression of IL-6, Stat3, p-Stat3 and Snail1 were also examined in TSCC cells treated by ISO with or without ICI-118,551. Activation of β2-AR promoted the expression of IL-6, p-Stat3 and Snail1, while blockade of β2-AR signaling with ICI-118,551 inhibited β2-AR-mediated activation of IL-6/Stat3/Snail pathway (Fig. 7A). When cells were treated with anti-IL-6 monoclonal antibody (mAb), ISO-induced EMT was partially reversed by blocking IL-6/Stat3/Snail pathway (Fig. 7B) (24).