This study was approved by the Institutional Bioethics Committee of China Medical University (Shenyang, China) and informed consent was obtained from all participating individuals. A total of 51 pairs of samples, including a tumor sample (T) and a matched, cancer-free peritumor tissue sample (N), were examined in this study. All tissues were obtained from patients diagnosed with laryngeal carcinoma and receiving surgery (including partial laryngectomy in 29 patients and total laryngectomy in 22) at the First Hospital of China Medical University from 2011 to 2012. Immediately after surgical removal from the patients, 24 pairs of tissue samples were snap-frozen in liquid nitrogen and then transferred to −80°C until further use (for western immunoblot); another 27 pairs of tissue samples were fixed in formalin before being embedded into paraffin and processed into 4-μm thick sections (for immunohistochemistry).

Human laryngeal carcinoma cell line, Hep-2, was purchased from the Cell Bank, Chinese Academy of Sciences (Shanghai, China) and cultured in RPMI-1640 medium (Gibco, Carlsbad, CA, USA) containing 10% fetal bovine serum (FBS; HyClone, Logan, UT, USA).

IHC staining was performed using the IHC staining kit (Maxim, Fuzhou, China), following the manufacturer's instructions. Briefly, the tissue sections were blocked with H₂O₂ (to block endogenous peroxidase) followed by nonimmune serum (reagent A) for 20 min, and then incubated with mouse anti-human EMP1 polyclonal antibody (Abnova, Taiwan, China) at 4°C overnight. After being washed with phosphate-buffered saline (PBS) 3 times, the sections were incubated with biotinylated secondary antibody (reagent B) at room temperature for 15 min, followed by streptavidin-horseradish peroxidase (reagent C) for 15 min. The signals were detected using 3,3′-diaminobenzidine tetrahydrochloride (DAB; Maxim) and counter stained with hematoxylin.

The tissue sections were scored by two independent pathologists blinded to the study, as detailed below. The 'positive rate' (P) was defined as follows: 0 means <10% of the tissues showed positive staining; 1, 10–25%; 2, 26–50%; 3, 51–75%; and 4, 76–100%. The 'intensity of staining' (I) was defined as: 0 indicated negative staining; 1, light yellow; 2, brown; and 3, dark brown. The total staining score (S) was calculated as P × I.

Total RNA was isolated from frozen tissues (~100 mg each) using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instructions. Reverse transcription into cDNA was then performed on total RNA, using the Reverse Transcription system (A3500; Promega, Madison, WI, USA) per the manufacturer's instructions. Real-time PCR was performed using the following primers: EMP1 forward, 5′-AGGGAATACATGGTTTACTCCA-3′ and reverse, 5′-AGAGAGATTGGCCAGCAAAA-3′; GAPDH (internal control), forward, 5′-ACGGATTTGGTCGTATTGGG-3′, reverse, 5′-CGCTCCTGGAAGATGGTGAT-3′. The thermal cycling conditions were 95°C for 30 sec, followed by 40 cycles of 95°C for 5 sec and 60°C for 34 sec.

To measure the protein level of EMP1 in tissues, 24 pairs of samples (~200 mg each) were individually packaged in sterilized tinfoil, frozen in liquid nitrogen for 10 min, and broken into powder with a pulverizer. The tissue powder was then homogenized in PBS at 0°C followed by digestion in lysis buffer on ice for 30 min. After centrifugation at 14,000 × g for 5 min, protein-containing supernatant was collected and the protein concentration was measured using the BCA protein assay kit (Beyotime, Jiangsu, China), according to the manufacturer's instructions. A total of 80 μg protein from each sample was separated on SDS-PAGE gel, transferred onto PVDF membrane, blocked in Tris-HCl and Tween-20 (TBST) buffer containing 5% fat-free milk, and incubated with either rabbit anti-human EMP1 antibody (1:500; Abcam, Cambridge, MA, USA) or anti-β-actin antibody (internal control; Beyotime) at 4°C overnight. After being washed with TBST 3 times (5 min each), the membrane was incubated with a horseradish peroxidase (HRP)-conjugated goat anti-rabbit-IgG (1:2,000, ZSGB-Bio Origene, Beijing, China) at 37°C for 90 min. Detection was achieved using enhanced chemiluminescence (ECL) substrate (Thermo Fisher Scientific, Waltham, MA, USA) by ECL, and the images were acquired and quantified by the Gel Imaging system (Chishun, Nanjing, China).

For transient transfection of EMP1 into Hep-2 cells, the mammalian expression vector pcDNA3.1 expressing no proteins (control, pcDNA3.1-empty) and that expressing human EMP1 (pcDNA3.1-EMP1; GenScript, Nanjing, China) were transfected into Hep-2 cells using Lipofectamine 2000 (Invitrogen), according to the manufacturer's instructions.

Hep-2 cells in log phase were seeded into a 96-well plate at 5×10³/ml, with each condition set up in quintuplicate. Then cells were transfected with either pcDNA3.1-empty or pcDNA3.1-EMP1 for 24, 48, 72, 96 or 120 h. Then MTT (5 mg/ml, 10 μl/well; Sigma, St. Louis, MO, USA) reagent was added into each well and incubated with cells for a further 4 h, after which, DMSO (150 μl/well) was added to the cells and the plate was gently shaken for 15 min. Absorbance was measured at 490 nm.

Colony formation assay was performed as previously described (25). Briefly, at 24 h, after transfecting Hep-2 cells with either pcDNA3.1-empty or pcDNA3.1-EMP1, 4,000 cells were seeded into the 6-well tissue culture plate and cultured at 37°C for two weeks. The colonies were fixed in 95% ethanol, stained with crystal violet (0.1% w/v), and counted under a microscope.

For the Transwell migration assay, Hep-2 cells at 24 h after transfection (8×10⁴/ml) were seeded into the upper chamber of the Transwell insert (pore size 8 μm; Corning, Lowell, MA, USA). RPMI-1640 medium with 10% FBS was added into the lower chamber. After 24 h migration at 37°C, the insert was fixed with 10% methanol and washed 3 times with PBS. The cells remaining on the upper surface of the insert membrane were gently removed by wiping with cotton swabs. The insert was stained with hematoxylin for 40 sec, washed with PBS 3 times, and stained with eosin for 3 min. After another three washes with PBS, the cells were re-stained with hematoxylin for 30 sec. Each experimental condition was set up in quintuplicate. The migrated cells were imaged under a light microscope and the number of migrated cells was averaged from five fields (upper, lower, left, right and center) in each image.

At 24 h and 48 h after transfection of Hep-2 cells with either pcDNA3.1-empty or pcDNA3.1-EMP1, the cells were detached with trypsin, washed with PBS 3 times, and stained with 5 μl fluorescein isothiocyanate (FITC)-conjugated Annexin V and 10 μl propidium iodide (PI) (both from Keygen, Jiangsu, China) staining solution at room temperature for 15 min. After staining, the cells were analyzed by a FACScan flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA).

All experiments were independently repeated at least 3 times. Data were presented as mean ± SD. A χ² test was used to examine the possible associations between EMP1 expression and clinicopathological factors. Student's t-test was used to compare differences between control and treatment cells. SPSS 16.0 for Windows was used for statistical analyses. P<0.05 was considered statistically significant.

To assess the status of EMP1 in laryngeal carcinoma, we first compared its expression on both the steady-state mRNA and the protein levels in laryngeal carcinoma tissues to that in matched cancer-free peritumor tissues. As shown in Fig. 1A, out of the 51 sample pairs examined, in only 3 pairs was the EMP1 mRNA level more than twice as high in T than in N tissues; while in 24 pairs, the level was reduced by >50% in the T sample than in the matched N sample. The differences between the T and N samples were statistically significant (P<0.05) (Fig. 1B). Consistently, the EMP1 protein was significantly lower in the T samples than in matched N samples (P<0.05) (Fig. 1C and D).

IHC in tissues revealed strong EMP1 expression in peritumor normal epithelium, predominantly on the cell membrane and in the cytoplasm (Fig. 2A, left panel), where the average total staining score (S) was 11.60±1.26. In contrast, EMP1 staining was markedly reduced in the matched laryngeal carcinoma tissue (Fig. 2A, right panel), which was associated with a significantly lower S score (7.55±3.87; P<0.05). In addition, we observed that staining of EMP1 decreased with the histological grading of the tumor, with the highest S appearing in well-differentiated tumors, followed by intermediately differentiated tumors, and the lowest in poorly differentiated tumors (Fig. 2B). Further correlation analysis confirmed a significant correlation between the EMP1 S score and pathological grade (P<0.05) (Table I), but did not confirm a correlation with other clinicopathological features, including age, gender, TNM stage or lymph node metastasis (P>0.05) (Table I).

The significant reduction of EMP1 expression in the laryngeal carcinoma tissues compared to that in the matching peritumor laryngeal epithelia and its correlation with the differentiation status of the tumors suggest the tumor suppressor activity of EMP1. To characterize the functional significance of EMP1 in laryngeal carcinoma cells, we applied a gain-of-function approach and ectopically expressed EMP1 in laryngeal carcinoma Hep-2 cells. As shown in Fig. 3A, the transient transfection of Hep-2 cells with pcDNA3.1-EMP1 significantly boosted the expression of EMP1 in the cells, compared to cells transfected with the control vector pcDNA3.1-empty. Quantification analysis confirmed significant upregulation of EMP1 expression in pcDNA3.1-EMP1 cells compared to that in pcDNA3.1-empty cells at 24, 48 and 72 h after transfection (Fig. 3B).

Upon the overexpression of EMP1 in Hep-2 cells, we first examined the cell viability by MTT assay. As shown in Fig. 3C, both the mock-transfected (MOCK) and pcDNA3.1-empty-transfected (control) cells presented similar viability over 120 h after the transfection, suggesting that the control vector did not generate an obviously unfavorable effect on cell viability. In contrast, starting from 48 h after transfection, viability in EMP1-expressing cells was significantly lower than in the other two groups (P<0.05).

Cell viability results from a balance between cell proliferation and apoptosis. The marked decrease in cell viability following EMP1 ectopic expression prompted us to assess its effect on cellular apoptosis. By staining cells with Annexin V and PI, we found that at 48 h after EMP1 overexpression, there was a marked increase in the percentage of Annexin V⁺ and PI⁺ apoptotic cells (11.63±0.40%), compared to cells undergoing mock transfection (3.77±0.31%) or transfected with the control vector (2.53±0.11%) (P<0.05) (Fig. 4).

The capability of tumor cells to generate a large number of clonogenic progeny (clonogenicity/cancer stemness) is an important feature of cancer stem cells (26). To measure the effects of EMP1 on cancer clonogenic capacity, we performed colony formation assay. EMP1 overexpression in Hep-2 cells significantly reduced the number of colonies (675.00±43.66) formed, compared to mock-transfected cells (1055.67±161.55) or cells transfected with the control vector (1161.67±139.08) (P<0.05) (Fig. 5A and B).

Migration is another important feature for cancer cells, closely related to tumor metastasis, the most lethal phenotype of cancers. Using Transwell migration assay, we revealed that EMP1 overexpression significantly reduced the number of migrated cells (13.67±0.33/field), compared to mock-transfected cells (37.67±3.84/field), or cells transfected with the control vector (34.33±2.40/field; P<0.05) (Fig. 5C and D).