Seventy-three NPC and 30 normal nasopharyngeal tissues were obtained from the Department of Otolaryngology Head and Neck Surgery, The First Affiliated Hospital of Bengbu Medical College from January 2003 to December 2010. The NPC patients received no perioperative radiotherapy or chemotherapy prior to surgery. Written informed consent was obtained from all patients enrolled in the present study. The demographic and clinicopathological features of all patients are documented in Table I. The Ethics Committee of Bengbu Medical College approved all protocols involving the patient samples according to the Declaration of Helsinki (as revised in Tokyo 2004).

Human NPC cell lines 5–8F, 6-10B, CNE-1 and CNE-2 were purchased from the Cancer Center of Sun Yat-Sen University (Guagzhou, China) and cultured in RPMI-1640 medium containing 10% fetal bovine serum (FBS) (both from Gibco, Grand Island, NY, USA), 100 U/ml penicillin and 100 U/ml streptomycin. The human immortalized nasopharyngeal epithelial cell line NP69 was cultured in serum-free medium (Invitrogen, Carlsbad, CA, USA) supplemented with all necessary growth factors (Gibco). All cells were maintained in a humidified cell incubator with 5% CO₂.

The miR-212 mimics and inhibitor, and the corresponding negative control vectors were obtained from GeneCopoeia (Guangzhou, China). SOX4 siRNA and SOX4 overexpression vectors were obtained from Addgene (Cambridge, MA, USA). All the vectors were transfected into NPC cells to overexpress or inhibit the expression of miR-212 or SOX4 in NPC cells based on the instructions provided for Lipofectamine 2000.

TPIzol (Invitrogen) was used to extract the RNA from clinical specimens and NPC cells following the manufacturer's instructions. PCR amplification and quantification for miR-212 were performed using the TaqMan miRNA reverse transcription kit and the TaqMan human miRNA assay kit (both from Applied Biosystems, Foster City, CA, USA). Primers for miRNA miR-212 and U6 (HmiRQP9001) were purchased from GeneCopoeia. U6 was used as the internal control for measuring the relative level of miR-212.

Cellular proteins extracted from NPC cells using RIPA lysis buffer (BioMed, Beijing, China) were separated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and were transferred to polyvinylidene fluoride (PVDF) membranes. The blots were incubated with the following primary antibodies overnight: GAPDH (1:2,500; Cell Signaling Technologies, Danvers, MA, USA) and SOX4 (1:500; Santa Cruz Biotechnology, Santa Cruz, CA, USA). After incubating with the primary antibody, the membranes were incubated with secondary antibodies (1:5,000; Bio-Rad, Hercules, CA, USA) at room temperature for 2 h, and then the protein signals were detected using the Bio-Rad Gel imaging system.

NPC cells (1×10⁵) transfected with the corresponding vectors were re-suspended in serum-free medium, and were seeded into the Transwell inserts of 8-µm pore size (Millipore, Billerica, MA, USA). Serum-containing medium (650 µl) (20% FBS) was added to the lower chamber as the attractant. Regarding the invasion assay, each upper chamber was coated with a mixture of Dulbeccos modified Eagles medium (DMEM) and Matrigel (Becton-Dickinson Labware, Franklin Lakes, NJ, USA) at a ratio of 8:1. Twenty-four hours later, the NPC cells that had migrated or invaded through the Transwell membranes on the lower surface were stained with 0.1% crystal violet, and the cell numbers were counted from 10 different fields of the lower surface of the filter. Three independent experiments were performed.

Wild-type SOX4 3′-UTR sequence and the mutated SOX4 3′-UTR sequence were constructed into the pGL3 control vector (Promega, Madison, WI, USA) to obtain the wt SOX4-3′-UTR and mt SOX4-3′-UTR vector, respectively. Primers for SOX4 3′UTR are listed as follows: wild-type SOX4 3′UTR forward, GAGCTCCTCCGCCTTCT TTTCTAC and reverse, CTCGAGCACGTCTTCTCATTTA CACC; mutant SOX4 3′UTR forward, GAGCATTTGATGTG GTACAGGGGCAG and reverse, TCCTCTCCTCCACGCC TCCGGGGTC. For the luciferase reporter assay, NPC cells were co-transfected with the wild-type or mutant construct, and miR-212 mimics (CNE-2 cells), or inhibitor (6-10B cells), or control, or negative control vector. Forty-eight hours after transfection, the cells were harvested and lysed. The Dual-Luciferase reporter assay system (Promega, Shanghai, China) was used to measure the firefly and Renilla luciferase activities.

Statistical analysis was conducted with GraphPad Prism 5 software (GraphPad Software, Inc., La Jolla, CA, USA). Statistical analysis including the Pearson's Chi-square test, the Spearman's rank correlation coefficient, the two-tailed Student's t-test and Kaplan-Meier plots were used in the present study. P<0.05 was considered to indicate statistically significant difference.

The expression level of miR-212 in clinical tissues derived from NPC patients was evaluated by qRT-PCR. As shown in Fig. 1A, the expression level of miR-212 in NPC tissues was significantly lower than that in normal tissues (P<0.01, Fig. 1A). Moreover, the expression level of miR-212 was significantly decreased in patients with metastasis (P<0.05, Fig. 1B) and patients in tumor-node-metastasis (TNM) stage III–IV (P<0.05, Fig. 1C). Next, we compared the expression of miR-212 among 4 NPC cell lines (6-10B, 5–8F, CNE1 and CNE2) and NP69 a nasopharyngeal epithelial cell line. Compared with the NP69 cells, all NPC cells had a significantly decreased miR-212 level (P<0.05, Fig. 1D). These data suggest that miR-212 plays a tumor-suppressive role in NPC and is involved in the progression of NPC.

We investigated the clinical significance of the decreased expression level of miR-212 in NPC. We divided the NPC patients into two groups based on the cut-off value which was defined as the median value of the miR-212 level: miR-212 low expression group (n=36) and miR-212 high expression group (n=37). Then, the correlation between the clinicopathological features of the NPC patients and miR-212 level was evaluated. As shown in Table I, a decreased expression level of miR-212 was significantly associated with advanced TNM stage (P=0.013), and the occurrence of metastasis of NPC (P<0.001). Furthermore, Kaplan-Meier analysis showed that patients with a low expression level of miR-212 had a significantly lower overall survival rate (P=0.0158, Fig. 2A) and disease-free survival rate (P=0.0092, Fig. 2B).

After confirming the expression status and clinical significance of miR-212 in NPC, we examined the biological functions of miR-212 in NPC cells. In addition, a significant association between miR-212 and TNM stage and metastasis motivated us to investigate whether miR-212 modulates the metastatic behaviors of NPC cells. Transfection of the miR-212 mimic into the CNE-2 cells significantly increased the expression level of miR-212 (P<0.01, Fig. 3A). Subsequently, overexpression of miR-212 in the CNE-2 cells led to significantly decreased migration (P<0.05, Fig. 3B) and invasion (P<0.01, Fig. 3C) of CNE-2 cells. To further confirm functional influences of miR-212 on the migration and invasion of NPC cells, we downregulated the expression of miR-212 in 6-10B cells with the miR-212 inhibitor. Transfection of the miR-212 inhibitor significantly decreased the level of miR-212 in the 6-10B cells (P<0.01, Fig. 3D), and led to significantly increased migration (P<0.01, Fig. 3E) and invasion (P<0.01, Fig. 3F) of 6-10B cells.

To elucidate the molecular mechanisms responsible for the functional influence of miR-212 in NPC cells, we searched publically available database TargetScan to identify the downstream target of miR-212. Data from TargetScan showed that SOX4, a well-known oncogenic protein in human cancers (16–19), was a possible downstream target of miR-212. As suggested by Fig. 4A, the sequences complementary to the binding sites of miR-212 were found in the 3′-UTR of SOX4. Then, we performed luciferase reporter assay to further confirm that miR-212 could bind to the 3′-UTR of SOX4. The results showed that forced expression miR-212 significantly decreased the luciferase activity of wild-type SOX4 3′-UTR (P<0.01, Fig. 4B) while had no influence on that of the mutant SOX4 3′-UTR (Fig. 4B). Furthermore, inhibition of miR-212 expression increased the luciferase activity of wild-type SOX4 3′-UTR (P<0.05, Fig. 4C) and did not affect the luciferase activity of mutant SOX4 3′-UTR. After confirming that miR-212 affected the luciferase activity of SOX4 3′-UTR by interacting with the 3′-UTR of SOX4, we performed qRT-PCR and western blotting to confirm that miR-212 modulates the expression of SOX4 in NPC cells. The results of qRT-PCR and western blotting demonstrated that forced expression of miR-212 significantly reduced the mRNA (P<0.05, Fig. 5A) and protein (P<0.01, Fig. 5B) levels of SOX4 in the CNE-2 cells. In contrast, knockdown of miR-212 significantly increased the mRNA (P<0.05, Fig. 5C) and protein (P<0.01, Fig. 5D) levels of SOX4 in the 6-10B cells. These data demonstrated that SOX4 is a direct downstream target of miR-212 in NPC cells.

To further clarify whether SOX4 is not only a downstream target of miR-212, but also a functional mediator of miR-212 in NPC cells, we performed the transfection of the SOX4-expressing vector into CNE-2 cells with miR-212 overexpression (CNE-2-miR-212 cells). Transfection of the SOX4 overexpression vector significantly increased SOX4 expression in the CNE-2-miR-212 cells (P<0.05, Fig. 6A). Subsequently, SOX4 overexpression abrogated the inhibitory effects of miR-212 on the migration (P<0.05, Fig. 6B) and invasion (P<0.01, Fig. 6C) of CNE-2 cells. In contrast-, SOX4 siRNA transfection into 6-10B cells with miR-212 knockdown (6-10B-anti-miR-212 cells) significantly reduced the expression of SOX4 (P<0.05, Fig. 6D). Knockdown of SOX4 reversed the promoting effects of the miR-212 inhibitor on the migration (P<0.01, Fig. 6E) and invasion (P<0.01, Fig. 6F) of 6-10B cells. These findings indicate that SOX4 is not only a downstream target of miR-212, but also a functional mediator of miR-212 in NPC cells.