This study was approved by the Research Ethics Committee of Xi’an Jiaotong University. Written informed consent was obtained from all the patients. The specimens were handled and made anonymous according to the ethical and legal standards. A total of 247 patients were enrolled in this study. Patients received curative resection for CRC at the Second Affiliated Hospital, Xi’an JiaoTong University (Shaanxi, China) between 2001–2008. None of the patients enrolled in this research received blood transfusion or chemotherapy before surgery. The clinicopathological information of the patients is shown in Table I. The follow-up information of all participants was updated every 3 months by telephone. The overall survival was defined as the time elapsed from surgery to death. Information regarding the death of patients was ascertained from their family. Patients were followed after surgical treatment until May 2013, with a median follow-up of 83 months (range, 8–139 months). During the follow-up period, 121 patients (49.0%) died of the disease. The median overall and progression-free survival of patients was 37 and 34 months, respectively.

The expression of miR-494 in CRC and corresponding adjacent tissues were detected by RT-qPCR assay. Briefly, total RNA was extracted from tissues using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s protocol. Then, miRNA expression levels were quantitated using TaqMan microRNA real-time RT-PCR kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s protocol. Data were analyzed with 7500 Software v.2.0.1 (Applied Biosystems), with the automatic Ct setting for adapting baseline and threshold for Ct determination. The universal small nuclear RNA U6 (RNU6B) was used as an endogenous control for miRNAs. Each sample was examined in triplicate and the amounts of PCR products produced were non-neoplasticized to RNU6B.

Human CRC cell lines SW620, SW480, HCT116 and normal colon epithelium cell line CCD-18Co were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China), where they were characterized by mycoplasma detection, DNA fingerprinting, isozyme detection and cell vitality detection. They were cultured in RPMI-1640 (Invitrogen Life Technologies) mediums supplemented with 10% fetal bovine serum (FBS) (HyClone, Logan, UT, USA) and cultured in a humidified incubator at 37°C in 5% CO₂.

miR-494 mimics and inhibitors were chemically synthesized by Shanghai GenePharma (Shanghai, China). Once the cells were 80% confluent, miR-494 mimics or inhibitor was transfected into osteosarcoma cells with Lipofectamine 2000 (Invitrogen Life Technologies) according to the manufacturer’s instructions. Cells were also transfected with scramble oligonucleotide as negative control (NC). The expression level of miR-494 in the transfected osteosarcoma cells were identified by RT-qPCR.

Cell migration and invasion capacity were measured using Transwell migration assays (Millipore, Billerica, MA, USA) in vitro. The CRC SW620 cells were transfected with miR-494 inhibitors and NCs for 48 h, the CRC SW480 cells were transfected with miR-494 mimics and NCs for 48 h, and then the cells were suspended in RPMI-1640 with 10 g/l BSA at a density of 1×10⁶ cells/ml. Then, cell suspensions (150 μl) were seeded in the upper chamber with aporous membrane coated with (for the Transwell invasion assay) or without (for the migration assay) Matrigel (BD Biosciences, San Diego, CA, USA). To attract the cells, 500 μl of RPMI-1640 with 10% serum was added to the bottom chamber. After allowing the cells to migrate for 24 h or to invade for 48 h, the penetrated cells on the filters were fixed in dried methanol and stained in 4 g/l crystal violet. The numbers of migrated or invasive cells were determined from five random fields using a microscope (Olympus, Tokyo, Japan) at ×10 magnification.

CRC cells were seeded in a 96-well plate at 60% confluence. After 24 h, cells were transfected with 120 ng of wild-type (WT) or mutant (MT) 3′-UTR of PTEN mRNA expression vector pGL3-target-3′UTR and 30 ng of miR-494 mimics using Lipofectamine 2000. Cells were collected 48 h after transfection, and luciferase activity was measured using a dual luciferase reporter assay system according to the manufacturer’s protocol (Promega Corporation, Madison, WI, USA).

Cells were harvested in lysis buffer (50 mM NaCl, 50 mM EDTA, 1% Triton X-100) containing protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN, USA). The cell lysates (30 μg) were separated using 10% SDS-PAGE gels and then transferred onto nitrocellulose membranes (Millipore, Bedford, MA, USA). The membranes were blocked with 5% non-fat milk diluted in PBS for 2 h at room temperature before the addition of the appropriate primary antibody. The antibodies used in this study included anti-PTEN (1:1,000, no. 9188; Cell Signaling Technology, Inc., Danvers, MA, USA) and anti-β-actin (1:3,000, A2066; Sigma, St. Louis, MO, USA). The membranes were then washed with PBS containing 0.05% Tween-20 and incubated with the appropriate HRP-conjugated secondary antibody (1:5,000; Sigma) for 1 h at room temperature. The bands were visualized using a chemiluminescence reagent (New England Nuclear, Boston, MA, USA).

Statistical analysis was performed using IBM SPSS statistical software (version 20.0). Survival curves were estimated using the Kaplan-Meier method, and distributions were evaluated by the log-rank test. Cox proportional hazard models of factors related to survival were used to calculate RRs and identify the factors that affect survival. The differences in characteristics between the two groups were examined by the χ² test and Fisher’s exact test. All P-values were determined from two-sided tests, and statistical significance was based on a P-value of 0.05.

Expression of miR-494 was analyzed in 247 pairs of CRC and adjacent tissues by RT-qPCR and normalized against an endogenous control (U6 RNA). As shown in Fig. 1A the expression level of miR-494 in the CRC tissues was found to be distinctly upregulated when compared to that in adjacent tissues. Furthermore, in comparison to the non-metastatic CRC tissues, the miR-494 levels were significantly higher in metastatic CRC tissues (Fig. 1B). Moreover, miR-494 levels were upregulated in the tumor tissues obtained from the patients who suffered CRC recurrence (Fig. 1C). Collectively, these data indicated that significant upregulation of miR-494 expression occurred in CRC and was correlated with CRC relapse and metastasis. To further evaluate the association of miR-494 with CRC metastasis, we analyzed miR-494 levels in three CRC cell lines (SW620, SW480, and HCT116) and normal colon epithelium cell line CCD-18Co. Similarly, we found the expression of miR-494 was much higher in all three CRC cell lines than that in the normal colon cell line. It is noteworthy that miR-494 level was higher in metastatic CRC cells (SW620) than its paired metastatic cell line (SW480) (Fig. 1D). Taken together, the above findings suggest that miR-494 levels correlate with metastatic potential of CRC.

To determine the clinical significance of miR-494 in CRC, we analyzed the association of miR-494 expression with various clinicopathological parameters of CRC tissues. The median miR-494 expression in all 247 patients with CRC was 3.84. The patients were divided into two groups according to their expression levels of miR-494, using the median level as a cutoff: the high miR-494 expression group (n=129) and the low miR-494 expression group (n=118). As shown in Table I, miR-494 was significantly upregulated in CRC patients with advanced clinical stage (P=0.027), metastasis status (P=0.001), and positive lymph node metastasis (P=0.006). We also found a trend of increased miR-494 level from well to poor differentiation (P=0.022). However, miR-494 expression was not significantly correlated with gender, age, tumor size, tumor location and depth of invasion.

To determine the prognostic value of miR-494 expression in CRC, we analyzed the relationship between the miR-494 and clinical outcome. The relationship between miR-494 expression overall survival or progression-free survival was investigated using Kaplan-Meier analysis and log-rank test. Statistically significant differences in overall survival and progression-free survival were found between the high miR-494 level group and low miR-494 level group (Fig. 2A and B; log-rank test: P=0.0192 and P=0.0097, respectively). The patients with high miR-494 expression tended to have shorter overall and progression-free survival time when compared to patients with low miR-494 expression. Univariate analysis identified advanced clinical stage, metastasis status, lymph node status, differentiation status and high expression of miR-494 as indicators for poor prognosis for overall survival and progression-free survival (all P<0.05), whereas age, gender and tumor location were not significantly associated with overall survival and progression-free survival (Tables II and III). To test whether the prognostic value of high miR-494 expression was independent of other risk factors for poor overall and progression-free survival, a multivariate analysis was performed using a Cox proportional hazard model. Multivariate analyses including age, gender, tumor location, tumor stage, miR-494 expression, differentiation status, lymph node status, depth of invasion and metastasis status demonstrated that high miR-494 expression was an independent predictor for poor overall and progression-free survival in CRC patients (HR=2.874, CI=1.148–3.294, P=0.017 and HR=3.712, CI=1.513–5.272, P=0.009, respectively). Statistically significant results were also obtained for advanced tumor stage, and metastasis status, whereas all other parameters were not significant for overall and progression-free survival (Tables II and III).

As patients with recurrent metastatic CRC present with poor outcome, and our previous results indicated that miR-494 was significantly associated with progression-free and overall survival, we aimed to ascertain whether miR-494 affects the cell migration and invasion of CRC cells. The SW480 cell line was derived from the primary adenocarcinoma and SW620 cell line was derived from metastatic adenocarcinoma of the same patients (24). Our previous results proved that SW480 had lower expression of miR-494 than SW620. So they were used to investigate the role of miR-494 on cell migration and invasion. SW480 was transiently transfected with miR-494 mimics, whereas SW620 was transiently transfected with miR-494 inhibitors. As expected, transfection of miR-494 mimics or miR-494 inhibitors resulted in an increased or decreased miR-494 expression in SW480 or SW620, respectively (Fig. 3A and B). Moreover, the migration and invasion assays demonstrated that miR-494 overexpression significantly increased the migration and invasion ability of SW480 (Fig. 3C and E). Conversely, knockdown of miR-494 in SW620 decreased the migration and invasion ability of SW620 (Fig. 3D and F). These results indicate that miR-494 functions as oncogenic miRNA and contributes to the progression and metastasis of CRC cells.

To characterize the mechanism by which miR-494 promotes CRC metastasis, we searched for potential target genes of miR-494 using three publicly available databases, TargetScan, PicTar and miRanda. We were particularly interested in tumor suppressor gene PTEN, because PTEN was a crucial factor in various central processes of cancer development and loss of PTEN was proved to be associated with advanced CRC, liver metastasis, and poor survival (25,26). It has been proven that PTEN was the direct target of miR-494 in other cells (19,27). Considering the tissue-specific and developmental stage-specific manner of miRNAs, we investigate the relationship of PTEN and miR-494 in CRC cell lines. In order to confirm PTEN is a target gene for miR-494 in CRC cells, RT-qPCR was used to detect the expression of PTEN in CRC cell line SW480 and SW620. The expression of PTEN at mRNA and protein level was significantly downregulated after overexpression of miR-494 in SW480 cell line. Conversely, the expression of PTEN was significantly upregulated after knockdown miR-494 expression in SW620 cell line (Fig. 4A and B). Furthermore, we assessed the significance of the miR-494 and PTEN correlation in CRC tissues. We determined the PTEN mRNA and miR-494 expression in the same CRC specimens by RT-qPCR. As shown in Fig. 4C, a statistically significant inverse correlation was revealed by Spearman’s correlation analysis between mRNA levels of miR-494 and PTEN (r=−0.798; P<0.001). Taken together, our results suggest that miR-494 negatively regulates the expression of its potential target gene PTEN in CRC cell lines and tissues.

We further performed luciferase reporter assay to verify whether miR-101 directly targeted the 3′-UTR of PTEN in CRC cells. The target sequence of EZH2 3′-UTR (WT 3′-UTR) or the MT sequence (MT 3′-UTR) was cloned into a luciferase reporter vector (Fig. 4D). SW480 and SW620 cells were then transfected with WT or MT 3′-UTR vector and the miR-494 mimic. As shown in Fig. 4E, a significant decrease in luciferase activity was noted between the PTEN WT 3′-UTR group and the NC group in SW480 cells (P<0.05). The repressive effect was abrogated by point mutation in the core binding sites of the EZH2 3′-UTR. A similar trend was also found in the SW620 cells (Fig. 4F). These results indicate that miR-494 exerts an inhibitory effect on PTEN expression via interaction with the 3′-UTR of PTEN in CRC cells.