MiR-32 functions as a tumor suppressor and directly targets EZH2 in human oral squamous cell carcinoma

Abstract

Background: MicroRNA-32 (miR-32) is dysregulated in certain human malignancies and correlates with tumor progression. However, its expression and function in oral squamous cell carcinoma (OSCC) remain unclear. Thus, the aim of this study was to explore the effects of miR-32 expression on OSCC tumorigenesis and development.

Material/methods: Real-time quantitative PCR was applied to evaluate the expression level of miR-32 in OSCC cell lines and primary tumor tissues. The association of miR-32 expression with clinicopathological factors and prognosis was also analyzed. In vitro cell proliferation, apoptosis, invasion, and migration assays were executed to elucidate biological effects of miR-32. Western blotting and luciferase assays were performed to confirm the regulation of EZH2 by miR-32.

Results: Down-regulation of miR-32 was found in OSCC tissues compared with corresponding noncancerous tissues (P<0.001). Decreased miR-32 expression was significantly associated with advanced T classifications, positive N classification, advanced TNM stage, and shorter overall survival (all P<0.05). Multivariate regression analysis corroborated that low-level expression of miR-32 was an independent unfavorable prognostic factor for OSCC patients. In vitro functional assays showed that overexpression of miR-32 reduced OSCC cell proliferation, migration, and invasion, and promoted cell apoptosis. In contrast, miR-32 knock-down resulted in an increase in cell growth and invasiveness. Finally, we identified EZH2 as the functional downstream target of miR-32 by directly targeting the 3'-UTR of EZH2.

Conclusions: These findings indicate that miR-32 may act as a tumor suppressor in OSCC and could serve as a novel therapeutic agent for miR-based therapy.

Figure 1

Expression of miR-32 and EZH2 in oral squamous cell carcinoma (OSCC) tissues and cell lines. (A) MiR-32 expression was significantly lower in OSCC tissues than in the corresponding non-cancerous tissues. MiR-32 expression levels were calculated by the 2^−ΔCt method and normalized to U6 small nuclear RNA. (B) miR-32 expression was down-regulated in OSCC cell lines SCC-4, SCC-9, SCC-25, and Tca8113, compared to human normal oral keratinocyte cells. * p<0.05; ** p<0.01. (C) Relative EZH2 protein levels in OSCC and corresponding non-cancerous tissues. EZH2 protein levels were measured by Western blot analysis and normalized to β-actin. (D) EZH2 protein levels in OSCC cells were higher than in normal lung epithelial cells. (E) The inverse correlation of EZH2 protein levels with miR-32 expression was examined by Pearson correlation analysis.

Figure 2

Overall survival curves for 2 groups defined by low and high expression of miR-32 in patients with oral squamous cell carcinoma. Low miR-32 expression levels were significantly associated with poor outcome (P<0.001, log-rank test).

Figure 3

Effects of miR-32 mimics or inhibitors transfection on biological behaviors of Tca8113 and SCC-4 cells. (A) qRT-PCR analysis confirmed increased miR-32 expression in Tca8113 cells transfected with miR-32 mimics, and decreased miR-32 expression in SCC-4 cells transfected with miR-32 inhibitors. U6 RNA was used as an internal control. (B) MTT assay showed that miR-32 reduced cell proliferation in vitro. Data represent the mean ±SD of the experiments performed in triplicate. ** p<0.01. (C) Cell apoptosis was detected by flow cytometric analysis after transfection with miR-32 mimics, miR-32 inhibitors, or negative control. (D) Transwell invasion assay showed that up-regulation of miR-32 impeded the invasion of Tca8113 cells, while transfection of SCC-4 cells with miR-32 inhibitors promoted cell invasion. (E) Scratch migration assay confirmed the inhibitory effect of miR-32 on OSCC cell migration. ** p<0.01.

Figure 4

EZH2 is a direct target of miR-32. (A) miR-32-binding sites in EZH2 3′UTR region. EZH2-mut indicates the EZH2 3′UTR with mutation in miR-32-binding sites. (B) Western blot showed that transfection of miR-32 decreased EZH2 protein expression. (C) Relative luciferase assay comparing the pGL3-EZH2 and pGL3-EZH2-MUT vectors in Tca8113 cells. Firefly luciferase activity was normalized to Renilla luciferase activity. * p<0.05.