EBV-miR-BART10-3p facilitates epithelial-mesenchymal transition and promotes metastasis of nasopharyngeal carcinoma by targeting BTRC

Abstract

Epstein-Barr virus (EBV) infection is closely associated with tumorigenesis and development of nasopharyngeal carcinoma (NPC), but the underlying molecular mechanisms remain poorly understood. It has been recently reported that EBV encodes 44 mature miRNAs, some of which were found to promote tumor development by targeting virus-infected host genes or self-viral genes. However, few targets of EBV encoded-miRNAs that are related to NPC development have been identified to date. In this study, we revealed that in NPC cells, EBV-miR-BART10-3p directly targets BTRC gene that encodes βTrCP (beta-transducin repeat containing E3 ubiquitin protein ligase). We found that EBV-miR-BART10-3p expression in clinical samples from a cohort of 106 NPC patients negatively correlated with BTRC expression levels. Over-expression of EBV-miR-BART10-3p and down-regulation of BTRC were associated with poor prognosis in NPC patients. EBV-miR-BART10-3p promoted the invasion and migration cabilities of NPC cells through the targeting of BTRC and regulation of the expression of the downstream substrates β-catenin and Snail. As a result, EBV-miR-BART10-3p facilitated epithelial-mesenchymal transition of NPC. Our study presents an unreported mechanism underlying EBV infection in NPC carcinogenesis, and provides a potential novel biomarker for NPC diagnosis, treatment and prognosis.

Keywords: BTRC; EBV-miR-BART10-3p; Epstein-Barr virus (EBV); epithelial-mesenchymal transition (EMT); nasopharyngeal carcinoma (NPC).

Figure 1. The correlation between the expression of BTRC mRNA and EBV-miR-BART10-3p was analyzed by real-time PCR data obtained from 28 NPC tissues and 9 non-tumor nasopharyngeal epithelial tissues

N, non-tumor nasopharyngeal epitheliums; T, NPC. N, n = 9; T, n = 28, *, p < 0.05; ***, p < 0.001).

Figure 2. The inverse correlation between high expression of EBV-miR-BART10-3p and low expression of βTrCP in NPC and their expression was associated with poor survival of NPC patients

A. Comparison of the expression of EBV-miR-BART10-3p between 106 NPC tissue samples and adjacent epithelial tissues was performed by in situ hybridization (ISH). As shown in representative images, high expression of EBV-miR-BART10-3p was detected in NPC tissues, as compared to adjacent epithelial tissues. B. βTrCP expression was inversely correlated with EBV-miR-BART10-3p in the same cohort of NPC tissues and adjacent epithelial tissues, detected by immunohistochemistry (IHC). C. Overexpression of EBV-miR-BART10-3p in NPC was associated with lymph-vascular invasion (p < 0.05). D. The highly expressed EBV-miR-BART10-3p was correlated with in situ relapse (n = 27) or distant metastasis (n = 61) in NPC patients (p < 0.05). E. and F. The highly expressed EBV-miR-BART10-3p was correlated with shorter disease free survival (DFS, p = 0.030, E) or overall survival (OS, p = 0.010, F) of NPC patients. G. and H. The low expressed βTrCP expression was correlated with shorter disease free survival (DFS, p = 0.013, G) or overall survival (OS, p = 0.006, H) of NPC patients.

Figure 2. The inverse correlation between high expression of EBV-miR-BART10-3p and low expression of βTrCP in NPC and their expression was associated with poor survival of NPC patients

A. Comparison of the expression of EBV-miR-BART10-3p between 106 NPC tissue samples and adjacent epithelial tissues was performed by in situ hybridization (ISH). As shown in representative images, high expression of EBV-miR-BART10-3p was detected in NPC tissues, as compared to adjacent epithelial tissues. B. βTrCP expression was inversely correlated with EBV-miR-BART10-3p in the same cohort of NPC tissues and adjacent epithelial tissues, detected by immunohistochemistry (IHC). C. Overexpression of EBV-miR-BART10-3p in NPC was associated with lymph-vascular invasion (p < 0.05). D. The highly expressed EBV-miR-BART10-3p was correlated with in situ relapse (n = 27) or distant metastasis (n = 61) in NPC patients (p < 0.05). E. and F. The highly expressed EBV-miR-BART10-3p was correlated with shorter disease free survival (DFS, p = 0.030, E) or overall survival (OS, p = 0.010, F) of NPC patients. G. and H. The low expressed βTrCP expression was correlated with shorter disease free survival (DFS, p = 0.013, G) or overall survival (OS, p = 0.006, H) of NPC patients.

Figure 3. EBV-miR-BART10-3p targeted BTRC gene and inhibited its expression in NPC cells

EBV negative NPC cell lines HNE2 and 5-8F were transfected by EBV-miR-BART10-3p mimics (BART10-3p) or negative control (NC) respectively. Expression of exogenous BART10-3p was detected by real-time PCR A. or Northern blotting B.. NS: no signaling. C666-1 was served as positive control in Northern blotting, T1, T2, T3 and T4 are four NPC primary biopsies, the U6 RNA probe was used as an internal loading control. The expression of BTRC at the mRNA C. or protein D. levels were decreased in the EBV-miR-BART10-3p mimics transfected NPC cells, as compared to one with negative control (NC), detected by real-time PCR or western blotting. E. EBV-miR-BART10-3p expression was significantly inhibited by a synthesized inhibitor (BART10-3p In) in C666-1, a EBV-positive NPC cell line, as compared to negative control (NC) one. The mRNA F. and protein G. expression levels of BTRC were increased in C666-1 cells transfected with EBV-miR-BART10-3p inhibitor. BTRC as a direct target of EBV-miR-BART10-3p was confirmed in HNE2 H. and 5-8F I. cells by co-transfection with EBV-miR-BART10-3p mimics and luciferase reporter containing either wild type (BTRC-WT) or mutated (BTRC-mutant) EBV-miR-BART10-3p binding sites in BTRC 3′-UTR. EBV-miR-BART10-3p mimics attenuated the luciferase activity of BTRC-WT, rather than BTRC-mutant (*, p < 0.05; ***, p < 0.001, Figures are representative of three experiments).

Figure 4. EBV-miR-BART10-3p promoted invasion and migration of NPC cells by reducing BTRC expression

A. The invasion ability was evaluated by transwell assay in EBV negative NPC cells HNE2 and 5-8F or EBV positive cells C666-1. EBV-miR-BART10-3p mimics (BART10-3p), BTRC overexpression vector (BTRC), BART10-3p+BTRC or BTRC siRNA were transfected into HNE2 and 5-8F cells, respectively. EBV-miR-BART10-3p inhibitors (BART10-3p In), BTRC overexpression vector (BTRC), BTRC siRNA (siBTRC), or BART10-3p In+siBTRC were transfected into C666-1 cells, respectively. EBV-miR-BART10-3p mimics could significantly promote invasion of NPC cells, whereas the EBV-miR-BART10-3p mimics-enhanced tumor cell invasion and migration were rescued by overexpression of BTRC. B. Wound healing assay showed that both EBV-miR-BART10-3p mimics and BTRC siRNA accelerated would gap closure, as compared with those with negative control (NC). Overexpression of BTRC in HNE2 and 5-8F cells reduced the migration ability, leading to a delayed wound gap closure. Wound healing assay were also performed after EBV-miR-BART10-3p inhibitors (BART10-3p In), BTRC overexpression vector (BTRC), BTRC siRNA (siBTRC), or BART10-3p In+siBTRC transfection in C666-1 cells. The migration ability could be blocked by BART10-3p In or BTRC expression vector. The BART10-3p In-blocked migration ability of C666-1 cells was rescued by siBTRC, and siBTRC alone also increased the migration ability. The cells in five randomly selected fields were counted and the data were shown as the mean ± SD (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

Figure 4. EBV-miR-BART10-3p promoted invasion and migration of NPC cells by reducing BTRC expression

A. The invasion ability was evaluated by transwell assay in EBV negative NPC cells HNE2 and 5-8F or EBV positive cells C666-1. EBV-miR-BART10-3p mimics (BART10-3p), BTRC overexpression vector (BTRC), BART10-3p+BTRC or BTRC siRNA were transfected into HNE2 and 5-8F cells, respectively. EBV-miR-BART10-3p inhibitors (BART10-3p In), BTRC overexpression vector (BTRC), BTRC siRNA (siBTRC), or BART10-3p In+siBTRC were transfected into C666-1 cells, respectively. EBV-miR-BART10-3p mimics could significantly promote invasion of NPC cells, whereas the EBV-miR-BART10-3p mimics-enhanced tumor cell invasion and migration were rescued by overexpression of BTRC. B. Wound healing assay showed that both EBV-miR-BART10-3p mimics and BTRC siRNA accelerated would gap closure, as compared with those with negative control (NC). Overexpression of BTRC in HNE2 and 5-8F cells reduced the migration ability, leading to a delayed wound gap closure. Wound healing assay were also performed after EBV-miR-BART10-3p inhibitors (BART10-3p In), BTRC overexpression vector (BTRC), BTRC siRNA (siBTRC), or BART10-3p In+siBTRC transfection in C666-1 cells. The migration ability could be blocked by BART10-3p In or BTRC expression vector. The BART10-3p In-blocked migration ability of C666-1 cells was rescued by siBTRC, and siBTRC alone also increased the migration ability. The cells in five randomly selected fields were counted and the data were shown as the mean ± SD (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

Figure 5. EBV-miR-BART10-3p up-regulated the expression of β-catenin and Snail through inhibiting βTrCP

A. Western blot analysis of the expression of BTRC and its substrates β-catenin and Snail in EBV negative NPC cell lines HNE2 and 5-8F transfected with EBV-miR-BART10-3p mimics (BART10-3p) or BTRC expression vector (BTRC). B. Western blot analysis of the expression of BTRC and its substrates β-catenin and Snail in EBV positive NPC cell lines C666-1 transfected with EBV-miR-BART10-3p inhibitors (BART10-3p In) or BTRC siRNA (siBTRC). C. and D. The effect of EBV-miR-BART10-3p mimics (BART10-3p) or BTRC expression vector (BTRC) on ubiquitination degradation rate of β-catenin C. and Snail D. in 5-8F cells was detected by western blotting at the indicated time point after treatment with Cycloheximide (CHX), an inhibitor of protein biosynthesis. GAPDH was used as an internal loading control.

Figure 6. The effect of EBV-miR-BART10-3p on EMT in NPC cells was confirmed by immunofluorescence

The expression levels of Vimentin and β-catenin were examined by immunofluorescence assay in 5-8F A. or C666-1 B. cells transfected with EBV-miR-BART10-3p mimics (BART-10-3p), or BTRC expression vector (BTRC), EBV-miR-BART10-3p inhibitors (BART-10-3p In), or BTRC siRNA (siBTRC). Up-regulation of Vimentin and β-catenin by both EBV-miR-BART10-3p mimics and BTRC siRNA, as well as down-regulation of them by ectopic BTRC or EBV-miR-BART10-3p inhibitors were also confirmed by this assay (NC: negative control). Five randomly selected areas were scanned and data were shown as the mean ± standard deviation (right panel, *, p < 0.05; **, p <0.01).

Figure 7. EBV-miR-BART10-3p promoted EMT through BTRC

The expression levels of epithelial markers ZO-1, E-cadherin and claudin-1, as well as mesenchymal markers ZEB1, N-cadherin, Vimentin and Slug were examined by western blot analysis in HNE2 and 5-8F cell lines that were transfected with EBV-miR-BART10-3p mimics (BART10-3p) or BTRC expression vector (BTRC). GAPDH was used as an internal control in western blotting.

Figure 8. Graphical representation illustrated the role of EBV-miR-BART10-3p mediated pathway in the EMT of NPC