HOXC10 promotes migration and invasion via the WNT-EMT signaling pathway in oral squamous cell carcinoma

Abstract

As a master regulator of embryonic morphogenesis, homeodomain-containing gene 10 (HOXC10) has been found to promote progression of human cancers and indicate poor survival outcome. Therefore, we concentrate on elucidating the role of HOXC10 in progression of oral squamous cell carcinoma (OSCC). In our study, the expression of HOXC10 was significantly increased in human OSCC samples and was significantly correlated with TNM stage and lymph node metastasis. Upregulation of HOXC10 indicated a poor overall survival of OSCC patients according to the Kaplan-Meier survival curves. Furthermore, HOXC10-knockdown dramatically suppressed migration, invasion, and expression of N-Cadherin, Vimentin and Snail, as well as increased E-cadherin level both in vivo and in vitro. Bioinformatics and cellular study further confirmed that HOXC10 may promote invasion and migration of OSCC cells by regulating the WNT/epithelial-mesenchymal transition (EMT) signaling pathway. These findings suggest that HOXC10 plays a pivotal role in the metastasis of OSCC and highlight its usefulness as a potential prognostic marker or therapeutic target in human OSCC.

Keywords: HOXC10; WNT10B, oral squamous cell carcinoma, epithelial-mesenchymal transition.

Figure 1

HOXC10 expression was up regulated in oral squamous cell carcinoma (OSCC) (A). Representative immunohistochemical staining of HOXC10 in human OSCC tissue compared with that in normal mucosa; scale bar: 50 μm. (B). Quantification of HOXC10 expression levels in human mucosa and OSCC tissue. (C). ROC curve analysis was adapted to determine cut-off score for dichotomizing as low expression and high expression of HOXC10. (D). Kaplan-Meier curve of overall survival in 57 patients with OSCC stratified by the expression level of HOXC10. The duration of survival was measured from the beginning of the treatment to the time of death or at 60 months. The cumulative survival for OSCC patients with positive HOXC10 expression was significantly lower than that for OSCC patients who were HOXC10-negative. (E). The relative HOXC10 mRNA level was detected by RT-PCR in OSCC patients in fresh, paired cancer and normal tissues. The data are presented as the means ± SEM. **P<0.01 versus the control group.

Figure 2

Expression analysis of HOXC10 in OSCC cell lines. (A). Western blot analysis was performed to assess the expression level of HOXC10 in OKC and OSCC cell lines; GAPDH served as a loading control. (B). Knockdown of HOXC10 by two different shHOXC10 sequences in the FaDu cell line; GAPDH served as a loading control. (C). Knockdown of HOXC10 by two different shHOXC10 sequences in the SCC4 cell line; GAPDH served as a loading control. The relative density data were calculated by ImageJ software. (D). Colony formation of control and shHOXC10 cells in soft agar after 14 days of culture. The data are presented as the means ± SEM. *P<0.05, **P<0.01 versus the control group. (n=6)

Figure 3

Depleting HOXC10 by RNA interference decreased OSCC cell line migration and invasion. Wound-healing assay showed that knockdown of HOXC10 suppressed cell mobility of FaDu (A) and SCC4 (B) cell lines, and the quantification of the wound closures show a statistically significant difference; scale bar: 200 μm. The transwell assay showed that the migration (C) and invasion (D) abilities of FaDu and SCC4 cells were impaired after knockdown of HOXC10 compared with those in the negative control group, and the quantification of cell numbers with the ImageJ; scale bar: 50 μm. The data are presented as the means ± SEM. *P<0.05, **P<0.01, ***P<0.001 versus the control group. (n=6).

Figure 4

HOXC10 was positively correlated with Wnt signaling pathways. (A-B). The main enriched pathways of the HOXC10-correlated genes identified by KEGG analysis though the DAVID online database. (C). FaDu cells and (D) SCC4 cells were treated with shHOXC10, and Wnt signaling pathway-related gene expression was determined. The data are presented as the means ± SEM. **P<0.01 versus the control group.

Figure 5

Knockdown of HOXC10 suppresses the WNT-EMT process in OSCC cell lines. (A). FaDu cells and (B) SCC4 cells were treated with shHOXC10, and Wnt10B, N-cadherin, E-cadherin, and Vimentin levels were determined. GAPDH served as an internal standard for protein loading. (C). FaDu cells and (D) SCC4 cells were treated with negative control (NC) and shHOXC10; representative immunofluorescence is shown, and fluorescence of N-Cadherin, Snail and E-cadherin was quantified; scale bar: 20 μm. The data are presented as the means ± SEM. **P<0.01 versus the control group.

Figure 6

Knockdown of HOXC10 blocks tumor growth, invasion and metastasis of OSCC in vivo. (A). Tumor growth curve for shHOXC10 and control mice. (B). The tumor volume and weight were measured. (C). Dissected tumors were photographed. (D). Representative images of immunohistochemical analysis of HOXC10, Wnt10B, DVL2, E-cadherin, N-cadherin and Vimentin in tumors; scale bar: 50 μm. The data are presented as the means ± SEM. *P<0.05, **P<0.01 versus the control group.

Figure 7

HOXC10 is positively correlated with Wnt-EMT signaling pathways in OSCC patients. (A). The protein levels of Wnt10b, E-cadherin and Vimentin were determined. (B). Correlation of HOXC10 and Wnt10b protein levels in OSCC samples. (C). Correlation of HOXC10 and Vimentin protein levels in OSCC samples. **P<0.01, versus the normal mucosa group;