Abstract. The prognosis of oral squamous cell carcinoma (OSCC) patients is affected by tumor recurrence and metastasis, and cancer stem cells are hypothesized to be involved in these processes. Thus, the aim of the present study was to determine whether the expression levels of five stem cell-related transcription factors, sex determining region Y-box 2 (SOX2), octamer-binding transcription factor 4 (Oct4), avian myelocytomatosis viral oncogene homolog (c-Myc), Krüppel-like factor 4 (KLF4) and brachyury, are associated with metastasis and survival in OSCC. Immunohistochemistry was performed to analyze the expression of these proteins in biopsy specimens obtained from 108 OSCC patients. The results revealed that the expression of SOX2, Oct4, KLF4 and brachyury were significantly associated with lymph node metastasis (P=0.002, P=0.031, P=0.003 and P=0.007, respectively). In addition, the expression of KLF4 and brachyury were significantly associated with distant metastasis (P=0.014 and P=0.012, respectively). Furthermore, multivariate analysis revealed that SOX2 and KLF4 are predictive factors for lymph node metastasis [odds ratios (ORs), 4.526 and 4.851, respectively], and KLF4 is also a predictive factor for distant metastasis (OR, 9.607). In addition, OSCC patients with low co-expression of SOX2, KLF4 and brachyury exhibited a significantly lower disease-specific survival rate (78.6 vs. 100%; P=0.025; χ 2 =5.033) and disease-free survival rate (60.7 vs. 90.9%; P=0.015; χ 2 =5.897) when compared with OSCC patients with high co-expression of these factors. The results indicate that SOX2, KLF4 and brachyury serve important roles in tumor progression, and these transcription factors may thus represent clinically useful prognostic markers for OSCC.

CD82/KAI1, a member of the tetraspanin superfamily, is a suppressor of metastasis and CD82 inhibits canonical Wnt signaling via downregulation of several Frizzled (FZD) isoforms, resulting in accumulation of β-catenin at the cell membrane. In this study, we investigated the mechanism through which CD82 inhibited FZD expression by examining the effects of microRNAs (miRNAs). The miRanda algorithm predicted 11 miRNAs from FZD sequences. Among these miRNAs, CD82 caused upregulation of miR-203 (by 2.095-fold) and downregulation of miR-338-3p (by 0.354-fold) as compared with control cells. Transfection with miR-203 and miR338-3p mimics or inhibitors revealed that miR-203 downregulated FZD2 mRNA (by 0.268-fold) and protein expression (by 0.701-fold). Moreover, transfection with the miR-203 mimic also inhibited cell migration. Therefore, these findings suggested that CD82 enhanced the expression of miR-203 and directly downregulate FZD2 expression, suppressing cancer metastasis by inhibition of the Wnt signaling pathway.