In this study, we observed that miR-195 expression was reduced in TSCC compared with adjacent nonmalignant tissues, and that decreased expression was correlated with cancer progression and prognosis. Moreover, we determined that decreased miR-195 expression was associated with poor overall survival in TSCC patients, independent of other clinicopathologic factors.
miR-195 could be a potential biomarker for prognosis prediction in TSCC patients. Except for their close association with patient outcomes, biomarkers should ideally be expressed at stable levels in the patient samples in question. In this respect, miRNAs are relatively stable as compared with other biological macromolecules. They can be well preserved in tissue samples even after formalin fixation and paraffin-embedding, and can be efficiently extracted and evaluated 
. Therefore, based on our observations that decreased miR-195 expression was associated with poor overall survival in TSCC patients, we anticipate that miR-195 could be a useful prognostic factor for TSCC and that its stability should allow the development of practical, economical methods for TSCC detection. Moreover, the development of cancers involves the altered expression of multiple genes, so the protein product of a single oncogene may not accurately reflect the status of the disease. However, just as other single miRNAs are known to target multiple messenger RNAs to regulate gene expression 
, miR-195 can also regulate multiple coding genes that are related to tumor growth 
. Thus, expression of miR-195 is likely to reflect altered physiology of TSCC more precisely and effectively than any of the target genes alone.
Although our data failed to establish evidence for Cyclin D1 and Bcl-2 expression as prognostic markers in TSCC, we did demonstrate for the first time that immunostaining of Cyclin D1 and Bcl-2 is inversely correlated with miR-195 levels in TSCC tissues. Because Cyclin D1 and Bcl-2 have been shown to be direct targets of miR-195 
and their expression in TSCC may account for the effect of miR-195, we examined the expression of these two proteins in paraffin sections of TSCC samples using immunohistochemistry. In this study, the expression of the Cyclin D1 was only statistically significantly associated with the tumor size of TSCC but not with other clinicopathological factors analyzed, whereas the expression of Bcl-2 in TSCC was not statistically significantly associated with any of the clinicopathological factors analyzed. However, according to previous studies, the role of Cyclin D1 as a prognostic marker remains controversial 
and there is no consensus on the use of Bcl-2 as a prognostic marker 
for squamous cell carcinomas among head and neck cancer patients either. Our results concerning Cyclin D1 and Bcl-2 were consistent with some of these publications 
. Variation in the prognostic significance of Cyclin D1 and Bcl-2 in previous studies may be attributable to differences in sample size, definitions of positive expression, the inclusion of tumors from different subsites of the oral cavity, and the diversity of treatments. More importantly, our data showed that the expression of Cyclin D1 and Bcl-2 in TSCC tissues is inversely correlated with miR-195 levels. These important observations not only support previous findings that Cyclin D1 and Bcl-2 are target genes silenced by miR-195 but also demonstrate that the expression of miR-195 is potentially a more accurate prognostic tumor marker than Cyclin D1 or Bcl-2 levels alone in TSCC patients.
The anti-tumor effect of miR-195 in TSCC could be at least partially via inhibition of Cyclin D1 and Bcl-2 expression. We performed a series of experiments using two TSCC cell lines (SCC-15 and CAL27) to investigate the function of miR-195. Our results demonstrate that ectopic overexpression of miR-195 reduces cell viability, inhibits cell cycle progression, and promotes cell apoptosis. Moreover, Cyclin D1 and Bcl-2 were shown to be direct targets of miR-195 by a dual-luciferase reporter assay and western blots, and their inhibition may account for the antitumor effect of miR-195 in TSCC. However, because TargetScan predicts hundreds of potential targets of miR-195 (http://www.targetscan.org
), we cannot exclude the possibility that other potential targets of miR-195 may govern additional cancer pathways that promote TSCC cancer development and that miR-195 may also target different molecules in different types of cancer.
Our study focused on a large series of patients who satisfied stringent recruitment criteria: (1) tumor location at the anterior body of the tongue, (2) squamous cell carcinoma, and (3) surgery as the primary treatment. We hope that this study will provide more accurate and clinically useful information on the prognostic significance of miR-195 expression. Several papers have described the involvement of miRNAs in head and neck squamous cell carcinoma 
. In these publications, which generally have included comparisons of normal and tumor samples, miRNA profiling was used to associate the expression of miRNAs with malignant progression and prognosis. Although these initial data have already suggested that miRNAs are involved in squamous cell carcinogenesis, the studies have always included heterogenous groups of patients with cancers from different subsites of oral cavity, and gene expression patterns from squamous cell carcinomas at different subsites of oral cavity may not be equally associated with cancer prognosis. For example, squamous cell carcinomas of the tongue have been shown to be different from those of the cheek in previous studies 
, perhaps because different molecular genetic pathways are involved.
In conclusion, our study has confirmed in a large and homogeneous patient population that miR-195 expression was decreased in 80.2% of TSCC tumor samples compared with adjacent nonmalignant tissues and has provided evidence that miR-195 may be an independent biomarker of clinical prognosis among TSCC patients. Moreover, the anti-tumor effects of miR-195 in TSCC may be partially mediated by its inhibition of Cyclin D1 and Bcl-2 expression. Because miR-195 appears to have an anti-tumor effect in TSCC cell lines and has potential as a prognostic biomarker, it will be interesting in future experiments to further define the role of miR-195 in TSCC development.