br Conclusion and future directions br Conflict

Conclusion and future directions
Conflict of interest
Acknowledgments
Introduction Epithelial-mesenchymal transition (EMT) is a process by which epithelial learning pathways lose their polarity and are converted to a mesenchymal phenotype; recently, this has been regarded as the critical event that can induce morphogenetic changes during embryonic development, organ fibrosis and tumor metastasis. Phenotypic changes of EMT include the downregulation of epithelial markers (e.g., E-cadherin, desmoplakin and plakoglobin) and upregulation of mesenchymal markers (e.g., vimentin, fibronectin and α-smooth muscle actin). The initiation of EMT is hallmarked by the repression of E-cadherin. Several mechanisms including transcriptional repression and promoter hypermethylation are shown to repress the expression of CDH, the gene encoding E-cadherin, and induce EMT. A number of transcription factors, including Snail (also known as Snail1), Slug (also known as Snail2), Twist1 (also known as Twist), Zeb1, SIP1 and E47, participate in the transcriptional repression of CDH1 during EMT. In addition, emerging evidence highlights the role of chromatin modification in CDH1 repression: Snail interacts with HDAC1-HDAC2-SIN3A complex to facilitate the CDH1 repression through chromatin modification. We previously also demonstrated that regulation of the chromatin modifier Bmi1 by Twist1 contributes to the suppression of CDH1 by Twist1. In human cancers, accumulated evidence suggests the critical role of EMT in the progression and dissemination of different types of cancers. Notably, the increased expression levels of EMT regulators (e.g., Twist1, Snail, Slug, SIP1, E47, Zeb1) have been reported to be associated with an aggressive phenotype and worse prognosis in different cancers, including HNSCC. Recent studies indicate that hypoxia is one of the major causes contributing to the progression, metastasis and treatment resistance of HNSCC. The linkage between EMT and hypoxia for promoting invasive phenotype of cancers has become clear in recent years. We recently demonstrated that direct regulation of TWIST1 by hypoxia-inducible factor-1 (HIF-1) promotes metastasis. Furthermore, activation of other EMT regulators, including Snail, SIP1, Zeb1 and E47 has also been shown under hypoxic conditions. Although the importance of hypoxia-induced EMT has been highlighted, the interplay between different EMT regulators and their unique roles in hypoxia-mediated EMT has never been previously addressed. MicroRNAs are small non-coding RNA molecules that suppress gene expression by interacting with the 3′-untranslated regions (3′-UTRs) of target mRNAs. MicroRNAs have been shown to be involved in a wide range of pivotal biological processes in mammalian cells, including cell growth, development, proliferation, differentiation and death. Regarding the role of microRNA in cancer, more than 50% of annotated human microRNA genes are located in fragile chromosomal regions that are susceptible to amplification, deletion or translocation in the tumor development process. Recent evidence has indicated that some microRNAs function either as oncogenes or as tumor suppressors, and expression profiling has revealed characteristic microRNA signatures in certain human cancers. The microRNA-29 family members, including miR-29a, miR-29b, and miR-29c, have been shown to play the critical role of tumorigenesis and metastasis in different kinds of human cancer. Some reports suggest the oncogenic role of miR-29 in cancer cells, in that miR-29a induces self-renewal capacity in hematopoiesis and initiates acute myeloid leukemia (AML). Also, miR-29a suppresses the epithelial polarity and promotes metastasis of breast cancer cells. In contrast, the tumor suppressive effect of miR-29 has also been reported. It is well-established that miR-29s enhances apoptosis of hepatocellular carcinoma (HCC) cells, and downregulation of miR-29s is associated with a worse survival of HCC patients. Furthermore, miR-29b induces global methylation through directly targeting DNMT3A, 3B, and indirectly targeting DNMT1 in AML; miR-29b modulates KIT expression by targeting SP1 and is also down-regulated by aberrant expression of KIT. Therefore, the definitive effect of the miR-29 family on different types of human cancers remains inconclusive.