Consistently, the high metastasis of cancer cells is the bottleneck in the process of tumor treatment. In this process of metastasis, a pivotal role is executed by epithelial–mesenchymal transition (EMT). The epithelial-to-mesenchymal transformation was first proposed to occur during embryonic development. Later, its important role in explaining embryonic developmental processes was widely reported. Recently, EMT and its intermediate state were also identified as crucial drivers in tumor progression with the gradual deepening of research. To gain insights into the potential mechanism, increasing attention has been focused on the EMT-related transcription factors. Correspondingly, miRNAs target transcription factors to control the EMT process of tumor cells in different types of cancers, while there are still many exciting and challenging questions about the phenomenon of microRNA regulation of cancer EMT. We describe the relevant mechanisms of miRNAs regulating EMT, and trace the regulatory roles and functions of major EMT-related transcription factors, including Snail, Twist, zinc finger E-box-binding homeobox (ZEB), and other families. In addition, on the basis of the complex regulatory network, we hope that the exploration of the regulatory relationship of non-transcription factors will provide a better understanding of EMT and cancer metastasis. The identification of the mechanism leading to the activation of EMT programs during diverse disease processes also provides a new protocol for the plasticity of distinct cellular phenotypes and possible therapeutic interventions. Here, we summarize the recent progress in this direction, with a promising path for further insight into this fast-moving field.

Nrf2 (nuclear factor E2-related factor 2, encoded by Nfe2l2) acts as a master transcriptional regulator in mediating antioxidant, detoxification, and cytoprotective responses against oxidative, electrophilic, and metabolic stress, but also plays a crucial role in cancer metabolism and multiple oncogenic pathways, whereas the redox sensor Keap1 functions as a predominant inhibitor of Nrf2 and, hence, changes in its expression abundance directly affect the Nrf2 stability and transcriptional activity. However, nuanced functional isoforms of Keap1 α and β have rarely been identified to date. Herein, we have established four distinct cell models stably expressing Keap1^−/−, Keap1β(Keap1^Δ1–31), Keap1-Restored, and Keap1α-Restored aiming to gain a better understanding of similarities and differences of two Keap1 isoforms between their distinct regulatory profiles. Our experimental evidence revealed that although Keap1 and its isoforms are still localized in the cytoplasmic compartments, they elicited differential inhibitory effects on Nrf2 and its target HO-1. Furthermore, transcriptome sequencing unraveled that they possess similar but different functions. Such functions were further determined by multiple experiments in vivo (i.e., subcutaneous tumour formation in nude mice) and in vitro (e.g., cell cloning, infection, migration, wound healing, cell cycle, apoptosis, CAT enzymatic activity, and intracellular GSH levels). Of note, the results obtained from tumourigenesis experiments in xenograft model mice were verified based on the prominent changes in the PTEN signaling to the PI3K-AKT-mTOR pathways, in addition to substantially aberrant expression patterns of those typical genes involved in the EMT (epithelial–mesenchymal transition), cell cycle, and apoptosis.