EMT process in bone metastasis

Introduction

Epithelial—mesenchymal transition (EMT) was originally described as a fundamental process essential for developmental programs including formation of mesoderm, neural crest, heart valve, palatogenesis, and myogenesis [ ]. During the EMT process, the epithelial cells transition from polarized, cobblestone-like cells to migratory, spindle-shaped mesenchymal cells. In addition to morphological changes, cells undergoing EMT also exhibit changes at the molecular level by losing expression of epithelial markers such as E-cadherin, occludin, and EpCAM, and gaining expression of mesenchymal markers such as N-cadherin, vimentin, and fibronectin. Several signaling pathways (including TGFs, BMPs, FGF, EGF, HGF, Wnt/beta-catenin, and Notch) regulate EMT and involve both transcriptional and posttranscriptional processes.

The role of EMT in development is firmly established. Recent studies have begun to demonstrate that the EMT program is hijacked by tumor cells and contributes to tumor progression especially during the development of metastasis and therapy resistance [ , , ]. Elegant studies have provided compelling evidence that EMT program is associated with multiple critical steps in tumor progression. However, controversies about exact contributions and necessities of EMT in metastasis formation have also been evoked [ , ]. In this chapter, we will focus on EMT occurring in cancer progression, with emphasis on the development of bone metastases, and discuss opportunities for novel antimetastatic approaches for cancer therapy.

EMT in physiological processes and cancer

Based on the physiological or pathophysiological context, EMT has been categorized into three subtypes: developmental (type I), fibrosis and wound healing (type II), and cancer (type III).

It is important to understand the physiological process of EMT in development since the markers and critical signaling pathways are usually shared in all EMT processes, which have been described in excellent reviews [ ]. In brief, the type I EMT occurs during embryo development including embryo implantation, placenta formation, gastrulation, and neural crest formation [ ]. Through EMT, the fertilized egg develops into multigerm layers; the epithelial cells of the epiblast migrate and differentiate into mesoderm, and subsequently endoderm cells. Of note, the type I EMT is a well-programmed change in embryo development and does not involve any pathological events. In addition, following the primary EMT, the mesenchymal cells are able to revert to epithelial phenotypes, through mesenchymal-to-epithelial transition (MET), which is critical for organ formation [ ]. The type II EMT is involved in tissue regeneration and organ fibrosis. This refers to a pathological condition where inflammatory cells release EMT-inducing factors such as TGF-β, PDGF, FGF, and MMPs, and stimulate normal functional epithelial cells to undergo EMT, leading to extensive organ fibrosis [ ]. Type III EMT is involved in cancer development and will be the focus of this chapter.

EMT in the primary tumor and metastatic dissemination

Since the first description of EMT in cancer progression, EMT has been inherently related to metastasis [ , ]. EMT bestows tumor cell features including enhanced mobility, invasion, and resistance to apoptotic stimuli. All these features will benefit tumor cells to invade, survive, and establish metastasis in distant organs. Furthermore, as a result of EMT, tumor cells acquire chemoresistance and exhibit stemness, with increased potential for initiating secondary tumors [ ].

Recently, there has been an intense debate about the necessity of EMT in metastasis formation in vivo . An intravital imaging approach was used to show that single breast cancer cells gained mobility for hematogenous metastasis by activating EMT-promoting TGF-β-Smad2/3 signaling [ ]. Indeed, EMT was also observed during metastasis in spontaneous tumor models in mice, where disseminated tumor cells (DTCs) in the lung of MMTV-PyMT transgenic mice expressed a mesenchymal marker, FSP-1, suggesting the involvement of EMT in tumor dissemination [ ]. Using the DMBA/TPA-induced skin tumor model, activation of Twist, which is one of the major EMT-promoting signals, contributed to tumor invasion, tumor cell dissemination, and extravasation [ ]. Direct evidence of EMT has also been shown in a K-Ras mediated spontaneous pancreatic tumor model, which develops liver metastases [ ]. At the early stage during the primary tumor development, even before malignancy could be detected by rigorous histologic analysis, EMT tumor cells were detected in the primary lesions. These post-EMT tumor cells gained expression of typical mesenchymal markers including fibronectin, Zeb-1, and FSP-1, and lost expression of E-cadherin. Importantly, the post-EMT tumor cells represent the majority of metastatic tumor cells that seeded the metastatic liver. However, these approaches have been limited to snapshots of EMT events in discrete windows of the metastatic cascade in vivo including the primary tumor site, peripheral circulation, and distant metastatic organs, have been used as evidence of a central role for EMT in the multistep cascade of metastasis, and the fate of the epithelial cancer cell from the primary tumor to the metastatic site in the context of epithelial-to-mesenchymal plasticity (EMP) warrants investigation. Accordingly, recent studies have begun to employ rigorous lineage tracing approaches to actually demonstrate the process of EMT in vivo . For example, using an EMT lineage tracing strategy of mesenchymal-specific Cre-mediated β-galactosidase activity, Trimboli et al. compared the incidence of EMT events in three different oncogene-driven breast tumor models [ ]. Significantly, post-EMT tumor cells were detected in the Myc-driven tumors, but not in the PyMT- or Neu-driven tumors. Notably, lung metastases were formed in almost all MMTV-PyMT and MMTV-neu mice, but not in MMTV-myc animals, suggesting that the contribution of EMT in metastasis may be tumor type specific. Recently, similar EMT lineage tracing models using Cre-mediated switch of fluorescent markers were established in MMTV-PyMT and MMTV-Neu mice [ ]. Surprisingly, lung metastases appeared to be mostly derived from tumor cells persisting with their epithelial phenotype, rather than the post-EMT cells. Consistently, Slug (Snai2) and Twist EMT programing was also found to be dispensable for metastasis formation in pancreatic cancer models that metastasize to the liver [ ]. Inhibiting EMT by overexpression of miR-200 family members or specific knockout EMT transcription factor (Snail/Snai1) did not impair metastasis [ , ]. These findings evoked vigorous debate about whether EMT is a prerequisite for metastasis [ , , , ]. Of note, it is undeniable that EMT-associated properties would contribute to tumor metastasis as evidenced by both in vitro and in vivo studies. However, due to the complexity of EMT program, variety of EMT status, and dynamic features EMT process, inconsistent observations have been acquired with different tumor models. The controversies about EMT contribution in metastasis are also partially due to the unstandardized definition of EMT and potential methodological thresholds, as discussed in the recent reviews [ , ]. Current data suggest that the EMP which defines the capabilities of tumor cells to adapt their EMT status according to microenvironment challenges may better fit the requirement for metastasis than EMT phenotypes themselves [ ]. Indeed, single-cell RNA-seq analysis of Tri-PyMT cells showed spectrum of EMT phenotypes, with EMT-related genes concomitantly expressed with the activation of GFP, and defined the pre-EMT and post-EMT compartments within the breast tumor [ ].

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here