Principles of stem cell biology and cancer: future applications and therapeutics. Edited by T. Regad, T. J. Sayers and R. C. Rees. John Wiley & Sons (2015)
Part II. Cancer stem cells
CSCs are important contributors to tumour recurrence and metastasis (Figure 6.3). They are functionally defined as a subset of tumour cells that are capable of self-renewal and that, when seeded at limiting dilutions in immunocompromised mice, will form new tumours with a cellular heterogeneity similar to that of the primary tumour (Chang and Mani, 2013; Marie-Egyptienne et al., 2013). CSCs are generally identified using characteristic cell-surface markers, the most common of which are CD44 (often in combination with CD24), CD133 and EpCAM (Hill et al., 2009; Scheel and Weinberg, 2012; Marie-Egyptienne et al., 2013). It is important to recognize, however, that universal CSC cell-surface markers do not exist, the functional relevance of these markers to the biology of CSCs is not understood and the precise markers used to label CSCs are tissue type-dependent (Chang and Mani, 2013; Marie-Egyptienne et al., 2013). Despite these complexities, cell-surface markers have been used to isolate CSCs from many solid tumours, including breast, brain, prostate, lung, pancreas and ovarian cancers (Marie-Egyptienne et al., 2013). The CSC phenotype has several characteristics, including the capacity for self-renewal, expression of stem cell markers, increased tumourigenicity in vivo, enhanced survival and resistance to chemotherapy and radiotherapy. Furthermore, it is now well recognized that CSCs exhibit many of the characteristics associated with cancer cells that have undergone EMT, providing a strong biological link between EMT, CSCs and metastasis (Scheel and Weinberg, 2012).
It is becoming increasingly clear that strong connections exist between the CSC phenotype and cancer cells that have undergone EMT (Scheel and Weinberg, 2012). Several studies have shown that the activation of EMT by the inducer TGF-β and the transcriptional activators Snail1, Twist1 and Zeb1 in normal epithelial cells results in the acquisition of CSC traits, including a stem cell surface-marker profile, a mesenchymal phenotype, self-renewal capacity and increased tumourigenicity (Chang and Mani, 2013; Tam and Weinberg, 2013). A number of molecular components common to the EMT programme and the CSC state have been identified. For example, Zeb1 inhibits the expression of miR-200 family members, which are involved in the suppression of genes involved in stemness and EMT, including the polycomb protein Bmi1 and the histone-modifying enzyme Suz12. Furthermore, Zeb1 and miR-200 family members negatively regulate each other through a double-feedback loop, suggesting that the inhibition of epithelial differentiation and the acquisition of CSC traits are interconnected (Scheel and Weinberg, 2012). Furthermore, the Wnt-β-catenin signalling pathway, which is a prototypic component of the EMT programme, is active in CSCs in several types of cancer, including breast and colon cancer. The Wnt-β-catenin signalling pathway is involved in regulating the activity of Snail, which in turn represses E-cadherin expression, resulting in a feedback loop that reinforces Wnt-β-catenin activity. It is a critical inducer of EMT in carcinomas, and is also important for the maintenance of CSCs, indicating a common mechanistic connection between CSCs and cells that have undergone EMT.
An inherent consequence of the EMT process is that it imparts epithelial cancer cells with traits that are associated with increasingly malignant cell behaviours, including motility, invasiveness and resistance to apoptosis, the combination of which allows for metastatic dissemination. In addition, passage through an EMT imparts tumour-initiating properties to cancer cells – traits that may be critical for the ability of cancer cells to seed distant sites, resulting in the growth of clinically significant macrometastases (Tam and Weinberg, 2013).
Like the concept of epithelial – mesenchymal plasticity (see Section 6.4), it is evident that stemness may also be a flexible or plastic state, and that the microenvironment may be an important regulator of this flexibility. Certainly, studies have linked the CSC phenotype with cancer cells undergoing EMT (Scheel and Weinberg, 2012). The question of whether bulk tumour cells can become CSCs remains hotly debated, but CSCs that arise from the bulk tumour population may get there via EMT (Scheel and Weinberg, 2012). Current evidence points to a model whereby cancer cells co-opt an already established cell biological programme present in normal epithelial cells to provide a source of CSCs by enabling dedifferentiation of cancer epithelial cells within carcinomas, suggesting that the stemness state of CSCs may exhibit considerable plasticity. However, the extent to which CSCs derived from progenitor stem cells and those derived from differentiated cells through an EMT are similar or different remains to be rigorously investigated (Nieto, 2013).