Introduction The mechanisms by which pluripotency regulators
The mechanisms by which pluripotency regulators dictate self-renewal versus differentiation are only recently being studied (Wagner et al., 2010; Niwa et al., 2009). In addition, various approaches are being developed to examine such processes (Wang et al, 2013). The transcriptional repressor REST was discovered as a major regulator of neuronal genes in non-neural cells, but it is now known to repress many genes with diverse functions and, thus, to regulate both normal development and disease states (Ballas and Mandel, 2005; Kagalwala et al., 2008; Gopalakrishnan, 2009; Buckley et al., 2010; Gao et al., 2011; Roopra et al., 2012; Hwang et al., 2013; Negrini et al., 2013). REST is a major buy PD 173074 modifier, and it accomplishes this role by interacting with various cellular cofactors in a context-dependent manner (Ballas and Mandel, 2005; Qureshi et al., 2010; Bithell, 2011; Gao et al., 2012). Its context-dependent function is also observed in its role in cancer; REST has an oncogenic function in neural cells, in which it is normally not expressed, and a tumor suppressor function in non-neural cells, in which it is normally expressed (Kagalwala et al., 2008). In mESCs, REST is expressed at very high levels (Ballas et al., 2005) and genome-wide analysis found it to be a part of the mESC pluripotency network (Boyer et al., 2005; Johnson et al., 2008; Kim et al., 2008; Wang et al., 2006). Experimentally, REST was also found to regulate mESC self-renewal and pluripotency by repressing transcription of the miR-21 gene, and loss of REST produced expression of differentiation markers (Singh et al., 2008). However, whereas some studies found that loss of REST did not cause a change in the expression of differentiation markers (Buckley et al., 2009; Jorgensen et al., 2009; Jorgensen and Fisher, 2010), others found that loss of REST actually caused decreased expression of differentiation markers (Yamada et al., 2010), suggesting a gap in our knowledge of REST-mediated mESC pluripotency. More recently, it has been shown that this apparent discrepancy results from the context-dependent function of REST (Singh et al., 2012). Here we show that, under permissive conditions, REST–microRNA (miR)-21–Sox2 signaling cascade maintains self-renewal in E14Tg2a.4 cells.
Discussion Here, we propose that under permissive conditions, miR-21 suppresses Sox2 expression and the REST–miR-21–Sox2 axis regulates self-renewal and pluripotency in E14Tg2a.4 mESCs. The suppressive effect of miR-21 has been recently seen in other cell types (Trohatou et al., 2014; Ni et al., 2014). Because pluripotency is a critical decision-making step for mESCs, one would expect that several counteracting mechanisms would modulate it so that the mESC pluripotency is maintained in equilibrium with environmental cues. We suggest that while Oct4, Sox2, and Nanog serve as core pluripotency factors, REST and miR-21 serve as maintenance factors in the mESC pluripotency network (Fig. 4E). This classification is based on the impact of these factors on the normal development of blastocysts (Supplementary Fig. S2). Mice haploinsufficient in either the core or maintenance factors are apparently normal. However, the distinction between the core and maintenance factors becomes apparent in null mice. Mouse embryos deficient in the core self-renewal regulators (Oct4−/−, Nanog−/−, and Sox2−/−) are immediately affected; these embryos develop to the blastocyst stage but fail to develop further under normal conditions (Avilion et al., 2003; Chambers et al., 2003; Mitsui et al., 2003; Nichols et al., 1998). In contrast, mouse embryos deficient in maintenance factors show an apparent lack of an immediate effect on blastocyst development. Rest (−/−) mice survive past the blastocyst stage but show progressive embryonic lethality between embryonic days 9.5 and 11 (Chen et al., 1998), and miR-21−/− mice are phenotypically normal (Ma et al., 2011). This scenario is very similar to the other maintenance factors, LIF, Stat3 and Myc. LIF−/− mice are viable and show only retarded postnatal growth (Stewart et al., 1992), LIFRβ−/− mice die only after birth (Li et al., 1995), gp130−/− mice die between 12.5days postcoitum and term (Nichols et al., 2001), Stat3−/− embryos develop until embryonic day 6 and then degenerate between embryonic days 6.5 and 7.5 (Takeda et al., 1997), c-Myc mice develop until embryonic day 10.5 (Davis et al., 1993). These results indicate that the maintenance factors are not needed for the regulation of blastocyst development under normal conditions. Our results suggest that loss of these maintenance factors can be compensated for by other known, or unknown, factors in the network, including extracellular cues provided by feeder cells or extracellular matrix components such as laminin (Singh et al., 2012).