Embryonic stem (ES) cells are pluripotent and capable of self-renewal, thus holding promise for regenerative medicine. Recent studies have begun to provide insights into the molecular mechanisms underlying pluripotency and self-renewal. In this article, we discuss the roles of transcriptional regulation, epigenetic regulation and miRNAs in the maintenance of pluripotency and the differentiation of ES cells.
Embryonic stem (ES) cells are derived from the inner cell mass (ICM) of the pre-implantation embryo (1–3) and can give rise to all tissue lineages of the three primary germ layers, a property known as pluripotency. Due to their pluripotency, ES cells hold great promise for basic studies of tissue formation and cell replacement therapy. To achieve the ultimate goal of clinical application, ES cells need to be differentiated efficiently into a specific lineage and undifferentiated ES cells have to be eliminated from the differentiated cells. Thus, it is important to understand the molecular mechanisms underlying the pluripotency of ES cells, to better appreciate, and make use of, the distinctions between the pluripotent and differentiated states.
Another property of ES cells is their continuous self-renewal, which requires that the unique transcriptional profile of the pluripotent state be maintained. In contrast, to differentiate into various cell lineages, ES cells must shift to alternative transcriptional profiles. Both transcriptional regulation and epigenetic regulation play pivotal roles in maintaining the existing transcriptional profile and controlling the plasticity of the transcriptional profile. Additionally, microRNAs (miRNAs), novel regulators of gene expression, are emerging as key regulators of pluripotency. Here, we discuss three aspects of the molecular mechanisms of pluripotency, transcriptional regulation, epigenetic regulation and miRNAs.
THE REGULATORY NETWORK OF PLURIPOTENCY FACTORS
In early experimental approaches to address the molecular mechanism of pluripotency, it was hypothesized that pluripotent cells express a unique set of factors that underlie their ‘stemness’. These so called pluripotency factors not only serve as markers of pluripotent cells, but also are functionally important for pluripotency maintenance. To test this idea, some early studies compared global transcriptional profiles of ES cells and their differentiated derivatives (4–6). However, results of these studies varied significantly and the majority of candidate genes have not been functionally validated (7,8). Although global microarray comparisons provide few clear insights into pluripotency, genetic studies focused on individual genes were successful in identifying critical pluripotency factors. Two homeodomain transcription factors, OCT4 and NANOG, were the first proteins identified as essential for both early embryo development and pluripotency maintenance in ES cells (9–11). To better understand the roles of OCT4 and NANOG in pluripotency maintenance, two groups performed genome-wide binding site analyses of OCT4 and NANOG in mouse and human ES cells (12,13). SOX2, an HMG-box transcription factor, was also subjected to chromatin immunoprecipitation coupled with microarray assay (ChIP-chip) in human ES cells, as Sox2 heterodimerizes with Oct4 to regulate several genes in mouse ES cells (12). OCT4, SOX2 and NANOG share a substantial fraction of their target genes. Interestingly, these three factors demonstrate autologous feedback and control one another's transcription in a large regulatory circuit. Many targets of OCT4, SOX2 and NANOG encode key transcription factors for differentiation and development, and are transcriptionally inactive in ES cells. OCT4, SOX2 and NANOG also regulate transcriptionally active genes involved in pluripotency maintenance (12,13).
In addition to Oct4, Sox2 and Nanog, many other factors required for pluripotency have been identified, including Sall4, Dax1, Essrb, Tbx3, Tcl1, Rif1, Nac1 and Zfp281 (13–15). These pluripotency factors regulate each other to form a complicated transcriptional regulatory network in ES cells (16). For example, Sall4, a spalt family member, interacts with Nanog and co-occupies Nanog and Sall4 enhancer regions. Additionally, Sall4 also regulates Oct4 expression by binding to the Oct4 promoter (17,18). Essrb and Rif1 are primary targets of both Oct4 and Nanog (13).
Beyond their DNA-binding activities, these pluripotency proteins are extensively interconnected via protein–protein association. Starting with the pluripotency factor Nanog, followed by iterative tagging and purification of Nanog-associated proteins, a protein interaction network in mouse ES cells has been constructed (14). This mini-interactome is highly enriched for proteins that are required for the survival or differentiation of the ICM and for early development. Many of the genes encoding proteins in the interaction network are targets of Nanog and/or Oct4, suggesting that the transcriptional network might have a feedback mechanism through the protein interaction network. The protein interaction network is linked to several cofactor pathways largely involved in transcriptional repression (14). These data support a model wherein essential factors maintain the pluripotent state by simultaneously activating genes involved in pluripotency and repressing genes important for development.
Recent success in reprogramming somatic cells with specific genes further confirms the essential roles of these factors in pluripotency. Oct4, Sox2, c-Myc and Klf4 together can reprogram mouse embryonic and adult fibroblast cells to a pluripotent state (19–21). These same four factors have also been proven capable of reprogramming human dermal fibroblasts (22,23), whereas another group showed that OCT4, SOX2, NANOG and LIN28 were sufficient to establish pluripotent cells from human somatic cells (24). How these proteins work together to induce pluripotency remains elusive, but in addition to their direct functions in transcriptional regulation, it is likely that they interact with chromatin remodeling factors and histone modifying enzymes to modulate chromatin conformation.
EPIGENETIC REGULATION OF CHROMATIN IN PLURIPOTENT CELLS
Given the close association of transcription factors with pluripotency, we are left to wonder whether other classes of proteins or non-protein factors contribute to pluripotency? As the substrate of transcription, chromatin is subjected to various forms of epigenetic regulation that modulate the transcriptional activity in specific genomic regions, including chromatin remodeling, histone modifications, histone variants and DNA methylation. For example, trimethylation of lysine 9 and lysine 27 of histone 3 (H3K9 and H3K27) correlate with inactive regions of chromatin, whereas H3K4 trimethylation, and acetylation of H3 and H4 are associated with active transcription (25), and DNA methylation generally represses gene expression (26). Given that ES and somatic cells contain the identical genomic DNA (with few exceptions), epigenetic regulation is one of the major influences on their differentiation potential and, furthermore, is important for pluripotency.
To maintain pluripotency in ES cells, genes whose up-regulation leads to differentiation should be inactive. Polycomb group proteins (PcG) play important roles in silencing these developmental regulators. The PcG proteins function in two distinct Polycomb Repressive Complexes, PRC1 and PRC2. Genome-wide binding site analyses have been carried out for PRC1 and PRC2 in mouse ES cells and for PRC2 in human ES cells (27,28). The genes regulated by the PcG proteins are co-occupied by nucleosomes with trimethylated H3K27. These genes are transcriptionally repressed in ES cells and are preferentially activated when differentiation is induced. Many of these genes encode transcription factors with important roles in development. Interestingly, the pluripotency factors OCT4, SOX2 and NANOG co-occupy a significant fraction of the PcG protein regulated genes (27,28). These data suggest that the PcG proteins may facilitate pluripotency maintenance by suppressing developmental pathways.
Developmental regulators inactive in ES cells require activation upon differentiation. ES cells possess specific mechanisms to ensure that these genes are potent for activation. The recently discovered ‘bivalent’ histone code keeps its target gene in a state ‘poised’ for transcription (29,30). The bivalent domain has both repressive and active histone markers: a large region of H3K27 trimethylation harboring a smaller region of H3K4 trimethylation. In ES cells, bivalent domains are frequently associated with developmentally regulated transcription factors that are expressed at low levels. Upon differentiation, most of the bivalent domains become either H3K4 methylated or H3H27 methylated, consistent with associated changes in gene expression (29). Although the bivalent histone code primarily regulates key developmental transcription factors, some tissue-specific genes, such as Ptcra, Il12b and Alb1, are controlled by windows of unmethylated CpG dinucleotides and putative ‘pioneer’ factors in ES cells. These tissue-specific genes are silenced in ES cells, and most of the CpG dinucleotides in their promoter and enhancer regions are methylated. The unmethylated windows are located in the silent enhancers where the binding of transcription factors is required for maintaining the unmethylated state. These unmethylated windows are necessary for the activation of tissue-specific genes in differentiated cells (31).
Beyond the specific regulations of development-related genes, ES cells maintain chromatin in a highly dynamic and transcriptionally permissive state. First, fewer heterochromatin foci are detected in ES cell nuclei, where they appear to be more diffuse than those in differentiated cells. Second, fluorescence recovery after photobleaching and biochemical analyses reveal that compared with differentiated cells, ES cells have an increased fraction of loosely bound or soluble architectural chromatin proteins, including core and linker histones, as well as the heterochromatin protein HP1. A hyperdynamic chromatin structure is functionally important for pluripotency maintenance, as restriction of the dynamic exchange of the linker histone H1 prevents ES cell differentiation (32). Third, the status of histone modifications also indicates that the chromatin in ES cells is more transcriptionally permissive than in differentiated cells. Consistent with the global dynamics of chromatin, ES cell differentiation is associated with a decrease in global levels of active histone marks, such as acetylated histone H3 and H4, and an increase in repressive histone marks, specifically histone H3 lysine 9 methylation (32,33). Such a highly dynamic and transcriptionally permissive chromatin environment may facilitate rapid transcriptional profile alternations upon differentiation and allow various transcriptional profiles to be established.
PLURIPOTENCY FACTORS AND THE EPIGENETIC REGULATION OF CHROMATIN
Both pluripotency factors and epigenetic regulators provide fundamental mechanisms underlying pluripotency. Both pathways also engage in cross-talk with one another in order to maintain pluripotency. First, pluripotency factors regulate genes encoding epigenetic control factors. It has been shown that OCT4, SOX2 and NANOG co-regulate certain genes encoding components of chromatin remodeling and histone modifying complexes, such as SMARCAD1, MYS3 and SET (12). Second, pluripotency factors also interact with histone modifying enzymes and chromatin remodeling complexes. Nanog and Oct4 interact directly or indirectly with the histone deacetylase NuRD (P66b and HDAC2), polycomb group (YY1, Rnf2 and Rybp) and SWI/SNF chromatin remodeling (BAF155) complexes (14). Finally, the genes of pluripotency factors are subjected to epigenetic regulation. Good examples of this are two histone demethylase genes, Jmjd1a and Jmjd2c, which are downstream targets of Oct4 (13,34). Jmjd1a acts as a positive regulator of the pluripotency-associated genes, Tcl1, Tcfcf2l1 and Zfp57, by demethylating H3K9Me2 at the promoters. Jmjd2c removes H3K9Me3 marks at the Nanog promoter to positively regulate Nanog expression (34).
MICRO-RNAs AND PLURIPOTENCY
Recent studies have discovered that non-coding RNA comprises a large fraction of vertebrate transcriptomes. Although not all non-coding RNAs are functional, many play important regulatory roles. MicroRNA is a family of small non-coding RNAs of ∼22 nt in length. They regulate gene expression via at least two distinct mechanisms: degradation of target mRNA transcripts and inhibition of mRNA translation (35). MicroRNA expression profiles in both human and mouse ES cells revealed that ES cells express a unique set of miRNAs, and that these miRNAs are down-regulated as ES cells differentiate into embryoid bodies. Some of these miRNAs are conserved between human and mouse and are clustered in the genome (36,37). These data suggest that miRNAs might play a role in the maintenance of pluripotency in ES cells.
MicroRNAs also appear to be important for ES cell differentiation. Knockout of Dicer, an RNase III-family nuclease critical for RNA interference (RNAi) and miRNA generation, compromises ES cell proliferation and renders ES cells deficient in differentiation (38,39). Because Dicer is essential for both RNAi and miRNA generation, this experiment does not define which pathway is responsible for the defect in ES cells. To investigate the specific role of miRNAs in pluripotency, DGCR8, an RNA-binding protein that assists the RNase III enzyme Drosha in the processing of miRNA, was knocked out in mouse ES cells. Loss of DGCR8 results in a complete absence of mature miRNAs, though the RNAi pathway is not affected. When induced to differentiate, DGCR8-deficient ES cells fail to fully down-regulate pluripotency markers and retain an ES cell colony morphology. Nevertheless, they do express some markers of differentiation, confirming the specific role of miRNAs in ES cell differentiation (40).
As some miRNAs promote ES cell differentiation, one would reasonably expect them to be regulated like development-related genes, i.e. to be absent in ES cells and then up-regulated upon differentiation. Interestingly, the expression of miRNAs is regulated at both the transcriptional and post-transcriptional level. Extensive post-transcriptional regulation of miRNA processing in mouse ES cells has been observed in vitro as well as during early mouse development in vivo. Many miRNA primary transcripts, including members of the Let-7 family, are present at high levels, but are not processed by Drosha in ES cells. As ES cells differentiate, primary miRNA transcripts are processed to make mature miRNAs which then facilitate differentiation (41,42). LIN28, one of the four factors used to reprogram human somatic cells, has been shown to be involved in blocking miRNA processing in ES cells (43).
Despite the importance of miRNA in pluripotency maintenance, detailed mechanisms describing how miRNAs regulate pluripotency remain elusive. MicroRNAs might facilitate differentiation by down-regulation of pluripotency-associated genes, thereby permitting cells to passively drift towards differentiation. It has been shown that the microRNA miR-134 promotes ES cell differentiation into the ectodermal lineage, partly due to its direct translational attenuation of Nanog and LRH1 (44). Further studies of miRNAs in the self-renewal and differentiation of ES cells are required to elucidate the functions of miRNAs in both the maintenance and inhibition of pluripotency.
Our knowledge of the molecular mechanisms of pluripotency has advanced greatly in the past few years. It has become increasingly clear that both transcriptional and epigenetic regulations are two essential mechanisms underlying pluripotency. Figure 1 summarizes the key regulatory networks of pluripotency. In undifferentiated ES cells, pluripotency factors work together with epigenetic regulators to activate genes involved in pluripotency maintenance and to suppress development-related genes. Although genes controlling differentiation are transcriptionally inactive, they are nevertheless maintained in a potent state for transcriptional activation. Also, a hyperdynamic chromatin structure in ES cells contributes to a rapidly plastic transcriptional profile, thereby facilitating their robust multi-lineage differentiation potential. Upon differentiation, a diminution of pluripotency factors facilitated by miRNAs leads to dramatic changes in the transcriptional profile, where genes required for pluripotency maintenance become silenced. Differentiation-related genes are regulated in multiple ways, where some are activated while the remainder remains inactive. Among the inactive genes, some will lose the unique epigenetic mark(s) associated with transcriptional activation potential and will thus become resistant to activation. Meanwhile, the chromatin in differentiated cells becomes more compact, which correlates with decreased differentiation potential and lineage restriction.
To further understand the molecular mechanisms of pluripotency, studies on how the chromatin remodeling and histone modifying complexes modulate the global and regional chromatin structure in pluripotent cells are required. Given that somatic cells can now be reprogrammed into a pluripotent state with defined factors, we are well poised to investigate pluripotency from a different angle, where fresh perspectives almost certainly shall offer new and potentially unexpected insights. Chief among the questions to be answered is how these factors regulate transcription during reprogramming and how they interact with the chromatin remodeling and histone modifying complexes to reprogram the chromatin into the pluripotent state. Taken together, improved understanding of pluripotency should allow us to derive pluripotent cells more efficiently. In addition to their continued use to explore fascinating questions of basic science within the field of development, our hope is that 1 day such insights will lead to clinical applications in regenerative medicine.
The principal investigator's laboratory is supported by grants from the NIH and the NIH Director's Pioneer Award of the NIH Roadmap for Medical Research, the Burroughs Wellcome Fund, the Leukemia and Lymphoma Society, the Harvard Stem Cell Institute, the Children's Hospital Stem Cell program, and the Howard Hughes Medical Institute.
We thank M.W. Lensch for comments on the manuscript.
Conflict of Interest statement. None declared.