The flurry of recent publications regarding reprogramming of mature cell types to induced pluripotent stem cells raises the question: what exactly is pluripotency? A functional definition is provided by examination of the developmental potential of pluripotent stem cell types. Defining pluripotency at the molecular level, however, can be a greater challenge. Here, we examine the emerging list of genes associated with induced pluripotency, with particular attention to their functional requirement in the mouse embryo. Knowledge of the requirement for these genes in the embryo and in embryonic stem cells will advance our understanding of how to reverse the developmental clock for therapeutic benefit.
What is pluripotency?
Pluripotency can be defined in both functional and molecular terms. Functionally, pluripotency refers to the ability of a cell to give rise to cell types of all three germ layers of the embryo: ectoderm, mesoderm, and endoderm, as well as the germline. As such, the very early embryo contains a succession of cells with pluripotent capacity from blastocyst to gastrula stages, but none of these cells behave as an ongoing source of stem cells in the later embryo. Embryonic stem (ES) cells are also pluripotent, but exhibit an additional feature: unlimited proliferation in a pluripotent state. Molecular definition of pluripotency requires identification of molecules that support these functional properties. Identification of these has been a major focus in the fields of developmental and stem cell biology. Below we discuss sources of pluripotent stem cells, strategies for identifying molecular regulators of their establishment and maintenance, and the role of these genes during mouse development.
Different kinds of pluripotent stem cells exist
Pluripotent stem cells hold promise for regenerative medicine, since they theoretically provide an unlimited source of new tissue for therapeutic intervention. Additionally, pluripotent stem cells provide a unique perspective into the establishment and development of the tissues from which they originate. Understanding the molecular regulation of the origins of pluripotent stem cells can thus provide insight into normal development.
Pluripotent stem cell lines have been derived from the mouse embryo at several stages. All are capable of differentiating into cells representative of a variety of adult tissue types in various assays, including embryoid body, teratoma, and some can contribute to mouse development in chimeras (Fig. 1). However, in spite of these similarities, many differences have been noted among pluripotent stem cell types. Differences include morphology, gene expression profiles, growth factor requirements, and the ability to contribute to chimeras (Rossant 2008). The best-studied pluripotent stem cell is the ES cell, derived from mouse or human inner cell mass (ICM) at the blastocyst stage (Evans & Kaufman 1981, Martin 1981, Thomson et al. 1998). In addition, self-renewing, pluripotent stem cells, called epiblast stem cells, have been derived from the mouse embryo at a slightly later stage, when the ICM has become the epiblast (Brons et al. 2007, Tesar et al. 2007). Pluripotent mouse embryonic germ cells can be derived from embryo-derived primordial germ cells (PGCs; Matsui et al. 1992), and pluripotent cells have been derived from spontaneously arising embryonal carcinomas (EC cells). In addition, multipotent germline stem cells have been derived from neonatal and adult mouse testes cells (Kanatsu-Shinohara et al. 2004, Guan et al. 2006). Whether a common pluripotency/self-renewal pathway underlies establishment, maintenance, and regulation of all of these cell lines is an area of active research.
To this end, much can be learned through efforts to induce pluripotency and self-renewal in mature cell types by nuclear reprogramming. Initial reprogramming strategies relied on changing the environment of the entire mature cell nucleus, for e.g. by transplantation to an activated oocyte (Wilmut et al. 1997) or fusion with an ES cell (Tada et al. 2001). More recently, genetic strategies initially developed in the laboratory of Shinya Yamanaka have been used for nuclear reprogramming. The first efforts involved four genes (Pou5f1 (Oct4), Sox2, Klf4, and Myc (cMyc)) introduced into mature cell types using retroviral vectors (Takahashi & Yamanaka 2006). Subsequent efforts have used alternate vectors or have substituted drugs for some of these genes (Maherali & Hochedlinger 2008, Feng et al. 2009b). Importantly, genetic reprogramming has been used to generate iPS cells in other species, including human, monkey, and rat (Takahashi et al. 2007, Yu et al. 2007, Liu et al. 2008, Li et al. 2009), arguing that genetic reprogramming is not a mouse-specific parlor trick.
The existence of different kinds of pluripotent stem cells and different reprogramming strategies raises the interesting notion that pluripotency can be achieved in different ways. Recognizing and describing these inherent differences is essential for understanding what makes cells competent to respond to pluripotency factors. Use and comparison of these different cells may thus lead to identification of new molecular regulators of development and pluripotency. Several methods have been used to identify molecular regulators of pluripotency in both stem cell lines and embryos, and these are summarized next.
Methods for identifying pluripotency factors
Several methods have been used for identifying pluripotency factors in both embryos and stem cells. Studies in embryos have relied on candidate gene analysis, with examination of phenotypes resulting from knockout alleles in vivo. Studies in cell lines have been either functional or descriptive, but have been more amenable to more large-scale, whole genome approaches. Large-scale functional strategies have included RNAi screens and overexpression screens (Chambers et al. 2003, Ivanova et al. 2006), while large-scale descriptive approaches have used gene expression profiling, epigenetic profiling, and protein expression profiling (Boyer et al. 2005, Bernstein et al. 2006, Loh et al. 2006, Wang et al. 2006, Walker et al. 2007). These approaches have rapidly expanded our knowledge of molecules associated with pluripotency. In some cases, the functional importance of new genes identified in these studies has been further validated in ES cells. However, much remains to be learned of the functions of and epistatic relationships among these genes, not just in established ES lines, but also during normal development. These efforts, in contrast with high throughput approaches, require detailed mechanistic analysis performed in the embryo using knockout mouse lines. Although in vivo genetic approaches can be more time consuming than high-throughput in vitro approaches, they provide essential knowledge for which there is no substitute.
While the list of pluripotency-associated genes continues to grow, genes involved in cellular reprogramming warrant special attention. The ability of these genes to uniquely initiate a cascade of events leading to the pluripotent stem cell state underscores their fundamental importance at the top of a pluripotency hierarchy. Much consideration has been given to how and why reprogramming even works. Initial speculation raised cautionary caveats about the role of viral integration or cell type in reprogramming. However, non-integrating vectors, excisable vectors, and even soluble proteins, have since been used for factor delivery (Okita et al. 2008, Stadtfeld et al. 2008b, Kaji et al. 2009, Soldner et al. 2009, Woltjen et al. 2009, Zhou et al. 2009). In addition, reprogramming may not involve unique rare, reprogramming-competent cells since many different cell types have been used (Nakagawa et al. 2008, Shi et al. 2008, Silva et al. 2008, Stadtfeld et al. 2008a, Guo et al. 2009), although it is hard to rule out this latter possibility completely. Regardless of the origin of the parental cell, exogenous reprogramming factors are clearly required for expansion of iPS cells. A better understanding of the requirement for these genes in embryos and in ES cells is essential to discussion of how reprogramming works.
What are the reprogramming factors?
It all began with a list of ES cell associated transcripts (ECATs), a list of about 20 genes identified as enriched in mouse ES cells compared to other non-pluripotent cell types (Mitsui et al. 2003). In the first successful reprogramming study, pools of ECATs were systematically introduced into mouse fibroblasts, while selecting for expression of another ES cell-expressed gene. Eventually, a minimum combination of four transcription factors was identified: POU5F1 (Oct4), SOX2, KLF4, and MYC (cMyc; Takahashi & Yamanaka 2006). Later, MYC was found to be inessential (Nakagawa et al. 2008), although it plays a role in the efficiency and speed of reprogramming, which is about twice as slow without MYC (Nakagawa et al. 2008). Since this time, human fibroblasts have been reprogrammed with the same four factors (Takahashi et al. 2007, Wernig et al. 2007, Yu et al. 2007, Park et al. 2008), or with a slightly different combination of factors, including POU5F1, SOX2, NANOG, and the RNA-binding protein LIN28 (Yu et al. 2007). Subsequently, POU5F1 and SOX2 alone have been shown to reprogram human fibroblasts in the presence of a chemical inhibitor of the histone deacetylase valproic acid (Huangfu et al. 2008), suggesting that chromatin modification plays an essential role in cellular reprogramming. More recently, many other chemical regulators of chromatin-modifying machinery have also been shown to substitute for SOX2 or KLF4 or to increase efficiency in reprogramming mouse fibroblasts (Feng et al. 2009b).
While reprogramming factors reset the cellular phenotype from the inside, reprogramming clearly also requires extrinsic signals provided by the cell culture milieu, since all reprogramming events have been performed in ES cell culture conditions. These conditions include growth factors, cytokines, and other signals provided by the cell culture medium, fetal bovine serum, and feeder cells. Together, intrinsic and extrinsic factors maintain pluripotency and guarantee self-renewal. How extrinsic signals are integrated with intrinsically acting factors is not entirely clear, but is an active area of investigation. Possibilities include activation of self-renewal/proliferation genes or repression of prodifferentiation genes. For example, signaling by the cytokine LIF leads to activation of the transcription factor STAT3, which may regulate a suite of ES cell genes (Sekkai et al. 2005, Chen et al. 2008). Bone morphogenetic protein (BMP) or serum suppresses expression of the inhibitors of differentiation (Id) family of transcription factors (Ying et al. 2003). In addition, parallel signaling pathways have been found to connect LIF signaling and transcription factors maintaining pluripotency in ES cells (Niwa et al. 2009). While ES cell signaling pathways will not be discussed further, it is important to keep in mind that they play an essential role in supporting pluripotency. Rather, we will focus on factors delivered into the cell during reprogramming.
Early embryonic development
To understand the role of reprogramming factors in development, it is useful to begin with an overview of mouse development. Following fertilization, the zygote undergoes cleavages that increase cell number without affecting the total size of the embryo. This lasts throughout preimplantation development, and culminates with formation of the blastocyst (Fig. 2). By the blastocyst stage, the first three lineages are established: embryo, placenta, and extraembryonic endoderm (Yamanaka et al. 2006). Shortly after this stage, the blastocyst hatches from its protein coat, the zona pellucida, and implants in the uterine wall. Subsequent growth, patterning, and reorganization of all three lineages produce mature tissues of the fetus and accompanying extraembryonic tissues through a series of stereotyped steps. Cells on the outside of the blastocyst or trophectoderm will form trophoblast structures such as giant cells, extraembryonic ectoderm, chorion, ectoplacental cone, and eventually the placenta (Simmons & Cross 2005). Inside cells of the blastocyst's inner cell mass are fated to become either embryo or extraembryonic endoderm (Chazaud et al. 2006). The embryonic lineage, or epiblast, will go on to produce three germ layers during gastrulation, as well as some extraembryonic structures, while the extraembryonic endoderm will give rise to the endoderm of the yolk sac. Since ES cells are derived from the inner cell mass of the blastocyst, ES cell genes may also play an essential early role in development.
The list of genes regulating pluripotency in ES cells continues to grow at an exciting pace. However, relatively few of these genes have been demonstrated to possess the unique ability to participate to induce pluripotency, and some have been found not to promote reprogramming (Feng et al. 2009a). Arguably, genes sufficient to reprogram mature cell types could play a special upstream role in establishment of ES or ICM-like properties and are thus of particular interest to developmental biologists. We will therefore focus on the subset of pluripotency genes that have been shown to be dominantly capable of reprogramming mature cell types during induced pluripotency. Here, we will examine requirements for these reprogramming factors in mouse embryonic development and ES cell homeostasis.
POU5F1 (also known as Oct3, Oct4 and Oct3/4) is a POU-domain transcription factor encoded by the Pou5f1 gene. Given that POU5F1 plays essential roles in both the mouse embryo and in ES cells, the involvement of POU5F1 in reprogramming was not surprising. In embryos lacking Pou5f1, the inner cell mass degenerates, and neither embryonic nor extraembryonic endoderm cells survive (Nichols et al. 1998). Not surprisingly, ES cells cannot be derived from Pou5f1 mutants (Nichols et al. 1998). Moreover, genetic reduction of Pou5f1 in established ES cell lines leads to the formation of trophoblast-like cells (Niwa et al. 2000). Together, these observations have led to the proposal that POU5F1 represses the trophoblast program of development in ES cells and during early lineage decisions in the embryo.
POU5F1 plays different roles in stem cells at later stages in mouse development. At later stages, Pou5f1 is expressed in multipotent stem cells, including PGCs (Scholer et al. 1990), where it is required for survival (Kehler et al. 2004). These observations led to the proposal that POU5F1 promotes ‘stem cell-ness’ in general. However, Pou5f1 is not required for self-renewal of somatic stem cells in a wide variety of organs (Lengner et al. 2007). Thus, although POU5F1 is an essential reprogramming factor, enabling mature somatic cells to revert to an ES-like state, it does not appear to be required for somatic stem cell self-renewal. This observation underscores intrinsic differences in gain and loss of function assays. Notably, POU5F1 is sufficient to prevent differentiation and expand progenitor cell proliferations when overexpressed in adult epithelia, presumably due to expansion of adult stem cell populations (Hochedlinger et al. 2005). However, in spite of the proposed role for Pou5f1 in maintaining ES cell fates, overexpression of Pou5f1 in ES cells does not maintain their self-renewal in the absence of self-renewal factors such as LIF, but leads to their differentiation (Niwa et al. 2000). This phenotype was proposed to result from titration of other factors by overexpressed Pou5f1 in ES cells. Thus, functional differences in response to POU5F1 exist between adult and ES cells, but the molecular basis for these differences is not understood. One testable hypothesis is that different POU5F1 cofactors exist within ES and adult stem cell populations, and this could alter the response to ectopic POU5F1.
Several key POU5F1 cofactors operating in ES cells are known. According to the current view, POU5F1 acts together with SOX2 and NANOG at the core of a transcription factor network regulating pluripotency in ES cells. These three proteins have been found to co-occupy putative enhancer elements of genes thought to promote pluripotency and repress differentiation (Boyer et al. 2005, Loh et al. 2006). POU5F1 and SOX2 bind transcriptional targets in a protein-DNA ternary complex (Yuan et al. 1995), and promote their own expression (Tomioka et al. 2002, Okumura-Nakanishi et al. 2005), as well as that of Nanog (Kuroda et al. 2005) in ES cells. However, many genes are not co-occupied by all three factors (Boyer et al. 2005, Loh et al. 2006), and thus each gene must have distinct targets as well. The central importance of SOX2 and NANOG to ES and iPS cell biology is recapitulated by the requirement for these genes in early embryogenesis.
The SRY-box containing transcription factor SOX2 has been used for most genetic reprogramming efforts, excluding those in which Sox2 was expressed endogenously in the starting cell population (Kim et al. 2008, 2009). As an POU5F1 cofactor, Sox2 mutants might be expected to phenocopy Pou5f1 mutants. However, Sox2 mutants are lethal at a slightly later stage than Pou5f1 mutants, suggesting that maternal SOX2 protein may participate in early lineage decisions (Avilion et al. 2003). Importantly, Sox2 mutants do exhibit defects in the embryonic lineage, or epiblast, although at a later stage than Pou5f1 mutants. Interestingly, the placental lineage is also affected, with defects detected in the chorion. Like Pou5f1, Sox2 is required for maintaining pluripotency and/or self-renewal in ES cells. ES cells cannot be derived from Sox2 null embryos (Avilion et al. 2003), even though embryos survive beyond the point at which ES cells are derived. This observation highlights an intrinsic difference between formation of ES cells and embryogenesis, i.e. establishment of pluripotency must proceed by a unique genetic program, which is distinct from programs overseeing normal development.
Analysis of the requirement for Sox2 in existing ES cell lines has been informative. Reduction of Sox2 levels in established ES cell lines leads to formation of trophoblast-like cells (Li et al. 2007, Masui et al. 2007), supporting the notion that POU5F1 and SOX2 cooperate to repress trophoblast fates in ES cells. Notably, Pou5f1 overexpression can rescue Sox2 deficiency in ES cells (Masui et al. 2007), consistent with the proposal that activation of Pou5f1 expression is a major role of SOX2. However, these observations also suggest that POU5F1/SOX2 dimerization is not essential for activation of pluripotency network genes, although this proposal has not been tested by direct interference of POU5F1/SOX2 binding. Similar to the consequences of Pou5f1 overexpression, overexpression of Sox2 also leads to rapid differentiation of ES cells (Kopp et al. 2008).
The role of Sox2 as a general supporter of stem cell behavior has been examined in other stem cell contexts as well. SOX2 is also highly expressed in progenitor cells throughout the CNS. Sox2 is not absolutely required for neural stem cell self-renewal or multipotency (Miyagi et al. 2008). However, genetic redundancy with other SOX factors may mask its role. Interestingly, in retinal progenitor cells, where Sox2 is the sole SOX factor expressed, Sox2 is required for proliferation and differentiation (Taranova et al. 2006). Beyond regulating proliferation, Sox2 plays important roles in embryo patterning, such as foregut patterning (Que et al. 2007). Thus, Sox2 appears to be regulated in a context-dependent manner to achieve multiple developmental outcomes.
The homeobox transcription factor NANOG, which has been used for reprogramming human cells (Yu et al. 2007), plays an essential role during lineage specification in the mouse embryo. In the mouse, Nanog mutant exhibit defects in lineage specification sometime between the blastocyst and embryonic day 5.5, and only extraembryonic endoderm appears to survive in the absence of Nanog (Mitsui et al. 2003). Thus, Nanog is essential for the embryonic versus extraembryonic endoderm lineage decision.
The requirement for Nanog in pluripotency has been examined to some extent as well. Deletion of Nanog from existing ES cells does not eliminate their ability to self-renew or contribute to embryos in chimeras (Mitsui et al. 2003, Chambers et al. 2007). Nanog null ES cells exhibit subtle morphological differences and are more prone to differentiate than wild type ES cells (Mitsui et al. 2003, Chambers et al. 2007). Normal populations of ES cells, in fact, appear to express variable levels of NANOG, with those expressing less NANOG being more prone to differentiation and less prone to self-renewal (Hatano et al. 2005, Chambers et al. 2007). Thus, NANOG levels may act as a switch, allowing ES cells to rapidly choose between self-renewal and differentiation. Importantly, Pou5f1 and Sox2 continue to be expressed in ES cells forcibly lacking NANOG, as are all other ES cell markers examined (Chambers et al. 2007). This observation is consistent with the fact that Pou5f1 and Sox2 are alone sufficient to initiate the reprogramming cascade in both mouse and human cells in the absence of ectopic Nanog (Takahashi & Yamanaka 2006, Takahashi et al. 2007). Nonetheless, overexpression of Nanog in mouse ES cells can prevent differentiation in the absence of the critical self-renewal cytokine LIF (Chambers et al. 2003, Mitsui et al. 2003). Thus, Nanog is sufficient, but not necessary for ensuring self-renewal.
The role of Nanog in other stem cell types of the mouse has been examined to a limited extent. Nanog is also expressed in the germline (Hatano et al. 2005), and Nanog mutant ES cells are not detected in the germline beyond embryonic day 11.5 (Chambers et al. 2007), suggesting a requirement for Nanog in germline stem cells. Analysis of the requirement for Nanog in the mouse at later stages should provide interesting insight into its role in adult stem cells, and how these differ from pluripotent stem cells.
KLF4 belongs to a large family of 17 Kruppel-like zinc finger domain transcription factors, several of which are expressed in the preimplantation mouse embryo. Loss of Klf4 does not result in embryonic lethality, but to defects in multiple lineages post-natally, including skin (Segre et al. 1999) and goblet cells in the colon (Katz et al. 2002). Reduction of Klf4 in ES cells does not alter self-renewal or developmental potential (Nakatake et al. 2006, Jiang et al. 2008). However, two other Klf factors (Klf2 and Klf5) are co-expressed with Klf4 in ES cells (Jiang et al. 2008). While reduction of each Klf alone, or each combination of two Klfs, did not perturb ES cell self-renewal, loss of the three Klf factors simultaneously led to differentiation and decreased proliferation (Jiang et al. 2008). Together, these results argue for genetic redundancy among the Klf genes during lineage specification.
KLF4 is sufficient to trigger reprogramming when combined with POU5F1 and SOX2, and supports other KLF factors during pluripotency and self-renewal. Consistent with this, overexpression of Klf4 can reduce ES cell differentiation in some assays (Li et al. 2005). In addition, other KLF factors, such as KLF1, KLF2, and KLF5 can substitute for KLF4 during reprogramming, albeit with lower efficiency (Nakagawa et al. 2008). Genome-wide location analysis revealed that the KLF factors bind putative regulatory elements of Pou5f1, Sox2, Nanog, and many other pluripotency-associated genes (Nakagawa et al. 2008). Thus, KLF may participate in regulating a pluripotency network.
Examination of the requirement for other KLF factors has led to some insight into the connection between KLF factors in ES cells and their establishment in the early embryo. While loss of Klf2 leads to embryonic lethality between embryonic days 12–14 owing to severe hemorrhage (Kuo et al. 1997), loss of Klf5 has a much earlier phenotype. Loss of Klf5 leads to lethality due to trophoblast defects around the blastocyst stage (Ema et al. 2008). ES cells cannot be derived from Klf5 null embryos, even when trophoblast tissues are rescued by tetraploid complementation (Ema et al. 2008). When deleted from existing ES cells, ES cells continue to self-renew, although they are more prone to spontaneously differentiate and proliferate more slowly (Ema et al. 2008). This phenotype was not observed in RNAi knockdown experiments. However, it is not currently clear whether this discrepancy is due to differences between ES cell establishment and maintenance, or simply to the assay in use. Overexpression of Klf4 can rescue Klf5-deficient ES cells Klf5 (Ema et al. 2008), again suggesting some redundancy between these Klf factors.
Since Klf4 and Klf5 play overlapping roles in pluripotency, study of Klf5 may provide insight into the mechanism by which Klf4 promotes pluripotency. One mechanism proposed for KLF5-mediated maintenance of ES cell self-renewal involves the TCL1–AKT pathway. Both TCL1 levels and AKT phosphorylation are reduced in ES cells in the absence of Klf5, and expression of a constitutively active form of AKT partially restores the rate of proliferation to wild-type levels. These observations suggest that KLF factors may regulate self-renewal through regulation of AKT signaling. Alternatively, KLF4 may upregulate Nanog during reprogramming by repressing p53 (Rowland et al. 2005). Elucidating the precise molecular mechanisms underlying the role of KLF factors in maintaining and driving pluripotency will be an exciting avenue for future research.
The orphan nuclear factor estrogen-related receptor beta (ESRRB) was recently shown to be capable of reprogramming mouse fibroblasts in conjunction with POU5F1 and SOX2 (Feng et al. 2009a). ESRRB can thus substitute for KLF4 in reprogramming. Consistent with this proposal, the authors found that ESRRB can also rescue self-renewal in ES cells in which all three Klf paralogs had been knocked down. In addition, genome-wide location analysis revealed a tendency for ESRRB-binding sites to colocalize with KLF-binding sites (Feng et al. 2009a). The authors suggest that ESRRB may promote reprogramming via KLF factors. However, the efficiency of reprogramming by ESRRB was lower than by KLF4 (Feng et al. 2009a). In addition, differences exist between the roles of these two genes in the embryo.
Both the timing and tissues affected differ between Esrrb and Klf gene family knockouts. Esrrb knockout embryos die around E10.5 with placental defects (Luo et al. 1997). Moreover, deletion of Esrrb strictly within the embryonic lineage rescues this phenotype (Luo et al. 1997), arguing for a specific requirement for Esrrb in the extraembryonic lineage. Genetic redundancy with the related genes Esrra or Esrrg could preclude observation of a requirement for Esrrb in the embryonic lineage. Interestingly, Esrrg, but not Esrra could also promote reprogramming in the presence of POU5F1 and SOX2 (Feng et al. 2009a). Analysis of Esrrb and Esrrg double knockout embryos would therefore be particularly interesting.
The basic helix–loop–helix/leucine zipper transcription factor MYC is neither essential for reprogramming nor for early lineage specifications. Rather, Myc null embryos die between embryonic days 9 and 10, with a phenotype suggesting requirements in growth and proliferation of many organ systems (Davis et al. 1993). However, much of this phenotype is likely to result indirectly from abnormal extraembryonic structures in these mutants, since embryo-specific loss of Myc enables normal development of all organs, with the exception of the hematopoietic system, up to day 12 of embryogenesis (Dubois et al. 2008). MYC has been independently shown to be important for regulating hematopoietic stem cell self-renewal and differentiation (Wilson et al. 2004). The related factor MYCN (nMyc) has been shown to be important in neural stem cells (Knoepfler et al. 2002), suggesting a more general role for MYC factors in proliferative populations. Whether simultaneous loss of MYC, MYCN, and/or related MYCL1 (lMyc) would result in earlier developmental or widespread stem cell defects remains an open question.
In addition to promoting proliferation and survival of somatic stem cells, MYC may play an important role in ES cell proliferation. Myc levels are decreased as ES cells undergo differentiation during LIF withdrawal, and overexpression of a dominant-negative MYC promotes differentiation (Cartwright et al. 2005). Interestingly, overexpression of stable, non-phosphorylable MYC variant can delay differentiation (Cartwright et al. 2005), suggesting a potential link between proliferation and potency.
Although it is still unclear what role MYC plays during reprogramming, several models have been proposed (Knoepfler 2008). For example, MYC may simply promote proliferation, which could indirectly alter the state of the cell or activity of the reprogramming factors, i.e. MYC-stimulated proliferation could lead to chromatin remodeling, facilitating access of reprogramming factors to pluripotency targets. This hypothesis is consistent with the finding that the related protein MYCN influences chromatin structure in neural stem cells (Knoepfler et al. 2006). Alternatively, MYC may have its own set of pluripotency targets. MYC and the LIF signaling target STAT3 co-occupy putative regulatory elements of many pluripotency-associated genes in ES cells (Kidder et al. 2008). Thus, MYC may help transduce pluripotency-promoting effects of signaling pathways.
While MYC can be omitted completely during reprogramming, its inclusion greatly increases efficiency (Nakagawa et al. 2008). Importantly, fibroblasts express Myc at around 20% of levels expressed in ES cells, and this is increased during reprogramming (Nakagawa et al. 2008), suggesting that endogenous Myc still participates in reprogramming. Nonetheless, it may be safer to avoid use of exogenous Myc, as tumors have been observed to arise in mice generated from iPS cells bearing MYC, but not from those that did not (Takahashi & Yamanaka 2006, Takahashi et al. 2007), although this appears to be cell type-specific (Nakagawa et al. 2008). As an alternative, other MYC paralogs can substitute in the reprogramming mix. For example, MYCL1 (Nakagawa et al. 2008) and MYCN have been used, resulting in reduced tumorigenicity (Blelloch et al. 2007, Nakagawa et al. 2008).
The microRNA pathway
Two studies have identified roles for the microRNA (miRNA) pathway in genetic reprogramming of fibroblasts to pluripotent stem cells. One of these studies identified the miRNA processing protein LIN28 as sufficient to reprogram human fibroblasts, in conjunction with POU5F1, SOX2, and NANOG (Yu et al. 2007). The other study identified a subset of miRNAs that improve reprogramming efficiency in the mouse, when combined with POU5F1, SOX2, and KLF4 (Judson et al. 2009). Interestingly, both pathways involve MYC, although at different levels, either upstream or downstream of MYC (Fig. 3).
LIN28, originally identified in Caenorhabditis elegans, is an RNA-binding protein that has been shown to participate in reprogramming of human cells in combination with POU5F1, SOX2, and NANOG (Yu et al. 2007). Lin28 exhibits dynamic expression patterns throughout mouse embryogenesis (Yang & Moss 2003), suggesting important and developmentally-regulated roles. Lin28 is required for differentiation of skeletal muscle from cultured myoblasts (Polesskaya et al. 2007). Lin28 knockdown in ES cells limits germ cell formation both in vivo and in vitro (West et al. 2009). Analysis of a null allele should provide exciting insight into the role of miRNAs in early lineage specifications.
Lin28 is expressed at high levels in both mouse and human ES cells, and is downregulated during their differentiation (Yang & Moss 2003, Richards et al. 2004). Lin28 has recently been found to be necessary and sufficient to inhibit processing of the miRNA let7 in mouse ES cells (Viswanathan et al. 2008). However, levels of pluripotency markers were unchanged (Viswanathan et al. 2008). Nonetheless, Lin28 is able to participate in cellular reprogramming in human cells (Yu et al. 2007) and promote proliferation of tumor cells (Guo et al. 2006). The explanation for this behavior may lie with the fact that LIN28 appears to be capable of regulating Myc through let7, i.e. the miRNA let7 (Mirlet7), which is regulated by LIN28, can target Myc for degradation (Koscianska et al. 2007, Kumar et al. 2007). Consistent with this observation, Lin28 can regulate Myc expression in cultured myoblasts (Polesskaya et al. 2007). These observations are consistent with Lin28 substituting for Myc during reprogramming in human cells. Examination of the requirement for Lin28 during mouse development should help reveal to what extent Lin28 activity is conserved between human and mouse, and whether there are additional Lin28 targets besides Myc.
More recently, a subset of miRNAs was found to increase efficiency of mouse fibroblast reprogramming by POU5F1, SOX2, and KLF4 (Judson et al. 2009). Importantly, these miRNAs, which include members of the miR-290 family, led to an increased proportion of correctly reprogrammed iPS colonies, than when MYC was used. More detailed analysis revealed that the expression of these miRNAs is normally activated during conventional 3 or 4 factor-based reprogramming, but not until relatively late in the process, suggesting that these miRNAs act fairly downstream in the reprogramming process.
Members of the miR-290 family are normally expressed in ES cells and downregulated during their differentiation (Wang et al. 2008). These miRNAs play an important role in cell cycle regulation in self-renewing ES cells (Wang et al. 2008). However, during reprogramming they may act downstream of MYC and subsequent to chromatin remodeling (Judson et al. 2009). The requirement for the miR-290 cluster during embryonic development has not been reported. Partial insight into this issue may be provided by analysis of loss of Dgcr8, an RNA-binding protein specific to the miRNA pathway (Wang et al. 2007). Knockout of Dgcr8 leads to abnormal development by E6.5, and lethality sometime prior to E10 (Wang et al. 2007), consistent with a possible role in early lineage specification.
Programming and reprogramming in reproduction and human health
We are only just beginning to understand how reprogramming factors induce pluripotency in differentiated cells. Systems-level analyses in iPS and ES cells have identified large lists of genes probably acting downstream of the reprogramming genes to effect cellular changes at different levels (Fig. 3). Examining the requirement of endogenous copies of these genes and targets during reprogramming will be essential for testing this and other models of induced pluripotency. For example, two studies have shown that a single factor can revert other stem cell types to ES-like cells (Guo et al. 2009, Kim et al. 2009), and these efforts were presumably successful owing to the endogenous expression of other reprogramming factors and/or targets. Reprogramming in cells lacking components of the pluripotency network should help formally test this assumption and organize lists of genes into discrete pathways promoting pluripotency.
While the list of genes sufficient to induce pluripotency in mature cell types continues to grow, so do the number of small molecules capable of substituting for one or more of them. Together, these studies will facilitate alternative and possibly safer mechanisms for cellular reprogramming, with the ultimate goal of generating patient-derived stem cells for therapeutic treatment of a variety of human health issues. Meanwhile, examination of the role of these genes during embryonic development in the mouse will help us close the gap between induced pluripotency and fetal health.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
This work was supported by grant no. FRN13426 from the Canadian Institutes of Health Research.
We are indebted to members of the Rossant Laboratory for discussion.
AvilionAANicolisSKPevnyLHPerezLVivianNLovell-BadgeR2003Multipotent cell lineages in early mouse development depend on SOX2 function. Genes and Development17126–140.
BernsteinBEMikkelsenTSXieXKamalMHuebertDJCuffJFryBMeissnerAWernigMPlathK2006A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell125315–326.
BlellochRVenereMYenJRamalho-SantosM2007Generation of induced pluripotent stem cells in the absence of drug selection. Cell Stem Cell1245–247.
BoyerLALeeTIColeMFJohnstoneSELevineSSZuckerJPGuentherMGKumarRMMurrayHLJennerRG2005Core transcriptional regulatory circuitry in human embryonic stem cells. Cell122947–956.
BronsIGSmithersLETrotterMWRugg-GunnPSunBChuva de Sousa LopesSMHowlettSKClarksonAAhrlund-RichterLPedersenRA2007Derivation of pluripotent epiblast stem cells from mammalian embryos. Nature448191–195.
CartwrightPMcLeanCSheppardARivettDJonesKDaltonS2005LIF/STAT3 controls ES cell self-renewal and pluripotency by a Myc-dependent mechanism. Development132885–896.
ChambersIColbyDRobertsonMNicholsJLeeSTweedieSSmithA2003Functional expression cloning of Nanog, a pluripotency sustaining factor in embryonic stem cells. Cell113643–655.
ChambersISilvaJColbyDNicholsJNijmeijerBRobertsonMVranaJJonesKGrotewoldLSmithA2007Nanog safeguards pluripotency and mediates germline development. Nature4501230–1234.
ChazaudCYamanakaYPawsonTRossantJ2006Early lineage segregation between epiblast and primitive endoderm in mouse blastocysts through the Grb2–MAPK pathway. Developmental Cell10615–624.
ChenXXuHYuanPFangFHussMVegaVBWongEOrlovYLZhangWJiangJ2008Integration of external signaling pathways with the core transcriptional network in embryonic stem cells. Cell1331106–1117.
DavisACWimsMSpottsGDHannSRBradleyA1993A null c-myc mutation causes lethality before 10.5 days of gestation in homozygotes and reduced fertility in heterozygous female mice. Genes and Development7671–682.
DuboisNCAdolpheCEhningerAWangRARobertsonEJTrumppA2008Placental rescue reveals a sole requirement for c-Myc in embryonic erythroblast survival and hematopoietic stem cell function. Development1352455–2465.
EmaMMoriDNiwaHHasegawaYYamanakaYHitoshiSMimuraJKawabeYHosoyaTMoritaM2008Kruppel-like factor 5 is essential for blastocyst development and the normal self-renewal of mouse ESCs. Cell Stem Cell3555–567.
FengBJiangJKrausPNgJHHengJCChanYSYawLPZhangWLohYHHanJ2009aReprogramming of fibroblasts into induced pluripotent stem cells with orphan nuclear receptor Esrrb. Nature Cell Biology11197–203.
FengBNgJHHengJCNgHH2009bMolecules that promote or enhance reprogramming of somatic cells to induced pluripotent stem cells. Cell Stem Cell4301–312.
GuanKNayerniaKMaierLSWagnerSDresselRLeeJHNolteJWolfFLiMEngelW2006Pluripotency of spermatogonial stem cells from adult mouse testis. Nature4401199–1203.
GuoYChenYItoHWatanabeAGeXKodamaTAburataniH2006Identification and characterization of lin-28 homolog B (LIN28B) in human hepatocellular carcinoma. Gene38451–61.
GuoGYangJNicholsJHallJSEyresIMansfieldWSmithA2009Klf4 reverts developmentally programmed restriction of ground state pluripotency. Development1361063–1069.
HatanoSYTadaMKimuraHYamaguchiSKonoTNakanoTSuemoriHNakatsujiNTadaT2005Pluripotential competence of cells associated with Nanog activity. Mechanisms of Development12267–79.
HochedlingerKYamadaYBeardCJaenischR2005Ectopic expression of Oct-4 blocks progenitor-cell differentiation and causes dysplasia in epithelial tissues. Cell121465–477.
HuangfuDOsafuneKMaehrRGuoWEijkelenboomAChenSMuhlesteinWMeltonDA2008Induction of pluripotent stem cells from primary human fibroblasts with only Oct4 and Sox2. Nature Biotechnology261269–1275.
IvanovaNDobrinRLuRKotenkoILevorseJDeCosteCSchaferXLunYLemischkaIR2006Dissecting self-renewal in stem cells with RNA interference. Nature442533–538.
JiangJChanYSLohYHCaiJTongGQLimCARobsonPZhongSNgHH2008A core Klf circuitry regulates self-renewal of embryonic stem cells. Nature Cell Biology10353–360.
KajiKNorrbyKPacaAMileikovskyMMohseniPWoltjenK2009Virus-free induction of pluripotency and subsequent excision of reprogramming factors. Nature458771–775.
Kanatsu-ShinoharaMInoueKLeeJYoshimotoMOgonukiNMikiHBabaSKatoTKazukiYToyokuniS2004Generation of pluripotent stem cells from neonatal mouse testis. Cell1191001–1012.
KatzJPPerreaultNGoldsteinBGLeeCSLaboskyPAYangVWKaestnerKH2002The zinc-finger transcription factor Klf4 is required for terminal differentiation of goblet cells in the colon. Development1292619–2628.
KehlerJTolkunovaEKoschorzBPesceMGentileLBoianiMLomeliHNagyAMcLaughlinKJScholerHR2004Oct4 is required for primordial germ cell survival. EMBO Reports51078–1083.
KimJBZaehresHWuGGentileLKoKSebastianoVArauzo-BravoMJRuauDHanDWZenkeM2008Pluripotent stem cells induced from adult neural stem cells by reprogramming with two factors. Nature454646–650.
KimJBSebastianoVWuGArauzo-BravoMJSassePGentileLKoKRuauDEhrichMvan den BoomD2009Oct4-induced pluripotency in adult neural stem cells. Cell136411–419.
KnoepflerPSChengPFEisenmanRN2002N-myc is essential during neurogenesis for the rapid expansion of progenitor cell populations and the inhibition of neuronal differentiation. Genes and Development162699–2712.
KoppJLOrmsbeeBDDeslerMRizzinoA2008Small increases in the level of Sox2 trigger the differentiation of mouse embryonic stem cells. Stem Cells26903–911.
KoscianskaEBaevVSkrekaKOikonomakiKRusinovVTablerMKalantidisK2007Prediction and preliminary validation of oncogene regulation by miRNAs. BMC Molecular Biology879.
KumarMSLuJMercerKLGolubTRJacksT2007Impaired microRNA processing enhances cellular transformation and tumorigenesis. Nature Genetics39673–677.
KuoCTVeselitsMLBartonKPLuMMClendeninCLeidenJM1997The LKLF transcription factor is required for normal tunica media formation and blood vessel stabilization during murine embryogenesis. Genes and Development112996–3006.
KurodaTTadaMKubotaHKimuraHHatanoSYSuemoriHNakatsujiNTadaT2005Octamer and Sox elements are required for transcriptional cis regulation of Nanog gene expression. Molecular and Cellular Biology252475–2485.
LengnerCJCamargoFDHochedlingerKWelsteadGGZaidiSGokhaleSScholerHRTomilinAJaenischR2007Oct4 expression is not required for mouse somatic stem cell self-renewal. Cell Stem Cell1403–415.
LiYMcClintickJZhongLEdenbergHJYoderMCChanRJ2005Murine embryonic stem cell differentiation is promoted by SOCS-3 and inhibited by the zinc finger transcription factor Klf4. Blood105635–637.
LiJPanGCuiKLiuYXuSPeiD2007A dominant-negative form of mouse SOX2 induces trophectoderm differentiation and progressive polyploidy in mouse embryonic stem cells. Journal of Biological Chemistry28219481–19492.
LiWWeiWZhuSZhuJShiYLinTHaoEHayekADengHDingS2009Generation of rat and human induced pluripotent stem cells by combining genetic reprogramming and chemical inhibitors. Cell Stem Cell416–19.
LiuHZhuFYongJZhangPHouPLiHJiangWCaiJLiuMCuiK2008Generation of induced pluripotent stem cells from adult rhesus monkey fibroblasts. Cell Stem Cell3587–590.
LohYHWuQChewJLVegaVBZhangWChenXBourqueGGeorgeJLeongBLiuJ2006The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nature Genetics38431–440.
LuoJSladekRBaderJAMatthyssenARossantJGiguereV1997Placental abnormalities in mouse embryos lacking the orphan nuclear receptor ERR-beta. Nature388778–782.
MartinGR1981Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. PNAS787634–7638.
MasuiSNakatakeYToyookaYShimosatoDYagiRTakahashiKOkochiHOkudaAMatobaRSharovAA2007Pluripotency governed by Sox2 via regulation of Oct3/4 expression in mouse embryonic stem cells. Nature Cell Biology9625–635.
MatsuiYZseboKHoganBL1992Derivation of pluripotential embryonic stem cells from murine primordial germ cells in culture. Cell70841–847.
MitsuiKTokuzawaYItohHSegawaKMurakamiMTakahashiKMaruyamaMMaedaMYamanakaS2003The homeoprotein Nanog is required for maintenance of pluripotency in mouse epiblast and ES cells. Cell113631–642.
MiyagiSMasuiSNiwaHSaitoTShimazakiTOkanoHNishimotoMMuramatsuMIwamaAOkudaA2008Consequence of the loss of Sox2 in the developing brain of the mouse. FEBS Letters5822811–2815.
NakagawaMKoyanagiMTanabeKTakahashiKIchisakaTAoiTOkitaKMochidukiYTakizawaNYamanakaS2008Generation of induced pluripotent stem cells without Myc from mouse and human fibroblasts. Nature Biotechnology26101–106.
NakatakeYFukuiNIwamatsuYMasuiSTakahashiKYagiRYagiKMiyazakiJMatobaRKoMS2006Klf4 cooperates with Oct3/4 and Sox2 to activate the Lefty1 core promoter in embryonic stem cells. Molecular and Cellular Biology267772–7782.
NicholsJZevnikBAnastassiadisKNiwaHKlewe-NebeniusDChambersIScholerHSmithA1998Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell95379–391.
NiwaHMiyazakiJSmithAG2000Quantitative expression of Oct-3/4 defines differentiation, dedifferentiation or self-renewal of ES cells. Nature Genetics24372–376.
NiwaHOgawaKShimosatoDAdachiK2009A parallel circuit of LIF signalling pathways maintains pluripotency of mouse ES cells. Nature460118–122.
OkitaKNakagawaMHyenjongHIchisakaTYamanakaS2008Generation of mouse induced pluripotent stem cells without viral vectors. Science322949–953.
Okumura-NakanishiSSaitoMNiwaHIshikawaF2005Oct-3/4 and Sox2 regulate Oct-3/4 gene in embryonic stem cells. Journal of Biological Chemistry2805307–5317.
ParkIHZhaoRWestJAYabuuchiAHuoHInceTALerouPHLenschMWDaleyGQ2008Reprogramming of human somatic cells to pluripotency with defined factors. Nature451141–146.
PolesskayaACuvellierSNaguibnevaIDuquetAMossEGHarel-BellanA2007Lin-28 binds IGF-2 mRNA and participates in skeletal myogenesis by increasing translation efficiency. Genes and Development211125–1138.
QueJOkuboTGoldenringJRNamKTKurotaniRMorriseyEETaranovaOPevnyLHHoganBL2007Multiple dose-dependent roles for Sox2 in the patterning and differentiation of anterior foregut endoderm. Development1342521–2531.
RowlandBDBernardsRPeeperDS2005The KLF4 tumour suppressor is a transcriptional repressor of p53 that acts as a context-dependent oncogene. Nature Cell Biology71074–1082.
ScholerHRDresslerGRBallingRRohdewohldHGrussP1990Oct-4: a germline-specific transcription factor mapping to the mouse t-complex. EMBO Journal92185–2195.
SegreJABauerCFuchsE1999Klf4 is a transcription factor required for establishing the barrier function of the skin. Nature Genetics22356–360.
SekkaiDGruelGHerryMMoucadelVConstantinescuSNAlbagliOTronik-Le RouxDVainchenkerWBennaceur-GriscelliA2005Microarray analysis of LIF/Stat3 transcriptional targets in embryonic stem cells. Stem Cells231634–1642.
ShiYDoJTDespontsCHahmHSScholerHRDingS2008A combined chemical and genetic approach for the generation of induced pluripotent stem cells. Cell Stem Cell2525–528.
SilvaJBarrandonONicholsJKawaguchiJTheunissenTWSmithA2008Promotion of reprogramming to ground state pluripotency by signal inhibition. PLoS Biology6e253.
SimmonsDGCrossJC2005Determinants of trophoblast lineage and cell subtype specification in the mouse placenta. Developmental Biology28412–24.
SoldnerFHockemeyerDBeardCGaoQBellGWCookEGHargusGBlakACooperOMitalipovaM2009Parkinson's disease patient-derived induced pluripotent stem cells free of viral reprogramming factors. Cell136964–977.
StadtfeldMBrennandKHochedlingerK2008aReprogramming of pancreatic beta cells into induced pluripotent stem cells. Current Biology18890–894.
StadtfeldMNagayaMUtikalJWeirGHochedlingerK2008bInduced pluripotent stem cells generated without viral integration. Science322945–949.
TadaMTakahamaYAbeKNakatsujiNTadaT2001Nuclear reprogramming of somatic cells by in vitro hybridization with ES cells. Current Biology111553–1558.
TakahashiKYamanakaS2006Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell126663–676.
TakahashiKTanabeKOhnukiMNaritaMIchisakaTTomodaKYamanakaS2007Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell131861–872.
TaranovaOVMagnessSTFaganBMWuYSurzenkoNHuttonSRPevnyLH2006SOX2 is a dose-dependent regulator of retinal neural progenitor competence. Genes and Development201187–1202.
TesarPJChenowethJGBrookFADaviesTJEvansEPMackDLGardnerRLMcKayRD2007New cell lines from mouse epiblast share defining features with human embryonic stem cells. Nature448196–199.
ThomsonJAItskovitz-EldorJShapiroSSWaknitzMASwiergielJJMarshallVSJonesJM1998Embryonic stem cell lines derived from human blastocysts. Science2821145–1147.
TomiokaMNishimotoMMiyagiSKatayanagiTFukuiNNiwaHMuramatsuMOkudaA2002Identification of Sox-2 regulatory region which is under the control of Oct-3/4-Sox-2 complex. Nucleic Acids Research303202–3213.
WalkerEOhishiMDaveyREZhangWCassarPATanakaTSDerSDMorrisQHughesTRZandstraPW2007Prediction and testing of novel transcriptional networks regulating embryonic stem cell self-renewal and commitment. Cell Stem Cell171–86.
WangJRaoSChuJShenXLevasseurDNTheunissenTWOrkinSH2006A protein interaction network for pluripotency of embryonic stem cells. Nature444364–368.
WangYMedvidRMeltonCJaenischRBlellochR2007DGCR8 is essential for microRNA biogenesis and silencing of embryonic stem cell self-renewal. Nature Genetics39380–385.
WangYBaskervilleSShenoyABabiarzJEBaehnerLBlellochR2008Embryonic stem cell-specific microRNAs regulate the G1-S transition and promote rapid proliferation. Nature Genetics401478–1483.
WernigMMeissnerAForemanRBrambrinkTKuMHochedlingerKBernsteinBEJaenischR2007In vitro reprogramming of fibroblasts into a pluripotent ES-cell-like state. Nature448318–324.
WestJAViswanthanSRYabuuchiACunniffKTakeuchiAParkIHSeroJEZhuHPerez-AtaydeAFrazierAL2009A role for Lin28 in primordial germ-cell development and germ-cell malignancy. Nature460909–913.
WilsonAMurphyMJOskarssonTKaloulisKBettessMDOserGMPascheACKnabenhansCMacdonaldHRTrumppA2004c-Myc controls the balance between hematopoietic stem cell self-renewal and differentiation. Genes and Development182747–2763.
WoltjenKMichaelIPMohseniPDesaiRMileikovskyMHamalainenRCowlingRWangWLiuPGertsensteinM2009piggyBac transposition reprograms fibroblasts to induced pluripotent stem cells. Nature458766–770.
YingQLNicholsJChambersISmithA2003BMP induction of Id proteins suppresses differentiation and sustains embryonic stem cell self-renewal in collaboration with STAT3. Cell115281–292.
YuJVodyanikMASmuga-OttoKAntosiewicz-BourgetJFraneJLTianSNieJJonsdottirGARuottiVStewartR2007Induced pluripotent stem cell lines derived from human somatic cells. Science3181917–1920.
YuanHCorbiNBasilicoCDaileyL1995Developmental-specific activity of the FGF-4 enhancer requires the synergistic action of Sox2 and Oct-3. Genes and Development92635–2645.
ZhouHWuSJooJYZhuSHanDWLinTTraugerSBienGYaoSZhuY2009Generation of induced pluripotent stem cells using recombinant proteins. Cell Stem Cell4381–384.