Progress in understanding reprogramming to the induced pluripotent state

Abstract

Induction of pluripotency by transcription factors has become a commonplace method to produce pluripotent stem cells. Great strides have been made in our understanding of the mechanism by which this occurs--particularly in terms of transcriptional and chromatin-based events--yet only a small part of the complete picture has been revealed. Understanding the mechanism of reprogramming to pluripotency will have important implications for improving the efficiency and quality of reprogramming and advancing therapeutic application of induced pluripotent stem cells. It will also help to reveal the machinery that stabilizes cell identity and to instruct the design of directed differentiation or lineage switching strategies. To inform the next phase in understanding reprogramming, we review the latest findings, highlight ongoing debates and outline future challenges.

Figure 1. The generation of iPSCs is a multistep process

Known events occurring in early, middle and late phases of mouse embryonic fibroblasts reprogramming are depicted. Events in the early and middle phases may be less clearly separated from each other compared to those that take place late. Even though the initial response to reprogramming factor expression, i.e. downregulation of somatic expression and changes in H3K4me2, may occur population-wide, none of the early and middle steps alone are sufficient for the induction of pluripotency and only a subset of cells makes it from one step to the next, accounting for the low overall efficiency of the process. While events that eventually lead to the reprogrammed state are initiated early, successful reprogramming requires the expression of the four transcription factors – Oct4 (O), Klf4 (K), Sox2 (S) and cMyc (M) - until the iPSC state is established, otherwise cells revert back to a differentiated state. The partially reprogrammed state of pre-iPSCs is often identified in reprogramming cultures as ESC-like colonies that don't express Nanog and other pluripotency-related genes that are induced only during the last step of reprogramming. The dashed line indicates that pre-iPSCs can be converted to iPSCs with treatments that enhance the late phase of reprogramming. The close cooperation between reprogramming factors and the extracellular milieu during reprogramming is indicated by the fact that certain signaling pathways affect the reprogramming process at the indicated places. For references please see ^,–,,,,

Figure 2. Roles of the reprogramming factors and their interaction with chromatin during the final step of reprogramming

Scheme illustrating the different function of the reprogramming factors in the late phase of reprogramming. Oct4, Sox2, and Klf4 are implicated in mediating the upregulation of the pluripotency network and only bind many of these genes during the final step of reprogramming to allow their transcriptional activation^,. In contrast, many cMyc targets are bound and activated at an intermediate step prior to activation of the pluripotency network^,. It appears that the inability of Oct4, Sox2, and Klf4 to bind and activate pluripotency-related genes is, at least to some cases, associated with repressive chromatin signatures found in promoters of these genes in the partially reprogrammed state, which is reset to active chromatin in iPSCs^,. The loss of these repressive marks appears to be required for efficient reprogramming (see Box 2). Overexpression of the pluripotency transcription factor Nanog synergistically acts with DNA demethylating agents to enhance the final transition to the iPSC state^,. Nanog co-binds many of the regulatory regions with Oct4, Sox2, and Klf4 and may promote binding of Oct4, Sox2, and Klf4, and recruitment of co-activators such as the histone acetyltransferase p300^–,. In addition, ESC-specific chromatin remodeling complexes have been implicated the activation pluripotency-related genes. For instance, overexpression of components of the BAF chromatin-remodeling complex enhances reprogramming and may facilitate binding of Oct4 to pluripotency target genes. Similarly, Chd1, another chromatin remodeling enzyme thought to be important for the maintenance of an open chromatin state in mouse ESCs, is required for reprogramming and may act during the late phase of the process.

Figure 3. X chromosome inactivation (XCI) and reprogramming

a) Reactivation of the Xi is observed during female mouse pre-implantation development. Here, XCI occurs first in an imprinted fashion that exclusively inactivates the paternally-inherited X (Xp) in all cells of the pre-implantation embryo. Reactivation of the Xp in epiblast cells of the blastocyst allows a second round of XCI, where upon differentiation each cell has a random chance of inactivating the paternally or maternally-inherited X chromosome (Xm). Mouse ESCs, derived from epiblast cells of the blastocyst, therefore carry two active X's (XaXa). Reprogramming of female mouse cells with Oct4 (O), Klf4 (K), Sox2 (S) and cMyc (M) leads to reactivation of the Xi such that mouse iPSCs, like female mouse ESCs, carry two Xa's, but the underlying mechanism remains unknown. b) In mouse ESCs, XCI is closely coupled to pluripotency by linking pluripotency transcription factors to the regulation of Xist, which encodes a non-coding RNA that is the key mediator of XCI. To maintain XaXa status in ESCs, different transcription factors are proposed to directly repress Xist by binding to its intronic region and activate Tsix via binding to its regulatory regions, which in turn also blocks Xist accumulation^–. These findings may explain why Xi reactivation during reprogramming occurs simultaneously with establishment of the pluripotency network, but the exact timing of Xi-reactivation relative to the pluripotency network remains to be established. c) Reprogramming of female human fibroblasts to the typical human iPSC state with basic fibroblast growth factor (bFGF) culture conditions does not lead to Xi reactivation. In each female human iPSC line, the XCI status of the single fibroblast cell it originated from is therefore propagated, resulting in non-random XCI. It is thought that human iPSCs represent the mouse EpiSC-state. Reprogramming under LIF cell culture conditions with 2i, a small molecule cocktail that inhibits mitogen-activated protein kinases and glycogen synthase kinase 3, leads to the establishment of metastable mouse ESC-like human iPSCs with XaXa.