You are here

| Pluripotent Stem Cells

The Expanding Talents of Oct4 and Sox2: Pluripotency Factors in Embryonic Stem Cells Regulate Differentiation into Germ Layers

From Cell
By Stuart P. Atkinson

While much is known about the relative contribution to pluripotency by key transcription factors such as Oct4, Nanog and Sox2, little is known about the role they may have, if any, in lineage-specific differentiation. Detailed transcriptional maps have been generated which show the integration of these factors into pluripotency networks (Chen et al, Lu et al, Marson et al, Wang et al) but how this map alters and how transcription factor function alters when a cell leaves the pluripotent state and becomes lineage restricted is unknown. This question has now begun to be answered in a Cell paper from the laboratory of Sharad Ramanathan at Harvard University, Cambridge, USA using directed differentiation of mouse embryonic stem cells (mESCs) as a model system to study the relative contributions of Oct4 and Sox2 to germ layer development and show that, in addition to their roles in pluripotency, they also act to regulate lineage choice (Thomson et al).

Initial analysis studied the expression of pluripotency associated factors (PAFs) in mESCs and progenitor cells of the mesendoderm (ME) or neural ectoderm (NE) fate using previously generated microarray data (Shen et al). Oct4, Nanog, Klf5, Tbx3 and Klf9 were shown to be expressed in both mESCs and ME cells but were downregulated in NE cells (ME class genes), and Sox2, Foxp1 Rbpj, Dnmt3a and Zfp532 were expressed in mESCs and NE cells, while being downregulated in the ME cells (NE class genes), overall suggesting that PAFs may have a role in lineage choice. This hypothesis was then tested by detailed studies of mESC differentiation using a defined system in which mESCs were maintained in a defined serum-free medium (containing neural supplements N2 and B27) under feeder-free conditions with the withdrawal of LIF and BMP for 48 hr to allow the cells to respond to differentiation signals (Jackson et al). The addition of Wnt3a or CHIR (a Wnt agonist) promoted an ME cell fate with differentiated cells expressing Brachyury and Foxa2, while retinoic acid (RA) treatment promoted the NE cell fate triggering the expression of Sox1, Brn2 (Pou3f2) and Nestin. Although both differentiation techniques led to the differentiation of approximately 70% of cells, it was also noted that under ME-directed differentiation in the presence of Wnt3a or CHIR, some cells expressed Sox1 (NE fate) and not Brachyury, demonstrating that this method of differentiation provides an excellent system for the study of the role of PAFs during both ME and NE lineage choice.

Microarray analysis of the differentiating cells demonstrated that after the signal-withdrawal pre-treatment 87% of genes changed by less than two-fold, 9% (including PAFs) were downregulated more than two-fold, while 4% were induced by more than two-fold. Nanog, Oct4, Sox2, Klf4, Klf5 and Tbx3 mRNA levels all decreased as compared to ESC (5%, 74%, 30%, 14%, 19% and 12% of pre-treatment only respectively) with an associated drop in protein levels. Nanog was studied in more detail which showed that Nanog protein levels fell exponentially with time, and only cells that had lost Nanog responded to CHIR treatment by up-regulating Brachyury, while loss of Nanog led to a corresponding loss of Oct4 and Sox2, overall suggesting that Nanog is an early and necessary event in the loss of the ESC state and allows for differentiation.

The differences in PAF dynamics between pluripotent cells and lineage-specified cells observed from the microarray data suggests that this result should be reflected in the protein levels of these factors in individual cells during differentiation. Therefore, cells were studied after directed ME and NE differentiation. The authors discovered that Tbx3, Klf4, Klf9 and Rbpj protein were absent from differentiated cells, suggesting that these ME/NE factors identified from the microarray analysis were not reactivated during differentiation suggesting that they do not play a role in lineage choice. However, Oct4, Sox2, Nanog, Jarid2, Klf5, Foxp1 and Dnmt3a protein were all present in differentiating cells, suggesting that these pluripotency-associated factors play a role in lineage choice during differentiation. Jarid2 and Dnmt3a may be required for the modulation of the epigenetic environment during development thereby allowing for transcription factor binding and the activation/repression of certain genes. Further, more detailed analysis of the differentiated cell types demonstrated that Oct4 and Klf5 were present in Brachyury positive cells (ME cells), while in Sox1 positive cells (NE cells) Oct4 was missing and Klf5 was sequestered to a non-nuclear sub-compartment. Sox2 and Brachyury were mutually exclusive while Sox2 and Sox1 co-expression correlated strongly to NE lineage induction, with Foxp1, Nanog, Dnmt3a and Jarid2 present in both ME and NE cells. This suggests that Oct4 and Sox2 may have important roles in the differentiation of mESC towards the ME and NE fate respectively.

Whether Oct4 and Sox2 patterns are established prior to lineage choice was next studied using FACS and live cell imaging. FACS analysis demonstrated that after the addition of CHIR, Sox2 protein levels dropped drastically with only a slight decrease in Oct4, but after 12 hours the onset of Brachyury expression occurred alongside an increase in Oct4 expression. Conversely, after RA addition, Sox2 protein levels decreased slightly while Oct4 levels decreased significantly, with subsequent Sox1 activation after 17 to 19 hours alongside further Sox2 activation and Oct4 repression. This data was corroborated using live single cell imaging using a Oct4-mCitrine fusion reporter cell line which showed that Oct4 and Sox2 levels change with relation to time and the differentiation status of the cell and overall suggests that while Oct4 may inhibit the NE lineage and Sox2 may inhibit the ME lineage.

This was further investigated by analysing Sox2 and Oct4 DNA binding dynamics in differentiating mESCs using ChIP-qPCR (chromatin immunoprecipitation linked with quantitative PCR analysis) at regulatory regions of Oct4, Sox2, Nanog and Brachyury. Unfortunately, ChIP analysis of Sox1 regulatory regions could not be undertaken as reproducible ChIP-qPCR results could not be generated. In mESCs, both Oct4 and Sox2 bound to known regulatory regions of Nanog, Sox2, Oct4 and Brachyury and as differentiation occurred (towards both ME and NE lineages) Oct4 and Sox2 enrichment fell at Nanog and Sox2 regulatory regions. However in ME cells, Oct4 became enriched at a regulatory site of the Sox2 gene, while Sox2 itself became depleted across this region and in NE cells; Sox2 became enriched at the same Sox2 regulatory region and also at a site on the Brachyury gene promoter. Overall this suggests that during differentiation Oct4 and Sox2 regulatory region binding is decreased, but a small fraction of sites can be enriched for these transcription factors to power lineage specific differentiation, and also reinforces the Oct4-ME and Sox2-NE association. This hypothesis was confirmed by the perturbation of Oct4 and Sox2 expression through plasmid mediated over-expression; Oct4 over-expression during RA-mediated NE differentiation repressed Sox1 expression but did not block Brachyury expression, while Sox2 over-expression during CHIR-mediated ME differentiation repressed Brachyury expression, and those cells which expressed Brachyury also expressed Oct4.

Together these experiments highlight the fact that well know pluripotency factors have additional important roles in lineage choice during mESC differentiation. These results reveal that Oct4 represses Sox1 and the neuroectodermal lineage choice, while Sox2 represses Brachyury and the mesendodermal lineage choice and so the differential activation of these genes regulates cell fate choice. The revelation that pluripotency maintenance and lineage choice are intrinsically linked will allow us to integrate what we already know of the pluripotent state and start to build dynamic models of how this network changes when signals enter into this network to initiate cellular changes, which may also give more insight into the reprogramming of somatic cells for induced pluripotent stem cell (iPSC) generation. However, the adjoining preview (Iovino and Cavalli) highlights that many questions remain to be addressed, such as how the differentiation signals induce specific repression, how Oct4 and Sox2 become differentially targeted in pluripotent and differentiated cell types, and how indeed each factor can have both activating and repressing roles. This type of defined system will be useful in answering such questions and will perhaps lead to the discovery of new co-factors, protein complexes, genetic and epigenetic regulatory mechanisms or signalling pathways that allow for specific differentiation along the different germ layers.



Chen X, Xu H, Yuan P, Fang F, Huss M, Vega VB, Wong E, Orlov YL, Zhang W, Jiang J, Loh YH, Yeo HC, Yeo ZX, Narang V, Govindarajan KR, Leong B, Shahab A, Ruan Y, Bourque G, Sung WK, Clarke ND, Wei CL, Ng HH.
Integration of external signaling pathways with the core transcriptional network in embryonic stem cells.
Cell. 2008 Jun 13;133(6):1106-17.

Lu R, Markowetz F, Unwin RD, Leek JT, Airoldi EM, MacArthur BD, Lachmann A, Rozov R, Ma'ayan A, Boyer LA, Troyanskaya OG, Whetton AD, Lemischka IR.
Systems-level dynamic analyses of fate change in murine embryonic stem cells.
Nature. 2009 Nov 19;462(7271):358-62.

Marson A, Levine SS, Cole MF, Frampton GM, Brambrink T, Johnstone S, Guenther MG, Johnston WK, Wernig M, Newman J, Calabrese JM, Dennis LM, Volkert TL, Gupta S, Love J, Hannett N, Sharp PA, Bartel DP, Jaenisch R, Young RA.
Connecting microRNA genes to the core transcriptional regulatory circuitry of embryonic stem cells.
Cell. 2008 Aug 8;134(3):521-33.

Wang J, Rao S, Chu J, Shen X, Levasseur DN, Theunissen TW, Orkin SH.
A protein interaction network for pluripotency of embryonic stem cells.
Nature. 2006 Nov 16;444(7117):364-8.

Thomson M, Liu SJ, Zou LN, Smith Z, Meissner A, Ramanathan S.
Pluripotency factors in embryonic stem cells regulate differentiation into germ layers.
Cell. 2011 Jun 10;145(6):875-89.

Shen X, Liu Y, Hsu YJ, Fujiwara Y, Kim J, Mao X, Yuan GC, Orkin SH.
EZH1 mediates methylation on histone H3 lysine 27 and complements EZH2 in maintaining stem cell identity and executing pluripotency.
Mol Cell. 2008 Nov 21;32(4):491-502.

Jackson SA, Schiesser J, Stanley EG, Elefanty AG.
Differentiating embryonic stem cells pass through 'temporal windows' that mark responsiveness to exogenous and paracrine mesendoderm inducing signals.
PLoS One. 2010 May 19;5(5):e10706.

Further Reading

Iovino N, Cavalli G.
Rolling ES Cells Down the Waddington Landscape with Oct4 and Sox2.

Cell. 2011 Jun 10;145(6):815-7.