You are here

| Pluripotent Stem Cells

Differential Recruitment of Methyl CpG-Binding Domain Factors and DNA Methyltransferases by the Orphan Receptor Germ Cell Nuclear Factor Initiates the Repression and Silencing of Oct4

From the July Edition of Stem Cells
By Stuart P. Atkinson

Oct4 (or Pou5f1) is one of a few genes recognised as being vitally important for the pluripotent nature of embryonic stem cells (ESCs) and is commonly used for induced pluripotent stem cells (iPSC) generation. Its downregulation during differentiation is essential, and indeed it has been suggested that a failure to properly down-regulate Oct4 could have detrimental effects for the application of cell types differentiated from iPSCs (Zhao et al). Gcnf (also known as Nr6a1), an orphan nuclear receptor germ cell nuclear factor, has been previously linked to Oct4 repression through binding to specific sequences, and the loss of Gcnf in mouse ESCs (mESCs) leads to the failure of Oct4 repression during differentiation (Gu et al, 2005). While the epigenetic contribution to pluripotency and differentiation has been the focus of intense study, specific studies of the epigenetic mechanisms regulating Oct4 expression have been lacking, and so now Gu et al, 2011, from the lab of Austin J. Cooney at the Baylor College of Medicine, Houston, Texas in a study published in the July edition of Stem Cells, attempt to delineate the mechanisms by which Gcnf mediates DNA methylation and repression of Oct4.

Initial analysis of Oct4 expression and promoter methylation status in Gcnf knockout (KO) mESCs during retinoic acid (RA) -mediated differentiation demonstrated that Oct4 does not become repressed and this is linked to the loss of DNA methylation at Oct4 promoter sequences. The authors also demonstrate that DNA methylation of the Oct4 promoter near the Gcnf binding site was also disrupted in Gcnf KO mouse embryos and this is linked to the loss of repression of Oct4, suggesting that in vitro analysis also reflects the in vivo situation. These two experiments suggest that Gcnf’s repressive function is mediated through the modulation of DNA methylation and therefore Gcnf should interact with proteins known to modulate this modification. To test this hypothesis, yeast two-hybrid protein interaction assays were established and the authors found that the Gcnf ligand binding domain specifically interacted with Mbd2, Mbd3, Dnmt1, Dnmt3a and Dnmt3b, strongly suggesting that Gcnf mediates the silencing of Oct4 through the recruitment of the DNA methylation machinery to promoter sequences during differentiation. Detailed analysis of these genes demonstrated that expression of Mbd3 and 4 was not significantly altered at the RNA or protein level in the Gcnf KO ESCs versus wild-type (WT) ESCs or during differentiation. Dnmt3a and b showed different dynamics, with Dnmt3a being induced upon differentiation and Dmnt3b being repressed in the WT ESCs, while in Gcnf KO ESCs, Dnmt3a expression did not change, but the rate of loss of Dnmt3b slowed. The ability of these factors to bind the Oct4 promoter was validated by chromatin immunoprecipitation (ChIP) analysis and showed binding of Gcnf, Mbd3 and Dnmt3a at the initial stages of Oct4 repression (Day 1-2)), with Mbd2 binding at a later stage of differentiation (Day 3), when Mbd3 and Dnmt3a binding is becoming reduced. In Gcnf KO ESCs, Gcnf, Mbd2, Mbd3 and Dnmt3a binding is abolished suggesting Gcnf recruits all three of these proteins.

The authors then went on to study these interactions further using KO ESCs for the DNA methylation factors that appear to be recruited by Gcnf. In Mbd3 KO ESCs, Oct4 is not properly silenced during differentiation but does not reach the levels seen in differentiating Gcnf KO ESCs and in undifferentiated WT ESCs, while Gcnf levels are enhanced. Mbd2 KO ESCs however showed similar levels of Oct4 and Gcnf as undifferentiated WT ESCs. Further studies utilised a LIF-rescue experiment which entails 4 days of treatment with LIF following the 6 days of RA-mediated differentiation to study the dynamics of methylation at the Oct4 promoter. In WT ESCs, Oct4 was silenced and silencing was maintained after the addition of LIF, with little or no alteration in the levels of DNA methylation. Conversely, although Oct4 repression was observed in Mbd2 and Mbd3 KO ESCs, LIF allowed for the re-expression of Oct4 in both these cell types. A lack of Mbd3 led to the loss of methylation of the Oct4 promoter during RA-mediated differentiation compared to WT ESCs which was unchanged upon LIF treatment. At 6 days of RA-mediated differentiation, DNA methylation in the Mbd2 KO ESCs was only slightly retarded, but the addition of LIF led to the loss methylation at the Oct4 promoter, so suggesting that Mbd3 may be required for the initiation DNA methylation while loss of Mbd2 leads to the failure to maintain DNA methylation established during differentiation. Further studies of Dnmt KO ESCs demonstrated that the loss of Dnmt1 led to a 50% reduction in DNA methylation during differentiation when compared to WT ESC at the Oct4 promoter. Absence of Dnmt3a led to a complete loss of DNA methylation upon differentiation which was not observed with of the knockdown of Dnmt3b, suggesting that Gcnf preferentially binds to Dnmt3a to initiate DNA methylation of the Oct4 promoter during differentiation. Analysis of Dnmt3a and 3b double KO ESCs that lack the ability to de novo methylate DNA demonstrated that Mbd3 remained capable of binding to the Oct4 promoter in a Gcnf-dependent manner to regulate Oct4 repression, while Mbd2 recruitment requires DNA methylation and Gcnf.

Overall, this data suggests a model for the initiation of Oct4 repression by the recruitment of Dnmt3a, Mbd2 and 3 by Gcnf with Mbd3 initiating silencing, Dnmt3a mediating do novo DNA methylation and Mbd2 maintaining the DNA methylation leading to long term repression. The works also demonstrates the relative importance of Gcnf in Oct4 repression during differentiation and elegantly defines the initiation of Oct4 repression under conditions of RA-mediated repression in mESC.

 

References

Zhao T, Zhang ZN, Rong Z, Xu Y.
Immunogenicity of induced pluripotent stem cells.
Nature. 2011 May 13;474(7350):212-5. doi: 10.1038/nature10135.

Gu P, LeMenuet D, Chung AC, Mancini M, Wheeler DA, Cooney AJ.
Orphan nuclear receptor GCNF is required for the repression of pluripotency genes during retinoic acid-induced embryonic stem cell differentiation.
Mol Cell Biol. 2005 Oct;25(19):8507-19.

Gu P, Xu X, Le Menuet D, Chung AC, Cooney AJ.
Differential Recruitment of Methyl CpG-Binding Domain Factors and DNA Methyltransferases by the Orphan Receptor Germ Cell Nuclear Factor Initiates the Repression and Silencing of Oct4.
Stem Cells. 2011 Jul;29(7):1041-51.