On 2016 Jan 25, Jacob H. Hanna commented:

Thank you for the valuable information.

We highlight that the gene expression data in Chd4 null embryos presented in this paper, and the immuno-staining for Oct4/Nanog on Mbd3 null embryos in Kaji et al. Development 2007 <PMID 17287250>, both show formation of hallmarks of pluripotency in vivo at the pre-implantation epiblast ( http://imgur.com/lsM6kbV ). This is in contrast to Nanog null embryos for example, that cannot sustain Oct4+ cells at E3.5-E4.5 ICMs ( http://imgur.com/lsM6kbV ).

As Mbd3 KO ESCs derivation in vitro is not compromised under optimized growth conditions (e.g. 2i/LIF (Rais et al. Nature 2013), KSR/LIF or high quality FBS/LIF conditions), we will be also testing Chd4 KO ESC derivation from null mice to finalize the discussion and provide a more complete and solid answer.

On 2016 Jan 25, Brian Hendrich commented:

This paper is about the function of Chd4 in cells of preimplantation stage embryos. It is not about ES cells. Anyone wishing to attempt derivation of Chd4-null ES cells can obtain our Chd4 floxed mouse line which has been deposited with MRC Harwell (http://www.har.mrc.ac.uk/) for distribution.

On 2016 Jan 23, Jacob H. Hanna commented:

I would first like to congratulate the authors on this elegant and important paper. The Hendrich group has previously purported in Kaji et al. Development 2007 paper that it is absolutely impossible to derive Mbd3-/- ESCs from Mbd3 null E3.5 embryos. This result was surprising considering that the authors show in the same paper presence of Oct4+/Nanog+ cells in Mbd3-/- ICMs. Further, Mbd3-/- ESCs are "hyper-naive" and resist differentiation even in the absence of LIF (Kaji et al. Nature Cell Biology 2006, Reynolds et al. Cell Stem Cell 2012). Our group has revisited this result in Rais et al. Nature 2013, and was able to efficiently derive Mbd3-/- ESCs in serum free enriched 2i/LIF conditions. This suggests that the main conclusion of Kaji et al. 2007 Development paper Kaji K, 2007 titled "Mbd3 is required for development of pluripotent cells", is invalid and constitutes an artifact possibly due to using a low quality fetal bovine serum (FBS) batch.

In O'Shaughnessy-Kirwan et al. Development 2015 have generated Chd4 null mouse embryos and show that Oct4+/Nanog+ ICM is formed in vivo, just like in Mbd3-/- embryos Kaji K, 2007. This indicates again that Mbd3/Chd4/NuRD is dispensable for the formation of pluripotent cells in vivo. It is disappointing that the authors did not describe attempts to derive Chd4 knockout ESCs in this work and explain their ambiguous and controversial results on NuRD complex null ESC derivations Kaji K, 2007.

Jacob (Yaqub) Hanna M.D. Ph.D.

Department of Molecular Genetics

Weizmann Institute of Science | 234 Herzl St, Rehovot 7610001, Israel

Email: jacob.hanna@weizmann.ac.il

Lab website: http://hannalabweb.weizmann.ac.il/

Twitter: @Jacob_Hanna

On 2016 Jan 23, Jacob H. Hanna commented:

I would first like to congratulate the authors on this elegant and important paper. The Hendrich group has previously purported in Kaji et al. Development 2007 paper that it is absolutely impossible to derive Mbd3-/- ESCs from Mbd3 null E3.5 embryos. This result was surprising considering that the authors show in the same paper presence of Oct4+/Nanog+ cells in Mbd3-/- ICMs. Further, Mbd3-/- ESCs are "hyper-naive" and resist differentiation even in the absence of LIF (Kaji et al. Nature Cell Biology 2006, Reynolds et al. Cell Stem Cell 2012). Our group has revisited this result in Rais et al. Nature 2013, and was able to efficiently derive Mbd3-/- ESCs in serum free enriched 2i/LIF conditions. This suggests that the main conclusion of Kaji et al. 2007 Development paper Kaji K, 2007 titled "Mbd3 is required for development of pluripotent cells", is invalid and constitutes an artifact possibly due to using a low quality fetal bovine serum (FBS) batch.

In O'Shaughnessy-Kirwan et al. Development 2015 have generated Chd4 null mouse embryos and show that Oct4+/Nanog+ ICM is formed in vivo, just like in Mbd3-/- embryos Kaji K, 2007. This indicates again that Mbd3/Chd4/NuRD is dispensable for the formation of pluripotent cells in vivo. It is disappointing that the authors did not describe attempts to derive Chd4 knockout ESCs in this work and explain their ambiguous and controversial results on NuRD complex null ESC derivations Kaji K, 2007.

Jacob (Yaqub) Hanna M.D. Ph.D.

Department of Molecular Genetics

Weizmann Institute of Science | 234 Herzl St, Rehovot 7610001, Israel

Email: jacob.hanna@weizmann.ac.il

Lab website: http://hannalabweb.weizmann.ac.il/

Twitter: @Jacob_Hanna

On 2016 Jan 25, Brian Hendrich commented:

This paper is about the function of Chd4 in cells of preimplantation stage embryos. It is not about ES cells. Anyone wishing to attempt derivation of Chd4-null ES cells can obtain our Chd4 floxed mouse line which has been deposited with MRC Harwell (http://www.har.mrc.ac.uk/) for distribution.