Differential regulation of lineage commitment in human and mouse primed pluripotent stem cells by NuRD

Background

Stem cell differentiation is accompanied by extensive changes in gene expression as cells transition from pluripotency to differentiated states. These gene expression changes are influenced by the activities of chromatin remodelling complexes that alter the positions of nucleosomes and/or modify histones. The Nucleosome Remodelling and Deacetylase (NuRD) complex is one such chromatin modifying complex. This complex is well conserved in vertebrates and is known to have roles in coordinating differentiation at various stages of development. In pluripotent cells, the NuRD complex is thought to act to facilitate the exit from the self-renewing state and regulate the expression of differentiation related genes. Despite being well-studied in model organisms, the NuRD complex has not been fully characterised in human stem cells. In this study, Ragheb et al. investigate the role of NuRD in human pluripotent cells and uncover some surprising differences between the role of NuRD in human and mouse.

 

Key findings

NuRD complex structure is conserved in mouse and human pluripotent stem cells

The authors began their investigation by comparing the composition of the NuRD complex in mouse and human pluripotent cells. Genome editing was used to add a FLAG epitope tag to the core MBD3 NuRD subunit in human induced pluripotent stem cells (hiPSCs). This tag was used to purify interacting proteins from the human cells, as well as from mouse stem cells containing an identical modification.

Orthologues of known NuRD complex components were purified from both cell lines along with other previously characterised interacting proteins. Overall, very few differences were observed in the biochemical composition of the NuRD complex between human and mouse cells, reflecting the fact that the complex is very highly conserved between the two species.

 

NuRD mutant hiPS cells are unable to maintain a stable pluripotency state

Ragheb et al. went on to study the functional consequences of disrupting the NuRD complex in human cells. Using genome editing, they created an MDB3 knockout cell line. Initial characterisation of these cells showed a surprising result. Rather than observing a homogenous culture of stem cells as seen with an equivalent knockout in mice cells, some of the human stem cells appeared to display features of differentiated cells. Given that mouse MDB3-KO cells are resistant to differentiation, the finding that human MDB3-KO cells would instead display markers of premature differentiation is unexpected. However, when these cells were induced to differentiate, they also failed to display some of the characteristic morphological markers of differentiated cells that were present in control cells or cells rescued with an MBD3 transgene. From these data, the authors conclude that in human cells, NuRD is required to not only maintain pluripotency, but also to coordinate cell fate during differentiation.

 

Human NuRD activity is required for appropriate transcriptional response to differentiation signals

To better understand the role of human NuRD in differentiation and pluripotency, the authors next profiled the transcriptomes of control, MDB3-KO, and rescued cell lines at various timepoints throughout induced differentiation. In support of their previous observations, they saw that many pluripotency related genes were downregulated before differentiation was induced.

Analysis of gene expression changes across the differentiation time-course resulted in the identification of several clusters of genes that were differentially expressed. The largest cluster represented genes that were normally silenced as differentiation progressed in control cells. In the MDB3-KO cells, these genes started at reduced levels reflecting the premature differentiation state of these cells, yet were reduced further as differentiation progressed. The remaining clusters contained genes involved in differentiation which failed to respond appropriately to differentiation signals across all timepoints. Furthermore, no common pathways could be identified in the genes comprising these clusters, indicating that the cells were failing to respond to a range of signals, rather than just one specific pathway.

Overall, it seemed that whilst human NuRD deficient cells were able to respond to signals associated with the loss of pluripotency, they did not appropriately activate the gene expression programme required to coordinate differentiation.

 

MBD3/NuRD controls lineage commitment differently in human and mouse primed PSC

Since conflicting results had been observed between the role of NuRD in mouse ESCs and human stem cells, the authors sought to compare these cell types more closely. Since the human cells used were in a “primed” pluripotent state, whilst previous results from mouse cells were conducted largely on “naïve” cells, it was unclear whether the differences observed were due to the species or the cell state. To address this, NuRD knockouts in primed pluripotent mouse cells (mEpiSCs) were examined. These cells resembled the naïve knockout cells, showing no signs of premature differentiation.

Next the transcriptomes of mouse naïve pluripotent cells were profiled at the same timepoints during differentiation as examined previously for human cells. In contrast to human cells which showed an overall lack of response to differentiation signals, the mouse cells up or downregulated differentiation or pluripotency related genes, but at inappropriate levels to correctly coordinate differentiation.

 

Summary

Ragheb et al. conclude from these experiments that the role of NuRD in regulating pluripotency and differentiation is subtly different in human and mouse. In the graphical abstract (Fig. 1) the authors liken this to an orchestra in which NuRD is the conductor. In this analogy, the human and mouse orchestras respond to the loss of NuRD in different ways. The mouse musicians fail to harmonise, but still make the right kind of noises, whilst the humans just produce a terrible racket!

Overall, this preprint provides compelling evidence that despite being closely conserved between the two species, human and mouse NuRD activity in regulating differentiation displays markedly different characteristics. I think this study is important because it provides a nice reminder that the models we use are just that – models, and that even the most highly conserved systems may behave differently in different contexts.

 

 

doi: https://doi.org/10.1242/prelights.17179

Read preprint (No Ratings Yet)