Mcm2 promotes stem cell differentiation via its ability to bind H3-H4

  1. Xiaowei Xu
  2. Xu Hua
  3. Kyle Brown
  4. Xiaojun Ren
  5. Zhiguo Zhang

  • Curated by eLife

    eLife logo

    eLife assessment

    This manuscript reports a novel role of Mcm2 licensing factor and helicase subunit of the Mcm2-Mcm7 complex in the differentiation of embryonic stem cells into neuronal lineages. A series of compelling experimental manipulations dissect the abnormalities in the formation of heterochromatin at pluripotent genes and the resolution of bivalent chromatin domains at lineage-specific genes in differentiation in response to mutation of the histone binding domain of Mcm2. These findings provide new insights into the replication-independent roles of Mcm2. This paper will be of interest to scientists working on development and embryonal cell differentiation.

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Mcm2, a subunit of the minichromosome maintenance proteins 2–7 (Mcm2-7) helicase best known for its role in DNA replication, contains a histone binding motif that facilitates the transfer of parental histones following DNA replication. Here, we show that Mcm2 is important for the differentiation of mouse embryonic stem (ES) cells. The Mcm2-2A mutation defective in histone binding shows defects in silencing of pluripotent genes and the induction of lineage-specific genes. The defects in the induction of lineage-specific genes in the mutant cells are likely, at least in part, due to reduced binding to Asf1a, a histone chaperone that binds Mcm2 and is important for nucleosome disassembly at bivalent chromatin domains containing repressive H3K27me3 and active H3K4me3 modifications during differentiation. Mcm2 localizes at transcription starting sites and the binding of Mcm2 at gene promoters is disrupted in both Mcm2-2A ES cells and neural precursor cells (NPCs). Reduced Mcm2 binding at bivalent chromatin domains in Mcm2-2A ES cells correlates with decreased chromatin accessibility at corresponding sites in NPCs. Together, our studies reveal a novel function of Mcm2 in ES cell differentiation, likely through manipulating chromatin landscapes at bivalent chromatin domains.

Article activity feed

  1. Version published to 10.7554/elife.80917 on eLife
  2. eLife

    Author Response

    Reviewer #1 (Public Review):

    Xu et al show that mutants in three DNA replication proteins, Mcm2, Pole3, and Pole4 have defects in differentiation in a mouse embryonic stem cell (ESC) model. The Mcm2 mutant (called Mcm2-2A), which specifically blocks the interaction of Mcm2 with histones, has defects in multilineage differentiation and neural differentiation, despite having minimal effect on ESC proliferation or gene expression. Mcm2-2A fails to fully silence ESC genes and activate appropriate differentiation genes. Chromatin profiling analyses show Mcm2 binds many promoters. During differentiation, the Mcm2-2 mutant retains K3K27me3 at differentiation gene promoters and reduced accessibility, consistent with the observed defects in gene expression.

    The findings that Mcm2-2A has minimal effect on proliferation and gene expression in ESCs, but impairs differentiation are interesting, particularly since this mutant seems to separate the histone binding roles of Mcm2 and its roles in DNA replication. Furthermore, the fact the histone binding function is only necessary when cells exit the pluripotent state is of interest. The studies were reasonably thorough and generally support the conclusions that Mcm2 is important for reshaping histone modifications during differentiation, although the details by which this occurs are not clear. Although the authors used two different strategies for identifying the direct binding sites of Mcm2 on chromatin, Mcm2 enrichment at individual loci was relatively weak, suggesting Mcm2 may localize somewhat diffusely. This somewhat weakens the conclusions about the direct vs indirect effects of Mcm2 on chromatin structure and gene expression.

    Overall, this paper reports an interesting set of findings that have a few caveats/limitations regarding how Mcm2 mediates these effects on chromatin during ESC differentiation.

    My biggest question is about the Mcm2 CUT&RUN data, which appears to have low signal-to-noise. The authors appear to be aware of this issue, as they also used an Mcm2-FLAG line for CUT&RUN studies, with similarly low signal to noise. To be clear, this may be due to the binding properties of Mcm2, which may bind chromatin relatively broadly, causing few highly enriched peaks to be observed (similar to cohesin complex in the absence of CTCF). However, it makes the Mcm2 binding data difficult to interpret. First, most Mcm2 peaks seem to be near promoters. Promoters often have a small amount of signal in negative control (IgG or irrelevant antibody) CUT&RUN experiments, presumably due to their MNase accessibility. It is not clear to what extent Mcm2 peaks exceed background because no negative control CUT&RUN was performed. The high correlation of FLAG and Mcm2 CUT&RUN libraries might still be evident if some of this signal is due to background at TSSs. Second, the authors call 13,742 peaks, but browser tracks of some example peaks at the Pax6 and Nanog promoters show minimal enrichment relative to surrounding regions (Fig. 5I, 5S1B). I have concerns that some of these peaks called statistically significant are not biologically meaningful.

    We thank the reviewer for his/her time to review this story and for his/her positive comments. We shared the reviewers’ concern about low signal to noise for Mcm2 CUT&RUN. However, the Mcm2 CUT&RUN signals most likely reflect Mcm2 binding.

    Reviewer #2 (Public Review):

    It is established that different histone chaperones not only facilitate the assembly of DNA into nucleosomes following DNA replication and transcription but also are essential to stem cell maintenance and differentiation. Here the authors Xiaowei Xu et al. propose a novel role for Mcm2 DNA helicase, a subunit of the origin licensing complex Mcm2-7 in stem cell differentiation in addition to or in connection to maintaining genomic integrity in DNA replication. This study is a continuation of the authors' previously published work implicating Mcm2-Ctf4-Polα axis in the parental histone H3-H4 transfer to lagging strands. The present study is elegantly executed with a systemic analysis of the role of Mcm2 in the ES differentiation to neuronal lineage.

    We thank the reviewer for his/her time to review the manuscript and for his/her positive comments.

    Major questions

    1. Mouse ES cells with a mutation at the histone binding motif of Mcm2 (Mcm2-2A) grew normally, but exhibited defects in differentiation. Also, the Mcm2-2A mutation linked global changes in gene expression, chromatin accessibility and histone modifications were not apparent to the similar degree in mouse ES cells compared to NPCs. The authors suggest that the excessive amount of Mcm2 in ES cells, similar to DNA replication, safeguards the chromatin accessibility and gene expression in mouse ES cells resulting in Mcm2-2A mutant ES cells being able to restore the symmetric distribution of parental histones before cell division. What is underlying the mechanism of this difference since overabundant Mcm2 is present in both ES cells and NPCs?

    This is an excellent good question that we can only speculate. As discussed above and below, our results indicate that Mcm2 functions with Asf1a to resolve the bivalent chromatin domains during pluripotency exit. Therefore, it is highly likely that Mcm2’s role in differentiation is independent of its role in DNA replication. Therefore, in the revised manuscript, we downplayed this possibility and suggested that the differentiation defects in Mcm2-2A mutant cells may arise from the involvement of Mcm2 in resolving bivalent chromatin domains (p24).

    1. CAF-1, Asf1a, and Mcm2 partake in similar or redundant chromatin regulation during differentiation with silencing of pluripotent genes and induction of lineage-specific genes. These processes were found commonly dysregulated in both Mcm2-2A cells and Asf1a KO ES cells, albeit with varying degrees. How can authors exclude the possibility of Mcm2 affecting the differentiation via Asf1 with which it forms a complex, as a potentially redundant mechanism in the deposition of newly synthesized or recycled histones?

    To address this question, we performed the following experiment. First, we overexpressed Asf1a in both WT and Mcm2-2A mutant ES cells and determined whether Asf1a overexpression suppress the differential defects in Mcm2-2A mutant cells (Figure 2- figure supplement 2A). We observed that Asf1a overexpression did not rescue the differential defects of Mcm2-2A mutant cells based on analysis of cell morphology (Figure 2- figure supplement 2B) as well as the expression of Oct4 and lineage specific genes during differentiation (Figure 2- figure supplement 2C-E).

    Second, we knocked out Asf1a in both WT cells and Mcm2-2A mutant cells using CRISPR/Cas9 (Figure 2- figure supplement 2F and 2G). and compared the effects of Asf1a KO, Mcm2-2A and Mcm2-2A Asf1a KO double mutation on differentiation. As detailed above, these results indicate that Mcm2’s function in the induction of lineage specific genes is dependent on Asf1a. However, Mcm2 also has independent role on the regulation of pluripotency genes which might through its unique roles on parental histone deposition and gene expression regulation. We discussed these points in the results (p10-13) and discussion (p22-23).

    It is known that CAF-1 and Mcm2 are involved in deposition of new H3-H4 and parental H3-H4, respectively. Further, there is little evidence that CAF-1 interacts with Mcm2 in the literature. Therefore, we did not analyze the relationship between CAF-1 and Mcm2 during differentiation. In the revised manuscript, we discussed these points to address the reviewer’s concern.

    Can authors test potential redundancy between Mcm2 and other histone chaperones and modifiers? Can the authors rescue the NPC phenotype induced by Mcm2 -2A mutant? Can the authors rescue the Mcm2-2A phenotype by overexpression of another histone chaperone or modifier?

    As stated above, we overexpressed Asf1a, which is known to interact with Mcm2, and found that overexpression of Asf1a did not rescue differentiation defects of Mcm2-2A mutant cells. On the other hand, overexpression of Mcm2 in Mcm2-2A cells did rescue defects in differentiation (Figure 2E-G). As discussed above, our results indicate that Mcm2 and Asf1a function in the same pathway for resolving bivalent chromatin domains based on analysis of Asf1a KO Mcm2-2A double mutant as well as RNA-seq datasets of Asf1a KO and Mcm2-2A during differentiation. However, the defects of Mcm2-2A on silencing of Oct4 was not observed in Asf1a mutant cells. Together, these results indicate that the defects in differentiation of Mcm2-2A cells are, at least in part, due to a reduced interaction with Asf1a. Furthermore, Mcm2 also has its unique role in promoting the silencing of pluripotency genes.

    1. Authors argue that Mcm2 may regulate the deposition of newly synthesized or recycled histones via the ability to recycle 1. parental H3.1 and H3.3, 2. via binding directly H3-H4, and/or via 3. Pol II transcription. Which of these mechanisms may be more unique to Mcm2 compared to the other histone chaperones and modifiers?

    This is a very interesting, but challenging question to address for the following reasons. First, while Mcm2-2A mutant showed defects in binding to both H3.1 and H3.3, it is almost impossible to identify a Mcm2 mutant that bind H3.1 and H3.3 differently. Based on our recent studies, our results indicate that the defects in the induction of lineage specific genes are likely due to a loss of Asf1a interaction. However, the defects in silencing of pluripotent gene such as Oct4 is unlikely due to a loss of interaction with Asf1a. Therefore, we suggest that defects in silencing of pluripotent genes in Mcm2-2A mutant cells are likely due to Mcm2’s role in parental histone transfer and/or gene transcription. In the revised manuscript, we dramatically modified the discussion section to reflect the new results as well as to further mitigate the concerns of the reviewer.

    1. Authors observed that in the ES cells the majority of Mcm2 CUT&RUN peaks were enriched with H3K4me3 CUT&RUN signals and ATAC-seq peaks and a small fraction of Mcm2 CUT&RUN peaks were engaged at the bivalent chromatin domains (H3K4me3+ and H3K27me3+). In contrast, in wild-type NPCs all the Mcm2 peaks co-localized with H3K4me3 and ATAC-seq peaks (H3K4me3+, H3K27me3-). The authors thus argued that Mcm2 binding to chromatin is rewired during differentiation citing this differential engagement of Mcm2 with the bivalent chromatin domains in ES and NPCs. What is the mechanism of Mcm2 differential engagement with the bivalent chromatin domains?

    As stated above, the original discussion may be misleading. In the revised manuscript, we dramatically rewrote the discussion based on the new results indicating that Mcm2 and Asf1a function similarly for the induction of lineage specific genes marked by bivalent promoters during pluripotency exit.

    1. Authors indicated that in mouse ES cells Mcm2 CUT&RUN peaks exhibited low densities at the origins. DNA replication origins are licensed by the MCM2-7 complexes, with most of them remaining dormant. Dormant origins rescue replication fork stalling in S phase and ensure genome integrity. It is reported that ESs contain more dormant origins than progenitor cells such as NPCs and that may prevent the replication stress. Also, partial depletion of dormant origins does not affect ECs self-renewal but impairs their differentiation, including toward the neural lineage. Moreover, reduction of dormant origins in NPCs impairs their self-renewal due to accumulation of DNA damage and apoptosis. Can authors exclude the role of reduced dormant origins reflected in the reduced density of Mcm2 at the origins in the differentiation to neuronal lineages?

    Thank the reviewer for excellent suggestions. We have now discussed these points about the potential role of Mcm2 in dormant origins and differentiation defects in the discussion (p24). However, I would like to point out that based on the new results, this is an unlikely mechanism. Supporting this idea, it is known that Mcm2-2A mutant cells from yeast and mouse ES cells are not sensitive to replication stress, such as HU (Foltman, Evrin et al. 2013, Huang, Stromme et al. 2015).

  3. eLife

    eLife assessment

    This manuscript reports a novel role of Mcm2 licensing factor and helicase subunit of the Mcm2-Mcm7 complex in the differentiation of embryonic stem cells into neuronal lineages. A series of compelling experimental manipulations dissect the abnormalities in the formation of heterochromatin at pluripotent genes and the resolution of bivalent chromatin domains at lineage-specific genes in differentiation in response to mutation of the histone binding domain of Mcm2. These findings provide new insights into the replication-independent roles of Mcm2. This paper will be of interest to scientists working on development and embryonal cell differentiation.

  4. eLife

    Reviewer #1 (Public Review):

    Xu et al show that mutants in three DNA replication proteins, Mcm2, Pole3, and Pole4 have defects in differentiation in a mouse embryonic stem cell (ESC) model. The Mcm2 mutant (called Mcm2-2A), which specifically blocks the interaction of Mcm2 with histones, has defects in multilineage differentiation and neural differentiation, despite having minimal effect on ESC proliferation or gene expression. Mcm2-2A fails to fully silence ESC genes and activate appropriate differentiation genes. Chromatin profiling analyses show Mcm2 binds many promoters. During differentiation, the Mcm2-2 mutant retains K3K27me3 at differentiation gene promoters and reduced accessibility, consistent with the observed defects in gene expression.

    The findings that Mcm2-2A has minimal effect on proliferation and gene expression in ESCs, but impairs differentiation are interesting, particularly since this mutant seems to separate the histone binding roles of Mcm2 and its roles in DNA replication. Furthermore, the fact the histone binding function is only necessary when cells exit the pluripotent state is of interest. The studies were reasonably thorough and generally support the conclusions that Mcm2 is important for reshaping histone modifications during differentiation, although the details by which this occurs are not clear. Although the authors used two different strategies for identifying the direct binding sites of Mcm2 on chromatin, Mcm2 enrichment at individual loci was relatively weak, suggesting Mcm2 may localize somewhat diffusely. This somewhat weakens the conclusions about the direct vs indirect effects of Mcm2 on chromatin structure and gene expression.

    Overall, this paper reports an interesting set of findings that have a few caveats/limitations regarding how Mcm2 mediates these effects on chromatin during ESC differentiation.

    My biggest question is about the Mcm2 CUT&RUN data, which appears to have low signal-to-noise. The authors appear to be aware of this issue, as they also used an Mcm2-FLAG line for CUT&RUN studies, with similarly low signal to noise. To be clear, this may be due to the binding properties of Mcm2, which may bind chromatin relatively broadly, causing few highly enriched peaks to be observed (similar to cohesin complex in the absence of CTCF). However, it makes the Mcm2 binding data difficult to interpret. First, most Mcm2 peaks seem to be near promoters. Promoters often have a small amount of signal in negative control (IgG or irrelevant antibody) CUT&RUN experiments, presumably due to their MNase accessibility. It is not clear to what extent Mcm2 peaks exceed background because no negative control CUT&RUN was performed. The high correlation of FLAG and Mcm2 CUT&RUN libraries might still be evident if some of this signal is due to background at TSSs. Second, the authors call 13,742 peaks, but browser tracks of some example peaks at the Pax6 and Nanog promoters show minimal enrichment relative to surrounding regions (Fig. 5I, 5S1B). I have concerns that some of these peaks called statistically significant are not biologically meaningful.

  5. eLife

    Reviewer #2 (Public Review):

    It is established that different histone chaperones not only facilitate the assembly of DNA into nucleosomes following DNA replication and transcription but also are essential to stem cell maintenance and differentiation. Here the authors Xiaowei Xu et al. propose a novel role for Mcm2 DNA helicase, a subunit of the origin licensing complex Mcm2-7 in stem cell differentiation in addition to or in connection to maintaining genomic integrity in DNA replication. This study is a continuation of the authors' previously published work implicating Mcm2-Ctf4-Polα axis in the parental histone H3-H4 transfer to lagging strands. The present study is elegantly executed with a systemic analysis of the role of Mcm2 in the ES differentiation to neuronal lineage.

    Major questions
    1. Mouse ES cells with a mutation at the histone binding motif of Mcm2 (Mcm2-2A) grew normally, but exhibited defects in differentiation. Also, the Mcm2-2A mutation linked global changes in gene expression, chromatin accessibility and histone modifications were not apparent to the similar degree in mouse ES cells compared to NPCs.
    The authors suggest that the excessive amount of Mcm2 in ES cells, similar to DNA replication, safeguards the chromatin accessibility and gene expression in mouse ES cells resulting in Mcm2-2A mutant ES cells being able to restore the symmetric distribution of parental histones before cell division.
    What is underlying the mechanism of this difference since overabundant Mcm2 is present in both ES cells and NPCs?

    2. CAF-1, Asf1a, and Mcm2 partake in similar or redundant chromatin regulation during differentiation with silencing of pluripotent genes and induction of lineage-specific genes. These processes were found commonly dysregulated in both Mcm2-2A cells and Asf1a KO ES cells, albeit with varying degrees.
    How can authors exclude the possibility of Mcm2 affecting the differentiation via Asf1 with which it forms a complex, as a potentially redundant mechanism in the deposition of newly synthesized or recycled histones?
    Can authors test potential redundancy between Mcm2 and other histone chaperones and modifiers? Can the authors rescue the NPC phenotype induced by Mcm2 -2A mutant? Can the authors rescue the Mcm2-2A phenotype by overexpression of another histone chaperone or modifier?

    3. Authors argue that Mcm2 may regulate the deposition of newly synthesized or recycled histones via the ability to recycle 1. parental H3.1 and H3.3, 2. via binding directly H3-H4, and/or via 3. Pol II transcription. Which of these mechanisms may be more unique to Mcm2 compared to the other histone chaperones and modifiers?

    4. Authors observed that in the ES cells the majority of Mcm2 CUT&RUN peaks were enriched with H3K4me3 CUT&RUN signals and ATAC-seq peaks and a small fraction of Mcm2 CUT&RUN peaks were engaged at the bivalent chromatin domains (H3K4me3+ and H3K27me3+). In contrast, in wild-type NPCs all the Mcm2 peaks co-localized with H3K4me3 and ATAC-seq peaks (H3K4me3+, H3K27me3-). The authors thus argued that Mcm2 binding to chromatin is rewired during differentiation citing this differential engagement of Mcm2 with the bivalent chromatin domains in ES and NPCs. What is the mechanism of Mcm2 differential engagement with the bivalent chromatin domains?

    5. Authors indicated that in mouse ES cells Mcm2 CUT&RUN peaks exhibited low densities at the origins. DNA replication origins are licensed by the MCM2-7 complexes, with most of them remaining dormant. Dormant origins rescue replication fork stalling in S phase and ensure genome integrity. It is reported that ESs contain more dormant origins than progenitor cells such as NPCs and that may prevent the replication stress. Also, partial depletion of dormant origins does not affect ECs self-renewal but impairs their differentiation, including toward the neural lineage. Moreover, reduction of dormant origins in NPCs impairs their self-renewal due to accumulation of DNA damage and apoptosis.
    Can authors exclude the role of reduced dormant origins reflected in the reduced density of Mcm2 at the origins in the differentiation to neuronal lineages?

  6. Version published to 10.1101/2022.06.16.496513v1 on bioRxiv