DOT1L activity affects cell lineage progression in the developing brain by controlling metabolic programs

  1. Bismark Appiah
  2. Camilla L. Fullio
  3. Christiane Haffner
  4. Patrice Zeis
  5. Martin Treppner
  6. Patrick Bovio
  7. Arquimedes Cheffer
  8. Ilaria Bertani
  9. Harald Binder
  10. Dominic Grün
  11. Nereo Kalebic
  12. Elena Taverna
  13. Tanja Vogel

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Cortical neurogenesis depends on the tight balance between self-renewal and differentiation of apical progenitors (APs), the key progenitor type generating all other neural cells including neocortical neurons. We here report the activity of the histone methyltransferase DOT1L as a gatekeeper for AP cell identity. Combining lineage tracing with single-cell RNA sequencing of clonally related cells, we explore consequences of DOT1L inhibition on AP lineage progression during neurogenesis in the embryonic mouse neocortex. At the cellular level, DOT1L inhibition led to increased neurogenesis driven by a shift from asymmetric self-renewing to symmetric neurogenic divisions of APs. At the molecular level, we show that DOT1L activity preserved AP identity by promoting transcription of a gene set involved in AP metabolism. On a mechanistic level, DOT1L inhibition increased expression of metabolic genes, including microcephaly-associated Asparagine synthetase ( Asns ) and overexpression of ASNS in APs resulted in increased neuronal differentiation. Asns expression was predicted to be controlled through EZH2 and we show that DOT1L activity allows PRC2-mediated repression of Asns expression. Importantly, inhibition of ASNS activity rescued increased AP differentiation upon DOT1L inhibition. Our data show that DOT1L activity/PRC2 crosstalk controls AP lineage progression by regulating AP metabolism, and they provide a mechanistic view on how DOT1L activity might affect neocortical neurogenesis.

Article activity feed

  4. preLights

    Excerpt

    Enough neurons at the right moment: how cell division, epigenetics and metabolism interact to ensure that the brain is big enough.

  5. Review Commons

    Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    Manuscript number: RC-2022-01528

    Corresponding author(s): Elena Taverna and Tanja Vogel

    1. General Statements [optional]

    We thank the reviewers for the comments and points they raised. We think what we have been asked is a doable task for us and we are confident we will manage to address all points in a satisfactory manner.

    2. Description of the planned revisions

    Reviewer #1 (Evidence, reproducibility and clarity (Required)):

    Reviewer’s comment: The manuscript investigated the role of DOT1L during neurogenesis especially focusing on the earlier commitment from APs. Using tissue culture method with single-cell tracing, they found that the inhibition of DOT1L results in delamination of APs, and promotes neuronal differentiation. Furthermore, using single cell RNA-seq, they seek possible mechanisms and changes in cellular state, and found a new cellular state as a transient state. Among differentially expressed genes, they focused on microcephaly-related genes, and found possible links between epigenetic changes led by DOT1L inhibition and epigenetic inhibition by PRC2. Based on these findings, they suggested that DOT1L could regulate neural fate commitment through epigenetic regulation. Overall, it is well written and possible links from epigenetic to metabolic regulation are interesting. However, there are several issues across the manuscript.

    Response to Reviewer and planned revision:

    We thank the reviewer’s 1 for her/his comments and constructive criticism.

    We hope the revision plan will address the points raised by the reviewer in a satisfactory manner.

    Major issues:

    *Reviewer’s comment:

    1. It is not clear whether the degree of H3K79 methylation (or other histones) changes during development, and whether DOT1L is responsible for those changes. It is necessary to show the changes in histone modifications as well as the levels of DOT1L from APs to BPs and neurons, and to what extent the treatment of EPZ change the degree of histone methylation.

    Response to Reviewer and planned revision:

    • As for the level of DOT1L protein We tried several commercially available antibodies, but they do not work in the mouse, even after multiple attempts and optimization. So, unfortunately we will not be able to provide this piece of information.

    • As for the level of DOT1L mRNA We will provide info regarding the DOT1L mRNA level in APs, BPs and neurons by using scRNAseq data from E12, E14, E16 WT cerebral cortex.

    • As for the levels of H3K79methylation, we did not intend to claim that the histone methylation is responsible for the reported fate transition. We will edit the text to avoid any possible confusion. If it is deemed to be necessary to address the point raised by the reviewer, we do have 3 options, that we here in order of priority and ease of execution from our side.

    • immunofluorescence with an Ab against H3K79me2 using CON and EPZ-treated hemispheres.

    • FACS sort APs, BPs and neurons from CON and EPZ-treated hemispheres, followed by immunoblot for H3K79me2 to assess the H3K79me2 levels. As for the FACS sorting, we will use a combinatorial sorting in the lab on either a TUBB3-GFP or a GFP-reporter line using EOMES-driven mouse lines. This strategy has already been employed in the lab by Florio et al., 2015 and we will use it with minor modifications.

    • scCut&Tag for H3K79me2 from CON and EPZ-treated hemispheres. This option entails a collaboration with the Gonzalo Castelo-Branco lab in Sweden and might therefore require additional time to be established and carried out. Reviewer’s comment:

    Furthermore, the study mainly used pharmacological bath application. DOT1L has anti-mitotic effect, thus it is not clear whether the effect is coming from the inhibition of transmethylation activity.

    Response to Reviewer and planned revision:

    In a previous work we used a genetic model (DOT1L KO mouse) that showed microcephaly (Franz et al. 2019). For this study, we wanted to fill a gap in knowledge by understating if the DOT1L effect was mediated by its enzymatic activity. For this reason, we choose to use the pharmacological inhibition with EPZ, whose effect on DOT1L activity has been extensively reported and documented in literature (EPZ is a drug currently in phase clinical 3 studies).

    The stringent focus of this study on the pharmacological inhibition is thus a step toward understanding what specific roles DOT1L can play, both as scaffold or as enzyme.

    Here, we concentrate on the enzymatic function and the scaffolding function is beyond the scope of this specific study. We can further discuss and elaborate on the rationale behind this in the revised manuscript.

    Reviewer’s comment:

    In addition, the study assumed that the effect of EPZ is cell autonomous. However, if EPZ treatment can change the metabolic state in a cell, it would be possible that observed effects was non-cell autonomous. It would be important to address if this effect is coming in a cell-autonomous manner by other means using focal shRNA-KD by IUE.

    Response to Reviewer and planned revision:

    We did not claim that the effect of EPZ is cell autonomous, we are actually open on this point, as we consider both explanations to be potentially valid. We will edit the text to avoid any possible confusion on what we assume and what not.

    As a general consideration, it is entirely possible that the effects are non-cell autonomous. We will comment and elaborate on that in the revised manuscript.

    If the reviewer/journal considers this a point that must be addressed experimentally, then we will proceed as follows:

    • DOT1L shRNA-KD via in utero electroporation, followed by either
    • in situ hybridization for ASNS to check if ASNS transcript is increased upon DOT1L shRNA-KD compared to CON
    • FACS sorting of the positive electroporated cells (CON and DOT1L shRNA-KD), followed by qPCR to assess the levels of ASNS
    • If the reviewer wants us to check for a more downstream effect on fate, then we will immuno-stain the DOT1L shRNA-KD and CON with TUBB3 AB and/or TBR1 AB (as already done in the present version of the manuscript). Reviewer’s comment:
    1. The possible changes in cell division and differentiation were found by very nice single-cell tracing system. However, changes in division modes occurring in targeted APs such as angles of mitotic division and the expression of mitotic markers were not addressed. These information is critical information to understand mechanisms underlying observed phenotype, delamination, differentiation and fate commitment.

    Response to Reviewer and planned revision:

    Previous effects of DOT1L manipulation on the mitotic spindle were observed in a previous paper, using DOT1L KO mouse (Franz et al. 2019). Considering that in our experiments we do use a pharmacological inhibition, we will address this point by quantifying the spindle angle in CON and EPZ-treated cortical hemispheres.

    We will co-stain for DAPI to visualize the DNA/chromosomes, and for phalloidin (filamentous actin counterstain) that allows for a precise visualization of the apical surface and of the cell contour, as it stains the cell cortex.

    Of note, the protocols we are referring to are already established in the lab, based on published work from the Huttner lab (Taverna et al, 2012; Kosodo et al, 2005).

    Reviewer’s comment:

    1. The scRNA-seq analysis indicated interesting results, but was not fully clear to explain the observed results in histology. In fact, in single cell RNA-seq, the author claimed that cells in TTS are increased after EPZ treatment, which are more similar to APs. However, in histological data, they found that EPZ treatment increased neuronal differentiation. These data conflicts, thus I wonder whether "neurons" from histology data are actually neurons? Using several other markers simultaneously, it would be important to check the cellular state in histology upon the inhibition/KD of DOT1L.

    Response to Reviewer and planned revision:

    The reviewer’s comment is valid, and we indeed found that TTS cells are an intermediate state between APs and neurons in term of transcriptional profile. This is the reason why we called this cell cluster transient transcriptional state.

    We plan to address this point by staining for TBR1 and/or CTIP2 in CON and EPZ-treated hemispheres and to expand with this EOMES and SOX2 co-staining.

    Minor issues:

    Reviewer’s comment: Figure 1

    • It is not clear delaminated cells are APs, BPs or some transient cells (Sox2+ Tubb3+??). It is important to use several cell type-specific and cell cycle markers simulnaneously to characterize cell-type specific identity of the analysed cells by staining. These applied to Fig1B,D,E,F,G,as well as Fig2,3.

    Response to Reviewer and planned revision:

    We will address this point by using a combinatorial staining scheme for several fate markers such as TUBB3, EOMES and SOX2, as suggested by the reviewer.

    Reviewer’s comment:

    • Please provide higher magnification images of labelled cells (Fig 1H)

    Response to Reviewer and planned revision:

    In the revised manuscript, we will provide higher magnification for the staining.

    Reviewer’s comment:

    • Please provide clarification on the criteria of Tis21-GFP+ signal thresholding.

    Response to Reviewer and planned revision:

    In the revised manuscript, we will provide a clarification on the criteria of Tis21-GFP+ signal thresholding.

    Reviewer’s comment:

    • Splitting the GFP signal between ventricular and abventricular does not convincingly support the "more basal and/or differentiated" states after EPZ treatment.

    Response to Reviewer and planned revision:

    We will provide a clarification regarding this point.

    Reviewer’s comment:

    • Please explain the presence of Tis21-GFP+ cells at the apical VZ.

    Response to Reviewer and planned revision:

    Tis21-GFP+ cells at the apical VZ has been extensively reported in the literature, since the first paper by Haubensak et al. regarding the generation of the Tis21-GFP+ line. In a nutshell, T Tis21-GFP+ cells are present throughout the VZ (therefore also in the apical portion) as neurogenic, Tis21-GFP positive cells are undergoing mitosis at the apical surface. Indeed, the presence of Tis-21 GFP signal have been extensively used by the Huttner lab and collaborators to score apical neurogenic mitosis. In addition, since AP undergo interkinetic nuclear migration, it follows that Tis21-GFP+ nuclei are going to be present throughout the entire VZ.

    In the revised manuscript, we will explain this point and cite additional literature.

    Reviewer’s comment:

    • Order the legends in same order as the bars.

    Response to Reviewer and planned revision:

    We will follow reviewers’ recommendation and order the legends accordingly.

    Reviewer’s comment: Figure 2* *-Fig 2B) The difference between CON and EPZ apical contacts is not clear and does not match with the graph in Fig 2E.

    Response to Reviewer and planned revision:

    We will explain Fig. 2B in more detail and provide additional images in the revised manuscript.

    Reviewer’s comment: -Supp Fig 2 - are these injected slices cultured in control conditions? Please include this in the text and figure/figure legend

    Response to Reviewer and planned revision:

    In the revised manuscript, the text will be changed to address this point and provide clearer info.

    Reviewer’s comment: Fig 2C) The EPZ-treated DxA555+ cells exhibit morphological change of cell shape. Is this phenotype? please comment on the image shown for EPZ treatment panel.

    Response to Reviewer and planned revision:

    We thank the reviewer for having raised this point.

    The change in morphology might be a consequence of delamination and or of cell fate. In the revised manuscript, we will certainly better comment on this very relevant point and expand the discussion accordingly.

    Reviewer’s comment: Fig 2F - 2G) Data presented on EOMES+ and TUBB3+ % are counterintuitive. The authors claimed that TUBB3+ cells are increased and neuronal differentiation is promoted. However, no changes in EOMES+ are observed. What is the explanation? Did the author check the double positive cells? These could be TSS cells?

    Response to Reviewer and planned revision:

    We thank the reviewer to have raised this point.

    As envisioned by the reviewer, we suspect that the counterintuitive data might be due to TSS cell, which based on our scRNAseq data are expressing at the same time several cell type specific markers. It is possible that, since the treatment with EPZ is 24h long, cells (like the TTS cluster) have no time to completely eliminate the EOMES protein. If that were to be the case, then we would expect to still detect (as we indeed do) EOMES immunoreactivity.

    To address this point, we will:

    • analyze scRNA-seq data and check which is the extent of co-expression of Eomes and Tubb3 mRNAs in the TTS population.
    • Check for EOMES and TUBB3 double positive cells in the microinjection experiment. Reviewer’s comment: Figure 2 and Figure 3) the number of pairs analyzed for EPZ is twice as that of Con for comparison of the parameters taken into account. Please include n of each graph in the figure legend of the specific panel if not the same for all panels in that figure (i.e. for figure 3)

    Response to Reviewer and planned revision:

    We will revise the text accordingly.

    Reviewer’s comment:

    Figure 3) The data indicated that the number of daughter cell pairs in EPZ samples is almost double than Control. Is this the phenotype? More numbers of daughter cells in EPZ treated samples were observed from the same number of injections? or the number of injected cells were different?

    Response to Reviewer and planned revision:

    Due to technical reasons, we indeed performed a higher number of injections in EPZ-treated slices. We think this is the main reason behind the difference in number.

    If the reason were to be biological, one would expect to see the same trend in IUE experiments, but this is actually not the case. This does suggest/corroborate the idea that the reason behind the difference is mainly technical.

    Reviewer’s comment: Figure 4)

    • Please clarify if the single cell transcriptomic analysis has been performed only once, and if yes, how statistical testing to compare the cell proportion is carried out with only one batch. Fig 4G)

    Response to Reviewer and planned revision:

    As for the scRNAseq on microinjected cells:

    the scRNA-seq analysis was done once using cells pooled from 3 different microinjection experiments performed in 3 different days.

    As for the scRNAseq on IUE cells:

    The scRNA-seq analysis was done once using cells pooled from 2-3 different IUE experiments performed in 3 different days.

    For all scRNAseq experiments the statistical testing is achieved by intrasample comparisons according to established bioinformatics pipelines. We will better explain this point in the revised manuscript.

    Reviewer’s comment: Figure 4 and 5)

    • Figures are not supportive of the statement regarding APs' neurogenic potential upon DOT1L inhibition. TSS transcriptomic profile resembles more progenitors than neurons. Please comment on TSS neurogenic capacity taking into account the provided GO and RNAseq.

    Response to Reviewer and planned revision:

    We thank Reviewer 1 for raising this point, It is indeed true that TTS resemble more AP than neurons (as indicated in the Fig. S5B, C). We took that to indicate the fact that these cells are transient and therefore still maintain some AP features. Interestingly, TTS downregulate cell division markers, suggesting a restriction of proliferative potential, as one would expect for cells with an increased neurogenic potential. We will discuss this point in the revised manuscript.

    Reviewer’s comment:

    • Please provide GO analysis for APs and BPs.

    Response to Reviewer and planned revision:

    Following the reviewer’s suggestion, we will incorporate a more careful and in-depth analysis in the revised version of the manuscript.

    Reviewer’s comment:

    • Reconstruct figure 5A by listing genes in the same order in both Con and EPZ and prioritize EPZ-Con differences instead of cell-cell differences.

    Response to Reviewer and planned revision:

    We will revise Figure 5A based on the reviewer’s comment.

    Reviewer’s comment:

    Moreover, the presented genes in the heatmap is not the same in two conditions (i.e. NEUROG1 is present in EPZ but absent in Con). Please justify.

    Response to Reviewer and planned revision:

    This observation is based on different activities of transcription factor networks in the control and EPZ condition. They are not supposed to be the same as the cell states are altered and different TF are expressed and active upon the treatment in the diverse cell types. In a revised manuscript we will justify this point.

    Reviewer’s comment: Fig 5D)

    • Please explain why binding of EZH2 on the promoter of Asns is strongly reduced in comparison to a mild significant reduction of H3K79me/H3K27me3 in EPZ compared to Control.

    Response to Reviewer and planned revision:

    Several explanations are possible

    First, the variation can be due to batch effects.

    Second, the acute reduction of EZH2 might not be directly accompanied by a reduced histone mark, which is reduced either by cell division or by demethylases. The two processes of getting rid of the mark might be slower than the reduction of EZH2 presence at the respective site.

    Based on the reviewer’s comment, we will explain this point in the revised manuscript.

    Reviewer’s comment:

    Also is the changed directly medicated by DOT1L?

    Please test whether DOT1L can bind the promoter of Asns.

    Response to Reviewer and planned revision:

    To address this relevant issue we will proceed with the following protocol:

    • electroporate a tagged version of DOT1L into ESCs
    • select ESCs and differentiate them into NPC_48h.
    • treat NPC with DMSO (Con) or EPZ
    • harvest CON and EPZ-treated NPC
    • perform ChIP-qPCR DOT1L at the Asns promoter Reviewer’s comment: Please provide the expression patterns of DOT1L and Asns during neuronal differentiation.

    Response to Reviewer and planned revision:

    As for Dot1l

    Dot1l expression was shown in Franz et al 2019, by ISH from E12.5 to E18.5.

    As for Asns

    We will provide E14.5 in situ staining of Asns in the developing mouse brain using the Gene Paint database (see Figure below).

    We will also show immunostainings for ASNS at mid-neurogenesis, provided that Ab against ASNS works in the mouse.

    Other General comments:

    Reviewer’s comment: Please Indicate VZ, SVZ and CP on the side of the pictures/ with dot lines in the pictures both for primary figures and supplementary.

    Response to Reviewer and planned revision:

    We will revise the figures accordingly.

    Reviewer’s comment:

    • The Results and figures sometimes do not support the statement made by the authors

    Response to Reviewer and planned revision:

    We will carefully check on this and eliminate any overinterpretation or non-supported statements from the text.

    • Schemes are not informative/explanatory enough, i.e. time windows of treatment and sample collection, culture conditions details.

    Response to Reviewer and planned revision:

    We will revise the schemes to include more details. In particular, we plan to add a supplementary figure with a detailed visual description of the protocol, to match the detailed description presented in the materials and methods.

    Reviewer’s comment:

    • A more extensive characterization of TTS cells in terms of differentiation progression and integration would be enlightening

    Response to Reviewer and planned revision:

    In general, we are facing two main challenges while studying the TTS population: one is the lack of a specific marker gene for TTS, the other is the relatively small size of the TTS subpopulation.

    For these reasons, our ability to carry on an in-depth analysis of this cell state is limited.

    Considering the reviewer’s comment, in the revised manuscript we will expand the analysis ad characterization of the differentiation potential of TTS using RNA velocity trajectory.

    We can also expand the discussion on this point.

    Reviewer’s comment:

    • Picture quality can be improved, provide high magnification images.

    Response to Reviewer and planned revision:

    We will revise the figures to include higher magnification images.

    Reviewer #1 (Significance (Required)):

    Reviewer’s comment: The study could be important for the specific field in neural development. It aims to understand mutations in respective genes and brain malformation. If the link between epigenetic and metabolic changes is clearly shown, it will be interesting. However, the current manuscript is still rather descriptive, and clear mechanistic insights were not provided. The study have potentials and additional data will strength the value of study.

    Response to Reviewer and planned revision:

    We will address the direct impact of DOT1L and H3K79me2 on the Asns gene locus during the revision (see the rationale of the experimental strategy also in the revision plan above). We hope we will thus provide a mechanistic link between epigenetics and altered metabolome.

    Reviewer #2 (Evidence, reproducibility and clarity (Required)):

    Reviewer’s comment: Appiah et al. present a concise manuscript that provides details and possible mechanisms of their previous work (Franz et al., 2019; Ferrari et al., 2020). The study uses diverse lines of investigation to arrive at most conclusions. However, as interesting as the data is, we find that at the present state, it is not sufficient to prove that, indeed, the asparagine metabolism is regulated by DOTL1/PRC2 crosstalk. The neurogenic shift presented in the first part of the paper is not comprehensive and, therefore, not very convincing. The quality of images provided in the main and supplementary data is less than ideal. Additional data analysis and interpretation of the scRNA seq data may be needed. The authors finally conclude with rescue experiments done in culture and in-vivo, which we believe is the stand-out part of this study. Overall the manuscript has some interesting observations that are often over-interpreted with less supporting data. The manuscript reads well but requires additional data and changes in the claims/interpretation to be suited for publication.

    Response to Reviewer and planned revision:

    In the revised manuscript, we hope we will address the comments and concerns raised by the reviewer in a satisfactory manner. Comments

    Reviewer’s comment:

    1. Abstract: Is this statement correct: "DOT1L inhibition led to increased neurogenesis driven by a shift from asymmetric self-renewing to symmetric neurogenic divisions of APs. AP undergoes symmetric division for self-renewal and asymmetric neurogenic divisions.

    Response to Reviewer and planned revision:

    Based on the current literature (cit. Huttner and Kriegstein), AP undergo:

    • symmetric division for proliferative division at early stages of neurogenesis
    • asymmetric self-renewing division, generating an AP and a BP at mid neurogenesis. This division is also described as neurogenic, as it produces a BP, that is a step further than AP in term of neurogenic potential.
    • symmetric consumptive division at late neurogenesis To avoid any possible confusion, we will re-phrase the sentence to include the adjective “consumptive” and specify the composition of the progeny.

    In the revised manuscript, the sentence will read as follow:

    "DOT1L inhibition led to increased neurogenesis driven by a shift of APs from asymmetric self-renewing (generating one AP and one BP) to symmetric consumptive divisions (generating two neurons)"

    Reviewer’s comment: All the data is based on treatments with EPZ (DOTL1 inhibitor), yet no information is shown to support its targeted activity in this system. A proof of principle in the chosen experimental system is missing; for instance, examining the activity or protein level of DOTL1 and decreased methylation of the target(s) is essential.

    Response to Reviewer and planned revision:

    EPZ is a well characterized drug, that has been used previously in our lab and by others as well.

    As for our lab, the information regarding the inhibitor, its activity and efficiency in inhibiting DOT1L towards H3K79me2 was shown in Franz et al. Supplementary Fig. S6 D, E.

    In the present manuscript, an additional confirmation that EPZ targets DOT1L in regard to its H3K79me2 activity is shown in Fig. 5D.

    We would refer to this information more explicitly in a revised manuscript.

    Reviewer’s comment:

    1. Figure 1: The scoring of centrosomes and cilia is insufficient to conclude delamination and increase in basal fates. The effect could be on ciliogenesis or centrosome tethering to the apical end-feet of the AP, and other possible explanations for this observation also exist. The images are too small; larger images or graphic representations could be helpful in addition to the data.

    Response to Reviewer and planned revision:

    We did not intend to claim that the change in centrosome location demonstrate delamination, but only that it suggests delamination. This criterion has been extensively used as a proxy for delamination by several labs working on the cell biology of neurogenesis, such Huttner and Gotz labs. If the issue persists, we can re-phrase in a more cautious way the text referring to Figure 1 to highlight that the data only suggest delamination.

    Response to Reviewer and planned revision:

    To make a statement regarding delamination, I would like to see either the dynamics of delamination (organotypic slices images), staining with BP markers, or morphological changes of AP (staining that will reveal loss of adherence) or comparable data to support the observation. In my opinion Supp. Figure 1 is insufficient; the single image is not convincing; I would like to see 3D reconstruction and better-quality images.

    Response to Reviewer and planned revision:

    We can certainly provide better images and co-stain with relevant markers.

    We think it is beyond the scope of the manuscript embarking in live imaging as we are not studying the dynamics of delamination per se.

    Reviewer’s comment: Tis21 data (1H), again of low quality, is only a single piece of evidence and the conclusion "suggesting that the acquisition of a basal fate was paralleled by a switch to neurogenesis" is premature. I think other cell cycle exit reporters, Fucci markers, pHis, BrdU, NeuroD, or Tbr2 reporters (Li et al., 2020, (Haydar and Sestan labs)) to name a few, are necessary to establish the conclusions. The authors should show other markers such as PAX6, EOMES, or other upper-layer markers upon cell cycle exit in the SVZ/CP. These additional experiments will assist in cell fate analysis.

    Response to Reviewer and planned revision:

    We completely understand the points raised by the reviewer, and we plan to address them by co-staining with PAX6/SOX2, PH3 and/or EOMES.

    We think establishing the Fucci or EOMES mouse system is beyond the scope of the manuscript. In addition, given the present setting of all labs involved, it would be logistically unattainable (see also comments in the section below).

    We think the co-staining scheme and plan will be informative enough to satisfactory address the concerns raised by the reviewer.

    Reviewer’s comment:

    1. Figure 2: The microinjection experiments are elegant; the images, however, do not complement the experiment. The images of the microinjected cells seem not to be reconstructed from z-stacked optical slices, so often, processes are not continuous (panel B, for example); therefore, it is not clear if an apical process is indeed missing or just not seen.

    Response to Reviewer and planned revision:

    The mentioned images are reconstructed from continuous Z-stacks, as we always do given the type of data. We can provide better reconstructions and/or additional images.

    Reviewer’s comment:

    The data analysis should include other parameters; BrdU staining could have given information on cell cycle exit, PAX6, SOX2, and EOMES on the location of the cells in the VZ/sVZ. The quality of images showing EOMES and TUBB3 staining is so low that it makes the reader doubt the validity of the quantifications. "Taken together, these data suggest that the inhibition of DOT1L might favor the acquisition of a neuronal over BP cell fate" This interpretation should be subjected to more investigations. It is possible that this treatment just accelerates the AP-> BP -> Neuronal fate. The author's claim needs to be backed by additional experiments or be changed.

    Response to Reviewer and planned revision:

    To address this point, we will include in the revised manuscript staining and co-staining with PAX6, SOX2 (see also response above) and provide a BrdU labeling experiment.

    Reviewer’s comment:

    1. Figure 3: The experiment concept and its performance are impressive, yet the data is insufficient. The images in A that are supposed to be representative show two cells; their location is not clear, and the expression of GFP is not clear; in fact, both pairs seem to be GFP negative (not clear what is the threshold for background). Staining with anti-GFP and a second method to follow neurogenesis is necessary.

    Response to Reviewer and planned revision:

    We did use different staining methods and schemes to follow neurogenesis. As specified above, we will deepen our analysis by using additional markers, such as TBR1.

    Reviewer’s comment:

    1. On page 9, lines 8-10, the authors claim that their number of cells was "sufficient" for single-cell analysis; the numbers are Response to Reviewer and planned revision:

    In the revised manuscript, we will include the analysis of how many cells are needed to identify cluster of 6 cell types in this paradigm, based for example on the algorithms developed in Treppner et al. 2021.

    Reviewer’s comment:

    1. The authors use Seurat and RaceID without their appropriate citations in the first mention during the results. The authors also stop immediately after DEG analysis along with clustering. The authors could analyze their RNA-seq data with a trajectory; to say the least, the identification/characterization of TTS and neurons as Neurons I, II, and III are insufficient. There could be multiple ways to show the "fate" of cells in the isolated FACS, which the authors have missed.

    Response to Reviewer and planned revision:

    We will include the respective citations in a revised manuscript. We provide already differentiation trajectories but will include other methods, including scVelo of FateID to extend the trajectory analyses. We kindly ask the reviewer to also refer to the comments above regarding the TTs cluster characterization as part of our effort to provide a better picture of the different clusters.

    Reviewer’s comment:

    1. The authors detected candidates like Fgfr3, Nr2f1, Ofd1, and Mme as part of their treated (different approaches) datasets (from their DEG analysis). They correctly cite Huang et al., 2020 but fail to give us a sense of the consequences of these gene dysregulations. The authors can also validate if these proteins are expressed in their treated cells.

    Response to Reviewer and planned revision:

    In the revised manuscript we will comment on the function of the four genes mentioned.

    In addition, we will validate the expression of these genes on protein and transcriptional level through immunostainings -provided that antibodies are working in our system- or smFISH, respectively.

    Reviewer’s comment:

    1. The authors list a few GO terms (page 10, lines 1-10) and associate them with reduced proliferation; they must cite relevant studies. The authors can also add supplementary data showing which genes in their data correspond to these GO terms.

    Response to Reviewer and planned revision:

    We thank the reviewer for pointing out the missing citations.

    We of course agree on the need to add them, and we will do so in the revised manuscript.

    Reviewer’s comment:

    1. On Page 11, lines 3-7, the authors describe their method to arrive at the 17 targets with TF activity from the previous analysis. Can the authors describe the method used to correlate the two? The reviewer understands this could be MEME analysis or analysis of earlier datasets of Ferrari et al. 2020. But it must be explicitly stated, and a few examples in supplementary need to be exemplified as this analysis is key to discovering the three metabolic genes.

    Response to Reviewer and planned revision:

    In the revised manuscript, we will clarify the exact analysis that resulted in the identification of the 17 target genes, using the specific tool for gene network analysis, that is based on our scRNA-seq data alone, but not on the Ferrari et al 2020 data set.

    3. Description of the revisions that have already been incorporated in the transferred manuscript

    n/a

    4. Description of analyses that authors prefer not to carry out

    Reviewer’s comment: Tis21 data (1H), again of low quality, is only a single piece of evidence and the conclusion "suggesting that the acquisition of a basal fate was paralleled by a switch to neurogenesis" is premature. I think other cell cycle exit reporters, Fucci markers, pHis, BrdU, NeuroD, or Tbr2 reporters (Li et al., 2020, (Haydar and Sestan labs)) to name a few, are necessary to establish the conclusions. The authors should show other markers such as PAX6, EOMES, or other upper-layer markers upon cell cycle exit in the SVZ/CP. These additional experiments will assist in cell fate analysis.

    Response to Reviewer and planned revision:

    As pointed out above, we think establishing the Fucci or EOMES mice system is beyond the scope of the manuscript as it will not provide more information than the ones we will obtain from systematic and extensive co-staining experiments. In addition, all labs involved are facing a logistic issue (animal house not ready yet, construction works etc) that made the importing and setting up of the colony unattainable for the next 6-10months. If the reviewer and/or the editorial board think this is a major point compromising the entire revision, we kindly ask to contact us again so that we can discuss the issue and arrive to a shared conclusion.

  6. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    Appiah et al. present a concise manuscript that provides details and possible mechanisms of their previous work (Franz et al., 2019; Ferrari et al., 2020). The study uses diverse lines of investigation to arrive at most conclusions. However, as interesting as the data is, we find that at the present state, it is not sufficient to prove that, indeed, the asparagine metabolism is regulated by DOTL1/PRC2 crosstalk. The neurogenic shift presented in the first part of the paper is not comprehensive and, therefore, not very convincing. The quality of images provided in the main and supplementary data is less than ideal. Additional data analysis and interpretation of the scRNA seq data may be needed. The authors finally conclude with rescue experiments done in culture and in-vivo, which we believe is the stand-out part of this study. Overall the manuscript has some interesting observations that are often over-interpreted with less supporting data. The manuscript reads well but requires additional data and changes in the claims/interpretation to be suited for publication.

    Comments

    1. Abstract: Is this statement correct: "DOT1L inhibition led to increased neurogenesis driven by a shift from asymmetric self-renewing to symmetric neurogenic divisions of APs". AP undergoes symmetric division for self-renewal and asymmetric neurogenic divisions.

    All the data is based on treatments with EPZ (DOTL1 inhibitor), yet no information is shown to support its targeted activity in this system. A proof of principle in the chosen experimental system is missing; for instance, examining the activity or protein level of DOTL1 and decreased methylation of the target(s) is essential.

    1. Figure 1: The scoring of centrosomes and cilia is insufficient to conclude delamination and increase in basal fates. The effect could be on ciliogenesis or centrosome tethering to the apical end-feet of the AP, and other possible explanations for this observation also exist. The images are too small; larger images or graphic representations could be helpful in addition to the data.

    To make a statement regarding delamination, I would like to see either the dynamics of delamination (organotypic slices images), staining with BP markers, or morphological changes of AP (staining that will reveal loss of adherence) or comparable data to support the observation. In my opinion Supp. Figure 1 is insufficient; the single image is not convincing; I would like to see 3D reconstruction and better quality images.

    Tis21 data (1H), again of low quality, is only a single piece of evidence and the conclusion "suggesting that the acquisition of a basal fate was paralleled by a switch to neurogenesis" is premature. I think other cell cycle exit reporters, Fucci markers, pHis, BrdU, NeuroD, or Tbr2 reporters (Li et al., 2020, (Haydar and Sestan labs)) to name a few, are necessary to establish the conclusions. The authors should show other markers such as PAX6, EOMES, or other upper-layer markers upon cell cycle exit in the SVZ/CP. These additional experiments will assist in cell fate analysis.

    1. Figure 2: The microinjection experiments are elegant; the images, however, do not complement the experiment. The images of the microinjected cells seem not to be reconstructed from z-stacked optical slices, so often, processes are not continuous (panel B, for example); therefore, it is not clear if an apical process is indeed missing or just not seen. The data analysis should include other parameters; BrdU staining could have given information on cell cycle exit, PAX6, SOX2, and EOMES on the location of the cells in the VZ/sVZ. The quality of images showing EOMES and TUBB3 staining is so low that it makes the reader doubt the validity of the quantifications.
      "Taken together, these data suggest that the inhibition of DOT1L might favor the acquisition of a neuronal over BP cell fate" This interpretation should be subjected to more investigations. It is possible that this treatment just accelerates the AP-> BP -> Neuronal fate. The author's claim needs to be backed by additional experiments or be changed.
    2. Figure 3: The experiment concept and its performance are impressive, yet the data is insufficient. The images in A that are supposed to be representative show two cells; their location is not clear, and the expression of GFP is not clear; in fact, both pairs seem to be GFP negative (not clear what is the threshold for background). Staining with anti-GFP and a second method to follow neurogenesis is necessary.
    3. On page 9, lines 8-10, the authors claim that their number of cells was "sufficient" for single-cell analysis; the numbers are <500 for all samples. The authors need to justify this statement or articles that carefully analyze the number required for such a conclusion as references.
    4. The authors use Seurat and RaceID without their appropriate citations in the first mention during the results. The authors also stop immediately after DEG analysis along with clustering. The authors could analyze their RNA-seq data with a trajectory; to say the least, the identification/characterization of TTS and neurons as Neurons I, II, and III are insufficient. There could be multiple ways to show the "fate" of cells in the isolated FACS, which the authors have missed.
    5. The authors detected candidates like Fgfr3, Nr2f1, Ofd1, and Mme as part of their treated (different approaches) datasets (from their DEG analysis). They correctly cite Huang et al., 2020 but fail to give us a sense of the consequences of these gene dysregulations. The authors can also validate if these proteins are expressed in their treated cells.
    6. The authors list a few GO terms (page 10, lines 1-10) and associate them with reduced proliferation; they must cite relevant studies. The authors can also add supplementary data showing which genes in their data correspond to these GO terms.
    7. On Page 11, lines 3-7, the authors describe their method to arrive at the 17 targets with TF activity from the previous analysis. Can the authors describe the method used to correlate the two? The reviewer understands this could be MEME analysis or analysis of earlier datasets of Ferrari et al. 2020. But it must be explicitly stated, and a few examples in supplementary need to be exemplified as this analysis is key to discovering the three metabolic genes.

    Significance

    Appiah et al. present a concise manuscript that provides details and possible mechanisms of their previous work (Franz et al., 2019; Ferrari et al., 2020). The study uses diverse lines of investigation to arrive at most conclusions. However, as interesting as the data is, we find that at the present state, it is not sufficient to prove that, indeed, the asparagine metabolism is regulated by DOTL1/PRC2 crosstalk. The neurogenic shift presented in the first part of the paper is not comprehensive and, therefore, not very convincing. The quality of images provided in the main and supplementary data is less than ideal. Additional data analysis and interpretation of the scRNA seq data may be needed. The authors finally conclude with rescue experiments done in culture and in-vivo, which we believe is the stand-out part of this study.

    Overall the manuscript has some interesting observations that are often over-interpreted with less supporting data. The manuscript reads well but requires additional data and changes in the claims/interpretation to be suited for publication.

  7. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    The manuscript investigated the role of DOT1L during neurogenesis especially focusing on the earlier commitment from APs. Using tissue culture method with single-cell tracing, they found that the inhibition of DOT1L results in delamination of APs, and promotes neuronal differentiation. Furthermore, using single cell RNA-seq, they seek possible mechanisms and changes in cellular state, and found a new cellular state as a transient state. Among differentially expressed genes, they focused on microcephaly-related genes, and found possible links between epigenetic changes led by DOT1L inhibition and epigenetic inhibition by PRC2. Based on these findings, they suggested that DOT1L could regulate neural fate commitment through epigenetic regulation. Overall, it is well written and possible links from epigenetic to metabolic regulation are interesting. However there are several issues across the manuscript.

    Major issues:

    1. It is not clear whether the degree of H3K79 methylation (or other histones) changes during development, and whether DOT1L is responsible for those changes. It is necessary to show the changes in histone modifications as well as the levels of DOT1L from APs to BPs and neurons, and to what extent the treatment of EPZ change the degree of histone methylation. Furthermore, the study mainly used pharmacological bath application. DOT1L has anti-mitotic effect, thus it is not clear whether the effect is coming from the inhibition of transmethylation activity. In addition, the study assumed that the effect of EPZ is cell autonomous.However, if EPZ treatment can change the metabolic state in a cell, it would be possible that observed effects was non-cell autonomous. It would be important to address if this effect is coming in a cell-autonomous manner by other means using focal shRNA-KD by IUE.
    2. The possible changes in cell division and differentiation were found by very nice single-cell tracing system. However, changes in division modes occurring in targeted APs such as angles of mitotic division and the expression of mitotic markers were not addressed. These information is critical information to understand mechanisms underlying observed phenotype, delamination, differentiation and fate commitment .
    3. The scRNA-seq analysis indicated interesting results, but was not fully clear to explain the observed results in histology. In fact, in single cell RNA-seq, the author claimed that cells in TTS are increased after EPZ treatment, which are more similar to APs. However, in histological data, they found that EPZ treatment increased neuronal differentiation. These data conflicts, thus I wonder whether "neurons" from histology data are actually neurons? Using several other markers simultaneously, it would be important to check the cellular state in histology upon the inhibition/KD of DOT1L.

    Minor issues:

    Figure 1

    • It is not clear delaminated cells are APs, BPs or some transient cells (Sox2+ Tubb3+??). It is important to use several cell type-specific and cell cycle markers simulnaneously to characterize cell-type specific identity of the analysed cells by staining.These applied to Fig1B,D,E,F,G,as well as Fig2,3.

    • Please provide higher magnification images of labelled cells (Fig 1H)

    • Please provide clarification on the criteria of Tis21-GFP+ signal thresholding.

    • Splitting the GFP signal between ventricular and abventricular does not convincingly support the "more basal and/or differentiated" states after EPZ treatment.

    • Please explain the presence of Tis21-GFP+ cells at the apical VZ.

    • Order the legends in same order as the bars.

    Figure 2

    • Fig 2B) The difference between CON and EPZ apical contacts is not clear and does not match with the graph in Fig 2E.

    • Supp Fig 2 - are these injected slices cultured in control conditions? Please include this in the text and figure/figure legend

    Fig 2C) The EPZ-treated DxA555+ cells exhibit morphological change of cell shape. Is this phenotype? please comment on the image shown for EPZ treatment panel.

    Fig 2F - 2G) Data presented on EOMES+ and TUBB3+ % are counterintuitive. The authors claimed that TUBB3+ cells are increased and neuronal differentiation is promoted. However, no changes in EOMES+ are observed. What is the explanation? Did the author check the double positive cells? These could be TSS cells?

    Figure 2 and Figure 3) the number of pairs analyzed for EPZ is twice as that of Con for comparison of the parameters taken into account. Please include n of each graph in the figure legend of the specific panel if not the same for all panels in that figure (i.e. for figure 3)

    Figure 3)

    •  The data indicated that the number of daughter cell pairs in EPZ samples is almost double than Control. Is this the phenotype? More numbers of daughter cells in EPZ treated samples were observed from the same number of injections? or the number of injected cells were different?

    Figure 4)

    • Please clarify if the single cell transcriptomic analysis has been performed only once, and if yes, how statistical testing to compare the cell proportion is carried out with only one batch. Fig 4G)

    Figure 4 and 5)

    • Figures are not supportive of the statement regarding APs' neurogenic potential upon DOT1L inhibition. TSS transcriptomic profile resembles more progenitors than neurons. Please comment on TSS neurogenic capacity taking into account the provided GO and RNAseq.
    • Please provide GO analysis for APs and BPs.

    Figure 5)

    • Reconstruct figure 5A by listing genes in the same order in both Con and EPZ, and prioritize EPZ-Con differences instead of cell-cell differences. Moreover, the presented genes in the heatmap is not the same in two conditions (i.e. NEUROG1 is present in EPZ but absent in Con). Please justify. Fig 5D)
    • Please explain why binding of EZH2 on the promoter of Asns is strongly reduced in comparison to a mild significant reduction of H3K79me/H3K27me3 in EPZ compared to Control. Also is the changed directly medicated by DOT1L? Please test whether DOT1L can bind the promoter of Asns.

    Please provide the expression patterns of DOT1L and Asns during neuronal differentiation.

    Other General comments:

    • Please Indicate VZ, SVZ and CP on the side of the pictures/ with dot lines in the pictures both for primary figures and supplementary.
    • The Results and figures sometimes do not support the statement made by the authors
    • Schemes are not informative/explanatory enough, i.e. time windows of treatment and sample collection, culture conditions details..
    • A more extensive characterization of TTS cells in terms of differentiation progression and integration would be enlightening
    • Picture quality can be improved, provide high magnification images.

    Significance

    The study could be important for the specific field in neural development. It aims to understand mutations in respective genes and brain malformation. If the link between epigenetic and metabolic changes is clearly shown, it will be interesting. However, the current manuscript is still rather descriptive, and clear mechanistic insights were not provided. The study have potentials and additional data will strength the value of study.

  8. Version published to 10.1101/2022.04.08.487591v2 on bioRxiv
  9. Version published to 10.1101/2022.04.08.487591v1 on bioRxiv