The ZMYND8 chromatin factor protects cardiomyocyte identity and function in the mouse heart

  1. Andrew Kekūpaʻa Knutson
  2. Abigail Avelar
  3. Ralph V. Shohet

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Appropriate gene expression within cardiomyocytes is coordinated by chromatin factors and is essential for heart function. We investigated the role of the chromatin reader ZMYND8 in the mouse heart using null and conditional knockouts ( Zmynd8-cKO) . While full-length Zmynd8 is not required for cardiomyocyte development, Zmynd8-cKO mice develop cardiomegaly, decreased cardiac function, and premature death compared to controls. Transcriptome analysis of Zmynd8-cKO cardiomyocytes reveals illegitimate expression of transcripts normally limited to skeletal muscle. Additionally, we observe integration of TNNI2 skeletal troponin into cardiac sarcomeres of mutant mice. We conclude that ZMYND8 is necessary to maintain appropriate cardiomyocyte gene expression and cardiac function.

Article activity feed

  1. Review Commons

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

    Learn more at Review Commons

    Reply to the reviewers

    REVIEWER #1:

    The authors identify ZMYND8 as a bromodomain protein: is there evidence the actions described in this paper involve interaction of ZMYND8 with acetylated lysines? Does this mechanism play a role in ZMYND8's transcriptional regulatory activities?

    ZMYND8 is recruited to chromatin via its Bromo, PHD, and PWWP domains which recognize H3K4me1 and/or H3K14ac marks. Methyl marks on H3K4 are regulated by several lysine methyltransferases (e.g., MLL family and SETD1A/B) and demethylases (e.g., KDM5A-D) while H3K14ac is regulated by GCN5/PCAF, p300/CBP and/or Myst3. ZMYND8 also recruits histone deacetylases to chromatin including members of the highly conserved Nucleosome Remodeling and Deacetylase (NuRD) complex, HDAC1 and HDAC2. NuRD primarily deacetylates H3K27ac marks, however it is possible other acetyl moieties are affected by this complex.

    Using ChIP-seq, we now show that *Zmynd8-cKO *cardiomyocytes retain H3K27ac marks at misexpressed genes. Interestingly, while some of these genes have altered H3K27ac at their promoters (and therefore have full-length misexpressed transcripts; i.e., Casq1, Cdh16) other genes (e.g., Lamb3, Chst3) show changes in H3K27ac in the middle of the gene and this tracks with gene expression changes. We interpret this unusual transcript and H3K27ac pattern as evidence of potential ZMYND8-regulated intragenic enhancer elements. We include the following in our resubmission:

    1. Figure 5 which shows changes in H3K27ac levels at different genes, showing examples of genome browser tracks at the following genes Casq1, Cdh16, Camk1g, and Chst3.
    2. Supplemental Figure S5 showing H3K27ac and H3K27me3 marks at the cardiac myosin locus (i.e., *Myh6 *and Myh7) and surrounding genes in control and *Zmynd8-cKO * We also show retention of H3K27ac at the *Zmynd8 *gene in *Zmynd8-cKO *cardiomyocytes, again supporting an autoregulatory mechanism of *Zmynd8 *expression.
    3. An additional section in Results titled “H3K27 acetylation marks are retained at specific loci in *Zmynd8-cKO *cardiomyocytes”
    4. New “ChIP-seq and Analysis” section in Materials and Methods
    5. An updated model in Figure 6 that includes ZMYND8’s activities in modulating H3K27ac levels This first analysis on H3K27ac and H3K27me3 deposition in *Zmynd8-cKO *cardiomyocytes is not comprehensive and genome-wide analysis on these datasets will ultimately be performed in combination with additional datasets including ZMYND8 ChIP-seq from isolated cardiomyocytes. However, given the pertinence to ZMYND8’s transcriptional activities and in response to this reviewer’s critique, we include this pertinent H3K27ac and H3K27me3 ChIP-seq data.

    Given the newness of this model and multiple isoform issues, the authors should show the entire gel for the westerns in SFigure 1C.

    We now show the entire blots for all western blots in Supplementary Figure 1.

    Nuclear staining is in SFigure 1E (typo in text): most of the staining in the control is non-myocyte and non-nuclear, making the statement about IHC showing depletion less convincing for Nkx lines.

    We have fixed the typo in the text on page 5 line 128 and now correctly refer to this figure as Supplemental Figure S2. To better visualize nuclear ZMYND8 staining in this figure, we now show an adjusted image with increased contrast and brightness settings on both control and *Zmynd8-cKO *images and added arrowheads to indicate nuclei in the isolated cardiomyocytes. We also note that the flox sites only span the nuclear localization sequence for the protein so cytoplasmic ZMYND8 may still be present in *Zmynd8-cKO *cells.

    Regarding perinuclear ZMYND8 staining: am I accurate in observing the perinuclear staining is still present in the KO? What do the authors make of this?

    We do not observe perinuclear staining of ZMYND8 in KO cells. In Figure 1C, we believe the reviewer is observing potential staining in the cytoplasm, not perinuclear staining of ZMYND8 that we see in the control Myh6-CreTg/0 cardiomyocytes. We have added yellow arrowheads in Figure 1C to delineate perinuclear ZMYND8 staining we describe in the text.

    What is the protein amount in the Zmynd8fl/+ mice? Do the hearts upregulate the protein to compensate?

    We have added a gel in Supplemental Figure 1 that now shows protein isolated from Myh6-CreTg/0; Zmynd8fl/+ hearts and *Myh6-CreTg/0 *controls (Supplemental Figure 1C, right gel). It does not appear that Myh6-CreTg/0; Zmynd8fl/+ cardiomyocytes upregulate ZMYND8 to compensate for loss of one allele, as determined by Western blotting. However, our analysis shows differing ratios of the detected bands between conditional heterozygous mice, underscoring the need to further study the different ZMYND8 species present in cardiomyocytes. We state this in the results section (page 5, lines 123-124).

    Do the individual cardiomyocytes hypertrophy in the Zymnd8 cKO mice? Do they proliferate?

    Our analysis of cardiomyocyte morphology does reveal hypertrophy. The results we report include a new observation of variation in cell shape and are likely at least as sensitive as WGA staining which we find to be confounded by sectioning artifacts, cell identity, and position of the sections in the heart. We do not observe changes in H3S10ph staining between wild type and knockout hearts (data not shown) however we acknowledge further analysis of this may be warranted via other cell proliferation markers.

    Regarding this statement: "These results show that ZMYND8 is necessary to prevent the onset of contractile dysfunction that leads to heart failure and death." I think what the authors showed is that loss of ZMYND8 causes contractile dysfunction, heart failure and death.

    We acknowledge the difference in these statements and have now changed the text on page 7, lines 160-162 to “…these results show that loss of ZMYND8 from cardiomyocytes leads to contractile dysfunction, heart failure, and death.”

    The switch like up regulation of skeletal muscle genes is an interesting observation. Do the authors have any evidence how this works? Other studies with EZH2 are mentioned, and if ZYMND8 is in fact acting as a bromodomain, the mechanism might involve regulation of enhancer methylation/acetylation at K27. This is testable, certainly at the target genes the investigators have identified (Casq1 and Tnni2), by ChIP-PCR.

    As described above, we now include ChIP-seq data of H3K27ac and H3K27me3 marks in control and Zmynd8-cKO cardiomyocytes. As the reviewer suggests, there is retention of H3K27Ac marks in cKO cardiomyocytes, suggesting that ZMYND8 is necessary to recruit histone deacetylases to specific loci to remove acetyl moieties from H3K27. Regarding specific skeletal muscle genes, we do find a difference in histone acetylation at the promoter of the *Casq1 *gene and show this in Figure 5.

    The model in Figure 4C makes sense, but the authors do not present any data to support this molecular mechanism. If the authors ChIP for localization of TFs in KO vs control and/or examine histone marks, they could build support for this model, particularly since they have already identified target genes.

    We have now updated our model in Figure 6 to include ZMYND8’s role in modulating H3K27ac levels at target loci, leading to upregulation of mRNA transcripts. We add consideration of the implications of this in the Discussion.

    Reviewer #1 (Significance (Required)):

    The authors identify ZMYND8 as a bromodomain protein: is there evidence the actions described in this paper involve interaction of ZMYND8 with acetylated lysines? Does this mechanism play a role in ZMYND8's transcriptional regulatory activities? We include new data to demonstrate this. Please see above.

    REVIEWER #2:

    The study is reporting the role of ZMYND8 chromatin factor in the mouse heart. Mutations have been previously identified in genetic studies of atrioventricular septal defects and syndromic congenital cardiac abnormalities. Therefore the authors perform cardiomyocyte specific knockout of exon 4 (with the nuclear localisation signal) using Myh6 and Nkx2.5 cre. Full length protein seems to be removed from the nucleus. The knockout doesn't seem to affect embryonic development, but leads to hypertrophy and premature death. The authors perform transcriptome analysis and find 55 upregulated and 4 downregulated genes that are mainly related to contraction and ion transport. especially they find skeletal muscle proteins including fast-twitch troponin I upregulated. Tnni2 seems to be integrated into the sarcomeres, albeit the antibody staining is not in the expected location (see below). Shape of cardiomyocytes was apparently different, although this is seemingly not related to Tnni2 expression.

    Specific points:

    • ZMYND8 has been previously linked to atrioventricular septal defects, but the authors do not explore if this is the case also in their model; could the authors please expand

    We have not seen obvious septal defects in any *Zmynd8-cKO *mice. We now state this explicitly in the Results section on page 7, lines 159-160 and discuss this discrepancy from the observations in humans in our Discussion on page 12. The human study analyzing families carrying *Zmynd8 *mutations reported a variety of heart malformations in 7 of the 11 individuals. The septal defects observed in these individuals were not consistent and may be incidental to the molecular function of ZMYND8 within cardiomyocytes. One possibility is that these malformations are caused by stress during development, with *Zmynd8 *mutations sensitizing the heart to these defects. We acknowledge in the discussion that further analyses of septal defects in this knockout model could be useful in the future with more stringent stereoscopic techniques.

    • the initial section is difficult to follow. Especially, the authors seem surprised regarding the size of the bands. They should make clear what the expected band size should be after removal of exon 4 and if this doesn't fit, explore the reasons experimentally if possible.

    Rigorous analyses of the different *Zmynd8 *isoforms in cardiomyocytes will be a focus of future work as this may explain the mosaicism seen in cKO cardiomyocytes and the discrepancy between TNNI2 expression and cell shape (see below). We have reorganized the section and discuss potential explanations for our observed band sizes.

    • the authors explore the shape of the cardiomyocytes and find cells that are shorter and thicker. It would be meaningful to include other metrics including, sarcomere length, contractility measurements and calcium transients (especially in light of the change ion transporters).

    We agree that an investigation of the effects of the mutation and the skeletal muscle proteins on cellular contractility could be very interesting. Here we have contented ourselves with evaluating the effects at a physiological level through assessment of cardiac function.

    • it is unclear why Tnni2 stains for the M-band (where in fact should be no actin and troponin) and not a typical double band with the H zone excluded (see here for good staining example: https://www.biorxiv.org/content/10.1101/2020.09.09.288977v1.full.pdf). also the staining looks very fuzzy. can the authors provide evidence that the antibody is staining troponin I in skeletal muscle at the correct localisation to demonstrate the specificity of the antibody?

    We thank the reviewer for raising this point and do agree that there are instances where we observe TNNI2 staining colocalizing with MYOM1 staining. After closer examination of our images, we believe we do also see TNNI2 staining between M-lines and attribute this discrepancy to our antibody staining and/or biological differences between cells however, further analysis with better microscopy and immunostaining techniques is warranted. We have added an additional image to Figure 4A and have modified this results section on page 9, lines 217-222.

    • it is interesting why Tnni2 is detectable only in a subfraction of cells, but this remains unexplored. Could this e.g. be right vs left ventricular cardiomyocytes? or is this related to the remaining isoforms of ZMYND8? The authors should try to identify the source of this variability

    We agree that the TNNI2 mosaicism is an interesting phenotype and thank the reviewer for possible explanations. We favor the model of mosaicism being an effect of compensatory mechanisms by other ZMYND8 isoforms and discuss this in the discussion on page 8, line 228-229. This will be a focus of future work.

    • if Tnni2 is unrelated to the changes in hypertrophic phenotype of the cardiomyocytes, then the authors should aim to identify if one of the other differentially regulated proteins might be related (e.g. ion transporter). The experiments above might help to identify this

    We agree that identifying the causal agents of hypertrophy in this model would be interesting. It is however possible that we are simply seeing the expected effect of reduced contractility leading to hypertrophic compensation. Sorting this out will require additional mutant analyses and/or siRNA experiments all of which come with their own caveats and are outside of the scope of this initial analysis. Our aim for this manuscript was to report on the effects of ZMYND8 removal from cardiomyocytes. Additionally, it is certainly possible that phenotypes we report in this article are independent of the gene expression changes we have detected in the mutant and could be caused by other roles for ZMYND8 such as the DNA damage response. We include this possibility in our discussion.

    Reviewer #2 (Significance (Required)):

    Overall the manuscript is interesting in principle - it documents the role of a disease linked protein that hasn't been explored in the heart in detail, however at this point it seems premature and doesn't follow through on a solid detailed analysis.

    The change in transcription profiles and especially the upregulation of skeletal muscle isoforms is intriguing, but should be further explored. There seems a lack of hypothesis and instead the authors analyse Tnni2 and cell shape, but while the cell shape is different they don't find a correlation with Tnni2. so if the authors suggest that cell shape is important (as indeed might be), how is this regulated?

    Our goal for this initial paper is to describe the physiological and molecular phenotypes of the *Zmynd8-cKO *mouse model. It would be interesting to pursue a study directed at this question, perhaps of cell sorted "fat" and "thin" myocytes, but that would be beyond the scope of this report.

    The study could be of interest to cardiovascular researchers, but needs to be expanded on the points above.

    My expertise is in cardiovascular research

    REVIEWER #3:

    Reviewer #3 (Evidence, reproducibility and clarity (Required)):

    Summary:

    Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate). Please place your comments about significance in section 2.

    The authors found that Zmynd8-cKO mice develop dilated hearts, decreased cardiac function, and illegitimate expression of skeletal muscle genes. They concluded that ZMYND8 is necessary to maintain appropriate cardiomyocyte gene expression and cardiac function.

    Major comments:

    • Are the claims and the conclusions supported by the data or do they require additional experiments or analyses to support them? The claim that "Zmynd8 is dispensable for cardiac development" is not supported by the lethality of Zmynd8 D/D mice.

    We interpret our observation that viable Nkx2.5-CreTg/0; Zmynd8fl/fl mice are born and grow to adulthood as evidence that Zmynd8 is not necessary for establishment of the cardiac lineage. However, we do agree that labeling Zmynd8 as dispensable is not supported by the experiments using Zmynd8D/D mice. We hypothesize that the lethality of the Zmynd8D/*D *mice is due to early embryonic events since empty egg sacs were observed at E8.0, however we do agree that ZMYND8’s role in cardiac development cannot be assessed using this line. We state that empty yolk sacs are found in mother uteri 8 days after mating on page 4, lines 94-96.

    • Please request additional experiments only if they are essential for the conclusions. Alternatively, ask the authors to qualify their claims as preliminary or speculative, or to remove them altogether. The claim should be changed into "function of Zmynd8 in cardiac development can not be fully assessed in Zmynd8 D/D mice".

    We agree that the lethality of Zmynd8D/*D * mice prevents any analysis of early embryonic roles for the establishment of the cardiac lineage. This is additionally confounded by the fact that other partial-length isoforms of Zmynd8 may still be present in our knockout model. We have modified our interpretation and have further discussed the potential role of ZMYND8 in early cardiac development on page 4, line 96.

    • If you have constructive further reaching suggestions that could significantly improve the study but would open new lines of investigations, please label them as "OPTIONAL". OPTIONAL: What about the phenotype of Nkx2-5 Cre mediated knockout of Zmynd8? Is it more severe than Myh6 Cre mediated knockout? At more earlier embryonic stage when cardiomyocytes are differentiated, are the skeletal muscle developmental genes ectopically upregulated in heart tube?

    This is an interesting observation and deserves further investigation. Our initial analysis of *Nkx2.5-CreTg/0; Zmynd8fl/fl *mice reveals that these mice do not die earlier than *Myh6-CreTg/0; Zmynd8fl/fl *mice or have a more severe phenotype. In fact, mice with Nkx2.5-Cre mediated cKO mice live longer than Myh6-Cre mediated cKO mice. We show that these mice do have ZMYND8 depleted from their cardiomyocyte nuclei and ectopically express TNNI2.

    This discrepancy in phenotype has been observed recently in mice lacking Kdm8 (Ahmed et al, 2023) and has been attributed to a lower efficiency of the Nkx2.5-Cre recombinase compared to Myh6-driven Cre.

    • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated time investment for substantial experiments. Yes.

    • Are the data and the methods presented in such a way that they can be reproduced? Yes.

    • Are the experiments adequately replicated and statistical analysis adequate? Yes.

    Minor comments:

    • Specific experimental issues that are easily addressable. Have the female Zmynd8-cKO mice always died before their male siblings been pregnant with heart overload?

    All lifespan data are of non-pregnant females. All mice (i.e., both males and females) used in these analyses were not used for mating. We now explicitly say this in the mouse husbandry section of our Materials and Methods section.

    • Are prior studies referenced appropriately?

    This paper "De Novo ZMYND8 variants result in an autosomal dominant neurodevelopmental disorder with cardiac malformations" should be referenced.

    Thank you. We have referenced this paper (Dias et al. 2022) on page 3, line 61 as well as in the Discussion on page 9, line 211.

    • Are the text and figures clear and accurate? Description of "cardiomegaly, preventing a compact myocardium phenotype, heart enlargement and thinning of the ventricular" should be more accurate and professional. We have changed the following in the text:

    Page 6, line 150 “preventing a compact myocardium phenotype” to “during later stages of cardiac development” on

    Page 6, line 153 “heart enlargement” to “The heart weight of Zmynd8-cKO mice”

    Page 7, line 158 “thinning of the ventricular” to “dilated cardiomyopathy”

    • Do you have suggestions that would help the authors improve the presentation of their data and conclusions? GSEA analysis of RNA-seq can be used to show the enrichment of cardiac and skeletal genes.

    Because GSEA analysis requires at least three replicates per group to have the appropriate statistical power, we opted to show Gene Ontology analysis using DAVID software.

    Reviewer #3 (Significance (Required)):

    • General assessment: provide a summary of the strengths and limitations of the study. What are the strongest and most important aspects? What aspects of the study should be improved or could be developed? This study show that Zmynd8-cKO mice develop dilated hearts, decreased cardiac function, and illegitimate expression of skeletal muscle genes. However, the genes regulated by Zmynd8 during early developmental stage have not been identified and the functional mechanism of Zmynd8 during heart development remains unclear.

    • Advance: compare the study to the closest related results in the literature or highlight results reported for the first time to your knowledge; does the study extend the knowledge in the field and in which way? Describe the nature of the advance and the resulting insights (for example: conceptual, technical, clinical, mechanistic, functional,...). Genetic mutations of Zmynd8 have been identified in congenital heart diseases with cardiac structural defects. And this study further shows that dysfunction/weaker mutations of Zmynd8 as a reason for dilated cardiomyopathy with decreased function.

    • Audience: describe the type of audience ("specialized", "broad", "basic research", "translational/clinical", etc...) that will be interested or influenced by this research; how will this research be used by others; will it be of interest beyond the specific field? This study shows that dysfunction of Zmynd8 as a reason for dilated cardiomyopathy with decreased function. Researchers of "basic research" and "clinical" may be interested in this study.

    • Please define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate. heart development, dilated cardiomyopathy, epigenetics

  2. 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 #3

    Evidence, reproducibility and clarity

    Summary:

    Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate). Please place your comments about significance in section 2.

    The authors found that Zmynd8-cKO mice develop dilated hearts, decreased cardiac function, and illegitimate expression of skeletal muscle genes. They concluded that ZMYND8 is necessary to maintain appropriate cardiomyocyte gene expression and cardiac function.

    Major comments:

    • Are the claims and the conclusions supported by the data or do they require additional experiments or analyses to support them?

    The claim that "Zmynd8 is dispensable for cardiac development" is not supported by the lethality of Zmynd8 / mice.

    • Please request additional experiments only if they are essential for the conclusions. Alternatively, ask the authors to qualify their claims as preliminary or speculative, or to remove them altogether.

    The claim should be changed into "function of Zmynd8 in cardiac development can not be fully assessed in Zmynd8 / mice".

    • If you have constructive further reaching suggestions that could significantly improve the study but would open new lines of investigations, please label them as "OPTIONAL".

    OPTIONAL: What about the phenotype of Nkx2-5 Cre mediated knockout of Zmynd8? Is it more severe than Myh6 Cre mediated knockout? At more earlier embryonic stage when cardiomyocytes are differentiated, are the skeletal muscle developmental genes ectopically upregulated in heart tube?

    • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated time investment for substantial experiments.

    Yes.

    • Are the data and the methods presented in such a way that they can be reproduced?

    Yes.

    • Are the experiments adequately replicated and statistical analysis adequate?

    Yes.

    Minor comments:

    • Specific experimental issues that are easily addressable.

    Have the female Zmynd8-cKO mice always died before their male siblings been pregnant with heart overload?

    • Are prior studies referenced appropriately?

    This paper "De Novo ZMYND8 variants result in an autosomal dominant neurodevelopmental disorder with cardiac malformations" should be referenced.

    • Are the text and figures clear and accurate?

    Description of "cardiomegaly, preventing a compact myocardium phenotype, heart enlargement and thinning of the ventricular" should be more accurate and professional.

    • Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

    GSEA analysis of RNA-seq can be used to show the enrichment of cardiac and skeletal genes.

    Significance

    • General assessment: provide a summary of the strengths and limitations of the study. What are the strongest and most important aspects? What aspects of the study should be improved or could be developed? This study show that Zmynd8-cKO mice develop dilated hearts, decreased cardiac function, and illegitimate expression of skeletal muscle genes. However, the genes regulated by Zmynd8 during early developmental stage have not been identified and the functional mechanism of Zmynd8 during heart development remains unclear.
    • Advance: compare the study to the closest related results in the literature or highlight results reported for the first time to your knowledge; does the study extend the knowledge in the field and in which way? Describe the nature of the advance and the resulting insights (for example: conceptual, technical, clinical, mechanistic, functional,...). Genetic mutations of Zmynd8 have been identified in congenital heart diseases with cardiac structural defects. And this study further shows that dysfunction/weaker mutaions of Zmynd8 as a reason for dilated cardiomyopathy with decreased function.
    • Audience: describe the type of audience ("specialized", "broad", "basic research", "translational/clinical", etc...) that will be interested or influenced by this research; how will this research be used by others; will it be of interest beyond the specific field? This study shows that dysfunction of Zmynd8 as a reason for dilated cardiomyopathy with decreased function. Researchers of "basic research" and "clinical" may be interested in this study.
    • Please define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate. heart development, dilated cardiomyopathy, epigenetics
  3. 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

    The study is reporting the role of ZMYND8 chromatin factor in the mouse heart. Mutations have been previously identified in genetic studies of atrioventricular septal defects and syndromic congenital cardiac abnormalities. Therefore the authors perform cardiomyocyte specific knockout of exon 4 (with the nuclear localisation signal) using Myh6 and Nkx2.5 cre. Full length protein seems to be removed from the nucleus. The knockout doesn't seem to affect embryonic development, but leads to hypertrophy and premature death. The authors perform transcriptome analysis and find 55 upregulated and 4 downregulated genes that are mainly related to contraction and ion transport. especially they find skeletal muscle proteins including fast-twitch troponin I upregulated. Tnni2 seems to be integrated into the sarcomeres, albeit the antibody staining is not in the expected location (see below). Shape of cardiomyocytes was apparently different, although this is seemingly not related to Tnni2 expression.

    Specific points:

    • ZMYND8 has been previously linked to atrioventricular septal defects, but the authors do not explore if this is the case also in their model; could the authors please expand
    • the initial section is difficult to follow. Especially, the authors seem surprised regarding the size of the bands. They should make clear what the expected band size should be after removal of exon 4 and if this doesn't fit, explore the reasons experimentally if possible.
    • the authors explore the shape of the cardiomyocytes and find cells that are shorter and thicker. It would be meaningful to include other metrics including, sarcomere length, contractility measurements and calcium transients (especially in light of the change ion transporters)
    • it is unclear why Tnni2 stains for the M-band (where in fact should be no actin and troponin) and not a typical double band with the H zone excluded (see here for good staining example: https://www.biorxiv.org/content/10.1101/2020.09.09.288977v1.full.pdf). also the staining looks very fuzzy. can the authors provide evidence that the antibody is staining troponin I in skeletal muscle at the correct localisation to demonstrate the specificity of the antibody?
    • it is interesting why Tnni2 is detectable only in a subfraction of cells, but this remains unexplored. Could this e.g. be right vs left ventricular cardiomyocytes? or is this related to the remaining isoforms of ZMYND8? The authors should try to identify the source of this variability
    • if Tnni2 is unrelated to the changes in hypertrophic phenotype of the cardiomyocytes, then the authors should aim to identify if one of the other differentially regulated proteins might be related (e.g. ion transporter). The experiments above might help to identify this

    Significance

    Overall the manuscript is interesting in principle - it documents the role of a disease linked protein that hasn't been explored in the heart in detail, however at this point it seems premature and doesn't follow through on a solid detailed analysis.

    The change in transcription profiles and especially the upregulation of skeletal muscle isoforms is intriguing, but should be further explored. There seems a lack of hypothesis and instead the authors analyse Tnni2 and cell shape, but while the cell shape is different they don't find a correlation with Tnni2. so if the authors suggest that cell shape is important (as indeed might be), how is this regulated?

    The study could be of interest to cardiovascular researchers, but needs to be expanded on the points above.

    My expertise is in cardiovascular research

  4. 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 authors identify ZMYND8 as a bromodomain protein: is there evidence the actions described in this paper involve interaction of ZMYND8 with acetylated lysines? Does this mechanism play a role in ZMYND8's transcriptional regulatory activities?

    Given the newness of this model and multiple isoform issues, the authors should show the entire gel for the westerns in SFigure 1C. Nuclear staining is in SFigure 1E (typo in text): most of the staining in the control is non-myocyte and non-nuclear, making the statement about IHC showing depletion less convincing for Nkx lines.

    Regarding perinuclear ZMYND8 staining: am I accurate in observing the perinuclear staining is still present in the KO? What do the authors make of this?

    What is the protein amount in the Zmynd8fl/+ mice? Do the hearts upregulate the protein to compensate?

    Do the individual cardiomyocytes hypertrophy in the Zymnd8 cKO mice? Do they proliferate?

    Regarding this statement: "These results show that ZMYND8 is necessary to prevent the onset of contractile dysfunction that leads to heart failure and death." I think what the authors showed is that loss of ZMYND8 causes contractile dysfunction, heart failure and death.

    The switch like up regulation of skeletal muscle genes is an interesting observation. Do the authors have any evidence how this works? Other studies with EZH2 are mentioned, and if ZYMND8 is in fact acting as a bromodomain, the mechanism might involve regulation of enhancer methylation/acetylation at K27. This is testable, certainly at the target genes the investigators have identified (Casq1 and Tnni2), by ChIP-PCR.

    The model in Figure 4C makes sense, but the authors do not present any data to support this molecular mechanism. If the authors ChIP for localization of TFs in KO vs control and/or examine histone marks, they could build support for this model, particularly since they have already identified target genes.

    Significance

    The authors identify ZMYND8 as a bromodomain protein: is there evidence the actions described in this paper involve interaction of ZMYND8 with acetylated lysines? Does this mechanism play a role in ZMYND8's transcriptional regulatory activities?

  5. Version published to 10.1101/2022.10.06.510015v2 on bioRxiv
  6. Version published to 10.1101/2022.10.06.510015v1 on bioRxiv