Chromatin Organization during C. elegans Early Development

Round 1

Reviewer 1 Report

This review summarizes work conducted in C. elegans embryonic development and early larval stages (early development) that investigate several aspects of chromatin regulation. The authors describe data ranging from histone post translational modification to the spatial organization of the genome. While the review covers an important topic and does a nice job in highlighting the recent advances in the field, some citations are not correct and some sentences contain conclusions that don´t match with the cited paper (as outlined below in the major points). In other cases, the sentences were difficult to understand (as outline below).

Major points:

-        The review would greatly benefit from a description of the main points discussed (H3K9me regulation, H3K27me3, active marks, 3D genome architecture and so on..) that is divided according to the timing of development. For example, describing what chromatin changes are present/deposited/erased (as appropriate): i) in germ cells and are transmitted (or not) to the embryos, ii) in very early embryos (including ZGA and up to gastrulation-up to the authors to decide if they want to focus on ZGA in a separate section), iii) post gastrulation embryos and, eventually, iv) late embryos (post comma stage or post 1.5 fold). When it is too much work to reorganize the text in such way, then I think that at the minimum, the authors should make a figure that summarizes the different chromatin organization changes that occur as development of the embryo progresses, as explained above. Indeed, the two tables present in the current version of the review are somewhat redundant, while no figure is presented. I would suggest to keep the first table, maybe adding the info on HTZ-1 and H1.1 that is present in the second one, but to definitely make a schematic representation that summarizes what happens as development progresses. Highlighting this aspect with a dedicated figure would not only help the reader to better understand when in development the main chromatin reorganization events take place in worms, but also it would help to draw parallels with very recent research conducted in mammals and that shows how chromatin organization is regulated differently during the different phases of very early embryogenesis with respect to ZGA. For example, DOI: 10.1101/gad.350799.123, DOI: 10.1038/s41586-023-06872-1.

-        Lines 85-86: “MET-2 is localized in the cytosol in early embryos and is transported inside the nucleus at the onset of gastrulation in 20-50 cell embryos by its co-factor LIN-65 [18].” I don´t think that the correct citation for this is Delaney 2019 that states “Previous results suggested that an overexpressed mCherry::MET-2 fusion protein localized to the cytoplasm of early C. elegans embryos (Towbin et al., 2012). In contrast, we show here, by tagging the endogenous met-2 locus to generate a fully functional fusion, that MET-2 is enriched in the nucleus at the nuclear periphery in irregular-sized foci, throughout C. elegans development.” The correct citation for the authors´sentence is Mutlu 2018. On the contrary, both works agree on the requirement of LIN-65 for MET-2 nuclear foci formation.

-        Lines 99-102: “In lin-65 and arle-14 mutants, repetitive DNA elements are significantly de-repressed, indicating that heterochromatin formation is not as robust in these mutants.” The correct reference for this statement is missing as in the Mutlu 2018 paper they only tested 2 repetitive elements and for one of them, there's only a bit of de-repression in lin-65 mutant and almost no increase in arle-14. More repeats are analyzed in the Delaney 2019 paper, however there it appears that only lin-65 and not arle-14 is required for silencing. The sentence should be rephrased and the authors should not state a role for arle-14 in silencing repeats, unless they know of another work that I am unaware of. In that case, it should be cited.

-        Lines 106-107: “Increasing and decreasing the length of the S phase in early embryos by growing worms at higher temperatures led to precocious nuclear accumulation of all three binding partners and growing worms at lower temperatures led to delayed accumulation [20].” This sentence is unclear: high temperature decreases, while low temperature increases the length of S- phase. Moreover, the authors conclude the opposite from what is written in the cited paper that states “The cell cycle at 15°C is approximately twice as long as that at 25°C,…. We observed that, at 15°C, MET-2::GFP was already detectable in nuclei by the two- to four-cell stage (Fig. S5A). At 25°C, there was not much enrichment for nuclear MET-2::GFP at the two- to four-cell stage, but it started to accumulate in the nucleus at the eight-cell stage.” Thus, it is not true that at higher temperatures there is a precocious accumulation of MET-2 in the nucleus.

-        Lines 159-173. It is perhaps worth mentioning that C. elegans has eight linker histone variants (HIS-24 and HIL-1 to HIL-7) and mention whether anything is known or not on the roles of HIL-1 to HIL-7 in regulating chromatin.

-        Lines 222-225: “Activating and repressive histone marks tend to be enriched on mutually exclusive genomic regions….”.  I think here the authors need to be careful with what they state. While it is true across species that H3K36me3 and H3K27me3 are mutually exclusive on genomic regions, this is not the case for H3K4me3 and H3K27me3, which instead constitute so-called bivalent domains which are particularly relevant in pluripotent cells. See review: DOI:https://doi.org/10.1016/j.tig.2019.11.004. I think the authors need to decide whether they are discussing mutual exclusiveness at the level of genomic regions or of the same histone tail/nucleosome.

-        In table 1, the use of the word “activating” to define active chromatin, or euchromatin, marks can be misleading in my opinion. The direct role for these marks in activating transcription is not clear yet. And actually, in the case of the H3K36me3 mark deposited by MES-4, which the authors have in the table, it was actually proven to be present in absence of ongoing transcription. See reference 42 in the reference list (10.1371/journal.pgen.1001091). Furthermore, concerning H3K36me3, I think the authors should add MET-1, which is a major H3K36me3 depositing enzyme (See Extended data figure in Cabianca 2019, their reference number 17). In particular, it was shown that it is MET-1 the HMT depositing H3K36me3 with elongating Pol II (again reference 42 in the reference list).

-        Line 275: “LADs are tethered to the inner nuclear lamina through their interaction with LEM-2, an inner membrane transmembrane protein [55].” This sentence is not correct. LADs are tethered to the nuclear lamina, in C. elegans, via H3K9 methylation being recognized by the protein CEC-4 (as cited by the authors in ref. 54 from Gonzalez-Sandoval 2015). In fact, in Gonzalez-Sandoval et al., heterochromatin detaches from the nuclear periphery in embryos lacking CEC-4, despite the presence of LEM-2 that was used to perform the ChIPseq. LEM-2 is a nuclear lamina protein that, similarly to LMN-1 (Towbin 2012) and EMR-1 (doi: 10.1186/gb-2014-15-2-r21 and ref 17, Cabianca 2019), can be used to determine which genomic sequences interact with the nuclear lamina by proximity to these proteins. However, to the best of my knowledge, no data in C. elegans shows any direct role for LEM-2 in actively tethering heterochromatin to the nuclear envelope.

-        Lines 283-294. I think here it is fundamental to make a clear distinction between heterochromatin arrays and endogenous heterochromatin. While SET-25 and MET-2 are redundant in anchoring an array (refs 16 and 18), in ref 18, Delaney 2019 clearly shows in Figure 7 that met-2 mutants strongly detach ENDOGENOUS heterochromatin from the nuclear periphery, indicating that the presence of SET-25 cannot compensate. The difference between array and endogenous chromatin most likely lies in the fact that SET-25 requires MET-2 to interact with most endogenous sequences but binds the heterochromatic array independently of MET-2 (as shown in ref 21 in this review doi:10.1101/gad.344234.120).

-        Lines 295-296: similar to what said above, when referring to a heterochromatic array the word LAD should not be used as this typically refers to endogenous chromatin. Cec-4 mutants detach both endogenous heterochromatin and heterochromatic arrays in embryos (Gonzalez-Sandoval), but this is no longer the case in L1s for an array (Gonzalez-sandoval 2015, Cabianca 2019) while it does have a partial effect on endogenous chromatin, although the effect is stronger in double cec-4 mrg-1 mutants (Cabianca 2019).

-        Lines 306-308: currently, to the best of my knowledge, there is no evidence that CBP/p300 forms a complex with MRG-1 or any of its orthologs in metazoans, and indeed I could not find data supporting this statement in the two cited papers. This sentence should be eliminated or heavily revised.

-        Lines 313-314: No ChIP data in the study mentioned shows that H3K9me3 is affected in absence of MRG-1. While this is a possibility, it was not tested and hence the sentence should be revised.

-        Lines 315-317: “MET-1 and MES-4-mediated deposition of H3K36me2/me3 marks on euchromatin is also required for proper tethering of heterochromatin through an unknown mechanism [17].” While there are certainly many aspects that remain undefined, the paper cited (Cabianca 2019) links the effect of H3K36me2/3 by MES-4 and MET-1 to MRG-1, which has a chromobarrel domain capable of binding this mark. Therefore MES-4 and MET-1 are likely acting on the MRG-1 pathway, possibly upstream of it, similarly to what the authors mention in lines 305-306.

-        Lines 349-350: “global TAD 349 distribution through development.” Can the authors be more specific and state when TADs are formed in embryonic development? Is it known?

-        Lines 387-390: While a heterochromatic array was shown to be decompacted in cec-4 mutants in Gonzalez-Sandoval (ref 54) and cabianca et al, no data in these cited article shows anything on the compaction of chromosomes. Thus, the sentence needs to be rephrased or the citations corrected.

Minor points

-        When the authors refer to germline, it might be useful to highlight the cases in which it is known whether the contribution is maternal or paternal. For example, see this paper from the Strome lab https://doi.org/10.1038/s41467-018-06236-8

-        Having a dedicated paragraph on histone variants would help the clarity of the review, rather than having them scattered throughout. Can be avoided when histone variants are included in a general figure of chromatin organization changes during development, as mentioned in the major points section.

-        Lines 370-374 describe an approach of chromosome tracing that was developed in the Mango lab, as correctly cited in their ref 15. However, it is confusing that they now put ref 70, which is the HiC, downstream of the sentence discussing chromosome tracing.

Unclear/not precise sentences:

-        Lines 26-28 “As embryogenesis progresses, the chromatin architecture begins transforming into two distinct states: the actively transcribed euchromatin and the transcriptionally silent heterochromatin”. It is not chromatin architecture, but rather chromatin, that segregates into euchromatin and heterochromatin according to the acquisition of certain features. I also find the verb transform into not very accurate, maybe the authors can rephrase.

-        Lines 34 “Two other mechanisms are the establishment of topologically associated domains (TADs) and lamina-associating domains (LADs)….”. Two other mechanisms of what? To start a paragraph like this is confusing. Do the authors refer to two additional layers of chromatin organization?

-        Line 47: Equivalize is a verb I couldn’t find in the dictionary.

-        Line 52: the second of three “the” is probably a “that”.

-        Lines 57-58: it is unclear why a random sentence starting from “and heterochromatin….. chromatin. As zygotic” is in bold.

-        Line 219: unclear what “a host of developmental defects” means.

-        Line 293: it is LIN-65 and not LIN-61 the cofactor of MET-2

-        Line 311: the array used in the cited study is not extrachromosomal. It is integrated. Please remove extrachromosomal.

Some sentences are convoluted other contain typos. They were all highlighted in the minor comments section.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

In their review article untitled “Chromatin Organization during C. elegans Early Development”, the authors cover the formation of key features of the chromatin architecture and its organization (TADs, LADs, heterochromatin and A/B compartments) during C. elegans early development.

The review explores this topic in some detail by presenting first the heterochromatin formation, second the euchromatin, then the antagonism between both of them, their spatial nuclear localization and structural features (TADs), to finish by presenting the situation upon loss of developmental plasticity and the establishment of dosage compensation on the X chromosome.

Overall, this is an interesting, detailed and timely review touching topics of broad interest for the reader (chromatin organization, epigenetic, embryo development). The review is well structured and balanced and the different parts of the text are written in a clear manner. However, several parts of the review, particularly the description of the heterochromatin formation (paragraph 2) and the spatial organization (paragraph 6) could be improved by being illustrated with some figures and or schematic cartoons. In addition several points require additional informations. Below are several concerns that should be addressed, as follows:

Major comments:

1) As general comments:

a) In the whole review (text and tables), the authors wrote “activating and inactivating” related to histone marks. It would be more correct to replace it by “active “ and “ repressive” or “inactive” instead of.

b) Similarly in the whole review, the authors defined TAD as “topologically associated domains”, however the correct name is "Topologically associating domains".

2) a) line 38-39, the authors mention :” DNA elements within a self-interacting domain

share more interactions inside their domain compared to interactions with DNA elements

outside their domains”. Indeed, this is true but it would be important in addition to mention that these interactions are highly stochastic and occur at low frequencies (see references Cattoni et al.,2017; Flyamer et al., 2017; Nagano et al., 2017; Stevens et al.,2017).

b) line 38, I propose to replace “DNA elements” by “chromosomal contacts”.

3) As an important point, I suggest to the authors adding a figure and / or a cartoon summarizing the establishment of heterochromatin during early development in C.elegans. Doing that will definitively help the reader and improve the quality of the review. This cartoon could illustrate the paragraphs 2-3-4. Authors could take example from the Figure 2 of the review https://doi.org/10.1111/nyas.14303. (Of course replacing Drosophila development by early stage C. elegans embryo development, then describing the different levels of active and repressive histone modifications and their respective enzymes. Finally, below replacing the TADs /LADs by EDRs formation, localization, etc.)

4) line 245-247: the authors should explain more what could be the hypothetical mechanism that might explain that MET-2 nuclear foci itself are excluding the active histone marks at mes-2 targeted loci ? Are they thinking about LLPS process, or something else ?

5) To help the reader and improve the clarity of the review, I suggest to the authors adding a figure and / or a cartoon illustrating the LADs formation in nematodes by mentioning the different factors involved (histone modifications, enzymes triggering the modifications, readers, etc). This would be greatly appreciated.

6) line 274: I think it is also important to mention that LADs are also enriched in H3K9me2. H3K9me2-marked heterochromatin localization to the nuclear periphery is evolutionary conserved from C. elegans to humans (see ref https://doi.org/10.7554/eLife.49278).

7) In the paragraph 7 “Emergence of topologically associated domains (TADs)”, I suggest that it is important to tell that although the autosomal genome is not segmented by TADs, there are anyway more than 30 000 sequences with characteristic enhancer chromatin features identified. In this context, a recent study (Biorxiv https://doi.org/10.1101/2023.07.14.549011) described that enhancer loci correlate with 3D hairpin-like structures extending 10-50 kb from the enhancers and named fountains. In this paper, the authors suggest that fountains could play a similar role as TADs and direct enhancer-promoter interactions.

8) line 338-341: the authors wrote :”TADs are composed of chromatin domains that have a high frequency of interaction within the DNA sequences inside the domain and have relatively low interaction frequencies with DNA sequences in other domains.“

I do not completely agree with this statement of "high frequency". It has been shown in Drosophila that the association frequencies within TADs and between TAD borders are below ~10%, independently of TAD size, epigenetic state, or cell type. TAD structure being conserved this observation might be similar in C. elegans. Furthermore, the confinement of chromatin into TADs may require only small differences in absolute contact probabilities (~2-fold).

What is important to highlight is that TAD are organized by multiple, small-frequency, but yet specific interactions that are regulated by epigenetics and transcriptional state.

Minor comments:

1) line 51: In the sentence “mechanisms the regulate”, please correct “the” by “that”.

2) line 159: Please correct the capital letter in “Linker”.

3) line 186: Replace “at the level of microscopy” by “at the microscopic level.

4) line 191: Could you please define what do you mean by AB descendants or AB lineage.

5) line 212: Remove “in” in the following sentence “histone marks in during embryogenesis”.

6) line 352: Perhaps it is the word “although” instead of “through”.

7) line 420: Replace “SET1DB” by “SETDB1”.

The text review is written in a clear manner and the quality of the English is good.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This review article summarizes a wealth of published findings on chromatin organization in early development of the nematode C. elegans. The text is well-written, and the summary tables are a nice compilation of the many marks of histone modification that occur and their functions. I do not have the necessary breadth of knowledge in this topic to be assured of its complete accuracy, however the detail in which the paper citations and evidence are presented (and the remaining editorial marks from Track Changes) give me confidence that the authors have checked the text thoroughly for accuracy.

This is a very nice review and very thorough and will be very helpful for the research community. I have only a few suggestions that would make this already through review perhaps a bit better.

1. Can the last paragraph be offset in its own paragraph that is titled ‘Future Work’ or similar? Although there are places in the text where there are mechanisms or phenotypes that are not yet fully described, it would be nice to have the authors present their take on where the next 5-10 years of research will go.

2. C. elegans is of course but one species within a genus with over 50 known species. Are there any studies of chromatin organization in any other nematodes? For example, some studies in dosage compensation have been done in other species by the Meyer lab. Are there any insights into flexibility in dosage compensation mechanisms?

3. In the same evolutionary theme, is there a way (perhaps in a table) to relate the findings in C. elegans to other species, in particular invertebrates or mammals? At least in broad themes to know whether there are some features that are widely conserved and some that may be novel or found only in nematodes.

4. The article lacks any images demonstrating the type of beautiful cell biology that can be done with the nearly transparent C. elegans embryos, and it also lacks any diagrams. Is there a way to provide an image or two of a C. elegans embryo showing the localization of chromatin domains in situ by confocal microscopy or similar? And is there a way to have a diagram showing a temporal progression of chromatin marks as the embryo moves from the zygote through to initiation of zygotic gene expression by the 4-cell stage or later?

5. Watch where the tables start and make sure these don’t have clumsy breaks like the start of table 2, which breaks to another page just after its column headings (at least in the draft sent for review).

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

This study comprehensively reviewed dynamic chromatin organization during C. elegans embryogenesis. The authors meticulously summarized the intricate interactions between histone modifications and molecular mechanisms that regulate the formation of chromatin architecture, heterochromatin and euchromatin, the chromatin states associated with transcription activity, providing significant insights into future studies. I fully recommend the publication of this paper in DNA. Please see two suggestions for a minor revision below.

 1, Change “the mechanisms the regulate” to “the mechanisms that regulate” at line 52.

2, Remove the duplicate “in” in “repressive histone marks in during embryogenesis” at line 212

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I thank the authors for addressing the points I raised before. I find the manuscript much better.
I am left with few minor comments:

- Lines 321 Heterochromatic arrays require both H3K9me2 and me3 to be peripheral and not only H3K9me2. In fact, both SET-25 and MET-2 need to be mutated to see detachment. And this conditions depletes worms from all H3K9me (mass spec measurements of histone PTMs by Towbin et al 2012).

-Lines 331: I would like to point out that the "conflicting reports" mentioned by the authors are solved  by the fact that SET-25 was shown to methylate some targets independently of MET-2, and the heterochromatin arrays are among these (PMID: 33303642), therefore H3K9me at the arrays still happens in absence of MET-2, due to the action of SET-25. On the contrary, at the majority of endogenous sequences, with few exceptions (PMID: 33303642), SET-25 requires MET-2 for binding. Thus, loss of MET-2 leads to the strong detachment observed on endogenous sequences by Delaney et al.

Line 349: in the paper cited ChIP-qPCR at selected genes was performed, not ChIPseq.

Lines 482-484: Only a percentage of cec-4 mutants continue to differentiate. This has to be clarified as, as it is written now, the sentence gives the wrong impression that all embryos don´t arrest, which is not correct. Moreover, some cec-4 embryos develop into L1-like larvae. This also should be clarified in the text as these larvae die and are not like normal L1s.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

I would like to congratulate the authors who have largely modified and improved the review "Chromatin Organization during C. elegans Early Development".I'm fully satisfied by the author's answers and by the novel informations (new Figures and paragraphs) added in this revised version.

I support that this review can be accepted  in the present form.

Author Response

Thank you, your comments were very helpful in the first round of revisions.