Polycomb Repressive Complex 2 in Eukaryotes—An Evolutionary Perspective
Alaguraj Veluchamy
Round 1
Reviewer 1 Report
Polycomb repressive complexes are histone modifying complexes in which multiple components of these complexes are evolutionarily conserved across multiple species.
In this review article, the authors have covered extensively on evolutionary aspects of PRC2. This paper was a joy to read. Coherent review of PRC2 in different taxonomic groups are provided.
- Keeping in line with the title of this article, authors could include phylogenetic analysis/tree on PCR2 components, either an aggregated tree on all PRC2 components or separate phylogenetic/evolutionary tree on individual components could be included.
- Structural aspects of these PRC components are not included in “2. Features of PRC2 core composition, structure and function”.
- Authors could reduce the citation of large number of reviews and incorporate the original findings.
- Figure1, Table1 and Supplementary Table1 are essentially providing the same information. Redundancy could be avoided.
- In general, there is a particular focus on unicellular and multicellular green lineage than discussions on animal models such as drosophila, humans, C. elegans and mouse. For example, PRC2 is an emerging target for inhibition in case of certain cancer.
Author Response
Reviewer 1
Comment: Polycomb repressive complexes are histone modifying complexes in which multiple components of these complexes are evolutionarily conserved across multiple species.
In this review article, the authors have covered extensively on evolutionary aspects of PRC2. This paper was a joy to read. Coherent review of PRC2 in different taxonomic groups are provided.
Response: Thank you for your comments on our review. We attempted to resolve your concerns about the article one by one.
Comment1: Keeping in line with the title of this article, authors could include phylogenetic analysis/tree on PCR2 components, either an aggregated tree on all PRC2 components or separate phylogenetic/evolutionary tree on individual components could be included.
Response: We computed phylogenetic trees for described PRC2 subunits of model organisms, but we realised they do not provide sufficient resolution for making a conclusion beyond the known separation of eukaryotic clades or the known divergence between the different protein subunits. This we thought had limited added value. As existing articles that we cite (Huang et al (2017)- https://0-doi-org.brum.beds.ac.uk/10.1093/bfgp/elw007, Sharaf et al (2020) - https://www.biorxiv.org/content/10.1101/2021.07.16.452543v1) feature refined phylograms for PRC2 and PRC1 (Huang et al.(2019)://doi.org/10.1186/s12864-019-5905-9), we finally decided not to include the trees. Since a suggestion of including alignments to indicate the similarity of the different subunits was brought up by another reviewer, we instead indicate the level of identity/similarity of full-length protein sequences of described PRC2 subunit – this can be found in the modified Supplementary Table 1. (For your information, we also upload several slides with the alignments/identity matrices and NJ-trees - please see attachment.)
Comment 2: Structural aspects of these PRC components are not included in “2. Features of PRC2 core composition, structure and function”.
Response: Thank you for pointing out this mistake. In the final version before first submission, we removed structural information and incorporated the relevant information in the Figure 2-caption to minimize the text and improve the flow, but we forgot to modify the title accordingly. Now this heading is modified to ’Features of PRC2 core composition and function’ – line 81.
Comment 3: Authors could reduce the citation of large number of reviews and incorporate the original findings.
Response: Due to the breadth of the topic, we struggled to incorporate references to as many original studies as possible, while keeping the refence list within reasonable limits (still we feel it is very long). Original articles were cited for crucial particular findings, especially wherever we felt that the statement directly links to the major focus of the article. In cases where more information was needed for support or where we refer to an opinion, recent reviews from different research groups are cited to maintain balance/acknowledge existing work.
Comment 4: Figure1, Table1 and Supplementary Table1 are essentially providing the same information. Redundancy could be avoided.
Response: In Figure 1, we aim to illustrate the increasing diversity of PRC2 components, in line with increasing body plan complexity. It contains information on PRC2 in better-studied model species. Table 1 provides more complete comparative information including all eukaryotic supergroups and species where PRC2 has been studied. Supplementary table 1 is an extension of Table 1 that we though could be useful for researchers in the PRC2-field but not of much relevance for general readership. In Supplementary Table 1, we newly included the information of %identity/similarity of the listed subunits compared to the PRC2 subunits of D. melanogaster. In this light, we do not feel that the three components are redundant.
Comment 5: In general, there is a particular focus on unicellular and multicellular green lineage than discussions on animal models such as drosophila, humans, C. elegans and mouse. For example, PRC2 is an emerging target for inhibition in case of certain cancer.
Response: We agree with this comment and realise that our review does not cover the topic completely. Our main goal was to focus on the comparison between PRC2 composition and function in different eukaryotic groups, especially focusing on the green lineage and unicellular species. For this reason, we decided to limit the topics that we included to (i) information that was fundamental for understanding the concepts and (ii) information where comparative statements could be made. Although we realise that the function of PRC2 in cancer formation is a major topic of study and interest, and that its inhibition offers potent emerging tool for cancer treatment, the possibilities for comparative discussion in clades outside animals are limited. Excellent reviews on this topic are available that provide information, detail and opinion that greatly exceed what we could cover in our review.
We agree however that the reference to this major topic of PRC2 research is missing, and for this reason, we added the following:
Line46-49: “In line with its function in cell identity maintenance in animals, PRC2 dysfunction is frequently associated with cancer development and PRC2 is a potent target for anticancer therapy (reviewed in [20–22]).”
- Wang, S.; C. Ordonez-Rubiano, S.; Dhiman, A.; Jiao, G.; Strohmier, B.P.; Krusemark, C.J.; Dykhuizen, E.C. Polycomb Group Proteins in Cancer: Multifaceted Functions and Strategies for Modulation. NAR Cancer 2021, 3, zcab039, doi:10.1093/narcan/zcab039.
- Dockerill, M.; Gregson, C.; O’ Donovan, D.H. Targeting PRC2 for the Treatment of Cancer: An Updated Patent Review (2016 - 2020). Expert opinion on therapeutic patents 2021, 31, 119–135, doi:10.1080/13543776.2021.1841167
- Piunti, A.; Shilatifard, A. The Roles of Polycomb Repressive Complexes in Mammalian Development and Cancer. Nature Reviews Molecular Cell Biology 2021, 22, 326–345, doi:10.1038/s41580-021-00341-1.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
I consider the review is well written and introduces the topic properly.
However, I would like to suggest a few points to improve the final manuscript:
* New text on enhancers could enrich the article, which is now intensively focused on promoters:
- PRC1 Ring1b is known to participate in active enhancers (Nat Commun. 2018 Aug 23;9(1):3377. doi: 10.1038/s41467-018-05728-x. PMID: 30139998)
- PRC2, on the contrary, seems to be associated to poised enhancers (Nat Commun. 2021 Jul 16;12(1):4344. doi: 10.1038/s41467-021-24641-4. PMID: 34272393)
* Figure1/2 can be complemented by showing (in a new figure) the multiple sequence alignment of some of these
PRC2 core components (and also selected accessory factors) to highlight the degree of conservation at amino
acid level
* A table with all the genome-wide experiments described along the review would be very useful for readers
(including GEO codes of each sample)
* Also a new figure to summarize differences in the H3K27me3 patterns of ChIPseq occupancy (promoter, telomer, etc.)
accross different species/clades would be interesting to discuss whether they are also conserved or not
* The known dual relationship with H3K36me3 could be also discussed (Nat Struct Mol Biol. 2012 Dec;19(12):1257-65. doi: 10.1038/nsmb.2434. Epub 2012 Oct 28. PMID: 23104054)
- Minor: Figure 1 - "Accessary units" should be "Accessory units"?
Author Response
Reviewer 2
I consider the review is well written and introduces the topic properly.
However, I would like to suggest a few points to improve the final manuscript:
Response: Thank you for taking the time to comment on our review. We tried to address each of your comments separately.
Comment 1: New text on enhancers could enrich the article, which is now intensively focused on promoters: - PRC1 Ring1b is known to participate in active enhancers (Nat Commun. 2018 Aug 23;9(1):3377. doi: 10.1038/s41467-018-05728-x.PMID:30139998)
- PRC2, on the contrary, seems to be associated to poised enhancers (Nat Commun. 2021 Jul 16;12(1):4344. doi: 10.1038/s41467-021-24641-4. PMID: 34272393).
Response: Thank you for this relevant point, we have mentioned PRC involvement at enhancers on lines 165-168: “In addition to genic loci, H3K27me3 and PRC2 reside in poised enhancers (PEs) that often associate with bivalent genes in vertebrate pluripotent cells [152]. While PRC1 contributes to the PE marking globally [152] and also targets active enhancers in cancer cells [153], PRC2 is involved at PEs at specific loci [152]. “
Comment 2: Figure1/2 can be complemented by showing (in a new figure) the multiple sequence alignment of some of these PRC2 core components (and also selected accessory factors) to highlight the degree of conservation at amino acid level
Response: We have carried out the multiple sequence alignments using subunits of the different model species. We are not including the alignments for spatial reasons (please see attachment), but we newly indicate the level of amino acid sequence identity/similarity to full-length D. melanogaster subunits in modified Supplementary Table 1. It should be noted, however, that E(z) and Su(z)12 homologs are primarily conserved in the SET and VEFS domain regions, respectively.
Comment 3: A table with all the genome-wide experiments described along the review would be very useful for readers. (including GEO codes of each sample)
Response: We have considered your suggestion carefully. Finally, however, we decided not to include the GEO accession numbers. The reason was the large number of experiments associated with H3K27me3 in the different species, and we think it is not possible to incorporate them all. We try to refer to the seminal works and presume that the reader (or potential data user) would anyways first consult the source literature before using the dataset.
Comment 4: Also a new figure to summarize differences in the H3K27me3 patterns of ChIPseq occupancy (promoter, telomer, etc.) across different species/clades would be interesting to discuss whether they are also conserved or not.
Response: To address this comment, we mapped existing publicly available data generated by the articles cited onto the respective genome assemblies to create an additional Figure 3, depicting chromosome-wide distribution of H3K27me3 in the different described model species. We felt that a simple schematic that would separate the chromosome into the subtelomeric and telomeric regions may raise impression of H3K27me3 location to telomeric repeats per-se, which is not likely to be identified by mapping NGS reads to existing genome assemblies. Location to telomeric repeats is discussed in the review text on lines 626-656.
Comment 5: The known dual relationship with H3K36me3 could be also discussed (Nat Struct Mol Biol. 2012 Dec;19(12):1257-65. doi: 10.1038/nsmb.2434. Epub 2012 Oct 28. PMID: 23104054).
Response: Thank you for this comment. We have added the following sentence on lines 249-252: “In addition to opposing PRC2, H3K36me3 may promote H3K27me3. In mouse embryonic stem cells, Phf19, a PCL ortholog, binds to H3K36me2/H3K36me3, recruiting PRC2 and lysine demethylases to promote PRC2 activity [202].”
Comment 6: Minor: Figure 1 - "Accessary units" should be "Accessory units"?
Response: Thank you for pointing this out – we have corrected the spelling error in Figure 1 to “Accessory units”.
Author Response File: Author Response.pdf
Reviewer 3 Report
The importance of Polycomb regulation in animals and plants is clear from the recent explosion of papers on this topic. This manuscript is therefore a useful compilation of Polycomb proteins throughout the tree of life and the implications those evolutionary relationships have on function.
There is an extraordinary amount of information presented and the figures and tables are generally very helpful to the reader.
There are small mistakes throughout so the manuscript needs diligent checking eg. Physcomitrium patens should be Physcomitrella patens, Lycopidiophytes should be Lycopodiophytes
Figure 1 accessary =accessory
Line 161 in plants H3K27me1 is catayzed by.... ATXR5 and ATXR6. This has only been established for the pericentromeric heterochromatin, not the whole genome.
Line 477/478 says “VRN2 homologs have only been found in dicots” whereas in line 485ff “the studied VRN2 homologs across angiosperms including monocots”.
“MC-dipeptide is present in EMF2 N-terminal region in multiple angiosperms” from line 487 could be misunderstood as EMF2 being subject to N-end rule degradation, which is not the case.
Author Response
Reviewer 3
Comment: The importance of Polycomb regulation in animals and plants is clear from the recent explosion of papers on this topic. This manuscript is therefore a useful compilation of Polycomb proteins throughout the tree of life and the implications those evolutionary relationships have on function.
There is an extraordinary amount of information presented and the figures and tables are generally very helpful to the reader.
Response: Thank you for the feedback to our manuscript. Below, we tried to respond to each of your comments.
Comment 1: There are small mistakes throughout, so the manuscript needs diligent checking eg. Physcomitrium patens should be Physcomitrella patens, Lycopidiophytes should be Lycopodiophytes
Response: The moss P. patens has been suggested to be re-named to Physcomitrium patens instead of Physcomitrella patens based on refined phylogenetic position: Rensing et al. (2020) (https://0-academic-oup-com.brum.beds.ac.uk/plcell/article/32/5/1361/6115584). According to the current NCBI taxonomy classification (https://0-www-ncbi-nlm-nih-gov.brum.beds.ac.uk/Taxonomy/Browser/wwwtax.cgi?id=3218), Physcomitrium patens can be used, hence we opted to use the updated name.
Additionally, thank you for pointing out our mistake - we corrected the spelling error in Lycopidiophytes to “Lycopodiophytes” (line 489).
Comment 2: Figure 1 accessary =accessory
Response: Thank you for pointing out the error, we have corrected the spelling error in Figure 1 from “accessary” to “accessory”.
Comment 3: Line 161 in plants H3K27me1 is catayzed by.... ATXR5 and ATXR6. This has only been established for the pericentromeric heterochromatin, not the whole genome.
Response: Thank you for this comment, it is a relevant point and we have modified lines 158-160 in the following way: “In animals, PRC2 catalyzes H3K27me1/2/3 [50] while in plants, H3K27me1 in centromeres and pericentromeres is catalyzed by the ARABIDOPSIS TRITHORAX-RELATED HMTs (ATXR5 and ATXR6) [150].”
Comment 4: Line 477/478 says “VRN2 homologs have only been found in dicots” whereas in line 485ff “the studied VRN2 homologs across angiosperms including monocots”.
Response: Thank you for pointing this out - it is our mistake - VRN2 is present both in monocots and dicots, but not in non-vascular plants - we have corrected it on line 500.
Comment 5: “MC-dipeptide is present in EMF2 N-terminal region in multiple angiosperms” from line 487 could be misunderstood as EMF2 being subject to N-end rule degradation, which is not the case.
Response: We agree that the sentence was misleading, thank you for pointing it out. We rephrased the sentence on lines 512-515: ‘In multiple angiosperms, including Amborella trichopoda, but not basal land plants (lycophytes and bryophytes), EMF2 contains internally-located MC-dipeptide not subject to N-end rule degradation, supporting VRN2 origin by EMF2 duplication and N-terminal truncation [263].’
