Nanog organizes transcription bodies

Version published to 10.1016/j.cub.2022.11.015

Jan 1, 2023

Review coordinated via ASAPbio’s crowd preprint review

This review reflects comments and contributions by Richa Arya, Samuel Lord, Arthur Molines and Sónia Gomes Pereira. Review synthesized by Richa Arya.

In the manuscript titled “Nanog organizes transcription bodies” Kuznetsova et al. discuss how the transcriptional bodies are assembled. They show that Nanog and Sox19b cluster before the transcription actually starts and initiate the formation of transcription bodies.

The following comments and suggestions were raised to help strengthen the manuscript:

Result: RNA Pol II transcription localizes to two transcription bodies when zebrafish genome activates…

The heading of the section is a bit confusing. Does it refer to the transcription of RNA pol II or transcripts of RNA pol II?

Within the result section: ‘…as 64-cell stage…’ Would be good for the reader to clarify that this is division number 6 to make it clearer that it is way earlier than what is reported in the previous sentence.

‘...productive transcription starts in two transcription bodies in the nucleus. These transcription bodies are isolated, large, long-lived, and appear at a predictable time during development…’. At this stage the study has reported localization data but not activity data (this is included later). The formation of clusters (which is what is detected) might suggest but cannot conclude about the activity of the enzyme or whether RNA is actually being produced.

Figure 1

‘…the percentage of nuclei with at least one Pol II (Ser5P or Ser2P) cluster is indicated…’. The injection is happening at the 1-cell stage. Then observations are made at the 64/128/256 cell stage. Are all the nuclei labelled at these stages? or only a subset? Recommend providing some clarification about the percentages reported? Are they a consequence of the embryo being mosaic, some cells containing the label injected at the 1-cell stage and some not? Or is it biological noise? Or a combination of both? Is this the ratio of (# of cells with puncta)/(total # of cells) or is it (# of cells with puncta)/(# of cells containing some labelled Pol II)?

'C. Tracks of transcription bodies at 64-, 128-, and 256-cell stage. The presence of Pol II Ser5P, Ser2P, or both, is indicated by red, blue, and white circles, respectively. Time on the x-axis in minutes after mitosis.’. The sample size seems too small, can some clarification be provided to help with the interpretation of this data:

In the first plot, one embryo is 64 cells. Even taking in consideration the fact that the embryo might be a mosaic with 50% of the cells labelled and that one can not image all of the cells due to the thickness of the sample, it should leave a few cells imaged per embryo (5-10 cells). It would be good if one experimental replication was made of multiple embryos injected in parallel. So, with all these considerations, the 20-ish tracks displayed on the first plots seem like a small number. If one experimental replicate is 3-4 embryos and 5 cells can be imaged per embryo then around 20 tracks would be the result of 1 replicate (vs 3 as indicated in the methods). If 10% of the cells can be imaged at 256-cell stage, with 3 replicates each made of multiple embryos, it would give more than 60-ish tracks.

‘For wt, N=3, n=111; for mir430 mutant N=3, n=72…’. Please clarify what the two n refer to in the figure legend?

The methods state "A minimum of 3 biological and 3 technical replicates was generated for each experiment. The number of experimental replicates (N) as well as the number of measured nuclei (n) are reported for each conducted experiment individually in the respective figure legend." - Recommend including a shorter but similar clarification in the figure legend.

Result: Transcription factors cluster prior to, and independent of transcription

'and visualized each transcription factor in combination with the initiating form of RNA Pol II (Figure 2A, and Movies S4-6…’ Suggest adding a clarification about when zygotic translation starts in zebrafish and whether translation starts before transcription in zygotes.

‘We conclude that transcription factors cluster prior to, and independent of transcription elongation….’ From the data it should be possible to estimate a mean delta T from TF clustering to Pol II clustering, it may be relevant to report such a number.

Figure 2: Pairwise non-parametric Wilcoxon tests: There is a concern about the use of a pairwise test as the two conditions CTRL and amanitin are two different conditions.
Result: RNA accumulation results in dissociation of transcription factor clusters

‘... the appearance of RNA Pol II Ser5P (initiation) clusters was also delayed in the absence of transcription elongation (Figure 2E)…’. Suggest calculating the delta in time between TF cluster appearance and Pol II cluster (as suggested above). It appears the "delay" in the apparition of the Pol II puncta is the delay observed for the TF, which would indicate that with or without transcription Pol II joins the TF cluster at the same time.

‘…while accumulation of RNA causes them to dissolve…’. Is this based on the observation that inhibition of transcription results in a longer cluster lifetime? RNA accumulation might promote clusters to dissolve, but whether it is the "cause" of their dissolution has not been tested. Recommend reframing the fragment to avoid conclusions about RNA accumulation.

Result: Nanog organizes transcription bodies

'…cycle (Figure 3A)…'. There is a concern about comparing Nanog and Sox apparition time if they are not observed within the same embryo / nuclei. The present data are convolved by variations between embryos and between nuclei, recommend providing some clarification and looking at the time difference between each TF and the corresponding Pol II cluster.

‘...Nanog RNA Pol II Ser5P could still be detected…’. Suggest re-phrasing this part as "to determine if RNA Pol II Ser5P could still be detected in the absence of Nanog".

Figure 3.

‘In C-D, the percentage of nuclei with the indicated pattern is indicated…’. Suggest some further clarification about the percentages reported. In C does this indicate that 9 % of the cells form Sox clusters in absence of Nanog? And in D that 27 % of cells form Pol II clusters in absence of Nanog? If that is the case, recommend discussing it as it might impact the conclusion that Nanog is "required" for Pol II clustering.

'Pairwise non-parametric Wilcoxon tests were performed, ns indicates P > 0.05…'. Reconsider the use of pairwise tests, as noted above.

Figure 4 - ‘percentages indicate how often the shown phenotype is observed. For D and E, N ≥ 3 and n ≥ 18.’ - Please clarify how these percentages are calculated. Is this the percentage of nuclei with the described phenotype per embryo? Or the percentage of embryos with at least one nucleus with the depicted phenotype? In Figure 1 the percentage for Pol II in WT at 128-cell stage is 80%. Figure 4 reports 100%, is it evaluating the same thing? If it is preferable not to write exactly N and n values for all the conditions, maybe these could be shown in the figure itself.
Result: Nanog DBD as well as IDR are required to organize transcription bodies

‘In this study we analyzed the assembly of two transcription bodies…’. Recommend placing this under a separate Discussion/conclusions section.

‘…and RNA accumulates, transcription factor clusters disassemble.’ It is not clear that the statement is supported by the data, consider reframing the fragment.

STAR METHODS

Please provide additional details about the different aspects of methodology. Also consider depositing the custom scripts to a public platform such as github or zenodo where these materials can be publicly accessed and referenced, supporting reproducibility.

‘Preparation of embryos for use in live-cell microscopy….At 16- to 32-cell stage…’. In the movies (or at least their legends), the embryos shown are at the latter cell stages. Would it be possible to clarify whether later staged embryos were prepared and how? If the approach involved waiting until the desired development stage was achieved, please indicate so.

‘Image pre-processing with Noise2Void… The network was trained on and applied to the raw spinning disk confocal data in full 3D with both color channels being present…’. Are there specific parameters that should be specified? How many stacks / movies were used for training? How was it evaluated that the training was sufficient?

‘Signal normalization…..The denoised and max-projected 2D image data was normalized…’. Please report the details of the process e.g what was the normalization?

‘Determining developmental stage and mid-point between interphases……This method is very reliable as the inter-nuclear distances in these early stages are highly stereotypic…’. Was this method previously described and/or used? If so, please provide references.

Read the original source

Nanog organizes transcription bodies

This article has been Reviewed by the following groups

Listed in

Abstract

Article activity feed

Multifaceted roles of cohesin in regulating transcriptional loops

Chromatin protein complexes involved in gene repression in lamina-associated domains

A phosphorylation code coordinating transcription condensate dynamics with DNA replication