Her6 and Prox1 are novel regulators of photoreceptor regeneration in the zebrafish retina
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Review Commons)
Abstract
Damage to light-sensing photoreceptors (PRs) occurs in highly prevalent retinal diseases. As humans cannot regenerate new PRs, these diseases often lead to irreversible blindness. Intriguingly, animals, such as the zebrafish, have the ability to regenerate PRs efficiently and restore functional vision. Upon injury, mature Müller glia (MG) undergo reprogramming to adopt a stem cell-like state. This process is similar to cellular dedifferentiation, and results in the generation of progenitor cells, which, in turn, proliferate and differentiate to replace lost retinal neurons. In this study, we tested whether factors involved in dedifferentiation of Drosophila CNS are implicated in the regenerative response in the zebrafish retina. We found that hairy-related 6 ( her6 ) negatively regulates of PR production by regulating the rate of cell divisions in the MG-derived progenitors. prospero homeobox 1 ( prox1 ) is expressed in differentiated PRs, and likely promotes PR differentiation through phase separation. Interestingly, upon Her6 downregulation, Prox1 is precociously upregulated in the PRs, to promote PR differentiation; conversely, loss of Prox1 also induces a downregulation of Her6. Together, we identified two novel candidates of PR regeneration that cross regulate each other, and may be exploited to promote human retinal regeneration and vision recovery.
Article activity feed
-
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
The authors do not wish to provide a response at this time.
-
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
In this manuscript, Veen and colleagues investigate a role for hairy-related 6 (her6) and prospero homeobox 1 (prox1) in the regeneration of photoreceptors in the zebrafish retina. The candidate genes, her6 and prox1 were chosen for investigation based on initial analyses of genes implicated in the dedifferentiation of cells in the Drosophila brain. Using a chemogenetic approach to ablate long-wavelength cone photoreceptors (red cones), the authors investigate the role of her6 and prox1 in mediating the regenerative response of Müller glia and progenitors derived from these cells in the production of photoreceptors. They use …
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
In this manuscript, Veen and colleagues investigate a role for hairy-related 6 (her6) and prospero homeobox 1 (prox1) in the regeneration of photoreceptors in the zebrafish retina. The candidate genes, her6 and prox1 were chosen for investigation based on initial analyses of genes implicated in the dedifferentiation of cells in the Drosophila brain. Using a chemogenetic approach to ablate long-wavelength cone photoreceptors (red cones), the authors investigate the role of her6 and prox1 in mediating the regenerative response of Müller glia and progenitors derived from these cells in the production of photoreceptors. They use morpholinos to identify that her6 and prox1 likely cross regulate each other.
The authors state that the loss of Her6 leads to an increase in the number of photoreceptors after injury. However, the marker they use is a Zpr1 antibody, which marks both red and green cones but no other cone type nor rods. Thus what they show is an increase in red and green cones, not all photoreceptors. More markers would be needed to make that point. It also remains possible that the loss of Her6 leads to an increase in the number of overall cells generated after red cone ablation and thus Her6 is not exclusively linked with photoreceptor regeneration. Previous work has shown that while the regenerative response is biased toward the production of cell-types lost as a result of injury, other cell-types are also generated (Powell et al., 2016). It would be important to follow up whether other retinal cell-types are also generated. While her6 and prox1 in conjunction might regenerate (some?) photoreceptors, her6 alone might have a broader effect.
The segue between the first and second sections of the results felt abrupt to me. I wonder whether the authors could instead follow this sequence:
- Identifying candidate genes that regulate de-differentiation in the Drosophila. 2.Investigating whether the candidate genes play a role in the dedifferentiation that Müller cells must go through in their regenerative response.
- To test point 2, set up an ablation paradigm - MTZ mediated red cone ablation etc. etc.
The authors noted several genes as being mediators of dedifferentiation in the Drosophila. What was their reasoning for choosing the 2 they ended up pursuing in the zebrafish retina?
I do not understand why an increase in the number of PH3+ cells in Her6 MO treated retinae indicates a faster cell-cycle? Could it also mean that cells are stuck in the G2-M phase (marked by PH3) longer? There are other approaches to measure whether cell-cycle is sped up, as the authors suggest.
Minor suggestions to improve clarity:
Introduction: Noel et al., 2021, in my view, is not necessarily the reference most appropriate for a broad statement of how vision works. It appears to be specifically about zebrafish photoreceptors.
Methods:
Fish husbandry and set up
The authors state that the light/dark cycle in their fish facility is 12/12h. To my knowledge, this is not the standard practice; it is usually 14/10h light/dark respectively. (See Aleström et al., Laboratory Animals 2020, Vol. 54(3) 213-224).No references are given for the Tg(lws2:nfsb-mCherry) and the Tg (her4.1:dRFP) lines. Additionally, the reference for Tg(gfap:eGFP) is incorrect. This should be Bernardos and Raymond 2006 Gene Expr Patterns 2006 Oct;6(8):1007-13. doi: 10.1016/j.modgep.2006.04.006.
Immunostaining fish samples
'The slides were covered with Parafilm and tissue was dampened.....'
Here a reader may be confused by whether the authors refer to retinal tissue - presumably the authors mean paper tissue to create a humidified chamber.Microscopy and analysis
A bit more detail here would help - for example, are the 40x and 60x objective air or immersion objectives?Analysis of zebrafish retinal sections
It is unusual to quantify the percentage of cells expressing specific markers by counting voxels. Can the authors clarify why they took this approach?Are the micrographs depicted in the figures, single confocal planes, max. intensity projections of several confocal planes?
mCherry labelled photoreceptors are sometimes difficult to appreciate in the micrographs of the main figures - eg. Fig 2C. Here it appears that mCherry is not expressed by the entire cell. It would help to show just the mCherry on its own.
Fig 4 -GHI - it would help to see the double labelled cells prox1 and zpr1 at higher magnification
Supplementary Fig 3 - PCNA, not GS is depicted in pink. GS not GFAP is in green?
Significance
This is an interesting study that pursues candidate genes derived from Drosophila screens to highlight pathways that might operate in the regeneration of cells that were previously ablated in the vertebrate CNS.
Approaches such as the ones used in this study contribute to our overall knowledge about coaxing regeneration in the context of the mammalian CNS.
The work here should be of interest to biologists interested in regeneration. -
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 manuscript by Veen and colleagues assesses two transcription factors, and makes the novel conclusion that they regulate each other in a manner that is required for photoreceptor regeneration in zebrafish. The work is potentially exciting, because similar findings from zebrafish have found traction in translation to mammals, where regeneration of photoreceptors has surprising promise to treat blindness.
The authors have been ambitious in there approach to the problem by disrupting these genes in the adult retina, which is the appropriate context required to assess photoreceptor regeneration. Because technologies for conditional …
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 manuscript by Veen and colleagues assesses two transcription factors, and makes the novel conclusion that they regulate each other in a manner that is required for photoreceptor regeneration in zebrafish. The work is potentially exciting, because similar findings from zebrafish have found traction in translation to mammals, where regeneration of photoreceptors has surprising promise to treat blindness.
The authors have been ambitious in there approach to the problem by disrupting these genes in the adult retina, which is the appropriate context required to assess photoreceptor regeneration. Because technologies for conditional gene ablation are not very available in zebrafish, these authors use electroporation of morpholinos to accomplish their goals. Where most researchers have abandoned this very challenging approach, it seems these authors have found some success.
Together the technical feat and intriguing conclusions combine, in my opinion, to make this paper worthy of serious consideration for publication. I would hate to see it not be made available for public consumption. Its' merits are strong, but some shortcomings in communication and interpretation nevertheless should be addressed. I suggest Major and minor concerns below.
I suspect that doing further experiments would be asking a lot of the authors at this point, but I point out some possible experiments that would improve the manuscript if my suspicion is wrong. Without further experiments, I suggest much of the writing needs to be carefully qualified and less deterministic.
Major concerns:
- The authors need to quantify impacts of MO without MTZ (or with MTZ on wildtype fish without the nfsb Tg). Alternatively, the interpretation needs to be softened considerably. Observations made include increased proliferation and more PR, but these are not clearly connected by the data in a way that allows you to claim "more regeneration". A plausible alternative is that the MO was protective, and the MTZ did not kill as many PR cells when the genes were knocked down. Moreover, Figure 3 shows that only one of two proliferation markers is increased (how to explain?) and only at one timepoint, so this may be a fluke. I suggest softening the conclusions to state that gene knockdown increased proliferation and led to increased PR abundance, thus implying improved regeneration (and provide the alternative interpretation). E.g. the punchy titles of Figures 3, 5, 7 are not supported by the data; neither is text in Discussion bottom of page 15.
- Fig 5 quantification of proliferation is needed if the interpretation is about regeneration (see comment 1). Instead, the conclusions could be reworked to match the data.
- I'm unclear on why these experiments couldn't have been completed in mutant zebrafish. Are they not viable?
- Sequence and chemistry of the MO knockdown reagents must be provided. If they are similar to previously published MO reagents (several for both gene targets have been published) then this might be used to improve confidence of MO efficacy. Were the MOs modified to facilitate electroporation? The gene targets also must be listed with less ambiguity, e.g. when "prox1" is mentioned, do you mean prox1a? Without these details, the experiments fail to provide enough info to permit replication.
- A suggestion to improve the text [no need for new experiments]: The Discussion should address assumptions about MO knockdowns in regards to: a) efficacy, and b) specificity. E.g. (a) future experiments might challenge the efficacy by measuring the abundance of genes that are regulated by prox1 and her6. E.g. (b) future experiments should challenge the specificity of the MO reagents by testing to see if the same result is attained with disparate MO oligos, by phenocopy with CRISPR, performing the work in mutants (I assume rescuing the knockdown by replacing the target gene is not feasible by electroporation, but that would be ideal).
- The claim that Prox1 is in PR (Figure 4 title) is not convincing. Does the scRNA-Seq confirm this, and why not invoke this data to clarify more concretely? Figure 4A shows a lot of green prox1 signal, but that is very inconsistent with what is shown in Figure 4G, where no prox1 signal is observed in the PR. On page 12, which relates to this Figure, the authors instead say that Prox1 is detected in PR after injury (a big difference compared to title of Fig4!). Fig4I' shows some signal in the area of the PR, but the overlap of the signals is not convincing and it looks to mostly be adjacent to the zpr1 signal; maybe it is Muller glia or some other cone type, or rod cells. If it is Muller glia or rods, then the interpretation needs to be adjusted. Regardless, it is unclear if this is in LWS cones, which is presumably what regenerates after LWS cone ablation(?)
- Figures showing prox1 or Hes1 IHC (Fig 2, 4, 6, 7 & Supps) - how many replicates were evaluated (how many individual fish were assessed) to determine that these IHCs are representative.
- Some of the data, i.e. some photomicrographs of IHC, are used repeatedly in separate Figures. I cannot find a comment in the manuscript acknowledging this. Panel F is identical in Figures 3 and 5, and panel 7E is identical to Supp panel S5E. My opinion here is mixed: I think re-using these Figures is marginally ok if it is explicitly and repeatedly described (e.g. in Methods, Results, and Fig Legends), but I also think that if the authors have replicated the experiments sufficiently, then they will surely have some other micrographs to use. My opinion is tipped into grumpy and worried about good data integrity, because in both cases the lines that indicate retinal layers are drawn in different places between the replicated panels; that could happen out of sloppy-ness or instead could be a ploy to help hide the Figure recycling. I prefer to assume the authors are of good intent and have made an error (indeed the panels are all meant to represent the same control treatments) but I would not want the manuscript published without explicitly rectifying this issue. Minimally the replicate micrographs should be explicitly acknowledged. My search for other duplicated panels was not exhaustive.
Minor points:
- a) Page numbers and line numbers would make it less work to prepare a constructive critique of this paper. Similarly, the Figures need Figure numbers.
- b) On the histograms, does each dot represent an individual fish? (i.e. an independent biological replicate).
- c) It would be lovely to learn that left vs. right eyes were used as internal controls in each case, and then the authors could plot the difference between control & treatment within each individual. Perhaps this would allow normalizations or more powerful statistical tests, and then the PCNA data would be more aligned with the conclusions, for example.
- d) Figure 5: expected to see quantification of PH3 here, akin to Figure 3.
- e) P. 6 secondary antibodies probably did not come from ZIRC
- f) More should be done to acknowledge past papers examining Her6, Hes1 and Prox1 in vertebrate retina.
- g) I do not see how the final section of the manuscript (beginning with "Insulinoma" to the end of the Discussion is relevant to the paper. A very odd ending to this manuscript. Some sentences (especially beginning the section with a topic sentence) would be need to be added if this writing is to remain.
- h) The final two sentences of the Abstract were interesting - these ideas are unfortunately not Discussed again later in the manuscript.
- i) What is the source of the transgenic zebrafish line Tg(lws2:nfsb-mCherry) ? Is it maybe from Wang...Yan 2020 PLOS BIOL (PMID: 32168317)? If yes, it would be ideal to provide an allele number. If no, construction of this line should be described.
- j) Bottom page 4 says "Two transgenic lines used were crossed" but only one line is mentioned.
- k) Then on page 7, the text says "Zebrafish line Tg(her4.1:dRFP/gfap:GFP/lws2:nfsb-mCherry) for red cone ablation, ..." which muddies the waters even further.
- l) When the antibody zpr1 is described, it is mentioned as a "zinc finger" (many instances throughout). This is incorrect, and the words "zinc finger" can be removed.
- m) It would be useful to state in Methods, and at first occurrence in figure legends, that the antibody ZPR1 labels double cones (the red & green cones), and these make up about half of the cone photoreceptor population. (i.e. not all cones are evaluated in this work).
- n) Figure 2 desperately needs a panel describing methodological timeline, similar to Fig 1D. It is really hard to figure what happened when (e.g. when did ablation occur? When was the MO delivered?). This also should be described more explicitly in the Methods, which seem quite vague on this point: Electroporated fish went straight into MTZ?
- o) Throughout the authors refer to injury, e.g. hpi = hours post injury. I don't think this represents the methods very well at all, because they have ablated the cells, not injured them. Injured cells don't regenerate (because they are not dead). This miswording contributes to confusion interpreting the Figures, which are not decipherable as stand-alone items.
- p) There is a really weird yellow dotted line that spans between and ACROSS adjacent panels in Figure 2. It covers the white line separating panels F' & G', and then again in F" and G".
- q) Fig 2, it is evident that Hes1 protein is not eliminated so you cannot claim it is "not expressed". It is perhaps reduced in abundance, but signal is still obviously present.
- r) Title of Figure 6 needs to rewritten: LLPS may be occurring, but until you manipulate both LLPS and Prox1 together, you cannot claim that they act through one another.
- s) Figure 7 title needs to be rewritten: PR are not quantified here.
- t) Figure 6: I am deeply incredulous that applying any chemical to zebrafish for only two minutes can alter cell differentiation, except perhaps via toxicity. Perhaps examples of similar impacts can be provided from the literature to make it seem more credible that the mechanism here is LLPS in retinal cells.
The following minor comments are all captured under the notion that the Figure Legends all need to be re-written by a senior colleague. Figures+legends should be interpretable as stand-alone items. All these Figures fail this minimal standard. Below are some issues, but really I'd suggest starting with a blank slate.
- u) Figure 1 must mention Drosophila. So very very confusing to read this believing it is about zebrafish.
- v) Figure 1 what is "deadpan"?
- w) Fig 2 title, how do you know these progenitors are MG-derived?
- x) Fig 2, define abbreviation MG
- y) Fig 2 title, Hes1 is less abundant, but that might be from alternative mechanisms other than "reduced expression" (e.g. altered PTMs, increased clearance, LLPS, etc)
- z) Fig 3 legend is a jumble of oddity. At least three distinct signals are supposedly labelled in pink(?). separately, What about the mCherry - is it also in pink?
- aa) Most Figures: we know they are micrographs, so you don't need to lead the Description saying "micrographs of...". Instead, describe the logic of the experiment and the overall interpretations.
- bb) All Figures: it is really odd to list the data (averages & variances, including implausible significant digits on each) for every treatment - that is what the histograms are meant to convey.
- cc) Figure 4 should send the reader to the Supplemental so they know that the no-primary control experiment is available.
- dd) Fig 6B legend - explain what the chemicals are meant to do (e.g. "block LLPS").
- ee) Fig 6 define "2m" = 2 minutes?
- ff) Several more abbreviations are not defined: hpi, ONL, etc...
Significance
The manuscript by Veen and colleagues assesses two transcription factors, and makes the novel conclusion that they regulate each other in a manner that is required for photoreceptor regeneration in zebrafish. The work is potentially exciting, because similar findings from zebrafish have found traction in translation to mammals, where regeneration of photoreceptors has surprising promise to treat blindness.
Together the technical feat and intriguing conclusions combine, in my opinion, to make this paper worthy of serious consideration for publication. I would hate to see it not be made available for public consumption. Its' merits are strong, but some shortcomings in communication and interpretation nevertheless should be addressed
-
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 interrogate the roles of prox1a and her6 in red/green cone photoreceptor regeneration in zebrafish. Using scRNAseq, they find evidence that her6 is expressed in MG and suggest it is downregulated in Müller glia-derived progenitor cells during retinal regeneration. Further, they show that knockdown of her6 leads to an increase in the rate of cone PR regeneration and suggest that this is due to increased proliferation of retinal progenitors. The authors then demonstrate that prox1a is expressed in differentiated PRs and that knockdown inhibits PR regeneration. An exploration of the role of Liquid-Liquid Phase
Separation …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 interrogate the roles of prox1a and her6 in red/green cone photoreceptor regeneration in zebrafish. Using scRNAseq, they find evidence that her6 is expressed in MG and suggest it is downregulated in Müller glia-derived progenitor cells during retinal regeneration. Further, they show that knockdown of her6 leads to an increase in the rate of cone PR regeneration and suggest that this is due to increased proliferation of retinal progenitors. The authors then demonstrate that prox1a is expressed in differentiated PRs and that knockdown inhibits PR regeneration. An exploration of the role of Liquid-Liquid Phase
Separation (LLPS) in prox1a signaling is provided. Finally, the authors show that her6 knockdown leads to increased prox1a expression and that prox1a knockdown leads to loss of her6 expression, suggesting both factors impact expression of the other to regulate the kinetics of PR regeneration in the zebrafish retina.Major comments:
- As presented, the "downregulation" of her6 in MG-derived progenitors is not convincing due to a discrepancy between the single cell data and antibody staining. For "downregulation" to be the case, MG and/or progenitor cells would need to be expressing her6 and then lose it over time during regeneration. The single cell data are consistent with that showing expression of her6 in quiescent and proliferating MG, but the antibody staining is not. In addition, regarding the single cell data, the colors used to discriminate early PRs and PRs are nearly identical. The authors need to make clear which clusters correspond to MG-derived progenitor cells using clear labeling. Regarding antibody labeling, prior to cone PR ablation, the expression of Her6 (ie Hes1 staining) appears to be localized to interneurons and RGCs, not MG - yet the single cell data suggest this is not the case. Either the single cell data or the antibody labeling may be correct, but not both. The authors should provide a figure showing co-labeling of Her6/Hes1 and a MG marker such as GS or GFAP. In the absence of that result, the antibody data is most consistent with her6 expression either not being expressed in MG and MG-derived progenitor cells or expressed in only a subset of progenitors - both of which conflict with the single cell data.
- The her6 MO results are not convincing with respect to data interpretation. Knockdown of her6 results in more cone PRs at 72 hpi and more progenitor proliferation at the same time point (as assessed by pH3 immunostaining). For an increase in progenitor proliferation to explain the increase in cone PRs the observation would need to precede the increase in cones, not be coincident with it. Alternative hypotheses need to be explored, for instance, whether her6 downregulation alters cell fate choices in progenitor cells. The precocious increase in prox1 expression at 48 hpi (Figure 6A-E) would be consistent with that idea, leading either to precocious cone differentiation or a stronger bias toward the red/green cone PR fate.
- The connection to Liquid-liquid Phase Separation (LLPS) is the weakest component of the study given the following issues:
- a. Prox1 staining as punctate and/or liquid-like assembly associated is not convincingly demonstrated, this would require much higher magnification, quantification, and co-localization with factors known to aggregate in liquid-like assemblies.
- b. The reagents used to modulate LLPS are not specific to Prox1 and will alter all LLPS-associated signaling events.
- c. What assay was used to set "effective and consistent" outcomes at a concentration of 5% 1,6-HD anda treatment time of 2 minutes? Was toxicity or any other assay performed to assess the specificity of 1,6-HD effects in vivo at this concentration?
- d. Shouldn't the regenerated cone cells be expressing the lws2:nfsb-mCherry transgene? This is especially troubling given that the lws2:nfsb-mCherry transgene is evident at the periphery in both conditions whereas the zpr1antibody (i.e., Arr3a expression) appears to differentially label more central areas of the retina.
- e. Controls are needed to demonstrate that the expression of other transcription factors is normal following 1,6-HD treatment.
- f. Controls are needed to show that zpr1 staining / Arr3a expression is not selectively altered by 1,6-HD treatment, eg are other markers of red/cones similarly altered?
- Statistical methods are not provided for the Dros portion of the study. Please indicate which statistical tests were used and what corrections performed for multiple comparisons.
- A control is need to show that the antibody against mammalian Hes1 labels Her6 in zebrafish.
Minor Comments:
- Please provide alleles for all transgenic resources used as this will help to ensure others are able to replicate these findings and/or use equivalent resources for their own research.
- "Zebrafish injury model" section states that two transgenic lines were used to create a PR ablation line but only one line is listed? Please clarify. Also the larval ablation model should be included in the Methods.
- Montgomery et al ablated rods not cones. As cells exhibit differential sensitivity to cell ablation methods - including rods and cones - it would be preferable if a reference for NTR/Mtz-mediated cone cell ablation was cited.
- Figure S2: Since fish between 10-12 months were used, there is no need to mention the ablation results with the younger fish as it is unrelated to the study.
- 24 hpi as first observation of MG-derived progenitors (Fig. S3D-D') is very weak, data is not convincing. Please update the figure or revise the manuscript. Figure 2 is much better, simply referencing Figure 2E' would suffice.
- Edit needed: "They began proliferating at 48hpi, as indicated by clusters of PCNA+ cells across the inner nuclear cell layer (INL; Figure S3E-E'), also observed by at 72 and 96 hpi (Figure S3F-F' and G-G' EE', outlined).
- her6 has been previously implicated in retinal regeneration previously and references to these studies should be included in the intro and/or discussion. See: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8643038/
https://bmcdevbiol.biomedcentral.com/articles/10.1186/1471-213X-6-36
https://www.jneurosci.org/content/34/43/14403 - Please confirm her6 MO-based downregulation with a more quantitative method.
- "DamID" should be defined at the first instance (first paragraph of Results) as DNA adenine methyltransferase Identification.
- zpr-1 is an antibody not a protein and should not be capitalized (per "Zpr1+ cells").
- According to ZFIN, prox1 is now prox1a, the authors use both. It would be helpful to stay consistent and use the most recent nomenclature to refer to prox1a.
Referees cross-commenting
Reviewer #2 / Major concern #1: Unclear how one assesses the effects of gene knockdown on regeneration in the absence of cell ablation? The reviewer should clarify what "impacts" they would like quantified (e.g., developmental effects, etc) . For instance, assessing effects of gene knockdown on PR number developmentally (with egg injections of morpholino and/or sgRNAs for Crispants) would help to account for potential gene knockdown effects on progenitor differentiation (i.e., causing a fate bias toward red cones) or protection (e.g., absence of normal levels of cell death/TUNEL).
Reviewer #2 / Major concern #8: Repeated micrographs: In my view this should not be done at all. Please use different examples of control data between comparisons of her6 and prox1a knockdowns - even if these experiments were al performed in parallel there should be additional control samples from each assay and across a minimum of three biological repeats.
Reviewer #3, Paragraph 2 ("The authors state that the loss of Her6..."): Pulse-chase experiments demonstrating an increase in the number of newly generated red cones - by virtue of co-labeling with pulse marker and Lws2 transgene - would help to address this concern and clarify the degree of bias exhibited in the red cone ablation paradigm (an issue discussed by all reviewers and which the authors are familiar with, eg, DOI: 10.1186/s13064-017-0089-y)
Significance
This work is significant in interrogating the function of two classic factors associated with regulation of neurogenesis and neuronal differentiation in Drosophila in the context of cone photoreceptor regeneration in zebrafish. My expertise lies in the latter paradigm and I have restricted my evaluation to that portion of the study. The novelty of the study is diminished somewhat by recent reports of the role of Notch signaling in retinal regeneration in zebrafish - thus, it would follow that a major signaling component of the Notch pathway, Hes1/Her6, would likely be involved as well. Indeed, expression changes for her6 during regeneration have been previously reported (see refs above). However, the present study does add a functional characterization which is a significant advance. The advent of stimulation of retinal regenerative capacity in mice by Tom Reh and colleagues adds intrigue to the study of retinal regeneration in zebrafish as the insights generated are being applied in a way that suggests possible relevance to human disease. Thus defining conditions that lead to accelerated retinal regeneration is of widespread interest. However, several issues raised in the detailed review above diminish my enthusiasm for the study overall.
-