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

We thank the reviewer for their constructive comments and the fair and interesting discussion between reviewers.

__Reviewer #1 __

We are delighted to read that the reviewer finds the manuscript “very clear and of immediate impact […] and ready for publication” regarding this aspect. We have toned down the conclusion, proposing rather than concluding that “the incapacitation of Cmg2[KO] intestinal stem cells to function properly […] is due to their inability to transduce Wnt signals”.

We have addressed the 3 points that were raised as well as the minor comments.

Point #1

The mouse mutant is just described as 'KO', referring to the previous work by the authors. The cited work simply states that this is a zygotic deletion of exon 3, which somehow leads to a decrease in protein abundance that is almost total in the lung but not so clear in the uterus. Exon 3 happens to be 72 bp long [https://www.ncbi.nlm.nih.gov/nuccore/NM_133738], so its deletion (assuming there are no cryptic splicing sites used) leads to an internal in-frame deletion of 24 amino acids. So, at best, this 'KO' is not a null, but a hypomorphic allele of context-dependent strength.

Unfortunately, neither the previous work nor this paper (unless I have missed it!) contains information provided about the expression levels of Cmg2 in the intestine of KO mice - nor which cell types usually express it (see below). I think that using anti Cmg2 in WB and immunohistofluorescence of with ISC markers with intestine homogenate/sections of wild-type and mutant mice would be necessary to set the stage for the rest of the work.

We now provide and explanation and characterization the Cmg2KO mice. Exon 3 indeed only encodes a short 24 amino acid sequence. This exon however encodes a ß-strand that is central to the vWA domain of CMG2, and therefore critical for the folding of this domain. As now shown in Fig. S1c, CMG2Dexon3 is produced in cells but cleared by the ER associated degradation pathway, therefore it is only detectable in cells treated with the proteasome inhibitor MG132, at a slightly lower molecular weight than the full-length protein. This is consistent, and was inspired by the fact that multiple Hyaline Fibromatosis missense mutations that map to the vWA domain lead to defective folding of CMG2, further illustrating that this domain is very vulnerable to modifications. In Fig. S1c, we moreover now show immunoprecipitation of Cmg2 from colonic tissue of wild-type (WT) and knockout (KO) mice, which confirm the absence of Cmg2 protein in Cmg2KO samples.

Point #2

Connected to the previous point, the expression pattern of Cmg2 in the intestine is not described. Maybe this is already established in the literature, but the authors do not refer to the data. This is important when considering that the previous work of the authors suggests that Cmg2 might contribute to Wnt signalling transduction through physical, cis interactions with the Wnt co-receptor LRP6. Therefore, one would expect that Cmg2 would be cell-autonomously required in the intestinal stem cells.

The expression pattern of Cmg2 in the gut has not been characterized and is indeed essential to understanding its function. To address this gap, we now added a figure (Fig. 1) providing data from publicly available RNA-seq datasets and from our RNAscope experiments on Cmg2WT mice. Of note, we unfortunately have never managed to detect Cmg2 protein expression by immunohistochemistry of mouse tissue with any of the antibodies available, commercial or generated in the lab.

In the RESULTS section we now mention:

To investigate Cmg2 expression in the gut, we first analyzed publicly available spatial and scRNA-seq datasets to identify which cell types express Cmg2 across different gut regions. Spatial transcriptomic data from the mouse small intestine and colon revealed that Cmg2 is broadly expressed throughout the gut, including in the muscular, crypt, and epithelial layers (Fig. 1A–C). To validate these findings, we performed RNAscope in situ hybridization targeting Cmg2 in the duodenum and colon of wild-type mice. The expression pattern observed was consistent with the spatial transcriptomics data (Fig. 1D–E). We then analyzed scRNA-seq data from the same dataset to assess cell-type-specific expression in the mouse colon. Cmg2 was detected at varying levels across multiple cell types, including enterocytes and intestinal stem cells, as well as mesenchymal cells, notably fibroblasts.

Of note for the reviewer, not mentioned in the manuscript, this wide-spread distribution of Cmg2 across the different cell types is not true for all organs. We have recently investigated the expression of Cmg2 in muscle and found that it is almost exclusively expressed in fibroblasts (so-called fibro-adipocyte progenitors) and very little in any other muscle cells, in particular fibers.

Interestingly also, as now mentioned in the manuscript and shown in Fig. S1,the ANTXR1 protein, which is highly homologous to Cmg2 at the protein level and share its function of anthrax toxin receptor, displayed a much more restricted expression pattern, being confined primarily to fibroblasts and mural cells, and notably absent from epithelial cells. This differential expression highlights a potentially unique and epithelial-specific role for Cmg2 in maintaining intestinal homeostasis.

Point #3

The authors establish that the regenerating crypts of Cmg2[KO] mice are unable to transduce Wnt signalling, but it is not clear whether this situation is provoked by the DSS-induce injury or existed all along. Can Cmg2[KO] intestinal stem cells transduce Wnt signalling before the DSS challenge? If they were, it might suggest that the 'context-dependence' of the Cmg2 role in Wnt signalling is contextual not only because of the tissue, but because of the history of the tissue or its present structure. It would also suggest that Cmg2 mutant mice, unless reared in a germ-free facility for life, would eventually lose intestinal homeostasis, and maybe suggest the level of intervention/monitoring that HFS patients would require. It might also provide an explanation in case Cmg2 was not expressed in ISCs - if the state of the tissue was as important as the presence of the protein, then the effect on Wnt transduction could be indirect and therefore it might not be required cell-autonomously.

We agree that understanding whether Cmg2KO intestinal stem cells are intrinsically unable to transduce Wnt signals, or whether this defect is contextually induced following injury (such as DSS treatment), is a critical point.

As a first line of evidence, we show than under homeostatic condition, Wnt signaling appears largely intact in Cmg2KO crypts, with comparable levels of ß-catenin and expression levels of canonical Wnt target genes (e.g., Axin2, Lgr5) to those observed in WT animals (Figs. S1j-l and S3d-e). This indicates that Cmg2 is not essential for basal Wnt signaling under steady-state conditions.

These findings thus support the idea that the requirement for Cmg2 in Wnt signal transduction is context-dependent—not only at the tissue level but also temporally, being specifically required during regenerative processes or in altered microenvironments such as during inflammation or epithelial damage. This context-dependence may reflect changes in the composition or accessibility of Wnt ligands, receptors, or matrix components during repair, where Cmg2 could play a scaffolding or stabilizing role.

These aspects are now discussed in the text.

I think points 1 and 2 are absolutely fundamental in a reverse genetics investigation. Point 3 would be nice to know but the outcome would not change the tenet of the paper. I believe that the work needed to deal these points can be performed on archival material. I do not think the mechanism proposed can be taken from 'plausible' to 'proven' without proposing substantial additional investigation, so I will not suggest any of it, as it could well be another paper.

We have addressed points 1 and 2, and provided evidence and discussion for Point 3.

__Minor points __

1- Figure 1 legend says "In (c), results are mean {plus minus} SEM" - this seems applicable to (d) as (c) does not show error whiskers.

We thank the reviewer for picking up this error. We modified : “In (c), results are median” and “In (d, f and g) Results are mean ± SEM.”

2- Figure 1 legend says "(d) Body weight loss, (f) the aspect of the feces and presence of occult blood were monitored and used for the (e) DAI. Results are mean {plus minus} SEM. Each dot represents the mean of n = 12 mice per genotype". This part looks like has suffered some rearrangement of words. The first instance of (f) should be (e), I guess, and I am not sure what "(e) DAI" means. And for (e), "mean {plus minus} SEM" does not seem applicable. This needs some light revision.

The legend was clarified as followed : “(d) __Body weight loss, and (e) aspect of the feces and presence of occult blood were monitored and used to evaluate Disease activity index in (f).__

3 - Figure 1H legend does not say which statistical test was made in the survival experiment in (h) - presumably log-rank? A further comment on the survival statistics: euthanised animals should not be counted towards true mortality when that is what is recorded as an 'event'. They should be right-censored. However, in this case, reaching the euthanasia criterion is just as good an indicator of health as mortality itself. So, simply by changing the Y axis from 'survival' to 'event-free survival' (or something to that effect), where 'events' are either death or reaching the euthanasia criterion, leaves the analysis as it is, and authors do not need to clarify that figure 1H shows "apparent mortality", as it is straightforward "complication-free survival" (just not entirely orthogonal to weight loss).

The Y axis was changed from 'survival' to “percentage of mice not reaching the euthanasia criterion”.

4 - Some density measurements are made unnecessarily on arbitrary units (per field of view) - this should be simple to report in absolute measures (i.e. area of tissue screened or, better still, length of epithelium screened).

Because the aera of tissue can vary significantly between damages, regenerating and undamaged tissue, we reported the length of epithelium screened as suggested : “per 800um tissue screened” in Fig S1c and Fig 2b.

5 - Figure 2E should read "percent involvement"

This has been corrected.

6 - Figure 2J should read "lipocalin..."

This has been corrected.

7 - In section "CMG2 Is Dispensable for YAP/TAZ-Mediated Reprogramming to Fetal-Like Stem Cells", the authors write ""We measured the mRNA levels of two additional YAP target genes, Cyr61 and CTGF...". I presume the "additional" is because Ly6a is also a target of YAP/TAZ, but if the reader does not know, it is puzzling. I would suggest to make this link explicit.

We added : “In addition to the fetal-like stem cell marker Ly6a, which is a YAP/TAZ target gene, we measured the mRNA levels of two others YAP target genes, Cyr61 and CTGF”

8 - In Figures S2, 3 and S3, I think that the measures expressed as "% of homeostatic X in WT" really mean "% of average homeostatic X in WT". This should be made clear somewhere.

We added: “Dotted line represents the average homeostatic levels of Cmg2 WT” in figure legends

9 - In panel C, the nature of the data is not entirely clear. First, the corresponding part of the legend says "Representative images of n=4 mice per genotype" which I presume should refer to panel B. Then, the graph plots 4 data points, which suggests that they correspond to 4 mice - but how many fields of view? Also, the violin plot outline is not described - I presume it captures all the data points from the coarse-grained pixel analysis, but it should be clarified.

It was modified as suggested : “(c) Results are presented as violin plot of the Ly6a mean intensity of all data points from the coarse-grain analysis. Each symbol represents the mean per mice of n=4 mice per condition. Results are mean ± SEM. Dotted line represents the average homeostatic levels of Cmg2WT. P values obtained by two-tailed unpaired t test.”

10 - In Figure 3H and 3I, I would suggest to add the 7+3 timepoint where the data come from.

We unfortunately do not understand the suggestion of the reviewer, given that these panels show the 7+3 time point.

11 - In section "CMG2 Is Critical for Restoring the Lgr5+ Intestinal Stem Cell Pool", the authors say "...The mRNA levels of ... LRP6, β-catenin (Fig. S3a-b), and Wnt ligands (Wnt5a, 5b, and 2b) were comparable between the colons of Cmg2WT and Cmg2KO mice (Fig. S3c)..." without clarifying in which context - one needs to read the figure legend to realise this is "timepoint 7+3". I suggest to add "in the recovery phase" or "in regenerating colons" or something shorter, just to guide the reader.

We added : “Initially, we quantified the expression of key molecular components involved in Wnt signaling in mice colon 3 days after DSS withdrawal using qPCR.”

12 - Like with the previous point, it is not clear when the immunohistofluorescence of B-catenin is made - not even in the legend, as far as I could see. The only hint is that authors say "the nuclei of cells in the atrophic crypts of Cmg2KO..." with 'atrophic' probably indicating again the 7+3 timepoint.

We have changed the text and now mention “Next, we analyzed β-catenin activation in the colon of Cmg2WT and Cmg2KO mice during the recovery phase.”

13 - A typo in the discussion: tunning for tuning.

This has been corrected.

14 - In the discussion, the authors talk about the 'CMG2' protein (all caps - formatting convention for human proteins) but before they were referring to 'Cmg2' (formatting convention for mouse proteins). That is fine but some of the statements where "CMG2" is used clearly refer to observations made in the mouse.

We have now used Cmg2, whenever referring to the mouse protein.

15 - Typos in methods: "antigen retrieval by treating [with] Proteinase K"; "Image acquisition and analyze [analysis]"; "All details regarding code used for immunofluorescence analysis”.

This has been corrected.

__Reviewer #2 __

We are very pleased to read that the reviewer found the study “overall well designed, meticulously carried out, and with clear and convincing results that are most reasonably and thoughtfully interpreted”.

For this reader, one additional thought comes to mind. If I understand the field correctly it would be informative to know with greater confidence where - in what cell type, epithelial or mesenchymal - the CMG2-LRP6-WNT interaction occurs.

This point was also raised by Reviewer I, and we have now added a new Figure 1, that describes Cmg2 expression in the gut, based both on from publicly available RNA-seq datasets and our RNAscope experiments on Cmg2WT mice. Of note, we unfortunately have never managed to detect Cmg2 protein expression by immunohistochemistry of mouse tissue with any of the antibodies available, commercial or generated in the lab.

After injury the CMG2-KO mouse epithelium exhibits defective WNT signal transduction - as evidenced by failure of b-catenin to translocate into the nucleus. At first glance, this result is a disconnect with the paper by van Rijin that claims the defect in Hyaline Fibromatosis Syndrome cannot be due to loss of CMG2 expression/function in the barrier epithelial cell - a claim based on the mostly normal phenotypes of human CMG2 KO duodenal organoids. But the human organoids studied in the van Rijin paper, like all others, are established and cultured in very high WNT conditions, perhaps obscuring the lack of the CMG2-LRP6-WNT interaction. And in fact, the phenotypes of these human CMG2-KO duodenoids were not entirely normal - the CMG2-KO stem-like organoids (even when cultured in high WNT/R-spondin conditions) developed abnormal intercellular blisters consistent with a defect in epithelial structure/function - of unknown cause and not investigated.

We thank the reviewer for raising this point and we fully agree. We now specify in the text that the human CMG2-KO duodenoids showed blisters, indeed consistent with a defect in epithelial structure/function, and that they were grown on high Wnt media which likely obscure the CMG2 requirement.

I think it would be informative to prepare colon organoids (and duodenoids) from WT and CMG2-KO mice to quantify their WNT dependency during establishment and maintenance of the stem-like (and WNT-dependent) state. If CMG2 acts within the epithelial cell to affect WNT signaling (regardless of WNT source), organoids prepared from colons of CMG2-KO mice would require more WNT in culture media to establish and maintain the stem cell proliferative state - when compared to organoids prepared from WT mice. This can be quantified (and confirmed molecularly by transgene expression if successful). Enhanced dependency of high concentrations of exogenous WT would be evidence for a primary defect in WNT-(LRP2)-CMG2 signal transduction localized to the epithelial barrier cell - thus addressing the apparent discrepancy with the van Rijin paper - and for my part, advancing the field. And the discovery of a defect in the epithelium itself for WNT signal transduction would implicate a biologically most plausible mechanism for development of protein losing enteropathy.

By no means do I consider these experiments to be required for publication (especially if considered to be incremental or already defined - WNT-CMG2 is not my field of research). This study already makes a meaningful contribution to the field as I state above. But in the absence of new experimentation, the issue should probably be discussed in greater depth.

We are working out conditions to grow colon organoids that from WT and Cmg2 KO mice, indeed playing around with the concentrations of Wnt in the various media to identify those that would best mimic the regeneration conditions. This is indeed a study in itself. We have however included a discussion on this point in the manuscript as suggested.

__Reviewer #3: __

We thank the reviewer for her/his insightful comments.

The premise is that the causative germline mutated gene, CMG2/ANTRX2, may have a functional role in colonic epithelium in addition to controlling the ECM composition. There is little background information but one study has shown no primary defect in epithelial organoids grown from patients with the syndrome. This leads the authors to wonder if non-homeostatic, conditions might reveal a function role for the gene in regeneration.

Reviewer 2 commented on the fact that “human organoids studied in the van Rijin paper, like all others, are established and cultured in very high WNT conditions, perhaps obscuring the lack of the CMG2-LRP6-WNT interaction. And in fact, the phenotypes of these human CMG2-KO duodenoids were not entirely normal - the CMG2-KO stem-like organoids (even when cultured in high WNT/R-spondin conditions) developed abnormal intercellular blisters consistent with a defect in epithelial structure/function - of unknown cause and not investigated”.

We have now added a discussion on this point in the manuscript.

The authors' approach to test the hypothesis is to use a mouse germline knockout model and to induce colitis and regeneration by the established protocol of introducing dextran sodium sulfate (DSS) into the drinking water for five days. In brief there is no phenotype apparent in the untreated knockout (KO) but these animals show a more severe response to DSS that requires them to be killed by 10 days after the start of treatment. This effect following phenotypic characterisation of the colonic epithelium is interpreted as showing the CMG2 is a Wnt modifier required for the restoration of the intestinal stem cell population in the final stages of repair.

The experiment and analysis seem reasonably well executed - although a few specific comments follow below. The narrative is simple and easy to understand. However, there are significant caveats that cast doubts on the interpretation made that loss of CMG2 impairs the transition of colonic epithelial cells from a fetal like state to adult ISCs.

First there is only a single approach and single type of experiment performed. There is a lack of independent validation of the phenotype and how it is mediated.

We do not fully understand what type of independent validation of the phenotype the reviewer would have liked to see. Is it the induction of intestinal damage using a stress other than DSS?

The DSS dose in this kind of experiment is often determined empirically in individual units. Here the 3% used is within published range but at upper end. The control animals show a typical response with symptoms of colitis worsening for 2-3 days after the removal of DSS and then recovery commonly over another 5-7 days. Here the CMG2 KO mice fail to recover and are killed by 9 or 10 days. The authors attempt to exploit the time course by identifying normal initial (7days) and defective late (10days) repair phases in KO animals when compared to controls. It is from this comparison that conclusions are drawn. However, the alternative interpretation might be that the epithelium of KO animals is so badly damaged, and indeed non-existent (from viewing Fig2a), that it is incapable of mounting any other response other than death and that the profiling shown is of an epithelium in extremis. The repair capability and dynamics of the KO would have been better tested under more moderate DSS challenge, if this experiment had been regarded as a pilot rather than as definitive.

The choice of 3% DSS was in fact based on a pilot experiment. As now shown in Fig. S4, we tested different concentrations and found that 3% DSS was the lowest concentration that reliably induced the full spectrum of colitis-associated symptoms, including significant body weight loss, diarrhea, rectal bleeding (summarized in the Disease Activity Index), as well as macroscopic signs such as colon shortening and spleen enlargement. Based on these criteria, we selected 3% DSS for the study described in the manuscript.

In this model, WT mice showed a typical progression: body weight stabilized rapidly after DSS withdrawal, with resolution of diarrhea and rectal bleeding. Histological analysis at day 9 revealed signs of epithelial regeneration, including hypertrophic crypts and increased epithelial proliferation.

In contrast, Cmg2KO mice failed to initiate this recovery phase. Clinical signs such as weight loss, diarrhea, and bleeding persisted after DSS withdrawal, ultimately necessitating euthanasia at day 9–10 due to humane endpoint criteria. Unfortunately, this prevented us from exploring later timepoints to determine whether regeneration was delayed or completely abrogated in the absence of Cmg2.

Regarding the severity of epithelial damage, as raised by Reviewer 1, we now provide detailed histological scoring in the supplementary data. This analysis shows that the severity of inflammation and crypt damage was similar between WT and KO animals, as were inflammatory markers such as Lipocalin-2. The key difference lies in the extent of tissue involvement. While the lesions in WT mice were more localized, Cmg2KO mice displayed widespread and diffuse damage with no sign of regeneration as shown by the absence of hypertrophic crypts and a marked reduction in both epithelial coverage and proliferative cells. Importantly, at day 7, the percentage of epithelial and proliferating cells was comparable between genotypes, further supporting the idea that Cmg2KO mice failed to initiate this recovery phase and present a defective repair response.

The animals used were young (8 weeks) and lacked any obvious defect in collagen deposition. Does this change with treatment? Even if not, is it possible that there is a defect in peristalsis or transit time of gut contents, resulting in longer dwell times and higher effective dose of DSS to the KO epithelium?

Collagen deposition, particularly of collagen VI, is known to increase in response to intestinal injury and plays a critical role in promoting tissue repair following DSS-induced damage (Molon et al., PMID: 37272555). As suggested, we investigated whether Cmg2KO mice exhibit abnormal collagen VI accumulation following DSS treatment.

Our results show that, consistent with published data, WT mice exhibit a marked increase in collagen VI expression during the acute phase of colitis, with levels returning toward baseline following DSS withdrawal. A similar expression pattern was observed in Cmg2KO mice, with no significant differences in Col6a1 mRNA levels between WT and KO animals throughout the entire time course of the experiment. This observation was further confirmed at the protein level by western blot and immunohistochemistry analyses, suggesting that the impaired regenerative capacity observed in Cmg2KO mice is independent of Collagen VI.

Regarding the possibility of altered peristalsis or intestinal transit time contributing to increased DSS exposure in KO mice, this is indeed a possibility. Although we did not directly measure gut motility in this study, we did not observe any signs of intestinal obstruction or fecal retention in Cmg2KO mice. Indeed, during the experiment, animals were single caged for 30min in order to collect feces and no difference in the amount of feces collected was observed between WT and KO mice, arguing against a substantial difference in transit time (see figure below). The possible altered peristalsis and these observations are now mentioned in the discussion.

Is CMG2 RNA and protein expressed in the colonic epithelium? It is not indicated or tested in the submitted manuscript. This reviewer struggled to find evidence, notably it did not seem to be referenced in the organoid paper they reference in introduction (ref 13).

This very valid point was also raised by Reviewers 1 and 2. The expression pattern of Cmg2 in the gut has indeed not been characterized and is essential to understanding its function. To address this gap, we added a figure (Fig. 1) providing data from publicly available RNA-seq datasets and from our RNAscope experiments on Cmg2WT mice. Of note, we unfortunately have never managed to detect Cmg2 protein expression by immunohistochemistry of mouse tissue with any of the antibodies available, commercial or generated in the lab.

__Specific comments: __

Figure 3 c-e and associated text are confusing. In c the Y scale seems inappropriate to show percentages up to 15,000%.

In this graph values are normalized to homeostatic level of WT mice which represent 100%

In d and e the use of percentages may by correct. However, it is claimed in text that Cty61 and CTFG are upregulated in the KO. That is not what the plots appear to show as the compare to WT untreated cells, in which case the KO have not downregulated these genes in the way the controls have.

As clarified in the text, under regenerative conditions, a transient activation of YAP signaling is crucial to induce a fetal-like reversion of intestinal stem cells. However, in a subsequent phase, the downregulation of YAP and the reactivation of Wnt signaling are necessary to complete intestinal regeneration. Several studies have highlighted a strong interplay between the Wnt and YAP pathways, suggesting that their coordinated regulation is essential for effective gut repair. Nevertheless, the precise mechanisms governing this interaction remain incompletely understood.

In our model, this critical transition—YAP downregulation and Wnt reactivation—appears to be impaired. CMG2 may either hinder Wnt reactivation directly, or lead to sustained YAP signaling, which in turn suppresses activation of the Wnt pathway. Further studies, using in-vivo model and organoid models, will be necessary to understand the mechanistic role of Cmg2 in this regulatory process.

A precision of the figure has been updated as followed: both of which were significantly upregulated in the injured colons of Cmg2KO mice compared to DSS-injured Cmg2WT mice

__**Referees cross-commenting** __

Rev2 Points 1 and 2 made by Referee 1 (and point 4 of Referee 3) appear most reasonable, and if not already done should be.

We have indeed addressed these 2 points.

I also noted the more severe morphology of DSS damaged epithelium shown in Fig 2a noted by Referee 3 - and this I agree is a confounding factor. […] For my part, the concern is understandable but likely not operating in a confounding way. And the evidence for the reprogramming of the damaged epithelium into "fetal-like stem cells" (the 1st step in restitution of lost stem cells) occurs in both WT and KO mice - and these data are strong. For this reader, the block convincingly shows up for KO mouse at the WNT dependent step

The representative image has been updated, and a transverse section has been added to better illustrate that, although both epithelium and crypt structures can be present, the epithelial morphology differs significantly. Indeed, the regenerating epithelium of Cmg2WT mice displays a thick epithelial layer with well-polarized epithelial cells, whereas in cmg2KO mice, the epithelium appears atrophic, characterized by a thinner epithelial layer and elongated epithelial cells.

__Rev 3 __

This reviewer remains sceptical. I agree the authors performed the experiment well to confirm that DSS dosing was as equivalent as possible across the study. But DSS acts to induce colitis because it is concentrated in the colonic lumen as water is absorbed. Also ECM responses and remodelling are a central part of colitis models. And my concern is that the actual exposure in the KO group is influenced by transit of faeces/DSS is secondary to the known action of CMG2 on collagen deposition. The consequence of this being a protracted damage phase in which a restoration of adult stem cells would not be expected and leading to epithelial failure.

However, we differ. I might propose that the authors are asked to investigate and confirm expression of CMG2 in the epithelium and to repeat the analysis of collagen levels they performed on untreated CMG2 KO mice on colons from CMG2 KO mice having received DSS to see if these differ from controls.

This has now been done.

__Rev 1 __

Both reviewer #2 and reviewer #3 make relevant points, from the point of view of extracting as much biological knowledge as we can from the observations reported in the manuscript.

Reviewer #2 suggestion to use Cmg2[KO] organoids to investigate the dependence of Wnt transduction on Cmg2 is the type of experiments I refrained to propose. However, I think the "skeleton" of the mechanism is there and is reasonably solid. Fleshing it out may well be another paper.

I agree with Reviewer #3 objections to the timing and severity of the DSS damage. However, I am not sure how much they invalidate the main tenet of the paper:

• DSS may affect Cmg2[KO] more severely, but the overall disease score is comparable during the DSS treatment. If this severity was enough to be the main driver of the phenotype, it should have left a mark in the Histological and Disease activity scores. In this regard, I think it would be helpful if the authors provided an expanded version of Figure 2A with examples of the different levels of "Crypt damage" scored, and the proportions for each. This could be in the supplementary material and would balance the impressions induced by a single image.

As suggested, we included a detail of histological score including the crypt damage score in Supplementary Fig 3i showing no significant differences in crypt damage between Cmg2WT and Cmg2KO mice.

• If DSS affected the recovery, this would also be compatible with having a more severe histological phenotype (which is not shown overall, just in Fig 2A) because one would also expect the tissue to attempt regeneration during the 7 days of DSS treatment.

This is an interesting point, and we now allude to this aspect in the manuscript.

• The only objection that I find difficult to argue is the effective duration of the treatment. If indeed peristalsis is affected, it may be that during the 'recovery' phase there is still DSS in the intestine. This could be perhaps verified using a DS detection assay (e.g. https://arxiv.org/pdf/1703.08663) on the intestinal contents or the faeces of the mice during the 3-day recovery period.

We have attempted to obtain and purchase Heparin Red to perform this assay. Unfortunately, we have not obtained the reagent, which has never been delivered. We now also mention the following in the Discussion:

One could envision that Cmg2KO mice have a defect in peristalsis resulting in longer dwell times and possibly higher effective dose of DSS to the KO epithelium. We however did not observe any signs of intestinal obstruction or fecal retention in Cmg2KO mice. Animals were single-caged for 30 min to collect feces. We did not observe any difference in amounts collected from WT and KO mice, arguing against a substantial difference in transit time of gut contents. Moreover, if DSS affected the recovery, one would have expected a more severe histological phenotype in the colon of Cmg2KO since the tissue likely already attempts regeneration during the 7 days of DSS treatment. But this was not the case. Therefore, while we cannot formally rule out the presence of residual DSS in Cmg2KO mice during the DSS withdrawal phase, there is currently no indication that this was the case.

I think of what the aim of scholarly publication is, with this paper, and I find myself going back to a statement of the authors' discussion - that this work suggests that infants risking death may be offered (compassionate, I guess) IBD treatment. What does this hinge upon? I think, on the basic observation that diarrhoea (in the mouse model) is not intrinsic but caused by an inflammation-promoting insult. Is this substantiated? I think it is. Could we learn more biology from this disease model, about Wnt and about how ECM affects tissue regeneration? Certainly. Can this learning wait? I believe it can.

We thank the reviewer for this statement.

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

This manuscript has a good rationale in trying to understand why infants with an inherited condition, Hyaline Fibromatosis Syndrome, that is primarily associated with turnover and deposition of extracellular collagen also develop severe diarrhoea that can contribute to their premature death. The premise is that the causative germline mutated gene, CMG2/ANTRX2, may have a functional role in colonic epithelium in addition to controlling the ECM composition. There is little background information but one study has shown no primary defect in epithelial organoids grown from patients with the syndrome. This leads the authors to wonder if non-homeostatic, conditions might reveal a function role for the gene in regeneration.

The authors' approach to test the hypothesis is to use a mouse germline knockout model and to induce colitis and regeneration by the established protocol of introducing dextran sodium sulfate (DSS) into the drinking water for five days. In brief there is no phenotype apparent in the untreated knockout (KO) but these animals show a more severe response to DSS that requires them to be killed by 10 days after the start of treatment. This effect following phenotypic characterisation of the colonic epithelium is interpreted as showing the CMG2 is a Wnt modifier required for the restoration of the intestinal stem cell population in the final stages of repair.

The experiment and analysis seem reasonably well executed - although a few specific comments follow below. The narrative is simple and easy to understand. However, there are significant caveats that cast doubts on the interpretation made that loss of CMG2 impairs the transition of colonic epithelial cells from a fetal like state to adult ISCs.

Significance

1. First there is only a single approach and single type of experiment performed. There is a lack of independent validation of the phenotype and how it is mediated.
2. The DSS dose in this kind of experiment is often determined empirically in individual units. Here the 3% used is within published range but at upper end. The control animals show a typical response with symptoms of colitis worsening for 2-3 days after the removal of DSS and then recovery commonly over another 5-7 days.

Here the CMG2 KO mice fail to recover and are killed by 9 or 10 days. The authors attempt to exploit the time course by identifying normal initial (7days) and defective late (10days) repair phases in KO animals when compared to controls. It is from this comparison that conclusions are drawn.

However, the alternative interpretation might be that the epithelium of KO animals is so badly damaged, and indeed non-existent (from viewing Fig2a), that it is incapable of mounting any other response other than death and that the profiling shown is of an epithelium in extremis. The repair capability and dynamics of the KO would have been better tested under more moderate DSS challenge, if this experiment had been regarded as a pilot rather than as definitive. 3. The animals used were young (8 weeks) and lacked any obvious defect in collagen deposition. Does this change with treatment? Even if not, is it possible that there is a defect in peristalsis or transit time of gut contents, resulting in longer dwell times and higher effective dose of DSS to the KO epithelium? 4. Is CMG2 RNA and protein expressed in the colonic epithelium? It is not indicated or tested in the submitted manuscript. This reviewer struggled to find evidence, notably it did not seem to be referenced in the organoid paper they reference in introduction (ref 13).

Specific comments:

Figure 3 c-e and associated text are confusing. In c the Y scale seems inappropriate to show percentages up to 15,000%. In d and e the use of percentages may by correct. However, it is claimed in text that Cty61 and CTFG are upregulated in the KO. That is not what the plots appear to show as the compare to WT untreated cells, in which case the KO have not downregulated these genes in the way the controls have.

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 paper uses mice lacking Capillary Morphogenesis Gene 2 (CMG2- KO) mice to investigate the pathogenic mechanism underlying the protein losing enteropathy seen in children with severe Hyaline Fibromatosis Syndrome. Significance of the work is further enhanced as the intestinal phenotype induced by CMG2-KO provided a model system (with robust validated tools) for testing newly emerging (and paradigm shifting) ideas in mechanisms of tissue regeneration after injury - generalizable to tissue restitution beyond the intestine.

The study shows that in the mouse colon CMG2 plays a critical role in recovery from mucosal/epithelial damage chemically induced by dextran-sulfate-sodium (DSS). Mice lacking CMG2 failed to recover from DSS colitis with no evidence for restitution of the DSS-damaged epithelium. WT mice recovered after DSS removal.

The first step in restitution of epithelial damage in the intestine, when the epithelial stem-cell populations are depleted as in this model of DSS colitis, occurs by the transformation of surviving differentiating/differentiated epithelial cells back into a stem-cell-like (fetal-cell-like) state. This step in the process was found to occur normally in the CMG2 KO mouse. The block in restitution was located to the step where de-differentiated (fetal-cell-like) colonocytes are induced back into their WNT-dependent proliferative state - thus replenishing the normally proliferating stem (LGR5+) cells of the colonic crypt. The reason for this failure is explained by a defect in WNT signaling in the injured colons of CMG2 KO mice, as assessed by failure of -catenin translocation into the nucleus of barrier epithelial cells - a down-stream effect of WNT signaling and consistent with the dependence on CMG2 for WNT signaling in other experimental systems.

The study is overall well designed, meticulously carried out, and with clear and convincing results that are most reasonably and thoughtfully interpreted. The paper makes a meaningful contribution to the field. It models an experiment of nature to test, delineate, and verify disease pathogenesis and a newly revised mechanism for mucosal tissue repair.

For this reader, one additional thought comes to mind. If I understand the field correctly it would be informative to know with greater confidence where - in what cell type, epithelial or mesenchymal - the CMG2-LRP6-WNT interaction occurs.

After injury the CMG2-KO mouse epithelium exhibits defective WNT signal transduction - as evidenced by failure of -catenin to translocate into the nucleus. At first glance, this result is a disconnect with the paper by van Rijin that claims the defect in Hyaline Fibromatosis Syndrome cannot be due to loss of CMG2 expression/function in the barrier epithelial cell - a claim based on the mostly normal phenotypes of human CMG2 KO duodenal organoids. But the human organoids studied in the van Rijin paper, like all others, are established and cultured in very high WNT conditions, perhaps obscuring the lack of the CMG2-LRP6-WNT interaction. And in fact, the phenotypes of these human CMG2-KO duodenoids were not entirely normal - the CMG2-KO stem-like organoids (even when cultured in high WNT/R-spondin conditions) developed abnormal intercellular blisters consistent with a defect in epithelial structure/function - of unknown cause and not investigated.

I think it would be informative to prepare colon organoids (and duodenoids) from WT and CMG2-KO mice to quantify their WNT dependency during establishment and maintenance of the stem-like (and WNT-dependent) state. If CMG2 acts within the epithelial cell to affect WNT signaling (regardless of WNT source), organoids prepared from colons of CMG2-KO mice would require more WNT in culture media to establish and maintain the stem cell proliferative state - when compared to organoids prepared from WT mice. This can be quantified (and confirmed molecularly by transgene expression if successful). Enhanced dependency of high concentrations of exogenous WT would be evidence for a primary defect in WNT-(LRP2)-CMG2 signal transduction localized to the epithelial barrier cell - thus addressing the apparent discrepancy with the van Rijin paper - and for my part, advancing the field. And the discovery of a defect in the epithelium itself for WNT signal transduction would implicate a biologically most plausible mechanism for development of protein losing enteropathy.

By no means do I consider these experiments to be required for publication (especially if considered to be incremental or already defined - WNT-CMG2 is not my field of research). This study already makes a meaningful contribution to the field as I state above.

But in the absence of new experimentation, the issue should probably be discussed in greater depth.

Significance

The study is overall well designed, meticulously carried out, and with clear and convincing results that are most reasonably and thoughtfully interpreted. The paper makes a meaningful contribution to the field. It models an experiment of nature to test, delineate, and verify disease pathogenesis and a newly revised mechanism for mucosal tissue repair.

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

In this work, Bracq and colleagues provide clear evidence that the persistent diarrhoea seen in a mouse model of Hyaline Fibromatosis Syndrome is related to the inability of their intestinal epithelium to properly regenerate. This is very clear and of immediate impact. This aspect of the paper, I think, is ready for publication, and would merit immediate dissemination on its own. It is great that the manuscript is in bioRxiv already.

I am not so thoroughly convinced about the mechanism that the author propose to explain the incapacitation of Cmg2[KO] intestinal stem cells to function properly. The authors propose that it is due to their inability to transduce Wnt signals, and while this is plausible, I think there are few things that the paper should contain before this can be proposed firmly:

Point #1

The mouse mutant is just described as 'KO', referring to the previous work by the authors. The cited work simply states that this is a zygotic deletion of exon 3, which somehow leads to a decrease in protein abundance that is almost total in the lung but not so clear in the uterus. Exon 3 happens to be 72 bp long [https://www.ncbi.nlm.nih.gov/nuccore/NM_133738], so its deletion (assuming there are no cryptic splicing sites used) leads to an internal in-frame deletion of 24 amino acids. So, at best, this 'KO' is not a null, but a hypomorphic allele of context-dependent strength. Unfortunately, neither the previous work nor this paper (unless I have missed it!) contains information provided about the expression levels of Cmg2 in the intestine of KO mice - nor which cell types usually express it (see below). I think that using anti Cmg2 in WB and immunohistofluorescence of with ISC markers with intestine homogenate/sections of wild-type and mutant mice would be necessary to set the stage for the rest of the work.

Point #2

Connected to the previous point, the expression pattern of Cmg2 in the intestine is not described. Maybe this is already established in the literature, but the authors do not refer to the data. This is important when considering that the previous work of the authors suggests that Cmg2 might contribute to Wnt signalling transduction through physical, cis interactions with the Wnt co-receptor LRP6. Therefore, one would expect that Cmg2 would be cell-autonomously required in the intestinal stem cells.

Point #3

The authors establish that the regenerating crypts of Cmg2[KO] mice are unable to transduce Wnt signalling, but it is not clear whether this situation is provoked by the DSS-induce injury or existed all along. Can Cmg2[KO] intestinal stem cells transduce Wnt signalling before the DSS challenge? If they were, it might suggest that the 'context-dependence' of the Cmg2 role in Wnt signalling is contextual not only because of the tissue, but because of the history of the tissue or its present structure. It would also suggest that Cmg2 mutant mice, unless reared in a germ-free facility for life, would eventually lose intestinal homeostasis, and maybe suggest the level of intervention/monitoring that HFS patients would require. It might also provide an explanation in case Cmg2 was not expressed in ISCs - if the state of the tissue was as important as the presence of the protein, then the effect on Wnt transduction could be indirect and therefore it might not be required cell-autonomously.

I think points 1 and 2 are absolutely fundamental in a reverse genetics investigation. Point 3 would be nice to know but the outcome would not change the tenet of the paper. I believe that the work needed to deal these points can be performed on archival material. I do not think the mechanism proposed can be taken from 'plausible' to 'proven' without proposing substantial additional investigation, so I will not suggest any of it, as it could well be another paper.

A few minor points picked along the way:

1. Figure 1 legend says "In (c), results are mean {plus minus} SEM" - this seems applicable to (d) as (c) does not show error whiskers.
2. Figure 1 legend says "(d) Body weight loss, (f) the aspect of the feces and presence of occult blood were monitored and used for the (e) DAI. Results are mean {plus minus} SEM. Each dot represents the mean of n = 12 mice per genotype". This part looks like has suffered some rearrangement of words. The first instance of (f) should be (e), I guess, and I am not sure what "(e) DAI" means. And for (e), "mean {plus minus} SEM" does not seem applicable. This needs some light revision.
3. Figure 1H legend does not say which statistical test was made in the survival experiment in (h) - presumably log-rank? A further comment on the survival statistics: euthanised animals should not be counted towards true mortality when that is what is recorded as an 'event'. They should be right-censored. However, in this case, reaching the euthanasia criterion is just as good an indicator of health as mortality itself. So, simply by changing the Y axis from 'survival' to 'event-free survival' (or something to that effect), where 'events' are either death or reaching the euthanasia criterion, leaves the analysis as it is, and authors do not need to clarify that figure 1H shows "apparent mortality", as it is straightforward "complication-free survival" (just not entirely orthogonal to weight loss).
4. Some density measurements are made unnecessarily on arbitrary units (per field of view) - this should be simple to report in absolute measures (i.e. area of tissue screened or, better still, length of epithelium screened).
5. Figure 2E should read "percent involvement"
6. Figure 2J should read "lipocalin..."
7. In section "CMG2 Is Dispensable for YAP/TAZ-Mediated Reprogramming to Fetal-Like Stem Cells", the authors write ""We measured the mRNA levels of two additional YAP target genes, Cyr61 and CTGF...". I presume the "additional" is because Ly6a is also a target of YAP/TAZ, but if the reader does not know, it is puzzling. I would suggest to make this link explicit.
8. In Figures S2, 3 and S3, I think that the measures expressed as "% of homeostatic X in WT" really mean "% of average homeostatic X in WT". This should be made clear somewhere.
9. In panel C, the nature of the data is not entirely clear. First, the corresponding part of the legend says "Representative images of n=4 mice per genotype" which I presume should refer to panel B. Then, the graph plots 4 data points, which suggests that they correspond to 4 mice - but how many fields of view? Also, the violin plot outline is not described - I presume it captures all the data points from the coarse-grained pixel analysis, but it should be clarified.
10. In Figure 3H and 3I, I would suggest to add the 7+3 timepoint where the data come from.
11. In section "CMG2 Is Critical for Restoring the Lgr5+ Intestinal Stem Cell Pool", the authors say "...The mRNA levels of ... LRP6, β-catenin (Fig. S3a-b), and Wnt ligands (Wnt5a, 5b, and 2b) were comparable between the colons of Cmg2WT and Cmg2KO mice (Fig. S3c)..." without clarifying in which context - one needs to read the figure legend to realise this is "timepoint 7+3". I suggest to add "in the recovery phase" or "in regenerating colons" or something shorter, just to guide the reader.
12. Like with the previous point, it is not clear when the immunohistofluorescence of B-catenin is made - not even in the legend, as far as I could see. The only hint is that authors say "the nuclei of cells in the atrophic crypts of Cmg2KO..." with 'atrophic' probably indicating again the 7+3 timepoint.
13. A typo in the discussion: tunning for tuning.
14. In the discussion, the authors talk about the 'CMG2' protein (all caps - formatting convention for human proteins) but before they were referring to 'Cmg2' (formatting convention for mouse proteins). That is fine but some of the statements where "CMG2" is used clearly refer to observations made in the mouse.
15. Typos in methods: "antigen retrieval by treating [with] Proteinase K"; "Image acquisition and analyze [analysis]"; "All details regarding code[s] used for immunofluorescence analysis"

Referees cross-commenting

*this session contains comments from ALL the reviewers"

Rev2

Points 1 and 2 made by Referee 1 (and point 4 of Referee 3) appear most reasonable, and if not already done should be.

I also noted the more severe morphology of DSS damaged epithelium shown in Fig 2a noted by Referee 3 - and this I agree is a confounding factor. But overall, multiple lines of evidence were assembled to show that the KO mice and WT mice suffered DSS-induced colitis with equal severity - and with closely equal severity of damage to the intestinal epithelium (though the image in Fig 2a is disturbing). For my part, the concern is understandable but likely not operating in a confounding way. And the evidence for the reprogramming of the damaged epithelium into "fetal-like stem cells" (the 1st step in restitution of lost stem cells) occurs in both WT and KO mice - and these data are strong. For this reader, the block convincingly shows up for KO mouse at the WNT dependent step

Rev 3 This reviewer remains sceptical. I agree the authors performed the experiment well to confirm that DSS dosing was as equivalent as possible across the study. But DSS acts to induce colitis because it is concentrated in the colonic lumen as water is absorbed. Also ECM responses and remodelling are a central part of colitis models. And my concern is that the actual exposure in the KO group is influenced by transit of faeces/DSS is secondary to the known action of CMG2 on collagen deposition. The consequence of this being a protracted damage phase in which a restoration of adult stem cells would not be expected and leading to epithelial failure.

However, we differ. I might propose that the authors are asked to investigate and confirm expression of CMG2 in the epithelium and to repeat the analysis of collagen levels they performed on untreated CMG2 KO mice on colons from CMG2 KO mice having received DSS to see if these differ from controls.

Rev 1 Both reviewer #2 and reviewer #3 make relevant points, from the point of view of extracting as much biological knowledge as we can from the observations reported in the manuscript.

Reviewer #2 suggestion to use Cmg2[KO] organoids to investigate the dependence of Wnt transduction on Cmg2 is the type of experiments I refrained to propose. However, I think the "skeleton" of the mechanism is there and is reasonably solid. Fleshing it out may well be another paper.

I agree with Reviewer #3 objections to the timing and severity of the DSS damage. However, I am not sure how much they invalidate the main tenet of the paper:

• DSS may affect Cmg2[KO] more severely, but the overall disease score is comparable during the DSS treatment. If this severity was enough to be the main driver of the phenotype, it should have left a mark in the Histological and Disease activity scores. In this regard, I think it would be helpful if the authors provided an expanded version of Figure 2A with examples of the different levels of "Crypt damage" scored, and the proportions for each. This could be in the supplementary material and would balance the impressions induced by a single image.

• If DSS affected the recovery, this would also be compatible with having a more severe histological phenotype (which is not shown overall, just in Fig 2A) because one would also expect the tissue to attempt regeneration during the 7 days of DSS treatment.

• The only objection that I find difficult to argue is the effective duration of the treatment. If indeed peristalsis is affected, it may be that during the 'recovery' phase there is still DSS in the intestine. This could be perhaps verified using a DS detection assay (e.g. https://arxiv.org/pdf/1703.08663) on the intestinal contents or the faeces of the mice during the 3-day recovery period.

I think of what the aim of scholarly publication is, with this paper, and I find myself going back to a statement of the authors' discussion - that this work suggests that infants risking death may be offered (compassionate, I guess) IBD treatment. What does this hinge upon? I think, on the basic observation that diarrhoea (in the mouse model) is not intrinsic but caused by an inflammation-promoting insult. Is this substantiated? I think it is. Could we learn more biology from this disease model, about Wnt and about how ECM affects tissue regeneration? Certainly. Can this learning wait? I believe it can.

Significance

In this work, Bracq and colleagues provide clear evidence that the persistent diarrhoea seen in a mouse model of Hyaline Fibromatosis Syndrome is related to the inability of their intestinal epithelium to properly regenerate. This is very clear and of immediate impact. For instance, the authors themselves point at the possibility of applying treatments for Inflammatory Bowel Disease to HFS patients. While what happens in a mouse model is not necessarily the same as in human patients, the fact that persistent diarrhoea is a life-threatening symptom in HFS make this proposal, at least in compassionate use of the therapies and until its efficacy is disproven, very plausible. This is a clear gap of knowledge that addresses an unmet medical need.

I find that the work shows clearly that HFS mouse model subjects have normal intestinal function until challenged with a standard chemically-induced colitis. Then, the histological and health deterioration of the HFS mouse model is clear in comparison with normal mice, which can regenerate appropriately. This is shown with a multiplicity of orthogonal techniques spanning molecular, histological and organismal, which are standard and very well reported in the paper.

The authors propose a specific cellular and molecular mechanism to explain the incapacity of the intestinal epithelium in the mouse model of HFS to regenerate. According to this mechanism, the protein Cmg2, whose mutation causes HFS in humans, would be necessary for intestinal stem cells to transduce the signal of Wnt ligands and therefore support their behaviour as regenerative cells. This mechanism is plausible, but more basic and advanced work would be needed to take it as proven.

This work would be of interest to both the clinical, biomedical, and basic research communities interested in rare diseases, the gastrointestinal system, collagen and extracellular matrix, and Wnt signalling.

My general expertise is in developmental and stem cell biology using reverse genetics, transgenesis and immunohistological and molecular methods of data production, and lineage tracing, digital imaging and bioinformatic analytical methods; I work with Drosophila melanogaster and its adult gastrointestinal system.

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

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 the manuscript entitled " The hepatitis E virus capsid protein ORF2 counteracts cell-intrinsic antiviral responses to enable persistence in hepatocytes ", Ann-Kathrin Mehnert interrogated that HEV pORF2 can inhibit host antiviral response. They found interaction of HEV ORF2 and TBK1. The finding is interesting and echoed with some previous studies that ORF2 can inhibit innate immunity.

The study could benefit from the consideration of some major and specific points, as indicated below:

Major issues:

1. The researchers used p6, a cell-adapted clone, which was isolated form a chronic HEV patient. As previous studies suggested, p6 may behave differently than wild-type strains. Did the authors tried other HEV strains, as they used ips-induced model that was reported supportive to wild-type HEV?
2. Figure 1F, ORF2 can interact with TBK1 as showed. But the prediction from Alphafold is weak. Also, could the author more evidence than the co-IP?
3. Figure 2C and 2D, at 5 dpi, one can observed a stronger antiviral response, but at 7 dpi, no obvious difference was observed. Could the authors comment on this? 4．Figure 2H and 2I, detailed description of how the authors measured the positive cells should be provided. Did the authors selected whole plate of cells for counting? As showed in Figure 2H, the signals of IF were stronger at 5 and 7 dpi when compared at 3 dpi, but why the proportion of positive cells was reduced in Figure 2I?
4. The study emphasized the function of ORF2 on HEV "persistence". However, this cannot be fully supported by cell models. In future, study on chronic HEV infection animal models may be conducted.
5. The authors study ORF2 in whole. It will be of benefit to the readers that the authors could specified the function of secreted ORF2 and ORF2 capsid in the current study.

Minor issues:

1. Figure 3A, this is an elegant design. More data may provide for the validation of the formation of the virions.
2. Figure 1, data should be provided for the successful expression of HEV-1 or HEV-3 ORF2, and ORF3.
3. line 219, the current evidence that supported this statement is weak, especially for ORF2.
4. Suppl Figure 3F-3H, statistical analysis is needed
5. Suppl Figure 3F-3H, it seems that when no treatment was admistrated, the level of ISG15 in ΔORF2 group was higher than those of the WT and ΔORF3 group. Could the authors comment?
6. Figure 3D and 3E, the starting time of the detection is not aligned.
7. Figure 3F, scale bar is missing.
8. In M&M, statistical method should be provided with more details and cover all the experiments used.

Significance

In the manuscript entitled " The hepatitis E virus capsid protein ORF2 counteracts cell-intrinsic antiviral responses to enable persistence in hepatocytes ", Ann-Kathrin Mehnert interrogated that HEV pORF2 can inhibit host antiviral response. They found interaction of HEV ORF2 and TBK1. The finding is interesting and echoed with some previous studies that ORF2 can inhibit innate immunity.

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

Summary: Authors described the protective mechanism mediated by ORF2 that protects viral replication from the antiviral responses. They have utilized the advanced single-cell RNA sequencing to decipher the dampened antiviral responses in the presence of ORF2 HEV. I believe the study is important for the HEV literature and believe that the manuscript can be considered for publication after authors (1) rewrites the results and discussions separately until the journal wants it to be together. (2) answer the below questions.

Minor comments:

Line 69, 71 - I have never seen in any paper including reference in this way!

Line 72 and 73 - missing reference!

Line 92, 93 - missing reference!

Line 95 to 99 - missing references!

Major comments:

I would like the authors to answer few questions: 1. Did the authors study only the P6 HEV genome? Have they done anything comparative with the other strain to understand if the proposed mechanism is not the strain specific? 2. Can the authors explain why we do not see any band in the Fig. 1F B-actin?

Significance

The paper uses advanced technique as single cell RNA seq to understand the mechanism of ORF2 assisting in the HEV replication.

The study is well designed.

This study will add up to understand some of the persistence infection seen in solid organ transplant patients. This study gives a mechanistic overview of HEV avoidance of antiviral response.

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

In this study, the authors investigated how HEV ORF2 interferes with host antiviral responses to sustain viral infection. They employed several models, including pattern recognition receptors (PRRs) KO cell lines, immunodeficient cells and stem cell-derived models, to prove that: 1) ORF2 is essential for viral replication and 2) ORF2 dampens the interferon and inflammatory signaling pathways. They confirmed the interaction between ORF2 and TBK1, a central mediator of innate immune responses and identified residues in ORF2 that affect its interaction with TBK1. Finally, through single cell RNAseq, they demonstrated that ORF2 is a viral antagonist that inhibits host ISG expression in both infected and bystander cells. Interestingly, the sets of genes that are upregulated in WT vs ORF2-deficient virus infected cells are not entirely identical, suggesting that ORF2 may also modulate host gene expression in addition to suppressing the immune response. This research confers new immune antagonism mechanisms mediated by HEV capsid for sustainable HEV replication in host cells and provides potential therapeutic targets for HEV treatments.

Major comments:

1. The authors conclude from Figure 1 that the HEV ORF2 protein antagonizes both antiviral and inflammatory signaling pathways. The authors comprehensively investigated PRRs-mediated activation of type I interferon by viruses or poly(I:C) through overexpression of MDA5, RIG-I and TLR3. However, they only investigated the impact of ORF2 on host inflammatory response through evaluating the levels of TNFAIP3 RNA in the presence of MDA5 overexpression. It would be informative if the authors also check for NFkB activation/phosphorylation and expression of classical pro-inflammatory cytokines such as IL-1b and IL6. Interestingly, changes in IFNB secretion after ORF2 overexpression appear more dramatic compared to changes in IFNB1 RNA levels (compare Figure 1A-C with Supplementary Figure 1A and C). Are the IFN-beta protein expression changes statistically significant in Supplementary Figure 1?
2. Changes in the IFN response do not always translate into changes in the viral RNA levels. In Figure 2B-D, the authors attributed the higher induction of IFNL1 and ISG15 on day 5 to the absence of ORF2 inhibition. However, the expression of these two genes drops to the same levels as the ones in WT viral RNA-electroporated cells on day 7, which is strange as ORF2-deficient viral RNA levels continue to be inhibited on day 7. This is different from the stem cell derived hepatocytes infected with the trans-complementation viruses in Figure 3G-H where there are significant differences in ISG15 levels between WT and ORF2-deficient virus infected cells on both days 5 and 7. To support their hypothesis, the authors need to further confirm the sudden upregulated antiviral activity on day 5 in electroporated HepG2/C3A cells by testing JAK/STAT phosphorylation and type I interferon secretion.
3. The authors used different hepatocyte systems coupled with viral RNA electroporation or trans-complementation virus infection to investigate ORF2-mediated interference of the IFN pathway, which is highly complementary. However, while the electroporation of viral RNA into HepG2/C3A (Figure 2B-D) and infection of stem cell-derived hepatocytes with trans-complementation viruses (Figure 3F-H) result in similar upregulation of ISG expression on day 5, that wasn't observed in HepG2/C3A cells infected with trans-complementation viruses (Figure 3C-E) on day 5. The authors need to discuss the discrepancy among these different systems. Since the ORF2-deficient trans-complementation virus still brings in ORF2 proteins from the producer cells but cannot generate new ORF2 proteins, do ORF2 proteins from these two different sources have different functions in different hepatocyte systems? In addition, other than the data points that are shown to be not significantly different in Figure 3D-E, are any of the other data points significantly different?
4. The single cell RNAseq data are very informative and revealed two interesting groups of genes. First, the ISGs that are further induced in the cells infected with ORF2-deficient HEV compared to cells infected with WT HEV (Figure 4N) are likely suppressed by ORF2. Second, the ISGs that are uniquely induced in the absence of ORF2 are different from the genes that are uniquely induced by WT HEV (Supplementary Table 2), suggesting that ORF2 may also modulate host gene expression. The authors can further characterize these two groups of ISGs by performing gene knockdown or knockout and investigating whether ORF2 directly interacts with these ISG products to determine the functional consequences of their upregulation. Related to that, are there other gene expression changes beyond ISG signatures which would suggest that ORF2 can regulate host gene expression? Figure 4A-C only shows comparisons for WT or ORF2-deficient vs. uninfected cells. The authors can perform GO and KEGG analyses to see if certain biological processes/pathways are enriched among the WT vs ORF2-deficient HEV induced genes. Further characterization of these genes (ISGs or not) would shed light on the novel roles of ORF2 in both immune antagonism and gene regulation and greatly increase the significance of the study.
5. In Supplementary Figure 3F-H, the authors used BX795 to inhibit TBK1 (a target of ORF2) and found decreases in IFNL1 and ISG15 expression whether cells are electroporated with WT, ORF2-deficient, or ORF3-deficient viral RNA. However, this does not correlate with the data in Figure 2E-G where TBK1 inhibition results in significant differences in viral RNA levels only in the absence of ORF2 or ORF3. These results would suggest that the effects of TBK1 inhibition on viral RNA levels is independent of changes in the IFN/ISG expression levels.

Significance

The study addresses a long-standing question in the field about the immune antagonism activities of HEV ORF2 and ORF3 which previous studies have conflicting results on. The strength of this study is the use of complementary approaches such as ORF2 trans complementation system and single cell sequencing, and more relevant models such as stem cell derived hepatocytes to rigorously dissect the role of newly synthesized ORF2 protein in immunocompetent cell context. The manuscript is well written and would appeal to researchers in the HEV and innate immunity fields. However, the significance of the study is dampened by changes in the IFN response not always correlate with the inconsistency of ORF2-mediated inhibitory effects in different models and the still poorly defined mechanism of ORF2 suppression of the IFN pathway. The study would make conceptual advance if the authors can address the discrepancies in their findings and perform additional characterization to determine the functional consequences of ORF2-mediated immune suppression and gene regulation.

My expertise is in innate immunity and host-virus interactions.

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

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 (Evidence, reproducibility and clarity (Required)): __ Summary In this work, the authors present a careful study of the lattice of the indirect flight muscle (IFM) in Drosophila using data from a morphometric analysis. To this end, an automated tool is developed for precise, high-throughput measurements of sarcomere length and myofibril width, and various microscopy techniques are used to assess sub-sarcomeric structures. These methods are applied to analyze sarcomere structure at multiple stages in the process of myofibrillogenesis. In addition, the authors present various factors and experimental methods that may affect the accurate measurement of IFM structures. Although the comprehensive structural study is appreciated, there are major issues with the presentation/scope of the work that need to be addressed: Major Comments 1. The main weakness of the paper is in its claim of presenting a model of the sarcomere. Indeed, the paper reports a structural study that is drawn onto a 3D schematic. There is no myofibrillogenesis model that would provide insights into mechanisms. Therefore, the use of the word model is grossly overstated.

In biology, the term “model” is used in various contexts, but it generally refers to a simplified representation of a biological system, a structure or a process. Accordingly, we consider “model” the most fitting phrase for what we present in Figure 4 (Figure 7 in the revised manuscript). These are not arbitrary 3D schematics; they are scaled representations in which the length, the number and the relative three-dimensional arrangement of thin and thick filaments are based on measurements. These measurements are primarily based on our own data (presented in the main text and provided in the supplementary materials), as published data were either lacking or inconsistent. Moreover, we would like to highlight that we do not claim to present a conceptual or mechanistic model of myofibrillogenesis, but we do present structural reconstructions or models for four developmental time points. Therefore, we disagree with the remark that “the use of the word model is grossly overstated”, as our wording fully corresponds to the common sense.

In general, the major focus and contribution of the work is unclear. How does the comprehensive nature of the measurements contribute to existing literature?

We significantly revised the text to highlight the main points more firmly, and added an additional section to help non-specialist readers to better understand our aims and findings.

Figure labels are often rather confusing - for example it is unclear why there is a B, B', B' etc instead of B,C,D, etc.

The figure labels have been revised in accordance with the reviewer’s recommendation.

Some comments in the text are not clearly tied to the figures. For example, in lines 108-109, are the authors referring to the shadow along the edges of the myofibril when saying they are not clearly defined (Figure 1C)?

The lines refer to the fact that identifying the boundary of an “object” in a fluorescence microscopy image is inherently challenging - even under ideal conditions where the object’s image is not affected by nearby signals or background noise. To improve clarity, we revised this section and now it reads: The other key parameter - myofibril diameter - is typically measured using phalloidin staining. However, accurately delineating their boundaries in micrographs is difficult - even under optimal conditions (high signal‑to‑noise ratio, no overlapping fibers, etc.; Fig. 1C). This limitation arises from the fundamental nature of light microscopy as the image produced is a blurred version of the actual structure, due to convolution with the microscope’s point spread function.

In line 116, it is unclear what "surrounding structures" the authors are referring to if the myofibrils are isolated.

We revised the text for clarity. It now states: Once isolated, myofibrils lie flat on the coverslip, aligning with the focal plane of the objective lens. This orientation allows for high-resolution, undistorted imaging and accurate two-dimensional measurements, free from interference by neighboring biological structures (e.g.: other myofibrils).

In lines 141-142, there is no reference of data to back up the claim of validation.

We addressed this mistake by including a reference to Fig. S1E (Fig. S1D in the revised manuscript).

In line 170, the authors mention the mef2-Gal4/+ strain as a Gal4 driver line but do not clearly state how this strain is different from the wildtypes or how this impacts their results.

Mef2-Gal4 is a muscle-specific Gal4 driver, often used in Drosophila muscle studies. It is a convention between Drosophila geneticists that presence of a transgene (i.e. Mef2-Gal4) changes the genetic background, and although it does not necessariliy cause any phenotypic effect, it is clearly distinguished from the wild type situation, and whenever relevant, Mef2-Gal4/+ is the preferred choice (if not the correct choice) as a control instead of wild type. As clear from our data, presence of the Mef2-Gal4 driver line does not affect the length or width of IFM sarcomeres as compared to wild type.

In lines 182-185, the authors discuss the effects of tissue embedding on morphometrics. Were factors such as animal sex, age, fiber type, etc. conserved in these experiments? If not, any differences in results may be confounding.

We fully agree with the reviewer that when testing the effect of a single variable, all other variables should remain constant. This is actually one of the main points emphasized in the results section. Additionally, this information is already provided in the Source Data files for each panel.

In lines 199-201, the authors discuss results of myofibril diameter using different preparation methods, yet no data is cited to support the claims. In line 220, the phrase "6 independent experiments" is unclear. Is each independent experiment performed using a different animal? Furthermore, are 6 experiments performed for each time point?

We substantially revised the relevant paragraphs and ensured that the corresponding data (Figure 2A in the revised manuscript) is cited each time when it is discussed. We conducted six independent experiments at each time point. This is consistently indicated in the figures and can be verified in the SourceData files (specifically, Fig3SourceData in this case). To clarify what we mean by "independent experiments," we added the following sentence to the Methods section: Experiments were considered independent when specimens came from different parental crosses, and each experiment included approximately six animals to capture individual variability.

In line 254, the authors refer to "number of sarcomeres". It must be clearly stated if this refers to sarcomeres per myofibril, image area, etc.

It is now clearly stated as: "number of sarcomeres per myofibril".

In line 274, the authors refer to "myofilament number". It must be clearly stated if this refers to myofilaments per myofibril, image area, etc.

We counted the number of myofilaments in developing myofibrils, and this is now clearly stated in the text and in the legend of Figure 3 (Figure 4 in the revised manuscript).

In line 299, the authors mention that thin filaments measured less than 560 nm in length, yet no data is cited to support this.

The previously missing reference to Figure 4 (Figure 7 in the revised manuscript) has now been added in addition to the revised Supplementary Figure 5.

In the "Quantifying sarcomere growth dynamics" section of the summary (starting from line 402) the authors introduce data that would be more naturally placed in the results and discussion section.

As suggested by the reviewer, we incorporated the key aspects of sarcomere growth dynamics into the Results and Discussion section.

In lines 422-423, it is not mentioned what the controls are for.

This was already explained in the main text between lines 167 and 173.

In the caption of Figure 1C, it is not mentioned what the red dashed lines in the microscope images represent.

The caption has been updated to include the following clarification: The red dashed lines border the ROI used for generating the intensity profiles.

In the caption of Figure 1D, the difference between the lighter and darker grey points is not mentioned.

This was already explained in each relevant figure legend. In this specific case, it is stated between lines 850 and 852: “Light gray dots represent individual measurements of sarcomere length and myofibril diameter, while the larger dots indicate the mean values from independent experiments.”

In line 849, the stated p-value (0.003) does not match that mentioned in the figure (0.0003).

We thank the reviewer for noticing this small mistake; correction was made to display the accurate p-value of 0.0003 at both places.

In line 874, it is not clear what an "independent experiment" refers to (different animal, etc.?).

We refer the reviewer to point 9, where this question has already been addressed.

Figure 2A is hard to read. Using different colored dots for different time points might help.

As suggested by the reviewer, we generated a plot with the individual points color-coded by time.

The significant figures presented in Figure 4 give a completely inaccurate representation of the variability of the measurements achieved with these techniques.

Certainly, each measured parameter exhibits inherent biological and technical variability. We have made all the raw data available to the reader through the SourceData files, and this variability is also evident in Figures 1, 2, 3, Supplementary Figure 1, 3, and 5 (Figure 1, 2, 3, 4, 6, and Supplementary Figure 1 in the revised manuscript). Also we have included an additional plot (Supplementary Figure 5 in the revised manuscript) that presents the calculated thin and thick filament lengths and their uncertainty. However, in Figure 4 (Figure 7 in the revised manuscript), our goal was to present an easily understandable visual representation of the sarcomeric structures for each time point, based on the averages of the relevant measurements.

In line 877, it should be mentioned that the number of filaments is counted per myofibril. The y-axes in the figure should also be adjusted to clarify this.

As suggested by the reviewer, both the figure legend and the plot have been updated to clearly indicate that the filament count refers to the number per myofibril.

In line 883, it is not clear what an "independent experiment" refers to (different animal, etc.?).

We refer the reviewer to point 9, where this question has already been addressed.

The statement of sample sizes in all figures is a little confusing.

Following general guidelines, we used SuperPlots to effectively present the data, as nicely demonstrated in the JCB viewpoint article by Lord et al., 2020 (PMID: 32346721). Individual measurements are shown as pooled data points, allowing readers to appreciate the spread, distribution and number of measurements. Overlaid on these pooled dot plots are the mean values from each independent experiment, with error bars representing variability between independent experiments. Sample sizes are provided for both individual measurements and independent experiments. This is now clearly explained in the Materials and Methods section, and we corrected the legends to improve clarity (“n” indicates the number of independent experiments/individual measurements).

In lines 1007-1008, the authors imply that the lattice model is needed for calculation of myofilament length. However, from the equations and previous data, it seems that this can be estimated using the confocal and dSTORM images.

As the reviewer correctly noted, myofilament length can be estimated using measurements from confocal and dSTORM images, following the equations provided. However, constructing even a simplified model requires multiple constraints to be defined and applied in a specific order. In practice, one must first determine the number and arrangement of myofilaments in a cross-sectional view of an “average sarcomere” before attempting to build a longitudinal model, where length calculations become relevant. This is now clarified in the text.

A more specific discussion of future directions is needed to put this paper in context. For example: Can anything from the overall process be used to better understand sarcomere dynamics in larger animals/humans? Can this be applied to disease modelling?

To address these questions, we have added a section titled STUDY LIMITATIONS, which states: “Our study is focused on describing the growth of IFM sarcomeres during myofibrillogenesis at the level of individual myofilaments. Additionally, we developed a user-friendly software tool for precise sarcomere size measurements and demonstrate that these measurements are sensitive to varying conditions. Whereas, this tool can be used successfully on whole muscle fiber preparations as well, our pipeline was intentionally optimized for individual IFM myofibrils ensuring higher measurement precision in our hands than other type of preparations. Thus, we predict that future work will be required to extend it to sarcomeres from other muscle tissues or species. Nevertheless, our study exemplifies a workflow how to measure sarcomere dimensions precisely. With some variations, it should be possible to adopt it for other muscles, including vertebrate and human striated muscles. To facilitate this and to enhance the accessibility and usability of this dataset, we welcome any feedback and suggestions from researchers in the field.”

One of the major claims of the paper is that there is a measurable variability with sex and other parameters. However, this data is never clearly summarized, presented (except for supplement), or discussed for its implications.

We followed the suggestion of the reviewer, and we moved this supplementary data into a main figure, and thoroughly revised the corresponding paragraphs to present and discuss the findings more clearly.

Minor Comments: 1. Lines 60-65 seem to break the flow of the introduction. As the authors discuss existing methods in literature for IFM analysis in the previous couple sentences, the following sentences should clearly state the limitations of existing methods/current gap in literature and a general idea of what the current work is contributing.

We agree with this remark, and we substantially revised the Introduction to clearly define the existing gap in the literature and to articulate how our work addresses this gap.

In line 104, the acronym for ZASPs is not spelled out.

The acronym has now been spelled out for clarity.

**Referee Cross-commenting**

I agree as well.

Reviewer #1 (Significance (Required)):

In summary, this paper provides a multi-scale characterization of Drosophila flight muscle sarcomere structure under a variety of conditions, which is potentially a significant contribution for the field. However, the paper scope is overstated in that it does not provide an actual sarcomere model. Further, there are multiple issues with data presentation that impact the readability of the manuscript.

Although it is somewhat unclear what would be “an actual sarcomere model” for the reviewer, but we cannot accept that we made on overstatement by using the word “model”, because one of the main outcomes of our work are indeed the myofilament level sarcomere models depicted in Figure 4 (Figure 7 in the revised manuscript). As said above, we do not claim that these would be molecular models, or mechanistic models or developmental models, but it makes absolutely nonsense (even in common terms!) that our scaled graphical representations (based on a wealth of measurements) should not be or cannot be called models.

As to the comment with data presentation, we thank the reviewer for the numerous suggestions, and we substantially revised the manuscript to increase clarity and overall readability.

__Reviewer #2 (Evidence, reproducibility and clarity (Required)): __ Summary: In this manuscript titled "A myofilament lattice model of Drosophila flight muscle sarcomeres based on multiscale morphometric analysis during development," Görög et al. perform a detailed analysis of morphological parameters of the indirect flight muscle (IFM) of D. melanogaster. The authors start by illustrating the range of measurements reported in the literature for mature IFM sarcomere length and width, showing a need to revisit and determine a standardized measurement. They develop a new Python-based tool, IMA, to analyze sarcomere lengths from confocal micrographs of isolated myofibrils stained with phalloidin and a z-disc marker. Using this tool, they demonstrate that sample preparation (especially mounting medium), as well as fiber type, sex, and age influence sarcomere measurements. Combining IMA, TEM, and STORM data, they measure sarcomere parameters across development, providing a comprehensive and up-to-date set of "standardized" sarcomere measurements. Using these data, they generate a model integrating all of the parameters to model sarcomeres at four discrete timepoints of development, recapitulating key phases of sarcomere formation and growth.

Major comments: Line 200 & 901 - Figure S1B - The authors make a strong statement about the use of liquid versus hardening media, and it is clear from the image provided in Figure S1 that there is a difference in the apparent sarcomere width. The identity of the "liquid media" versus the "hardening media" should be clearly identified in the Results, in addition to the legend for Figure S1. The authors show that "glycerol-based solutions" increase sarcomere width, but the Materials only list 90% glycerol and PBS. However, a frequently used liquid mounting media is Vectashield. Based on the literature, measurements in liquid Vectashield show diameters significantly less than 2.2 microns observed here with presumably 90% glycerol or PBS. Can the authors qualify this statement, or provide data that all forms of liquid mounting media cause this effect? Does this also apply to hemi-thorax and sectioned preparations, or just isolated myofibrils?

We used a PBS-based solution containing 90% glycerol as our liquid medium, as now stated in the main text. In response to the reviewer’s suggestion, we also tested a non-hardening version of Vectashield (H-1000). Myofibrils in Vectashield were significantly thicker than those in ProLong Gold but still thinner than those in the 90% glycerol–PBS solution, shown in Figure 2B. The mechanisms that could potentially explain these observations have been described in several studies (Miller et al., 2008; Tanner et al., 2011, 2012). Briefly, IFM is a densely packed macromolecular assembly. Upon removal of the cell membrane, myofibrillar proteins attract water, leading to overhydration of the myofilament lattice. This increases the spacing between filaments, resulting in an expansion of overall myofibril diameter. The extent of hydration depends on the osmolarity of the surrounding medium, as the system eventually reaches osmotic equilibrium. While both liquid media induced significant swelling, the observed differences likely reflect variations in their osmotic properties. In contrast, dehydration - an essential step in electron microscopy sample preparation - reduces the spacing between filaments, making myofibrils appear thinner. This explains why EM micrographs consistently show significantly smaller myofibril diameters (Chakravorty et al., 2017).

        Hardening media such as ProLong Gold introduce additional artifacts: during polymerization, these media shrink, exerting compressive forces on the tissue (Jonkman et al., 2020). We therefore propose that isolated myofibrils first expand due to overhydration in the dissection solution, and are then compressed back toward their *in vivo* dimensions during incubation in ProLong Gold. The average *in vivo* diameter of IFM myofibrils can be estimated without direct measurements, as it is determined by two key factors: (i) the number of myofilaments, which has been quantified in EM cross-sections in several studies (Fernandes & Schöck, 2014; Shwartz et al., 2016; Chakravorty et al., 2017) including our own, and (ii) the spacing between filaments, which can be measured by X-ray diffraction even in live *Drosophila* or under various experimental conditions (Irving & Maughan, 2000; Miller et al., 2008; Tanner et al., 2011, 2012). Our findings suggest that the effects of lattice overhydration and media-induced shrinkage are most pronounced in isolated myofibrils. In larger tissue preparations, the inter-myofibrillar space likely acts as a mechanical and osmotic buffer, reducing the extent of such distortions

Can the authors comment on whether the length of fixation or fixation buffer solution, in addition to the mounting medium, make a difference on sarcomere length and diameter measurements? This is another source of variation in published protocols.

The effect of fixation time on sarcomere morphometrics in whole-mount IFM preparations has been previously demonstrated by DeAguero et al. (2019), as briefly noted in our manuscript. To extend these findings, we performed a comparison using isolated myofibrils, assessing morphometric parameters after fixation for 10, 20 (standard) and 60 minutes. We found no difference between the 10- and 20-minute fixation conditions; however, fixation for 60 minutes resulted in significantly increased myofibril diameter (and these data are now shown in Supplementary Figure 1C). A comparable increase in thickness was also observed when using a glutaraldehyde-based fixative. These results suggest that more extensively fixed myofibrils may better resist the compressive forces exerted by hardening media.

Line 237-238. The authors conclude that premyofibrils are much thinner than previously measured. The use of Airyscan to more accurately measure myofibril width at this timepoint is a good contribution, as indeed diffraction and light scatter likely contribute to increased width measured in light microscopy images. I also wonder, though, how well the IMP software performs in measuring width at 36h APF, given how irregular the isolated myofibrils at this stage look (wide z-lines but thinner and weaker H and I bands as shown in Fig. 2B)?

The reviewer is correct that measurements during the early stages of myofibrillogenesis require additional effort. However, in addition to its automatic mode, IMA can also operate in semi-automatic or manual modes, ensuring complete control over the measurements. Myofibril width is determined from the phalloidin channel at the Z-line (as described in the software’s User Guide and Supplementary Figure 2), where it is at its thickest.

Also, how much of the difference in sarcomere width arises due to effects of "stripping" components off of the sarcomere at the earliest timepoint (for example alpha-actinin or Zasp proteins)?

A comparison between isolated myofibrils and those from microdissected muscles (Supplementary Figure 3B, Figure 3C in the revised manuscript) shows that the isolation process does not alter the morphometric measurements of sarcomeres. Moreover, the measured myofibril width aligns well with what we expect based on the number of myofilaments observed in TEM cross-sections of myofibrils at 36 hours APF (Figure 3A, now Figure 4A in the revised manuscript), supporting the consistency of our model.

Myofibrils at early timepoints do contain more than 4-12 sarcomeres in a line (they extend the full length of the myofiber), so it is possible they are breaking due to the detergent and mechanical disruption induced by the isolation method.

The reviewer is correct - myofibrils likely span the full length of the myofiber from the onset of myofibrillogenesis. However, during the isolation of individual myofibrils, they often break, and even mature myofibrils typically fragment into pieces of about 300 µm in length (illustrated in Figure 1E, now Figure 2A in the revised manuscript). Importantly, our measurements show that this fragmentation does not affect the assessed sarcomere length or width (as shown in Supplementary Figure 3B, now Figure 3C in the revised manuscript).

Line 312 - What does "stable association" mean in this context? The authors mention early timepoints lack stable association of alpha-Actinin or Zasp52, and they reference Fig. S4C, but this figure only shows 72h and 24 AE, not 36h and 48 h APF. Previous reports have seen localization of both alpha-Actinin and Zasp52, so presumably the detergent or mechanical isolation is stripping these components off of the isolated myofibrils up until 72h.

In agreement with previous reports, we also detected both α-Actinin (as shown in former Supplementary Figure 3B, now Figure 3C) and Zasp52 in microdissected IFM starting from 36 hours APF. However, these markers were largely absent from the isolated myofibrils of young pupae (36 to 60 hours APF). By 60 hours APF, strong α-Actinin and Zasp52 staining became evident in isolated myofibrils, whereas dTitin epitopes were clearly detectable from the earliest time point examined. This indicates that some proteins, such as α-Actinin and Zasp52, can be lost during the isolation process, whereas others like dTitin are retained and this differential sensitivity appears to depend on developmental stage. A likely explanation is that α-Actinin and Zasp52 are recruited early to Z-bodies but are only fully incorporated as more mature Z-disks form between 48 and 60 hours APF. This incomplete incorporation at the earlier stages could account for their loss during the isolation process. This interpretation is supported by our morphological analysis of the Z-discs, as shown in the dSTORM dataset (former Figure 3B, B’’, now Figure 4C, E) and in longitudinal TEM sections (former Supplementary Figure 5B, now in Figure 6B). Because α-Actinin and Zasp52 are not detected in isolated myofibrils at 36 and 48 hours APF, they are not included in Figure S4C (Figure 5C in the revised manuscript). This is explained in the updated figure legend.

This same type of issue comes up again in Lines 325-334, where the authors talk about 3E8 and MAC147. They state that 3E8 signal significantly declines in later stages and that MAC147 is not suitable to label myofibrils in young pupae, but they only show data from 72 APF and 24 AE (which looks to have decent staining for both 3E8 and MAC147). A clearer explanation here would be helpful.

To put it simply: we used one myosin antibody to label the A-band in the IFM of 36h APF and 48h APF animals, and a different antibody for the 72h APF and 24h AE stages. In more detail: Myosin 3E8 is a monoclonal antibody targeting the myosin heavy chain and labels the entire length of mature thick filaments except for the bare zone (former Supplementary Figure 4D, now in Figure 5D), suggesting its epitope is near the head domain. As a result, we expect a uniform A-band staining - excluding the bare zone - which is exactly what we observe in the IFM of young pupae (36h APF and 48h APF; formerly Figure 3B, now Figure 4C in the revised manuscript). However, at 72h APF and 24h AE, Myosin 3E8 produces a different staining pattern: two narrow stripes flanking the bare zone and two broader, more diffuse stripes near the A/I band junction (former Supplementary Figure 4D, now Figure 5D). This change is likely due to restricted antigen accessibility at these later developmental stages - a common issue in the densely packed IFM - making this antibody unsuitable for reliably measuring thick filament length in these stages.

MAC147 is another monoclonal antibody against Mhc that recognizes an epitope near the head domain. However, it only works reliably in more mature myofibrils (72h APF and 24h AE; formerly Figure 3B, now Figure 4C in the revised manuscript), likely due to its specificity for a particular Mhc isoform. This is why we do not include images from earlier developmental stages using this antibody. We added a revised, concise explanation in the main text for general readers, and provided a more detailed description for specialist readers in the legend of Supplementary Figure 4D (updated as Figure 5D in the revised manuscript).

Figure 3B. The authors show the H, Z, and I lengths in B', B', and B' and discuss these lengths in the text (lines 305-320). It would also be nice to actually have the plots showing the measured/calculated lengths for thin and thick filaments. These are mentioned in the results, but I cannot find the plots in the figures and there is no panel reference.

A summary table of the measured and calculated parameters is provided in Fig4SourceData (Fig7Source Data in the revised manuscript). However, following the reviewer’s suggestion, we also generated an additional plot (Supplementary Figure 5 in the revised manuscript) that displays the calculated thin and thick filament lengths.

Line 400. Does the model in Figure 4 actually have molecular resolution as the authors claim? From these views, thick and thin filaments appear to be represented by cylindrical objects. Localization of specific molecules would require further modeling with individual proteins. Or do the authors mean localization from STORM imaging relative to the ends of the thick and/or thin filaments? The model itself is a useful contribution, but based on Figure 4, resolution of individual molecules is not evident.

The reviewer is correct; and we fully agree that we do not present a molecular model of sarcomeres in this study - nor do we claim to. Instead we present a myofilament level model. Nevertheless, the scaled myofilament lattice model we introduce could serve as a geometric constraint when constructing supramolecular models of sarcomeres. As the reviewer rightly notes, implementing such an approach would require additional effort.

The main Results section of the text is condensed into 4 figures. However, I found myself flipping back and forth between the main figures and the supplement continuously, especially parts of Supplemental Figures 1, 3, 4, and 5. With such large amounts of detail in the Results relying on the supplement, it may be worth considering reorganizing the main and supplemental figures, and having 7 main figures, to include important panels that are currently in the supplement (esp. Fig S1B, S1C, S1D, S3B, S4, S5).

We found it a very useful suggestion, and we substantially reorganized the figures in the revised manuscript according to the recommendations of the reviewer.

Minor comments: On the plots in Fig. S1B, D, and F, it is hard to see the color of the dots because the red error bars are on top of them. Can the other distribution dots be tinted the correct color or the x-axis labels be added, so it is clear which dataset is which?

We significantly enlarged the dots to enhance visual clarity.

Line 142 needs a reference to Figure S1, Panel E, which shows the accuracy and precision measurements.

The requested panel reference has now been included in the revised manuscript.

Lines 198 - is this range from the above publications? Needs to be clearly cited.

The range has indeed been estimated using measurements from the aforementioned publications, and this point is now further clarified in the revised text.

Figure S3B is confusing - why do the blow-ups overlap both the top (presumably microdissected) and the bottom (presumably isolated) images? The identity of microdissected images should be labeled, as they are hard to see underneath of the blown-up images and the identity of individual image planes wasn't immediately obvious.

We refined the panel structure of Figure S3B (Figure 3C in the revised manuscript) to enhance clarity as the reviewer suggested.

Line 298. By "misaligned," do the authors mean the pointed ends are not uniformly anchored in the z-disc, leading to the wide z-disc measurements? At this early stage, I'm not sure "misaligned" is the right word - perhaps "were not yet aligned in register at the z-disc" or something similar.

We revised the text for clarity. It now reads: At 36 hours APF, thin filaments had not yet aligned in perfect register at the Z-disc, with most measuring less than 560 nm in length - and exhibiting considerable variability.

Figure S6 - spelling mistake in label of panel A, "sarcomer" should be "sarcomere"

The typo is corrected.

Line 487. Spelling "Zaps52" should be "Zasp52"

The typo is corrected.

Line 887. Spelling "Myofilement" should be "Myofilament"

The typo is corrected.

Line 946-947. In the legend for Supp. Fig. 3., the authors should specify which published datasets on sarcomere length are shown in the figure by including the references in the legend. Presumably the "isolated individual myofibrils" are the blue "this study" lines, leaving the "microdissected muscles" as the magenta "previous reports" on the figure. Without the reference, it is not clear if these are microdissected, isolated myofibrils, hemi-thorax sections, cryosections, or another preparation method for the "previous reports" data.

The references have now been added to both the figure and its legend.

**Referee Cross-commenting**

I agree with the comments from the other reviewers. Many of the major themes are consistent across the reviews, including regarding the model, preparation methods, and the software tool.

Reviewer #2 (Significance (Required)):

Strengths: This manuscript is an important contribution to the field of sarcomere development. The authors use modern technologies to revisit variation in morphometric measurements in the literature, and they identify parameters that influence this variation. Notably, sex-specific differences, DLM versus DVM measurements, and mouting media are potential contributors to the variability. Combining TEM and STORM with a confocal timecourse of isolated myofibrils, they refine previously published values of sarcomere length and width, and add more comprehensive data for filament length, number and spacing. This highly accurate timecourse demonstrates continual growth of sarcomeres after 48 h APF, and correct some inconsistencies from previous large-scale timecourse datasets. These data are very valuable to the field, especially Drosophila muscle biologists, and will serve as a comparative resource for future studies. Weaknesses: At early timepoints, loss of sarcomere components through mechanical or detergent-mediated artifacts may influence the authors' measurements. In addition, isolating myofibrils is not always the most ideal approach, as it loses information on myofiber structure as well as organization and structure of the myofibrils in vivo.

We believe that the control experiments we presented here adequately demonstrate that sarcomere measurements are not affected by the myofibril isolation process at early timepoints (Figure 3C). Nevertheless, we certainly agree with the reviewer that isolated myofibrils alone cannot capture the entire complexity of muscle tissues, and additional approaches should also be applied in complex projects. Yet, we are confident that our approach offers the most reliable and efficient method for precise morphometric analysis of the sarcomeres, and although alone it is very unlikely to be sufficient to address all questions of a muscle development project, it can still be applied as a very useful and robust tool.

The point regarding liquid versus hardening mounting media is valuable, but remains to be tested and validated with the diverse liquid and hardening media used by other labs.

Whereas it would not be feasible for us to test all possible liquid and hardening media used by others in all possible conditions, we tested the effect of Vectashield (the most commonly used liquid media) according to the suggestion of the reviewer, and the results are now included in the manuscript. We think that this is a valuable extension of the list of the materials and conditions we tested, although we need to point out that our primary goal was not necessarily to test as many conditions as possible (because the number of those conditions is virtually endless), rather to raise awareness among colleagues that these variables can significantly impact the data obtained and affect their comparability.

The IMA software seems to be designed specifically for analysis of isolated myofibrils, and it is unclear if it would work for other types of IFM preparations.

As stated in the manuscript, IMA is a specialized tool designed for the analysis of individual myofibrils. While it can also process other types of IFM preparations in semi-automatic or manual modes, we believe these approaches compromise both efficiency and accuracy. This is further clarified in the revised manuscript.

A last point is that TEM and STORM may not be available on a regular basis to many labs, hindering wide implementation of the approach used in this manuscript to generate very accurate and detailed measurements of sarcomere morphometrics.

Regarding the availability of TEM and STORM, we acknowledge that these techniques are not universally accessible. However, that is exactly one major value of our work that our open-source software tool now allows researchers to generate valuable data using only a confocal microscope in combination with our published datasets.

Audience: Scientists who study sarcomerogenesis or Drosophila muscle biology.

My expertise: I study muscle development in the Drosophila model.

__Reviewer #3 (Evidence, reproducibility and clarity (Required)): __ Summary: This manuscripts presents a computational tool to quantify sarcomere length and myofibril width of the Drosophila indirect flight muscles, including developmental samples. This tool was applied to confocal and STORM super-resolution images of isolated myofibrils from adult and developing flight muscles. Thick filament numbers per myofibril were counted during development of flight muscles. A myofilament model of developing flight muscle myofibrils is presented that remains speculative for the early developmental stages.

Major comments: 1. The title of the manuscript appears unclear. What is a lattice model? Lattice is an ordered array. The filament array parameters for mature flight muscles was aready measured. It appears that the authors speculate how this order might be generated during sarcomere assembly, which is not studied in this manuscript as it is limited to periodic arrays after 36h APF.

As the reviewer correctly points out, a lattice refers to an ordered array - in the case of IFM sarcomeres, this includes both thin and thick filaments. Therefore, the phrase "myofilament lattice model of Drosophila flight muscle sarcomeres" specifically describes a model representing the spatial organization of these filament arrays within the sarcomere. To provide additional clarity for readers, we have revised the title to include more context. It now reads: Developmental Remodeling of Drosophila Flight Muscle Sarcomeres: A Scaled Myofilament Lattice Model Based on Multiscale Morphometrics

To create a model of these arrays, three essential pieces of information are required:

1) The length of the filaments,

2) The number of filaments, and

3) The relative position of the filaments.

While some direct measurements are available in the literature, and others can be used to calculate the necessary values, available data is often contradictory or simply different from each other (as described in our ms) making them unsuitable for constructing scaled models of the myofilament arrays. In contrast to that, here we present a comprehensive and consistent set of measurements that enabled us to build models not only of mature sarcomeres but also of sarcomeres at three other significant developmental time points.

Regarding the mention of "sarcomere assembly" in line 37, we intended it to refer to the growth of the sarcomeres, not their initial formation. We do not speculate about sarcomere assembly anywhere in the text. In fact, we have clearly stated multiple times that our focus is on the growth of the IFM myofilament array during myofibrillogenesis. Nevertheless, to avoid confusion, we revised the phrase in line 37 to "sarcomere growth".

The authors review the flight muscle sarcomere length literature and conclude it is variable because of imprecise measurements. Likely this is partially true, however, more importantly is that the sarcomere length and width changes during isolation methods of the myofibrils, as well as by various embedding methods, as the authors show here as well in Figure 1B-E.

We dedicated two sections of the Results - “An automated method to accurately measure sarcomeric parameters” and “IFM sarcomere morphometrics are affected by sex, age, fiber type, and sample preparation” - to exploring potential sources of variability in published IFM sarcomere measurements. Based on these analyses, we conclude that such variability stems from both measurement imprecision and biological or technical factors, including sex, age, fiber type and, of foremost, sample preparation. Because it is difficult to quantify the relative impact of each variable across published studies, we have refrained from speculations about the relative contribution of the different factors in the revised manuscript.

Hence, I find the strongly claims the authors make here surprising, while they are isolating the myofibrils. Hence, these myofibrils are ruptured at the ends, relaxed or contracted, depending on buffer choice and passive tension is released. On page 8, the authors correctly state that the embedding medium causes shrinkage of the myofibrils. While isolation is state of the art for electron microscopy techniques, other methods including sectioning or even whole mount preparation have been developed for high resolution microscopy of IFMs that avoid these artifacts. Unfortunately, this manuscript only uses isolated myofibrils that were fixed and then mechanically dissociated by pipetting. This method likely induces variations as seen by the large spread of sarcomere length reported in Figure 1C (2.8-3.9µm?) and even bigger spreads for myofibril widths. Are these also seen in tissue without dissections? Unfortunately, no comparision to intact flight muscles are reported with the here presented quantification tool. The sarcomere length spread in the developmental samples is even larger.

The major issue raised in this paragraph is the use of isolated myofibril versus intact flight muscle preparations. The reviewer claims that the latter might be superior because the isolated myofibrils are ruptured at their ends. Clearly, the intact IFMs cannot be imaged in vivo by light microscopy because the adult fly cuticle is opaque. To visualize these muscles, one must open the thorax, but neither microdissection nor sectioning preserves them perfectly, even the cleanest longitudinal cuts sever some myofibrils, and dissection itself can damage the tissue. Although published images often show only the most pristine regions, the practice of selective cropping cannot be taken as a scientific argument. Here, by comparing sarcomere lengths measured in isolated myofibrils with those from whole-mount longitudinal DLM sections and microdissected IFM myofibers, we demonstrate that isolation does not alter sarcomere length (Figure 1E, now Figure 2A in the revised manuscript). As to myofibril width, it is determined by two parameters: the number of myofilaments and the spacing between them. In vivo filament spacing has been measured directly, and filament counts can be obtained from EM cross-sections of DLM fibers. Combining these values gives an expected in vivo myofibril diameter. While isolated myofibrils measure thinner than those in whole-mount or microdissected samples (Figure 1E, now Figure 2A in the revised manuscript), their diameter closely matches this in vivo estimate (see manuscript, lines 187–198). Therefore, we conclude that isolated myofibrils (even if it seems counterintuitive for this reviewer) are superior for sarcomere measurements than whole-mount preparations - and that is why we primarily rely on them here.

Despite that, we certainly recognize that isolated myofibrils cannot recapitulate every aspect of an IFM fiber, and the need for whole-mount preparations during our IFM studies is not questioned by us.

        In addition to this general answer to the issues raised in the above paragraph of the reviewer, we would like to specifically reflect for some of the remarks:

„Unfortunately, this manuscript only uses isolated myofibrils that were fixed and then mechanically dissociated by pipetting.”

This is a false statement that “this manuscript only uses isolated myofibrils” as we used different preparation methods for initial comparisons (see Figure 1E, now Figure 2A in the revised manuscript). Additionally, unlike the reviewer assumed, the myofibrils were first dissociated and then fixed, and not vice versa (as described in the Materials and Methods section).

„This method likely induces variations as seen by the large spread of sarcomere length reported in Figure 1C (2.8-3.9µm?) and even bigger spreads for myofibril widths. Are these also seen in tissue without dissections?”

This remark makes absolutely no sense, as we do not report sarcomere length values in Figure 1C at all. By assuming that the reviewer meant to refer to Figure 1B, it still remains a misunderstanding or a false statement, because that panel refers to the variations found in published data (not in our current data), and this is clearly explained both in the figure legend and the main text. Regardless of that, the stated spread does not appear unusual. In the article by Spletter et al. (2018), the authors report a similar spread (2.576–3.542 µm) for sarcomere length in mature IFM using whole-mount DLM cross-sections. As to the second question here, we do observe a comparable spread in other preparations as well (see Figure 1E, now Figure 2A in the revised manuscript), which is again the opposite conclusion as compared to the (clearly false) assumption of the reviewer.

„Unfortunately, no comparision to intact flight muscles are reported with the here presented quantification tool. „

This is also a false statement; as we do report comparison to whole mount cross sections which we belive the reviewer considers „intact” in Figure 1E (Figure 2A in the revised manuscript).

„The sarcomere length spread in the developmental samples is even larger.”

The spread is not larger at all than in previous reports, as clearly shown in Supplementary Figure 3A.

The authors suggest that there are sex differences in sarcomere length and pupal development duration. This is potentially interesting, unfortunately they then use mixed sex samples to analyse sarcomeres during flight muscle development.

In the revised manuscript, we now provide a more detailed description of a subtle post-eclosion difference in IFM sarcomere metrics between male and female Drosophila. We attribute this variation to the well-established observation that female pupae develop slightly faster than males, a property that may last till shortly after eclosion. Confirming this experimentally would require considerable effort with limited scientific benefit. Nonetheless, the subtle nature of this sex-linked variation reinforced our decision to include IFM sarcomeres from both male and female flies in our comprehensive developmental analysis.

The IMA software tool lacks critical assessment of its performance compared to other tools and the validation presented is too limited. IMA seems to generate systematic errors, based on Fig S1E, as it does not report the ground truth. These have to be discussed and compared to available tools. The principles of fitting used in IMA seem well adapted to IFM myofibrils in low noise conditions, but may not be usable in other situations. This should be assessed and discussed.

IMA is a specialized software tool developed to address a specific need, notably, to accurately and efficiently measure sarcomere length and myofibril diameter in individual IFM myofibril images labeled with both phalloidin and Z-disc markers. For our purposes, it remains the most suitable and reliable option, and we are confident that IMA outperforms all other available tools. To demonstrate this, we have included a table comparing the few alternatives (MyofibrilJ, SarcGraph, and sarcApp) capable of both measurements, which further supports our conclusion. Given IMA's focused application, extensive validation under artificially low signal-to-noise conditions is unnecessary. While IMA may introduce minor systematic errors (~0.01 µm for sarcomere length and ~0.03 µm for myofibril diameter), these are negligible errors relative to the limitations of the simulated ground truth data used for benchmarking. This point is now addressed in the manuscript.

It is claimed that validation was achieved on simulated IFM images: do the authors rather mean simulated isolated IFM myofibril images? This is not quite the same in terms of algorithm complexity and this should be corrected if this is the case.

Indeed, we used simulated individual IFM myofibril images, where both phalloidin labeling and Z-disc labeling are present. This is clearly shown in Supplementary Figure 1A, and stated in the text when first introduced: „we generated artificial images of IFM myofibrils with known dimensions, simulating the image formation process”

The authors need to revise their comparison to other tools. It is incomplete and seemingly incorrect. It should be clearly stated that IMA is limited to isolated myofibrils, which is a far easier segmentation task than what other tools can do, such as sarcApp (Neininger-Castro et al. 2023, PMID: 37921850). Defining the acronym would be valuable in that sense. The claim line 129-130 "none can adequately measure myofibril diameter from regular side view images" is unclear. What do the authors refer to as "side view images"? Sarc-Graph from Zhao et al 2021, PMID: 34613960, and sarcApp from Neininger-Castro et al. 2023 provide sarcomere width, in conditions that are very similar to what IMA does, e.g. on xy images based on the documentation provided on github. A performance comparison with these tools would be valuable. Does installation and use of IMA require computational skills?

Motivated by the reviewer’s comments, we revised the section introducing IMA. However, we chose not to include an extensive comparison with other software tools, as this would divert the manuscript’s focus without impacting the main conclusions. Instead, we added a summary table highlighting the key requirements for analyzing IFM sarcomere morphometrics from Z-stacks of phalloidin- and Z-line-labeled individual myofibrils and compared the available tools accordingly. In our experience, most software tools are developed to address very specific problems, even those marketed as general-purpose solutions. Consequently, applying them beyond their intended scope often results in reduced efficiency and suboptimal performance. Although sarcApp was initially available as a free tool, one of its dependencies (PySimpleGUI 5) has since adopted a commercial license model. Using a trial version of PySimpleGUI 5, we evaluated sarcApp on our dataset. The software is limited to single-plane image input, hence raw image stacks must be preprocessed into a suitable format, which is a time consuming step. Furthermore, implementation requires basic programming proficiency, as parameter adjustments must be performed directly within the source code to accommodate dataset-specific configurations. Once appropriately configured, sarcApp reliably quantifies both sarcomere length and myofibril width with accuracy comparable to that of IMA. However, it lacks built-in diagnostic feedback or visualization tools to facilitate measurement verification or troubleshooting during batch processing. SarcGraph also supports only single-plane image inputs and requires prior image preprocessing. Additionally, images must be loaded manually one by one, which further reduces processing efficiency. Parameter optimization relies on direct code modification through a trial-and-error process, demanding a certain level of programming proficiency. Even with these adjustments, the software frequently introduces artifacts - such as Z-line splitting - when applied to our dataset. Even when segmentation is successful, sarcomere length is often overestimated, whereas myofibril diameter is consistently underestimated. As compared to these issues, IMA was designed for ease of use and does not require any programming experience to install or operate. It can automatically handle raw microscopic image formats without the need for preprocessing. Segmentation is fully automated, with no requirement for parameter tuning. The tool provides visual feedback during both the segmentation and fitting steps, allowing users to confidently assess and validate the results. IMA produces accurate and precise measurements of sarcomere length and diameter. Batch processing is enabled by default, significantly improving efficiency when analyzing multiple images. Finally, unlike the reviewer stated, IMA is not limited to isolated myofibrils. It is optimized for isolated myofibrils (i.e. full performance is achieved on these samples), but it can also work on whole-mount preparations in semi-automatic and manual mode, which still allow precise measurements (with some reduction in processing efficiency).

As to the minor comments, the acronym IMA was already defined in lines 541 and 917–918 of the original submission, as well as on the software’s GitHub page. Additionally, we replaced the phrase "side view images" with "longitudinal myofibril projections" to improve clarity.

How do the authors know that the bright phallodin signal visible that the Z-disc at 36h and 48h APF is due to actin filament overlap, as suggested? An alternative solution are more short actin filaments at the early Z-discs.

It is widely accepted that the bright phalloidin signal at the Z-line in mature sarcomeres reflects actin filament overlap (e.g., Littlefield and Fowler, 2002; PMID: 11964243). Accordingly, in slightly stretched myofibrils, this bright signal diminishes, and in more significantly stretched myofibrils, a small gap appears (e.g., Kulke et al., 2001; PMID: 11535621). The width of this bright phalloidin signal corresponds to the electron-dense band seen in longitudinal EM sections (Figure 3B and Supplementary Figure 5B, now Figure 4B and Figure 6B in the revised manuscript) and matches the actin filament overlap observed in Z-disc cryo-EM reconstructions from other species (Yeganeh et al., 2023; Rusu et al., 2017), where individual thin filaments can be resolved. By extension, we interpret the bright phalloidin signals at the Z-discs observed at 36 h and 48 h APF as arising from similar actin filament overlaps, given their comparable width to the electron-dense Z-bodies described both in our study (Supplemantary Figure 5B, now Figure 6B in the revised manuscript) and by Reedy and Beall (1993). While we cannot fully rule out the reviewer’s alternative interpretation, for the time being it remains a bold speculation without supporting evidence, and therefore we prefer to stay with the conventional view.

The authors seem to doubt their own interpretation that actin filaments shrink when reading line 304 and following. This is obviously critical for the "model" presented.

Unlike the reviewer implies, we certainly do not doubt our own interpretation, but to avoid confusion we revised the corresponding paragraph in the manuscript and provided more details on our explanation, and we also provide a brief overview of it here. Between 36 h and 48 h APF we observe a pronounced structural transition in the IFM sarcomeres. In EM cross-sections, the previously irregular myofilament lattice becomes organized into a regular hexagonal pattern (Figure 3A, now Figure 4A in the revised manuscript) with filament spacing typical of mature myofibrils (Supplementary Figure 5A, now Figure 6A in the revised manuscript). In longitudinal EM sections, the elongated, amorphous Z-bodies condense along the myofibril axis to form well-defined, adult-like Z-discs (Supplementary Figure 5B, now Figure 6B in the revised manuscript). Similarly, dSTORM imaging shows that the Z-disc associated D-Titin epitopes become more compact and organized during this period (Supplementary Figure 4E, now Figure 5E in the revised manuscript). The edges of the thick filament arrays also become more sharply defined, and the appearance of a distinct bare zone indicates the establishment of a regular register (Figure 3B, now Figure 4B in the revised manuscript). By assuming that a similar reorganization occurs within the thin filament array, the apparent length of the thin filament array would decrease—not due to shortening of individual filaments, rather due to improved alignment. Although we cannot directly resolve single thin filaments, this reorganization offers the most plausible explanation for the observed change.

Minor comments: 1. Figure S1B is not called out in the text.

The reviewer might have missed this, but in fact, it is explicitly called out in line 181.

Fig. 1: Please state whenever images are simulations?

We appreciate the reviewer’s observation that the simulated IFM myofibril images are indistinguishable from the real ones, as this confirms the adequacy of these images for testing our software tool. However, this is already clearly indicated: Figure 1B features simulated images, as noted in the figure legend (line 824), and Supplementary Figure 1A similarly shows simulated images, as stated both in the legend (line 886) and in the figure.

Fig. 2: Length-width correlation - please provide individual points color-coded by time point?

As suggested by the reviewer, we generated a plot with the individual points color-coded by time.

"newly eclosed males and females, we observed that males have slightly shorter sarcomeres and narrower myofibrils". Please provide a statistical test supporting the difference.

In the revised manuscript, we compared sarcomere length and myofibril width between males and females from 0 to 96 hours AE using a two-way ANOVA with Sidak’s multiple comparisons test. We expanded our description of these observations in the main text, and details of the statistical analysis are now included in the revised figure legend (Figure 1E). Briefly, newly eclosed males showed slightly shorter sarcomeres than females - a consistent but non-significant trend (p = 0.9846) - which resolved by 12 h AE, with sarcomere lengths remaining similar thereafter (p = 0.1533; Figure 1E). In contrast, myofibril width was significantly narrower in the newly eclosed males (p = 0.0374), but this difference disappeared between 24 and 48 h AE as myofibrils expanded in diameter during post-eclosion development (p

Were statistical tests performed using animals as sample numbers? Please clarify in the images what are animal and what are sarcomere numbers.

Following standard guidelines, statistical tests were performed using the means of independent experiments, as noted in the figure legends. For each experiment, we used approximately 6 animals, and this information is now included in the Materials and Methods section.

mef2-Gal4 should be spelled Mef2-GAL4 according to Flybase.

This has been corrected in the revised text and figures.

Are the images shown in Figure 2B representative? 96h AE appears thicker than 24h AE but the graph reports no difference.

We aimed to show representative images, however, in the case of 96h APF we may have selected a wrong example. We now changed the image for a more appropriate one.

The authors only found Zasp52 and alpha-Actinin at the Z-discs from 72h APF onwards, which is different to what others have reported.

Similarly to former reports, we detected both α-Actinin (see Supplementary Figure 3B, now Figure 3C in the revised manuscript) and Zasp52 in microdissected IFMs as early as 36 hours APF. However, these markers were largely absent in isolated myofibrils from the early pupal stages (36–60 hours APF). By 60 hours APF, strong α-Actinin and Zasp52 signals were clearly visible in isolated myofibrils (the closest timepoint captured by dSTORM is 72h APF). As discussed in the manuscript, a likely explanation is that α-Actinin and Zasp52 are recruited to developing Z-bodies early on but are only fully incorporated into mature Z-discs between 48 and 60 hours APF. Their incomplete integration at earlier stages may lead to their loss during the isolation procedure.

Thick filament length during development has also been estimated by Orfanos and Sparrow, which should be cited (PMID: 23178940)

Contrary to the reviewer’s claim, the article 'Myosin isoform switching during assembly of the Drosophila flight muscle thick filament lattice' does not provide any measurements or estimates of thick filament length; it only includes a schematic illustration where the length of the thick filaments is not based on empirical data.

**Referee Cross-commenting**

I also agree with my colleagues comments, which are largely consistent.

Reviewer #3 (Significance (Required)):

This paper introduces a tool to measure sarcomere length. Easy to use tools that do this as well already exist. The tool can also measure sarcomere width, which it claims as unique point, which is not the case, see above comment.

We are aware that other tools exist to measure sarcomere parameters (and we did not claim the opposite in our ms), nevertheless, we need to emphasize that based on our comparisons, IMA is superior to all three alternatives. Three software tools could, in principle, be used to measure both sarcomere length and myofibril diameter: MyofibrilJ, SarcGraph, and sarcApp. However, two of them - MyofibrilJ and SarcGraph - consistently under- or overestimate these values. The only tool capable of performing these measurements reliably, sarcApp, is no longer freely available, it requires programming expertise, and it does not support raw image file formats, making it difficult to use in practice (see above comments for more details). In contrast, IMA is user-friendly and does not require any programming expertise to install or operate. It can automatically process raw microscopic image formats without the need for preprocessing. Segmentation is fully automated, and no parameter tuning is necessary. The tool offers visual feedback on both the segmentation and fitting processes, enabling users to validate results with confidence. IMA delivers accurate and precise measurements of sarcomere length and diameter. Additionally, batch processing is enabled by default, significantly enhancing workflow efficiency.

This manuscript shows that depending on the isolation and embedding media sarcomere and myofibrils width changes and hence artifacts can be introduced. While this is not suprising, it has not been well controlled in a number of previous publications.

Furthermore, this paper measures sarcomere length and width during flight muscle development and consolidates what was already known from previous publications. Sarcomeres are added until 48 h APF, then they grow in diameter. Despite strong claims in the text, I do not see any significant novel findings how sarcomeres grow in length or width or any significant deviations from what has been published before. This is even documented in the supplementary graphs by comparing to published data. It is close to identical.

The overall process has been quantitatively described in four previous studies (Reedy and Beall, 1993, Orfanos et al., 2015, Spletter et al., 2018, Nikonova et al., 2024). While there is general agreement on the pattern of sarcomere development, significant discrepancies exist among these datasets; differences that become particularly problematic when attempting to build structural models. More specifically: Reedy and Beall (1993) report substantially shorter sarcomeres compared to all other datasets, including ours. This discrepancy likely stems from two factors: (i) their use of longitudinal EM sections, where sample preparation is known to cause considerable tissue shrinkage; and (ii) the maintenance of their flies at 23 °C, a temperature that clearly delays development relative to the more commonly used 25 °C. Interestingly, Spletter et al. (2018) and Nikonova et al. (2024) conducted their experiments at 27 °C, which also deviates from standard conditions and may complicate comparisons. Orfanos et al. (2015) suggested that mature sarcomere length is reached by approximately 88 hours after puparium formation (APF). In contrast, our measurements show that sarcomeres continue to elongate beyond this point, reaching mature length between 12 and 24 hours post-eclosion. All four earlier studies report a mature sarcomere length around 3.2-3.3 µm, only slightly longer than the ~3.2 µm length of thick filaments (Katzemich et al., 2012; Gasek et al., 2016). This would imply an I-band length below ~100 nm, which is an implausibly short distance. In contrast, our data, along with several recent studies (González-Morales et al., 2019; Deng et al., 2021; Dhanyasi et al., 2020; DeAguero et al., 2019), support a mature sarcomere length of approximately 3.45 µm, placing the length of the I-band at around 250 nm. This estimate is more consistent with high-resolution structural observations from longitudinal EM sections and fluorescent nanoscopy (Szikora et al., 2020; Schueder et al., 2023). Although Reedy and Beall (1993) provide limited data on myofibril diameter during myofibrillogenesis, a more detailed quantitative analysis is presented by Spletter et al. (2018) and by Nikonova et al. (2024). Interestingly, Spletter et al. report two separate datasets - one based on longitudinal sections and another on cross-sections of DLM fibers. While the measurements are consistent during early pupal stages, they diverge significantly in mature IFMs (1.116 ± 0.1025 µm vs. 1.428 ± 0.0995 µm), a discrepancy that is not addressed in their publication. Nikonova et al. (2024) report even narrower myofibril widths (0.9887 ± 0.1273 µm). Moreover, the reported diameters of early myofibrils in all three datasets are nearly twice as large as those reported by Reedy and Beall (1993) and in our own measurements, directly contradicting the reviewer's claim that the values are “close to identical.” Finally, our data clearly demonstrate that both the length and diameter of IFM sarcomeres reach a plateau in young adults, which is a key developmental feature not examined in previous studies.

In summary, we did not and we do not intend to claim that our conclusions are novel as to the general mechanisms of myofibril and sarcomere growth. Rather, our contribution lies in providing a high-precision, robust analysis of the growth process using a state-of-the-art toolkit, resulting in a comprehensive description that aligns with structural data obtained from TEM and dSTORM. We therefore believe that expert readers will recognize numerous valuable aspects of our approaches that will advance research in the field.

Counting the total number of thick filaments during myofibril development is nice, however, this also has been done (REEDY, M. C. & BEALL, C. 1993, PMID: 8253277). In this old study, the authors reported the amount of filament across one myofibril. How does this compare to the new data here counting all filaments? Unfortunatley, this is not discussed.

Indeed, the study by Reedy and Beall (1993) was primarily based on longitudinal DLM sections, which were used to estimate myofibril width and count the number of thick filaments on this lateral view images (e.g., ~15 thick filaments wide at 75 hours APF), but total thick filament numbers were not provided. While such data could theoretically be used to estimate the number of myofilaments per myofibril, these estimations would depend on the unverified assumption that the section includes the full width of the myofibril. Additionally, the study did not provide standard deviations or the number of measurements, limiting the interpretability and reproducibility of their findings. These points highlight the need for a more rigorous and quantitative approach. For these reasons, we chose to quantify myofilament number using cross-sections, providing more accurate and reliable assessments.

Besides the difference between the lateral versus cross sections, a direct comparison of our studies is further complicated by differences in the developmental time points and experimental conditions used. Reedy and Beall (1993) reports data from pupae aged 42, 60, 75 and 100 hours, as well as from adults, whereas we present data from 36, 48, and 72 hours APF, and from 24 hours after eclosion, which corresponds to approximately 124 hours APF. Moreover, their experiments were carried out at 23 °C, a temperature that somewhat slows down pupal development and results in adult eclosion at around 112 hours APF, as stated in their study. In contrast, our experiments were carried out at the more commonly used 25 °C, where adults typically emerge around 100 hours APF.

Collectively, these differences prevented meaningful comparisons between the two datasets, and therefore we preferred to avoid lengthy discussions on this issue.

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:

This manuscripts presents a computational tool to quantify sarcomere length and myofibril width of the Drosophila indirect flight muscles, including developmental samples. This tool was applied to confocal and STORM super-resolution images of isolated myofibrils from adult and developing flight muscles. Thick filament numbers per myofibril were counted during development of flight muscles. A myofilament model of developing flight muscle myofibrils is presented that remains speculative for the early developmental stages.

Major comments:

1. The title of the manuscript appears unclear. What is a lattice model? Lattice is an ordered array. The filament array parameters for mature flight muscles was aready measured. It appears that the authors speculate how this order might be generated during sarcomere assembly, which is not studied in this manuscript as it is limited to periodic arrays after 36h APF.

2. The authors review the flight muscle sarcomere length literature and conclude it is variable because of imprecise measurements. Likely this is partially true, however, more importantly is that the sarcomere length and width changes during isolation methods of the myofibrils, as well as by various embedding methods, as the authors show here as well in Figure 1B-E.

Hence, I find the strongly claims the authors make here surprising, while they are isolating the myofibrils. Hence, these myofibrils are ruptured at the ends, relaxed or contracted, depending on buffer choice and passive tension is released. On page 8, the authors correctly state that the embedding medium causes shrinkage of the myofibrils. While isolation is state of the art for electron microscopy techniques, other methods including sectioning or even whole mount preparation have been developed for high resolution microscopy of IFMs that avoid these artifacts. Unfortunately, this manuscript only uses isolated myofibrils that were fixed and then mechanically dissociated by pipetting. This method likely induces variations as seen by the large spread of sarcomere length reported in Figure 1C (2.8-3.9µm?) and even bigger spreads for myofibril widths. Are these also seen in tissue without dissections? Unfortunately, no comparision to intact flight muscles are reported with the here presented quantification tool. The sarcomere length spread in the developmental samples is even larger.

1. The authors suggest that there are sex differences in sarcomere length and pupal development duration. This is potentially interesting, unfortunately they then use mixed sex samples to analyse sarcomeres during flight muscle development.

2. The IMA software tool lacks critical assessment of its performance compared to other tools and the validation presented is too limited. IMA seems to generate systematic errors, based on Fig S1E, as it does not report the ground truth. These have to be discussed and compared to available tools. The principles of fitting used in IMA seem well adapted to IFM myofibrils in low noise conditions, but may not be usable in other situations. This should be assessed and discussed.

It is claimed that validation was achieved on simulated IFM images: do the authors rather mean simulated isolated IFM myofibril images? This is not quite the same in terms of algorithm complexity and this should be corrected if this is the case.

1. The authors need to revise their comparison to other tools. It is incomplete and seemingly incorrect. It should be clearly stated that IMA is limited to isolated myofibrils, which is a far easier segmentation task than what other tools can do, such as sarcApp (Neininger-Castro et al. 2023, PMID: 37921850). Defining the acronym would be valuable in that sense. The claim line 129-130 "none can adequately measure myofibril diameter from regular side view images" is unclear. What do the authors refer to as "side view images"? Sarc-Graph from Zhao et al 2021, PMID: 34613960, and sarcApp from Neininger-Castro et al. 2023 provide sarcomere width, in conditions that are very similar to what IMA does, e.g. on xy images based on the documentation provided on github. A performance comparison with these tools would be valuable. Does installation and use of IMA require computational skills?

2. How do the authors know that the bright phallodin signal visible that the Z-disc at 36h and 48h APF is due to actin filament overlap, as suggested? An alternative solution are more short actin filaments at the early Z-discs. The authors seem to doubt their own interpretation that actin filaments shrink when reading line 304 and following. This is obviously critical for the "model" presented.

Minor comments:

1. Figure S1B is not called out in the text.

2. Fig. 1: Please state whenever images are simulations?

3. Fig. 2: Length-width correlation - please provide individual points color-coded by time point?

4. "newly eclosed males and females, we observed that males have slightly shorter sarcomeres and narrower myofibrils". Please provide a statistical test supporting the difference.

5. Were statistical tests performed using animals as sample numbers? Please clarify in the images what are animal and what are sarcomere numbers.

6. mef2-Gal4 should be spelled Mef2-GAL4 according to Flybase.

7. Are the images shown in Figure 2B representative? 96h AE appears thicker than 24h AE but the graph reports no difference.

8. The authors only found Zasp52 and alpha-Actinin at the Z-discs from 72h APF onwards, which is different to what others have reported.

9. Thick filament length during development has also been estimated by Orfanos and Sparrow, which should be cited (PMID: 23178940)

Referee Cross-commenting

I also agree with my colleagues comments, which are largely consistent.

Significance

This paper introduces a tool to measure sarcomere length. Easy to use tools that do this as well already exist. The tool can also measure sarcomere width, which it claims as unique point, which is not the case, see above comment.

This manuscript shows that depending on the isolation and embedding media sarcomere and myofibrils width changes and hence artifacts can be introduced. While this is not suprising, it has not been well controlled in a number of previous publications.

Furthermore, this paper measures sarcomere length and width during flight muscle development and consolidates what was already known from previous publications. Sarcomeres are added until 48 h APF, then they grow in diameter. Despite strong claims in the text, I do not see any significant novel findings how sarcomeres grow in length or width or any significant deviations from what has been published before. This is even documented in the supplementary graphs by comparing to published data. It is close to identical.

Counting the total number of thick filaments during myofibril development is nice, however, this also has been done (REEDY, M. C. & BEALL, C. 1993, PMID: 8253277). In this old study, the authors reported the amount of filament across one myofibril. How does this compare to the new data here counting all filaments? Unfortunatley, this is not discussed.

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

Summary:

In this manuscript titled "A myofilament lattice model of Drosophila flight muscle sarcomeres based on multiscale morphometric analysis during development," Görög et al. perform a detailed analysis of morphological parameters of the indirect flight muscle (IFM) of D. melanogaster. The authors start by illustrating the range of measurements reported in the literature for mature IFM sarcomere length and width, showing a need to revisit and determine a standardized measurement. They develop a new Python-based tool, IMA, to analyze sarcomere lengths from confocal micrographs of isolated myofibrils stained with phalloidin and a z-disc marker. Using this tool, they demonstrate that sample preparation (especially mounting medium), as well as fiber type, sex, and age influence sarcomere measurements. Combining IMA, TEM, and STORM data, they measure sarcomere parameters across development, providing a comprehensive and up-to-date set of "standardized" sarcomere measurements. Using these data, they generate a model integrating all of the parameters to model sarcomeres at four discrete timepoints of development, recapitulating key phases of sarcomere formation and growth.

Major comments:

• Line 200 & 901 - Figure S1B - The authors make a strong statement about the use of liquid versus hardening media, and it is clear from the image provided in Figure S1 that there is a difference in the apparent sarcomere width. The identity of the "liquid media" versus the "hardening media" should be clearly identified in the Results, in addition to the legend for Figure S1. The authors show that "glycerol-based solutions" increase sarcomere width, but the Materials only list 90% glycerol and PBS. However, a frequently used liquid mounting media is Vectashield. Based on the literature, measurements in liquid Vectashield show diameters significantly less than 2.2 microns observed here with presumably 90% glycerol or PBS. Can the authors qualify this statement, or provide data that all forms of liquid mounting media cause this effect? Does this also apply to hemi-thorax and sectioned preparations, or just isolated myofibrils?

• Can the authors comment on whether the length of fixation or fixation buffer solution, in addition to the mounting medium, make a difference on sarcomere length and diameter measurements? This is another source of variation in published protocols.

• Line 237-238. The authors conclude that premyofibrils are much thinner than previously measured. The use of Airyscan to more accurately measure myofibril width at this timepoint is a good contribution, as indeed diffraction and light scatter likely contribute to increased width measured in light microscopy images. I also wonder, though, how well the IMP software performs in measuring width at 36h APF, given how irregular the isolated myofibrils at this stage look (wide z-lines but thinner and weaker H and I bands as shown in Fig. 2B)? Also, how much of the difference in sarcomere width arises due to effects of "stripping" components off of the sarcomere at the earliest timepoint (for example alpha-actinin or Zasp proteins)? Myofibrils at early timepoints do contain more than 4-12 sarcomeres in a line (they extend the full length of the myofiber), so it is possible they are breaking due to the detergent and mechanical disruption induced by the isolation method.

• Line 312 - What does "stable association" mean in this context? The authors mention early timepoints lack stable association of alpha-Actinin or Zasp52, and they reference Fig. S4C, but this figure only shows 72h and 24 AE, not 36h and 48 h APF. Previous reports have seen localization of both alpha-Actinin and Zasp52, so presumably the detergent or mechanical isolation is stripping these components off of the isolated myofibrils up until 72h. This same type of issue comes up again in

• Lines 325-334, where the authors talk about 3E8 and MAC147. They state that 3E8 signal significantly declines in later stages and that MAC147 is not suitable to label myofibrils in young pupae, but they only show data from 72 APF and 24 AE (which looks to have decent staining for both 3E8 and MAC147). A clearer explanation here would be helpful. Figure 3B. The authors show the H, Z, and I lengths in B', B', and B' and discuss these lengths in the text (lines 305-320). It would also be nice to actually have the plots showing the measured/calculated lengths for thin and thick filaments. These are mentioned in the results, but I cannot find the plots in the figures and there is no panel reference.

• Line 400. Does the model in Figure 4 actually have molecular resolution as the authors claim? From these views, thick and thin filaments appear to be represented by cylindrical objects. Localization of specific molecules would require further modeling with individual proteins. Or do the authors mean localization from STORM imaging relative to the ends of the thick and/or thin filaments? The model itself is a useful contribution, but based on Figure 4, resolution of individual molecules is not evident.

• The main Results section of the text is condensed into 4 figures. However, I found myself flipping back and forth between the main figures and the supplement continuously, especially parts of Supplemental Figures 1, 3, 4, and 5. With such large amounts of detail in the Results relying on the supplement, it may be worth considering reorganizing the main and supplemental figures, and having 7 main figures, to include important panels that are currently in the supplement (esp. Fig S1B, S1C, S1D, S3B, S4, S5).

Minor comments:

• On the plots in Fig. S1B, D, and F, it is hard to see the color of the dots because the red error bars are on top of them. Can the other distribution dots be tinted the correct color or the x-axis labels be added, so it is clear which dataset is which?

• Line 142 needs a reference to Figure S1, Panel E, which shows the accuracy and precision measurements.

• Lines 198 - is this range from the above publications? Needs to be clearly cited.

• Figure S3B is confusing - why do the blow-ups overlap both the top (presumably microdissected) and the bottom (presumably isolated) images? The identity of microdissected images should be labeled, as they are hard to see underneath of the blown-up images and the identity of individual image planes wasn't immediately obvious.

• Line 298. By "misaligned," do the authors mean the pointed ends are not uniformly anchored in the z-disc, leading to the wide z-disc measurements? At this early stage, I'm not sure "misaligned" is the right word - perhaps "were not yet aligned in register at the z-disc" or something similar.

• Figure S6 - spelling mistake in label of panel A, "sarcomer" should be "sarcomere"

• Line 487. Spelling "Zaps52" should be "Zasp52"

• Line 887. Spelling "Myofilement" should be "Myofilament"

• Line 946-947. In the legend for Supp. Fig. 3., the authors should specify which published datasets on sarcomere length are shown in the figure by including the references in the legend. Presumably the "isolated individual myofibrils" are the blue "this study" lines, leaving the "microdissected muscles" as the magenta "previous reports" on the figure. Without the reference, it is not clear if these are microdissected, isolated myofibrils, hemi-thorax sections, cryosections, or another preparation method for the "previous reports" data.

Referee Cross-commenting

I agree with the comments from the other reviewers. Many of the major themes are consistent across the reviews, including regarding the model, preparation methods, and the software tool.

Significance

Strengths: This manuscript is an important contribution to the field of sarcomere development. The authors use modern technologies to revisit variation in morphometric measurements in the literature, and they identify parameters that influence this variation. Notably, sex-specific differences, DLM versus DVM measurements, and mouting media are potential contributors to the variability. Combining TEM and STORM with a confocal timecourse of isolated myofibrils, they refine previously published values of sarcomere length and width, and add more comprehensive data for filament length, number and spacing. This highly accurate timecourse demonstrates continual growth of sarcomeres after 48 h APF, and correct some inconsistencies from previous large-scale timecourse datasets. These data are very valuable to the field, especially Drosophila muscle biologists, and will serve as a comparative resource for future studies.

Weaknesses: At early timepoints, loss of sarcomere components through mechanical or detergent-mediated artifacts may influence the authors' measurements. In addition, isolating myofibrils is not always the most ideal approach, as it loses information on myofiber structure as well as organization and structure of the myofibrils in vivo. The point regarding liquid versus hardening mounting media is valuable, but remains to be tested and validated with the diverse liquid and hardening media used by other labs. The IMA software seems to be designed specifically for analysis of isolated myofibrils, and it is unclear if it would work for other types of IFM preparations. A last point is that TEM and STORM may not be available on a regular basis to many labs, hindering wide implementation of the approach used in this manuscript to generate very accurate and detailed measurements of sarcomere morphometrics.

Audience: Scientists who study sarcomerogenesis or Drosophila muscle biology.

My expertise: I study muscle development in the Drosophila model.

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

Summary

In this work, the authors present a careful study of the lattice of the indirect flight muscle (IFM) in Drosophila using data from a morphometric analysis. To this end, an automated tool is developed for precise, high-throughput measurements of sarcomere length and myofibril width, and various microscopy techniques are used to assess sub-sarcomeric structures. These methods are applied to analyze sarcomere structure at multiple stages in the process of myofibrillogenesis. In addition, the authors present various factors and experimental methods that may affect the accurate measurement of IFM structures. Although the comprehensive structural study is appreciated, there are major issues with the presentation/scope of the work that need to be addressed:

Major Comments

1. The main weakness of the paper is in its claim of presenting a model of the sarcomere. Indeed, the paper reports a structural study that is drawn onto a 3D schematic. There is no myofibrillogenesis model that would provide insights into mechanisms. Therefore, the use of the word model is grossly overstated.

2. In general, the major focus and contribution of the work is unclear. How does the comprehensive nature of the measurements contribute to existing literature?

3. Figure labels are often rather confusing - for example it is unclear why there is a B, B', B' etc instead of B,C,D, etc.

4. Some comments in the text are not clearly tied to the figures. For example, in lines 108-109, are the authors referring to the shadow along the edges of the myofibril when saying they are not clearly defined (Figure 1C)?

5. In line 116, it is unclear what "surrounding structures" the authors are referring to if the myofibrils are isolated.

6. In lines 141-142, there is no reference of data to back up the claim of validation.

7. In line 170, the authors mention the mef2-Gal4/+ strain as a Gal4 driver line but do not clearly state how this strain is different from the wildtypes or how this impacts their results.

8. In lines 182-185, the authors discuss the effects of tissue embedding on morphometrics. Were factors such as animal sex, age, fiber type, etc. conserved in these experiments? If not, any differences in results may be confounding.

9. In lines 199-201, the authors discuss results of myofibril diameter using different preparation methods, yet no data is cited to support the claims. In line 220, the phrase "6 independent experiments" is unclear. Is each independent experiment performed using a different animal? Furthermore, are 6 experiments performed for each time point?

10. In line 254, the authors refer to "number of sarcomeres". It must be clearly stated if this refers to sarcomeres per myofibril, image area, etc.

11. In line 274, the authors refer to "myofilament number". It must be clearly stated if this refers to myofilaments per myofibril, image area, etc.

12. In line 299, the authors mention that thin filaments measured less than 560 nm in length, yet no data is cited to support this.

13. In the "Quantifying sarcomere growth dynamics" section of the summary (starting from line 402) the authors introduce data that would be more naturally placed in the results and discussion section.

14. In lines 422-423, it is not mentioned what the controls are for.

15. In the caption of Figure 1C, it is not mentioned what the red dashed lines in the microscope images represent.

16. In the caption of Figure 1D, the difference between the lighter and darker grey points is not mentioned.

17. In line 849, the stated p-value (0.003) does not match that mentioned in the figure (0.0003).

18. In line 874, it is not clear what an "independent experiment" refers to (different animal, etc.?).

19. Figure 2A is hard to read. Using different colored dots for different time points might help.

20. The significant figures presented in Figure 4 give a completely inaccurate representation of the variability of the measurements achieved with these techniques.

21. In line 877, it should be mentioned that the number of filaments is counted per myofibril. The y-axes in the figure should also be adjusted to clarify this.

22. In line 883, it is not clear what an "independent experiment" refers to (different animal, etc.?).

23. The statement of sample sizes in all figures is a little confusing.

24. In lines 1007-1008, the authors imply that the lattice model is needed for calculation of myofilament length. However, from the equations and previous data, it seems that this can be estimated using the confocal and dSTORM images.

25. A more specific discussion of future directions is needed to put this paper in context. For example: Can anything from the overall process be used to better understand sarcomere dynamics in larger animals/humans? Can this be applied to disease modelling?

26. One of the major claims of the paper is that there is a measurable variability with sex and other parameters. However, this data is never clearly summarized, presented (except for supplement), or discussed for its implications.

Minor Comments

1. Lines 60-65 seem to break the flow of the introduction. As the authors discuss existing methods in literature for IFM analysis in the previous couple sentences, the following sentences should clearly state the limitations of existing methods/current gap in literature and a general idea of what the current work is contributing.

2. In line 104, the acronym for ZASPs is not spelled out.

Referee Cross-commenting

I agree as well.

Significance

In summary, this paper provides a multi-scale characterization of Drosophila flight muscle sarcomere structure under a variety of conditions, which is potentially a significant contribution for the field. However, the paper scope is overstated in that it does not provide an actual sarcomere model. Further, there are multiple issues with data presentation that impact the readability of the manuscript.

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

Manuscript number: RC-2025-02953

Corresponding author(s): Andreas, Villunger

1. General Statements [optional]

*We would like to thank the reviewers for their constructive input and overall support. We appreciate to provide a provisional revision plan, as outlined here, and are happy to engage in additional communication with journal editors via video call, in case further clarifications are needed. *

2. Description of the planned revisions

Insert here a point-by-point reply that explains what revisions, additional experimentations and analyses are planned to address the points raised by the referees.

Reviewer #1

__Evidence, reproducibility and clarity __

Summary: This manuscript by Leone et al describes the role of the PIDDosome in cardiomyocytes. Using a series of whole body and cardiomyocyte specific knockouts, the authors show that the PIDDosome maintains correct ploidy in these cells. It achieves this through inducing cell cycle arrest in cardiomyocytes in a p53 dependent manner. Despite this effect on ploidy, PIDDosome-deficient hearts show no structural or functional defects. Statistics and rigor appear to be adequate.

We thank this referee for taking the time to evaluate our work and their valuable comments. We assume that this reviewer by mistake indicates that the phenomenon we describe, depends on p53. As outlined in the abstract and throughout the manuscript, the effect is independent of p53, but may additionally still involve p21, acting along or parallel to the PIDDosome.

Major comments: 1. Figure 1 uses fluorescent intensity of a nuclear stain to determine ploidy per nucleus and they further separate the results into mononucleated, binucleated or multinucleated cells. It is hard to know how to interpret these results without further information or controls. Is there a good positive control that can be used to help to show whether this assay is quantitative? The differences are larger with the Raidd and caspase-2 knockouts than with the Pidd knockouts but this is not addressed.

*We appreciate this concern. Regarding a "good positive control" we can say that we follow state-of the art in the cardiomyocyte field and studies by the Evans (PMID: 36622904), Kuhn (PMID: 32109383), Bergmann (PMID: 26544945) and Patterson labs (PMID: 28783163, 36912240) all use the identical approach to discriminate 2n from 4n nuclei in microscopy images at the cellular level. The fact that the majority of rodent CM nuclei is indeed diploid (PMID: 31175264, 31585517 and 32078450) and a large number of nuclei has been evaluated to assess their mean fluorescence intensity (MFI) reduces the risk of a systematic bias in our analysis. Moreover, we have used an orthogonal approach that is indeed quantitative to define DNA content, i.e,. flow-cytometry based evaluation of DNA content in isolated CM nuclei (Fig. 1C). We hence are confident our assays are quantitative. *

Regarding the fact that loss of Pidd1 causes a more saddle phenotype, we can offer to discuss this in light of the fact that Pidd1 has additional functions, outside the PIDDosome (PMID: 35343572), and that we made similar observations when analyzing ploidy in hepatocytes (PMID: *31983631). Given the fact that all components of the PIDDosome show a similar phenotype, and that this phenotype is mimicked by loss of the protein that connects PIDD1 and centrosomes, ANKRD26 (Fig. 4a), we are confident that this biological variation in our analysis is not affecting our conclusions. *

On line 459 the authors state that the increase in polyploidy in PIDDosome knockouts occurs in adult hood but this is not directly tested. In fact, in the next section the polyploidy is assessed in early postnatal development. This statement should be explained or removed.

We see that we have made an unclear statement here. In fact, we first noted increases in ploidy in adult heart and then define the time window in development when this happens. This sentence will be rephrased.

In Figure 4. The authors obtained RNAseq data for P1, P7 and P14 but only show the differences with and without caspase-2 at P7. Given that the differences in ploidy are more significant at P14 (Fig 3D), all the comparisons should be shown along with analysis of whether the same genes/gene families are altered in the absence of caspase-2.

The reason why we focus on postnatal day 7 (P7) is that data from Alkass et al (PMID: 26544945) and other labs (PMID: 31175264 ) document that on this day the initial wave of binucleation peaks. Hence, we hypothesized that the PIDDosome must be active in most CM, which aligns well with the increased mRNA levels of all of its components (Figure 3). Interestingly, it seems that its action is tightly regulated, as mRNA of PIDDosome components drop on P10, suggesting PIDDosome shut-down or downregulation. Similar findings have been noted in the liver (PMID: *31983631). Alkass and colleagues also show that very few CMs enter another round of DNA synthesis between P7 and P14, and hence possible transcriptome changes in the absence of the PIDDosome will be strongly diluted. *

Please note that on P1, there is no difference between genotypes to be expected as all CM are mononucleated diploids and cytokinesis competent, as previously demonstrated (PMID: *26544945). Moreover, PIDDosome expression levels are extremely low (Fig. 3A). As such, no difference between genotypes are expected on P1. In addition, on P14 the ploidy phenotype observed in PIDDosome knockout mice reaches the maximum and ploidy increases are comparable to adult tissue. Thus, at this time the trigger for PIDDosome activation (cytokinesis failure) is no longer observed as the majority of CMs are post-mitotic, (PMID: 26247711). As such the impact of PIDDosome activation on the P14 transcriptome is most likely negligible. However, if desired, we can expand our bioinformatics analysis summarizing findings made related to DEGs over time in wt animals by comparing genotypes also on day 1 and day 14. In light of the above, analysis between genotypes on P7 holds still appears as the one most meaningful. *

Some validation of the RNAseq and/or proteomics results would be an important addition to this study

We agree with this notion and propose to validate key candidates related to cardiomyocyte proliferation and polyploidization, some of which we found to be differentially expressed at the mRNA level on day 7in the RNAseq data (e.g., p21, Foxm1, Kif18a, Lin37 and others)

Regarding the proteomics results, we face the challenge that we can only try to confirm if candidate proteins are likely caspase substrates in silico using DeepCleave*, and potentially pick one or two candidates linked to CM differentiation for further analysis in vitro and in heterologous cell based assays (e.g. 293T cells), as no bona-fide ventricular cardiomyocyte cell lines exist. Primary postnatal CMs are extremely difficult to transfect, nor they proliferate without drug-treatment, or fail cytokinesis ex vivo. *

Figure 4D: the authors make the conclusion that p21 is downstream of PIDD (et p53 independent). However, this is not supported by the data because the increase in 4N cells/decrease in 2N cells, although statistically significant, is nowhere near that of caspase-2 KO and caspase-2/p21 KO. Statistics should also compare p32KO with c2KO. In the absence of any other data, the more likely conclusion is that p21 is not involved.

*We agree that the findings related to the impact seen upon loss of p21 suggest that it is not the only effector involved in ploidy control and it may not even be an effector engaged by caspase-2, as C2/p21 DKO mice have an even higher ploidy increase, albeit not statistically significant. However, it is important to highlight that p21 (Cdkn1a) was found to be downregulated in our transcriptomic analysis suggesting an involvement in the caspace2-cascade. We are happy to highlight this when presenting the results and in the discussion. *

*We assume that this referee refers to p73 KO data that should be compared to Casp2 KO data (could be read as p73 or p53, but the latter we compare side by side with Casp2 in Fig. 4 already). As p73 KO mice were not found to be viable beyond day 7 (our attempt to find animals on day 10 failed, in line with published literature (PMID: 24500610, 10716451)), we can only offer to compare this data set to the data presented in Figure 3C, where we have analyzed ploidy increases on day 7 from wt and PIDDosome mutant mice. This re-analysis will show that only Caspase-2 mutant mice display a significant ploidy increase on P7, when compared to wt or p73 mutant animals, while no difference are noted between wt and p73 mutant mice (to be included in new Suppl. Fig. 3C) *

Minor comments: Suggest moving Figure 4A to Figure 3 as it seems to fit better there based on the citation of this figure in the text

*We can see some benefit in this recommendation and included panel 4A now in an updated version of Figure 3. *

Recommend enhancing the brightness of microscopy images in Figure 1E and 2D

We will try to improve image quality, may have been due to PDF conversion

Significance

This study provides interesting information for the role of the PIDDosome in protecting from polyploidy and adds to the body of work by this same group studying this pathway in the liver.

The main weakness in terms of significance is the lack of a phenotype in the hearts of these animals. Therefore, it is clear that ploidy (or at least PIDDosome dependent ploidy) has minimal impact on cardiac development.

We respectfully disagree with the comment that the lack of impact on cardiac function constitutes a weakness of our findings. Several studies on ploidy control in the liver (PMID 34228992) but importantly also heart (PMID: 36622904) have failed to document a clear impact of increased ploidy on organ function. This does not infer insignificance, but maybe rather that the context where this becomes relevant has not been identified. We are happy expand on this in our discussion

The authors mention that they have not tried giving these mice an myocardial infarct (MI) or inducing any other type of cardiac damage. Although it is understood that these experiments are likely outside of the scope of the present study, without this information the impact of this study is moderate. I recommend expanding the discussion to provide a more in-depth possible rationale as to why ploidy perturbations do not lead to structural changes like in the liver.

Despite this, the insights to the pathway itself are interesting to investigators in the caspase-2 field if a little underdeveloped, especially concerning the role of p21.

My expertise is in cell death and caspase biology (especially caspase-2). I have sufficient expertise to evaluate all parts of this paper.

*As mentioned above, we will amend our conclusions on p21, in light of potential findings made when validating DEG candidates, as stated above. *

*We hope that the changes and amendments proposed here will be satisfactory to this referee to recommend publication of a revised manuscript. *

Reviewer #2

__Evidence, reproducibility and clarity: __

__Summary: __

In this study, the authors investigated the role of the PIDDosome during cardiomyocyte polyploidization. PIDDosome is a multi-protein complex activating the endopeptidase Caspase-2, and shown to be involved in eliminating cells with extra centrosomes or in response to genotoxic stress (Burigotto & Fava, 2021, Sladky and Villunger, 2020). In both cases, the PIDDosome is recruited in a ANKRD26-dependent manner at the centrosomes leading to p53 stabilization and cell death (Burigotto & Fava, 2021; Evans et al., 2020; Burigotto et al., 2021).

Here, by studying mouse cardiomyocyte differentiation, the authors showed that PIDDosome is imposing ploidy restriction during cardiomyocyte differentiation. Importantly, in contrast to a previous report in the liver (Sladky et al., 2020), they showed that PIDDosome acts in a p53-independent manner in cardiomyocytes. Indeed, they suggested that PIDDosome controls ploidy in cardiomyocytes through p21 activation.

We want to thank this reviewer for the time taken to evaluate our work and provide critical feedback that will help to improve our revised manuscript.

__Major comments: __

In general the conclusions of the authors are well supported by the experiments. However, I would suggest the following experiments/analysis to strengthen the paper:

The authors should improve the Figure 1 to help the readers who are not familiar with cardiomyocyte polyploidization. For instance, I would suggest to add a scheme to summarize cardiomyocyte polyploidization (in terms of nuclear size, mono vs multi and so on).

We agree that a visual summary of the postnatal timing of CM polyploidization will be helpful for the generalist not familiar with the topic and have added a scheme, adapted from a study by Alkass et al. (PMID: *26544945), who elegantly defined the timing of this process during postnatal mice life (now Fig. 1A). *

Based on the images they presented in 1B, the authors should also measure the nuclear area or volume in the different conditions in which components of the PIDDosome were depleted. Indeed, these two parameters should be easier to conceptualize for the readers (instead of the fluorescence nuclear intensity). This could help to understand if the nuclear size is maintained between the different conditions and if this is comparable between mono, bi or multinucleated cardiomyocytes.

We have acquired this data and it can be used to provide additional information on nuclear area and/or volume. We propose to focus on re-analyzing data from wt, Casp2 and XMLC2CRE/Casp2f/f mice. The additional information can be included in Figures 1 & 2, respectively.

• In Figure 2A, the authors presented cross section of heart from animals showing that PIDDosome depletion has no effect on heart size. This is a surprising result since cardiomyocytes have higher ploidy levels and this could have an effect on their function. Since the importance of this observation, the authors should present a quantification of the heart size in the different conditions shown in Figure 2A.

We agree with this comment. We can measure the heart vs. body weight ratio or tibia length in adult Casp2-/- vs. WT (3 month old) in order to indirectly evaluate possible increases in CM size linked to increased ploidy.

Also, the percentage of cardiomyocytes presenting higher levels of ploidy seems quite low. The authors should discuss this point. In particular because this could explain the absence of consequences on heart size and function at steady state.

We agree with this conclusion and will expand on this in our discussion. It is important to note that as opposed to findings made in liver (PMID: *31983631), genetic manipulation of ploidy regulators such as E2f7/8 (PMID: 36622904), only led to modest changes in CM ploidy, suggesting that either a small band-width compatible with normal heart function exists, or that additional mechanisms exist that take control when these thresholds set by the PIDDosome or E2f7/8 are exceeded. These mechanisms could involve Cyclin G (PMID: 20360255), or TNNI3K (PMID: 31589606). Importantly, a recent publication has shown that overexpression of Plk1(T210D) and Ect2 from birth causes increased heart weight coupled with a minor decrease in CM size. These mice undergo to premature death (PMID: 39912233) suggesting that CM polyploidization is a tight regulated process regulated by several independent mechanisms during heart development. *

In Figure 2D, the authors measured the cardiomyocyte cross-sectional area and concluded that removing PIDDosome components have no effect on cardiomyocyte cell size. Since it has been shown that ploidy increase is normally associated with an increase in cell area, the authors should measure cell area of cardiomyocytes analyzed in Figure 1B. It could be then interesting to establish a correlation with nuclear area and the mono, bi or multinucleated status. This will strengthen the results showing that ploidy increases without affecting cell area.

Indeed, studies in PIDDosome deficient livers suggest that tissue is containing fewer but bigger cells (PMID: *31983631). As opposed to the liver the percentage of cardiomyocytes presenting higher levels of ploidy is relatively low. Thus, a possible increase in CM size in PIDDosome deficient mice may be masked in heart cross-sections. In order to better correlate the ploidy with cell size, we propose to reanalyze our microscopy images used to extract the data displayed in Fig. 1D. We may run into the problem though that the number of cells acquired may become limiting to achieve sufficient statistical power. In this case we could pool data from different PIDDosome mutant CM to increase statistical power. Again, we propose to initially prioritize wt vs. Casp2 vs. XMLC2/Casp2f/f mice. In addition, we can offer to quantify heart to body weight ratio or tibia length as an additional read-out (see answer to a previous reviewer comment). *

The authors should discuss the fact that PIDDosome depletion lead only to a mild increase in ploidy levels (4N) in a small percentage of cardiomyocyte. If the PIDDosome is controlling ploidy, one could expect that removing it should lead to a drastic increase in the ploidy levels. Is PIDDosome depletion leading to cell death in some cardiomyocyte? The authors should discuss this point in the discussion or if relevant show a staining with an apoptosis marker. Is another mechanism compensating to prevent higher ploidy levels in cardiomyocytes?

These are valid thoughts, some of which we contemplated before. In part, we have addressed them in our response to Reviewer#1, above, discussing similar findings made in E2f7/8 deficient hearts (PMID: 36622904), or Cyclin G overexpressing hearts (PMID: 20360255), where also only modest changes in ploidy were achieved. Together these observations are suggesting alternative control mechanism able to act, or limited tolerance towards larger shifts in ploidy, incompatible with proper cell function and survival. Towards this end, we can offer to test if we find increased signs of cell death in PIDDosome mutant hearts by TUNEL staining of histological sections. Of note, we did not find evidence for such a phenomenon in the liver (PMID: 31983631).

Even if the authors presented RNAseq data suggesting that the PIDDosome is activated during cardiomyocyte differentiation, they should clearly demonstrate this point to strengthen the message of the paper. Indeed, the conclusions are based on the absence of PIDDosome components triggering higher ploidy in cardiomyocytes. However, we don't know whether (and when) the PIDDosome is activated during cardiomyocyte differentiation to control their ploidy levels. I would suggest to analyze PIDDosome activation markers by immunofluorescence in *cardiomyocytes at different developmental stages. *

*We agree with this referee that direct proof of PIDDosome activation would be helpful and that we only infer back from loss of function phenotypes when and where the PIDDosome becomes activated. However, several technical issues prevent us from collecting more direct evidence of PIDDosome activation in the developing heart. 1) Polyploidization in heart CM appears to happen gradually in CM from day 3 on with a peak at day 7 (PMID: 26544945). Hence, this is not a synchronous process, where we could pinpoint simultaneous activation of the PIDDosome in all cells at the same time, which would facilitate biochemical analysis, e.g., by western blotting for signs of Caspase-2 activation (i.e. the loss of its pro-form, PMID: 28130345). 2) Our most reliable readout, MDM2 cleavage by caspase-2 giving rise to specific fragments detectable in western, is not applicable to mouse tissue, as the antibody we use only detects human MDM2 (PMID: 28130345) and no other MDM2 Ab we tested gave satisfactory results. Independent of that, 3) we do not see involvement of p53 in CM ploidy control (arguing against a role of MDM2). *

*As such, we can only offer to look at extra centrosome clustering in postnatal binucleated CM (as also suggested further below), as a putative trigger for PIDDosome activation. However, this has been published by the first author of this study before (PMID 31301302). Given that we have made the significant effort to time resolve the increase in ploidy in postnatal mice (please note that several hearts needed to be pooled for each time point, analyzed in multiple biological replicates), we think that our conclusions are well-justified based on the genetic data provided. *

Concerning the methods, the authors must add the references for each product they used and not only the origin. When relevant, the RRID should be indicated. Without this information the method and the data cannot be reproduced.

We will update this information where relevant to reproduce our results

Minor comments:

In general, the text and the figures are clear. Nevertheless, I would suggest the following changes:

• Figures 1B, 2B and 2C: the y-axis must start at 0.

We will adopt axes accordingly

Figure 4A: The authors should stain centrosomes in cardiomyocytes. This should strengthen the conclusion taken by the authors based on the results obtained in mice depleted for ANKRD26. Indeed, for the moment they are insufficient to conclude about the role of the centrosomes. The authors should show that centrosomes cluster in cardiomyocytes (a condition necessary for PIDDosome activation in polyploid cells) and if possible that component of the PIDDosome are recruited here.

*This point is well taken and addressed in part above. Clustering of extra centrosomes has been documented and published by the first author of this study in rat polyploid cardiomyocytes (PIMID; cited). We can offer to show clustering of centrosomes in mouse CM isolated from day 7 hearts, but while PIDD1 can be detected well in MEF, we repeatedly failed to stain fro PIDD1 in primary CMs. *

Figure 4F: I would suggest to modify the working model to emphasize more the differences between WT and PIDDosome KO.

We will aim to improve this cartoon/graphical abstract

The prior studies are referenced appropriately.

Reviewer #2 (Significance (Required)):

How polyploid cells control their ploidy levels during differentiation remains poorly understood. The data presented here represent thus an advance concerning this question. The actual model concerning PIDDosome activation relies on the presence of extra centrosomes that drives the ANKDR26-dependent recruitment of the PIDDosome. Then, Caspase 2 is activated leading to a p53-p21 dependent cell cycle arrest (Burigotto & Fava, 2021, Sladky and Villunger, 2020; Janssens & Tinel, 2012; Evans et al., 2020; Burigotto et al., 2021). In this study, the authors showed that similar pathway takes place during cardiomyocyte differentiation to control ploidy levels. These data are reminiscent of previous work showing PIDDosome involvement during hepatocyte polyploidization (Sladky et al. 2020). Together, these data highlight the prominent role of the PIDDosome complex in controlling ploidy levels in physiological context. Importantly, this study identified that the classical p53-dependent cell cycle arrest described after PIDDosome activation is not involved here. Instead, the data established that independently of p53, p21 contribute to control cardiomyocyte ploidy. In consequence, this study extends the initial pathway associated with PIDDosome activation and suggest that other mechanisms could take place to restrain cell proliferation upon PIDDosome activation. Overall, this makes this paper significant and of interest for the following fields: polyploidy, heart/cardiomyocyte development and PIDDosome.

My field of expertise includes polyploidy, cell cycle and genetic instability.

We thank this reviewer for the time taken and the positive feedback provided.

3. Description of the revisions that have already been incorporated in the transferred manuscript

Please insert a point-by-point reply describing the revisions that were already carried out and included in the transferred manuscript. If no revisions have been carried out yet, please leave this section empty.

N/A

4. Description of analyses that authors prefer not to carry out

Please include a point-by-point response explaining why some of the requested data or additional analyses might not be necessary or cannot be provided within the scope of a revision. This can be due to time or resource limitations or in case of disagreement about the necessity of such additional data given the scope of the study. Please leave empty if not applicable.

• *As outlined above, limited tools are available to validate putative caspase-2 substrates, identified in proteomics analysis, in an impactful manner. *
• *Also, as discussed above, we deem myocardial infarction experiments in mice as unsuitable to improve our work, as with all likely-hood, they will yield negative results. *

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

Summary:

In this study, the authors investigated the role of the PIDDosome during cardiomyocyte polyploidization. PIDDosome is a multi-protein complex activating the endopeptidase Caspase-2, and shown to be involved in eliminating cells with extra centrosomes or in response to genotoxic stress (Burigotto & Fava, 2021, Sladky and Villunger, 2020). In both cases, the PIDDosome is recruited in a ANKRD26-dependent manner at the centrosomes leading to p53 stabilization and cell death (Burigotto & Fava, 2021; Evans et al., 2020; Burigotto et al., 2021).

Here, by studying mouse cardiomyocyte differentiation, the authors showed that PIDDosome is imposing ploidy restriction during cardiomyocyte differentiation. Importantly, in contrast to a previous report in the liver (Sladky et al., 2020), they showed that PIDDosome acts in a p53-independent manner in cardiomyocytes. Indeed, they suggested that PIDDosome controls ploidy in cardiomyocytes through p21 activation.

Major comments:

In general the conclusions of the authors are well supported by the experiments. However, I would suggest the following experiments/analysis to strengthen the paper:

• The authors should improve the Figure 1 to help the readers who are not familiar with cardiomyocyte polyploidization. For instance, I would suggest to add a scheme to summarize cardiomyocyte polyploidization (in terms of nuclear size, mono vs multi and so on). Based on the images they presented in 1B, the authors should also measure the nuclear area or volume in the different conditions in which components of the PIDDosome were depleted. Indeed, these two parameters should be easier to conceptualize for the readers (instead of the fluorescence nuclear intensity). This could help to understand if the nuclear size is maintained between the different conditions and if this is comparable between mono, bi or multinucleated cardiomyocytes.
• In Figure 2A, the authors presented cross section of heart from animals showing that PIDDosome depletion has no effect on heart size. This is a surprising results since cardiomyocytes have higher ploidy levels and this could have an effect on their function. Since the importance of this observation, the authors should present a quantification of the heart size in the different conditions shown in Figure 2A. Also, the percentage of cardiomyocytes presenting higher levels of ploidy seems quiet low. The authors should discuss this point. In particular because this could explain the absence of consequences on heart size and function at steady state.
• In Figure 2D, the authors measured the cardiomyocyte cross-sectional area and concluded that removing PIDDosome components have no effect on cardiomyocyte cell size. Since it has been shown that ploidy increase is normally associated with an increase in cell area, the authors should measure cell area of cardiomyocytes analyzed in Figure 1B. It could be then interesting to establish a correlation with nuclear area and the mono, bi or multinucleated status. This will strengthen the results showing that ploidy increases without affecting cell area.
• The authors should discuss the fact that PIDDosome depletion lead only to a mild increase in ploidy levels (4N) in a small percentage of cardiomyocyte. If the PIDDosome is controlling ploidy, one could expect that removing it should lead to a drastic increase in the ploidy levels. Is PIDDosome depletion leading to cell death in some cardiomyocyte? The authors should discuss this point in the discussion or if relevant show a staining with an apoptosis marker. Is another mechanism compensating to prevent higher ploidy levels in cardiomyocytes?
• Even if the authors presented RNAseq data suggesting that the PIDDosome is activated during cardiomyocyte differentiation, they should clearly demonstrate this point to strengthen the message of the paper. Indeed, the conclusions are based on the absence of PIDDosome components triggering higher ploidy in cardiomyocytes. However, we don't know whether (and when) the PIDDosome is activated during cardiomyocyte differentiation to control their ploidy levels. I would suggest to analyze PIDDosome activation markers by immunofluorescence in cardiomyocytes at different developmental stages.

Concerning the methods, the authors must add the references for each product they used and not only the origin. When relevant, the RRID should be indicated. Without this information the method and the data cannot be reproduced.

The statistics are well indicated in the figures and in the figure legends.

Minor comments:

In general, the text and the figures are clear. Nevertheless, I would suggest the following changes:

• Figures 1B, 2B and 2C: the y-axis must start at 0.
• Figure 4A: The authors should stain centrosomes in cardiomyocytes. This should strengthen the conclusion taken by the authors based on the results obtained in mice depleted for ANKRD26. Indeed, for the moment they are insufficient to conclude about the role of the centrosomes. The authors should show that centrosomes cluster in cardiomyocytes (a condition necessary for PIDDosome activation in polyploid cells) and if possible that component of the PIDDosome are recruited here.
• Figure 4F: I would suggest to modify the working model to emphasize more the differences between WT and PIDDosome KO.

The prior studies are referenced appropriately.

Significance

How polyploid cells control their ploidy levels during differentiation remains poorly understood. The data presented here represent thus an advance concerning this question.

The actual model concerning PIDDosome activation relies on the presence of extra centrosomes that drives the ANKDR26-dependent recruitment of the PIDDosome. Then, Caspase 2 is activated leading to a p53-p21 dependent cell cycle arrest (Burigotto & Fava, 2021, Sladky and Villunger, 2020; Janssens & Tinel, 2012; Evans et al., 2020; Burigotto et al., 2021). In this study, the authors showed that similar pathway takes place during cardiomyocyte differentiation to control ploidy levels. These data are reminiscent of previous work showing PIDDosome involvement during hepatocyte polyploidization (Sladky et al. 2020). Together, these data highlight the prominent role of the PIDDosome complex in controlling ploidy levels in physiological context.

Importantly, this study identified that the classical p53-dependent cell cycle arrest described after PIDDosome activation is not involved here. Instead, the data established that independently of p53, p21 contribute to control cardiomyocyte ploidy. In consequence, this study extend the initial pathway associated with PIDDosome activation and suggest that other mechanisms could take place to restrain cell proliferation upon PIDDosome activation.

Overall, this makes this paper significant and of interest for the following fields: polyploidy, heart/cardiomyocyte development and PIDDosome.

My field of expertise includes polyploidy, cell cycle and genetic instability.

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

Summary:

This manuscript by Leone et al describes the role of the PIDDosome in cardiomyocytes. Using a series of whole body and cardiomyocyte specific knockouts, the authors show that the PIDDosome maintains correct ploidy in these cells. It achieves this through inducing cell cycle arrest in cardiomyocytes in a p53 dependent manner. Despite this effect on ploidy, PIDDosome-deficient hearts show no structural or functional defects. Statistics and rigor appears to be adequate.

Major comments:

1. Figure 1 uses fluorescent intensity of a nuclear stain to determine ploidy per nucleus and they further separate the results into mononucleated, binucleated or multinucleated cells. It is hard to know how to interpret these results without further information or controls. Is there a good positive control that can be used to help to show whether this assay is quantitative. The differences are larger with the Raidd and caspase-2 knockouts than with the Pidd knockouts but this is not addressed.
2. On line 459 the authors state that the increase in polyploidy in PIDDosome knockouts occurs in adult hood but this is not directly tested. In fact in the next section the polyploidy is assessed in early postnatal development. This statement should be explained or removed
3. In Figure 4. The authors obtained RNAseq data for P1, P7 and P14 but only show the differences with and without caspase-2 at P7. Given that the differences in ploidy are more significant at P14 (Fig 3D), all the comparisons should be shown along with analysis of whether the same genes/gene families are altered in the absence of caspase-2
4. Some validation of the RNAseq and/or proteomics results would be an important addition to this study
5. Figure 4D: the authors make the conclusion that p21 is downstream of PIDD (et p53 independent). However, this is not supported by the data because the increase in 4N cells/decrease in 2N cells, although statistically significant, is nowhere near that of caspase-2 KO and caspase-2/p21 KO. Statistics should also compare p32KO with c2KO. In the absence of any other data, the more likely conclusion is that p21 is not involved.

Minor comments:

Suggest moving Figure 4A to Figure 3 as it seems to fit better there based on the citation of this figure in the text Recommend enhancing the brightness of microscopy images in Figure 1E and 2D

Significance

This study provides interesting information for the role of the PIDDosome in protecting from polyploidy and adds to the body of work by this same group studying this pathway in the liver. The main weakness in terms of significance is the lack of a phenotype in the hearts of these animals. Therefore, it is clear that ploidy (or at least PIDDosome dependent ploidy) has minimal impact on cardiac development. The authors mention that they have not tried giving these mice an MI or inducing any other type of cardiac damage. Although it is understood that these experiments are likely outside of the scope of the present study, without this information the impact of this study is moderate. I recommend expanding the discussion to provide a more indepth possible rationale as to why ploidy perturbations do not lead to structural changes like in the liver. Despite this, the insights to the pathway itself are interesting to investigators in the caspase-2 field if a little underdeveloped, especially concerning the role of p21.

My expertise is in cell death and caspase biology (especially caspase-2). I havesufficient expertise to evaluate all parts of this paper.

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

The response appears in a PDF document, which will be easier to read than plain text

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

This article investigates the phenomenon of intracellular protein agglomeration. The authors distinguish between agglomeration and aggregation, both physically characterising them and developing a simple but elegant assay to differentiate the two. Using microscopy and structural analysis, this research demonstrates that unlike aggregates, agglomerates retain their folded structures (and are not misfolded), and do not colocalise with chaperones or interact with the proteostasis machinery which targets and breaks down misfolded proteins. The inert nature of agglomerates was further confirmed in fitness assays, though they were observed to disrupt the yeast proteome. Overall, agglomerated proteins were described and characterised, and shown to be largely neutral in vivo.

The claims and conclusions were well supported by the data. Microscopy and CD spectra (previously published) were used to confirm the nature of agglomerates and to rule out colocalistion with proteostasis machinery. This was confirmed by testing ubiquitination.

The fitness of yeast cells carrying enzymatically-inactive agglomerates was assayed by generating growth curves over 24 hours. The growth rate and doubling time were taken from these growth curves as a proxy for relative fitness. The authors mention not wanting to mask differences in lag, log or stationary phases between mutants. This could be achieved by using the area under each growth curve, rather than growth rate or doubling time alone. No further experimentation would be needed, and area under the curve may provide a more holistic metric to measure fitness by.

The results indicate that agglomerates confer a slight fitness advantage. The authors do not speculate on a reason for this. I would be interested to know why they thought this might be.

Referees cross-commenting

I have read the reports from the other reviewers and agree with their comments.

Significance

Protein filamentation is observed across the tree of life, and contributes greatly to cell structure and organisation (Wagstaff, J., Löwe, J. Prokaryotic cytoskeletons: protein filaments organizing small cells. Nat Rev Microbiol 16, 187-201 (2018).). Recent work in this field has shown that self-assembly is also important for enzyme function (S. Lim, G. A. Jung, D. J. Glover, D. S. Clark, Enhanced Enzyme Activity through Scaffolding on Customizable Self-Assembling Protein Filaments. Small 2019, 15, 1805558.). Previous work from several of these authors demonstrated that the ability of a protein to filament is subject to selection (Garcia-Seisdedos H, Empereur-Mot C, Elad N, Levy ED. Proteins evolve on the edge of supramolecular self-assembly. Nature. 2017 Aug 10;548(7666):244-247. doi: 10.1038/nature23320. Epub 2017 Aug 2. PMID: 28783726.). It has become increasingly clear that protein assemblies are ubiquitous, evolvable and perhaps overlooked in research.

This research explores a specific type of filamentation, named agglomeration, unique in that the protein which assemble are not misfolded (Romero-Romero ML, Garcia-Seisdedos H. Agglomeration: when folded proteins clump together. Biophys Rev. 2023;15: 1987-2003.). This is particularly of biomedical interest due to its role in disease, such as sickle cell anaemia (J. Hofrichter, P.D. Ross, & W.A. Eaton, Kinetics and Mechanism of Deoxyhemoglobin S Gelation: A New Approach to Understanding Sickle Cell Disease*, Proc. Natl. Acad. Sci. U.S.A. 71 (12) 4864-4868, https://doi.org/10.1073/pnas.71.12.4864 (1974).) The current research adds to the field by specifically exploring agglomerates in the most detailed methodology to date.

The novelty of this research lies especially in two areas; (1) establishing a method for distinguishing between aggregation and agglomeration, and (2) the finding that agglomerates are largely innocuous in vivo. The method established for defining agglomerates is simple, elegant and well-described in this paper's methods. The authors then probe cellular responses to agglomeration via both proteostasis machinery and cellular fitness. They noted no disruption to fitness and observed little targeting of agglomerates by chaperones. The experiments were thorough, conclusive, and resulted in interesting findings.

The inertia of this type of protein filament is unexpected; agglomerates are large and have been associated with disease. The results of this study, however, indicate that agglomerates are non-toxic and well-tolerated in vivo. The authors speculate that agglomerates may have evolved in a non-adaptive process, which is evolutionary very interesting. They also posit that these results could lead to synthetic biology applications such as a tracking expression or as a molecular sensor. This work is of great interest and impact both in cell biology, biomedicine and in-vivo biology.

Personal note: I come from a background of enzyme evolution and have viewed the work in this light.

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

This is a very interesting paper investigating the fitness and cellular effects of mutations that drive dihedral protein complex into forming filaments. The Levy group have previously shown that this can happen relatively easily in such complexes and this paper now investigates the cellular consequences of this phenomenon. The study is very rigorous biophysically and very surprisingly comes up empty in terms of an effect: apparently this kind of self-assembly can easily be tolerated in yeast, which was certainly not my expectation. This is a very interesting result, because it implies that such assemblies may evolve neutrally because they fulfill the two key requirements for such a trajectory: They are genetically easily accessible (in as little as a single mutation), and they have perhaps no detrimental effect on fitness. This immediately poses two very interesting questions: Are some natural proteins that are known to form filaments in the cell perhaps examples of such neutral trajectories? And if this trait is truly neutral (as long as it doesn't affect the base biochemical function of the protein in question), why don't we observe more proteins form these kinds of ordered assemblies.

I have no major comments about the experiments as I find that in general very carefully carried out. I have two more general comments:

1. The fitness effect of these assemblies, if one exists, seems very small. I think it's worth remembering that even very small fitness effects beyond even what competition experiments can reveal could in principle be enough to keep assembly-inducing alleles at very low frequencies in natural populations. Perhaps this could be acknowledged in the paper somewhere.
2. The proteins used in this study I think were chosen such that they do not have an important function in yeast that could be disrupted by assembly This allows the effect of the large scale assemblies to be measured in isolation. If I deduced this correctly, this should probably be pointed out agin in this paper (I apologise if I missed this).
3. The model system in which these effects were tested for is yeast. This organism has a rigid cell wall and I was wondering if this makes it more tolerant to large scale assemblages than wall-less eukaryotes. Could the authors comment on this?

Minor points:

In Figure 2D, what are the fits? And is there any analysis that rules out expression effects on the mutant caused by higher levels of the wild-type? The error bars in Figure 2E are not defined.

Significance

This is a remarkably rigours paper that investigates whether self-assembly into large structures has any fitness effect on a single celled organism. This is very relevant, because a landmark paper from the Levy group showed that many proteins are very close in genetic terms to forming such assemblies. The general expectation I think would have been that this phenomenon is pretty harmful. This would have explained why such filaments are relatively rare as far as we know. This paper now does a large number of highly rigours experiments to first prove beyond doubt that a range of model proteins really can be coaxed into forming such filaments in yeast cells through a very small number of mutations. Its perhaps most surprising result is that this does not negatively affect yeast cells.

From an evolutionary perspective, this is a very interesting and highly surprising result. It forces us to rethink why such filaments are not more common in Nature. Two possible answers come to mind: First, it's possible that filamentation is not directly harmful to the cell, but that assembling proteins into filaments can interfere with their basic biochemical function (which was not tested for here).

Second, perhaps assembly does cause a fitness defect, but one so small that it is hard to measure experimentally. Natural selection is very powerful, and even fitness coefficients we struggle to measure in the laboratory can have significant effects in the wild. If this is true, we might expect such filaments to be more common in organisms with small effective population sizes, in which selection is less effective.

A third possibility is of course that the prevalence of such self-assembly is under-appreciated. Perhaps more proteins than we currently know assemble into these structures under some conditions without any benefit or detriment to the organism.

These are all fascinating implications of this work that straddle the fields of evolutionary genetics and biochemistry and are therefore relevant to a very wide audience. My own expertise is in these two fields. I also think that this work will be exciting for synthetic biologists, because it proves that these kinds of assemblies are well tolerated inside cells.

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

In this work, the authors used yeast cell as a model system to study the abovementioned question. They established a model protein system using fluorescently labeled proteins that can form both agglomerates and aggregates. Using imaging experiments, they arguably showed that agglomerates do not colocalize with the proteostasis machinery, echoing what was observed by proteomics results. The proteomics results after pull down assay to study the interactome revealed that agglomerate-size-dependent changes were dependent on the cell-wall and plasma-membrane proteins. On the other hand, as expected, the misfolded proteins (aggregates) showed heavy involvement of proteostasis network components.

Although the experiments still lack some controls and failed to support some of the conclusions, I found this work is a nice complement of the field to emphasize the point that "aggregates" and "agglomerates" are two different states, which is often mistaken by lots of researchers in recent years, in particular with the membraneless organelles (LLPS). I support its publication after the authors may consider the following suggestions and make necessary improvement.

Major concerns:

My major concern was raised by the lack of evidence to support the model system's folding state in the cell. 1. In Figure 1 and 2, I found the evidence to distinguish the folded state of proteins in the cells was limited. The concept of using hybrid imaging technique to prove the folding state is not a common experiment. The description of Figure 2 was very limited. I am sure the general audience can be convinced that the model proteins were actually folded and form agglomeration. 2. In addition, for mutants formed aggregates, the authors may consider to perform fractionation or crosslinking or native page experiment to show the evidence of protein misfolding and aggregation. 3. Have the authors considered to use FRAP assay to distinguish "aggregates" and "agglomerates" states in the cell? Does each of the state display different dynamics in the cell?

Minor concerns:

1. In Figure 3, it is very interesting to see such patten. I wonder why some of the chaperones were not responsive to misfolded proteins but some were very addicted to proteostasis. Could you elaborate more on this point? Are they chaperone sensitive, namely selective to 60/10, 70/40 or 90 system?
2. In Figure 6, I suggest to add GO analysis and KEGG analysis to distinguish pathways and functional mismatch between "aggregates" and "agglomerates" interactomics.
3. The quantity control for the proteomics studies is needed, namely the reproducibility of 3 repeats?
4. This may beyond the scope of this work. I am interested whether the authors could point out whether similar works can be done in mammalian cells. What is the model system for mammalian cell that can form "agglomerates".

Referees cross-commenting

I read through the other two reviewers' comments, which I found reasonable. It seems like all reviewers agreed that this work is of enough significance for the field only with several technical concerns.

Significance

The submitted manuscript emphasized on a very important but often misleading concept: "aggregates" and "agglomerates" are two different states of protein structures in the cell with distinct physiological roles. However, these two states are of very similar phenotype: punctate structure in the cell. While the proteostasis network has been well-established for its central role of protein quality control and coping with misfolded and aggregated proteome, the authors attempted to profile the mechanism and physiological impact of mutation-induced folded-state protein filamentation, namely a model of "agglomerates". Such overarching goal of this work clearly pointed out the novelty of this work. Clearly, this is a new angle and aspect remained to be clarified for the field.

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

Manuscript number: RC-2025-02887

Corresponding author(s): Philippe Bastin

1. General Statements [optional]

• *

We thank the reviewers for their constructive suggestions. We are delighted to see that they appreciated our work and its interest for the broad cell biology community, as well as the potential impact of the inducible expression of tagged tubulin as a new tool to investigate microtubule assembly at large.

We are now providing a full revision that contains two major modifications and that addresses all the minor points detailed below. The two major modifications are:

• A simplification and a shortening of the text as requested by reviewers 1 and 3
• The addition of a new experiment evaluating the role of the locking protein CEP164C to gain insight into the mechanism, as suggested by reviewers 1 and 2 Briefly, CEP164C is a protein localised to the transition fibres (structures that dock the basal body of the flagellum to the membrane) of only the old flagellum. Its depletion leads to an excessive elongation of the old flagellum and the production of a shorter new flagellum, suggesting competition between the two flagella for tubulin incorporation (Atkins et al., 2021). In the new figure 5, we have expressed tagged tubulin in the CEP164CRNAi cell line and formally demonstrated simultaneous incorporation in both flagella. Unexpectedly, the new flagellum incorporated more tubulin than the old one, suggesting a bias of tubulin targeting in favour of the new flagellum and the existence of additional contributors to the Grow-and-Lock model.

2. Point-by-point description of the revisions

This section is mandatory. *Please insert a point-by-point reply describing the revisions that were already carried out and included in the transferred manuscript. *

• *

Reviewer #1

Evidence, reproducibility and clarity

The manuscript by Daniel Abbühl on "A novel approach to tagging tubulin reveals MT assembly dynamics of the axoneme in Trypanosoma brucei" uses an innnovative approach to label tubulin, which allows the authors to unveil new mechanisms in flagellar length regulation.

The manuscript is very nice and will be very interesting for the cell biology community and therefore should be accepted. In some parts it becames a bit complex with all the models and complex phrasing, I wonder whether the text could be simplified to be more appealing. I have a few minor comments:

We agree that some of the explanations are lengthy and complex. We have simplified the explanations and hopefully made the models more accessible. Complexity comes from the fact that trypanosomes do not have a synchronized cell cycle.

-From the model the authors show in Figure 8- there should be a way of pulsing the cells in G1 for a short amount of time -2 hours- and getting both flagella tips labelled. But the authors seem to require longer labelling to get that result. This should be better explained.

We are not quite sure what is meant here with both flagella as in G1-phase, all cells are mono-flagellated. We do see mono-flagellated cells with a labelled tip after 2 hours, both with the HALO-tag or the Ty-1-tubulin system.

In regard to bi-flagellate cells, we believe that incorporation in the OF happened at the beginning of G1-phase when the cell was mono-flagellated. If tubulin is present at that point, it will be incorporated at the tip. This cell then approaches the end of G1-phase and starts to initiate NF assembly. Since tagged tubulin is already present it will be incorporated along the whole length of the NF.

A short induction of 2h would not suffice as it wouldn't cover the duration of the G1-phase and the initiation of a NF (duration of G1-phase is ~4h). We attempted to explain this in Fig. 4 and reworked the text to make this clearer.

-Why do some cells not express the construct? Weren´t they all selected?

We never managed to get a cell line where inducible expression is present in 100% of cells. Here, around 95% of cells were positive for Ty-1-tubulin after 24h of induction. Non-expression is not a phenomenon restricted to this tubulin cell line but also observed with other ectopically expressed proteins (e.g. Sunter et al. JCS 2015, Bastin et al. MCB 1999). All these cell lines represent clonal populations and are resistant to antibiotic treatment, however not all cells express the respective protein. For each experiment where we believed the number of expressing cells matter (for example the washout), we quantified in how many cells Ty-1-tubulin was present in the cell body microtubules.

-"The linear regression line in Fig. 3C was corrected by subtracting 45 minutes from each timepoint due to the previously reported delay between addition of tetracycline and the expression of the respective protein". However, in the authors data the delay may amount to one hour (western analysis- S4). Shouldn´t they use their data.

Indeed, the western blot shows expression after 1-hour, however we did not take a 45-minute timepoint, so we don't know if the protein was detectable at that time. In addition, IFA is more sensitive than western blot. We cannot say exactly when the average cell starts to express the induced protein.

-Fig 3: To measure the timepoints of flagella growth, wouldn´t it be better to do it with NF that started to grow before induction, rather than starting to grow after induction, to be sure that the timing of incorporation is fully accounted for?

We indeed did consider only NFs, which started to grow before induction, as suggested by the reviewer. In the revised version the description of the experiment can be found on page 9 line 22 - 28.

-Although it is not the focus of the manuscript it would have been very interesting to use the CEP164C mutant to see whether it would change the dynamics of incorporation and fully test their model and discussion.

This is a great suggestion, so we performed some experiments to address this issue. When CEP164C was knocked down before Ty-1-tubulin expression, integration is seen at the distal tip of both NF and OF. This is coherent with the idea of removal of the locking protein from the OF. However, lengths of the green segments in NF and OF do not have the same length (NF ~6 µm, OF ~2 µm), which indicates that CEP164C might not be the only protein involved in regulating flagellum length. A new figure explaining this experiment was added (Fig. 5, Fig. S6). We believe this data provides novel insight on the locking mechanism and strengthens the manuscript.

-In some parts of the manuscript/supplemental material the authors say they insert the Ty-1- tag one aminoacid after the acetylated lysine- other parts they say two aminoacids after- this should be consistent.

We thank the reviewer for spotting these mistakes, we have changed the text accordingly.

-Fig. S1: 'Binding epitope of the TAT-1 antibody is highlighted in red'. There is no highlighting in red in this figure?

This sentence was removed.

-Fig. S2: Western blots are not very clear. What is the 'X' present in the C (first lane)? Weight of markers should be shown also in S4.

Molecular weight markers have been added. X is an empty lane, we have now indicated this in the figure legend.

-Fig 5: 'C: Frequency of bi-flagellated cells grouped by the different types of' The authors didn't finish the sentence.

Previous Fig. 5 is now Fig. 6. Sentence has been completed. "Frequency of bi-flagellated cells grouped by different types of old flagella"

-Fig. S7: The 'B' is missing in both picture and legend.

This has been added

---

Significance

This study advances our knowledge of flagellar length regulation and maintenance. Moreover, the tools designed in this work will be very useful for the cell biology community in general.

---

Reviewer #2

Evidence, reproducibility and clarity

Summary: The length of the old flagellum of Trypanosome is constant during G1 phase as well as during cell cycle progression when the new flagellum is assembled. The authors have previously proposed a "Grow and Lock" model for the flagellar length control in which no flagellar building blocks are incorporated. To test this hypothesis, the authors used a tagging strategy for alpha-tubulin and tracking its incorporation. The authors showed that the new flagellum incorporates new tubulins, as is expected. For the mature flagellum, tubulins are incorporated at the flagellar tip and only when the cells start to assemble the new flagellum. Thus, it shows that old flagellum is stable but not completely locked for the incorporation of tubulins.

Major comments: The study is methodologically rigorous, integrating fluorescence microscopy, biochemical approaches, and proteomic analyses to validate the functionality of the tagged tubulin. The use of both inducible expression and endogenous protein tagging (HaloTag) strengthens the conclusions. This study has supported the "Grow-and-Lock" model" that the authors previously proposed. In addition, they have revealed that the stability of the old flagellum is temporally controlled.

The data showed that brief incorporation of tubulins at the tip of the old flagellum occurs when the cells start to form the new flagellum. I thought the assembly of the new flagellum occurs during the cell division. However, in the abstract, it says that "The restriction is lifted briefly after the bi-flagellated cell has divided." Is my understanding wrong?

We believe incorporation at the tip of the "OF" occurred after the cell has divided, when the OF daughter is mono-flagellated. It happens before this daughter cells starts assembling its new flagellum is formed. Of course, when looking at biflagellated cells, the NF as well as the tip of the OF will be green, but our data supports that incorporation happened in G1-phase and not during the biflagellated stage as the lock seals the OF before the NF emerges. To clarify on terminology: The bi-flagellate stage begins when basal bodies are duplicated, shortly after the beginning of S-phase and ends with cytokinesis. This means G1-phase and the mono-flagellated stage are nearly the same (Woodward and Gull, JCS1990) and occupy ~40% of the cell cycle.

P12, "The cartoon in Fig. 5A illustrates the progression of the cells in scenario 2 (Fig. 4A) over the duration of one cell cycle (~9 hours)" I thought that one cell cycle should start with cell with only one flagellum, followed by assembly of a new flagellum during cell division, the cell then divides when the new flagellum is almost completely assembled. If my understanding is correct, perhaps the cartoon should be modified accordingly.

Indeed, the cell cycle starts with a cell in G1-phase. Here, we have chosen the initiation of a NF assembly as our starting point because we focused the investigation on bi-flagellated cells. We have now illustrated the cell cycle (adapted from Woodward and Gull 1990) and when cells are biflagellated in Fig. 6A (revised version).

Minor comments:

1) Several references are not correctly formatted. P3: (Flavin and Slaughter, 1974) (Rosenbaum 1969). P10, (Sherwin et al., 1987)(Sheriff et al., 2014) 2) In several places there are no space between the number and the unit. For eample, P3, 9 - 24µm/h. 7, 1μg/m; P8, 50kDa; P9, 1M; 8-9h; P11, 2.9µm/h and etc. 3) P11, Flagella were extracted. I thought the cells were extracted.

Thank you for pointing these out, we have changed these in the text.

---

Significance

Cilia and eukaryotic flagella are considered dynamic structures in which the flagellar components especially tubulins under constant turnovers even in steady state. This work demonstrates that in Trypanosome the stable old flagellum is temporally controlled for tubulin turnovers, suggesting a tight regulation of microtubule dynamics. Future elucidation of the regulatory mechanism will be more interesting. This work will be interesting to the field of cilia and microtubules. In addition, the new technique used for tracking tubulins will also be interesting.

I am an expert on ciliary biology.

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

Summary:

This study seeks to investigate the mechanism by which the length of an eukaryotic cilium is set and maintained in a constant state. The flagellated protist Trypanosoma brucei serves as the study model and the authors take advantage of the genetic tools that allow precise modification and tagging of flagellar proteins and they build on prior knowledge about the well-characterised flagellar assembly cycle, which allows tracking the assembly of a new flagellum alongside an existing old one in the course of one cell cycle. The group of Bastin has previously reported a very interesting "Grow-and-Lock Model for the Control of Flagellum Length in Trypanosomes" and this current manuscript provides a test of this model, and a refinement. Key to this is an advance in technique, reported here, namely expression of an epitope tagged version of alpha tubulin. The epitope is inserted in an internal loop, which apparently for the first time provides a traceable tubulin that is reliably incorporated into the cytoskeleton (subpellicular array, spindle and cilium). Expressing an inducible version of this Ty-1-tubulin allows for a set of experiments that measure the place and timing of tubulin incorporation into cilia. The results are largely confirmatory of previous findings (incorporation exclusively into the new flagellum, at the distal end, linear growth rate that matches previous estimates). Examination of tubulin incorporation patterns then reveal additional information about the old flagellum: evidence from Ty-1-tubulin labelling, corroborated by incorporation patterns of another ciliary protein (RSP 4/6) suggest that the "lock" on the old flagellum is relieved for short periods after cell division, leading to a refined model presented in Figure 8.

Major comments:

This study provides an elegant test of the grow-and-lock model and the major conclusions are supported by the data. I have no major concerns.

Minor comments:

There are several minor points that could be addressed to make the manuscript easier to follow (and adding line numbers to the manuscript would help with reviewing).

The introduction is quite long. Some of the well-established background information on the T. brucei cell cycle could be shortened. If the paper is intended for a broader audience, it would be valuable instead to cite studies that have succeeded in tagging tubulin and tracing its incorporation in other cilia. Could the Ty-1-tubulin approach be relevant more broadly or are simpler methods already established?

The introduction has been shortened, we now also cite two published studies that tracked tubulin integration in Chlamydomonas and C. elegans respectively.

On p.6 the rationale for endogenous tagging was to "reduce the risk of artifacts portentially due to untimely expression or unnatural protein levels". However most of the experiments were done with ectopically expressed inducible Ty-1-tubulin. For the experiments it is crucial to use an inducible system but the authors may wish to comment why the risk of artifacts was no longer a concern.

The reasoning here was that in case the Ty-1-tubulin would not have been incorporated into MTs, we could have attributed it solely to the presence of the tag and no other factors, but this was not the case. This therefore allowed us to move to the inducible expression system.

On p.7 / Fig S2A-B there appears to be a mistake in the presentation. Spindles are mentioned in the text - I can't see any in the figure. Fig S2A and B both show cytoskeletons, but the text suggests only B is about cytoskeletons. None of the blot shows BB2 staining of different cell fractions, contrary to statements in the text. The letter codes in the panel (T, C, D) don't match the codes in the legend (T, P, S).

We thank the reviewer for spotting the mistakes. A panel with the spindle was added in Fig. S2. We did not stain fraction blots of the in-situ tagged cell lines with BB2. However, this was done with the inducible cell line and is shown in Fig. 1D. Letter code in the legend was adapted to match the figure.

Figure 1. The evidence for incorporation into spindles is not strong. The structure indicated by the arrive could be a spindle but it's not very clear. There is a great example of a labelled spindle only in figure S5A. Here, at the start, it would be good to show a panel of cells in successive cell cycle stages (best, whole cells and cytoskeletons) to clearly show the structures that are labelled with Ty-1-tubulin.

The current Fig. 1B (Fig. 1A before) depicts whole cells of an induced and a non-induced culture; we show whole cells to provide a complete picture of tubulin integration. A panel with detergent extracted cytoskeletons from the in situ tagged cell line has been added to Fig. 1A. We chose to show cytoskeletons or isolated flagella instead of whole cells because (1) the flagella are easier to see and (2) it formally demonstrates that tagged tubulin is incorporated in MTs.

In general, tubulin labelling of the spindle was more consistently observed in whole cells as we did not use spindle preserving extraction buffers when preparing cytoskeletons. However, we did observe clear spindles in cytoskeletons as well (see Fig. S5 for example). The same was observed for the beta-tubulin specific KMX1 antibody in the past which is the gold standard to visualize the spindle (Sasse and Gull JCS1988). Regardless, a panel depicting spindle progression through mitosis using staining of Ty-1-tubulin has been added in Fig. S2 (The panel is a mix of whole cells and cytoskeletons).

On p.8 (end of first paragraph) there is reference to cell cycle analyses, but no data is shown. Also on p.8, please clarify what the evidence is that "a fraction of cells did not respond to tetracycline". The fact that they remain unstained by Ty-1-tubulin is not in itself evidence they did not respond to tetracycline.

We did not show the cell cycle data as it was similar to non-induced and does not provide any new information in our opinion. Hence, the sentence has been removed.

The reviewer is correct that we do not have evidence that these cells did not respond to tetracycline. Some cells remained completely devoid of Ty-1-tubulin even after multiple days of induction. This was typically between 5-10% of cells. In experiments where the exact number is important, we counted the amount of "non-expressers" in whole cells.

Figure S4A. The blot for the soluble fraction is not of great quality. I don't see how the conclusion was reached that the Ty-1-tubulin bands were faint.

The blot of the soluble fraction that was stained with BB2 had to be exposed a lot longer compared to the blot stained with TAT-1. The soluble blots were repeated with the same result (lots of background noise when using BB2, a clear blot with TAT-1). In the TAT-1 blot only the endogenous tubulin band is clearly visible, with some very faint signal above corresponding to the Ty-1-tubulin. Soluble Ty-1-tubulin with BB2 or TAT-1 is visible in Fig. 1D after longer inductions.

On p.11, it would be interesting to compare measured elongation rates with previously measured estimates for flagellum growth, comparing the growth rates, and relating them to cell cycle times in the corresponding experiments (which vary slightly between labs and studies).

We attempted to address this in the discussion by comparing our experiments to the assembly rate measured with the PFR as reporter (Bastin et al. 1999). We could mention the corresponding doubling times in correlation to how many cells are bi-flagellated, but this was only done with the Ty-1-tubulin cell line and not with the PFR. In our experiments the average doubling time was ~9 hours with 52% of cells being bi-flagellated. This was measured with FTZC (marker of the transition zone at the base of the flagellum) and Mab25 (marker of the axoneme of the flagellum) which will lead to a slight underestimate of the real number of bi-flagellated cells, as the NF is initially very close which makes it difficult to notice/differentiate from the old one.

Figure S6. I find the presentation of this figure confusing. It should be revised with clearer labelling of "cell cycle 1", "cell cycle 2", and the precise meaning of "type 3" should be clarified. There are two instances of "type 1" in the drawing, but one of these seems to fulfil the criteria of "type 3" (OF 1-4µm).

We agree with the reviewer and therefore decided to remove this figure. We also considered the comments of the other two reviewers about complexity of the manuscript and changed the text of figure 5 to make it more approachable. This includes a simpler explanation for the expected amounts of flagella.

Figure 7. In panel A, the absence of label at the NF distal end is not total, a purple line is still visible. Was any quantitation attempted (signal intensity, changes in length of labelled fragments over time?). Minimally, say how many cells were analysed for the numbers in panels D and E, and how many times this experiment was done.

We agree with the reviewer that the decrease in the TMR signal in the NF of the cell in the original Fig. 7A (currently Fig. 8A) is gradual and not abrupt. Similarly to the Ty-1-tubulin experiments where the tagged protein becomes progressively more available (increasing intensity), the intensity of TMR-ligand becomes progressively less abundant (gradually decreasing intensity) as new (not TMR labelled) protein gets synthesized during the period of NF construction, progressively diluting the initially fully labeled population of RSP4/6. The slope of the gradient may differ between axonemal constituents, as it reflects the kinetics of protein synthesis, degradation, its incorporation into the axoneme, as well as the size of the soluble protein pool in the cytosol. We classify this type of signal as gradients, as opposed to the sharp decrease. At initial times after TMR-ligand washout (e.g. 4 hours in Fig. 8C), this long gradient is observed at the distal end of NFs and in some uniflagellated cells (NF-inheriting daughters). The distal ends of OFs in these experiments (if not fully labelled) display a sharp decrease, as do frequent uniflagellated cells, likely OF-inheriting daughters. The existence of these two different patterns demonstrates that two different mechanisms are responsible for incorporation of fresh RSP4/6 into the NF and OF axoneme, respectively. While incorporation into the NF is gradual, incorporation into the distal region of the OF is stepwise (restricted in time). Numbers of cells quantified for the table in Fig. 8 have been added. The NFs and OFs displaying the patterns of the gradient and sharp decrease, respectively, were observed in multiple experiments.

Reviewer #3 (Significance (Required)):

• General assessment: strengths and limitations

Strengths: Trypanosoma brucei is a powerful model system in which to ask detailed questions about the assembly dynamics and hierarchy of microtubule-based cytoskeletal structures in general and cilia in particular. This elegant and well-designed study overcomes a previous technical limitation by allowing for the direct labelling of alpha tubulin, one of the main building blocks of the ciliary axoneme. The study sets out to test a specific hypothesis (grow-and-lock model) and provides evidence in support, leading to a refined model for cilia length regulation in trypanosomes.

Limitations: With this system, visualisation of new tubulin incorporation requires de novo synthesis. There is a time lag between inducing expression of Ty-1-tubulin with tetracycline and being able to visualize the tagged proteins that needs to be taken into consideration. This time lag was estimated based on previous studies and the relatively quick appearance of Ty-1-tubulin on Western blots (within hours). This inevitably creates a situation where levels of tagged tubulin change rapidly, creating gradients of signal intensity (and variations in levels) that lead to some uncertainty in estimations of length of labelled microtubule fragments. Furhtermore, the epitope label is not compatible with live cell imaging, restricting analyses to fixed cells. The Ty-1-tubulin data is well ducmented; the RSP4/6 data appear to corroborate these findings but are less extensively documented.

• Advance: The results succeed in integrating several recent findings from different research groups into a refined coherent model about cilia length regulation in trypanosomes. The tubulin tagging method could be gainfully transferred to other systems (although the state of the field in tubulin tagging in other systems is not clearly laid out in the paper).

This paper could be of interest to a broad cell biology community interested in cilia and cytoskeletal dynamics.

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:

This study seeks to investigate the mechanism by which the length of an eukaryotic cilium is set and maintained in a constant state. The flagellated protist Trypanosoma brucei serves as the study model and the authors take advantage of the genetic tools that allow precise modification and tagging of flagellar proteins and they build on prior knowledge about the well-characterised flagellar assembly cycle, which allows tracking the assembly of a new flagellum alongside an existing old one in the course of one cell cycle. The group of Bastin has previously reported a very interesting "Grow-and-Lock Model for the Control of Flagellum Length in Trypanosomes" and this current manuscript provides a test of this model, and a refinement. Key to this is an advance in technique, reported here, namely expression of an epitope tagged version of alpha tubulin. The epitope is inserted in an internal loop, which apparently for the first time provides a traceable tubulin that is reliably incorporated into the cytoskeleton (subpellicular array, spindle and cilium). Expressing an inducible version of this Ty-1-tubulin allows for a set of experiments that measure the place and timing of tubulin incorporation into cilia. The results are largely confirmatory of previous findings (incorporation exclusively into the new flagellum, at the distal end, linear growth rate that matches previous estimates). Examination of tubulin incorporation patterns then reveal additional information about the old flagellum: evidence from Ty-1-tubulin labelling, corroborated by incorporation patterns of another ciliary protein (RSP 4/6) suggest that the "lock" on the old flagellum is relieved for short periods after cell division, leading to a refined model presented in Figure 8.

Major comments:

This study provides an elegant test of the grow-and-lock model and the major conclusions are supported by the data. I have no major concerns.

Minor comments:

There are several minor points that could be addressed to make the manuscript easier to follow (and adding line numbers to the manuscript would help with reviewing).

The introduction is quite long. Some of the well-established background information on the T. brucei cell cycle could be shortened. If the paper is intended for a broader audience, it would be valuable instead to cite studies that have succeeded in tagging tubulin and tracing its incorporation in other cilia. Could the Ty-1-tubulin approach be relevant more broadly or are simpler methods already established?

On p.6 the rationale for endogenous tagging was to "reduce the risk of artifacts portentially due to untimely expression or unnatural protein levels". However most of the experiments were done with ectopically expressed inducible Ty-1-tubulin. For the experiments it is crucial to use an inducible system but the authors may wish to comment why the risk of artifacts was no longer a concern.

On p.7 / Fig S2A-B there appears to be a mistake in the presentation. Spindles are mentioned in the text - I can't see any in the figure. Fig S2A and B both show cytoskeletons, but the text suggests only B is about cytoskeletons. None of the blot shows BB2 staining of different cell fractions, contrary to statements in the text. The letter codes in the panel (T, C, D) don't match the codes in the legend (T, P, S).

Figure 1. The evidence for incorporation into spindles is not strong. The structure indicated by the arrive could be a spindle but it's not very clear. There is a great example of a labelled spindle only in figure S5A. Here, at the start, it would be good to show a panel of cells in successive cell cycle stages (best, whole cells and cytoskeletons) to clearly show the structures that are labelled with Ty-1-tubulin.

On p.8 (end of first paragraph) there is reference to cell cycle analyses, but no data is shown. Also on p.8, please clarify what the evidence is that "a fraction of cells did not respond to tetracycline". The fact that they remain unstained by Ty-1-tubulin is not in itself evidence they did not respond to tetracycline.

Figure S4A. The blot for the soluble fraction is not of great quality. I don't see how the conclusion was reached that the Ty-1-tubulin bands were faint.

On p.11, it would be interesting to compare measured elongation rates with previously measured estimates for flagellum growth, comparing the growth rates, and relating them to cell cycle times in the corresponding experiments (which vary slightly between labs and studies).

Figure S6. I find the presentation of this figure confusing. It should be revised with clearer labelling of "cell cycle 1", "cell cycle 2", and the precise meaning of "type 3" should be clarified. There are two instances of "type 1" in the drawing, but one of these seems to fulfil the criteria of "type 3" (OF 1-4µm).

Figure 7. In panel A, the absence of label at the NF distal end is not total, a purple line is still visible. Was any quantitation attempted (signal intensity, changes in length of labelled fragments over time?). Minimally, say how many cells were analysed for the numbers in panels D and E, and how many times this experiment was done.

Significance

General assessment: strengths and limitations

Strengths: Trypanosoma brucei is a powerful model system in which to ask detailed questions about the assembly dynamics and hierarchy of microtubule-based cytoskeletal structures in general and cilia in particular. This elegant and well-designed study overcomes a previous technical limitation by allowing for the direct labelling of alpha tubulin, one of the main building blocks of the ciliary axoneme. The study sets out to test a specific hypothesis (grow-and-lock model) and provides evidence in support, leading to a refined model for cilia length regulation in trypanosomes.

Limitations: With this system, visualisation of new tubulin incorporation requires de novo synthesis. There is a time lag between inducing expression of Ty-1-tubulin with tetracycline and being able to visualize the tagged proteins that needs to be taken into consideration. This time lag was estimated based on previous studies and the relatively quick appearance of Ty-1-tubulin on Western blots (within hours). This inevitably creates a situation where levels of tagged tubulin change rapidly, creating gradients of signal intensity (and variations in levels) that lead to some uncertainty in estimations of length of labelled microtubule fragments. Furhtermore, the epitope label is not compatible with live cell imaging, restricting analyses to fixed cells. The Ty-1-tubulin data is well ducmented; the RSP4/6 data appear to corroborate these findings but are less extensively documented.

Advance: The results succeed in integrating several recent findings from different research groups into a refined coherent model about cilia length regulation in trypanosomes. The tubulin tagging method could be gainfully transferred to other systems (although he state of the field in tubulin tagging in other systems is not clearly laid out in the paper).

This paper could be of interest to a broad cell biology community interested in cilia and cytoskeletal dynamics.

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

Summary:

The length of the old flagellum of Trypanosome is constant during G1 phase as well as during cell cycle progression when the new flagellum is assembled. The authors have previously proposed a "Grow and Lock" model for the flagellar length control in which no flagellar building blocks are incorporated. To test this hypothesis, the authors used a tagging strategy for alpha-tubulin and tracking its incorporation. The authors showed that the new flagellum incorporates new tubulins, as is expected. For the mature flagellum, tubulins are incorporated at the flagellar tip and only when the cells start to assemble the new flagellum. Thus, it shows that old flagellum is stable but not completely locked for the incorporation of tubulins.

Major comments:

The study is methodologically rigorous, integrating fluorescence microscopy, biochemical approaches, and proteomic analyses to validate the functionality of the tagged tubulin. The use of both inducible expression and endogenous protein tagging (HaloTag) strengthens the conclusions. This study has supported the "Grow-and-Lock" model" that the authors previously proposed. In addition, they have revealed that the stability of the old flagellum is temporally controlled.

The data showed that brief incorporation of tubulins at the tip of the old flagellum occurs when the cells start to form the new flagellum. I thought the assembly of the new flagellum occurs during the cell division. However, in the abstract, it says that "The restriction is lifted briefly after the bi-flagellated cell has divided." Is my understanding wrong?

P12, "The cartoon in Fig. 5A illustrates the progression of the cells in scenario 2 (Fig. 4A) over the duration of one cell cycle (~9 hours)" I thought that one cell cycle should start with cell with only one flagellum, followed by assembly of a new flagellum during cell division, the cell then divides when the new flagellum is almost completely assembled. If my understanding is correct, perhaps the cartoon should be modified accordingly.

Minor comments:

1. Several references are not correctly formatted. P3: (Flavin and Slaughter, 1974) (Rosenbaum 1969). P10, (Sherwin et al., 1987)(Sheriff et al., 2014)
2. In several places there are no space between the number and the unit. For eample, P3, 9 - 24µm/h. 7, 1μg/m; P8, 50kDa; P9, 1M; 8-9h; P11, 2.9µm/h and etc.
3. P11, Flagella were extracted. I thought the cells were extracted.

Significance

Cilia and eukaryotic flagella are considered dynamic structures in which the flagellar components especially tubulins under constant turnovers even in steady state. This work demonstrates that in Trypanosome the stable old flagellum is temporally controlled for tubulin turnovers, suggesting a tight regulation of microtubule dynamics. Future elucidation of the regulatory mechanism will be more interesting. This work will be interesting to the field of cilia and microtubules. In addition, the new technique used for tracking tubulins will also be interesting.

I am an expert on ciliary biology.

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 manuscript by Daniel Abbühl on "A novel approach to tagging tubulin reveals MT assembly dynamics of the axoneme in Trypanosoma brucei" uses an innnovative approach to label tubulin, which allows the authors to unveil new mechanisms in flagellar length regulation.

The manuscript is very nice and will be very interesting for the cell biology community and therefore should be accepted. In some parts it becames a bit complex with all the models and complex phrasing, I wonder whether the text could be simplified to be more appealing. I have a few minor comments:

• From the model the authors show in Figure 8- there should be a way of pulsing the cells in G1 for a short amount of time -2 hours- and getting both flagella tips labelled. But the authors seem to require longer labelling to get that result. This should be better explained.
• Why do some cells not express the construct? Weren´t they all selected?
• "The linear regression line in Fig. 3C was corrected by subtracting 45 minutes from each timepoint due to the previously reported delay between addition of tetracycline and the expression of the respective protein". However, in the authors data the delay may amount to one hour (western analysis- S4). Shouldn´t they use their data.
• Fig 3: To measure the timepoints of flagella growth, wouldn´t it be better to do it with NF that started to grow before induction, rather than starting to grow after induction, to be sure that the timing of incorporation is fully accounted for?
• Although it is not the focus of the manuscript it would have been very interesting to use the CEP164C mutant to see whether it would change the dynamics of incorporation and fully test their model and discussion.
• In some parts of the manuscript/supplemental material the authors say they insert the Ty-1- tag one aminoacid after the acetylated lysine- other parts they say two aminoacids after- this should be consistent.
• Fig. S1: 'Binding epitope of the TAT-1 antibody is highlighted in red'. There is no highlighting in red in this figure?
• Fig. S2: Western blots are not very clear. What is the 'X' present in the C (first lane)? Weight of markers should be shown also in S4.
• Fig 5: 'C: Frequency of bi-flagellated cells grouped by the different types of' The authors didn't finish the sentence.
• Fig. S7: The 'B' is missing in both picture and legend.

Significance

This study advances our knowledge of flagellar length regulation and maintenance. Moreover the tools designed in this work will be very useful for the cell biology community in general.

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

We thank the reviewers for providing valuable comments and suggestions for improving the manuscript.

Response to reviewer comments:

Reviewer-1

Comment 1: Major concern is the study lacks rigor in several areas where n=2, results are not quantified with statistics. They need to run power analysis and increase their samples sizes. Please include statistics on all measurements. Filamentous actin staining and alpha-sma is used to visualize mechanosensing but also in other cell activities such as cell contractility for movement, cell to substrate adhesion, cell division, etc. They need to query more mechanosensing related pathways (Piezo1/2, Yap/taz-Hippo, integrin-Focal Adhesion Kinase, etc) to show that mechanosensing changed.

Response: We have increased the sample size to a minimum of n=3 in most cases. However, a few experiments will require more time to increase sample size, as mentioned below.

Our data emphasized the role of Rac1 and SRF. We understand that other molecular players may also be involved in sensing or responding to mechanical forces, but surveying multiple families of candidates without a specific hypothesis or functional experiment is beyond the scope of this study.

__Comment 2: __Fig. 1: In panel E, the cranial bone area measurement is not normalized to mitigate the possible effect of individual differences.

Response: We have re-quantified the data with normalization to the length of the skull.

__Comment 3: __In Fig. 2 the authors mentioned many phenotypical changes (bone length changes, gap thickness change, apex thickness change, etc.) based on histology stain, none of them are quantified to show a significant difference between Rac1-WT and Rac1-KO.

Response: In Fig. 2A, we present the gross morphology of the Rac1-KO embryos and only discuss the tissue defects like edema, hematoma, and hypoplasia, confirmed through H&E as shown in Fig. 2C. We also show the apical limits of the intact calvaria in Fig. 2D, consistent with the calvaria defects observed at birth. In fact, we do not discuss any “bone length changes, gap thickness, or apex thickness change” in this section as suggested by the reviewer. To address the request for more quantification we have added measurement of the edematous area of the apical mesenchyme at E14.5 (Fig. 2C), and this is now shown in Suppl. Fig. 1E. We also added quantification of embryo genotypes and Chi-square tests, now shown in Suppl. Fig. 1D.

Comment 4: Fig. 2 In panel D, with only 2 embryos per group is not enough for quantitation

Response: We plan to increase the number of embryos during the revision period.

Comment 5: Fig. 2 In panel D, the two arrows in the Rac1-KO mutants are not easy to catch.

Response: We made the arrows bigger and bolder.

Comment 6: Fig. 3 The thickness quantification is not performed.

Response: We added quantification in Fig. 3D.

Comment 7: Fig. 3 The images show an obvious curve change of the apex between the control and mutant. Such change is not discussed in the results. Is it due to histology issue?

Response: We do not think it is due to technical issues but reflects a real change in the shape of the apex of the head. We modified the graphical representation in Figure 3E to reflect this change in curvature. We also added the following sentence to the results on page 7: “We also noted a loss of curvature in the apex of the Rac1-KO head at E13.5, which correlated with loss of aSMA+ mesenchymal cells and thinning of the EMM (Fig. 3E).”

__Comment 8: __The merged layer did not show S100a6. While the authors are showing apical expansion of the mesenchyme toward the dermis and meninges, it is hard to track where they are without a merged image.

Response: We added merged images.

Comment 9: Fig. 4 In panel B, 2 biological replicates per genotype are very low.

__Response: __The effect of Rac1-KO on cell cycle is already known (Moore et al. 1997; Nikolova et al. 2007; Gahankari et al. 2021), and our result is supported by in vivo quantification of Tom+Edu+ cells in different regions of the embryonic head shown in Fig. 4A. We prefer not to repeat this assay.

Comment 10: Fig. 4 There is no cell death data.

Response: We will generate data on cell death during the revision period.

__Comment 11: __Fig. 5 In panel B, the GAPDH western plot bands in the mutants seem to be thinner than those of controls.

Response: We verified equal loading with a Ponceau stain, so this minor change in the GAPDH level could be due to biological differences in the protein level. Nevertheless, by our estimation this minor difference does not explain away the major difference in Rac1 and Srf levels.

__Comment 12: __Though the immunostain showed a decrease in signal intensity, it is hard to know whether the decrease is significant enough across all Rac1-KO mutants. They need to measure the fluorescence intensity and perform statistics.

Response: We will generate better images of SRF staining and quantify the difference between Rac1-WT and Rac1-KO during the revision period.

Comment 13: Fig. 6: Similar as Fig. 2, there is no quantification and n=1 per genotype is not enough

Response: During the revision period we will increase the number of E12.5 Srf-KO and Srf-WT embryos to n=3 for Figure 6G. All other panels currently have n=7 or greater.

Comment 14: Fig. 7: Need quantification between Srf-KO and Rac1-KO with statistics to show they are not different, but both significantly different from WTs

Response: In Figure 7D we have added quantification of aSMA area in Srf-KO and Rac1-KO. These results show that both mutants have a similar phenotype with reduced aSMA expression compared to their respective WT littermates, which supports the conclusion that they work in the same pathway. We do not agree with the reviewer that the two mutants should show no statistical difference, because Rac1 and Srf are different genes with overlapping but also non-overlapping functions. During the revision period we will add more Srf-KO embryos and repeat the statistical analysis.

Comment 15: Supplement Fig.2: No image showing the time point before E11.5.

Response: We will add an E10.5 time point during the revision period.

Comment 16: Supplement Fig.3: The ventral view of Rac1-WT does not have the same angle as it shows in Rac1-KO. Makes harder to see the difference between control and mutant.

Response: We adjusted the brightness/contrast to make the difference clearer.

Comment 17: Supplement Fig.4 &7: The alkaline phosphatase stained area needs to be normalized to some other metric because the embryos could be different size.

Response: We normalized to the width of the eye and is now represented in Suppl. Fig. 4 and 7.

Comment 18: Supplement Fig 6 A: The legend and figure don't match. Is it E13.5 or 14.5. Panel 6B needs better images without curling of the tissue.

Response: This has been fixed. The immunostaining images in Suppl. Fig. 6A is E14.5. Panel B is now replaced with better images in the revised manuscript.

---

Reviewer-2

__Comment 1.1: __In Fig. 5, links between Rac1, SRF, αSMA, and contractility in mesenchymal cells are shown. Molecular analyses (Western blot and qPCR) were performed using primary cultured mesenchymal cells (prepared after freed from the epidermal population). Although use of cells prepared from E18.5 embryos may have been chosen by the authors for the safe isolation of the mesenchymal population without contamination of epidermal cells, this reviewer finds that anti-SRF immunoreactivity is weaker at E13.5 than at E12.5 (throughout the section including the mesencephalic wall) and therefore wonder whether SRF expression changes in a stage-dependent manner. So, simply borrowing results obtained from E18.5-derived cells for describing the scenario around E12.5 and E13.5 is a little disappointing point found only here in this study.

Response: In fact, the reason we chose E18.5 was to get enough cells to do the experiments in Figure 5A-D without extensive passaging and/or immortalization, which would undoubtedly cause the cells to deviate from their in vivo character as they become adapted to growing on plastic with 10% serum. Therefore, we prefer not to change the cells as suggested by the reviewer.

__Comment 1.2: __In Fig. 5F, it is difficult to clearly see "reduction" of SRF immunoreactivity in Rac1-KO. Therefore, quantification of %SRF+/totalTomato+ would be desired.

Response: __We will generate better images of SRF staining and quantify the difference between Rac1-WT and Rac1-KO during the __revision period.

__Comment 1.3: __Separately, direct comparison of spontaneous centripetal shrinkage of the apical/dorsal scalp tissues, which will occur in 30 min when prepared at E12.5 or E13.5 (Tsujikawa et al., 2022), between WT and Rac1-KO would strengthen the results in Fig. 5D. As KO is specific to the mesenchyme, the authors do not have to worry about removal of the epidermal layer (which would be much more difficult at E12.5-13.5 than E18.5). If the degree of centripetal shrinkage of the "epidermis plus mesenchyme" layers were smaller in Rac1-KO, it would be interpreted to be mainly due to poorer recoiling activity and contractility of the Rac1-KO mesenchymal tissue.

Response: __We will try to perform the centripetal shrinkage assays as shown by Tsujikawa et al., during the __revision period.

Comment 2: The authors favor "apical" vs. "basolateral" to tell the relative positions in the embryonic head, not only in the adult head. But "apical" vs. "basolateral" should be accompanied with dorsal vs. ventral at least at the first appearance. Apical-to-basal axis or apex vs. basolateral by itself can provide, in many contexts, impressions that epithelial layers/cells are being discussed. Please note that the authors also use "caudal" (in the embryonic head). Usually, a universally defined anatomical axis perpendicular to the rostral-to-caudal axis is the dorsal-to-ventral axis.

Response: Apologies for confusing terminology. The terminology is now defined uniformly according to the anatomical axis.

Comment 3: One of the authors' statements in ABSTRACT "In control embryos, α-smooth muscle actin (αSMA) expression was spatially restricted to the apical mesenchyme, suggesting a mechanical interaction between the growing brain and the overlying mesenchyme" and a similar one in RESULTS "αSMA was not detected in the basolateral mesenchyme of either genotype from E12.5-E14.5 (Suppl. Fig. 4A), suggesting restriction of the mechanosensitive cell state to the apical mesenchyme" need to be at least partly revised, taking previous publication about the normal αSMA pattern in the embryonic head into account more carefully. Tsujikawa et al. (2022) described "Low-magnification observations showed superficial immunoreactivity for alpha smooth muscle actin (αSMA), which has been suggested to function in cells playing force-generating and/or constricting roles; this immunoreactivity was continuously strong throughout the dorsal (calvarial) side of the head but not ventrally toward the face, producing a staining pattern similar to a cap (Figure 2A)" . Therefore, in this new paper, descriptions like "we observed ...., consistent with ....(2022)" or "we confirmed .... (2022)" would be more accurate and appropriate regarding this specific point. Such a minor change does not reduce this study's overall novelty at all.

Response: Thank you for the correction. We have replaced the terminology and cited the article (Tsujikawa et al., 2022) appropriately, crediting their finding.

Comment 4: It would be very helpful if the authors provide a schematic illustration in which physiological and pathological scenarios (at the molecular, cellular, and tissue levels found or suggested by this study) are shown.

Response: We have added a schematic representation of the molecular changes happening in the apical head development because of Rac1- and Srf-KO, and it is represented in Suppl. Fig. 7C.

---

Comment 5: Despite being put in the title, "mechanosensing" by mesenchymal cells is not directly assessed in this study. If appropriate, something like "mechano-functioning" would be closer to what the authors demonstrated.

__Response: __We changed the title to refer to “mechano-responsive mesenchyme”. We think this is appropriate because the cells of interest have reduced aSMA and reduced proliferation, both of which are known to occur, at least in part, as responses to mechanical inputs.

Reviewer-3

Comment 1: Prrx1-Cre targets calvarial mesenchyme and Suzuki et al., 2009 showed that Prrx1-Cre mediated loss of Rac1 lead to calvarial bone phenotype due to incomplete fusion of the skull. While this phenotype was not studied in detail, the statement in the intro and discussion that the calvarial phenotype has not been recapitulated in mice is incorrect.

Response: Suzuki et al showed incomplete fusion of the skull. Although the skull is a tissue that is affected in AOS, it is not akin to the scalp and calvaria aplasia that typifies AOS. Our result stands apart from this. We clarified our position as such:

Introduction (page 4): “Nevertheless, the calvaria phenotype seen in AOS individuals has not been explored in detail or fully recapitulated in mice.”

Discussion (page 11): “Previous studies have demonstrated the role of Rac1 in mesenchyme-derived tissues, but they did not recapitulate AOS phenotypes.”

Comment 2: The authors show that Pdgfra-Cre induced knockout of Rac1 leads to lower-than-expected numbers of Rac1-cKO embryos at E18.5 and P1. Phenotypic analysis shows that the earliest phenotype is blebbing and hematoma in the nasal region at E11.5/12.5. It is stated that this was resolved at E18.5. It is unclear if this is truly a resolution of the phenotype or that these embryos fail to survive until E18.5. Do 100% of the Rac1-cKO embryos exhibit the blebbing/hematoma at E11.5/12.5? What is the observed number/percentage of Rac1-cKO embryos at E11.5/12.5? If the observed percentage of Rac1-cKO is similar to that at E18.5 (lower than the expected 25%), this would support resolution. If the observed ratio is as expected at E11.5/12.5, then this would support embryonic loss before E18.5 rather than phenotypic resolution.

Response: Please note that 100% (n=12) of E12.5 Rac1-KO embryos displayed nasal and mild caudal edema as exhibited in Fig. 2A, but none (n=16) had blebbing/hematoma by E18.5. We added tables for the number of embryos recovered at E12.5 and E18.5 to Supplemental Figure 1. These results show that the percentage of mutants at E12.5 was 21.42%, not significantly different from the expected frequency (p = 0.5371). At E18.5, the percentage dropped slightly to 18.3%, but still not significantly different from expected (p = 0.1545). The significant change in frequency of blebbing/hematoma from E12.5 to E18.5, without any significant change in the frequency of mutants, supports phenotypic resolution of the early blebbing/hematoma.

Comment 3: It is stated that brain shape is altered in Rac1-cKO embryos at E14.5 and E18.5 and concluded that these shape differences are secondary to the cranial defects. Pdgfra+ cells gives rise to the meninges and if the Pdgfra-Cre line recapitulates this expression, then loss of the ubiquitously expressed Rac1 in the meninges could lead to a primary defect in the brain, which may lead to secondary defects in the calvarium and scalp. Their conclusion should recognize other possibilities.

Response: We agree it is possible that there are meninges defects that secondarily change the shape of the brain, and we added a mention of this possibility. It is highly unlikely that scalp defects are only secondary to brain changes because the first observable phenotypes are in the EMM that gives rise to the scalp.

Comment 4: The TdTom staining in wholemount at E13.5 (Supplemental Figure 2B) is difficult to appreciate in the image shown.

Response: At E11.5 there is good contrast between labeled cranial structures and non-labeled body. At E13.5, Tomato appears in most of the mesenchymal cells in the embryo, so there is not as much contrast. The lack of contrast at E13.5 may cause the reviewer think there is something wrong with the image.

Comment 5: The idea that the EMM laminates into the meninges and scalp layers is not new and should be properly cited (Vu et al., 2021, Scientific Reports). The following paper should also be cited on the use of alpha-SMA (Acta2) as a marker of the anterior calvaria mesenchyme: Holms et al., 2020 Cell Reports.

Response: Thank you. We are happy to add those citations.

Comment 6: It is concluded that meningeal development is maintained in the cKO; however, this conclusion was based on a single marker (S100a6) that is both expressed in the presumptive meninges and dermis and greatly reduced overall in the cKO. This conclusion should be softened or other markers used to show that the meninges is indeed normal.

Response: We softened the conclusion on the meninges in the revised manuscript, as this part of the phenotype is was not our focus but it would be a good thing to look at in the future.

Comment 7: The overlap of S100a6 and alpha-SMA is difficult to appreciate in the images shown in Figure 3. Since this is important to the conclusion, co-staining should be done. If co-staining cannot be done due to the primary antibodies' origins, then ISH should be done.

Response: We added merged images.

Comment 8: It is concluded that reduced alpha-SMA suggests an early failure of Rac-cKO cells to respond to the mechanical environment. While this is one possibility, the reduction of alpha-SMA may simply be due to a reduction of these cells resulting from failed differentiation, decreased proliferation, or increased apoptosis.

Response: We think the fact that aSMA is downregulated in cultured cells strongly argues against it being a trivial consequence of reduce proliferation etc. Nevertheless, we softened our conclusion to allow for some of these things to also contribute to the reduced aSMA expression. We will check apoptosis during the revision period.

Comment 9: The conclusion that alpha-SMA is a transient population only present in apical cranial mesenchyme between E12.5-14.5 is not consistent with prior studies: Holms et al., 2020 Cell Reports; Holms et al., 2021 Nature Communications; Farmer et al., 2021 Nature Communications; Takeshita et al., 2016 JBMR.

Response: There is no contradiction. Our statements are based on antibody staining where it is very evident that a-SMA-expressing cells are detectable throughout the apical mesenchyme between E12.5 and E14.5. But at E18.5 we do not see this kind of broad aSMA expression the apical head, suggesting a transient and spatially restricted population of cells in the apical mesenchyme. This is consistent with the studies from Tsujikawa et al., 2022 and Angelozzi et al., 2022. The papers mentioned by the reviewer are only focused on the suture mesenchyme. They do not claim there is broad aSMA/Acta2 expression in the apical head, but only in a spatially restricted subpopulation of suture mesenchymal cells.

Comment 10: In the SRF immunostaining results in control and Rac1-cKO embryos, it is difficult to appreciate the nuclear localization at E12.5 in Figure 5E, as the DAPI is over saturated, and the image quality is poor. The image quality is also poor in Figure 5F.

Response: We will generate better images of SRF staining and quantify the difference between Rac1-WT and Rac1-KO during the revision period.

Comment 11: To what extent is the expression/localization of MRTF, the transcriptional co-activator of SRF, altered in the calvarial mesenchyme of Rac1-cKO embryos? Changes in MRTF would strengthen the link between Rac1 and SRF.

Response: We do not know how MRTF expression/localization changes in the embryo tissue, but western blot data on Rac1-KO fibroblasts revealed a reduction in expression/nuclear localization of MRTF-A/B that mirrored the changes in SRF. We added these blots to Figure 5A. However, as noted at the end of the discussion, MRTF is not always required for SRF function in vivo ( Dinsmore, Elife 2022). The MRTFA/B-KO is a possibility for future work.

Comment 12: Hypoplasia of the apical mesenchyme (Figure 6G, inset 1) in Srf-cKO is difficult to see.

Response: During the revision period we will increase the number of E12.5 Srf-KO and Srf-WT embryos to n=3 for Figure 6G and replace the picture with a better one.

Comment 13: Generally, the organization of the data into many main and supplemental Figures makes the flow difficult to follow.

__Response____: __We understand the concern, but we have tried our best to organize the most important data into main figures and the relevant but less essential data into supplemental figures.

Comment 14: SFR interacts with Pdgfra interacts genetically with Srf in neural crest cells in craniofacial development, with Srf being a target of PDGFRa signaling (Vasudevan and Soriano, 2015, Dev Cell). Since the Pdgfra-Cre line used here is hemizygous, is important that the control used to look at SRF expression in the Rac1-cKO is Pdgfra-Cre+.

Response: It is standard practice to include some Cre+ mice in the control set to reveal whether Cre has toxic effects in the cells of interest. To the reviewer’s concern about genetic interactions between the Pdgfra gene and Srf, this should not be relevant here because the Pdgfra-Cre used in our study is a transgene and does not affect the endogenous Pdgfra gene.

Comment 15: The text size in all figures is too small and varies throughout, making it difficult to read.

Response: To fit the panel in the Word document, the figure is resized. This should not be an issue in the final manuscript.

Comment 16: Details about the pulse-chase timing of the EdU experiments should be included in the results. Also, does n = 3 for each stage and each genotype? I would be helpful to include a representative section for a control and cKO littermate pair.

Response: The details are now included in the methods section. Yes, n=3 in each stage and genotype (Fig. 4A). The representative images are also included.

Comment 17: The relative sizing of the panels within and between figures is haphazard. Some are very large and others very small (Figure 2, 6, Supplemental Figure 1, 2, 6, 7).

Response: The image panels are fixed in the revised manuscript.

Comment 18: In Figure 5A and F, the titles "E12.5" and "E13.5" are in italics.

Response: The fonts for the figures are fixed in the revised manuscript.

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: This manuscript by Rathnakar et al. examines the role of the small GTPase Rac1 in apical closure of the scalp and skull. Rac1 activity is regulated the guanine nucleotide exchange factor DOCK6 and the GTPase AHGAP31. Loss of function variants in DOCK6 and gain of function variants in AHGAP31 lead to sustained inactivation of Rac1 in Adams-Oliver syndrome (AOS), which is characterized by aplasia cutis congenita, underlying calvarial defects, and limb abnormalities. While Rac1 is thought to be a key in the pathogenesis of AOS, how decreased in Rac1 activity impact development of the head is not well-understood. The authors find that conditional loss of Rac1 in cranial mesenchyme (using Pdgfra-Cre), leads to AOS-like abnormalities in the scalp and skull. They go on to show that these abnormalities are linked to reduced alpha-SMA expression in the early migrating mesenchyme (EMM), decreased osteoprogenitor cells in the supraorbital mesenchyme (SOM), decreased proliferation, and the contractile function of fibroblasts. They also find that Rac1 cKO leads to reduced expression of the mechanosensitive transcription factor SRF. Finally, they show that loss of SRF in cranial mesenchyme (using Pdgfra-Cre) leads to an AOS-like scalp and skull phenotype that has mechanistic overlap with their findings in the Rac1 cKO.

Major:

1. Prrx1-Cre targets calvarial mesenchyme and Suzuki et al., 2009 showed that Prrx1-Cre mediated loss of Rac1 lead to calvarial bone phenotype due to incomplete fusion of the skull. While this phenotype was not studied in detail, the statement in the intro and discussion that the calvarial phenotype has not been recapitulated in mice is incorrect.
2. The authors show that Pdgfra-Cre induced knockout of Rac1 leads to lower-than-expected numbers of Rac1-cKO embryos at E18.5 and P1. Phenotypic analysis shows that the earliest phenotype is blebbing and hematoma in the nasal region at E11.5/12.5. It is stated that this was resolved at E18.5. It is unclear if this is truly a resolution of the phenotype or that these embryos fail to survive until E18.5. Do 100% of the Rac1-cKO embryos exhibit the blebbing/hematoma at E11.5/12.5? What is the observed number/percentage of Rac1-cKO embryos at E11.5/12.5? If the observed percentage of Rac1-cKO is similar to that at E18.5 (lower than the expected 25%), this would support resolution. If the observed ratio is as expected at E11.5/12.5, then this would support embryonic loss before E18.5 rather than phenotypic resolution.
3. It is stated that brain shape is altered in Rac1-cKO embryos at E14.5 and E18.5 and concluded that these shape differences are secondary to the cranial defects. Pdgfra+ cells gives rise to the meninges and if the Pdgfra-Cre line recapitulates this expression, then loss of the ubiquitously expressed Rac1 in the meninges could lead to a primary defect in the brain, which may lead to secondary defects in the calvarium and scalp. Their conclusion should recognize other possibilities.
4. The TdTom staining in wholemount at E13.5 (Supplemental Figure 2B) is difficult to appreciate in the image shown.
5. The idea that the EMM laminates into the meninges and scalp layers is not new and should be properly cited (Vu et al., 2021, Scientific Reports). The following paper should also be cited on the use of alpha-SMA (Acta2) as a marker of the anterior calvaria mesenchyme: Holms et al., 2020 Cell Reports.
6. It is concluded that meningeal development is maintained in the cKO; however, this conclusion was based on a single marker (S100a6) that is both expressed in the presumptive meninges and dermis and greatly reduced overall in the cKO. This conclusion should be softened or other markers used to show that the meninges is indeed normal.
7. The overlap of S100a6 and alpha-SMA is difficult to appreciate in the images shown in Figure 3. Since this is important to the conclusion, co-staining should be done. If co-staining cannot be done due to the primary antibodies' origins, then ISH should be done.
8. It is concluded that reduced alpha-SMA suggests an early failure of Rac-cKO cells to respond to the mechanical environment. While this is one possibility, the reduction of alpha-SMA may simply be due to a reduction of these cells resulting from failed differentiation, decreased proliferation, or increased apoptosis.
9. The conclusion that alpha-SMA is a transient population only present in apical cranial mesenchyme between E12.5-14.5 is not consistent with prior studies: Holms et al., 2020 Cell Reports; Holms et al., 2021 Nature Communications; Farmer et al., 2021 Nature Communications; Takeshita et al., 2016 JBMR.
10. In the SRF immunostaining results in control and Rac1-cKO embryos, it is difficult to appreciate the nuclear localization at E12.5 in Figure 5E, as the DAPI is over saturated, and the image quality is poor. The image quality is also poor in Figure 5F.
11. To what extent is the expression/localization of MRTF, the transcriptional co-activator of SRF, altered in the calvarial mesenchyme of Rac1-cKO embryos? Changes in MRTF would strengthen the link between Rac1 and SRF.
12. Hypoplasia of the apical mesenchyme (Figure 6G, inset 1) in Srf-cKO is difficult to see.
13. Generally, the organization of the data into many main and supplemental Figures makes the flow difficult to follow.
14. SFR interacts with Pdgfra interacts genetically with Srf in neural crest cells in craniofacial development, with Srf being a target of PDGFRa signaling (Vasudevan and Soriano, 2015, Dev Cell). Since the Pdgfra-Cre line used here is hemizygous, is important that the control used to look at SRF expression in the Rac1-cKO is Pdgfra-Cre+.

Minor:

1. The text size in all figures is too small and varies throughout, making it difficult to read.
2. Details about the pulse-chase timing of the EdU experiments should be included in the results. Also, does n = 3 for each stage and each genotype? I would be helpful to include a representative section for a control and cKO littermate pair.
3. The relative sizing of the panels within and between figures is haphazard. Some are very large and others very small (Figure 2, 6, Supplemental Figure 1, 2, 6, 7).
4. In Figure 5A and F, the titles "E12.5" and "E13.5" are in italics.

Significance

Overall, this is an interesting study that shares mechanistic insight into the scalp and skull deformities in AOS. The overall presentation of the work, particularly the figures, should be improved and streamlined to enhance clarity and better emphasize the novelty of the study. In addition, the conclusions are not always well-supported by the results and the interpretation of the results do not fully consider and cite previous studies.

Audience: Developmental Biologists

Expertise: Craniofacial development and disease

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

Summary

In mice lacking Rac1 in the PDGFRa+ mesenchymal cell lineage, the authors found Adams-Oliver syndrome (AOS)-like defects of the apical/dorsal scalp and calvaria, which was accompanied by the secondary brain protrusion by E18.5. The primary phenotype emerged at E11.5 and worsened from E12.5 to E14.5 in the apical/dorsal region of the embryonic head, with limited lateral expansion as well as reduced thickening/stratification of the mesenchymal layer expressing α-smooth muscle actin (αSMA). Very similar in vivo abnormalities were obtained when serum response factor (SRF), known as a mechanotransducing factor, was removed in PDGFRα+ mesenchymal cells. Rac1-lacking mesenchymal cells proliferated poorly in vivo and contracted weakly in culture, with reduced expression of SRF and αSMA. Based on these results and previously obtained understanding that the developing apical/dorsal mesenchyme is mechanically stretched by the underlying brain, the authors conclude that the mechanosensing-triggered morphogenetic behaviors of the apical/dorsal mesenchymal cells (i.e., proliferation, stratification, and contraction, which all lead to physical stability or mechanical resilience of that layer) is mediated by Rac1 and SRF. The authors also suggest that this molecular mechanism for the physiological maturation of the apical/dorsal mesenchyme may underlie the ventral-to-dorsal progression of osteogenesis, absence of which explains AOS pathogenesis.

Major comments:

In Fig. 5, links between Rac1, SRF, αSMA, and contractility in mesenchymal cells are shown. Molecular analyses (Western blot and qPCR) were performed using primary cultured mesenchymal cells (prepared after freed from the epidermal population). Although use of cells prepared from E18.5 embryos may have been chosen by the authors for the safe isolation of the mesenchymal population without contamination of epidermal cells, this reviewer finds that anti-SRF immunoreactivity is weaker at E13.5 than at E12.5 (throughout the section including the mesencephalic wall) and therefore wonder whether SRF expression changes in a stage-dependent manner. So, simply borrowing results obtained from E18.5-derived cells for describing the scenario around E12.5 and E13.5 is a little disappointing point found only here in this study. In Fig. 5F, it is difficult to clearly see "reduction" of SRF immunoreactivity in Rac1-KO. Therefore, quantification of %SRF+/totalTomato+ would be desired. Separately, direct comparison of spontaneous centripetal shrinkage of the apical/dorsal scalp tissues, which will occur in 30 min when prepared at E12.5 or E13.5 (Tsujikawa et al., 2022), between WT and Rac1-KO would strengthen the results in Fig. 5D. As KO is specific to the mesenchyme, the authors do not have to worry about removal of the epidermal layer (which would be much more difficult at E12.5-13.5 than E18.5). If the degree of centripetal shrinkage of the "epidermis plus mesenchyme" layers were smaller in Rac1-KO, it would be interpreted to be mainly due to poorer recoiling activity and contractility of the Rac1-KO mesenchymal tissue.

Minor comments:

1. The authors favor "apical" vs. "basolateral" to tell the relative positions in the embryonic head, not only in the adult head. But "apical" vs. "basolateral" should be accompanied with dorsal vs. ventral at least at the first appearance. Apical-to-basal axis or apex vs. basolateral by itself can provide, in many contexts, impressions that epithelial layers/cells are being discussed. Please note that the authors also use "caudal" (in the embryonic head). Usually, a universally defined anatomical axis perpendicular to the rostral-to-caudal axis is the dorsal-to-ventral axis.
2. One of the authors' statements in ABSTRACT "In control embryos, α-smooth muscle actin (αSMA) expression was spatially restricted to the apical mesenchyme, suggesting a mechanical interaction between the growing brain and the overlying mesenchyme" and a similar one in RESULTS "αSMA was not detected in the basolateral mesenchyme of either genotype from E12.5-E14.5 (Suppl. Fig. 4A), suggesting restriction of the mechanosensitive cell state to the apical mesenchyme" need to be at least partly revised, taking previous publication about the normal αSMA pattern in the embryonic head into account more carefully. Tsujikawa et al. (2022) described "Low-magnification observations showed superficial immunoreactivity for alpha smooth muscle actin (αSMA), which has been suggested to function in cells playing force-generating and/or constricting roles; this immunoreactivity was continuously strong throughout the dorsal (calvarial) side of the head but not ventrally toward the face, producing a staining pattern similar to a cap (Figure 2A)" . Therefore, in this new paper, descriptions like "we observed ...., consistent with ....(2022)" or "we confirmed .... (2022)" would be more accurate and appropriate regarding this specific point. Such a minor change does not reduce this study's overall novelty at all.
3. It would be very helpful if the authors provide a schematic illustration in which physiological and pathological scenarios (at the molecular, cellular, and tissue levels found or suggested by this study) are shown.
4. Despite being put in the title, "mechanosensing" by mesenchymal cells is not directly assessed in this study. If appropriate, something like "mechano-functioning" would be closer to what the authors demonstrated.

Significance

This study advances understanding of a key aspect of the molecular mechanisms underlying the normal mammalian craniofacial development, unveiling the role of Rac1 and SRF in the apical/dorsal mesenchymal layer which has inter-tissue mechanical relationships with the embryonic brain underneath. This study also advances understanding of Adams-Oliver Syndrome pathogenesis, demonstrating the biological significance of the normal inter-tissue mechanical relationships in the developing mammalian head. This study may have opened a door for the genetic/molecular dissection toward the tissue-level mechano-engineering, which would stimulate development of next-generation organoids or assembloids. Broad audience including developmental biologists/neuroscientists, molecular/cellular biologists, pathologists, clinical geneticists, and pediatricians would be interested in this work.

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

In this paper "Mouse scalp development requires Rac1 and SRF for the maintenance of mechanosensing mesenchyme", the authors demonstrated that deletion of Rac1 (Rac1-KO) with a PDGFRαCreTG mouse model led to absence of skull apex and a blebbing formation while the limbs were not impacted. Rac1-KO mice showed the Rac1 regulated expansion of the apical mesenchyme toward the very apex meningeal and dermis layer and the osteogenic differentiation of supra orbital arch mesenchyme. Rac1 also regulates the proliferation of apical mesenchyme, dermis differentiation, and mechanosensing of the cranial mesenchyme cells. The authors also indicated Rac1 was a regulator of Srf by showing the deletion of Rac1 lead to lower Srf mRNA level and SRF protein expression. Deletion of Srf showed similar phenotypes as Rac1-KO mice.

Major concern is the study lacks rigor in several areas where n=2, results are not quantified with statistics. They need to run power analysis and increase their samples sizes. Please include statistics on all measurements. Filamentous actin staining and alpha-sma is used to visualize mechanosensing but also in other cell activities such as cell contractility for movement, cell to substrate adhesion, cell division, etc. They need to query more mechanosensing related pathways (Piezo1/2, Yap/taz-Hippo, integrin-Focal Adhesion Kinase, etc) to show that mechanosensing changed.

Comments by figure.

Fig. 1: In panel E, the cranial bone area measurement is not normalized to mitigate possible effect of individual differences.

Fig. 2:

1. While the authors mentioned many phenotypical changes(bone length changes, gap thickness change, apex thickness change, etc) based on histology stain, none of them are quantified to show a siginificant difference between Rac1-WT and Rac1-KO.
2. In panel D, with only 2 embryos per group is not enough for quantitation.
3. In panel D, the two arrows in the Rac1-KO mutants are not easy to catch.

Fig. 3:

1. The thickness quantification is not performed.
2. The images show an obvious curve change of the apex between the control and mutant. Such change is not discussed in the results. Is it due to histology issue?
3. The merged layer did not show S100a6. While the authors are showing apical expansion of the mesenchyme toward the dermis and meninges, it is hard to track where they are without a merged image.

Fig.4:

1. In panel B, 2 biological replicates per genotype are very low
2. There is no cell death data.

Fig. 5:

1. In panel B, the GPDH western plot bands in the mutants seem to be thinner than those of controls.
2. Though the immunostain showed a decrease in signal intensity, it is hard to know whether the decrease is significant enough across all Rac1-KO mutants. They need to measure the fluorescence intensity and perform statistics.

Fig. 6: Similar as Fig. 2, there is no quantification and n=1 per genotype is not enough.

Fig. 7: Need quantification between Srf-KO and Rac1-KO with statistics to show they are not different but both significantly different with WTs.

Supplement Fig.2: No image showing the time point before E11.5.

Supplement Fig.3: The ventral view of Rac1-WT does not have the same angle as it shows in Rac1-KO. Makes harder to see the difference between control and mutant.

Supplement Fig.4 &7: The alkaline phosphatase stained area needs to be normalized to some other metric because the embryos could be different size.

Supplement Fig 6 A: The legend and figure don't match. Is it E13.5 or 14.5. Panel 6B needs better images without curling of the tissue.

Significance

Please see my comments above. This work is broadly of interest to developmental biologist, fracture healing, and human genetics fields.

The paper is easy to understand and follow. The massive amount of histology and immunostaining images make it easy to identify the point the authors want to show. All the figures are well-labeled and visually informative. The experiment sequence is logic. The gene deletion models provide solid and direct evidence on the necessity of their function during early head development. The discussion is thoughtfully written and clear. The authors discuss the connection of Rac1 and SRF with other signaling pathways, which makes them promising target toward Adams-Oliver syndrome.

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

We thank the reviewers for their comments

__Reviewer 1 __

This is review of the manuscript „A simple method to efficiently generate structural variation in plants" by Bechen et al. The manuscript presents a very interesting and innovative approach to generate structural variant mutations (including large ones) in the genome of Arabidopsis thaliana using a simple chemical treatment with TOPII inhibitor etoposide. Authors show that unlike chemical mutagens commonly used for induction of SNPs (EMS, sodium azide...), etoposide-treatment caused structural variants like DNA deletions, insertions, inversions and translocations. These mutations were identified by the whole genome short and long read sequencing that also indicated a WT-like frequency of SNPs. This finding can potentially help inducing mutations similar to high energy radiation in potentially any plant. First, the manuscript provides description of the unusual phenotypes found after etoposide treatment and their Mendelistic inheritance. Based on this, authors performed whole genome sequencing and mutation detection, validation. The experimental part ends by transcriptome analysis that authors use as the approach to identify the causal mutations. This part is, in my opinion, the weakest part of the manuscript and would benefit from further clarification or even additional experiments (see below). Overall the manuscript is very clear and contains all necessary information. The only part that was confusing to me, was the section focusing on the transcriptome analysis.

__Response: __Thank you for your appreciation of the study. As detailed below, we have changed our presentation of the RNA-seq results to better describe their purpose.

Major points: Line 222: In the section „RNA-Seq identifies genes that are associated with structural variation and mutant phenotypes", authors suggest that the changes in the transcript amount were used to identify causal mutations. I got confused by this section. Exach of the examples represents unique situation and thus only single cases are presented which makes it hard to estimate robustness of the presented approach. Also, the presented mutations have prominent phenotypes that were already heavily studied in the past and therefore the possible causal genes are mostly known. Therefore, I am not sure how this approach would stand in case of traits with unknown underlying genes.

__R____esponse: __Our intent was not to present a new method for mapping causative mutations. Like any other induced genetic mutation, there are many possible strategies for identifying the causative locus (loci), such as mapping-by-sequencing via a segregating F2 population (as mentioned below). Organization of the manuscript’s results has been changed to reflect that mRNA-seq was conducted to learn more about the phenotypes, not to definitively identify causal genes/mutations. We have also added additional text to the discussion to clarify candidate mutation mapping approaches on line 376: “How do we identify genetic changes that are causative for phenotypes of interest? Future studies will accelerate candidate-gene discovery by employing structural-variant callers, de novo genome assembly-based approaches, and RNA-Seq based mapping (Mahmoud et al 2019). Although our study did not aim to use RNA-Seq to identify mutations, it provides an example of how RNA-Seq data in tandem with genome sequencing can help shortlist potential causal mutations. In cases where a potential causative variant is not obvious, these strategies can be combined with traditional genetic mapping approaches. However, mapping-by-sequencing approaches might not be easily applicable to some classes of mutants. For example, inversions or translocations can suppress recombination and reduce the efficacy of mapping-by-sequencing.”

When refering to the case with the chromosomal inversion, I do not see how one will be able to map a candidate based on the relatively mild expression (but maybe I am missing something here). Similarly, the „mapping" approach applied to the variegated line would not be possible on a trait that is less studied and the candidates are not well known. I wonder why authors did not perform association mapping on a bulk of phenotypically mutant plants collected from a segregating F2 backcross population. This might be a more robust way of linking the phenotype with a mutation.

Response: Our primary goal with the manuscript was to demonstrate that etoposide-treatment induces mutant phenotypes and structural variation. Identifying the causal mutation for every phenotype is outside the scope of the present study. As described above, we have added additional text in the discussion to briefly describe candidate mutation mapping approaches that researchers can use.

Discussion section: I am missing discussion on how etoposide could be causing such structural variants.

Response: Etoposide’s mode of action is well-studied in animal systems and has been described in text that has been moved from results to introduction. Starting on line 90 it reads “Topo II relaxes torsional stress from DNA supercoiling generated during DNA replication or transcription by transiently breaking both strands and then ligating them after passing a DNA segment through the break. Between strand breakage and ligation, Topo II is covalently linked to DNA via a tyrosine residue, forming a topoisomerase cleavage complex [37]. This complex is stabilized by the inhibitor etoposide. A collision between covalently-linked Topo II and DNA polymerases during DNA replication, or with RNA polymerases during transcription, leads to removal of the Topo II enzyme, which results in the generation of double-stranded breaks (DSBs) [38–41]. The imprecise repair of DSBs leads to genomic rearrangements and structural variation in mouse spermatocytes, fibroblasts, and in human cells [42–44]. Previously, it was shown that treatment with etoposide inhibits plant growth[45,46] and causes fragmentation of chromosomes during meiosis in Arabidopsis [45]. However, its potential as a mutagen that can induce structural variation has not been investigated"

Minor points:

Line 70: Possibly add sodium azide. It is frequently used as mutagen for some plant species.

__Response: __Added sodium azide to line 76.

Line 122: „...etoposide is an excellent mutagen for efficiently creating large-effect mutations." This cannot be claimed at this point because the sequence analysis data were not shown yet. Please reformulate.

Response: We changed this line (now line 133) to: “The large proportion of plants showing visible phenotypes suggested that etoposide could be an excellent mutagen for efficiently creating large-effect mutations.”

**Referees cross-commenting** My main issue was the mapping protocol using transcriptomic changes. It is hard to believe that this approach would work well on unknown/less studied traits. What is your opinion?

Response: Identifying the causal mutation for every phenotype is outside the scope of the present study. Organization of the manuscript’s results has been changed to reflect that RNA-seq was conducted to learn more about the phenotypes, not to definitively identify causal genes/mutations. In some instances (BR-like dwarf), the RNA-seq data, combined with prior knowledge, suggested a causal variant (AS1), which was further bolstered by the identification of structural variants. Note that it is only for the variegated mutant that we definitively identified the causal mutation (IM) by genetic complementation.

Reviewer #1 (Significance (Required)):

Strengths - innovative way on how to induce structural variant mutations in plants.

Limitations - The approach on how to map the mutations needs more development. At this point i tis not clear how well the approach will work in other plant species.

Audience - basic and applied plant scientists

Response: We have adjusted our discussion of the role of mRNA-seq in the study, and added comments on approaches for mapping causative mutations. We hope this now clarifies the overall strategy. Topoisomerase’s sensitivity to etoposide inhibition is conserved amongst tested plants and animal species. We have changed the sentence and added references to introduction to show that etoposide acts on other plant species (line 100): “Previously, it was shown that treatment with etoposide impacts genome stability and inhibits plant growth in Arabidopsis thaliana, Allium cepa, and Lathyrus sativus [46-49] and causes fragmentation of chromosomes during meiosis in Arabidopsis [48]. “ In addition, preliminary (unpublished) work in our labs shows that etoposide has mutagenic impacts on legumes and Brassicaceae crop species. We therefore believe that this protocol should be widely applicable to other plant species.

===

Reviewer 2

• The manuscript describes mutagenesis of Arabidopsis by a topoisomerase II inhibitor. The method is effective, resulting in good density of SV and no detectable SNV. The authors provide a full characterization of the mutants, their phenotypes, and their genomes. The 34-sample selected for genomic analysis is sufficient to make firm conclusions.

• The manuscript is clearly written and illustrated.

• The manuscript does a very good job at covering the phenotypic and molecular analysis for this type of mutagenesis. For example, they highlight the difference between short and long reads in the identification of SV.

• The figures are very clear, with the exception of Fig. 3, which I found harder to follow. It would be enhanced by describing the candidate lesion(s) in the first panel of each mutant series. This would clarify the expectation. For example, larger indels (not examined here) should be associated with higher (insertion) or lower (deletion) expression of the affected genes. In the cases presented in Fig.3, the structural changes do not suggest obvious hypotheses. The authors examine the regions near breakpoint of inversions or near small indels. It makes sense, but it does not make the figure very digestible. The connected text in the results, on the other hand, is very clear. Perhaps, making the conclusions in the figure legend as well? As a connected thought, it would have been useful to provide expression data for a large indel exemplifying the cis/trans nature of regulatory changes.

__Response: __Thank you for your helpful comments. The RNA-seq data is now presented before the DNA sequencing data to clarify the role of the RNA-seq data in this study. The previous Figure 3 is now Figure 2. We have clarified the presentation by combining original Fig. 3C and 2E into a single panel (Fig. 4F); moving 3J to Figure 4E; removing what was 3I; and moving the original 3B, 3E-G to Figure S9. In Figure S9, a label for the type of SV was added to the scatterplots of expression of genes surrounding the SVs. For further clarity, the panel describing the inversion in BR-like dwarf is also now plotted in the same way as those of short-internode dwarf. We agree it would be informative to provide expression data for a large indel to determine the extent of cis or trans effects, but we did not find any large indels in the samples we sequenced with long-reads.

We agree with the reviewer that some of the mutants we have generated would be great material for further studying cis/trans nature of regulatory changes. We are strongly interested in this question; it is certainly a subject for a future publication.

• Deletions and other rearrangements may affect meiosis as noted by the authors. In addition, they can display gametophytic phenotypes and a deficit in transmission. The likelihood increases with the size of the indel. Large indels are not transmitted. Accordingly, for indels above a certain size, it is not possible to determine the number of causal loci from F2 ratios.

__Response: __We agree that the impact of very large SVs on meiosis alters segregation ratios and prevents determination of causal loci from F2 ratios. Impacts on meiosis will also likely impact our ability to use techniques based on bulk segregant analysis to finely map causal mutations. It is important to note that such mutations comprise only a fraction of all detected SVs.

Identification of multiple loci causing a phenotype in plants carrying large SVs will therefore require other approaches. For example, structural variation callers can be used to identify the boundaries of SVs like duplications or deletions. RNA-Seq can be used to identify genes at the SV whose expression is highly effected by cis-regulatory changes. To test if those SVs are responsible for the phenotype, genes at SV boundaries along with the novel promoters can then be reintroduced as transgenes. This should enable one to identify combinations of mutated genes that are responsible for the phenotypes.

• Although the use of topo II inhibitors for mutagenesis in plants is novel, the mutagenic effects described here are well documented in animals. This should be acknowledged (e.g. Heisig, Mutagen. 2009; Ferguson, Env Mol Mutagen. 1994)

__Response: __We have acknowledged prior work demonstrating the mutagenic impact of etoposide using primary literature as well as more recent references. These include references # 43-45.

**Referees cross-commenting** I also found the expression analysis confusing and in need of revision. One reason is that a set of clear expectations were not provided. I believe that the RNAseq analysis is expected to help identify the gene(s) that underlie a trait. For example, genes located on an indel are likely to display expression proportional to copy number. Also, a new junction or translocation could influence expression of the gene next to the break point. The authors should make this clear in the figure and the text.

Response: Organization of manuscript’s results has been changed to reflect that mRNA-seq was conducted to learn more about the phenotypes, not to definitively identify causal genes/mutations.

Reviewer #2 (Significance (Required)):

• I appreciated the description of the method. It should be widely applicable. In arabidopsis, it requires sustained growth in the presence of the inhibitor. This could limit its applicability. For example, it may not be effective with pollen because exposure by a short soaking period may not be sufficient. Culturing of large seeded species is possible, but adds complexity. In this context, radiations have advantages. I do agree with the authors on the difficulty in identifying a source. However, once one is found, radiation treatment is very simple and convenient.

• The manuscript describes a useful tool and the connected spectrum of mutations. It has the novelty, quality, and relevance to represent a significant contribution to plant biology and to be of broad interest.

__Response: __Thank you for your feedback. For Arabidopsis, we germinated and grew seeds on media containing etoposide for about two weeks (see Methods). In work that is not yet ready for publication, we have taken the same approach with a legume species and other Brassicaceae that have substantially larger seeds. We find that we need to use a higher dose of etoposide to induce phenotypes, but that it is easy to germinate and grow large-seeded plants for a couple of weeks on media containing etoposide. We don’t anticipate that seed size will be limiting for this method.

We agree that this technique will not work for pollen; it will only work for tissues with significant levels of DNA replication. However, this technique alleviates the need for collecting pollen. For example, pollen irradiation has been used to create poplar with structural variants. However, if using etoposide-based mutagenesis, one could grow poplar seeds, cuttings, explants, embryos, or calli on etoposide-containing media.

Reviewer 3

---

Summary:

The manuscript entitled "A simple method to efficiently generate structural variation in plants" by Bechen et al. investigated an efficient mutagen for inducing large structural variations in plants, replacing traditional irradiation methods with a chemical mutagenesis strategy. The study examined the effects of etoposide, a DNA topoisomerase II inhibitor, on structural variations and demonstrated that etoposide treatment induces a wide range of phenotypic and genome changes, including inversions, duplications, and deletions. Additionally, the authors analyzed the relationship between gene expression changes and genomic alterations to identify potential causal genes underlying specific phenotypes. While their findings provide clear and reliable evidence of structural variations induced by etoposide, I have several suggestions to enhance the clarity of their results, as detailed below.

__Response: __We thank the reviewer for their feedback. It has helped improve the presentation and clarity of our results.

Major comments:

1. Lines 166-169: My understanding is that you selected etoposide-treated M1 plants based on specific phenotypes, and observed their M2 and M3 progeny, categorizing them as either phenotype-positive or phenotype-negative. In Table S3, phenotypes other than BR-like dwarf, virescent, and short internode dwarf are not mentioned. Does this indicate that these other lines did not exhibit heritable phenotypic traits? If other lines showed some phenotype changes, could you incorporate progeny relationships along with phenotype information into Table S3? Additionally, in Figures S2 and S3, you reference 26A lines. Did they exhibit similar phenotypic changes among them?

__Response:____ __Unfortunately, we do not have detailed phenotypes of each chosen M1 line. Most lines had one or more of the phenotypes we mention in the results – “Those exposed to 160 µM of etoposide exhibited significantly more abnormal phenotypes than DMSO only or 80 µM etoposide plants, including loss of apical dominance, gnarled leaves, reduced plant size, seed abortion, and lower seed number at maturity (Figure 1A).” It is important to note that M2 phenotypes were not observed in M1.

Table S3 is now Table S7. Only some mutants lines or lineages that were sequenced had one of the scored phenotypes in M2 (10B,13B, 1A, 1B, 21A, 24B, 26A, 34C, 5A,9A). Of these, we have formally tracked inheritance of only 13B, 1A, 34C, and 5A over multiple generations. We do not have similar data for other sequenced lines . We also sequenced some lines (17B, 21B, 26C) without any mutant phenotypes to assess if plants lacking visible phenotypes still carried SVs. Indeed, as described in Table S9, all three of these lines carried small SVs, which might not affect genes that create an obvious visible phenotype under our growth or observation conditions.

Yes. 26A siblings all exhibited the same flat leaf phenotype.

Overall, our data suggests that mutant phenotypes and their causal SVs can be stably transmitted through multiple rounds of meiosis.

Lines 187-189 and Figure S4: The assessment of repeat copy number variation provides valuable insights. However, based on the figure, the conclusion that "etoposide treatment likely did not trigger genomic instability in repetitive DNA" is difficult to interpret. Could you modify the figure into a box plot with raw data points and include a statistical analysis to support this conclusion?

__Response: __The figure has been modified to a box plot and is now presented as main Figure 3. Wilcox test has been performed and shows no significant difference in read depth over NOR2, NOR4, and telomere regions between control and etoposide-treated lines.

Line 200 and Figure S8A: You state that SNV analysis identified a similar number of SNVs in treated and control plants. However, this is not easily interpretable from the figure. Could you include a statistical comparison between etoposide-treated and control plants? For example, EMS mutagenesis is known to induce specific G/C → A/T transitions. Did etoposide-treated and control plants exhibit the same types of nucleotide changes, or were there differences in the mutation spectrum?

Response: This figure has been modified to assess the entire mutational spectrum of SNVs, including statistical comparisons, and is now part of Figure 3. We have also added the following text on line 244: “However, SNV analysis identified a comparable spectrum and number of SNVs in etoposide-treated and control lines (Figure 3), suggesting that etoposide did not induce excess SNVs.”

Lines 219-220: Your conclusion clearly demonstrates the detection of numerous structural variations using both short- and long-read sequencing technologies. Could you provide a summary table listing the detected mutation positions? Since short-read sequencing is generally less effective in detecting large structural variations, I am particularly interested in evaluating the accuracy of Lumpy Express in identifying mutations.

Response: Short-read sequencing and Lumpy Express are unsurprisingly less effective in detecting large structural variations when compared with long-read based approaches. SVs detected by Nanopore were missed by short-read sequencing and Lumpy Express. However, it is hard to benchmark the efficacy of Lumpy Express as only a few lines were sequenced by both Nanopore long-read sequencing and short-read sequencing. After removing SVs that were also present in control lines, we could identify only one SV detected by both Lumpy Express and Nanopore sequencing; this SV is a deletion. In plant 1A_4_5, which was sequenced by short reads, Lumpy Express called a 70 bp deletion at Chr 5: 5776579. SV calling using Nanopore-generated sequence of a sibling plant, 1A_4_11, identified a 70 bp deletion at Chr 5:5776578. The SVs identified by Lumpy Express are presented in Table S9. Those identified from long-read data are in Table S11.

1. Figures 3E-G: To facilitate a clearer comparison of the effects of structural variations on gene expression between BR-like dwarf and short internode dwarf, could you add an average trend line to the figures, similar to Figure 3B?

Response: An average trend line has been added to these plots. These data are now presented in Figure S9.

Minor comments:

Line 105 and Figure 1A: In the manuscript, etoposide concentrations are stated as 0, 40, 80, and 160 μM, whereas Figure 1A labels the concentrations as 0, 80, 160, and 320 μM. Should the figure be updated to 0, 40, 80, and 160 μM for consistency?

__Response: __Thank you for the comment. We updated the results section to make concentrations listed consistent with methods (0, 20, 40, 80,160, 320, and 640 µM) and added additional description of seedling growth such that all concentrations are described. We also updated references to the figure such that only sentences regarding concentrations with photos reference Fig 1A.

Figure 1B legend: Typographical error: "roundsof" → "rounds of".

Response: Corrected.

Line 109: Do you have a summary table for the M1 generation? If so, could you provide it as a supplementary table?

Response: We regretfully do not have the data to populate a summary table for the M1 generation. As described above, most M1 plants have one or more of the following phenotypes mentioned in text: “loss of apical dominance, gnarled leaves, reduced plant size, seed abortion, and lower seed number at maturity “

Line 119: Figure 1B only defines developmental stages. To improve clarity, consider revising "Figure 1B" to "Figure 1B-F", allowing readers to easily understand the corresponding figures.

__Response: __We updated the Figure reference to Fig. 1B-F.

Line 121: The citation "(Figure S1, Table S1)" would be clearer if placed at the end of the sentence.

__Response: __The citation was moved to the end of the sentence.

Lines 137, 148, 167: To maintain consistency with Figure 1C-F and the manuscript's logical flow, could you standardize the order of phenotypes as "virescent, short internode dwarf, and BR-like dwarf" instead of the current variation?

__Response: __We have standardized the order of the figures and the discussions of phenotypes in the text as: BR-like dwarf, short-internode dwarf, virescent, then variegated.

Line 139: Why is "Figure 1B" referenced at this position? Would it be more appropriate to remove this reference?

Response: Figure 1B shows that the phenotypes were able to be transmitted at least until the M5 generation, thus the reference.

Figure S7 legend: Typographical error: "to to" → "to".

__Response: __Corrected.

Figure S8B (Chromosome 5 labels): Could you adjust the position labels to maintain a consistent format with other chromosomes?

Response: This figure has been modified on request of another reviewer such that this is no longer applicable.

Lines 262, 277, 279: "Figure S11" should be corrected to "Figure S10".

---

__Response: __Corrected.

Line 269: "Figure S10" should be corrected to "Figure S11B-H".

---

__Response: __Corrected.

Reviewer #3 (Significance (Required)):

In mutation studies aimed at inducing large-scale genomic variations, irradiation has traditionally been the primary method for mutagenesis. However, this study proposes a more efficient and accessible alternative using chemical mutagenesis with a DNA topoisomerase II inhibitor. Genomic analysis of mutants generated through this treatment revealed extensive genomic alterations, with a mutation frequency exceeding that of gamma irradiation-induced mutants. These findings suggest that this approach has the potential to advance mutation research for plant biologists and breeders seeking efficient methods for trait improvement. Furthermore, the authors integrate RNA-seq analysis for selected traits, demonstrating a systematic workflow for candidate gene identification and facilitating the determination of causal genes.

---

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:

The manuscript entitled "A simple method to efficiently generate structural variation in plants" by Bechen et al. investigated an efficient mutagen for inducing large structural variations in plants, replacing traditional irradiation methods with a chemical mutagenesis strategy. The study examined the effects of etoposide, a DNA topoisomerase II inhibitor, on structural variations and demonstrated that etoposide treatment induces a wide range of phenotypic and genome changes, including inversions, duplications, and deletions. Additionally, the authors analyzed the relationship between gene expression changes and genomic alterations to identify potential causal genes underlying specific phenotypes. While their findings provide clear and reliable evidence of structural variations induced by etoposide, I have several suggestions to enhance the clarity of their results, as detailed below.

Major comments:

1. Lines 166-169: My understanding is that you selected etoposide-treated M1 plants based on specific phenotypes, and observed their M2 and M3 progeny, categorizing them as either phenotype-positive or phenotype-negative. In Table S3, phenotypes other than BR-like dwarf, virescent, and short internode dwarf are not mentioned. Does this indicate that these other lines did not exhibit heritable phenotypic traits? If other lines showed some phenotype changes, could you incorporate progeny relationships along with phenotype information into Table S3? Additionally, in Figures S2 and S3, you reference 26A lines. Did they exhibit similar phenotypic changes among them?
2. Lines 187-189 and Figure S4: The assessment of repeat copy number variation provides valuable insights. However, based on the figure, the conclusion that "etoposide treatment likely did not trigger genomic instability in repetitive DNA" is difficult to interpret. Could you modify the figure into a box plot with raw data points and include a statistical analysis to support this conclusion?
3. Line 200 and Figure S8A: You state that SNV analysis identified a similar number of SNVs in treated and control plants. However, this is not easily interpretable from the figure. Could you include a statistical comparison between etoposide-treated and control plants? For example, EMS mutagenesis is known to induce specific G/C → A/T transitions. Did etoposide-treated and control plants exhibit the same types of nucleotide changes, or were there differences in the mutation spectrum?
4. Lines 219-220: Your conclusion clearly demonstrates the detection of numerous structural variations using both short- and long-read sequencing technologies. Could you provide a summary table listing the detected mutation positions? Since short-read sequencing is generally less effective in detecting large structural variations, I am particularly interested in evaluating the accuracy of Lumpy Express in identifying mutations.
5. Figures 3E-G: To facilitate a clearer comparison of the effects of structural variations on gene expression between BR-like dwarf and short internode dwarf, could you add an average trend line to the figures, similar to Figure 3B?

Minor comments:

Line 105 and Figure 1A: In the manuscript, etoposide concentrations are stated as 0, 40, 80, and 160 μM, whereas Figure 1A labels the concentrations as 0, 80, 160, and 320 μM. Should the figure be updated to 0, 40, 80, and 160 μM for consistency?

Figure 1B legend: Typographical error: "roundsof" → "rounds of".

Line 109: Do you have a summary table for the M1 generation? If so, could you provide it as a supplementary table?

Line 119: Figure 1B only defines developmental stages. To improve clarity, consider revising "Figure 1B" to "Figure 1B-F", allowing readers to easily understand the corresponding figures.

Line 121: The citation "(Figure S1, Table S1)" would be clearer if placed at the end of the sentence.

Lines 137, 148, 167: To maintain consistency with Figure 1C-F and the manuscript's logical flow, could you standardize the order of phenotypes as "virescent, short internode dwarf, and BR-like dwarf" instead of the current variation?

Line 139: Why is "Figure 1B" referenced at this position? Would it be more appropriate to remove this reference?

Figure S7 legend: Typographical error: "to to" → "to".

Figure S8B (Chromosome 5 labels): Could you adjust the position labels to maintain a consistent format with other chromosomes?

Lines 262, 277, 279: "Figure S11" should be corrected to "Figure S10".

Line 269: "Figure S10" should be corrected to "Figure S11B-H".

Significance

In mutation studies aimed at inducing large-scale genomic variations, irradiation has traditionally been the primary method for mutagenesis. However, this study proposes a more efficient and accessible alternative using chemical mutagenesis with a DNA topoisomerase II inhibitor. Genomic analysis of mutants generated through this treatment revealed extensive genomic alterations, with a mutation frequency exceeding that of gamma irradiation-induced mutants. These findings suggest that this approach has the potential to advance mutation research for plant biologists and breeders seeking efficient methods for trait improvement. Furthermore, the authors integrate RNA-seq analysis for selected traits, demonstrating a systematic workflow for candidate gene identification and facilitating the determination of causal genes.

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 describes mutagenesis of Arabidopsis by a topoisomerase II inhibitor. The method is effective, resulting in good density of SV and no detectable SNV. The authors provide a full characterization of the mutants, their phenotypes, and their genomes. The 34-sample selected for genomic analysis is sufficient to make firm conclusions.
• The manuscript is clearly written and illustrated.
• The manuscript does a very good job at covering the phenotypic and molecular analysis for this type of mutagenesis. For example, they highlight the difference between short and long reads in the identification of SV.
• The figures are very clear, with the exception of Fig. 3, which I found harder to follow. It would be enhanced by describing the candidate lesion(s) in the first panel of each mutant series. This would clarify the expectation. For example, larger indels (not examined here) should be associated with higher (insertion) or lower (deletion) expression of the affected genes. In the cases presented in Fig.3, the structural changes do not suggest obvious hypotheses. The authors examine the regions near breakpoint of inversions or near small indels. It makes sense, but it does not make the figure very digestible. The connected text in the results, on the other hand, is very clear. Perhaps, making the conclusions in the figure legend as well? As a connected thought, it would have been useful to provide expression data for a large indel exemplifying the cis/trans nature of regulatory changes.
• Deletions and other rearrangements may affect meiosis as noted by the authors. In addition, they can display gametophytic phenotypes and a deficit in transmission. The likelihood increases with the size of the indel. Large indels are not transmitted. Accordingly, for indels above a certain size, it is not possible to determine the number of causal loci from F2 ratios.
• Although the use of topo II inhibitors for mutagenesis in plants is novel, the mutagenic effects described here are well documented in animals. This should be acknowledged (e.g. Heisig, Mutagen. 2009; Ferguson, Env Mol Mutagen. 1994)

Referees cross-commenting

I also found the expression analysis confusing and in need of revision. One reason is that a set of clear expectations were not provided. I believe that the RNAseq analysis is expected to help identify the gene(s) that underlie a trait. For example, genes located on an indel are likely to display expression proportional to copy number. Also, a new junction or translocation could influence expression of the gene next to the break point. The authors should make this clear in the figure and the text.

Significance

• I appreciated the description of the method. It should be widely applicable. In arabidopsis, it requires sustained growth in the presence of the inhibitor. This could limit its applicability. For example, it may not be effective with pollen because exposure by a short soaking period may not be sufficient. Culturing of large seeded species is possible, but adds complexity. In this context, radiations have advantages. I do agree with the authors on the difficulty in identifying a source. However, once one is found, radiation treatment is very simple and convenient.
• The manuscript describes a useful tool and the connected spectrum of mutations. It has the novelty, quality, and relevance to represent a significant contribution to plant biology and to be of broad interest.

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

This is review of the manuscript „A simple method to efficiently generate structural variation in plants" by Bechen et al. The manuscript presents a very interesting and innovative approach to generate structural variant mutations (including large ones) in the genome of Arabidopsis thaliana using a simple chemical treatment with TOPII inhibitor etoposide. Authors show that unlike chemical mutagens commonly used for induction of SNPs (EMS, sodium azide...), etoposide-treatment caused structural variants like DNA deletions, insertions, inversions and translocations. These mutations were identified by the whole genome short and long read sequencing that also indicated a WT-like frequency of SNPs. This finding can potentially help inducing mutations similar to high energy radiation in potentially any plant. First, the manuscript provides description of the unusual phenotypes found after etoposide treatment and their Mendelistic inheritance. Based on this, authors performed whole genome sequencing and mutation detection, validation. The experimental part ends by transcriptome analysis that authors use as the approach to identify the causal mutations. This part is, in my opinion, the weakest part of the manuscript and would benefit from further clarification or even additional experiments (see below).

Overall the manuscript is very clear and contains all necessary information. The only part that was confusing to me, was the section focusing on the transcriptome analysis.

Major points:

Line 222: In the section „RNA-Seq identifies genes that are associated with structural variation and mutant phenotypes", authors suggest that the changes in the transcript amount were used to identify causal mutations. I got confused by this section. Exach of the examples represents unique situation and thus only single cases are presented which makes it hard to estimate robustness of the presented approach. Also, the presented mutations have prominent phenotypes that were already heavily studied in the past and therefore the possible causal genes are mostly known. Therefore, I am not sure how this approach would stand in case of traits with unknown underlying genes. When refering to the case with the chromosomal inversion, I do not see how one will be able to map a candidate based on the relatively mild expression (but maybe I am missing something here). Similarly, the „mapping" approach applied to the variegated line would not be possible on a trait that is less studied and the candidates are not well known. I wond why authors did not perform association mapping on a bulk of phenotypically mutant plants collected from a segregating F2 backcross population. This might be a more robust way of linking the phenotype with a mutation.

Discussion section: I am missing discussion on how etoposide could be causing such structural variants.

Minor points:

Line 70: Possibly add sodium azide. It is frequently used as mutagen for some plant species. Line 122: „...etoposide is an excellent mutagen for efficiently creating large-effect mutations." This cannot be claimed at this point because the sequence analysis data were not shown yet. Please reformulate.

Referees cross-commenting

My main issue was the mapping protocol using transcriptomic changes. It is hard to believe that this approach would work well on unknown/less studied traits. What is your opinion?

Significance

Strengths Innovative way on how to induce structural variant mutations in plants.

Limitations The approach on how to map the mutations needs more development. At this point i tis not clear how well the approach will work in other plant species.

Audience basic and applied plant scientists

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

Manuscript number: RC-2025-02888

Corresponding author(s): Christian, Fankhauser

General Statements

We were pleased to see that the three reviewers found our work interesting and provided supportive and constructive comments.

Our answers to their comments and/or how we propose to address them in a revised manuscript are included in bold.

1. Description of the planned revisions

Reviewer #1 (Evidence, reproducibility and clarity (Required)):

Summary: Plant systems sense shading by neighbors via the phytochrome signaling system. In the shade, PHYTOCHROME-INTERACTING FACTORS (PIFs) accumulate and are responsible for transcriptional reprogramming that enable plants to mobilize the "shade-avoidance response". Here, the authors have sought to examine the role of chromatin in modulating this response, specifically by examining whether "open" or "closed" chromatin regions spanning PIF target genes might explain the transcriptional output of these genes. They used a combination of ATAC-seq/CoP-qPCR (to detect open regions of chromatin), ChIP (to assay PIF binding) and RNA-seq (to measure transcript abundance) to understand how these processes may be mechanistically linked in Arabidopsis wild-type and pif mutant lines. They found that some chromatin accessibility changes do occur after LRFR (shade) treatment (32 regions after 1h and 61 after 25 h). While some of these overlap with PIF-binding sites, the authors found no correlation between open chromatin states and high levels of transcription. Because auxin is an important component of the shade-avoidance response and has been shown to control chromatin accessibility in other contexts, they examined whether auxin might be required for opening these regions of chromatin. They find that in an auxin biosynthesis mutant, there is a small subset of PIF target genes whose chromatin accessibility seems altered relative to the wild-type. Likewise, they found that chromatin accessibility for certain PIF targets is altered in phyB and pif mutant and propose that PIFs are necessary for changing the accessibility of chromatin in these genes. The authors thus propose that PIF occupancy of already open regions, rather than increased accessibility, underly the increase in transcript of abundance of these target genes in response to shade.

Major comments: *• I find that the data generally support the hypothesis presented in the manuscript that chromatin accessibility alone does not predict transcription of PIF target genes in the shade. That said, I think that a paragraph from the discussion (lines 321-332) would benefit from some careful rephrasing. I think it is perfectly reasonable to propose that PIF occupancy is more predictive of shade-induced transcriptional output than chromatin accessibility, but I think that calling PIF occupancy "the key drivers" (line 323) or "the main driving force" (line 76) risks ignoring the observation that levels of PIF occupancy specifically do not predict expression levels of PIF target genes (Pfeiffer et al., 2014, Mol Plant). For PIL1 and HFR1, the authors have shown that PIF promoter occupancy and transcript levels are correlated, but the central finding of Pfeiffer et al. was that this pattern does not apply to the majority of PIF direct target genes. Finding factors (i.e. histone marks) that convert PIF-binding information into transcriptional output appears to have been the impetus for the experiments devised in Willige et al., 2021 and Calderon et al., 2022. It is great that the authors have outlined in the discussion that there are a number of factors that modulate PIF transcriptional activating activity but I think that the emphasis on PIF-binding explaining transcript abundance should be moderated in the text. *

We appreciate the reviewers’ comments and will address it by introducing appropriate changes to the discussion. One element that should be pointed out is that the study of Willige et al., 2021 allows us to look at sites where PIF7 is recruited in response to the shade stimulus (a low R/FR treatment) and relate this to higher transcript abundance of the nearby genes. The study of Pfeiffer et al., 2014 which analyses PIF ChIP studies from several labs does not include this dynamic view of PIF recruitment in response to a stimulus. For example, this study re-analyses data from our lab, Hornitschek et al., 2012, in which we did PIF5 ChIP in low R/FR, but we did not compare that to high R/FR to enable an analysis of sites where we see recruitment of PIF5 in response to a shade cue. In the revised manuscript we will also include a new figure comparing PIF7 recruitment and changes in gene expression at direct PIF target genes.

• I think that the hypothesis could be further supported by incorporating the previously published ChIP-seq data on PIF1, PIF3 and PIF5 binding. Given these data are published/publicly available, I think it would be helpful to note which of the 72 DARs are bound by PIF1, PIF3 and/or PIF5. Especially so given that PIF5 (Lorrain et al., 2008, Plant J) and PIF1/PIF3 (Leivar et al., 2012, Plant Cell) contribute at least in some capacity to transcriptional regulation in response to shade. At the very least, it might help explain some of the observed increases in nucleosome accessibility observed for genes that don't have PIF4 or PIF7-binding.* This is a thoughtful suggestion. Our choice to focus on PIF7 target genes is dictated by two reasons. First, the finding that amongst all tested PIFs, PIF7 is the major contributor to the control of low R/FR (neighbor proximity) induced responses in seedlings (e.g. Li et al., 2012; de Wit et al., 2016; Willige et al., 2021). In addition, the PIF7 ChIP-seq and gene expression data from the Willige et al., 2021 paper was obtained using growth conditions very similar to the ones we used, hence allowing us to compare it to our data. As the reviewer suggests, other PIFs also contribute to the low R/FR response and hence looking at ChIP-seq for those PIFs in publicly available data is also informative. One limitation of this data is that ChIP-seq was not always done in seedlings grown in conditions directly comparable to the conditions we used (except for PIF5, see above). Nevertheless, we have performed this analysis with the available data suggested by the reviewer and intend to include the results in the revised version of the manuscript, presumably updated Figure 4B.

• In the manuscript, there are several instances where separate col-0 (wild type) controls have been used for identical experiments. Specifically, qPCR (Fig 3C, Fig S7C/D and Fig S8C/D), CoP-qPCR (Fig 5B/5C and Fig S8E/F) and hypocotyl measurements (Fig S7A/B and Fig S8A/B). In the cases of the hypocotyl measurements, there appear to be hardly any differences between col-0 controls indicating the measurements can be confidently compared between genotypes.

We appreciate this comment but to be comprehensive, we like to include a Col-0 control for each experiment (whenever possible) and hence also include the data when available.

• In some cases of qPCR and CoP-qPCR experiments however, the differences in values obtained from col-0 samples that underwent identical experimental treatments appear to vary significantly. In Figure 3C for example, the overall trend for PIL1 expression in col-0 is the same (e.g. HRFR levels are low, LRFR1 levels are much higher and LRFR25 levels drop down to some intermediate level) but the expression levels themselves appear to differ almost two-fold for the LRFR 1h timepoint (~110 on the left panel vs ~60 for the right panel). Given the size of the error bars, it appears that these data represent the mean from only one biological replicate. PIL1 expression levels at LRFR 1h as reported in Fig S7C and D also show similar ~2-fold differences. __This is a good comment. Having looked at PIL1 gene induction by low R/FR in dozens of similar experiments made us realize that indeed while the PIL1 induction is always massive, the extent is somewhat variable. Based on the data that we have (including from RNA-seq) we are convinced that this is due to the very low level of expression of PIL1 in high R/FR conditions. Given that induction by low R/FR is expressed as fold increase relative to baseline high R/FR expression, small changes in the lowly expressed PIL1* in high R/FR leads to seemingly significant differences in its induction by low R/FR across experiments.__

All qPCR data is represented by three biological replicates, and the variation between them per experiment is low, which is reflected in the size of the SD error bars. Data on technical and biological replicates in each panel will be clearly indicated in the revised figure legends.

• I would recommend that the authors explicitly describe the number of biological replicates used for each experiment in the methods section. If indeed these experiments were only performed once, I think the authors should be very careful in the language used in describing their conclusions and in assigning statistical significance. One possibility that could also be helpful would be normalizing LRFR 1h and LRFR 25h values to HRFR values and plotting these data somewhere in the supplemental data. If, for example, the relative levels of PIL1 are different between replicates but the fold-induction between HRFR and LRFR 1h are the same, this would at least allay any concerns that the experimental treatments were not the same. I understand that doing so precludes comparison between genotypes, but I do think it's important to show that at least the control data are comparable between experiments.

* All qPCR and CoP-qPCR experiments have been performed with three 3 biological replicates as described in Materials and Methods section, and these are represented in the Figures. Relative gene expression in the qPCR experiments was normalized to two housekeeping genes YLS8 and UBC21 and afterwards to one biological replicate of Col-0 control in HRFR. As indicated for the previous comment information about replicates will be included in the updated figure legends.

• Similarly, for the CoP-qPCR experiments presented in Fig 5B and 5C, the col-0 values for region P2 between Fig 5B and 5C shows that while HRFR and LRFR 1h look comparable, the values presented for LRFR 25h are quite different.

* This comment of the reviewer prompted us to propose a different way of representing the data that is clearer (new Figure 5B and 5C). We believe that this facilitates the comparison between the genotypes. Enrichment over the input was calculated for the chromatin accessibility of each region. Chromatin accessibility was further normalized against two open control regions on the promoters of ACT2 (AT3G18780, region chr3:6474579: 6474676) and RNA polymerase II transcription elongation factor (AT1G71080 region chr1:26811833:26811945). The difference between previous representation is that the regions are not additionally subtracted to Col-0 in HRFR. We will update the Materials and Methods and figure legend sections with this information.

Minor comments: • Presentation of Supplemental Figure 7A/7B and Supplemental Figure 8A/8B could be changed to make the data more clear (i.e. side-by-side rather than superimposed).

We propose changing the presentation of the hypocotyl length data to show the values for days side-by-side as the Reviewer suggests.

• I think that the paragraph introducing auxin (lines 25-37) could be reduced to 1-2 sentences and merged into a separate introductory paragraph given that the SAV3 work makes up a relatively minor component of the manuscript.

* We agree with the reviewer and will reduce the paragraph about auxin and merge it with the previous paragraph about transcription.

• • For Figure 3A, I would strongly encourage the authors to show some of the raw western blot data for PIF4, PIF5 and PIF7 protein abundance (and loading control), not just the normalized values. This could be put into supplemental data, but I think it should accompany the manuscript.

* We agree that presenting the raw data that was used for quantification is important. We will include the western blots used for quantifying PIF4, PIF5 and PIF7 protein abundance (and loading control DET3). This information will presumably be included to the Supplementary Figure 3C (figure number to be confirmed once we decide on all new data to be presented).

• Lines 145-147 "we observed a strong correlation between PIF4 protein levels (Figure 3A) and PIL1 promoter occupancy (Figure 3B), and a similar behavior was seen with PIF7 (Figure 3B)." According to Fig 3A, there is no statistically significant increase in PIF7 abundance after 1h shade. There is an apparent increase in PIF7 promoter occupancy, but the variation appears too large for it to be statistically significant. I agree that qualitatively there is a correlation for PIF4 but I think the description of the behavior of PIF7 should be rephrased.

* __As suggested by the reviewer, we will rephrase this paragraph to more accurately account for our data and also what was reported by others (e.g. Willige et al, 2021, in Li et al, 2012) regarding the regulation PIF7 levels and phosphorylation in response to a low R/FR treatment. __

• There appear to be issues in the coloring of the labels (light blue dots vs dark blue dots) for the PIF7 panels of Fig 3B and Supplemental Fig 3B.*

We thank the reviewer for pointing this out. This will be clarified by appropriate changes in the figure to avoid confusion in the revised version of Figure 3B.

Reviewer #1 (Significance (Required)):

This authors here have sought to examine the possibility that the transcriptional responses to shade mediated by the phy-PIF system might involve large-scale opening or closing of chromatin regions. This is an important and unanswered question in the field despite several studies that have looked at the role of histone variants (H2A.Z) and modifications (H3K4me3 and H3K9ac) in modulating PIF transcriptional activating activity. The authors have shown that, at least in the case of the transcriptional response to shade mediated by PIF7 (and to an extent PIF4), large-scale changes in chromatin accessibility are not occurring in response to shade treatment.

The results presented in this study support the hypothesis that large-scale changes in chromatin accessibility may have already occurred before plants see shade. This opens the possibility that perhaps the initial perception of light by etiolated (dark-grown seedlings) might trigger changes in chromatin accessibility, opening up chromatin in regions encoding "shade-specific" genes and/or closing chromatin in regions encoding "dark-specific" genes.

The audience for this particular manuscript encompasses a fairly broad group of biologists interested in understanding how environmental stimuli can trigger changes in chromatin reorganization and transcription. The results here are important in that they rule out chromatin accessibility changes as underlying the changes in transcription between the short-term and long-term shade responses. They also reveal that there are a few cases in which chromatin accessibility does change in a statistically-significant manner in response to shade. These regions, and the molecular players which regulate their accessibility, merit further exploration.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

The study by Paulisic et al. explores the variations in chromatin accessibility landscape induced by plant exposure to light with low red/far-red ratios (LRFR), which mimicks neighbor shade perception. The authors further compare these changes with the genome association of PIF4 and PIF7 transcription factors - two major actors of gene expression regulation in response to LRFR. While this is not highlighted in the main text, the analyses of chromatin accessibility are performed on INTACT-mediated nucleus sorting, presumably to ensure proper and clean isolation of nuclei.

Major comments

• Why is the experimental setup exposing plants to darkness overnight? Does this affect the response to LRFR, by a kind of reset of phytochrome signaling? I guess this choice was made to maintain a strong circadian rhythm. Yet, given that PIF genes are clock-regulated, I am afraid that this choice complicates data interpretation concerning the specific effects of LRFR exposure.

There appears to be some confusion which prompts us to better explain our protocol both by changing Figure 1A (that outlines the experimental conditions) and in the text.

Seedlings are grown in long day conditions because this is more physiologically relevant than growing them in constant light, which is a rather unnatural condition.

The reviewer is correct that PIF transcription is under circadian control and the shade avoidance response is gated by the circadian clock (e.g. Salter et al., 2003). To prevent conflating circadian and light quality effects, all samples that are compared are harvested at the same ZT (circadian time – hours after dawn). This allows us to focus our analysis on light quality effects specifically. We are therefore convinced that our protocol does not complicate the interpretation of the LRFR effects reported here.

• As a result of this setup, the 1h exposure to LRFR immediately follows HRFR while the 3h final LRFR exposure of the « 25h LRFR » samples immediately follows a long period of darkness. Can this explain why in several instances (e.g., at the ATHB2 gene) 1h LRFR seems to have stronger effects than 25h LRFR on chromatin accessibility?* Please check the explanation above. Both samples are harvested at the same ZT (ZT3, meaning 3 hours after dawn). The 1h LRFR seedlings went through the night, had 2 hours of HRFR then 1h of LRFR. The 25h are harvested at the very same ZT, meaning 3h after dawn. Importantly, the HRFR control was also harvested at ZT3, meaning 3h after dawn. As indicated above this protocol allows us to focus on the light quality effects by comparing samples that are all harvested at the same ZT.

We expect that the changes in Fig. 1A and associated text changes will clarify this issue.

• Lane 42 cites the work by Calderon et al 2022 as « Transcript levels of these genes increase before the H3K4me3 levels, implying that H3K4me3 increases as a consequence of active transcription ». Despite this previous study being reviewed and published, such a strong conclusion should be taken cautiously, and I disagree with it. The study by Calderon et al compares RNA-seq with ChIP-seq data, two methodologies with very different sensitivity, especially when employing bulk cells/whole seedlings as starting materials. For example, a gene strongly induced in a few cells will give a good Log2FC in RNA-seq data analysis (as new transcripts are produced after a low level of transcripts before shade) but, even though its chromatin variations would follow the same temporality or would even precede gene induction, this would be invisible in bulk ChIP-seq data analysis (which averages the signal of all cells together). I understand the rationale for relying on the conclusions made in an excellent lab with strong expertise in light signaling, but I recommend being cautious when relying on these conclusions to interpret new data.* We agree with this comment, and we will change the text to reflect this.

• The problem is that the same issue holds true when comparing ATAC-seq and RNA-seq data. ATAC signals reflect average levels over all cells while RNA-seq data can be influenced by a few cell highly expressing a given gene. Even though authors carefully sorted nuclei using an INTACT approach, this should be discussed, in particular when gene clusters (such as cluster C-D) show no match between chromatin accessibility and transcript level variations. In this regard, is PIF7 expressed in many cells or a small niche of cells upon LRFR exposure? The conclusions on its role in chromatin accessibility, analyzed here as mean levels of many different seedling cells, could be affected by PIF7 activity pattern (e.g., at lane 293). __This is a good comment. PIF7 is expressed in the cotyledons and leaves in LD conditions (Kidokoro et al, 2009, Galvao et al, 2019), and few available scRNA-seq datasets indicate an enrichment of PIF7 in the epidermis (Kim et al, 2021, Lopez-Anido et al, 2021). LRFR exposure only mildly represses PIF7* expression as seen in Figure 3A and also in our bulk RNA-seq study (Table S4). We will discuss this potential limitation to our study in a revised version of the manuscript.__

• Lane 89, the conclusion linking DNA methylation and DNA accessibility is unclear to me, this may be rephrased. Also, it should be noted that in gene-rich regions, most DNA methylation is located along the body of moderately to highly transcribing genes (gene-body methylation) while promoters of active and inactive genes are most frequently un-methylated.* We will rephrase to better reflect the presence or absence of DNA methylation on promoter regions of shade regulated genes that contain accessible sites.

• Figure 3B shows a few ChIP-qPCR results with important conclusions. Why not sequencing the ChIPped DNA to obtain a genome-wide view of the PIF4-PIF7 relationships at chromatin, and also consequently a more robust genome-wide normalization?

* Several studies have shown that in the conditions that we studied here: transfer of seedlings from high R/FR (simulated sun) to low R/FR (neighbor proximity), amongst all PIFs, PIF7 is the one that plays the most dominant function (e.g. Li et al., 2012; de Wit et al., 2016; Willige et al., 2021). PIF4 and PIF5 also contribute but to a lesser extent. Given that Willige et al., 2021 did extensive ChIP-seq studies for PIF7 using similar conditions to the ones we used, we decided to rely on their data (that we re-analyzed), rather than performing our own PIF7 ChIP-seq analysis. While also performing a ChIP-seq analysis for PIF4 in similar conditions might be useful (this data is not available as far as we know), we are not convinced that doing that experiment would substantially modify the message. In the revised version we will also include analysis of the data from Pfeiffer et al., 2014, which comprises a ChIP-seq. dataset for PIF5 (the closest paralog of PIF4) initially performed by Hornitschek et al., in 2012 in low R/FR conditions (see comment to reviewer 1 above). For new ChIP-seq, we would have to make this experiment from scratch with substantially more material than what we used for the targeted ChIP-qPCR analyses. We thus do not feel that such an investment (time and money) is warranted.

• • Given the known functional interaction between PIF7 and INO80, it would be relevant to test whether changes in chromatin accessibility at ATHB2 and other genes are affected in ino80 mutant seedlings. __We agree with the reviewer that this is potentially an interesting experiment. This will allow us to determine whether the nucleosome histone composition has an influence on nucleosome positioning at selected shade-regulated genes (e.g. ATHB2). We note that according to available data, the effect of INO80 would be expected once PIF7 started transcribing shade-induced genes. We therefore propose comparing the WT with an ino80 mutant for their seedling growth phenotype, expression of selected shade marker gene (e.g. ATHB2*) and chromatin accessibility before (high R/FR) and after low R/FR treatment at selected shade marker genes. This will allow us to determine whether INO80 influences chromatin accessibility prior to a low R/FR treatment and/or once the treatment started. Our plan is to include this data in a revised version of the manuscript. __
• On the same line, it would be interesting to test whether PIF7 target regions with pre-existing accessible chromatin would exist in ino80 mutant plants. In other words, testing a model in which chromatin remodeling by INO80 defines accessibility under HRFR to enable rapid PIF recruitment and DNA binding upon LRFR exposure.*

See our answer just above.

Minor comments

*• In Figure 1C, it seems that PIF7 target genes do not match the set of LRFR-downregulated genes (even less than at random). Why not exclude these 4 genes from the analyses? *

This is correct. There are indeed only 4 downregulated PIF7 target genes as we define them. Removing these genes from the analyses does not change our interpretation of the data and hence for completeness we propose keeping them in a revised version of the manuscript

• Figure 3A shows the quantification of protein blots, but I did not find the corresponding images. These should be shown in the figure or as a supplementary figure with proper controls.

* We will include the raw Westen blots used for quantification of PIF4, PIF5 and PIF7 in the revised version of the manuscript

• • Lane 102, it is unclear why PIF7 target genes were defined as the -3kb/TSS domains while Arabidopsis intergenic regions are on average much shorter. Gene regulatory regions, or promoters, are typically called within -1kb/TSS regions to avoid annotating a ChIP peak to the upstream gene or TE. A better proxy of PIF7 typical binding sites in gene regulatory regions could be determined by analysing the mean distance between PIF7 peak coordinates and the closest TSS. Typically, a gene meta-plot would give this information. __We agree that the majority of PIF7 binding peaks are close to the 5’ of the TSS based on the PIF7 binding distribution meta-plot. But several known PIF binding sites are actually further upstream than 1kb 5’ of the TSS (e.g. ATHB2 and HFR1). However, we re-analyzed the data using your suggestion with -2kb/TSS and -1kb/TSS and while the number of target genes is reduced, it does not change our conclusions about PIF7 binding sites being located on accessible chromatin regions. Importantly, some well characterized LRFR induced genes such as HFR1* would not be annotated correctly if only peaks closest to the gene TSS were taken into account, without flanking genes. In this case only the neighboring AT1G02350 would be annotated, hence missing some important PIF7 target genes. Taking this into consideration we will not modify this part of the analysis in a revised manuscript.__
• Figure 4B, what's represented in the ATAC-seq heatmap: does a positive z-score represent high accessibility?*

On the ATAC-seq heatmap we have represented z-scores of the average CPM (counts per million) for accessible chromatin regions. Z-scores are calculated by subtracting the average CPM from the median of averaged CPMs for each accessible chromatin region and then divided by the standard deviation (SD) of those averaged CPMs across all groups per accessible region (in our case a group is an average of three biological replicates for either HRFR, 1h or 25h of LRFR). In that sense, z-score indicates a change in accessibility, where higher z-score indicates opening of the region and lower z-score indicates a region becoming more closed when compared among the three light treatments (HRFR, 1h or 25h of LRFR). We will make sure that this is clear in the revised manuscript. Reviewer #2 (Significance (Required)):

Contradicting the naive hypothesis that PIFs may target shade-inducible genes to « open » chromatin of shade-inducible genes with the help of chromatin remodelers, such as INO80, the study highlights that PIF7 typically associates with pre-existing accessible chromatin states. Thus, even though this is not stated, results from this study indicate that PIF7 is not a pioneer transcription factor. The data seem very robust, and while some conclusions need clarification, it should be of great interest to the community of scientists studying plant light signaling and shade responses.

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

In their manuscript, Paulisic et al. investigate whether the transcriptional response of Arabidopsis seedlings to shade depends on chromatin accessibility, with a specific focus on PIF7-regulated genes. To this end, they perform ATAC-seq and RNA-seq, along with other experiments, on seedlings exposed to short and long shade and correlate the results with previously reported PIF7 and PIF4 ChIP-seq data. Based on their findings, they propose that shade-mediated transcriptional regulation may not require extensive remodeling of DNA accessibility. Specifically, they suggest that the open chromatin conformation allows PIFs to easily access and recognize their binding motifs, rapidly initiating gene expression in response to shade. This transcriptional response primarily depends on a transient increase in PIF stability and gene occupancy, with changes in chromatin accessibility occurring in only a small number of genes.

Major comments: * • I have one issue that, in my opinion, requires more attention. To define the PIF7 target genes, which were later used to estimate whether PIF7 binds to open or closed chromatin and affects DNA accessibility after its binding, the authors compared the 4h LRFR data point from Willige et al. (2021) ChIP-seq with their 1h RNA-seq data point. This comparison might have missed early genes where PIF7 binds before the 1h time point but is no longer present on DNA at 4h. I understand the decision to choose the 4h Willige et al. ChIP-seq data point, performed under LD conditions, as it matches the data in this study, rather than the 5min-30min data points, which were conducted in constant light. However, if possible, it would be interesting to also compare the RNA-seq data with the early PIF7 binding genes to assess how many additional PIF7 target genes could be identified based on that comparison and whether this might alter the conclusions. If the authors do not agree with this point, it should at least be emphasized that the ChIP-seq data and the RNA-seq/ATAC-seq data were performed under different LRFR conditions (R/FR 0.6 vs. 0.1), which may lead to the misidentification of PIF7 target genes in the manuscript.*

1) This is an interesting suggestion, we therefore reanalyzed 5, 10 and 30 min ChIP-seq timepoints from Willige et al, 2021 and compared them to 4h of LRFR (ZT4). We have crossed these lists of potential PIF7 targets with our 1h LRFR PIF457 dependent genes based on our RNA-seq. While some PIF7 targets appear only in early time points 5-10 min of LRFR exposure, overall, the number and composition of PIF7 target genes is rather constant across these timepoints. We propose to include these additional analyses in a revised version of the manuscript as a supplemental figure. However, these additional analyses do not influence our general conclusions.

2) The comment regarding the R/FR ratio is important. We will point this out although the conditions used by Willige et al., 2021 and the ones we used are similar, they are not exactly the same in terms of R/FR ratio. Importantly, in both studies the early transcriptional response largely depends on the same PIFs, many of the same response genes are induced (e.g. PIL1, AtHB2, HFR1, YUC8, YUC9 and many others) and the physiological response (hypocotyl elongation) is similar. This shows that this low R/FR response yields robust responses.

Minor comments: • In Fig. 1D, please describe the meaning of the blue shaded areas and the blue lines under the ATAC-seq peaks, as they do not always correlate.

The shaded areas and the bars define the extension of the ATAC-seq accessible chromatin peaks. We will add the meaning of the shaded areas and the blue bars in the Figure legend and correct the colors in a revised manuscript

• In Fig. 1E, it could be helpful to note that the 257 peaks in the right bar correspond to the peaks associated with the 177 genes in the left bar.* We will update Figure 1E and Figure legends for better understanding as the Reviewer suggested.

• In lines 116, 119, and 122, I believe it should read "Fig. 2" instead of "Fig. 2A."* We thank the Reviewer for noticing the error that we will correct.

• Lines 138-139: "PIF7 total protein levels were overall more stable, and only a mild and non-significant increase of PIF7 levels was seen at 1 h of LRFR." Since PIF7 usually appears as two bands in HRFR and only one band in LRFR, how was the protein level of PIF7 quantified in Fig. 3A? Additionally, I was wondering about the authors' thoughts on the discrepancy with Willige et al. (2021, Extended Data Fig. 1d), where PIF7 abundance seems to increased after 30 min and 2 h of LRFR.* PIF7 protein levels were quantified by considering both the upper and the lower band in HRFR (total PIF7) and normalizing its levels to DET3 loading control. We still observe an increase in the total PIF7 protein levels at 1h of LRFR, however this change was not statistically significant in these experiments. In our conditions as in Willige et al, 2021, the increase in PIF7 protein levels to short term shade seems consistent as is the pronounced shift or disappearance of the upper band (phosphorylated form) on the Western blots (raw data will be available in the revised manuscript). We will introduce text changes referring to the phosphorylation status of PIF7 in our conditions.

• Line 150: "... many early PIF target genes (Figure 3C)." Since only PIL1 is shown in Fig. 3C, I would recommend revising this sentence. Alternatively, the data could be presented, as in Fig. 2, for all the PIF7 target genes with transient expression patterns.

* We will introduce changes in the text to reflect that we only show PIL1 in the main Figure 3C.

• Line 204: I'm not sure if Supplementary Fig. 7C-D is correct here. If it is, could the order of the figures be changed so that Supplementary Fig. 7C-D becomes Supplementary Fig. 7A-B?*

The order of the panels A-B in the Supplementary Figure 7 follows the order of the text in the manuscript and is mentioned before panels C-D. It refers to the sentence “Overexpression of phyB resulted in a strong repression of hypocotyl elongation in both HRFR and LRFR, while the absence of phyB promoted hypocotyl elongation (Supplementary Figure 7A-B).”

• • Line 208: "In all three cases...". Please clarify what the three cases refer to. __We will change the text to more explicitly refer to the differentially accessible regions (DARs) of the genes ATHB2 and HFR1* shown in Figure 5A.__
• Line 231: Should Fig. 5C also be cited here in addition to Supplementary Fig. 7?* We will add the reference to Figure 5C that was missing.

*• In Supplementary Table 3, more information is needed. For example, it could mention: "This data is presented in Fig. 3 and is based on datasets from ChIP-seq, RNA-seq, etc."

*

The table will be updated with more information as suggested by the Reviewer.

• In the figure legend of Fig. 4B, please check the use of "( )".*

We will correct the error and include the references inside the parenthesis.

Reviewer #3 (Significance (Required)):

Paulisic et al. present novel discoveries in the field of light signaling and shade avoidance. Their findings extend our understanding of how DNA organization, prior to shade, affects PIF binding and how PIF binding remodels DNA accessibility. The data presented support the conclusions well and are backed by sufficient experimental evidence.

2. Description of the revisions that have already been incorporated in the transferred manuscript

The manuscript has not been modified yet.

3. Description of analyses that authors prefer not to carry out

• *

Reviewer 2 asked for new ChIP-seq analyses for PIF7 and PIF4. For reasons that we outlined above, we believe that such analyses are not required, and we currently do not intend performing these experiments.

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 their manuscript, Paulisic et al. investigate whether the transcriptional response of Arabidopsis seedlings to shade depends on chromatin accessibility, with a specific focus on PIF7-regulated genes. To this end, they perform ATAC-seq and RNA-seq, along with other experiments, on seedlings exposed to short and long shade and correlate the results with previously reported PIF7 and PIF4 ChIP-seq data. Based on their findings, they propose that shade-mediated transcriptional regulation may not require extensive remodeling of DNA accessibility. Specifically, they suggest that the open chromatin conformation allows PIFs to easily access and recognize their binding motifs, rapidly initiating gene expression in response to shade. This transcriptional response primarily depends on a transient increase in PIF stability and gene occupancy, with changes in chromatin accessibility occurring in only a small number of genes.

Major comments:

I have one issue that, in my opinion, requires more attention. To define the PIF7 target genes, which were later used to estimate whether PIF7 binds to open or closed chromatin and affects DNA accessibility after its binding, the authors compared the 4h LRFR data point from Willige et al. (2021) ChIP-seq with their 1h RNA-seq data point. This comparison might have missed early genes where PIF7 binds before the 1h time point but is no longer present on DNA at 4h. I understand the decision to choose the 4h Willige et al. ChIP-seq data point, performed under LD conditions, as it matches the data in this study, rather than the 5min-30min data points, which were conducted in constant light. However, if possible, it would be interesting to also compare the RNA-seq data with the early PIF7 binding genes to assess how many additional PIF7 target genes could be identified based on that comparison and whether this might alter the conclusions. If the authors do not agree with this point, it should at least be emphasized that the ChIP-seq data and the RNA-seq/ATAC-seq data were performed under different LRFR conditions (R/FR 0.6 vs. 0.1), which may lead to the misidentification of PIF7 target genes in the manuscript.

Minor comments:

- In Fig. 1D, please describe the meaning of the blue shaded areas and the blue lines under the ATAC-seq peaks, as they do not always correlate.

- In Fig. 1E, it could be helpful to note that the 257 peaks in the right bar correspond to the peaks associated with the 177 genes in the left bar.

- In lines 116, 119, and 122, I believe it should read "Fig. 2" instead of "Fig. 2A."

Lines 138-139: "PIF7 total protein levels were overall more stable, and only a mild and non-significant increase of PIF7 levels was seen at 1 h of LRFR."

Since PIF7 usually appears as two bands in HRFR and only one band in LRFR, how was the protein level of PIF7 quantified in Fig. 3A? Additionally, I was wondering about the authors' thoughts on the discrepancy with Willige et al. (2021, Extended Data Fig. 1d), where PIF7 abundance seems to increased after 30 min and 2 h of LRFR. Line 150: "... many early PIF target genes (Figure 3C)." Since only PIL1 is shown in Fig. 3C, I would recommend revising this sentence. Alternatively, the data could be presented, as in Fig. 2, for all the PIF7 target genes with transient expression patterns. - Line 204: I'm not sure if Supplementary Fig. 7C-D is correct here. If it is, could the order of the figures be changed so that Supplementary Fig. 7C-D becomes Supplementary Fig. 7A-B? - <br /> - Line 208: "In all three cases...". Please clarify what the three cases refer to. - <br /> - Line 231: Should Fig. 5C also be cited here in addition to Supplementary Fig. 7? - <br /> - In Supplementary Table 3, more information is needed. For example, it could mention: "This data is presented in Fig. 3 and is based on datasets from ChIP-seq, RNA-seq, etc." - <br /> - In the figure legend of Fig. 4B, please check the use of "( )".

Significance

Paulisic et al. present novel discoveries in the field of light signaling and shade avoidance. Their findings extend our understanding of how DNA organization, prior to shade, affects PIF binding and how PIF binding remodels DNA accessibility. The data presented support the conclusions well and are backed by sufficient experimental evidence.

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 by Paulisic et al. explores the variations in chromatin accessibility landscape induced by plant exposure to light with low red/far-red ratios (LRFR), which mimicks neighbor shade perception. The authors further compare these changes with the genome association of PIF4 and PIF7 transcription factors - two major actors of gene expression regulation in response to LRFR. While this is not highlighted in the main text, the analyses of chromatin accessibility are performed on INTACT-mediated nucleus sorting, presumably to ensure proper and clean isolation of nuclei.

Major comments

• Why is the experimental setup exposing plants to darkness overnight? Does this affect the response to LRFR, by a kind of reset of phytochrome signaling? I guess this choice was made to maintain a strong circadian rhythm. Yet, given that PIF genes are clock-regulated, I am afraid that this choice complicates data interpretation concerning the specific effects of LRFR exposure.
• As a result of this setup, the 1h exposure to LRFR immediately follows HRFR while the 3h final LRFR exposure of the « 25h LRFR » samples immediately follows a long period of darkness. Can this explain why in several instances (e.g., at the ATHB2 gene) 1h LRFR seems to have stronger effects than 25h LRFR on chromatin accessibility?
• Lane 42 cites the work by Calderon et al 2022 as « Transcript levels of these genes increase before the H3K4me3 levels, implying that H3K4me3 increases as a consequence of active transcription ». Despite this previous study being reviewed and published, such a strong conclusion should be taken cautiously, and I disagree with it. The study by Calderon et al compares RNA-seq with ChIP-seq data, two methodologies with very different sensitivity, especially when employing bulk cells/whole seedlings as starting materials. For example, a gene strongly induced in a few cells will give a good Log2FC in RNA-seq data analysis (as new transcripts are produced after a low level of transcripts before shade) but, even though its chromatin variations would follow the same temporality or would even precede gene induction, this would be invisible in bulk ChIP-seq data analysis (which averages the signal of all cells together). I understand the rationale for relying on the conclusions made in an excellent lab with strong expertise in light signaling, but I recommend being cautious when relying on these conclusions to interpret new data.
• The problem is that the same issue holds true when comparing ATAC-seq and RNA-seq data. ATAC signals reflect average levels over all cells while RNA-seq data can be influenced by a few cell highly expressing a given gene. Even though authors carefully sorted nuclei using an INTACT approach, this should be discussed, in particular when gene clusters (such as cluster C-D) show no match between chromatin accessibility and transcript level variations. In this regard, is PIF7 expressed in many cells or a small niche of cells upon LRFR exposure? The conclusions on its role in chromatin accessibility, analyzed here as mean levels of many different seedling cells, could be affected by PIF7 activity pattern (e.g., at lane 293).
• Lane 89, the conclusion linking DNA methylation and DNA accessibility is unclear to me, this may be rephrased. Also, it should be noted that in gene-rich regions, most DNA methylation is located along the body of moderately to highly transcribing genes (gene-body methylation) while promoters of active and inactive genes are most frequently un-methylated.
• Figure 3B shows a few ChIP-qPCR results with important conclusions. Why not sequencing the ChIPped DNA to obtain a genome-wide view of the PIF4-PIF7 relationships at chromatin, and also consequently a more robust genome-wide normalization?
• Given the known functional interaction between PIF7 and INO80, it would be relevant to test whether changes in chromatin accessibility at ATHB2 and other genes are affected in ino80 mutant seedlings.
• On the same line, it would be interesting to test whether PIF7 target regions with pre-existing accessible chromatin would exist in ino80 mutant plants. In other words, testing a model in which chromatin remodeling by INO80 defines accessibility under HRFR to enable rapid PIF recruitment and DNA binding upon LRFR exposure.

Minor comments

• In Figure 1C, it seems that PIF7 target genes do not match the set of LRFR-downregulated genes (even less than at random). Why not exclude these 4 genes from the analyses?
• Figure 3A shows the quantification of protein blots, but I did not find the corresponding images. These should be shown in the figure or as a supplementary figure with proper controls.
• Lane 102, it is unclear why PIF7 target genes were defined as the -3kb/TSS domains while Arabidopsis intergenic regions are on average much shorter. Gene regulatory regions, or promoters, are typically called within -1kb/TSS regions to avoid annotating a ChIP peak to the upstream gene or TE. A better proxy of PIF7 typical binding sites in gene regulatory regions could be determined by analysing the mean distance between PIF7 peak coordinates and the closest TSS. Typically, a gene meta-plot would give this information.
• Figure 4B, what's represented in the ATAC-seq heatmap: does a positive z-score represent high accessibility?

Significance

Contradicting the naive hypothesis that PIFs may target shade-inducible genes to « open » chromatin of shade-inducible genes with the help of chromatin remodelers, such as INO80, the study highlights that PIF7 typically associates with pre-existing accessible chromatin states. Thus, even though this is not stated, results from this study indicate that PIF7 is not a pioneer transcription factor. The data seem very robust, and while some conclusions need clarification, it should be of great interest to the community of scientists studying plant light signaling and shade responses.

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

Summary:

Plant systems sense shading by neighbors via the phytochrome signaling system. In the shade, PHYTOCHROME-INTERACTING FACTORS (PIFs) accumulate and are responsible for transcriptional reprogramming that enable plants to mobilize the "shade-avoidance response". Here, the authors have sought to examine the role of chromatin in modulating this response, specifically by examining whether "open" or "closed" chromatin regions spanning PIF target genes might explain the transcriptional output of these genes. They used a combination of ATAC-seq/CoP-qPCR (to detect open regions of chromatin), ChIP (to assay PIF binding) and RNA-seq (to measure transcript abundance) to understand how these processes may be mechanistically linked in Arabidopsis wild-type and pif mutant lines. They found that some chromatin accessibility changes do occur after LRFR (shade) treatment (32 regions after 1h and 61 after 25 h). While some of these overlap with PIF-binding sites, the authors found no correlation between open chromatin states and high levels of transcription. Because auxin is an important component of the shade-avoidance response and has been shown to control chromatin accessibility in other contexts, they examined whether auxin might be required for opening these regions of chromatin. They find that in an auxin biosynthesis mutant, there is a small subset of PIF target genes whose chromatin accessibility seems altered relative to the wild-type. Likewise, they found that chromatin accessibility for certain PIF targets is altered in phyB and pif mutant and propose that PIFs are necessary for changing the accessibility of chromatin in these genes. The authors thus propose that PIF occupancy of already open regions, rather than increased accessibility, underly the increase in transcript of abundance of these target genes in response to shade.

Major comments:

I find that the data generally support the hypothesis presented in the manuscript that chromatin accessibility alone does not predict transcription of PIF target genes in the shade. That said, I think that a paragraph from the discussion (lines 321-332) would benefit from some careful rephrasing. I think it is perfectly reasonable to propose that PIF occupancy is more predictive of shade-induced transcriptional output than chromatin accessibility, but I think that calling PIF occupancy "the key drivers" (line 323) or "the main driving force" (line 76) risks ignoring the observation that levels of PIF occupancy specifically do not predict expression levels of PIF target genes (Pfeiffer et al., 2014, Mol Plant). For PIL1 and HFR1, the authors have shown that PIF promoter occupancy and transcript levels are correlated, but the central finding of Pfeiffer et al. was that this pattern does not apply to the majority of PIF direct target genes. Finding factors (i.e. histone marks) that convert PIF-binding information into transcriptional output appears to have been the impetus for the experiments devised in Willige et al., 2021 and Calderon et al., 2022. It is great that the authors have outlined in the discussion that there are a number of factors that modulate PIF transcriptional activating activity but I think that the emphasis on PIF-binding explaining transcript abundance should be moderated in the text.

I think that the hypothesis could be further supported by incorporating the previously published ChIP-seq data on PIF1, PIF3 and PIF5 binding. Given these data are published/publicly available, I think it would be helpful to note which of the 72 DARs are bound by PIF1, PIF3 and/or PIF5. Especially so given that PIF5 (Lorrain et al., 2008, Plant J) and PIF1/PIF3 (Leivar et al., 2012, Plant Cell) contribute at least in some capacity to transcriptional regulation in response to shade. At the very least, it might help explain some of the observed increases in nucleosome accessibility observed for genes that don't have PIF4 or PIF7-binding.

In the manuscript, there are several instances where separate col-0 (wild type) controls have been used for identical experiments. Specifically, qPCR (Fig 3C, Fig S7C/D and Fig S8C/D), CoP-qPCR (Fig 5B/5C and Fig S8E/F) and hypocotyl measurements (Fig S7A/B and Fig S8A/B). In the cases of the hypocotyl measurements, there appear to be hardly any differences between col-0 controls indicating the measurements can be confidently compared between genotypes.

In some cases of qPCR and CoP-qPCR experiments however, the differences in values obtained from col-0 samples that underwent identical experimental treatments appear to vary significantly. In Figure 3C for example, the overall trend for PIL1 expression in col-0 is the same (e.g. HRFR levels are low, LRFR1 levels are much higher and LRFR25 levels drop down to some intermediate level) but the expression levels themselves appear to differ almost two-fold for the LRFR 1h timepoint (~110 on the left panel vs ~60 for the right panel). Given the size of the error bars, it appears that these data represent the mean from only one biological replicate. PIL1 expression levels at LRFR 1h as reported in Fig S7C and D also show similar ~2-fold differences.

I would recommend that the authors explicitly describe the number of biological replicates used for each experiment in the methods section. If indeed these experiments were only performed once, I think the authors should be very careful in the language used in describing their conclusions and in assigning statistical significance. One possibility that could also be helpful would be normalizing LRFR 1h and LRFR 25h values to HRFR values and plotting these data somewhere in the supplemental data. If, for example, the relative levels of PIL1 are different between replicates but the fold-induction between HRFR and LRFR 1h are the same, this would at least allay any concerns that the experimental treatments were not the same. I understand that doing so precludes comparison between genotypes, but I do think it's important to show that at least the control data are comparable between experiments.

Similarly, for the CoP-qPCR experiments presented in Fig 5B and 5C, the col-0 values for region P2 between Fig 5B and 5C shows that while HRFR and LRFR 1h look comparable, the values presented for LRFR 25h are quite different.

Minor comments:

Presentation of Supplemental Figure 7A/7B and Supplemental Figure 8A/8B could be changed to make the data more clear (i.e. side-by-side rather than superimposed).

I think that the paragraph introducing auxin (lines 25-37) could be reduced to 1-2 sentences and merged into a separate introductory paragraph given that the SAV3 work makes up a relatively minor component of the manuscript.

For Figure 3A, I would strongly encourage the authors to show some of the raw western blot data for PIF4, PIF5 and PIF7 protein abundance (and loading control), not just the normalized values. This could be put into supplemental data, but I think it should accompany the manuscript.

Lines 145-147 "we observed a strong correlation between PIF4 protein levels (Figure 3A) and PIL1 promoter occupancy (Figure 3B), and a similar behavior was seen with PIF7 (Figure 3B)." According to Fig 3A, there is no statistically significant increase in PIF7 abundance after 1h shade. There is an apparent increase in PIF7 promoter occupancy, but the variation appears too large for it to be statistically significant. I agree that qualitatively there is a correlation for PIF4 but I think the description of the behavior of PIF7 should be rephrased.

There appear to be issues in the coloring of the labels (light blue dots vs dark blue dots) for the PIF7 panels of Fig 3B and Supplemental Fig 3B.

Significance

This authors here have sought to examine the possibility that the transcriptional responses to shade mediated by the phy-PIF system might involve large-scale opening or closing of chromatin regions. This is an important and unanswered question in the field despite several studies that have looked at the role of histone variants (H2A.Z) and modifications (H3K4me3 and H3K9ac) in modulating PIF transcriptional activating activity. The authors have shown that, at least in the case of the transcriptional response to shade mediated by PIF7 (and to an extent PIF4), large-scale changes in chromatin accessibility are not occurring in response to shade treatment.

The results presented in this study support the hypothesis that large-scale changes in chromatin accessibility may have already occurred before plants see shade. This opens the possibility that perhaps the initial perception of light by etiolated (dark-grown seedlings) might trigger changes in chromatin accessibility, opening up chromatin in regions encoding "shade-specific" genes and/or closing chromatin in regions encoding "dark-specific" genes.

The audience for this particular manuscript encompasses a fairly broad group of biologists interested in understanding how environmental stimuli can trigger changes in chromatin reorganization and transcription. The results here are important in that they rule out chromatin accessibility changes as underlying the changes in transcription between the short-term and long-term shade responses. They also reveal that there are a few cases in which chromatin accessibility does change in a statistically-significant manner in response to shade. These regions, and the molecular players which regulate their accessibility, merit further exploration.

My fields of expertise are photobiology, photosynthesis and early seedling development.

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

Manuscript number: RC-2025-02953

Corresponding author(s): Andreas, Villunger

[The “revision plan” should delineate the revisions that authors intend to carry out in response to the points raised by the referees. It also provides the authors with the opportunity to explain their view of the paper and of the referee reports.

• *

The document is important for the editors of affiliate journals when they make a first decision on the transferred manuscript. It will also be useful to readers of the reprint and help them to obtain a balanced view of the paper.

• *

If you wish to submit a full revision, please use our "Full Revision" template. It is important to use the appropriate template to clearly inform the editors of your intentions.]

1. General Statements [optional]

*We would like to thank the reviewers for their constructive input and overall support. We appreciate to provide a provisional revision plan, as outlined here, and are happy to engage in additional communication with journal editors via video call, in case further clarifications are needed. *

2. Description of the planned revisions

Insert here a point-by-point reply that explains what revisions, additional experimentations and analyses are planned to address the points raised by the referees.

• *

Reviewer #1

__Evidence, reproducibility and clarity __

Summary: This manuscript by Leone et al describes the role of the PIDDosome in cardiomyocytes. Using a series of whole body and cardiomyocyte specific knockouts, the authors show that the PIDDosome maintains correct ploidy in these cells. It achieves this through inducing cell cycle arrest in cardiomyocytes in a p53 dependent manner. Despite this effect on ploidy, PIDDosome-deficient hearts show no structural or functional defects. Statistics and rigor appear to be adequate.

We thank this referee for taking the time to evaluate our work and their valuable comments. We assume that this reviewer by mistake indicates that the phenomenon we describe, depends on p53. As outlined in the abstract and throughout the manuscript, the effect is independent of p53, but may additionally still involve p21, acting along or parallel to the PIDDosome.

Major comments: 1. Figure 1 uses fluorescent intensity of a nuclear stain to determine ploidy per nucleus and they further separate the results into mononucleated, binucleated or multinucleated cells. It is hard to know how to interpret these results without further information or controls. Is there a good positive control that can be used to help to show whether this assay is quantitative? The differences are larger with the Raidd and caspase-2 knockouts than with the Pidd knockouts but this is not addressed.

*We appreciate this concern. Regarding a “good positive control” we can say that we follow state-of the art in the cardiomyocyte field and studies by the Evans (PMID: 36622904), Kuhn (PMID: 32109383), Bergmann (PMID: 26544945) and Patterson labs (PMID: 28783163, 36912240) all use the identical approach to discriminate 2n from 4n nuclei in microscopy images at the cellular level. The fact that the majority of rodent CM nuclei is indeed diploid (PMID: 31175264, 31585517 and 32078450) and a large number of nuclei has been evaluated to assess their mean fluorescence intensity (MFI) reduces the risk of a systematic bias in our analysis. Moreover, we have used an orthogonal approach that is indeed quantitative to define DNA content, i.e,. flow-cytometry based evaluation of DNA content in isolated CM nuclei (Fig. 1C). We hence are confident our assays are quantitative. *

Regarding the fact that loss of Pidd1 causes a more saddle phenotype, we can offer to discuss this in light of the fact that Pidd1 has additional functions, outside the PIDDosome (PMID: 35343572), and that we made similar observations when analyzing ploidy in hepatocytes (PMID: *31983631). Given the fact that all components of the PIDDosome show a similar phenotype, and that this phenotype is mimicked by loss of the protein that connects PIDD1 and centrosomes, ANKRD26 (Fig. 4a), we are confident that this biological variation in our analysis is not affecting our conclusions. *

On line 459 the authors state that the increase in polyploidy in PIDDosome knockouts occurs in adult hood but this is not directly tested. In fact, in the next section the polyploidy is assessed in early postnatal development. This statement should be explained or removed.

We see that we have made an unclear statement here. In fact, we first noted increases in ploidy in adult heart and then define the time window in development when this happens. This sentence will be rephrased.

In Figure 4. The authors obtained RNAseq data for P1, P7 and P14 but only show the differences with and without caspase-2 at P7. Given that the differences in ploidy are more significant at P14 (Fig 3D), all the comparisons should be shown along with analysis of whether the same genes/gene families are altered in the absence of caspase-2.

The reason why we focus on postnatal day 7 (P7) is that data from Alkass et al (PMID: 26544945) and other labs (PMID: 31175264 ) document that on this day the initial wave of binucleation peaks. Hence, we hypothesized that the PIDDosome must be active in most CM, which aligns well with the increased mRNA levels of all of its components (Figure 3). Interestingly, it seems that its action is tightly regulated, as mRNA of PIDDosome components drop on P10, suggesting PIDDosome shut-down or downregulation. Similar findings have been noted in the liver (PMID: *31983631). Alkass and colleagues also show that very few CMs enter another round of DNA synthesis between P7 and P14, and hence possible transcriptome changes in the absence of the PIDDosome will be strongly diluted. *

Please note that on P1, there is no difference between genotypes to be expected as all CM are mononucleated diploids and cytokinesis competent, as previously demonstrated (PMID: *26544945). Moreover, PIDDosome expression levels are extremely low (Fig. 3A). As such, no difference between genotypes are expected on P1. In addition, on P14 the ploidy phenotype observed in PIDDosome knockout mice reaches the maximum and ploidy increases are comparable to adult tissue. Thus, at this time the trigger for PIDDosome activation (cytokinesis failure) is no longer observed as the majority of CMs are post-mitotic, (PMID: 26247711). As such the impact of PIDDosome activation on the P14 transcriptome is most likely negligible. However, if desired, we can expand our bioinformatics analysis summarizing findings made related to DEGs over time in wt animals by comparing genotypes also on day 1 and day 14. In light of the above, analysis between genotypes on P7 holds still appears as the one most meaningful. *

Some validation of the RNAseq and/or proteomics results would be an important addition to this study

We agree with this notion and propose to validate key candidates related to cardiomyocyte proliferation and polyploidization, some of which we found to be differentially expressed at the mRNA level on day 7in the RNAseq data (e.g., p21, Foxm1, Kif18a, Lin37 and others)

Regarding the proteomics results, we face the challenge that we can only try to confirm if candidate proteins are likely caspase substrates in silico using DeepCleave*, and potentially pick one or two candidates linked to CM differentiation for further analysis in vitro and in heterologous cell based assays (e.g. 293T cells), as no bona-fide ventricular cardiomyocyte cell lines exist. Primary postnatal CMs are extremely difficult to transfect, nor they proliferate without drug-treatment, or fail cytokinesis ex vivo. *

Figure 4D: the authors make the conclusion that p21 is downstream of PIDD (et p53 independent). However, this is not supported by the data because the increase in 4N cells/decrease in 2N cells, although statistically significant, is nowhere near that of caspase-2 KO and caspase-2/p21 KO. Statistics should also compare p32KO with c2KO. In the absence of any other data, the more likely conclusion is that p21 is not involved.

*We agree that the findings related to the impact seen upon loss of p21 suggest that it is not the only effector involved in ploidy control and it may not even be an effector engaged by caspase-2, as C2/p21 DKO mice have an even higher ploidy increase, albeit not statistically significant. However, it is important to highlight that p21 (Cdkn1a) was found to be downregulated in our transcriptomic analysis suggesting an involvement in the caspace2-cascade. We are happy to highlight this when presenting the results and in the discussion. *

*We assume that this referee refers to p73 KO data that should be compared to Casp2 KO data (could be read as p73 or p53, but the latter we compare side by side with Casp2 in Fig. 4 already). As p73 KO mice were not found to be viable beyond day 7 (our attempt to find animals on day 10 failed, in line with published literature (PMID: 24500610, 10716451)), we can only offer to compare this data set to the data presented in Figure 3C, where we have analyzed ploidy increases on day 7 from wt and PIDDosome mutant mice. This re-analysis will show that only Caspase-2 mutant mice display a significant ploidy increase on P7, when compared to wt or p73 mutant animals, while no difference are noted between wt and p73 mutant mice (to be included in new Suppl. Fig. 3C) *

Minor comments: Suggest moving Figure 4A to Figure 3 as it seems to fit better there based on the citation of this figure in the text

*We can see some benefit in this recommendation and included panel 4A now in an updated version of Figure 3. *

Recommend enhancing the brightness of microscopy images in Figure 1E and 2D

We will try to improve image quality, may have been due to PDF conversion

---

Significance

This study provides interesting information for the role of the PIDDosome in protecting from polyploidy and adds to the body of work by this same group studying this pathway in the liver.

The main weakness in terms of significance is the lack of a phenotype in the hearts of these animals. Therefore, it is clear that ploidy (or at least PIDDosome dependent ploidy) has minimal impact on cardiac development.

We respectfully disagree with the comment that the lack of impact on cardiac function constitutes a weakness of our findings. Several studies on ploidy control in the liver (PMID 34228992) but importantly also heart (PMID: 36622904) have failed to document a clear impact of increased ploidy on organ function. This does not infer insignificance, but maybe rather that the context where this becomes relevant has not been identified. We are happy expand on this in our discussion

• *

The authors mention that they have not tried giving these mice an myocardial infarct (MI) or inducing any other type of cardiac damage. Although it is understood that these experiments are likely outside of the scope of the present study, without this information the impact of this study is moderate. I recommend expanding the discussion to provide a more in-depth possible rationale as to why ploidy perturbations do not lead to structural changes like in the liver.

Despite this, the insights to the pathway itself are interesting to investigators in the caspase-2 field if a little underdeveloped, especially concerning the role of p21.

My expertise is in cell death and caspase biology (especially caspase-2). I have sufficient expertise to evaluate all parts of this paper.

*As mentioned above, we will amend our conclusions on p21, in light of potential findings made when validating DEG candidates, as stated above. *

*We hope that the changes and amendments proposed here will be satisfactory to this referee to recommend publication of a revised manuscript. *

• *

---

---

Reviewer #2

__Evidence, reproducibility and clarity: __

__Summary: __

In this study, the authors investigated the role of the PIDDosome during cardiomyocyte polyploidization. PIDDosome is a multi-protein complex activating the endopeptidase Caspase-2, and shown to be involved in eliminating cells with extra centrosomes or in response to genotoxic stress (Burigotto & Fava, 2021, Sladky and Villunger, 2020). In both cases, the PIDDosome is recruited in a ANKRD26-dependent manner at the centrosomes leading to p53 stabilization and cell death (Burigotto & Fava, 2021; Evans et al., 2020; Burigotto et al., 2021).

Here, by studying mouse cardiomyocyte differentiation, the authors showed that PIDDosome is imposing ploidy restriction during cardiomyocyte differentiation. Importantly, in contrast to a previous report in the liver (Sladky et al., 2020), they showed that PIDDosome acts in a p53-independent manner in cardiomyocytes. Indeed, they suggested that PIDDosome controls ploidy in cardiomyocytes through p21 activation.

We want to thank this reviewer for the time taken to evaluate our work and provide critical feedback that will help to improve our revised manuscript.

__Major comments: __

In general the conclusions of the authors are well supported by the experiments. However, I would suggest the following experiments/analysis to strengthen the paper:

The authors should improve the Figure 1 to help the readers who are not familiar with cardiomyocyte polyploidization. For instance, I would suggest to add a scheme to summarize cardiomyocyte polyploidization (in terms of nuclear size, mono vs multi and so on).

We agree that a visual summary of the postnatal timing of CM polyploidization will be helpful for the generalist not familiar with the topic and have added a scheme, adapted from a study by Alkass et al. (PMID: *26544945), who elegantly defined the timing of this process during postnatal mice life (now Fig. 1A). *

Based on the images they presented in 1B, the authors should also measure the nuclear area or volume in the different conditions in which components of the PIDDosome were depleted. Indeed, these two parameters should be easier to conceptualize for the readers (instead of the fluorescence nuclear intensity). This could help to understand if the nuclear size is maintained between the different conditions and if this is comparable between mono, bi or multinucleated cardiomyocytes.

We have acquired this data and it can be used to provide additional information on nuclear area and/or volume. We propose to focus on re-analyzing data from wt, Casp2 and XMLC2CRE/Casp2f/f mice. The additional information can be included in Figures 1 & 2, respectively.

• In Figure 2A, the authors presented cross section of heart from animals showing that PIDDosome depletion has no effect on heart size. This is a surprising result since cardiomyocytes have higher ploidy levels and this could have an effect on their function. Since the importance of this observation, the authors should present a quantification of the heart size in the different conditions shown in Figure 2A.

We agree with this comment. We can measure the heart vs. body weight ratio or tibia length in adult Casp2-/- vs. WT (3 month old) in order to indirectly evaluate possible increases in CM size linked to increased ploidy.

Also, the percentage of cardiomyocytes presenting higher levels of ploidy seems quite low. The authors should discuss this point. In particular because this could explain the absence of consequences on heart size and function at steady state.

We agree with this conclusion and will expand on this in our discussion. It is important to note that as opposed to findings made in liver (PMID: *31983631), genetic manipulation of ploidy regulators such as E2f7/8 (PMID: 36622904), only led to modest changes in CM ploidy, suggesting that either a small band-width compatible with normal heart function exists, or that additional mechanisms exist that take control when these thresholds set by the PIDDosome or E2f7/8 are exceeded. These mechanisms could involve Cyclin G (PMID: 20360255), or TNNI3K (PMID: 31589606). Importantly, a recent publication has shown that overexpression of Plk1(T210D) and Ect2 from birth causes increased heart weight coupled with a minor decrease in CM size. These mice undergo to premature death (PMID: 39912233) suggesting that CM polyploidization is a tight regulated process regulated by several independent mechanisms during heart development. *

• *

In Figure 2D, the authors measured the cardiomyocyte cross-sectional area and concluded that removing PIDDosome components have no effect on cardiomyocyte cell size. Since it has been shown that ploidy increase is normally associated with an increase in cell area, the authors should measure cell area of cardiomyocytes analyzed in Figure 1B. It could be then interesting to establish a correlation with nuclear area and the mono, bi or multinucleated status. This will strengthen the results showing that ploidy increases without affecting cell area.

Indeed, studies in PIDDosome deficient livers suggest that tissue is containing fewer but bigger cells (PMID: *31983631). As opposed to the liver the percentage of cardiomyocytes presenting higher levels of ploidy is relatively low. Thus, a possible increase in CM size in PIDDosome deficient mice may be masked in heart cross-sections. In order to better correlate the ploidy with cell size, we propose to reanalyze our microscopy images used to extract the data displayed in Fig. 1D. We may run into the problem though that the number of cells acquired may become limiting to achieve sufficient statistical power. In this case we could pool data from different PIDDosome mutant CM to increase statistical power. Again, we propose to initially prioritize wt vs. Casp2 vs. XMLC2/Casp2f/f mice. In addition, we can offer to quantify heart to body weight ratio or tibia length as an additional read-out (see answer to a previous reviewer comment). *

The authors should discuss the fact that PIDDosome depletion lead only to a mild increase in ploidy levels (4N) in a small percentage of cardiomyocyte. If the PIDDosome is controlling ploidy, one could expect that removing it should lead to a drastic increase in the ploidy levels. Is PIDDosome depletion leading to cell death in some cardiomyocyte? The authors should discuss this point in the discussion or if relevant show a staining with an apoptosis marker. Is another mechanism compensating to prevent higher ploidy levels in cardiomyocytes?

These are valid thoughts, some of which we contemplated before. In part, we have addressed them in our response to Reviewer#1, above, discussing similar findings made in E2f7/8 deficient hearts (PMID: 36622904), or Cyclin G overexpressing hearts (PMID: 20360255), where also only modest changes in ploidy were achieved. Together these observations are suggesting alternative control mechanism able to act, or limited tolerance towards larger shifts in ploidy, incompatible with proper cell function and survival. Towards this end, we can offer to test if we find increased signs of cell death in PIDDosome mutant hearts by TUNEL staining of histological sections. Of note, we did not find evidence for such a phenomenon in the liver (PMID: 31983631).

Even if the authors presented RNAseq data suggesting that the PIDDosome is activated during cardiomyocyte differentiation, they should clearly demonstrate this point to strengthen the message of the paper. Indeed, the conclusions are based on the absence of PIDDosome components triggering higher ploidy in cardiomyocytes. However, we don't know whether (and when) the PIDDosome is activated during cardiomyocyte differentiation to control their ploidy levels. I would suggest to analyze PIDDosome activation markers by immunofluorescence in *cardiomyocytes at different developmental stages. *

*We agree with this referee that direct proof of PIDDosome activation would be helpful and that we only infer back from loss of function phenotypes when and where the PIDDosome becomes activated. However, several technical issues prevent us from collecting more direct evidence of PIDDosome activation in the developing heart. 1) Polyploidization in heart CM appears to happen gradually in CM from day 3 on with a peak at day 7 (PMID: 26544945). Hence, this is not a synchronous process, where we could pinpoint simultaneous activation of the PIDDosome in all cells at the same time, which would facilitate biochemical analysis, e.g., by western blotting for signs of Caspase-2 activation (i.e. the loss of its pro-form, PMID: 28130345). 2) Our most reliable readout, MDM2 cleavage by caspase-2 giving rise to specific fragments detectable in western, is not applicable to mouse tissue, as the antibody we use only detects human MDM2 (PMID: 28130345) and no other MDM2 Ab we tested gave satisfactory results. Independent of that, 3) we do not see involvement of p53 in CM ploidy control (arguing against a role of MDM2). *

*As such, we can only offer to look at extra centrosome clustering in postnatal binucleated CM (as also suggested further below), as a putative trigger for PIDDosome activation. However, this has been published by the first author of this study before (PMID 31301302). Given that we have made the significant effort to time resolve the increase in ploidy in postnatal mice (please note that several hearts needed to be pooled for each time point, analyzed in multiple biological replicates), we think that our conclusions are well-justified based on the genetic data provided. *

Concerning the methods, the authors must add the references for each product they used and not only the origin. When relevant, the RRID should be indicated. Without this information the method and the data cannot be reproduced.

We will update this information where relevant to reproduce our results

Minor comments:

In general, the text and the figures are clear. Nevertheless, I would suggest the following changes:

• Figures 1B, 2B and 2C: the y-axis must start at 0.

We will adopt axes accordingly

Figure 4A: The authors should stain centrosomes in cardiomyocytes. This should strengthen the conclusion taken by the authors based on the results obtained in mice depleted for ANKRD26. Indeed, for the moment they are insufficient to conclude about the role of the centrosomes. The authors should show that centrosomes cluster in cardiomyocytes (a condition necessary for PIDDosome activation in polyploid cells) and if possible that component of the PIDDosome are recruited here.

*This point is well taken and addressed in part above. Clustering of extra centrosomes has been documented and published by the first author of this study in rat polyploid cardiomyocytes (PIMID; cited). We can offer to show clustering of centrosomes in mouse CM isolated from day 7 hearts, but while PIDD1 can be detected well in MEF, we repeatedly failed to stain fro PIDD1 in primary CMs. *

Figure 4F: I would suggest to modify the working model to emphasize more the differences between WT and PIDDosome KO.

We will aim to improve this cartoon/graphical abstract

The prior studies are referenced appropriately.

Reviewer #2 (Significance (Required)):

How polyploid cells control their ploidy levels during differentiation remains poorly understood. The data presented here represent thus an advance concerning this question. The actual model concerning PIDDosome activation relies on the presence of extra centrosomes that drives the ANKDR26-dependent recruitment of the PIDDosome. Then, Caspase 2 is activated leading to a p53-p21 dependent cell cycle arrest (Burigotto & Fava, 2021, Sladky and Villunger, 2020; Janssens & Tinel, 2012; Evans et al., 2020; Burigotto et al., 2021). In this study, the authors showed that similar pathway takes place during cardiomyocyte differentiation to control ploidy levels. These data are reminiscent of previous work showing PIDDosome involvement during hepatocyte polyploidization (Sladky et al. 2020). Together, these data highlight the prominent role of the PIDDosome complex in controlling ploidy levels in physiological context. Importantly, this study identified that the classical p53-dependent cell cycle arrest described after PIDDosome activation is not involved here. Instead, the data established that independently of p53, p21 contribute to control cardiomyocyte ploidy. In consequence, this study extends the initial pathway associated with PIDDosome activation and suggest that other mechanisms could take place to restrain cell proliferation upon PIDDosome activation. Overall, this makes this paper significant and of interest for the following fields: polyploidy, heart/cardiomyocyte development and PIDDosome.

My field of expertise includes polyploidy, cell cycle and genetic instability.

We thank this reviewer for the time taken and the positive feedback provided.

• *

• *

3. Description of the revisions that have already been incorporated in the transferred manuscript

Please insert a point-by-point reply describing the revisions that were already carried out and included in the transferred manuscript. If no revisions have been carried out yet, please leave this section empty.

• *

N/A

4. Description of analyses that authors prefer not to carry out

Please include a point-by-point response explaining why some of the requested data or additional analyses might not be necessary or cannot be provided within the scope of a revision. This can be due to time or resource limitations or in case of disagreement about the necessity of such additional data given the scope of the study. Please leave empty if not applicable.

• *

• *As outlined above, limited tools are available to validate putative caspase-2 substrates, identified in proteomics analysis, in an impactful manner. *

• *Also, as discussed above, we deem myocardial infarction experiments in mice as unsuitable to improve our work, as with all likely-hood, they will yield negative results. *

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

Summary:

In this study, the authors investigated the role of the PIDDosome during cardiomyocyte polyploidization. PIDDosome is a multi-protein complex activating the endopeptidase Caspase-2, and shown to be involved in eliminating cells with extra centrosomes or in response to genotoxic stress (Burigotto & Fava, 2021, Sladky and Villunger, 2020). In both cases, the PIDDosome is recruited in a ANKRD26-dependent manner at the centrosomes leading to p53 stabilization and cell death (Burigotto & Fava, 2021; Evans et al., 2020; Burigotto et al., 2021).

Here, by studying mouse cardiomyocyte differentiation, the authors showed that PIDDosome is imposing ploidy restriction during cardiomyocyte differentiation. Importantly, in contrast to a previous report in the liver (Sladky et al., 2020), they showed that PIDDosome acts in a p53-independent manner in cardiomyocytes. Indeed, they suggested that PIDDosome controls ploidy in cardiomyocytes through p21 activation.

Major comments:

In general the conclusions of the authors are well supported by the experiments. However, I would suggest the following experiments/analysis to strengthen the paper:

• The authors should improve the Figure 1 to help the readers who are not familiar with cardiomyocyte polyploidization. For instance, I would suggest to add a scheme to summarize cardiomyocyte polyploidization (in terms of nuclear size, mono vs multi and so on). Based on the images they presented in 1B, the authors should also measure the nuclear area or volume in the different conditions in which components of the PIDDosome were depleted. Indeed, these two parameters should be easier to conceptualize for the readers (instead of the fluorescence nuclear intensity). This could help to understand if the nuclear size is maintained between the different conditions and if this is comparable between mono, bi or multinucleated cardiomyocytes.
• In Figure 2A, the authors presented cross section of heart from animals showing that PIDDosome depletion has no effect on heart size. This is a surprising results since cardiomyocytes have higher ploidy levels and this could have an effect on their function. Since the importance of this observation, the authors should present a quantification of the heart size in the different conditions shown in Figure 2A. Also, the percentage of cardiomyocytes presenting higher levels of ploidy seems quiet low. The authors should discuss this point. In particular because this could explain the absence of consequences on heart size and function at steady state.
• In Figure 2D, the authors measured the cardiomyocyte cross-sectional area and concluded that removing PIDDosome components have no effect on cardiomyocyte cell size. Since it has been shown that ploidy increase is normally associated with an increase in cell area, the authors should measure cell area of cardiomyocytes analyzed in Figure 1B. It could be then interesting to establish a correlation with nuclear area and the mono, bi or multinucleated status. This will strengthen the results showing that ploidy increases without affecting cell area.
• The authors should discuss the fact that PIDDosome depletion lead only to a mild increase in ploidy levels (4N) in a small percentage of cardiomyocyte. If the PIDDosome is controlling ploidy, one could expect that removing it should lead to a drastic increase in the ploidy levels. Is PIDDosome depletion leading to cell death in some cardiomyocyte? The authors should discuss this point in the discussion or if relevant show a staining with an apoptosis marker. Is another mechanism compensating to prevent higher ploidy levels in cardiomyocytes?
• Even if the authors presented RNAseq data suggesting that the PIDDosome is activated during cardiomyocyte differentiation, they should clearly demonstrate this point to strengthen the message of the paper. Indeed, the conclusions are based on the absence of PIDDosome components triggering higher ploidy in cardiomyocytes. However, we don't know whether (and when) the PIDDosome is activated during cardiomyocyte differentiation to control their ploidy levels. I would suggest to analyze PIDDosome activation markers by immunofluorescence in cardiomyocytes at different developmental stages.

Concerning the methods, the authors must add the references for each product they used and not only the origin. When relevant, the RRID should be indicated. Without this information the method and the data cannot be reproduced.

The statistics are well indicated in the figures and in the figure legends.

Minor comments:

In general, the text and the figures are clear. Nevertheless, I would suggest the following changes:

• Figures 1B, 2B and 2C: the y-axis must start at 0.
• Figure 4A: The authors should stain centrosomes in cardiomyocytes. This should strengthen the conclusion taken by the authors based on the results obtained in mice depleted for ANKRD26. Indeed, for the moment they are insufficient to conclude about the role of the centrosomes. The authors should show that centrosomes cluster in cardiomyocytes (a condition necessary for PIDDosome activation in polyploid cells) and if possible that component of the PIDDosome are recruited here.
• Figure 4F: I would suggest to modify the working model to emphasize more the differences between WT and PIDDosome KO.

The prior studies are referenced appropriately.

Significance

How polyploid cells control their ploidy levels during differentiation remains poorly understood. The data presented here represent thus an advance concerning this question.

The actual model concerning PIDDosome activation relies on the presence of extra centrosomes that drives the ANKDR26-dependent recruitment of the PIDDosome. Then, Caspase 2 is activated leading to a p53-p21 dependent cell cycle arrest (Burigotto & Fava, 2021, Sladky and Villunger, 2020; Janssens & Tinel, 2012; Evans et al., 2020; Burigotto et al., 2021). In this study, the authors showed that similar pathway takes place during cardiomyocyte differentiation to control ploidy levels. These data are reminiscent of previous work showing PIDDosome involvement during hepatocyte polyploidization (Sladky et al. 2020). Together, these data highlight the prominent role of the PIDDosome complex in controlling ploidy levels in physiological context.

Importantly, this study identified that the classical p53-dependent cell cycle arrest described after PIDDosome activation is not involved here. Instead, the data established that independently of p53, p21 contribute to control cardiomyocyte ploidy. In consequence, this study extend the initial pathway associated with PIDDosome activation and suggest that other mechanisms could take place to restrain cell proliferation upon PIDDosome activation.

Overall, this makes this paper significant and of interest for the following fields: polyploidy, heart/cardiomyocyte development and PIDDosome.

My field of expertise includes polyploidy, cell cycle and genetic instability.

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

Summary:

This manuscript by Leone et al describes the role of the PIDDosome in cardiomyocytes. Using a series of whole body and cardiomyocyte specific knockouts, the authors show that the PIDDosome maintains correct ploidy in these cells. It achieves this through inducing cell cycle arrest in cardiomyocytes in a p53 dependent manner. Despite this effect on ploidy, PIDDosome-deficient hearts show no structural or functional defects. Statistics and rigor appears to be adequate.

Major comments:

1. Figure 1 uses fluorescent intensity of a nuclear stain to determine ploidy per nucleus and they further separate the results into mononucleated, binucleated or multinucleated cells. It is hard to know how to interpret these results without further information or controls. Is there a good positive control that can be used to help to show whether this assay is quantitative. The differences are larger with the Raidd and caspase-2 knockouts than with the Pidd knockouts but this is not addressed.
2. On line 459 the authors state that the increase in polyploidy in PIDDosome knockouts occurs in adult hood but this is not directly tested. In fact in the next section the polyploidy is assessed in early postnatal development. This statement should be explained or removed
3. In Figure 4. The authors obtained RNAseq data for P1, P7 and P14 but only show the differences with and without caspase-2 at P7. Given that the differences in ploidy are more significant at P14 (Fig 3D), all the comparisons should be shown along with analysis of whether the same genes/gene families are altered in the absence of caspase-2
4. Some validation of the RNAseq and/or proteomics results would be an important addition to this study
5. Figure 4D: the authors make the conclusion that p21 is downstream of PIDD (et p53 independent). However, this is not supported by the data because the increase in 4N cells/decrease in 2N cells, although statistically significant, is nowhere near that of caspase-2 KO and caspase-2/p21 KO. Statistics should also compare p32KO with c2KO. In the absence of any other data, the more likely conclusion is that p21 is not involved.

Minor comments:

Suggest moving Figure 4A to Figure 3 as it seems to fit better there based on the citation of this figure in the text Recommend enhancing the brightness of microscopy images in Figure 1E and 2D

Significance

This study provides interesting information for the role of the PIDDosome in protecting from polyploidy and adds to the body of work by this same group studying this pathway in the liver. The main weakness in terms of significance is the lack of a phenotype in the hearts of these animals. Therefore, it is clear that ploidy (or at least PIDDosome dependent ploidy) has minimal impact on cardiac development. The authors mention that they have not tried giving these mice an MI or inducing any other type of cardiac damage. Although it is understood that these experiments are likely outside of the scope of the present study, without this information the impact of this study is moderate. I recommend expanding the discussion to provide a more indepth possible rationale as to why ploidy perturbations do not lead to structural changes like in the liver. Despite this, the insights to the pathway itself are interesting to investigators in the caspase-2 field if a little underdeveloped, especially concerning the role of p21.

My expertise is in cell death and caspase biology (especially caspase-2). I havesufficient expertise to evaluate all parts of this paper.

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

General Statements [optional]

There were several points that were raised by multiple reviewers, which we respond to as follows.

1. The reviewers pointed to a lack of clear comparison with experimental data. Perhaps this was insufficiently clear in the first submission, but the analysis of ECT-2 localization during cytokinesis was intended as a validation of the model, parameterized based on polarization and applied without further modification to cytokinesis. These situations differ in numerous respects: centrosome number, centrosome size, and we used several experimental conditions to control centrosome positioning. To address this more extensively, in the revised submission we analyzed our data further (Longhini and Glotzer, 2022) to extract profiles of ECT-2 and myosin. We used these profiles both to constrain model parameters (Appendix B.3) and to compare with model predictions for both polarization and cytokinesis (Figs. 3 and 5).
2. All of the reviewers pointed to our assumption that myosin indirectly recruits ECT-2. We apologize for a lack of clarity in the original draft about this. We had intended to convey the hypothesis that ECT-2 is recruited by a species that is advected with myosin, but for the sake of the minimal model we do not introduce any extra equations for this species and instead assume it colocalizes with myosin. In the revised manuscript, we address this by clearly listing the assumption (#2 on p. 7), and by comparing to an alternative model (Eq. (S4) and Fig. S7) that accounts directly for a third advected species. We also document specifically (second panel from left in Fig. 4) why the short residence time of ECT-2 makes patterning by pure advection impossible. That said, we still do not know the identity of this factor.
3. The reviewers pointed out that our use of the M4 term to limit contractility was dubious. This was a (probably misguided) attempt to use previously-published models to constrain our model. In the revised submission, we replaced this term with a more general nonlinear term Mk, where we first demonstrate that k = 1 is insufficient to match the data (p. 32), then consider k = 2,3. We present results in the main text for k = 2, while Fig. S5 shows that the corresponding results for k = 3 are not very different. Put another way, we empirically demonstrate that the specific form of this nonlinear term is not important, as long as it prevents contractile instabilities (as pointed out by one of the reviewers).
4. Apparently, our extension of the model to cytokinesis, and the evidence for validation of the model, was not clear in the original draft. Because of this, we reformulated the section (3.4) and figure (5) on cytokinesis. We identified four representative examples of centrosome positions, then compared the experimental profile of ECT-2 accumulation to the model result. For simplicity, we also eliminated the simulations of the non-phosphorylatable inactive copy of ECT-2 (“ECT-2 6A”). A more detailed analysis of that data revealed that the pattern of accumulation of ECT-2 6A at cleavage furrowing was more similar to the end of polarization, indicating that this copy of ECT-2 appears to have much slower turnover than the endogenous copy (as would expected from phosphorylation-dependent membrane displacement).
5. Fundamentally, our study addresses a similar question to (Illukkumbura et al., 2023), in the sense that we seek to understand how cortical flows could pattern ECT-2 and myosin, even though the residence time of ECT-2 is very low. Despite the similarities, it differs from the cited study in that ECT-2 is not an inert component that is asymmetrically distributed, but rather a component which regulates myosin levels and cortical flows, ultimately feeding back on its own accumulation. Due to these similarities and differences, we added an expository section in the discussion (p. 18) comparing our results to those of that study.

Point-by-point description of the revisions

This section is mandatory. Please insert a point-by-point reply describing the revisions that were already carried out and included in the transferred manuscript.

Reviewer 1

In this article, Maxian et al. propose a model combining 1-d simulations of ECT-2 and Myosin concentration at the cortex through binding/unbinding and advection at the cortex, with an input for AIR-1 cortical concentration based on the spatial localisation of the centrosomes in the cytoplasm. The objective of the authors is to recapitulate the role of (1) AIR-1, (2) its effector ECT-2 and (3) the downstream effector, driver of cortical flows, the molecular motors Myosin, in two key physiological processes, polarization and cell division. This is important as work over the last 10 years have emphasized the role of AIR-1 in embryo polarization. Previous biochemical-mechanical models have focused on RhoA/Myosin interactions (Nishikawa et al, 2017), the importance of a negative feedback and excitable RhoA dynamics (Michaux et al, 2018), or anterior PARs/posterior PARs/Myosin (Gross et al, 2019). The authors thus attempt to provide a new descriptive model in which RhoA is implicit, instead focusing on the role of centrosome localization on AIR-1 localization, and providing a framework to explore polarity establishment and cell division based on these 3 simple players. The first part of the model is very reminiscent of previously published models, while the second instead provides a link between the initial polarizing cue AIR-1 and polarization. Based on this description, the model is precisely tuned to achieve polarization while matching experimental observations of flow speed and ECT-2 A/P enrichment shape. The results are therefore certainly new and interesting.

Thank you for the positive assessment!

Major comments:

1. The authors use the position of the centrosomes as a static entry, resulting in a static AIR1 input. Is this true, or are the positions of the centrosomes dynamically modulated over the course of the different processes simulated here (for example as a consequence of cortical flows?), and if so, is the assumption of immobile position?

We assume that the centrosomes are fixed on the timescale of the cortical dynamics, and study how the cortex responds to a static AIR-1 signal (see clarifying comment on p. 4). In Fig. S4, we show that the cortex responds rapidly to changes in the existence or position of the AIR-1 signal. As such, slower dynamics might be the result of slowly moving centrosomes, as we show in supplementary simulations (Fig. S8).

1. While in its principle the model is quite simple and elegant, the detailed form of the equations describing the interactions between the players is more complex. Are all these required? If they are crucially important for the behavior of the model, these should be described more thoroughly, and if possible rooted more directly in experimental results:

Thank you for this comment. We agree that there were several non-trivial terms in our “minimal” model. Our guiding principle for the revision was to reduce complexity and better justify the terms that are included.

(a) kMEMEc _(Linear enhancement term): why would myosin impact E concentration? The authors state, p.7, ”There is a modest increase in the recruitment rate of ECT-2 due to cortical myosin (directly or indirectly), in a myosin concentration-dependent manner (Longhini and Glotzer, 2022).” I could not find the data supporting this assumption Longhini and Glotzer apparently rather point to a modulation of cortical flows. (”During anaphase, asymmetric ECT-2 accumulation is also myosin-dependent, presumably due to its role in generating cortical flows.”). Embedding this effect in the recruitment rate instead of expecting it from the model thus appears awkward. Could the authors specify how they came to this conclusion, which the authors might have derived from observations made in their previous work, but maybe did not fully document there?

This is an important issue. Since it was raised by all of the reviewers, we addressed it in our general comments. Throughout the manuscript (Figs. 4 and S4), we tried to highlight that cortical flows are insufficient to localize ECT-2, while the recruitment hypothesis provides a better match to the experimental data. The recruitment by an advected species was speculated upon in Longhini and Glotzer: ”Rather, we favor a model in which the association of ECT-2 with the cortex involves interactions with cortical component(s) that are concentrated by cortical flows.”

(b) kEME2Mc (ECT-2 non-linear impact on Myosin): does the specific form of the value to convey the enhancement (square form) have an impact on the results?

The specific form does not have an impact. In fact, in the revised version, our experimental data shows an asymmetry in myosin that is actually lower than ECT-2. As such, a nonlinear term here lacks justification, and we switched to a linear term of the form kEMEMc (see model equations on p. 6).

(c) KfbM4 ”The form of this term is a coarse-grained version of previously-published work (Michaux et al., 2018).” Myosin feedback on myosin localization proportionally to_ M4 _does not seem to directly derive from Michaux et al. Please detail this points more extensively and detail the derivation, in the supplements if not in the main text.

Based on this comment and that of reviewer 2, we decided to switch to a more general term for nonlinear negative feedback, as discussed in point 3 in general comments.

(d) P23. Parameter values: ”This is 1.5 times longer than the estimate for single molecules (Nishikawa et al., 2017; Gross et al., 2019) to reflect the more long-lived nature of myosin foci during establishment phase (Munro et al., 2004).” Not sure what the authors mean by more long-lived duration of foci during establishment phase. Seems rather arbitrary.

This was a misstatement on our part. A closer look at Gross et al. revealed that, under conditions similar to those we simulate (initial polarity establishment), the residence time of myosin is about 15 s (off rate 0.06 s−1). We modified our justification (p. 30) to include this. We also looked at the effect of longer myosin residence time on polarity establishment (Fig. S8).

1. It would be very helpful (and indeed more convincing) to include a direct comparison between modeling results and experimental counterpart whenever possible. This might not be possible for some data (e.g. Fig. 3d from Cowan et al), but should be possible for other, in particular Fig. 3c and Fig. 5b, for the flow speed and ECT-2 profiles. In Fig. 5b in particular, previously published experimental data could be produced to give the reader to compare model with experiments (possibly provided as an inset, at least for the wild type conditions).

We tried to bring in more data based on what was available from previous work (Longhini and Glotzer, 2022). Frame intervals of 10 s prohibited a PIV analysis for flow speeds, and punctate myosin profiles often made it difficult to measure myosin concentration. We were, however, able to extract the ECT-2 concentration from our previous movies and compare it to the model results. We included these comparisons in Figs. 3 and 5, with accompanying discussion in the text.

Minor comments: 1. Fig. 5b: ECT-2 C 6A(dhc-1) do not seem to be referenced or discussed in the main text.

Also, why present the results for the flow for 2 conditions and the ECT-2 localisation for 4? Or does the variation of ECT-2 not impact the flow profile?

As discussed in general comments, we decided to reformulate the cytokinesis figure to incorporate more experimental data. Since we have detailed data on ECT-2 localization, we presented these in Fig. 5 for four experimental conditions, comparing each to the model.

1. p.6: Given that the non-normalized data is used in the main text, and the normalized only appears in the supplemental, maybe star the dimensionless and remove all hats from the main for greater legibility?

We changed the notation to make the main text variables (dimensional) unadorned, while the dimensionless variables in the SI now have hats.

1. p.6: Eqn 1a: carrot missing on 3rd E?

This is now a moot point because of the previous comment.

1. p.14: replace_“embryo treatment” with ”experimental conditions”?

We changed “embryo treatment” to “experimental conditions” globally.

1. p.21, S4a: add_ A = A/A(Tot)

We added it in the last display on p. 28.

1. p.22: ”L = 134.6_ µ_m” - please write 134_ µ_m to retain the precision of original measurements

We made this change.

1. p.22: Please provide formula for all dimensionless values as a table at the end of the supplemental for the eager but less-mathematically proficient reader.

We added Table 1 to list the relationship between dimensional and dimensionless parameters.

Reviewer 2

The manuscript by Maxian, Longhini and Glotzer presents purely modeling work performed by the first author in conjunction with the already published experimental work by Longhini and Glotzer (eLife, 2022). The aim of the manuscript is to provide a mathematical model that connects the actomyosin contractility of the cell cortex in C. elegans zygote with the activity of the centrosomal kinase AurA (AIR-1 in C. elegans). The major claim of the authors is that their model, fitted to the experimental data pertaining to the zygote polarization, also describes dynamics during the zygote cytokinesis. In the model, the authors provide a heuristic approach to the biochemical dynamics, reducing their treatment to two variables: myosin and Ect2 Rho GEF. The biochemical model is integrated with a simple 1D active gel-type model for the cortical flow. The model uses static diffusive field of activity of AurA kinase in the cytoplasm as an input to their chemo-mechanical model.

Major concerns:

1. The biochemical model is highly heuristic and several major assumptions are poorly justified. Thus, the authors explicitly introduce recruitment of Ect2 by myosin, something apparently based on the experimental observations by Longhini and Glotzer in 2022, which had not been biochemically confirmed since with a clear molecular mechanism.

This is an important issue, and we appreciate your concern which was shared by the other reviewers. As discussed above on p. 1, we tried to justify this assumption better by (a) clearly stating it on p. 7, and (b) demonstrating that the dynamics we observe in live embryos are impossible without it. The model confirms what was pointed out by Longhini and Glotzer, that the short residence time of ECT-2, combined with in vivo flow speeds on the order of 10 µm/min, make it impossible for cortical flows alone to redistribute ECT-2.

1. The contribution of AurA is introduced highly schematically as a term based on enzyme inhibition biochemistry that increases the off rate of Ect2. The major assumption of the model is that AurA phosphorylates Ect2 strictly on the membrane (cortex) of the cell. Why? No molecular justification is given. If the authors cannot provide clear justification, this major assumption has to be clearly declared as such. The phosphorylation/dephosphorylation dynamics of Ect2 is not considered at all.

We clarified that the species we consider in the model (E) is unphosphorylated ECT-2, so that the negative flux comes from either unbinding or phosphorylation. Of course, AIR-1 phosphorylates ECT-2 in the cytoplasm as well, but our model only tracks the binding of unphosphorylated ECT-2 to the cortex. We clarified this on p. 6.

1. In the equation for myosin, the authors introduce disassembly/ inactivation term proportional to the fourth order of concentration of myosin. Why? This is a major assumption, which appears to be derived from the work by Michaux et al. 2018. There the authors (Michaux et al.) postulated that the rate of inactivation of RhoA GTPase was somehow proportional to the fourth power of RhoA concentration. It appears that Maxian et al. further assume that the myosin concentration is fast variable enslaved by Rho, so that_ M ∼ _[RhoA]. They then presumably assume that if the rate of degradation/ inactivation of Rho is proportional to the fourth power of Rho concentration, so is true for myosin (M). This is a logical error and is not justified. An important question, why do the current authors need this unusual assumption with such a high power of M disassembly/inactivation? Perhaps, this is because without this rather dubious term the cortex flow produces a blow-up of myosin concentration? This would be expected in their mechanical model - the continuous flow of actomyosin not compensated by cortex disassembly generally causes blow-up of biochemical concentrations transported by the flow, this is a known problem of the “simple” active gel model used by the authors. Maxian et al. have to provide clear derivation of the term −KfbM4 _and also demonstrate why they need this exotic assumption.

As mentioned above in general comments, this was a misguided attempt on our part to use previous literature to directly assign values to model parameters. In the revised manuscript, we considered a more general term for the nonlinear feedback. The fitting occurs in Fig. S3, where we impose the ECT-2 profile during pseudo-cleavage and try to fit the myosin profile. k = 1 is eliminated because the ECT-2 and myosin have different asymmetries. Higher order nonlinearities (k = 2,3) are successful in fitting the experimental data. In the main text, we present results from k = 2, then use Fig. S5 to present results on the k = 3 case.

1. The equation for myosin M has a membrane-binding term, which is second order in concentration of Ect2~E2, without which the model will not show the instability that the authors need. The only justification given is that ”some nonlinearity is required”. A proper derivation should be given here.

Our experimental data shows an asymmetry in myosin that is actually lower than ECT-2. As such, a nonlinear term in the binding rate lacks justification, and we switched to a linear term of the form kEMEMc (see model equations on p. 6).

1. The diffusion coefficients for Ect2 and myosin are chosen to be the same. Why? Clearly these molecules so different in size - myosin being a gigantic cluster monster of size_ 300nm _believed to be bound to actin, should have a much smaller diffusion coefficient?

Thank you for raising this point. We used the same diffusion coefficient for simplicity; because its dimensionless value is less than 10−4, diffusion is relatively unimportant in shaping the concentration fields. If we assume instead, for instance, that myosin cannot diffuse in the membrane, while ECT-2 has a ten-fold larger diffusion coefficient, the steady state profiles of ECT-2 and myosin are changed by at most 5% (see Fig. S6).

1. There are confusing statements regarding the role of actomyosin flows. In the beginning of the manuscript, the authors seem to state that since Ect2 has a high off rate, the effect of the flow on Ect2 localization is negligible in comparison with direct binding to myosin. Later, the authors state that flows are absolutely essential for the patterning. The authors need to clearly explain where and how the flows are important or not.

Thank you for pointing out this confusion. In the revised manuscript, we tried to be explicit that the combination of recruitment and flows is essential for patterning ECT-2. We did this in Figs. 4 and 5 by showing the results of simulations without recruitment (Fig. 4) and without recruitment and flows (Fig. 5).

Minor points:

1. page 9. Why is the rate of dephosphorylation of AurA is named Koff?

We changed the notation to kinac to reflect inactivation.

1. page 10. “Note that the model is calibrated to predict... which matches experimental observations” - this sentence needs changing. You want to say that you fit the model to experiments in the Longhini and Glotzer paper. There is no prediction here.

We removed this sentence.

1. page 14. “A plot of Ect-2 accumulation as a function of distance from the nearest cortex...” - clearly the word ”centrosome” is meant here instead of ”cortex”.

What was meant by this sentence was the distance from the centrosome to the nearest cortex pole (anterior or posterior). We modified it to make this more clear (p. 15).

1. page 16. ”Inactive, non-phosphorylatable version of Ect-2...” - non-phosphorylatable is clear, but why inactive?

As discussed in general comments we decided to simplify the cytokinesis figure and remove the simulations with non-phosphorylatable ECT-2. While it is not relevant, the ECT-2 6A variant represents a fragment of the protein that lacks the catalytic domain. Our original goal was to use these data to track the ECT-2 localization without perturbing the system biochemistry, but the data gave the hint of longer exchange kinetics, which confounded our analysis.

Reviewer 3

_Maxian et al. developed a mathematical model to explain the essential elements and interactions necessary and sufficient for the polarisation of the C. elegans zygote. The initiation of zygote polarisation has been extensively studied in recent years, highlighting the role of the centrosomal kinase Aurora-A (AIR-1) in controlling the cortical distribution of RhoGEF (ECT-2) and actomyosin contractility during polarisation. Although genetic experiments have demonstrated their function in this process, it remains to be tested whether these factors and their interactions are sufficient to induce polarisation.

This work has provided a theoretical framework to predict the activity of AIR-1 in the cytoplasm and at the cell cortex, and the cortical distribution of ECT-2 and myosin-II (NMY-2). This framework can recapitulate the dynamic rearrangement of ECT-2 and myosin-II during polarisation, with centrosomes positioned at the posterior pole of the zygote. This model can explain, at least in part, the asymmetric distribution of ECT-2 and myosin-II in the zygote undergoing cytokinesis, suggesting that the mechanism of AIR-1-mediated control of ECT-2 and myosin-II would regulate patterning during polarisation and cytokinesis. This theoretical framework is developed with reasonable assumptions based on previous genetic experiments (except for the myosin-dependent regulation of ECT-2; see comments below).

Thank you for the positive assessment!

Major issues:

1. The authors insist that this model correctly predicts the spatio-temporal dynamics of ECT-2 and myosin-II during polarisation and cytokinesis. However, the predicted results do not reproduce the in vivo pattern of ECT-2 in both phases. ECT-2 is cleared from the posterior cortex and establishes a graded pattern across the antero-posterior axis during polarisation (see their previous publication in eLife 2022, 11, e83992, Fig1A -480s) and cytokinesis (see eLife 2022, 11, e83992, Fig1C 60s and 120s). During both stages, ECT-2 does not show local enrichment at the boundary between the anterior and posterior cortical domains in vivo. In fact, when comparing the predicted results with the in vivo pattern of ECT-2 and cortical flow, the authors used non-quantitative descriptions such as ’in good agreement’, ’a realistic magnitude’,, ’resemble’. These vague descriptions should be revised and a quantitative assessment of ECT-2 distribution between in silico and in vivo should be included in a revised manuscript.

As mentioned on p. 1, in the revised manuscript we interacted with the data in a much stronger way. We first used data during pseudo-cleavage to infer the ECT-2/myosin relationship. We then examined (Fig. 3) quantitatively how the ECT-2 accumulation during polarization matches the experimental data (it matches early but not later stages). We repeated this for cytokinesis in Fig. 5, where we compared the ECT-2 profile across four experimental conditions to the model prediction.

1. I assume that the strange local enrichment of ECT-2 at the anteroposterior boundary is due to their assumption that the binding rate of ECT-2 is increased by a linear increase via cortical myosin-II (page 6). This assumption is not directly supported by experimental evidence. A previous study by the same group (eLife 2022, 11, e83992) showed that a progressive increase in ECT-2 concentration at the anterior cortex is partially accompanied by an increase in cortical flow and transport of myosin-II from the posterior pole to the anterior cortex. This observation supports the idea that ECT-2 may associate with cortical components transported by myosin-II based cortical flow. This unrealistic assumption makes the predicted distribution pattern of ECT-2 almost identical to that of cortical myosin-II, resulting in an increase in the concentration of ECT-2 at the anteroposterior boundary where myosin-II forms pseudocleavages and cleavage furrows. The authors should clarify why their mathematical model used this assumption and provide a comprehensive analysis and evaluation of the parameter value for an ECT-2-myosin-II interaction.

In the revised manuscript, we outlined the justification for this assumption after presenting the model equations. In the Appendix, we were able to constrain all parameters except the recruitment term. Then, we provided an analysis of how polarization changes when the recruitment term is increased. We show that the ECT-2 asymmetries with myosin flows are the same as those simply due to AIR-1 inhibition (since the lifetime of ECT2 is small). Adding indirect recruitment gives asymmetries that resemble experimental data from early establishment of polarity. We showed this both by assuming “myosin” (a species which colocalizes with myosin) recruits ECT-2 (Fig. S4) and by simulating an alternative model (Eq. (S4)) where an explicit species that is advected with cortical flows recruits myosin (Fig. S7).

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 (Evidence, reproducibility and clarity (Required)):

Summary: In this paper, the authors perform a screen by feeding C. elegans different E. coli genetic mutants and examining the effect on the expression of fat-7, a stearoyl-CoA 9-desturase, which has been associated with longevity. They identify 26 E. coli strains that decrease fat-7 expression, all of which slow development and increase lifespan. RNA sequencing of worms treated with 4 of these strains identified genes involved in defense against oxidative stress among those genes that are commonly upregulated. Feeding C. elegans these 4 bacterial strains results in increased ROS and activation of the mitochondrial unfolded protein response, which appears to contribute to lifespan extension as these bacterial strains do not increase lifespan when the mitochondrial unfolded protein response transcription factor ATFS-1 is disrupted. Finally, the authors demonstrate a role for iron levels in mediating these phenotypes: iron supplementation inhibits the phenotypes caused by the identified bacterial strains, while iron chelation mimics these phenotypes. Response: We thank the reviewer for an excellent summary of our work.

Major comments: The proposed model involves an increase in ROS levels activating the UPRmt and then leading to lifespan extension. If the elevation is ROS levels is contributing then treatment with antioxidants should prevent UPRmt activation and lifespan extension. Response: This is an excellent point. We will treat the FAT-7-suppressing diets with antioxidants and observe the effect on C. elegans UPRmt activation and lifespan.

The authors suggest that iron depletion may disrupt iron-sulfur cluster proteins. The Rieske iron-sulfur protein ISP-1 from mitochondrial electron transport chain complex III has previously been associated with lifespan. Point mutations affecting the function of ISP-1 or RNAi decreasing the levels of ISP-1 both result in increased lifespan (PMID 20346072, 11709184). Thus, iron depletion may be increasing ROS, activating UPRmt and increasing lifespan through decreasing ISP-1 levels.

Response: The reviewer has raised an intriguing possibility that the increased lifespan on the FAT-7-suppressing diets could be because of perturbation of ISP-1 function. While ISP-1 levels may not be directly affected by the mutant diets, ISP-1 function might be perturbed on these diets as ISP-1 function requires iron-sulfur clusters. Therefore, we will study the lifespan of isp-1(qm150) mutant on the FAT-7-suppressing diets to explore whether the lifespan extension on these diets is ISP-1 dependent.

All of the Kaplan-meier survival plots are missing statistical analyses. Please add p-values.

Response: The p-values for all the survival plots are included in the respective figure legends.

It would be helpful to include a model diagram of the proposed mechanisms in the main figures.

Response: We will make a model diagram after completing the experiments suggested by the reviewers.

Minor comments: Rather than "mutant diets" it would be more informative to call these "FAT-7-decreasing diets"

Response: We have changed “mutant diets” to “FAT-7-suppressing diets” throughout the manuscript.

Is it surprising that none of the bacterial strains increased FAT-7 levels? Why do you think this is?

Response: Yes, it was indeed surprising to find only bacterial strains that reduced FAT-7 levels and none that increased them. One possible explanation is that these bacterial mutants may not directly regulate fat-7 expression. Instead, they might alter the overall dietary composition, which is known to influence fat-7 levels. It appears that none of the tested mutants modified the diet in a manner that would lead to fat-7 upregulation.

Page 5. "We hypothesized that diets reducing FAT-7 might elevate oleic acid levels". Since FAT-7 converts stearic acid to oleic acid, wouldn't deceasing FAT-7 levels decrease oleic acid levels and increase stearic acid levels?

Response: FAT-7 expression is regulated by a feedback mechanism and is sensitive to the fatty acid composition within host cells; elevated levels of unsaturated fatty acids, such as oleic acid, suppress FAT-7 expression. There are two possible ways bacterial mutants could lead to reduced FAT-7 levels: (1) by directly inhibiting FAT-7 expression, which would be expected to result in increased stearic acid levels; or (2) by supplying higher amounts of oleic acid through their composition, thereby suppressing FAT-7 expression via feedback regulation. We focused on the second possibility, as elevated oleic acid levels—like those seen with FAT-7-suppressing diets—are known to promote C. elegans lifespan. To avoid confusion, we have revised the statement to: “We hypothesized that bacterial diets might reduce FAT-7 expression because they have elevated levels of oleic acid”.

Page 6. The authors cite Bennett et al. 2014 for the statement that "Activation of the UPRmt has been associated with lifespan extension". This paper reaches the opposite conclusion "Activation of the mitochondrial unfolded protein response does not predict longevity in Caenorhabditis elegans". Also, in the Bennett paper and PMID 34585931, it is shown that constitutive activation of ATFS-1 decreases lifespan. Thus, the relationship between the UPRmt and lifespan is not straightforward. These points should be mentioned.

Response: The reviewer has raised an important point. We have now included a paragraph in the discussion to highlight these points. The revised manuscript reads: “All 26 FAT-7-suppressing diets identified in our study elevated hsp-6p::GFP expression and extended C. elegans lifespan. Although UPRmt activation and lifespan extension were consistently observed across these diets, there was no strong correlation between hsp-6p::GFP levels and the degree of lifespan extension. The role of the UPRmt in promoting longevity remains controversial (Bennett et al., 2014; Soo et al., 2021; Wu et al., 2018). For instance, gain-of-function mutations in atfs-1 have been shown to reduce lifespan (Bennett et al., 2014; Soo et al., 2021). However, a recent study demonstrated that mild UPRmt activation can extend lifespan, whereas strong activation has the opposite effect (Di Pede et al., 2025). These findings suggest that UPRmt contributes to longevity only under specific conditions and at specific activation levels. In our study, lifespan extension on FAT-7-suppressing diets was dependent on ATFS-1, indicating that UPRmt activation was necessary for this effect.”

Page 6. "Our transcriptomic analysis suggested elevated ROS". Rather than refer to gene expression, it would be better to refer to the ROS measurements that were performed.

Response: We have changed it to the following sentence: “Our ROS measurement analysis suggested elevated ROS levels in worms fed FAT-7-suppressing diets.”

The long-lived mitochondrial mutants isp-1 and nuo-6 have increased ROS, UPRmt activation and increased lifespan. Multiple studies have examined gene expression in these long-lived mutant strains. How does gene expression in these mutants compare to worms treated with the FAT-7-decreasing E. coli mutants? While not necessary for this publication, it would be interesting to see whether the FAT-7-decreasing E. coli strains can increase isp-1 and nuo-6 lifespan.

Response: We will compare the gene expression changes observed in isp-1 and nuo-6 mutants with the gene expression changes observed in worms exposed to FAT-7-suppressing diets. Additionally, we will examine the lifespan of isp-1 mutants on the mutant diets. These data will be included in the revised manuscript.

SEK-1 is also involved in the p38-mediated innate immune signaling pathway, which has been shown to contribute to longevity in C. elegans. In fact, disruption of sek-1 using RNAi decreased the lifespan of several long-lived mutant strains PMID 36514863.

Response: We thank the reviewer for highlighting this point. We have now added that the role of SEK-1 in regulating lifespan on FAT-7-suppressing diets could also be because of its role in innate immunity. The revised manuscript reads: “Notably, SEK-1 also regulates innate immunity and is essential for the extended lifespan observed in several long-lived C. elegans mutants (Soo et al., 2023). Therefore, its effect on lifespan in response to FAT-7-suppressing diets may also stem from its role in innate immune regulation.”

Figure 2. Why were cyoA and ycbk chosen to show the full Kaplan-meier survival plot?

Response: These were selected randomly to show the range of the lifespan phenotype observed.

Figure 2, panel D. A better title may be "Mean Survival (Percent increase from control)"

Response: We have made this change.

While not necessary for this paper, it would be interesting to determine whether the FAT-7-decreasing E. coli strains alter resistance to oxidative stress.

Response: We will study the survival of worms on these diets upon supplementation with paraquat.

Figure 4. It may be interesting to include a correlation plot comparing hsp-6::GFP fluorescence and lifespan. It looks like the magnitudes of increase for each phenotype are not correlated.

Response: We have added a new Figure (Figure S4) to show the correlation between hsp-6::GFP fluorescence levels and percent change in mean lifespan. Indeed, there is no correlation between these phenotypes.

Reviewer #1 (Significance (Required)):

Overall, this is an interesting paper and the experiments are rigorously performed. The bacterial screen was comprehensive and was followed up by careful mechanistic experiments. This paper will be of interest to researchers studying the biology of aging. A diagram of the working model of the underlying mechanisms would enhance the paper. Response: We thank the reviewer for highlighting the significance of the study. We will include a model in the revised manuscript.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

In this manuscript, Das et al. investigate how different bacterial mutants affect the lifespan of C. elegans. The authors screened a library of E. coli mutants using a fat-7 reporter and identified 26 strains that reduce fat-7 expression, cause developmental delay, induce the mitochondrial unfolded protein response (using hsp-6 reporter), and increase worm lifespan. Among these, they focused on four strains and demonstrated that the effects of these mutants on developmental delay, fat-7 expression, and hsp-6 induction could be suppressed by iron supplementation. Furthermore, they showed that iron depletion alone is sufficient to induce fat-7 expression in worms. The lifespan extension observed in worms fed these mutant bacterial strains depends on SKN-1, SEK-1, and HLH-30. Overall, this is a well-written manuscript that highlights the role of iron in regulating fat-7 expression. However, the findings from the initial screen do not significantly expand upon what is already known in the literature. Many of the identified hits overlap with those reported by Zhang et al. (2019), which also highlighted the role of iron in developmental delay and hsp-6 induction. While the lifespan data and the role of fat-7 are novel aspects of this study, the authors have not conducted detailed mechanistic investigations to address key questions, such as: 1) How does the deletion of these bacterial genes alter the metabolic state of the diet? 2) How do these metabolic changes influence fat-7 expression in worms? 3) How does the downregulation of fat-7 contribute to longevity? Addressing these points would strengthen the mechanistic insights of the study.

Response: We thank the reviewer for a thoughtful summary of our work and for the valuable feedback provided to improve the manuscript. We would like to emphasize that the screening conditions and objectives of our study were fundamentally different from those of Zhang et al. (2019). Furthermore, Zhang et al. (2019) did not investigate the effects of the bacterial mutants identified in their screens on C. elegans lifespan. Notably, the 26 bacterial mutants identified in our screen do not overlap with those reported in previous studies that examined bacterial strains promoting C. elegans longevity. As detailed below, we will address the points raised by the reviewer that will certainly strengthen the mechanistic insights of the study.

Here are my detailed comments: 1. Suppressing FAT-7 levels in C. elegans does not inherently increase lifespan. To directly attribute this effect to FAT-7, it would be important to attempt a rescue experiment to restore FAT-7 expression and assess whether the lifespan extension persists. Additionally, measuring oleic acid levels in these mutants would help determine whether a high-oleic-acid diet is suppressing FAT-7 expression. The role of oleic acid cannot be ruled out using fat-2 mutants (Fig. 3B), as fat-2 mutants accumulate oleic acid when fed WT bacteria, but this may not translate to endogenous oleic acid accumulation in conditions where FAT-7 is suppressed.

Response: We thank the reviewer for these useful suggestions. We will overexpress FAT-7 under a pan-tissue promoter (eft-3) and study lifespan on FAT-7-suppressing diets. Moreover, to explore whether oleic acid has any role in enhancing lifespan on FAT-7-suppressing diets, we will study the lifespan of worms on these diets upon supplementing with oleic acid along with wild-type bacterium control.

To understand the host-microbe interaction in this study, it is important to determine what specific changes in the bacteria contribute to the observed phenotypes in worms. Identifying these bacterial factors will provide a clearer picture of their role in influencing worms stress signaling and lifespan.

Response: The phenotypes observed in C. elegans across all the identified bacterial mutants are remarkably consistent, including increased UPRmt activation, reduced FAT-7 levels, delayed development, and extended lifespan. This consistency suggests that a common underlying factor is driving these effects. Although the bacterial mutants appear genetically diverse, gene expression data from C. elegans, along with comparisons to the findings of Zhang et al. (2019), indicate that elevated levels of reactive oxygen species (ROS) may represent this shared factor. These results suggest that bacterial ROS play a central role in mediating the host-microbe interactions underlying the observed phenotypes. To further support this hypothesis, we will directly measure ROS levels in the identified bacterial mutants. Additionally, we will test whether antioxidant treatment can suppress the C. elegans phenotypes, thereby establishing a causal role for bacterial ROS.

It is important to rule out any changes in food consumption in worms fed these bacterial mutants, as differences in feeding amount could attribute to the observed lifespan effects.

Response: We will carry out pharyngeal pumping rate measurements to study whether there is any difference in food consumption in worms fed these bacterial mutants.

In figure 5A to 5G, please include the same-day controls to help clarify how iron supplementation effects these phenotypes relative to the control. For example, in Fig. 5F, it appears that iron extends the lifespan of worms fed the control diet. It would be clearer if appropriate controls were included in all of these figures or summarized in a table to help understand the impact of iron.

Response: We will include these controls in the revised manuscript.

How does iron depletion affect the levels of fat-7, and how does this contribute to the activation of the longevity pathways discussed in the manuscript.

Response: This is an intriguing question. There are at least two possible explanations: (1) oxidative stress may directly downregulate fat-7 expression, and (2) iron depletion could reduce ferroptosis, which in turn may influence fatty acid metabolism. In the revised manuscript, we will include data on how oxidative stress affects FAT-7 expression.

Minor comments 1. Please include a detailed table of the lifespan data for all replicates as a supplementary table.

Response: We have included the details of survival curves for all the data in the new Table S2.

In the Methods section, specify at what stage the worms were exposed to iron and the iron chelator for the lifespan experiments.

Response: The L1-synchronized worms were exposed to iron and iron chelator plates and allowed to develop till the late L4 stage before being transferred to lifespan assay plates that also contained the respective supplements. This information is now included in the Methods section.

Please clarify whether equal optical density (O.D.) of cells was seeded for both the WT and mutant strains, and mention if the mutants exhibit any growth defects.

Response: We have examined the growth of the bacterial mutants and found that they do not exhibit growth defects. Therefore, for all the assays, NGM plates were seeded with saturated cultures of all the bacterial strains. We have now included the growth curves data in the manuscript (Figure S4).

Reviewer #2 (Significance (Required)):

Significance General Assessment: This study by Das et al. explores the impact of bacterial mutants on C. elegans lifespan. A key strength of the study is the identification of bacterial mutants that influence the expression of the gene encoding fatty acid desaturase (fat-7) and lifespan in C. elegans. Furthermore, the study highlights the role of iron in regulating fat-7 expression, suggesting that iron imbalance may play a crucial role in modulating fatty acid metabolism. However, the study's main limitation is that it does not significantly extend the current understanding of the microbial modulation of host metabolism and aging, as many of the identified bacterial hits overlap with those previously reported in Zhang et al. (2019). The manuscript would benefit from more in-depth mechanistic exploration, especially with regard to how specific bacterial factors influence the metabolic state of the worms and how these changes ultimately modulate fat-7 expression and longevity.

Response: We thank the reviewer for highlighting the significance of our study. Once again, we would like to emphasize that the screening conditions and objectives of our study differed fundamentally from those of Zhang et al. (2019). Furthermore, Zhang et al. did not investigate the impact of the bacterial mutants identified in their screen on C. elegans lifespan. As outlined above, we will address the reviewer’s comments, which will undoubtedly strengthen the mechanistic insights of our study.

Advance: This study presents a conceptual advance by exploring the iron-dependent regulation of fat-7 expression and lifespan in C. elegans, linking bacterial mutations with key longevity pathways (SKN-1, SEK-1, and HLH-30). The novelty lies in the direct investigation of the bacterial-induced changes in fat-7 expression, though the role of iron in these mutants for development and induction of mito-UPR was previously shown in the literature. This study also adds to the growing body of work on C. elegans as a model for studying aging and host-microbe interactions, particularly in understanding how diet and microbial exposure affect metabolic processes and lifespan.

Response: We thank the reviewer for highlighting the advancement made by our study.

Audience: This research will primarily interest specialized audiences in aging research, microbiology, and metabolism, especially those focused on host-microbe interactions. Keywords of my expertise: Host-microbe interactions, metabolism, system biology, C. elegans, aging.

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

In this manuscript, Das et al. investigate how different bacterial mutants affect the lifespan of C. elegans. The authors screened a library of E. coli mutants using a fat-7 reporter and identified 26 strains that reduce fat-7 expression, cause developmental delay, induce the mitochondrial unfolded protein response (using hsp-6 reporter), and increase worm lifespan. Among these, they focused on four strains and demonstrated that the effects of these mutants on developmental delay, fat-7 expression, and hsp-6 induction could be suppressed by iron supplementation. Furthermore, they showed that iron depletion alone is sufficient to induce fat-7 expression in worms. The lifespan extension observed in worms fed these mutant bacterial strains depends on SKN-1, SEK-1, and HLH-30.

Overall, this is a well-written manuscript that highlights the role of iron in regulating fat-7 expression. However, the findings from the initial screen do not significantly expand upon what is already known in the literature. Many of the identified hits overlap with those reported by Zhang et al. (2019), which also highlighted the role of iron in developmental delay and hsp-6 induction. While the lifespan data and the role of fat-7 are novel aspects of this study, the authors have not conducted detailed mechanistic investigations to address key questions, such as: 1) How does the deletion of these bacterial genes alter the metabolic state of the diet? 2) How do these metabolic changes influence fat-7 expression in worms? 3) How does the downregulation of fat-7 contribute to longevity? Addressing these points would strengthen the mechanistic insights of the study.

Here are my detailed comments:

1. Suppressing FAT-7 levels in C. elegans does not inherently increase lifespan. To directly attribute this effect to FAT-7, it would be important to attempt a rescue experiment to restore FAT-7 expression and assess whether the lifespan extension persists. Additionally, measuring oleic acid levels in these mutants would help determine whether a high-oleic-acid diet is suppressing FAT-7 expression. The role of oleic acid cannot be ruled out using fat-2 mutants (Fig. 3B), as fat-2 mutants accumulate oleic acid when fed WT bacteria, but this may not translate to endogenous oleic acid accumulation in conditions where FAT-7 is suppressed.
2. To understand the host-microbe interaction in this study, it is important to determine what specific changes in the bacteria contribute to the observed phenotypes in worms. Identifying these bacterial factors will provide a clearer picture of their role in influencing worms stress signaling and lifespan.
3. It is important to rule out any changes in food consumption in worms fed these bacterial mutants, as differences in feeding amount could attribute to the observed lifespan effects.
4. In figure 5A to 5G, please include the same-day controls to help clarify how iron supplementation effects these phenotypes relative to the control. For example, in Fig. 5F, it appears that iron extends the lifespan of worms fed the control diet. It would be clearer if appropriate controls were included in all of these figures or summarized in a table to help understand the impact of iron.
5. How does iron depletion affect the levels of fat-7, and how does this contribute to the activation of the longevity pathways discussed in the manuscript.

Minor comments

1. Please include a detailed table of the lifespan data for all replicates as a supplementary table.
2. In the Methods section, specify at what stage the worms were exposed to iron and the iron chelator for the lifespan experiments.
3. Please clarify whether equal optical density (O.D.) of cells was seeded for both the WT and mutant strains, and mention if the mutants exhibit any growth defects.

Significance

General Assessment: This study by Das et al. explores the impact of bacterial mutants on C. elegans lifespan. A key strength of the study is the identification of bacterial mutants that influence the expression of the gene encoding fatty acid desaturase (fat-7) and lifespan in C. elegans. Furthermore, the study highlights the role of iron in regulating fat-7 expression, suggesting that iron imbalance may play a crucial role in modulating fatty acid metabolism. However, the study's main limitation is that it does not significantly extend the current understanding of the microbial modulation of host metabolism and aging, as many of the identified bacterial hits overlap with those previously reported in Zhang et al. (2019). The manuscript would benefit from more in-depth mechanistic exploration, especially with regard to how specific bacterial factors influence the metabolic state of the worms and how these changes ultimately modulate fat-7 expression and longevity.

Advance: This study presents a conceptual advance by exploring the iron-dependent regulation of fat-7 expression and lifespan in C. elegans, linking bacterial mutations with key longevity pathways (SKN-1, SEK-1, and HLH-30). The novelty lies in the direct investigation of the bacterial-induced changes in fat-7 expression, though the role of iron in these mutants for development and induction of mito-UPR was previously shown in the literature. This study also adds to the growing body of work on C. elegans as a model for studying aging and host-microbe interactions, particularly in understanding how diet and microbial exposure affect metabolic processes and lifespan.

Audience: This research will primarily interest specialized audiences in aging research, microbiology, and metabolism, especially those focused on host-microbe interactions.

Keywords of my expertise: Host-microbe interactions, metabolism, system biology, C. elegans, aging.

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

Summary:

In this paper, the authors perform a screen by feeding C. elegans different E. coli genetic mutants and examining the effect on the expression of fat-7, a stearoyl-CoA 9-desturase, which has been associated with longevity. They identify 26 E. coli strains that decrease fat-7 expression, all of which slow development and increase lifespan. RNA sequencing of worms treated with 4 of these strains identified genes involved in defense against oxidative stress among those genes that are commonly upregulated. Feeding C. elegans these 4 bacterial strains results in increased ROS and activation of the mitochondrial unfolded protein response, which appears to contribute to lifespan extension as these bacterial strains do not increase lifespan when the mitochondrial unfolded protein response transcription factor ATFS-1 is disrupted. Finally, the authors demonstrate a role for iron levels in mediating these phenotypes: iron supplementation inhibits the phenotypes caused by the identified bacterial strains, while iron chelation mimics these phenotypes.

Major comments:

The proposed model involves an increase in ROS levels activating the UPRmt and then leading to lifespan extension. If the elevation is ROS levels is contributing then treatment with antioxidants should prevent UPRmt activation and lifespan extension.

The authors suggest that iron depletion may disrupt iron-sulfur cluster proteins. The Rieske iron-sulfur protein ISP-1 from mitochondrial electron transport chain complex III has previously been associated with lifespan. Point mutations affecting the function of ISP-1 or RNAi decreasing the levels of ISP-1 both result in increased lifespan (PMID 20346072, 11709184). Thus, iron depletion may be increasing ROS, activating UPRmt and increasing lifespan through decreasing ISP-1 levels.

All of the Kaplan-meier survival plots are missing statistical analyses. Please add p-values.

It would be helpful to include a model diagram of the proposed mechanisms in the main figures.

Minor comments:

Rather than "mutant diets" it would be more informative to call these "FAT-7-decreasing diets"

Is it surprising that none of the bacterial strains increased FAT-7 levels? Why do you think this is?

Page 5. "We hypothesized that diets reducing FAT-7 might elevate oleic acid levels". Since FAT-7 converts stearic acid to oleic acid, wouldn't deceasing FAT-7 levels decrease oleic acid levels and increase stearic acid levels?

Page 6. The authors cite Bennett et al. 2014 for the statement that "Activation of the UPRmt has been associated with lifespan extension". This paper reaches the opposite conclusion "Activation of the mitochondrial unfolded protein response does not predict longevity in Caenorhabditis elegans". Also, in the Bennett paper and PMID 34585931, it is shown that constitutive activation of ATFS-1 decreases lifespan. Thus, the relationship between the UPRmt and lifespan is not straightforward. These points should be mentioned.

Page 6. "Our transcriptomic analysis suggested elevated ROS". Rather than refer to gene expression, it would be better to refer to the ROS measurements that were performed.

The long-lived mitochondrial mutants isp-1 and nuo-6 have increased ROS, UPRmt activation and increased lifespan. Multiple studies have examined gene expression in these long-lived mutant strains. How does gene expression in these mutants compare to worms treated with the FAT-7-decreasing E. coli mutants? While not necessary for this publication, it would be interesting to see whether the FAT-7-decreasing E. coli strains can increase isp-1 and nuo-6 lifespan.

SEK-1 is also involved in the p38-mediated innate immune signaling pathway, which has been shown to contribute to longevity in C. elegans. In fact, disruption of sek-1 using RNAi decreased the lifespan of several long-lived mutant strains PMID 36514863.

Figure 2. Why were cyoA and ycbk chosen to show the full Kaplan-meier survival plot?

Figure 2, panel D. A better title may be "Mean Survival (Percent increase from control)"

While not necessary for this paper, it would be interesting to determine whether the FAT-7-decreasing E. coli strains alter resistance to oxidative stress.

Figure 4. It may be interesting to include a correlation plot comparing hsp-6::GFP fluorescence and lifespan. It looks like the magnitudes of increase for each phenotype are not correlated.

Significance

Overall, this is an interesting paper and the experiments are rigorously performed. The bacterial screen was comprehensive and was followed up by careful mechanistic experiments. This paper will be of interest to researchers studying the biology of aging. A diagram of the working model of the underlying mechanisms would enhance the paper.

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 (Evidence, reproducibility and clarity (Required)):*

Summary: Chitin is a critical component of the extracellular matrix of arthropods and plays an essential role in the development and protection of insects. There are two chitin synthases in insects: Type A (exoskeletons) and Type B (for the peritrophic matrix in the gut). The study aims to investigate the specificity and mechanisms of the two chitin synthases in D. melanogaster and to clarify whether they are functionally interchangeable. Various genetic manipulations and fluorescence-based labeling were used to analyze the expression, localization, and function of Kkv and Chs2 in different tissues. Chs2 is expressed in the PR cells of the proventriculus and is required for chitin deposition in the peritrophic matrix. Kkv can deposit chitin in ectodermal tissues but not in the peritrophic matrix, whereas Chs2 can deposit chitin in the peritrophic matrix but not in ectodermal tissues. The subcellular localization of chitin synthases is specific to the tissues in which they are expressed. Kkv localizes apically in ectodermal tissues, whereas Chs2 localizes apically in the PR cells of the proventriculus. Altogether, Kkv and Chs2 cannot replace each other. The specificity of chitin synthases in D. melanogaster relies on distinct cellular and molecular mechanisms, including intracellular transport pathways and the specific molecular machinery for chitin deposition.*

• *

Congratulations on this incredible story and manuscript, which is straightforward and well-written. However, I have some comments that may help to improve it.

We thank the reviewer for this very positive comment. We have addressed all comments to clarify and improve our manuscript.

Major comments: 1.) Funny thing: the Chs2 mutant larva shows a magenta staining below the chitin accumulation of the esophagus, which looks like a question mark in 1H but cannot be found in control. Is that trachea reaching the pv?

We assume that the reviewer refers to Fig 1N. As the reviewer suspects, this corresponds to a piece of trachea. Figure 1N shows a single section, making it difficult to identify what this staining corresponds to. We are providing below a projection of several sections where it is easier to identify the staining as tracheal tissue (arrow).

We are now marking this pattern as trachea (tr) in the manuscript Figure 1N

2.) Also, though it is evident that the PM chitin is lost in Ch2 mutants, could it be that the region is disturbed and cells express somewhere else chitin? There are papers by Fuß and Hoch (e.g., Mech of Dev, 79, 1998; Josten, Fuß et al., Dev. Biol.267, 2004) using markers such as Dve, Fkh, Wg, Delta, and Notch, etc. for precisely marking the endodermal/ectodermal region in the embryonic foregut/proventriculus. It would be beneficial to show, along with chitin and Chs expression patterns, the ectoderm/endoderm cells. This is particularly important as the authors report endodermal expression of Chs2 in embryos but don't use co-markers of the endodermal cells.

We agree with the reviewer that this is an important issue and we note that Reviewer 2 also raised the same point. Therefore, we have addressed this issue.

We obtained an antibody against Dve, kindly provided by Dr. Hideki Nakagoshi. Dve marks the endodermal region in the proventriculus (Fuss and Hoch, 1998, Fuss et al., 2004, Nakagoshi et al., 1998).This antibody worked nicely in our dissected L3 digestive tracts and allowed us to mark the endodermal region. We also obtained an antibody against Fkh, kindly provided by Dr. Pilar Carrera. Fkh marks the ectodermal foregut cells (Fuss and Hoch, 1998, Fuss et al., 2004). While, in our hands, this antibody performed well in embryonic tissues, we observed no staining in our dissected L3 digestive tracts. The reason for this is unclear, but we suspect technical limitations may be responsible (the ectodermal region of the proventriculus is very internal, potentially hindering antibody penetration). To circumvent this inconvenience, we tested a FkhGFP tagged allele available in Bloomington Stock Center. Fortunately, we were able to detect GFP in ectodermal cells of L3 carrying this allele. Using this approach, we conducted experiments to detect Fkh and Dve in the wild type or in Df(Chs2) conditions (Fig S1). In addition, we used these markers to map the expression of Kkv and Chs2 in the proventriculus (Fig 4).

Altogether the results using these endodermal/ectodermal markers confirmed the presence of a cuticle adjacent to the FkhGFP-positive cells and a PM adjacent to the PR cells, marked by Dve. This PM is absent in Df(Chs2) L3 escapers, however, the general pattern of Fkh/Dve expression is not affected. Finally, we show that Chs2-expressing cells are positive for Dve while Kkv-expressing cells are not. We were unable to conduct an experiment demonstrating Kkv and Fkh co-expression due to technical incompatibilities, as both genes require the use of GFP-tagged alleles to visualise their expression. However, we believe that our imaging of Dve/Kkv clearly shows that Kkv expressing cells lack Dve expression and are localised in the internal (ectodermal) region of the proventriculus (Fig 4E).

3.) The origin of midgut chitin accumulation is unclear. Chitin can come from yeast paster. Can the authors check kkv and chs2 mutants for food passage and test starving L1 larvae to detect chitin accumulation in the midgut without feeding them?

This is a very interesting point that has also intrigued us.

We observed that, in addition to the PM layer lining the midgut epithelium, CBP staining also revealed a distinct luminal pattern. Our initial hypothesis was that this pattern corresponded to the PM. However, its presence in Df(Chs2) larval escapers clearly indicates that this is not the case. Unfortunately, we cannot assess this pattern in kkv mutants, as these die at eclosion and do not proceed to larva stages.

As the reviewer suggests, a likely possibility is that the luminal pattern originates from components in the food. These could correspond to yeast, as suggested by the reviewer, or possibly remnants of dead larvae present in the media (although Drosophila is considered herbivore in absence of nutritional stress).

To assess whether the luminal pattern originates from the food we conducted two independent experiments. In experiment 1, we collected larvae reared under normal food conditions. Newly emerged L3 larvae were transferred in small numbers to minimise cannibalism (Ahmad et al., 2015) to new Petri plates containing moist paper. Larvae were starved for 3,4 or 5 days. Larvae starved for more than 5 days did not survive. We then dissected the guts and analysed CBP staining. We observed the presence of luminal CBP staining in these larvae, along with the typical PM signal in the proventriculus and along the midgut. In experiment 2, we collected larvae directly on agar plates containing only agar (without yeast or any other nutrients). We allowed the larvae to develop. These larvae showed minimal growth. We dissected the guts of these small larvae (which were challenging to dissect) and analysed CBP staining. Again, we detected presence of luminal CBP staining.

These experiments indicate that, despite starvation, a luminal chitin pattern is still detected, suggesting that it is unlikely to originate from food. However, we cannot unequivocally rule out the possibility that the cannibalistic, detrivorous or carnivorous behavior of the nutrionally stressed larvae (Ahmad et al., 2015) in our experiments may influence the results. Therefore, more experiments would be required to address this point.

In summary, while we cannot provide a definitive answer to the reviewer's question, nor fully satisfy our own curiosity, we would like to note that this specific observation is unrelated to the main focus of our study, as we have confirmed that the luminal pattern is not dependent on Chs2 function.

Portions of midgut of starved larvae under the regimes indicated, stained for chitin (CBP, magenta). Note the presence of the luminal chitin pattern in the midgut

4.) Subcellular localization assays require improved analysis, such as a co-marker for the apical membrane and statistical analysis with co-localization tools, showing the overlap at the membrane and intracellularly with membrane co-markers and KDEL.

We have addressed the point raised by the reviewer. To analyse and quantify Chs2 subcellular localisation, particularly considering the observed pattern, we decided to use both a membrane and an ER marker. As a membrane marker we used srcGFP expressed in tracheal cells (see answer to point 7 of Reviewer 1) and as an ER marker we used KDEL. In this analysis, tracheal cells also expressed Chs2, which was visualised using the Chs2 antibody generated in the lab.

To assess the colocalisation of Chs2 with each marker we used the JaCop pluggin in Fiji. We analysed individual cells from different embryos stained for membrane/ER/Chs2 using single confocal sections (to avoid artificial colocalisation). Images were processed as described in Materials and Methods. We obtained the Pearson's correlation coefficient (r), which measures the degree of colocalisation, for Chs2/srcGFP and Chs2/KDEL, n=36 cells from 9 different embryos. The average r value for Chs2/srcGFP was 0,064, while the average for Chs2/KDEL was around 0,7. r ranges between -1 and 1, where 1 indicates perfect correlation, 0 no correlation, and -1 perfect anti-correlation. Typically, an r value of 0.7 and above is considered a strong positive correlation, whereas a value below 0,1 is regarded as very weak or no correlation. Thus, our colocalisation analysis supports the hypothesis that Chs2 is primarily retained in the ER when expressed in non-endogenous tissues, likely unable to reach the membrane.

We have reorganised the figures and now present an example of Chs2/srcGFP/KDEL subcellular localisation in tracheal cells and the colocalisation analysis in Fig 5H. The colocalisation analysis is described in the Materials and Methods section.

Minor comments:

5.) The authors used "L3 larval escapers." It would be interesting to know if the lack of Chs2 and the peritrophic matrix cause any physiological defects or lethality.

The point raised by the reviewer is very interesting and relevant. The peritrophic matrix is proposed to play several important physiological roles, including the spatial organisation of the digestive process, increasing digestive efficiency, protection against toxins and pathogens, and serving as a mechanical barrier. Therefore, it is expected that the absence of chitin in the PM of the Df(Chs2) larval escapers may cause various physiological effects.

Analysing these effects is a complex task, and it constitutes an entire research project on its own. In addressing the physiological requirements of the PM, we aim to analyse adult flies and assess various parameters, including viability, digestive transit dynamics, gut integrity, resistance to infections, fitness and fertility.

A critical initial challenge in conducting a comprehensive analysis of the physiological requirements of the PM is identifying a suitable condition to evaluate the absence of Chs2. In this work we are using a combination of two overlapping deficiencies that uncover Chs2, along with a few additional genes (as indicated in Fig S1F). This deficiency condition presents two major inconveniences: first, the observed defects could be caused or influenced by the absence of genes other than Chs2, preventing us from conclusively attributing the defects to Chs2 loss (unless we rescued the defects by adding Chs2 back as we did in the manuscript). Second, the larva escapers, which are rare, do not survive to adulthood (indicating lethality but preventing us from analysing specific physiological aspects).

To overcome these limitations, we are currently working to identify a genetic condition in which we can specifically analyse the absence of Chs2. We have identified several available RNAi lines and we are testing their efficiency in preventing chitin deposition in the PM. Additionally, we are characterising a putative null Chs2 allele, Chs2CR60212-TG4.0. This stock contains a Trojan-GAL4 gene trap sequence in the third intron, inserted via CRISPR/Cas9. As described in Flybase (https://flybase.org/), the inserted cassette contains a 'Trojan GAL4' gene trap element composed of a splice acceptor site followed by the T2A peptide, the GAL4 coding sequence and an SV40 polyadenylation signal. When inserted in a coding intron in the correct orientation, the cassette should result in truncation of the trapped gene product and expression of GAL4 under the control of the regulatory sequences of the trapped gene. We already know that, when crossed to a reporter line (e.g. UAS-GFP or UAS-nlsCherry) this line reproduces the Chs2 expression pattern, suggesting that the insertion may generate a truncated Chs2 protein. This line would represent an ideal tool to assess the absence of Chs2, and we are currently characterising it for further analysis

In summary, we fully agree with the reviewer that investigating the physiological requirements of the PM is a compelling area of research, and we are actively addressing this question. However, this investigation constitutes a substantial and independent research effort that we believe is beyond the scope of the current manuscript at this stage.

6.) The order identifiers are missing for materials and antibodies, e.g., anti-GFP (Abcam), but Abcam provides several ant-GFP; which was used? Please provide order numbers that guarantee the repeatability for others.

We have now added all identifiers for materials and reagents used, in the materials and methods section.

7.) Figure S5C, C', what marks GFP (blue) in the trachea? Maybe I have overlooked the description. What is UASsrcGFP? What is the origin of this line?

We apologise for not providing a more detailed description of the UASsrcGFP line. This line corresponds to RRID BDSC#5432, as now indicated in Materials and Methods section.

In this transgene, the UAS regulatory sequences drive the expression of GFP fused to Tag:Myr(v-src). As described in Flybase (https://flybase.org/), the P(UAS-srcEGFP) construct contains the 14 aa myristylation domain of v-src fused to EGFP. This tag is commonly used to target proteins of interest to the plasma membrane. The construct was generated by Eric Spana and is available in Drosophila stock centers.

We typically use this transgene as a plasma membrane marker to outline cell membrane contours. In our experiments, srcGFP, under the control of the btlGal4 promoter, was used to visualise the membrane of tracheal cells in relation to Chs2 accumulation. As indicated in point 4, we have now transferred the images of srcGFP/Chs2/KDEL to the main Figures and used it for colocalisation analyses.

8.) The authors claim that they validated the anti-Chs2 antibody. However, they show only that it recognizes a Cht2 epitope via ectopic expression. For more profound validation, immune staining is required in deletion mutants, upon knockdown, or upon expression of recombinant proteins, which is not shown.

We generated an antibody against Chs2. We found that the antibody does not reliably detect the endogenous Chs2 protein, and so we find no pattern in the proventriculus or any other tissue in our immunostainings. It is very possible that the combination of low endogenous levels of Chs2 with a sub-optimal antibody (or low titer) leads to this result. In any case, as the antibody does not detect endogenous Chs2, it cannot be validated by analysing the expression upon Chs2 knockdown. In contrast, our antibody clearly detects specific staining in various tissues (e.g. trachea, salivary glands, gut) when Chs2 is expressed using the Gal4/UAS system, confirming its specificity for Chs2. It is worth to point that it is not unusual to find antibodies that are not sensitive enough to detect endogenous proteins but can detect overexpressed proteins (e.g

(Lebreton and Casanova, 2016)).

As an additional way to validate the specificity of our antibody, we have used the chimeras generated, as suggested by the reviewer. As indicated in the Materials and Methods section, the Anti-Chs2 was generated against a region comprising 1222-1383 aa in Chs2, with low homology to Kkv. This region is present in the kkv-Chs2GFP chimera but absent in Chs2-KkvGFP (see Fig 7A). Accordingly, our antibody recognises kkv-Chs2GFP but does not recognise Chs2-KkvGFP (Fig S7).

We have revised the text in chapter 6 (6. Subcellular localisation of Chs2 in endogenous and ectopic tissues) to clarify these points and we have added the validation of the antibody using the chimeras in chapter 8 (8. Analysis of Chs2-Kkv chimeras) and Fig S7

9) The legend and text explaining Fig. 4 D-E' can be improved. The authors used the Crimic line, which is integrated into the third ("coding") intron. This orientation can lead to the expression of Gal4 and cause a truncated version of the protein (according to Flybase). Is Chs2 expression reduced in the crimic mutant? If the mutation causes expression of a truncated version, the Chs2 antibody may not be able to detect it as it recognizes a fragment between 1222 and 1383 aa? Also, I'm unsure whether the Chs2 antibody or GFP was used to detect expression in PR cells. The authors describe using Ch2CR60212>SrcGFP together with Chs2+ specific antibodies.

We apologise for the confusion.

As the reviewer points, Chs2CR60212-TG4.0 contains a Trojan-GAL4 gene trap sequence in the third intron, inserted via CRISPR/Cas9. As described in Flybase (https://flybase.org/), the inserted cassette contains a 'Trojan GAL4' gene trap element composed of a splice acceptor site followed by the T2A peptide, the GAL4 coding sequence and an SV40 polyadenylation signal. When inserted in a coding intron in the correct orientation, the cassette should result in truncation of the trapped gene product and expression of GAL4 under the control of the regulatory sequences of the trapped gene.

We found that when crossed to UAS-GFP or UAS-nlsCherry, this line reproduces a expression pattern that must correspond to Chs2. As the antibody that we generated is not suitable for detecting Chs2 endogenous expression, we resorted to using this combination, Chs2CR60212-TG4.0 crossed to a reporter line (such asUAS-GFP or UAS-nlsCherry), to visualise Chs2 expression by staining for GFP/Cherry in the intestinal tract and in the embryo (Figures 4 and S4).

We realise that the Figure labelling we used in our original submission is very misleading, and we apologise for this. In the original figures we had labelled the staining combination with Kkv, Chs2, Exp as if we had used these antibodies. However, in all cases, we used GFP to visualise the pattern of these proteins in the genetic combinations indicated in the figures. We have corrected this in our revised version. We have also updated the text (Chapter 5), figures and figure legends.

As the reviewer points, the insertion in Chs2CR60212-TG4.0 is likely to generate a truncated Chs2 protein. We cannot confirm this using the Chs2 antibody we generated because it does not recognise the endogenous Chs2 pattern. Nevertheless, as indicated in point 5, we are currently characterising this line. Our preliminary results indicate a high complexity of effects from this allele that require thorough analysis, as it may be acting as a dominant negative.

Reviewer #1 (Significance (Required)):

Significance: The manuscript's strength and most important aspects are the genetic analysis, expression, and localization studies of the two Chitin synthases in Drosophila embryos and larvae. However, beyond this manuscript, the development of mechanistic details, such as interaction partners that trigger secretion and action at the apical membranes and the role of the coiled-coil domain, will be interesting.

The manuscript uses "first-class" genetics to describe the different roles of the two Chitin synthases in Drosophila, comparing ectodermal chitin (tracheal and epidermal chitin) with endodermal (midgut) chitin. Such a precise analysis has not been investigated before in insects. Therefore, the study deeply extends knowledge about the role of Chitin synthases in insects.

The audience will specialize in basic research in zoology, developmental biology, and cell biology regarding - how the different Chitin synthases produce chitin. Nevertheless, as chitin is relevant to material research and medical and immunological aspects, the manuscript will be fascinating beyond the specific field and thus for a broader audience.

I'm working on chitin in the tracheal system and epidermis in Drosophila.

__Reviewer #2 (Evidence, reproducibility and clarity (Required)): __ Drosophila have two different chitin synthase enzymes, Kkv and Chs2, and due to unique expression patterns and mutant phenotypes, it is relatively clear that they have different functions in producing either the cuticle-related chitin network (Kkv) or the chitin associated with the peritrophic matrix (PM). However, what is unknown is whether the different functions in making cuticle vs PM chitin is related to differences in cellular expression and/or enzyme properties within the cell. The authors exploit the genetic tractability of Drosophila and their ability to image cuticle vs PM chitin production to examine whether these 2 enzymes can substitute each other. They conclude that these two proteins are not equivalent in their capacity to generate chitin. The data are convincing; however, it is currently presented in a subjective fashion, which makes it difficult to interpret. Additionally, in my opinion there is some interpretation that requires softening or alternatively interpreted.

We are pleased that the reviewer finds our data convincing. However, we acknowledge the reviewer's concern that our data was presented in a subjective manner, and we apologise for this. In response, we have carefully reviewed the entire manuscript and revised our data presentation to ensure a more objective tone. Numerous changes (including additional quantifications, new experiments and clarifications) have been incorporated throughout the text. These revisions are highlighted in the marked-up version. We hope that this revision provides a more accurate and objective presentation of our work.

Major Comments:

1- While the imaging is lovely, there are some things that are difficult to see in the figures. For example, the "continuous, thin and faint 'chitin' layer that lined the gut epithelium" is very difficult to visualise in the control images. Can they increase the contrast to help the reader appreciate this layer? This is particularly important as we are asked to appreciate a loss of this layer in the absence of Chs2.

We have tried to improve the figures so that the PM layer in the midgut region is more clearly visible. We have added magnifications of small sections at the midgut lumen/epithelium border in grey to help visualise the PM. These improvements have been made in Figures 1,2,S1,S2,S3 and we believe that they better illustrate our results.

2- All the mutant analysis is presented subjectively. For example, the authors state that they "found a consistent difference of CBP staining when they compared the 'Chs2' escapers to the controls". How consistent is consistent? Can this be quantified? What is the penetrance of this phenotype? They say that the thin layer is absent in the midgut and the guts are thinner. Could they provide more concrete data?

As indicated above, we have reviewed the text to provide a more objective description of the phenotypes.

We have quantified the defects in the Df(Chs2) mutant conditions. For this quantification we dissected intestinal tracts of control and Df(Chs2) larva escapers. We fixed, stained and mounted them together. The control guts expressed GFP in the midgut region as a way to distinguish control from mutants. We analysed the presence or absence of chitin in the PM. We found absence of chitin in the proventricular lumen and in the midgut in all Df(Chs2) guts and presence of chitin there in all control ones (n=12 Df(Chs2) guts, n=9 control guts, from 5 independent experiments). The results indicate a fully penetrant phenotype of lack of chitin in Df(Chs2) larva escapers (100% penetrance). We have added this quantification in the text, chapter 2 (2. Chs2 deposits chitin in the PM).

To quantify the thickness of the guts, we took measurements of the diameter in control and Df(Chs2) guts at two comparable distance positions from the proventriculus (position 1, position 2, see image). Our quantifications indicated thinner tubes in mutant conditions.

Image shows the anterior part of the intestinal tract, with the proventriculus encircled in white. Positions 1 and 2 indicate where the diameter quantifications were taken. Scatter plots quantifying the diameter at the two different positions in control and Chs2 larval escapers. Bars show mean {plus minus} SD. p=p value of unpaired t test two-tailed with Welch's correction.

However, we are aware that our analysis of the thickness of the gut is not accurate, because we have not used markers to precisely measure at the same position in all guts and because we have not normalised the measurement position in relation to the whole intestinal tract (mainly due to technical issues).

In relation to the fragility, we noticed that the guts of Chs2 larval escapers tended to break more easily during dissection than control guts, however, we have not been able to quantify this parameter in a reliable and objective manner.

Since we consider that the requirement of Chs2 for PM deposition is sufficiently demonstrated, and that aspects such as gut morphology or fragility relate to the physiological requirements of the PM, which we are beginning to address as a new independent project (see our response to point 5 of Reviewer 1), we have decided to remove the sentence 'We also noticed that the guts of L3 escapers were thinner and more fragile at dissection." from the manuscript to avoid subjectivity.

3- They state that Chs2 was able to restore accumulation of chitin in the PM of the proventriculus and the midgut. Please quantify. Additionally, does this restore the morphology of the guts (related to the comment above on the thinner guts in the absence of Chs2)?

We have quantified the rescue of chitin deposition in the PM when Chs2 is expressed in PR cells in a Df(Chs2) mutant background. For this quantification we used the following genetic cross: PRGal4/Cyo; Df(Chs2)/TM6dfdYFP (females) crossed to UASChs2GFP or UASChs2/Cyo; Df(Chs2)/TM6dfdYFP. We selected Df(Chs2) larval escapers by the absence of TM6 (recognisable by the body shape). Among these larval escapers, we identified the presence of Chs2 in PR cells by the expression of GFP or Chs2. We found absence of chitin in the proventriculus and in the midgut in all Df(Chs2) guts that did not express Chs2 in PR cells (n=8/8 Df(Chs2)). In contrast, chitin was present in those intestinal tracts where Chs2 expression was detected in PR cells (n=8/8 PRGal4-UASChs2; Df(Chs2) guts, from 5 independent experiments). The results indicate a full rescue of chitin deposition by Chs2 expression in PR cells in Df(Chs2) mutant larvae. We have added this quantification in the text, chapter 2 (2. Chs2 deposits chitin in the PM).

As requested by the reviewer, we have also conducted measurements to quantify gut thickness. We performed an analysis similar to the one described in point 2, this time comparing the diameter of Df(Chs2) and PRGal4-UASChs2;Df(Chs2) guts at positions 1 and 2 (see image in point 2 of Reviewer 2). Our quantifications indicated that guts were thicker when Chs2 is expressed in the PR region in Df(Chs2) larval escapers.

As discussed in point 2, we have decided not to include these results in the manuscript, as this type of analysis requires a more comprehensive investigation.

Scatter plots quantifying the diameter at the two different positions in Chs2 larval escapers and Chs2 larval escapers expressing Chs2 in PR cells. Bars show mean {plus minus} SD. p=p value of unpaired t test two-tailed with Welch's correction.

4- This may be beyond the scope of this paper, but I find it interesting that the PM chitin is deposited in the proventricular lumen. Yet it forms a thin layer that lines the entire midgut? Any idea how this presumably dense chitin network gets transported throughout the midgut to line the epithelium? I imagine that this is unlikely due to diffusion, especially if they see an even distribution across the midgut. Do they see any evidence of a graded lining (i.e. is it denser in the midgut towards the proventriculus and does this progressively decrease as you look through the midgut?)?

Insect peritrophic matrices have been classified into Type I and II (with some variations) depending on their origin (extensively reviewed in (Peters, 1992, Hegedus et al., 2019). Type I PMs are typically produced by delamination as concentric lamellae along the length of the midgut. Type II PMs, in contrast, are produced in a specialised region of the midgut that corresponds to the proventriculus and are typically more organised than Type I. In Type II PMs, distinct layers originate from distinct cell clusters in the proventriculus. It has been proposed that as food passes, it becomes encased by the extruded PM, which then slides down to ensheath the midgut. Drosophila larvae have been proposed to secrete a type II PM: through PM implantation experiments, Rizki proposed that the proventriculus is required to generate the PM in Drosophila larvae (Rizki, 1956). Our experiments confirmed this hypothesis: we show that expressing Chs2 exclusively in PR cells is sufficient to produce a PM along the midgut. Furthermore, we also show that expressing Chs2 in the midgut is not sufficient to produce a PM layer lining the midgut, at least at larval stages.

The type II PM in Drosophila is proposed to be fully organised into four layers in the proventricular region (also referred as PM formation zone) before reaching the midgut (Peters, 1992, King, 1988, Rizki, 1956, Zhu et al., 2024). However, the mechanism by which the PM is subsequently transported into the midgut remains unclear. PM movement posteriorly is thought to depend on to the pressure exerted by continuous secretion of PM material (Peters, 1992). Early work by Wigglesworth (1929, 1930) proposed that the PM is secreted into the proventricular lumen, becomes fully organised, and is then pushed down by a press mechanism involving the aposed ectodermal/endodermal walls of the proventriculus. Rizki suggested that muscular contractions of the proventriculus walls may play a role, and that peristaltic movements of the gut add a pulling force to push the PM into the midgut (Rizki, 1956). Nevertheless, to our knowledge, the exact mechanism is still not fully understood.

In response to the reviewer's question, the level of resolution of our analysis does not allow us to determine whether there is a graded PM lining along the midgut. However, available data using electron microscopy approaches suggest that the PM is a fully organised structure composed of four layers that is secreted and transported to line the midgut (King, 1988, Zhu et al., 2024).

5- The authors state that expression of kkv in tracheal cells of kkv mutants perfectly restores accumulation of chitin in the luminal filaments. Is this really 100% restoration? They also reference a paper here, which may have quantified this result.

We previously reported that the expression of kkv in tracheal cells restores chitin deposition in kkv mutants (Moussian et al,2015). However, our previous study did not quantify this rescue. As requested by the reviewer, we have now quantified the extent of the rescue.

To perform this quantification, we used the following genetic cross:

btlGa4/(Cyo); kkv/TM6dfdYFP (females) crossed to +/+; kkv UASkkvGFP/TM6dfdYFP (males)

We stained the resulting embryos with CBP (to detect chitin) and GFP. GFP staining allowed us to identify the kkv mutants (by the absence of dfdYFP marker) and to simultaneously identify the embryos that expressed kkvGFP in tracheal cells (through btlGal4-driven expression). Since btlGal4 is homozygous viable, most females carried two copies of btlGal4.

We compared the following embryo populations across 4 independent experiments:

1. Cyo/+; kkv/kkv UASkkvGFP (kkv mutants not expressing kkv in the trachea)

2. btlGal4/+; kkv/kkv UASkkvGFP (kkv mutants expressing kkv in the trachea) Results:

3. Cyo/+; kkv/kkv UASkkvGFP ---- 0/6 embryos deposited chitin in trachea

4. btlGal4/+; kkv/kkv UASkkvGFP ---- 27/27 embryos deposited chitin in trachea These results indicate complete restauration of chitin deposition in kkv mutants when kkv is expressed in tracheal cells (100% rescue).

To further investigate whether Chs2 can compensate for kkv function in ectodermal tissues, we performed a similar quantification using the following genetic cross:

btlGa4/(Cyo); kkv/TM6dfdYFP (females) crossed to UASChs2GFP/UASChs2GFP; kkv UASkkvGFP/TM6dfdYFP (males)

We compared the following embryo populations across 2 independent experiments:

1. Cyo/UASChs2GFP; kkv/kkv (kkv mutants not expressing Chs2 in the trachea)

2. btlGal4/ UASChs2GFP; kkv/kkv (kkv mutants expressing Chs2 in the trachea) Results:

3. Cyo/UASChs2GFP; kkv/kkv ---- 0/4 embryos deposited chitin in trachea

4. btlGal4/ UASChs2GFP; kkv/kkv ---- 0/16 embryos deposited chitin in trachea These results indicate no restauration of chitin deposition in kkv mutants expressing Chs2 in the trachea (0% rescue).

We have now incorporated these quantifications in the text, chapter 4 (4. Chs2 cannot replace Kkv and deposit chitin in ectodermal tissues.)

6- They ask whether Kkv overexpression in the proventriculus can rescue Chs2 mutants... and vice versa, whether Chs2 overexpression in ectodermal cells can rescue kkv mutants. They show that kkv overexpression leads to an intracellular accumulation of chitin in the proventriculus. However, Chs2 overexpression in the trachea did not lead to any accumulation of chitin in the cells. They tailored their experiments and the associated discussion to address the hypothesis that there is potentially some difference in trafficking of these components. However, another possibility, which they have not ruled out, is that the different ability of kkv and Chs2 to produce chitin inside cells of the proventriculus and ectoderm, respectively, is potentially related to different enzymatic activities and cofactors required for chitin formation in these different cell types. Is this another potential explanation for the differences that they observe?

We note that Kkv overexpression in any cell type (e.g. ectoderm, endoderm) consistently leads to chitin polymerisation. In ectodermal tissues, Kkv expression, in combination with Exp/Reb activity, results in extracellular chitin deposition. In the absence of Exp/Reb, Kkv expression leads to the accumulation of intracellular chitin punctae (De Giorgio et al., 2023, Moussian et al., 2015); this work). This correlates with the accumulation of Kkv at the apical membrane and presence of Kkv-containing vesicles, regardless of the presence of Exp/Reb (De Giorgio et al., 2023, Moussian et al., 2015); Figure 6, S6). In endodermal tissues, regardless of the presence of Exp/Reb, Kkv cannot deposit chitin extracellularly and instead produces intracellular chitin punctae. This correlates with a diffuse accumulation of Kkv in the endodermal cells (PR cells, or gut cells in the embryo) but presence of Kkv-containing vesicles (Figure 6, S6).

In previous work we showed that Kkv's ability to polymerise chitin is completely abolished when it is retained in the ER. Indeed, we found that a mutation in a conserved WGTRE region leads to ER retention, the absence of Kkv-containing vesicles in the cell, and absence of intracellular chitin punctae or chitin deposition (De Giorgio et al., 2023).

These findings indicate a correlation between Kkv subcellular localisation and chitin polymerisation/extrusion. Therefore, we hypothesise that intracellular trafficking and subsequent subcellular localisation play a crucial role in regulating Kkv activity (De Giorgio et al., 2023; this work).

We find that Chs2 is expressed in PR cells (Figure 4) and observe that only in these PR cells does Chs2 localise apically (Fig 5A-D, S5A,B). This localisation correlates with the ability of Chs2 to deposit chitin in the PM and the presence of intracellular chitin punctae in PR cells (Fig 1F). When Chs2 is expressed in other cells types, we detect it primarily in the ER and observed no Chs2-containing vesicles (vesicles are suggestive of trafficking). This localisation correlates with the inability of Chs2 to produce intracellular chitin punctae or extracellular chitin deposition.

Again, these results suggest a correlation between Chs2 subcellular localisation and chitin polymerisation/extrusion, aligning with the results observed for Kkv. Therefore, we hypothesise in this work that the intracellular trafficking and subsequent subcellular localisation of Chs2 play a crucial role in regulating its activity.

Our hypothesis is consistent with seminal work in yeast chitin synthases, which has demonstrated the critical role of intracellular trafficking, and particularly ER exit, in regulating chitin synthase activity (reviewed in (Sanchez and Roncero, 2022).

That said, we cannot exclude other explanations that are also compatible with the observed results. As pointed out by the reviewer, it is possible that Chs2 and Kkv require different enzymatic activities and/or cofactors for chitin polymerisation/deposition, which may be specific to different cell types. Indeed, we know that the auxiliary proteins Exp/Reb are specifically expressed in certain ectodermal tissues (Moussian et al., 2015). These mechanisms could act jointly or in parallel with the regulation of intracellular trafficking, or could even regulate this intracellular trafficking itself.

Identifying the exact mechanisms controlling Kkv and Chs2 intracellular trafficking would be necessary to determine whether additional mechanisms (specific cofactors or enzymatic activities) are also involved or even serve as the primary regulatory elements.

We have introduced these additional possibilities in the discussion section.

7- They co-express Chs2 and Reb and show that this does not lead to chitin production or secretion. In the discussion they conclude that Chs2 does not "seem to be dependent on 'Reb' activity". I think that this statement potentially needs softening. They show that Reb is not sufficient in to induce Chs2 chitin production in cells that do not normally make a PM. However, they do not show that it is not essential in cells that normally express Chs2 and make PM.

We fully agree with the reviewer's observation and thank her/him for pointing it out.

As indicated by the reviewer, we show that co-expression of Reb and Chs2 in different tissues does not lead to an effect distinct from that observed with Chs2 expression alone. In addition, in the discussion we mention that we could not detect expression of reb/exp in PR cells, which aligns with the findings from Zhu et al, 2024, indicating no expression of reb/exp in the midgut cells of the adult proventriculus, as assessed by scRNAseq. We found that exp is expressed in the ectodermal cells of the larval proventriculus (Fig S4D), correlating with kkv expression in this region and cuticle deposition. These findings led us to propose that Chs2 does not seem to be dependent on Exp/Reb activity.

However, in our original manuscript, we did not directly address whether Exp/Reb are required in the cells that normally express Chs2. As a result, we could not conclude that Chs2 relies on a set of auxiliary proteins different from Exp/Reb, and therefore a different molecular mechanism to that of Kkv in regulating chitin deposition.

To address this specific point, we have conducted a new experiment to test Exp/Reb requirement in PR cells. We co-expressed RNAi lines for Exp/Reb in these cells and found that chitin deposition in the PM was not prevented. This further supports the hypothesis that Exp/Reb activity is not necessary for Chs2 function. We have added this experiment to Chapter 4 and Fig S3I,J.

8- They looked at the endogenous expression pattern of kkv and Chs2 and say that they found accumulation of Kkv in the proventriculus and no accumulation in the midgut. Siimilarly, they look at the expression of Chs2 and detect it in cells of the proventriculus. Are there markers of these different cell types that they could use to colocalize these enzymes?

We agree with the reviewer that this is an important issue and we note that Reviewer 1 also raised the same point. Therefore, we have addressed this issue.

We obtained an antibody against Dve, kindly provided by Dr. Hideki Nakagoshi. Dve marks the endodermal region in the proventriculus (Fuss and Hoch, 1998, Fuss et al., 2004, Nakagoshi et al., 1998).This antibody worked nicely in our dissected L3 digestive tracts and allowed us to mark the endodermal region. We also obtained an antibody against Fkh, kindly provided by Dr. Pilar Carrera. Fkh marks the ectodermal foregut cells (Fuss and Hoch, 1998, Fuss et al., 2004, Nakagoshi et al., 1998). While, in our hands, this antibody performed well in embryonic tissues, we observed no staining in our dissected L3 digestive tracts. The reason for this is unclear, but we suspect technical limitations may be responsible (the ectodermal region of the proventriculus is very internal, potentially hindering antibody penetration). To circumvent this inconvenience, we tested a FkhGFP tagged allele available in Bloomington Stock Center. Fortunately, we were able to detect GFP in ectodermal cells of L3 carrying this allele. Using this approach, we conducted experiments to detect Fkh and Dve in relation to chitin accumulation in the wild type (Fig S1). In addition, we used these markers to map the expression of Kkv and Chs2 in the proventriculus (Fig 4). Our results using these endodermal/ectodermal markers confirmed the presence of a cuticle adjacent to the FkhGFP-positive cells and a PM adjacent to the PR cells, marked by Dve. Additionally, we show that Chs2-expressing cells are positive for Dve while Kkv-expressing cells are not. We could not conduct an experiment showing Kkv and Fkh co-expression due to technical incompatibilities, as we have to use GFP tagged alleles for both Kkv and Fkh to reveal their expression. However, we believe that our imaging of Dve/Kkv clearly shows that Kkv expressing cells lack Dve expression and localise in the internal (ectodermal) region of the proventriculus (Fig 4E).

9- They overexpress Chs2 in cells of the midgut and see that it colocalises with an ER marker. They conclude that it is retained in the ER, which again, for them suggests that it has a trafficking problem in these cells. However, they are overexpressing it in these cells and this strong accumulation that they observe in the ER could simply be due to the massive expression levels. Additionally, they cannot conclude that it doesn't get out of the ER at all. They could be correct in thinking that there may be a trafficking issue, but this experiment does not conclusively show that Chs2 is entirely retained in the ER when expressed in ectopic tissues. I wonder if their interpretation needs softening or whether they should potentially address alternative hypotheses.

The reviewer raises two distinct issues: 1) the localisation of overexpressed proteins 2) Chs2 ER retention.

We agree that massive overexpression can lead to artifactual subcellular localisation due to saturation of the secretory pathway, causing ER accumulation. In our experiments, we overexpressed Kkv and Chs2 in different tissues (trachea, salivary glands, embryonic gut, and larval proventriculus), inducing high levels of both chitin synthases.

For Kkv, we observed distinct subcellular localisation patterns in ectodermal versus endodermal tissues (illustrated in new Fig S6). In ectodermal tissues such as the trachea, large amounts of KkvGFP were detected, most of it localising apically. We also detected a more general KkvGFP distribution throughout the cell, including the ER, particularly at early stages. Additionally, we observed many KkvGFP-positive vesicles, reflecting exocytic and endocytic trafficking, as described previously (De Giorgio et al., 2023). The presence of these vesicles (as well as the apical localisation) indicates that KkvGFP is able to exit the ER. Indeed, our previous work demonstrated that when Kkv is retained in the ER, it does not localise apically or appear in vesicles (De Giorgio et al, 2023). In endodermal tissues, as described in our manuscript, KkvGFP did not exhibit polarised apical localisation and instead showed a diffuse pattern with some cortical enrichment. However, the presence of KkvGFP-containing vesicles still suggests that the protein is capable of exiting the ER also in these endodermal tissues.

We observed a different subcellular pattern when we overexpressed Chs2GFP. In tissues where Chs2 is not normally expressed (e.g., trachea, salivary gland, embryonic gut), we did not detect apical or membrane accumulation (see Fig. 5,S5, S6 and response to point 4 of Reviewer #1). Nor did we observe accumulation of Chs2GFP in intracellular vesicles. Instead, Chs2GFP showed strong colocalisation with an ER marker (see Fig. 5,S5, S6 and response to point 4 of Reviewer #1). In contrast, when overexpressed in PR cells, we detected apical enrichment (Fig 5A-D, S5A,B). This indicates that despite massive expression levels, Chs2 can exit the ER in particular tissues.

Taken together, our results strongly suggest that overexpressed Kkv can exit the ER in the different tissues analysed, whereas most Chs2GFP is retained in the ER in tissues other than PR cells. This correlates with the ability of overexpressed KkvGFP to polymerise chitin (either in intracellular puncta or deposited extracellularly depending on the presence of Exp/Reb) in all analysed tissues. Conversely, Chs2 was unable to polymerise chitin (either in intracellular puncta or extracellularly regardless of Exp/Reb presence) in tissues other than PR cells.

Nevertheless, we acknowledge that we cannot definitively conclude that all Chs2 protein is entirely retained in the ER. We have included this caveat in our revised manuscript (Chapter 6 and Discussion section).

Minor Comments: - No mention of Fig 3I in the results section and the order discussed in the results does not match the order in the figure.

We apologise for these inconsistencies. We have addressed this issue in the text, figure legend, and the image order in Figure 3 and Figure S3.

• In the results please provide some information on what the CRIMIC collection is and how it allows you to see Chs2 expression for non-experts.

We have addressed this point in chapter 5 in the revised version, and we now provide a more detailed explanation of the CRIMIC Chs2CR60212-TG4.0 allele.

Further details of this allele are also provided in our responses to points 5 and 9 of Reviewer 1.

Reviewer #2 (Significance (Required)):

Drosophila produce different types of chitinous structures that are required for either the exoskeleton of the animal or for proper gut function (peritrophic matrix). Additionally, most insects have two enzymes involved in the production of chitin and current data suggests that they have unique roles in producing either the exoskeleton or the peritrophic matrix. However, it is unclear whether their different functions are due to differences in cell type expression or differences in physiological activity of the enzymes. The authors exploit Drosophila to drive these 2 enzymes in different cell types that are known to produce the exoskeleton or the peritrophic matrix to determine whether they can functionally substitute mutant backgrounds. Their results give us a hint that these enzymes are not equivalent. What the authors were unable to address is why they are not equivalent. They hypothesise that the different physiological functions of the enzymes may be related to trafficking differences within their respective cell types. While this is an interesting hypothesis, the date are not really clear yet to make this conclusion.

This work will be of interest to anyone interested in chitinous structures in insects and the cell biology of chitin-related enzymes.

Literature

---

AHMAD, M., CHAUDHARY, S. U., AFZAL, A. J. & TARIQ, M. 2015. Starvation-Induced Dietary Behaviour in Drosophila melanogaster Larvae and Adults. Sci Rep, 5__,__ 14285.

DE GIORGIO, E., GIANNIOS, P., ESPINAS, M. L. & LLIMARGAS, M. 2023. A dynamic interplay between chitin synthase and the proteins Expansion/Rebuf reveals that chitin polymerisation and translocation are uncoupled in Drosophila. PLoS Biol, 21__,__ e3001978.

FUSS, B. & HOCH, M. 1998. Drosophila endoderm development requires a novel homeobox gene which is a target of Wingless and Dpp signalling. Mech Dev, 79__,__ 83-97.

FUSS, B., JOSTEN, F., FEIX, M. & HOCH, M. 2004. Cell movements controlled by the Notch signalling cascade during foregut development in Drosophila. Development, 131__,__ 1587-95.

HEGEDUS, D. D., TOPRAK, U. & ERLANDSON, M. 2019. Peritrophic matrix formation. J Insect Physiol, 117__,__ 103898.

KING, D. G. 1988. Cellular organization and peritrophic membrane formation in the cardia (proventriculus) of Drosophila melanogaster. J Morphol, 196__,__ 253-82.

LEBRETON, G. & CASANOVA, J. 2016. Ligand-binding and constitutive FGF receptors in single Drosophila tracheal cells: Implications for the role of FGF in collective migration. Dev Dyn, 245__,__ 372-8.

MOUSSIAN, B., LETIZIA, A., MARTINEZ-CORRALES, G., ROTSTEIN, B., CASALI, A. & LLIMARGAS, M. 2015. Deciphering the genetic programme triggering timely and spatially-regulated chitin deposition. PLoS Genet, 11__,__ e1004939.

NAKAGOSHI, H., HOSHI, M., NABESHIMA, Y. & MATSUZAKI, F. 1998. A novel homeobox gene mediates the Dpp signal to establish functional specificity within target cells. Genes Dev, 12__,__ 2724-34.

PETERS, W. 1992. Peritrophic Membranes, Springer Berlin, Heidelberg.

RIZKI, M. T. M. 1956. The secretory activity of the proventriculus of Drosophila melanogaster. Journal of Experimental Zoology, 131__,__ 203-221.

SANCHEZ, N. & RONCERO, C. 2022. Chitin Synthesis in Yeast: A Matter of Trafficking. Int J Mol Sci, 23.

ZHU, H., LUDINGTON, W. B. & SPRADLING, A. C. 2024. Cellular and molecular organization of the Drosophila foregut. Proc Natl Acad Sci U S A, 121__,__ e2318760121.

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

Drosophila have two different chitin synthase enzymes, Kkv and Chs2, and due to unique expression patterns and mutant phenotypes, it is relatively clear that they have different functions in producing either the cuticle-related chitin network (Kkv) or the chitin associated with the peritrophic matrix (PM). However, what is unknown is whether the different functions in making cuticle vs PM chitin is related to differences in cellular expression and/or enzyme properties within the cell. The authors exploit the genetic tractability of Drosophila and their ability to image cuticle vs PM chitin production to examine whether these 2 enzymes can substitute each other. They conclude that these two proteins are not equivalent in their capacity to generate chitin. The data are convincing; however, it is currently presented in a subjective fashion, which makes it difficult to interpret. Additionally, in my opinion there is some interpretation that requires softening or alternatively interpreted.

Major Comments:

• While the imaging is lovely, there are some things that are difficult to see in the figures. For example, the "continuous, thin and faint 'chitin' layer that lined the gut epithelium" is very difficult to visualise in the control images. Can they increase the contrast to help the reader appreciate this layer? This is particularly important as we are asked to appreciate a loss of this layer in the absence of Chs2.
• All the mutant analysis is presented subjectively. For example, the authors state that they "found a consistent difference of CBP staining when they compared the 'Chs2' escapers to the controls". How consistent is consistent? Can this be quantified? What is the penetrance of this phenotype? They say that the thin layer is absent in the midgut and the guts are thinner. Could they provide more concrete data?
• They state that Chs2 was able to restore accumulation of chitin in the PM of the proventriculus and the midgut. Please quantify. Additionally, does this restore the morphology of the guts (related to the comment above on the thinner guts in the absence of Chs2)?
• This may be beyond the scope of this paper, but I find it interesting that the PM chitin is deposited in the proventricular lumen. Yet it forms a thin layer that lines the entire midgut? Any idea how this presumably dense chitin network gets transported throughout the midgut to line the epithelium? I imagine that this is unlikely due to diffusion, especially if they see an even distribution across the midgut. Do they see any evidence of a graded lining (i.e. is it denser in the midgut towards the proventriculus and does this progressively decrease as you look through the midgut?)?
• The authors state that expression of kkv in tracheal cells of kkv mutants perfectly restores accumulation of chitin in the luminal filaments. Is this really 100% restoration? They also reference a paper here, which may have quantified this result.
• They ask whether Kkv overexpression in the proventriculus can rescue Chs2 mutants... and vice versa, whether Chs2 overexpression in ectodermal cells can rescue kkv mutants. They show that kkv overexpression leads to an intracellular accumulation of chitin in the proventriculus. However, Chs2 overexpression in the trachea did not lead to any accumulation of chitin in the cells. They tailored their experiments and the associated discussion to address the hypothesis that there is potentially some difference in trafficking of these components. However, another possibility, which they have not ruled out, is that the different ability of kkv and Chs2 to produce chitin inside cells of the proventriculus and ectoderm, respectively, is potentially related to different enzymatic activities and cofactors required for chitin formation in these different cell types. Is this another potential explanation for the differences that they observe?
• They co-express Chs2 and Reb and show that this does not lead to chitin production or secretion. In the discussion they conclude that Chs2 does not "seem to be dependent on 'Reb' activity". I think that this statement potentially needs softening. They show that Reb is not sufficient in to induce Chs2 chitin production in cells that do not normally make a PM. However, they do not show that it is not essential in cells that normally express Chs2 and make PM.
• They looked at the endogenous expression pattern of kkv and Chs2 and say that they found accumulation of Kkv in the proventriculus and no accumulation in the midgut. Siimilarly, they look at the expression of Chs2 and detect it in cells of the proventriculus. Are there markers of these different cell types that they could use to colocalize these enzymes?
• They overexpress Chs2 in cells of the midgut and see that it colocalises with an ER marker. They conclude that it is retained in the ER, which again, for them suggests that it has a trafficking problem in these cells. However, they are overexpressing it in these cells and this strong accumulation that they observe in the ER could simply be due to the massive expression levels. Additionally, they cannot conclude that it doesn't get out of the ER at all. They could be correct in thinking that there may be a trafficking issue, but this experiment does not conclusively show that Chs2 is entirely retained in the ER when expressed in ectopic tissues. I wonder if their interpretation needs softening or whether they should potentially address alternative hypotheses.

Minor Comments:

• No mention of Fig 3I in the results section and the order discussed in the results does not match the order in the figure.
• In the results please provide some information on what the CRIMIC collection is and how it allows you to see Chs2 expression for non-experts.

Significance

Drosophila produce different types of chitinous structures that are required for either the exoskeleton of the animal or for proper gut function (peritrophic matrix). Additionally, most insects have two enzymes involved in the production of chitin and current data suggests that they have unique roles in producing either the exoskeleton or the peritrophic matrix. However, it is unclear whether their different functions are due to differences in cell type expression or differences in physiological activity of the enzymes. The authors exploit Drosophila to drive these 2 enzymes in different cell types that are known to produce the exoskeleton or the peritrophic matrix to determine whether they can functionally substitute mutant backgrounds. Their results give us a hint that these enzymes are not equivalent. What the authors were unable to address is why they are not equivalent. They hypothesise that the different physiological functions of the enzymes may be related to trafficking differences within their respective cell types. While this is an interesting hypothesis, the date are not really clear yet to make this conclusion.

This work will be of interest to anyone interested in chitinous structures in insects and the cell biology of chitin-related enzymes.

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

Summary:

Chitin is a critical component of the extracellular matrix of arthropods and plays an essential role in the development and protection of insects. There are two chitin synthases in insects: Type A (exoskeletons) and Type B (for the peritrophic matrix in the gut). The study aims to investigate the specificity and mechanisms of the two chitin synthases in D. melanogaster and to clarify whether they are functionally interchangeable. Various genetic manipulations and fluorescence-based labeling were used to analyze the expression, localization, and function of Kkv and Chs2 in different tissues.

Chs2 is expressed in the PR cells of the proventriculus and is required for chitin deposition in the peritrophic matrix. Kkv can deposit chitin in ectodermal tissues but not in the peritrophic matrix, whereas Chs2 can deposit chitin in the peritrophic matrix but not in ectodermal tissues. The subcellular localization of chitin synthases is specific to the tissues in which they are expressed. Kkv localizes apically in ectodermal tissues, whereas Chs2 localizes apically in the PR cells of the proventriculus. Altogether, Kkv and Chs2 cannot replace each other. The specificity of chitin synthases in D. melanogaster relies on distinct cellular and molecular mechanisms, including intracellular transport pathways and the specific molecular machinery for chitin deposition. Congratulations on this incredible story and manuscript, which is straightforward and well-written. However, I have some comments that may help to improve it.

Major comments:

1. Funny thing: the Chs2 mutant larva shows a magenta staining below the chitin accumulation of the esophagus, which looks like a question mark in 1H but cannot be found in control. Is that trachea reaching the pv?
2. Also, though it is evident that the PM chitin is lost in Ch2 mutants, could it be that the region is disturbed and cells express somewhere else chitin? There are papers by Fuß and Hoch (e.g., Mech of Dev, 79, 1998; Josten, Fuß et al., Dev. Biol.267, 2004) using markers such as Dve, Fkh, Wg, Delta, and Notch, etc. for precisely marking the endodermal/ectodermal region in the embryonic foregut/proventriculus. It would be beneficial to show, along with chitin and Chs expression patterns, the ectoderm/endoderm cells. This is particularly important as the authors report endodermal expression of Chs2 in embryos but don't use co-markers of the endodermal cells.
3. The origin of midgut chitin accumulation is unclear. Chitin can come from yeast paster. Can the authors check kkv and chs2 mutants for food passage and test starving L1 larvae to detect chitin accumulation in the midgut without feeding them?
4. Subcellular localization assays require improved analysis, such as a co-marker for the apical membrane and statistical analysis with co-localization tools, showing the overlap at the membrane and intracellularly with membrane co-markers and KDEL.

Minor comments:

1. The authors used "L3 larval escapers." It would be interesting to know if the lack of Chs2 and the peritrophic matrix cause any physiological defects or lethality.
2. The order identifiers are missing for materials and antibodies, e.g., anti-GFP (Abcam), but Abcam provides several ant-GFP; which was used? Please provide order numbers that guarantee the repeatability for others.
3. Figure S5C, C', what marks GFP (blue) in the trachea? Maybe I have overlooked the description. What is UASsrcGFP? What is the origin of this line?
4. The authors claim that they validated the anti-Chs2 antibody. However, they show only that it recognizes a Cht2 epitope via ectopic expression. For more profound validation, immune staining is required in deletion mutants, upon knockdown, or upon expression of recombinant proteins, which is not shown.
5. The legend and text explaining Fig. 4 D-E' can be improved. The authors used the Crimic line, which is integrated into the third ("coding") intron. This orientation can lead to the expression of Gal4 and cause a truncated version of the protein (according to Flybase). Is Chs2 expression reduced in the crimic mutant? If the mutation causes expression of a truncated version, the Chs2 antibody may not be able to detect it as it recognizes a fragment between 1222 and 1383 aa? Also, I'm unsure whether the Chs2 antibody or GFP was used to detect expression in PR cells. The authors describe using Ch2CR60212>SrcGFP together with Chs2+ specific antibodies.

Referees cross-commenting I fully agree with the comments of Rev#2

Significance

The manuscript's strength and most important aspects are the genetic analysis, expression, and localization studies of the two Chitin synthases in Drosophila embryos and larvae. However, beyond this manuscript, the development of mechanistic details, such as interaction partners that trigger secretion and action at the apical membranes and the role of the coiled-coil domain, will be interesting.

The manuscript uses "first-class" genetics to describe the different roles of the two Chitin synthases in Drosophila, comparing ectodermal chitin (tracheal and epidermal chitin) with endodermal (midgut) chitin. Such a precise analysis has not been investigated before in insects. Therefore, the study deeply extends knowledge about the role of Chitin synthases in insects.

The audience will specialize in basic research in zoology, developmental biology, and cell biology regarding - how the different Chitin synthases produce chitin. Nevertheless, as chitin is relevant to material research and medical and immunological aspects, the manuscript will be fascinating beyond the specific field and thus for a broader audience.

I'm working on chitin in the tracheal system and epidermis in Drosophila.

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

Dear Editor,

Thank you for reviewing our article. We are happy to see that the reviewers are positive on our manuscript. We have tried to address nearly all their comments. Find below a point-by-point answer.

With best regards,

Bruno Lemaitre and Asya Dolgikh

Reviewer #1 (Evidence, reproducibility and clarity (Required)):

This work defines NimB1 protein as a PS binding bridging molecule but with a negative regulatory role in efferocytosis. Specifically, the authors demonstrate via a variety of genetic, cell biological, and other approaches that loss of NimB1 leads to Drosophila macrophages being more adherent to apoptotic targets and engulf them more robustly. The authors also nicely demonstrate that the function of NimB1 differs from NimB4, and the double mutant demonstrating PS-binding yet, distinct roles. Further, the authors show that NimB1 does not affect bacterial phagocytosis.

Overall, this is a well-done study. The authors have already done a very thorough job addressing the key points and I congratulate the authors.

My only minor comment is that the authors could try to make the comment better about whether or not such a 'negative regulatory' bridging molecules may exist in other species, and particularly mammals. If so, this is quite novel. The authors refer to CD47 but this is a membrane protein. The other minor comment is whether the authors ever tried express the PS binding domains as a fusion protein - this would provide a more direct evidence for the binding to PS (although the authors do competitive inhibition with Annexin V). This could be commented upon although testing this is not necessary if they have not already done so.

We greatly appreciate the reviewer’s positive feedback. In the revised manuscript, we have now included a more detailed discussion of mammalian proteins with analogous roles, specifically referencing Draper isoforms (I and II), the CD300 receptor family, and surfactant proteins A and B (see page 16).

Reviewer #1 (Significance (Required)):

The identification of the negative regulator bridging protein NimB1 is novel and could be broadly interesting to those studying efferocytosis.

Regarding the suggestion to overexpress just the putative PS-binding domain of NimB1, we agree this could strengthen the evidence for its PS-binding function. However, generating a new transgenic fly line would require significant additional time. Moreover, the presence of a PS-binding motif was also proposed in the recent study on Orion (Ji et al., 2023), which we have cited in our manuscript. The Orion binds PS through a conserved RRY motif. This motif is critical for Orion’s ability to directly interact with PS and facilitate its secretion. Mutagenesis experiments disrupting the RRY motif—specifically substituting arginine residues with alanines—abolished Orion’s PS-binding capacity, demonstrating the essential role of this sequence. Functional assays also validated that Orion competes with Annexin V, a well-established PS-binding protein, for access to PS-exposing surfaces (Ji et al., 2023).

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

Summary:

In this study, Dolgikh and colleagues propose a first investigation about the role of the drosophila Nimrod protein NimB1. Although the role of several members of the family in phagocytosis has been explored, the function of Nimrod type B proteins is less addressed. Within silico analysis, they first see a strong similarity between NimB1 and NimB4. They show that NimB1 is primarily expressed in phagocytes, and as NimB4 can bind phosphatidylserines (PS), leading to a possible shared role in efferocytosis. Using transgenic and null drosophila models, the authors then compare the impact of NimB1 overexpression or deficiency. They compare the effects shown to NimB4 and Draper deficient lines, as these two proteins were previously shown to play a role in efferocytosis. They propose that NimB1 is a secreted protein that binds apoptotic cells. They show that NimB1 deficiency changes the adhesion properties of macrophages. The major finding is that NimB1 delays the early stages of efferocytosis, contrary to NimB4 and Draper that on the contrary facilitate efferocytosis. In contrast, the authors propose that NimB1 increases the formation of phagosomes.

We appreciate the reviewer’s acknowledgment that our key discovery centered around NimB1 functioning as a negative regulator of efferocytosis. This finding highlights NimB1’s distinct role compared to NimB4 and Draper, which instead promote the process.

Major comments:

One of the major technical challenges here was to generate models to allow the detection of the protein in cellulo and in vivo. Although the results are convincing in transgenic lines NimB1 expression is driven by the endogenous promoter, one could still argue that the GFP tags would lead to changes in the localization of the protein.

We understand the concern regarding potential localization changes introduced by GFP tags. However, in previous studies, the same fosmid construct was applied to NimB4-sGFP, and produced a distinctly different expression pattern—NimB4-sGFP expression was more pronounced and clearly present in the glial cells in the brain (Petrignani et al, 2021: Figure EV1A). The fact that the NimB1-sGFP and NimB4-sGFP fosmids localized to different tissues suggests that possible any mis-localization changes due to the GFP tag do not override localization properties intrinsic to the proteins.

In line with the previous comment, to show that NimB1 is a secreted protein, the authors use an overexpression model. How to be sure, that overexpression itself does not lead to increased secretion, or shedding from the membrane?

The observation that uas-NimB1-RFP accumulates in the nephrocytes upon Lpp-Gal4 (fat body) expression, and the presence of a signal peptide suggests that this protein can be secreted.

We cannot exclude that in endogenous condition, NimB1, remains attached to hemocytes. We have confirmed that the Lpp driver is not expressed in nephrocytes.

Would an experiment with a control consisting in a known protein secreted by macrophages lead to the same staining pattern (positive control)? Could another methodology like a Western Blot on supernatants from hemocyte cell culture (over)expressing NimB1, with an anti-RFP staining, be envisaged?

We have already performed similar experiment with other secreted proteins such as NimB4-GFP (Petrignani et al., 2021: Figure: 1B). In the revised article, we have added Viking-RFP as a positive control of a secreted protein (Figure S1F). Figure S2 shows a Western blot with hemolymph extract. We detected NimB1-RFP at its expected molecular weight of 44 kDa, verifying that is present into the hemolymph (Supplementary Document S2 C).

It sems counterintuitive that phagocytes from Draper and NimB4 null mutants with defects in efferocytosis show increased load of apoptotic cells (Figure 6C and D in both unchallenged and injury condition). Do the authors have precedent data to cite going to the same direction? Are cell debris engulfed but not degraded efficiently?

The observation that Draper and NimB4 null mutants have an increased load of apoptotic cells has already been reported in the literature. Several studies have now shown that Draper is not always required for the initial uptake of apoptotic corpses but is critical for phagosome maturation (Meehan et al., 2016; Serizier et al., 2022; Serizier & McCall, 2017). In our article on NimB4 (Petrignani et al., 2021), we have previously shown that the accumulation of immature phagosomes that are not properly eliminated indirectly impairs the uptake of new apoptotic corpses. This explains why efferocytosis is then impaired only at late time points, when unresolved phagosomes have accumulated to the threshold that prevents further phagocytosis.

In Figure 6D it seems indeed that NimB4, NimB1/NimB4 and Draper mutants do not accumulate more apoptotic material upon injury. However, levels for NimB4 is close to the one obtained with NimB1 mutants. Is it statistically true? If yes, what could be the reason for this similarity? In any case, as some important conclusion relies on the comparison between UC and injury conditions, adequate statistics and representations could be proposed.

We thank the reviewer for this pertinent observation and the opportunity to clarify. In the unchallenged (UC) condition, NimB4sk2 and draperΔ5 mutants indeed exhibit significantly elevated levels of apoptotic cell (AC) content in macrophages compared to wild-type and NimB1 mutant genotypes (****p crimic and NimB1229/NimB1crimic* mutants show significantly lower levels in the UC condition, consistent with a role for NimB1 in early recognition or regulation of phagocytic initiation, not in corpse degradation.

In contrast, upon injury (90 minutes post-challenge) we observe a statistically significant increase in apoptotic material in NimB1 mutants compared to UC hemocytes of the same genotype (****p sk2 and draperΔ5* mutants between the UC and 90 min conditions (ns for NimB4). This is consistent with their known defect in corpse degradation, which results in saturation of phagocytic capacity at baseline, and an inability to respond further upon challenge with apoptotic cells.

While the absolute levels of apoptotic material in injured NimB1 and UC NimB4 mutants appear similar at first glance, statistical testing confirms that they are significantly different. NimB4 mutant macrophages retain apoptotic debris due to defective degradation, whereas NimB1 mutants have an increase in newly acquired apoptotic content due to enhanced uptake.

Additionally, NimB161, NimB4sk2 double mutants display a partial increase in apoptotic load upon injury (****p To directly address the reviewer’s suggestion, we have now recalculated and visualized key comparisons with appropriate statistical testing, as shown in Revision Figure 1. All statistical analyses were conducted using unpaired two-tailed Student’s t-tests. This additional figure allows clearer evaluation of genotype-specific differences at both baseline and post-injury conditions and supports our conclusions that NimB1 and NimB4 regulate distinct stages of phagocytosis. We have also clarified the text to better explain that both NimB4 and Draper mutants accumulate unresolved apoptotic material under baseline conditions, and do not accumulate further material upon challenge, due to a block in phagosome maturation.

Revisions Figure 1.

__Quantification of phagocytic events in wild-type and mutant macrophages under unchallenged and post-injury conditions __

(A) Comparison of phagocytic events per frame in w1118 (wild-type), NimB1crimic, NimB1229/NimB1crimic, NimB4sk2, NimB161,NimB4 sk2, and draperΔ5 larvae under unchallenged conditions (UC) and 90 minutes after injury (90 min). Data are presented as individual data points with means. Statistical significance was determined using Student's t-test (*P (B) Direct comparison of phagocytic events between NimB1crimic (red) and NimB4sk2 (gray), and between NimB1229/crimic (dark red) and NimB4sk2 (gray) under both unchallenged (UC) and post-injury (90 min) conditions.

The authors claim with analyses of Figure 8C and D, that NimB1 mutants show acidic vehicles normal in size and fluorescence intensity. However, statistical differences are still observed compared to control condition, which is also seen in representative images shown.

In Figure 8C and D, we provide two quantitative measures to clarify the size and intensity of acidic vesicles. First, we show that mean fluorescence in hemocytes is elevated for all NimB and draper mutants compared to wild type, indicating an overall increase in internalized material. However, we also quantified the number of vesicles per hemocyte and found that NimB1 mutants exhibit significantly more vesicles. Despite this increase, the representative images do not show an obvious enlargement of individual vesicles, suggesting that while more material is being taken up, the vesicles themselves are not enlarged. The enlarged vesicles in case of NimB4 or draper mutant would result from the unresolved cargo (Petrignani et al., 2021). This distinction underscores that higher fluorescence values reflect increased cargo internalization, rather than the larger vesicular structures that result from impaired degradation as in NimB4 or draper mutants.

Minor comments:

In figure 2D, what allows to say the expression is restricted in macrophages? Is it the colocalization with SIMU being a macrophage-specific marker?

In Figure 2D, we relied on SIMU as a macrophage-specific marker in Drosophila embryos to determine that NimB1 expression is restricted to macrophages. Previous research has demonstrated that SIMU is predominantly expressed in embryonic macrophages (where it is essential for apoptotic cell clearance) (Kurant et al., 2008; Roddie et al., 2019). Consequently, the colocalization of NimB1 signal with SIMU-positive cells strongly indicates that NimB1 is confined to macrophages during this developmental stage.

In figure S3B and C, it appears that double NimB1/NimB4 mutants exhibit less spreading than single ones (especially NimB4). Is it the case (statistical significance). If yes what could be the explanation?

Yes, the double NimB1, NimB4 mutants exhibit higher number of hemocytes and significantly reduced cell spreading compared to single mutants. The phenotype is similar to NimC1, eater double mutants (Melcarne et al., 2019) which also show higher number of hemocytes, reduced cell spreading and also diminished capacity to phagocytose apoptotic cells (and, in the case of NimC1, Eater, bacteria as well) (Melcarne et al., 2019). A likely explanation lies in impaired membrane remodeling critical for pseudopod extension and phagosome formation. Studies have shown that defects in actin polymerization or membrane tension can hinder pseudopod extension, reducing phagocytic efficiency (Lee et al., 2007; Masters et al., 2013). Same for the decreased ability of these mutants to form filopodium, a process essential for effective target engagement and engulfment. Filopodia play a significant role in capturing particles and directing them toward the macrophage body for engulfment (Horsthemke et al., 2017). Disruptions in these pathways lead to reduced phagocytic efficiency and a more rounded macrophage morphology, as the cells fail to spread properly (Horsthemke et al., 2017; Lillico et al., 2018). Other than these general observations, we do not have an explanation as to why NimB1, NimB4 double mutants specifically show a higher number of hemocytes and reduced cell spreading.

Several graphs are identical between figure 4 and S4. It is probably not useful and complicates reading.

We agree that duplicating these graphs complicates the presentation. Therefore, we have removed the redundant graphs in the supplementary materials, ensuring the data are shown only once to maintain clarity and ease of reading

As TEM images shown in Figure 8B do not lead to quantitative data, I would put it as supplementary file.

We agree that the TEM images in Figure 8B do not provide strictly quantitative data. To streamline the main manuscript, we have relocated these images to the supplementary section in the revised version

Reviewer #2 (Significance (Required)):

This study uses several approaches and models to address the role of NimB1 in efferocytosis. Both In Vitro and In Vivo approaches are proposed. They give insight into the role of this protein with unknown function so far. Some statistical analysis could be performed to improve the clarity of conclusions. One of the important aspects is the secreted nature of NimB1.However, additional approaches could be proposed to confirm this.

Basic immunologists and cell biologists would be interested in reading this article that highlights the delicate equilibrium between pro and anti-efferocytosis molecules.

I am an immunologist/cell biologist with expertise in lysosomal catabolism. As I work on mouse models or Human samples, my mastering of drosophila as a model is limited.

We thank the reviewer for the positive evaluation of our work. In this revision, we have added further detail to clarify the properties of NimB1 as a secreted protein and strengthen our data presentation. By providing additional clarity on methods and interpretations, we hope immunologists and cell biologists—including those who do not routinely work with Drosophila—will find our findings more accessible.

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

This paper investigates the role of NimB1, a secreted member of the Nimrod family in Drosophila, in the process of efferocytosis, the clearance of apoptotic cells by macrophages. Previous studies have identified NimB4, another secreted Nimrod protein, as a positive regulator of efferocytosis, enhancing both apoptotic cell binding and phagosome maturation. In contrast, the authors propose that NimB1 functions as a negative regulator, slowing down the early stages of apoptotic cell binding and internalization. This regulatory balance is suggested to fine-tune efferocytosis to maintain homeostasis.

The primary aim of this study was to characterize the function of NimB1 to better understand the roles of proteins within the NimB family.

This study identifies a novel function for NimB1 in modulating the early stages of efferocytosis, adding to our understanding of how Nimrod proteins fine-tune apoptotic cell clearance. The authors establish a clear phenotypic contrast between NimB1 and NimB4, which provides a compelling framework for understanding how positive and negative regulators coordinate phagocytosis. It also highlights the multiple roles of the secreted members of the Nimrod scavenger receptor family, which have remained so far poorly investigated.

This is an interesting study that could be strengthened by additional validation and broader experimental support. As the authors point out in the discussion, it is known that PS bridging molecules contribute to phagocytosis and that the contribution of positive and negative players finely tune phagocytosis in flies and mammals. Clarifying the mode of action of NimB1 in those processes would higher the impact of this interesting piece of work. For example, does NimB1 interact with NimB4 and if so, what is the role of this interaction? How does NimB1 integrate in the signaling cascade that allows scavenger receptors to bind PS? Does it act similar to Orion by enhancing the PS binding of a scavenger receptor?

Key Findings • NimB1 and NimB4 are structurally similar, as supported by AlphaFold2 modeling, suggesting functional relatedness. • NimB1 is expressed in macrophages, secreted into the hemolymph, and binds apoptotic cells in a phosphatidylserine (PS)-dependent manner. • NimB1 is induced by challenge. • NimB1 mutants display a hyper-phagocytic phenotype, with faster recognition and internalization of apoptotic cells. • NimB1 loss enhances macrophage adhesion and actin remodeling, while bacterial phagocytosis remains unaffected, suggesting a specific role in apoptotic clearance. • NimB1 acts early in the phagocytic process, while NimB4 functions at later stages, particularly in phagosome maturation.

We thank the reviewer for their positive assessment and are pleased that our findings identify NimB1 as a novel secreted negative regulator of efferocytosis, underscoring a greater level of regulatory complexity in apoptotic cell clearance.

Unfortunately, attempts to produce functional NimB1 protein were not successful, limiting our ability to address some of the reviewer’s suggestions experimentally. Despite these challenges, the evidence we present—particularly from our genetic assays—clearly indicates that NimB1 exerts an inhibitory influence during the early steps of apoptotic cell binding, distinguishing it from the late-stage promoting function of NimB4.

Major comments:

Figure 1: AlphaFold is a valuable tool for generating hypotheses, however predicted structures should not be presented as definitive evidence of similarity, particularly without complementary experimental validation. This section would be stronger if the structural predictions were explicitly framed as predictions. In the absence of such data, the interpretation should be toned down.

We agree with the reviewer and we have now framed our observation as prediction and toned down our interpretation. We also note that the similarities between NimB4 and NimB1 are also underlined by the phylogenetic analysis and expression pattern.

Figure 2DE: Given its basal level in homeostatic conditions, it would have been useful to look at the NimB1-GFP upon challenge. Also, the authors show only a single larval macrophage with no comparison point. To strengthen this result, the authors could include another protein quantification method, such as western blotting. Alternatively, labelling of NimB1>UASmRFP in embryo that present the highest expression levels would also strengthen this result.

Unfortunately, we cannot currently perform additional experiments on embryos within the scope of this project because those experiments were performed by our collaborators in Haifa (Estee Kurant Lab). Repeating them would require sending the lines to their lab and accommodating their experimental schedule and manpower constraints.

In supplementary Figure S1F: the authors overexpress NimB1-RFP using the fat body driver Lpp-Gal4 and show larvae with RFP in the nephrocyte. Could filet preparations be shown? Could the authors present evidence that the Lpp driver is not expressed in the nephrocytes (or refer to literature)?

The Lpp-Gal80 driver is described as fat body-specific and has been used to manipulate gene expression in the fat body in many other studies. We have checked Lpp-Gal80>UAS-GFP expression in larvae and did not observe expression in larval nephrocytes. The whole live larvae were observed under the microscope with no prior filet preparations. To provide the evidence that Lpp is not expressed in the nephrocytes we are providing the images of the whole larvae expressing GPF from the Lpp, as per genotype: Lgg-Gal80>UAS-GFP (see below, Revisions Figure 2).

---

Revisions Figure 2.

__Expression pattern of Lpp-Gal80>UAS-GFP in Drosophila larvae __

Representative fluorescence microscopy images showing GFP expression driven by the Lpp-Gal80 system in Drosophila larvae. The images display dorsal (top) and ventral (bottom) views of the same larva, demonstrating the pattern of expression throughout the fat body tissue. Green fluorescence indicates cells expressing the GFP reporter under the control of the Lpp promoter, which is predominantly active in the larval fat body.

The results on the increased number of hemocytes observed in the double NimB1, NimB4 mutant animals (Figure S3A) remains not only disconnected from the rest of the data but also unexplained. Providing a mechanistic view may require a significant amount of work that may indicate an additional role of the two NimBs but will not add to our understanding of the role of NimB1 in phagocytosis. Nevertheless, it would be at least useful to know whether in the double mutant the lymph gland is still intact, as its precocious histolysis could account for the elevated number of hemocytes. If that were the case, that could indicate that lacking the two NimBs triggers an inflammatory state that affects the lymph gland, meaning that the pathway controlling phagocytosis also has a systemic impact on development. When checking the representative Figure S4D, it seems that very large cells are present in the double mutants, even larger than in the single mutants. These could be (pre)lamellocytes, which constitute activated hemocytes, known to impact the status of the lymph gland. If the enhanced number of hemocytes does not depend on lymph gland histolysis, a simple immunolabeling with the anti-PH3 antibody would assess the proliferative phenotype of the double mutant hemocytes. At least this piece of data would provide a better explanation for the observed phenotype.

We thank the reviewer for this interesting comment. We cannot explain why NimB1, NimB4 double mutants have more hemocytes. It is unclear to us if this is a secondary consequence of defects in efferocytosis or linked to another function of these two NimBs, such as a role in adhesion. We did look at the lymph gland and our preliminary observations suggest that NimB1, NimB4 double mutants have an easily ruptured or fragile lymph gland, which could explain the higher number and the roundish shape of hemocytes in circulation as proposed by the reviewer. Lacking expertise on lymph gland, we prefer not to include this data, as they are not central to the main message of this article on role of NimB1 on efferocytosis. We have also noted the presence of lamellocytes in unchallenged NimB1, NimB4 double mutant larvae, as well as excessive lamellocyte production compared to controls upon clean injury (see below, Revisions Figure 3). We have mentioned the presence of lamellocytes in NimB1, NimB4 double mutants in the revised version. We prefer not include this new data directly in the article because this not central to the main message of the article.

---

__Revisions Figure 3. __

A.

B.

Lamellocyte recruitment following a clean injury in L3 Drosophila larvae:

(A) Quantification of lamellocytes per 50 frames of x63 microscopy lens in w1118 (wild-type), NimB1crimic, NimB4sk2, NimB161, NimB4sk2, and draperΔ5 larvae under unchallenged conditions (UC) and 3 hours after clean injury (3h). Arrowheads indicate lamellocytes.

(B) Representative confocal microscopy images of hemocytes isolated from challenged NimB161, NimB4sk2 larvae. Cells were fixed and stained with Phalloidin (green) to label F-actin and DAPI (blue) to visualize nuclei. The smaller inset (40x magnification) shows a detailed view of individual lamellocytes with characteristic morphology, while the larger field (20x magnification) displays the overall view on the hemocytes. Scale bar = 50 μm.

Figure 6: The connection between the ex-vivo (Figure 5) and in vivo (Figure 6) assays should be clarified. In the first type of assay, the lack of NimB4 results in reduced internalization (while lack of NimB1 enhances it). In the in vivo assay, more fragments are seen within the cell (hence internalized), using the NimB4 mutant. Also, in the ex-vivo assay, the lack of NimB1 does not affect the first steps ('attachment' and 'membrane'), while NimB4 does, yet it is proposed that NimB1 acts in the early steps (page 11-12). In that case, wouldn't we expect the double mutant NimB1 NmB4 to have the NimB1 phenotype?

The apparent discrepancy between our ex vivo and in vivo assays reflects the different methodologies and what each assay measures. In the ex vivo assay (Figure 4), we add exogenous fluorescently-labeled apoptotic cells to measure new engulfment events. Here, NimB4 mutant macrophages show reduced phagocytic index because they are already saturated with unresolved phagosomes, limiting their capacity to uptake additional corpses, as previously described by (Petrignani et al., 2021). This reduced uptake capacity is reflected in the decreased phagocytic index observed.

In contrast, our in vivo assay (Figure 6) uses DAPI staining to visualize all internalized material, including previously engulfed debris. As expected, we observe accumulation of DAPI signals in NimB4 mutant macrophages under unchallenged conditions, reflecting their inability to process and clear phagosomes rather than enhanced engulfment. This phenotype highlights the role of NimB4 in phagosome maturation rather than initial uptake.

Regarding the role of NimB1 in early phagocytic steps, while attachment and membrane measurements in the ex vivo assay don't show significant differences in NimBcrimic mutants individually, our other experiments demonstrate that NimB1 functions as a negative regulator during early recognition phases. The predominance of the NimB4 phenotype in the NimB1crimic, NimB4 double mutant parallels observations in draper mutants, where double mutants lacking both Draper I (positive regulator) and Draper II (negative regulator) display the Draper I phenotype (Logan et al., 2012). This suggests that phagosome maturation defects (the NimB4 phenotype) present a more severe bottleneck in the phagocytic process than enhanced early uptake (the NimB1crimic phenotype), explaining why the double mutant primarily exhibits accumulation of unresolved phagosomes rather than accelerated uptake. We have re-written this part of the article to clarify these points (see page 11).

Figure 8A: a definition of the phagocytic cup mentioned in the text (page 12, 2nd paragraph) as well as the homogenization of the scale bars in Figure 8A would clarify the interpretation of Figure 8A. The structures shown for w1118 seem quite distant from the structures highlighted for NimB1crimic.

According to reviewer 2, we have now moved this figure to the supplement. The reviewer is correct and we have modified the associated text to clarify the interpretation of the images (see page 12-13).

The same scale should be used across different panels in Figure 8. This is particularly important since the authors mention the size of the lysotracker vesicles to conclude on their levels of maturity. This data and conclusions would be strengthened by a quantification of the vacuole sizes and the combination with markers of phagosome/lysosome maturation levels. It would help disentangling the complementary roles of NimB1 and NimB4.

The scale bar has been homogenized.

Minor comments:

Figure 2BC: is there a particular reason to shift from Rp49 to Rpl32 as normalizing gene in Figure 2B and C? This prevents the comparison of NimB1 expression levels across the different graphs.

We thank the reviewer for highlighting this point. We changed the housekeeping gene from Rp49 to RPL32 in Figure 2C to unify the normalization strategy across all experiments and allow comparisons throughout the manuscript.

Page 9, 2nd paragraph and Figure S3C: the authors mention "Actin structure revealed an increased ratio of filopodia to lamellipodia across all mutants". A clear definition of the parameters defining filopodia and lamellipodia is required to fully appreciate the meaning of the ratio.

We thank the reviewer for the comment. To address this comment, we have included a clear definition of the parameters used to distinguish filopodia and lamellipodia on page 9. In particular, in the revised version we now specify that filopodia were defined as thin, spike-like actin-rich protrusions, while lamellipodia were defined as broad, sheet-like structures at the cell periphery. These criteria were applied consistently for quantification.

Figure S5B: a bar is missing in the right graph (% of cells containing AC, NimB1>UAS-NimB1-RFP). Page 10 2nd paragraph. The authors mention "draper mutants displayed impaired apoptotic cell binding and engulfment" referring to Figure 4. Figure S4 provide a more convincing illustration of this statement, since the decreased phagocytic index in Drpr KO is mostly due to less cells phagocytosing and not less material phagocytosed.

We thank the reviewer for the careful examination. In Figure S5B, the missing bar was due to its color being too close to the background color, making it difficult to distinguish. We have now corrected this by adjusting the color to ensure it is clearly visible.

Regarding the comment on page 10, we agree that Figure S4 more clearly illustrates the impaired apoptotic cell binding and engulfment observed in draper mutants, particularly through the reduced percentage of hemocytes engaging in phagocytosis. We have now clarified the statement in the text to ensure consistency and to guide the reader appropriately to Figure S4 (10).

Figure 6: not easy to distinguish the DAPI labelling relative to the nucleus vs. that of apoptotic fragments.

This is a good point. We have changed the images for clearer demonstration of the DAPI labelling. See Figure 6.

Figure 7B: the number of cells used to generate the violin plot should be indicated in the legend or the method section.

We have mentioned the number of cells used in the quantification (n-50 per genotype) in the figure legend.

A schematic figure recapitulating the data would help

We have added a schematic figure recapitulating the data. See Figure 9 with associated text.

Page 11 last line: homeostatic rather than hemostatic.

Thank you for this comment. We have changed it.

Reviewer #3 (Significance (Required)):

This study identifies a novel function for NimB1 in modulating the early stages of efferocytosis, adding to our understanding of how Nimrod proteins fine-tune apoptotic cell clearance. The authors establish a clear phenotypic contrast between NimB1 and NimB4, which provides a compelling framework for understanding how positive and negative regulators coordinate phagocytosis. It also highlights the multiple roles of the secreted members of the Nimrod scavenger receptor family, which have remained so far poorly investigated.

This is an interesting study that could be strengthened by additional validation and broader experimental support. As the authors point out in the discussion, it is known that PS bridging molecules contribute to phagocytosis and that the contribution of positive and negative players finally tune phagocytosis in flies and mammals. Clarifying the mode of action of NimB1 in those processes would higher the impact of this interesting piece of work. For example, does NimB1 interact with NimB4 and if so, what is the role of this interaction? How does NimB1 integrate in the signaling cascade that allows scavenger receptors to bind PS? Does it act similar to Orion by enhancing the PS binding of a scavenger receptor?

We thank the reviewer for the insightful comments and suggestions. Indeed, understanding the mode of action of NimB1 in the regulation of efferocytosis would significantly strengthen the impact of our findings. Our data, supported by structural and phylogenetic analyses, indicate that NimB1 and NimB4 share a conserved phosphatidylserine (PS)-binding motif, suggesting that these proteins may interact functionally. Preliminary biochemical observations, together with structural predictions, raise the possibility of a direct or indirect interaction between NimB1 and NimB4, although this remains to be experimentally confirmed.

Our observations from NimB1 and NimB4 double mutants reveal that the phenotype closely resembles that of NimB4 single mutants, indicating that NimB4 plays a dominant role in the downstream maturation steps of phagosome processing. These findings are consistent with a model in which NimB1 may modulate early phagocytic uptake, possibly by competing with NimB4 for PS binding or by limiting NimB4 accessibility to apoptotic cells, thereby fine-tuning the rate of efferocytosis.

Regarding the integration into the signaling cascade, while NimB1 and Orion both recognize PS, our data suggest that they function through distinct mechanisms. Orion enhances PS binding to Draper receptor isoforms to promote apoptotic corpse recognition. In contrast, NimB1 appears to act as an inhibitory modulator, potentially masking PS or limiting receptor engagement, thus slowing the phagocytic response. Further functional studies, including receptor-binding assays, will be important to determine whether NimB1 acts by altering receptor-ligand interactions or through a different regulatory pathway.

Future experiments investigating the potential direct interactions between NimB1 and NimB4, their respective affinities for PS, and their influence on phagocytic receptor dynamics will be necessary to better understand NimB1’s precise mode of action. Such studies will help clarify how secreted regulators fine-tune efferocytosis in Drosophila and may offer broader insights into conserved principles of phagocytic regulation across species.

__ __

List of References:

Horsthemke, M., Bachg, A. C., Groll, K., Moyzio, S., Müther, B., Hemkemeyer, S. A., Wedlich-Söldner, R., Sixt, M., Tacke, S., Bähler, M., & Hanley, P. J. (2017). Multiple roles of filopodial dynamics in particle capture and phagocytosis and phenotypes of Cdc42 and Myo10 deletion. The Journal of Biological Chemistry, 292(17), 7258–7273. https://doi.org/10.1074/jbc.M116.766923

Ji, H., Wang, B., Shen, Y., Labib, D., Lei, J., Chen, X., Sapar, M., Boulanger, A., Dura, J.-M., & Han, C. (2023). The Drosophila chemokine–like Orion bridges phosphatidylserine and Draper in phagocytosis of neurons. Proceedings of the National Academy of Sciences, 120(24), e2303392120. https://doi.org/10.1073/pnas.2303392120

Kurant, E., Axelrod, S., Leaman, D., & Gaul, U. (2008). Six-Microns-Under Acts Upstream of Draper in the Glial Phagocytosis of Apoptotic Neurons. Cell, 133(3), 498–509. https://doi.org/10.1016/j.cell.2008.02.052

Lee, W. L., Mason, D., Schreiber, A. D., & Grinstein, S. (2007). Quantitative Analysis of Membrane Remodeling at the Phagocytic Cup. Molecular Biology of the Cell, 18(8), 2883–2892. https://doi.org/10.1091/mbc.E06-05-0450

Lillico, D. M. E., Pemberton, J. G., & Stafford, J. L. (2018). Selective Regulation of Cytoskeletal Dynamics and Filopodia Formation by Teleost Leukocyte Immune-Type Receptors Differentially Contributes to Target Capture During the Phagocytic Process. Frontiers in Immunology, 9. https://doi.org/10.3389/fimmu.2018.01144

Masters, T. A., Pontes, B., Viasnoff, V., Li, Y., & Gauthier, N. C. (2013). Plasma membrane tension orchestrates membrane trafficking, cytoskeletal remodeling, and biochemical signaling during phagocytosis. Proceedings of the National Academy of Sciences, 110(29), 11875–11880. https://doi.org/10.1073/pnas.1301766110

Meehan, T. L., Joudi, T. F., Timmons, A. K., Taylor, J. D., Habib, C. S., Peterson, J. S., Emmanuel, S., Franc, N. C., & McCall, K. (2016). Components of the Engulfment Machinery Have Distinct Roles in Corpse Processing. PLOS ONE, 11(6), e0158217. https://doi.org/10.1371/journal.pone.0158217

Melcarne, C., Ramond, E., Dudzic, J., Bretscher, A. J., Kurucz, É., Andó, I., & Lemaitre, B. (2019). Two Nimrod receptors, NimC1 and Eater, synergistically contribute to bacterial phagocytosis in Drosophila melanogaster. The FEBS Journal, 286(14), 2670–2691. https://doi.org/10.1111/febs.14857

Petrignani, B., Rommelaere, S., Hakim-Mishnaevski, K., Masson, F., Ramond, E., Hilu-Dadia, R., Poidevin, M., Kondo, S., Kurant, E., & Lemaitre, B. (2021). A secreted factor NimrodB4 promotes the elimination of apoptotic corpses by phagocytes in Drosophila. EMBO Reports, 22(9), e52262. https://doi.org/10.15252/embr.202052262

Roddie, H. G., Armitage, E. L., Coates, J. A., Johnston, S. A., & Evans, I. R. (2019). Simu-dependent clearance of dying cells regulates macrophage function and inflammation resolution. PLoS Biology, 17(5), e2006741. https://doi.org/10.1371/journal.pbio.2006741

Serizier, S. B., & McCall, K. (2017). Scrambled Eggs: Apoptotic Cell Clearance by Non-Professional Phagocytes in the Drosophila Ovary. Frontiers in Immunology, 8, 1642. https://doi.org/10.3389/fimmu.2017.01642

Serizier, S. B., Peterson, J. S., & McCall, K. (2022). Non-autonomous cell death induced by the Draper phagocytosis receptor requires signaling through the JNK and SRC pathways. Journal of Cell Science, 135(20), jcs250134. https://doi.org/10.1242/jcs.250134

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

This paper investigates the role of NimB1, a secreted member of the Nimrod family in Drosophila, in the process of efferocytosis, the clearance of apoptotic cells by macrophages. Previous studies have identified NimB4, another secreted Nimrod protein, as a positive regulator of efferocytosis, enhancing both apoptotic cell binding and phagosome maturation. In contrast, the authors propose that NimB1 functions as a negative regulator, slowing down the early stages of apoptotic cell binding and internalization. This regulatory balance is suggested to fine-tune efferocytosis to maintain homeostasis.

The primary aim of this study was to characterize the function of NimB1 to better understand the roles of proteins within the NimB family.

This study identifies a novel function for NimB1 in modulating the early stages of efferocytosis, adding to our understanding of how Nimrod proteins fine-tune apoptotic cell clearance. The authors establish a clear phenotypic contrast between NimB1 and NimB4, which provides a compelling framework for understanding how positive and negative regulators coordinate phagocytosis. It also highlights the multiple roles of the secreted members of the Nimrod scavenger receptor family, which have remained so far poorly investigated.

This is an interesting study that could be strengthened by additional validation and broader experimental support. As the authors point out in the discussion, it is known that PS bridging molecules contribute to phagocytosis and that the contribution of positive and negative players finely tune phagocytosis in flies and mammals. Clarifying the mode of action of NimB1 in those processes would higher the impact of this interesting piece of work. For example, does NimB1 interact with NimB4 and if so, what is the role of this interaction? How does NimB1 integrate in the signaling cascade that allows scavenger receptors to bind PS? Does it act similar to Orion by enhancing the PS binding of a scavenger receptor ?

Key Findings

• NimB1 and NimB4 are structurally similar, as supported by AlphaFold2 modeling, suggesting functional relatedness.
• NimB1 is expressed in macrophages, secreted into the hemolymph, and binds apoptotic cells in a phosphatidylserine (PS)-dependent manner.
• NimB1 is induced by challenge.
• NimB1 mutants display a hyper-phagocytic phenotype, with faster recognition and internalization of apoptotic cells.
• NimB1 loss enhances macrophage adhesion and actin remodeling, while bacterial phagocytosis remains unaffected, suggesting a specific role in apoptotic clearance.
• NimB1 acts early in the phagocytic process, while NimB4 functions at later stages, particularly in phagosome maturation.

Major comments

1. Figure 1: AlphaFold is a valuable tool for generating hypotheses, however predicted structures should not be presented as definitive evidence of similarity, particularly without complementary experimental validation. This section would be stronger if the structural predictions were explicitly framed as predictions. In the absence of such data, the interpretation should be toned down.
2. Figure 2DE : Given its basal level in homeostatic conditions, it would have been useful to look at the NimB1-GFP upon challenge. Also, the authors show only a single larval macrophage with no comparison point. To strengthen this result, the authors could include another protein quantification method, such as western blotting. Alternatively, labelling of NimB1>UASmRFP in embryo that present the highest expression levels would also strengthen this result.
3. In supplementary Figure S1F : the authors overexpress NimB1-RFP using the fat body driver Lpp-Gal4 and show larvae with RFP in the nephrocyte. Could filet preparations be shown? Could the authors present evidence that the Lpp driver is not expressed in the nephrocytes (or refer to literature)?
4. The results on the increased number of hemocytes observed in the double NimB1, NimB4 mutant animals (Figure S3A) remains not only disconnected from the rest of the data but also unexplained. Providing a mechanistic view may require a significant amount of work that may indicate an additional role of the two NimBs but will not add to our understanding of the role of NimB1 in phagocytosis. Nevertheless, it would be at least useful to know whether in the double mutant the lymph gland is still intact, as its precocious histolysis could account for the elevated number of hemocytes. If that were the case, that could indicate that lacking the two NimBs triggers an inflammatory state that affects the lymph gland, meaning that the pathway controlling phagocytosis also has a systemic impact on development. When checking the representative Figure S4D, it seems that very large cells are present in the double mutants, even larger than in the single mutants. These could be (pre)lamellocytes, which constitute activated hemocytes, known to impact the status of the lymph gland. If the enhanced number of hemocytes does not depend on lymph gland histolysis, a simple immunolabeling with the anti-PH3 antibody would assess the proliferative phenotype of the double mutant hemocytes. At least this piece of data would provide a better explanation for the observed phenotype.
5. Figure 6: The connection between the ex-vivo (Figure 5) and in vivo (Figure 6) assays should be clarified. In the first type of assay, the lack of NimB4 results in reduced internalization (while lack of NimB1 enhances it). In the in vivo assay, more fragments are seen within the cell (hence internalized), using the NimB4 mutant. Also, in the ex-vivo assay, the lack of NimB1 does not affect the first steps ('attachment' and 'membrane'), while NimB4 does, yet it is proposed that NimB1 acts in the early steps (page 11-12). In that case, wouldn't we expect the double mutant NimB1 NmB4 to have the NimB1 phenotype?
6. Figure 8A : a definition of the phagocytic cup mentioned in the text (page 12, 2nd paragraph) as well as the homogenization of the scale bars in Figure 8A would clarify the interpretation of Figure 8A. The structures shown for w1118 seem quite distant from the structures highlighted for NimB1crimic.
7. The same scale should be used across different panels in Figure 8. This is particularly important since the authors mention the size of the lysotracker vesicles to conclude on their levels of maturity. This data and conclusions would be strengthened by a quantification of the vacuole sizes and the combination with markers of phagosome/lysosome maturation levels. It would help disentangling the complementary roles of NimB1 and NimB4.

Minor comments:

Figure 2BC : is there a particular reason to shift from Rp49 to Rpl32 as normalizing gene in Figure 2B and C? This prevents the comparison of NimB1 expression levels across the different graphs. Page 9, 2nd paragraph and Figure S3C: the authors mention "Actin structure revealed an increased ratio of filopodia to lamellipodia across all mutants". A clear definition of the parameters defining filopodia and lamellipodia is required to fully appreciate the meaning of the ratio. Figure S5B: a bar is missing in the right graph (% of cells containing AC, NimB1>UAS-NimB1-RFP). Page 10 2nd paragraph. The authors mention "draper mutants displayed impaired apoptotic cell binding and engulfment" referring to Figure 4. Figure S4 provide a more convincing illustration of this statement, since the decreased phagocytic index in Drpr KO is mostly due to less cells phagocytosing and not less material phagocytosed. Figure 6: not easy to distinguish the DAPI labelling relative to the nucleus vs. that of apoptotic fragments. Figure 7B: the number of cells used to generate the violin plot should be indicated in the legend or the method section. A schematic figure recapitulating the data would help. Page 11 last line: homeostatic rather than hemostatic.

Significance

This study identifies a novel function for NimB1 in modulating the early stages of efferocytosis, adding to our understanding of how Nimrod proteins fine-tune apoptotic cell clearance. The authors establish a clear phenotypic contrast between NimB1 and NimB4, which provides a compelling framework for understanding how positive and negative regulators coordinate phagocytosis. It also highlights the multiple roles of the secreted members of the Nimrod scavenger receptor family, which have remained so far poorly investigated.

This is an interesting study that could be strengthened by additional validation and broader experimental support. As the authors point out in the discussion, it is known that PS bridging molecules contribute to phagocytosis and that the contribution of positive and negative players finely tune phagocytosis in flies and mammals. Clarifying the mode of action of NimB1 in those processes would higher the impact of this interesting piece of work. For example, does NimB1 interact with NimB4 and if so, what is the role of this interaction? How does NimB1 integrate in the signaling cascade that allows scavenger receptors to bind PS? Does it act similar to Orion by enhancing the PS binding of a scavenger receptor ?

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

Summary:

In this study, Dolgikh and colleagues propose a first investigation about the role of the drosophila Nimrod protein NimB1. Although the role of several members of the family in phagocytosis has been explored, the function of Nimrod type B proteins is less addressed. With in silico analysis, they first see a strong similarity between NimB1 and NimB4. They show that NimB1 is primarily expressed in phagocytes, and as NimB4 can bind phosphatidylserines (PS), leading to a possible shared role in efferocytosis. Using transgenic and null drosophila models, the authors then compare the impact of NimB1 overexpression or deficiency. They compare the effects shown to NimB4 and Draper deficient lines, as these two proteins were previously shown to play a role in efferocytosis. They propose that NimB1 is a secreted protein that binds apoptotic cells. They show that NimB1 deficiency changes the adhesion properties of macrophages. The major finding is that NimB1 delays the early stages of efferocytosis, contrary to NimB4 and Draper that on the contrary facilitate efferocytosis. In contrast, the authors propose that NimB1 increases the formation of phagosomes.

Major comments:

• One of the major technical challenges here was to generate models to allow the detection of the protein in cellulo and in vivo. Although the results are convincing in transgenic lines NimB1 expression is driven by the endogenous promoter, one could still argue that the GFP tags would lead to changes in the localization of the protein.
• In line with the previous comment, to show that NimB1 is a secreted protein, the authors use an overexpression model. How to be sure, that overexpression itself does not lead to increased secretion, or shedding from the membrane ?
• Would an experiment with a control consisting in a known protein secreted by macrophages lead to the same staining pattern (positive control)? Could another methodology like a Western Blot on supernatants from hemocyte cell culture (over)expressing NimB1, with an anti-RFP staining, be envisaged?
• It sems counterintuitive that phagocytes from Draper and NimB4 null mutants with defects in efferocytosis show increased load of apoptotic cells (Figure 6C and D in both unchallenged and injury condition). Do the authors have precedent data to cite going to the same direction? Are cell debris engulfed but not degraded efficiently?
• In Figure 6D it seems indeed that NimB4, NimB1/NimB4 and Draper mutants do not accumulate more apoptotic material upon injury. However, levels for NimB4 is close to the one obtained with NimB1 mutants. Is it statistically true? If yes, what could be the reason for this similarity ? In any case, as some important conclusion relies on the comparison between UC and injury conditions, adequate statistics and representations could be proposed.
• The authors claim with analyses of Figure 8C and D, that NimB1 mutants show acidic vehicles normal in size and fluorescence intensity. However, statistical differences are still observed compared to control condition, which is also seen in representative images shown.

Minor comments:

• In figure 2D, what allows to say the expression is restricted in macrophages ? Is it the colocalization with SIMU being a macrophage-specific marker?
• In figure S3B and C, it appears that double NimB1/NimB4 mutants exhibit less spreading than single ones (especially NimB4). Is it the case (statistic significance). If yes what could be the explanation?
• Several graphs are identical between figure 4 ans S4. It is probably not useful and complicates reading.
• As TEM images shown in Figure 8B do not lead to quantitative data, I would put it as supplementary file.

Significance

This study uses several approaches and models to address the role of NimB1 in efferocytosis. Both In Vitro and In Vivo approaches are proposed. They give insight into the role of this protein with unknown function so far. Some statistical analysis could be performed to improve the clarity of conclusions. One of the important aspects is the secreted nature of NimB1.However, additional approaches could be proposed to confirm this.

Basic immunologists and cell biologists would be interested in reading this article that highlights the delicate equilibrium between pro and anti-efferocytosis molecules.

I am an immunologist/cell biologist with expertise in lysosomal catabolism. As I work on mouse models or Human samples, my mastering of drosophila as a model is limited.

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

This work defines NimB1 protein as a PS binding bridging molecule but with a negative regulatory role in efferocytosis. Specifically, the authors demonstrate via a variety of genetic, cell biological, and other approaches that loss of NimB1 leads to Drosophila macrophages being more adherent to apoptotic targets and engulf them more robustly. The authors also nicely demonstrate that the function of NimB1 differs from NimB4, and the double mutant demonstrating PS-binding yet, distinct roles. Further, the authors show that NimB1 does not affect bacterial phagocytosis.

Overall, this is a well-done study. The authors have already done a very thorough job addressing the key points and I congratulate the authors.

My only minor comment is that the authors could try to make the comment better about whether or not such a 'negative regulatory' bridging molecules may exist in other species, and particularly mammals. If so, this is quite novel. The authors refer to CD47 but this is a membrane protein. The other minor comment is whether the authors ever tried express the PS binding domains as a fusion protein - this would provide a more direct evidence for the binding to PS (although the authors do competitive inhibition with Annexin V). This could be commented upon although testing this is not necessary if they have not already done so.

Significance

The identification of the negative regulator bridging protein NimB1 is novel and could be broadly interesting to those studying efferocytosis.

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

1. General Statements [optional]

• *

*We thank the reviewers for finding the manuscript enjoyable and well-written, with experiments that were performed well, show solid results and provide useful data for the community. The reviewers have provided meaningful feedback to improve this study. We have addressed the comments point-by-point below. The main text will also be further modified to incorporate new analysis where it has not yet been done. *

• *

• *

2. Description of the planned revisions

Reviewer 1:

Summary OTX2 is a pivotal transcription factor that regulates the fate choice between somatic and primordial germ cell (PGC) lineages in early mouse development. In the current study, the authors use in vitro stem cell models to demonstrate that OTX2 mediates this developmental fate decision through controlling chromatin accessibility, whereby OTX2 helps to activate putative enhancers that are associated with somatic fate. By extension, those somatic-associated regulatory regions therefore become inaccessible in cells adopting PGC identity in which Otx2 is downregulated.

Comments I enjoyed reading this manuscript. The experiments have been carried out well and for the most part the results provide convincing evidence to support the claims and conclusions in the manuscript. I particularly liked the experiments using the inducible Otx2 transgene to examine the acute changes in chromatin accessibility following restoration of OTX2.

I include some suggestions below to the authors for additional analyses that I feel would further strengthen their study.

I also felt that the authors focus almost exclusively on the subset of OTX2-bound sites that lose accessibility in the absence of OTX2. But, as they show in several figure panels, these sites tend to be the minority and that most OTX2-occupied sites do not lose accessibility in Otx2-null cells (actually, more sites tend to gain accessibility). I encourage the authors to modify the text and some of the analyses to give a better balance to their study. We are pleased that this reviewer enjoyed our manuscript. As suggested by the reviewer, we included analyses on the regions that are bound by OTX2 but do not show an increase in accessibility (see section 3 reviewer 1 point 6). The text will be expanded to include the new data and to include the description of the subset of OTX2 sites that do not show accessibility changes in the absence of OTX2. We have responded to other points they raised as detailed in the sections below

• *

Figure 1: The authors write: "...OTX2 binds mostly to putative enhancers." Whether these distal sites are enhancers is not sufficiently evidenced in the manuscript, but it is important information to collect to support their model of OTX2 function. The authors should strengthen their analysis by examining whether OTX2 peaks are enriched at previously defined enhancer regions.

We plan to compare OTX2 bound regions with defined lists of enhancers identified in ESCs grown in Serum/LIF (e.g. Whyte et al 2013) and, if available, in 2i/LIF and EpiLCs. We will also analyse publicly available datasets for H3K4me1 (enhancer marker) and H3K27ac (marker of active regulatory regions) at the regions bound by OTX2 in ESCs and EpiLCs.

Figure 2: I'm still puzzled why the authors did not examine flow-sorted WT+cyto cells?

*We agree with the reviewer that it would be interesting to examine flow-sorted WT +cyto PGCLCs. Unfortunately, the expression of CD61 and SSEA1 only becomes visible from day 4 of PGCLC differentiation. Therefore, we were not able to isolate PGCLC at day 2 from WT cells differentiated in the presence of cytokines. We then used OTX2-/- cells at day 2 to model PGCLCs. This is based on the assumption that because day 6 Otx2-/- PGCLCs are transcriptionally similar to sorted day 6 WT cells (Zhang, Zhang et al Nature 2018), the same will be true at day 2. We will modify the text in the final version of this manuscript to clarify this point that has also been raised by reviewers 2 and 3. *

• *

Figure 3: I would be tempted to put Figure S3A and S3B into Figure 3. It would be better to show all 1246 DARs together, either ordered by OTX2 CT&RUN signal, or presented in two pre-defined groups (OTX2-bound vs unbound). I also suggest that the author show OTX2 signals and ATAC-seq signals for the 3028 DARs that gain accessibility in Otx2-null EpiLCs (this could be added to a supplemental figure).

Although the analysis has been carried out and the figures have been amended, the main text will be modified in a future updated version of the manuscript to incorporate these results.

• *

Figure 3: What is special about the 8% of OTX2-bound site that lose accessibility, versus the 92% of sites that do not?

*The 8% of the OTX2-bound regions that lose accessibility in the absence of OTX2 appear to be more sensitive to the loss of OTX2. One possible explanation is that the accessibility of the rest of OTX2 bound regions relies on other TFs, such as OCT4, that are expressed in EpiLCs. We will modify the main text to discuss this interesting point raised by the reviewer. *

Figure 6F: If the 4221 sites are split into those bound by OTX2 versus those that are not (related to Figure 6C) then is there a difference? i.e. are the OTX2-bound sites opening up?

We separated the 4,221 sites in OTX2 bound and unbound. The result is reported below:

*Although there is a slight increase in accessibility in the OTX2 bound subset, the average accessibility reaches less than ¼ of the accessibility of these regions when OTX2 is present from day 0 to day 4, while the OTX2 unbound regions do not show an increase in accessibility. Although we can not rule out that a longer treatment with tamoxifen may lead to higher accessibility in the OTX2 bound subset, the dynamics are extremely slower compared to the EpiLC regions where accessibility reaches 50% of the d0-d4 sample in just 1 hour of tamoxifen treatment. *

• *

Is there any evidence that OTX2 binds and compacts PGCLC enhancers in somatic cells? I appreciate this is different to the main thrust of the authors' model, but being able to show that OTX2 does not compact these sites lends further support to their preferred model of OTX2 opening sites of somatic lineages.

*Comparing the ATAC-seq in PGCLCs with ESCs and EpiLCs, we identify a subset of regions that are open in PGCLC only (PGCLC-specific accessible regions, see below). These regions do not show binding of OTX2 in WT EpiLCs or the d0-d2 Tam sample, suggesting that OTX2 does not bind and compact PGCLC-specific enhancers. *

• *

PGCLC-specific regions showing high accessibility only in PGCLCs.

• OTX2 CUT&RUN signal in WT EpiLC, OTX2-ERT2 PGCLCs in presence or absence of Tamoxifen, showing that OTX2 does not bind PGCLC-specific regions even when it is overexpressed in GK15 medium.*

*These analyses will be incorporated in the manuscript. *

• *

Discussion: Have prior studies established a connection between OTX2 and chromatin remodellers that can open chromatin? Or, if not, then perhaps this could be proposed as a line of future research.

We thank the reviewer for suggesting to amplify the discussion on the possible connection between OTX2 and chromatin remodellers. Although there is no evidence in the literature of a direct interaction between OTX2 and chromatin remodellers, this can not be excluded. The connection might also be indirect: OTX2 is known to interact with OCT4, which in turn interacts and recruits to chromatin the catalytic subunit of the SWI/SNF complex, BRG1. This point will be discussed in a modified version of the manuscript.

• *

• *

Reviewer 2:

Barbieri and Chambers explore the role of OTX2 on mouse pluripotency and differentiation. To do so, they examine how the chromatin accessibility and OTX2 binding landscape changes across pluripotency, the exit of pluripotency towards formative and primed states, and through to PGCLC/somatic differentiation. The work mostly represents a resource for the community, with possible implications for our understanding of how OTX2 might mediate the germline-soma switch of fates. While the findings of the work are modest, the results seem solid and the manuscript is clear and well-written.

*We are pleased that this reviewer found our results solid and the manuscript clear. *

I have some comments as indicated below:

1. The comparison between Otx2-/- cells in the presence of PGCLC cytokines compared to WT cells in the absence of cytokines seems like it is missing controls to me. I assume the authors wanted to enable homogeneous populations to facilitate their bulk sequencing methods, but it seems to me like they are comparing apples with oranges. It would have been better to have the reciprocal situations (Otx2-/- cells in basal differentiation medium, and WT cells in PGCLC cytokines) with a sorting strategy to better unpick the differences between the presence and absence of Otx2 in the 2 protocols. Having said that, the authors are careful not to draw many comparisons between those populations so I don't think this omission affects their current claims. They should however clarify whether the flow cytometry (Supp Fig2) was used for sorting cells or if all cells were taken for bulk sequencing. *We agree with the reviewer that it would be of interest to compare the PGCLC and somatic population derived from the OTX2-/- cells in GK15 without cytokines with the same populations derived from WT cells differentiated in the presence of cytokines. Our work aims to identify what happens at the stages of PGCLC differentiation when cells are still competent for both germline and somatic differentiation. Previous work from the lab showed that this dual competence is lost after day 2, therefore we focus our attention on this time of differentiation. Unfortunately, the two surface markers characteristics of PGCs (CD61 and SSEA1) are not expressed at day2 and, therefore we are not able to sort PGCLCs derived from OTX2-/- cells in GK15 without cytokines or WT cells differentiated in the presence of cytokines. As recognised by this reviewer, we aimed to obtain two homogenous populations that can model PGCLCs and somatic cells. This is based on data obtained at day 6 when Otx2-/- PGCLCs show a similar transcriptome to sorted day 6 WT cells (Zhang, Zhang et al Nature 2018) and the assumption that the same will be true at day 2. We will clarify that the supplementary Figure 2 is not a sorting strategy. As this point has been raised by reviewers 1 and 2 as well, we will modify the text to clarify the choice and the assumption behind using OTX2-/- cells in the presence of cytokines and WT cells in the absence of cytokines to model PGCLCs and somatic cells respectively. *

2. *

Throughout the text, the authors subject cells (WT / Otx2-/- /Otx2ER ) to different protocols to look at accessibility and Otx2 binding, but with no mention of the cell fate differences that occur in these different conditions. For instance, it is unclear to me to which fate the WT cells without PGCLC cytokines go - I presume this is neural but perhaps this is a mixed fate, given that they are in GK15 rather than N2B27. Likewise, the OTX2ER experiments may promote a mixed population between PGCLC/somatic fates, and this is never described. Ideally transcriptomic data would be collected, but failing that, qPCR data should be obtained to examine this more closely.

*We are planning to generate RT-qPCR data for germ layer markers (ectoderm, endoderm and mesoderm) in WT cells in GK15 without cytokines at day 2, as well as OTX2-ERT2 cells with and without Tamoxifen at day 2 (noTam, d0-d2) and day 4 (no Tam, d0-d4). *

The authors also state that "OTX2 facilitates Fgf5 transcription' (page10) but provide no transcriptional data to substantiate this claim. Again RT-qPCR would help make this point.

*We will analyse the level of Fgf5 by RT-qPCR in OTX2-ERT2 EpiLCs treated for 1 hour and 6 hours with Tamoxifen to show the effect of OTX2 on Fgf5 transcription. *

• *

It is unclear to me what the 'increase[d] accessibility' (eg abstract final sentence, Figure 3E) really means at the cellular level. Does this indicate that more cells have this site open, and does this have implications for the heterogeneity of cell fates observed? Since the authors are concerned with fate decisions, this seems like an important consideration that should at least be discussed.

The possibility that the increased accessibility is due to higher heterogeneity in the population is interesting and it will be included in the discussion in a revised version of the manuscript.

• *

• *

Reviewer 3:

In this manuscript, the authors perform OTX2 CUT&RUN and ATAC-seq in Otx2-null and WT ESCs, EpiLCs and PGCLCs to understand whether the role of OTX2 in restricting mouse germline entry that they previously described (Zhang Nature 2018) mechanistically depends on chromatin remodeling. They identify differentially accessible regions (DARs) between Otx2-null and WT cells at different stages of differentiation and show that many of these are OTX2 bound in WT. They then show using cells expressing OTX2-ER^T2 in Otx2-null Epiblast cells that when OTX2 is moved into the nucleus, the regions that were differentially closed in Otx2-null open within an hour, suggesting chromatin accessibility is directly controlled by OTX2 (rather than indirect effects involving transcription and translation which one would expect to take longer). The scope is narrow, but this is nice work and useful data for the mouse PGC field. However, there are a few places where the data could be strengthened, and the writing is a little confusing in places, for example by stating as fact in early sections what is not proven until later.

We thank the reviewer for finding our work nice and useful for the mouse PGC field, and for the useful comments to improve the manuscript. We have included new analysis and modified the text as suggested to improve the writing, avoiding early statements that were not fully proven until later in the manuscript. We have responded to other points they raised as detailed below and in the next section.

• *

1) "we compared Otx2-/- cells cultured in the presence of PGC-promoting cytokines with wild-type cells cultured in the absence of PGC-promoting cytokines. Under these conditions Otx2-/- cells produce an essentially pure (>90%) CD61+/SSEA1+ population that we refer to as PGCLCs, while wild-type cells yield a cell population from which PGCLCs are absent"

This is not a controlled comparison since one cannot separate the day 2 effect of cytokines from that of the Otx2 knockout. The manuscript would be strengthened if the authors include WT somatic and PGCLCs from the +cytokine conditions, which could be easily sorted out as shown in Supp. Fig. 2. Ideally they would also include Otx2-null somatic cells, although Supp. Fig. 2 shows those are rare under the conditions considered.

*This work aimed to analyse early stages of EpiLC to PGCLC differentiation when cells are still competent for both somatic and germline differentiation. This stage has been described previously to be at day 2 of differentiation in GK15 + cytokines (PGCLC differentiation medium, Zhang, Zhang et al, Nature 2018). Unfortunately, CD61 and SSEA1 are not expressed at day 2 of PGCLC differentiation, and they start to be expressed on the cell surface by day 4. Consequently, it is impossible to sort cells at day 2 using the CD61+/SSEA1+ strategy. To overcome this problem, we used WT cells grown in GK15 without cytokines to model a population of somatic cells and OTX2-/- cells grown in GK15+ cytokines to model a homogeneous population of PGCLCs. As explained in a similar point raised by reviewers 2 and 3, we assumed that, as OTX2-/- cells grown in the presence of cytokines are transcriptionally similar to sorted WT cells at day 6 (Zhang, Zhang et al, Nature 2018), OTX2-/- cells at day 2 are similar to their WT counterpart at day 2. The main text will be modified to clarify that we are using homogeneous populations to model both PGCLC and somatic cells and that Figure S2 does not show a sorting strategy. *

• *

3) "In ESCs, OTX2 binds We are planning to perform a statistical analysis to ascertain that the small number of DARs bound by OTX2 are or are not bound by chance by OTX2.

• *

4) It would be good if the discussion was broadened to include both human and other transcription factors that are involved. How much of these conclusions could one expect to carry over to human or other mammals? There is some work from the Surani lab considering OTX2 in human. One could even look at published ATAC or OTX2 chip-seq data in hPSCs and potentially learn something interesting. Furthermore, there are studies on other transcription factors modulating chromatin accessibility in the decision between germline and somatic cells, for example PRDM1, PRDM14 (refs in e.g. Tang et al Nat Rev Gen 2016) or TFAP2A (at least in human (Chen et al Cell Rep 2019)). Do these factors affect the same genes? Is a coherent picture emerging of their respective roles in germline entry?

*As suggested by the reviewer, we will discuss the role of OTX2 in human PGCLC formation and include studies on PGC-specific transcription factors concerning changes in chromatin accessibility in germline and somatic cells. This will be included in a revised version of the manuscript. *

• *

• *

3. Description of the revisions that have already been incorporated in the transferred manuscript

Reviewer 1:

1. Figure 1: The authors report in the methods that they performed OTX2 CUT&RUN in biological duplicates. It would strengthen their results if they showed in Figure S1 some representative data from each replicate separately to show the consistency. As suggested by the reviewer to show consistency between replicates, two representative tracks of the two CUT&RUN replicates at the Tet2 (ESCs) and Fgf5 (EpiLCs) loci have been included in Figure S1A. The corresponding tracks of the average bigwig files are reported in Figure 1E. The main text (page 5) and the figure legends have been amended to incorporate the new panels.

2. *

Figure 2: I think it would be helpful to remind the reader here that Otx2 is normally downregulated in PGCs, and that Otx2 expression is maintained (at least initially) in somatic cells. This would help explain the logic behind the choice of samples that were profiled.

We modified the text with the following sentence, as suggested by the reviewer, emphasising the level of OTX2 in early somatic vs early PGCLCs: “Otx2 expression is rapidly downregulated in the EpiLC to PGCLC transition while its expression is maintained longer in cells entering the somatic lineage [8]*” (page 7). *

• *

Figure 2D: I appreciate that the highlighted region at the Tet2 locus is a DAR, but from the genome tracks it looks as though the region still has high accessibility. Are there any other examples to exemplify a more obvious DAR? Additionally, since twice as many DARs gain accessibility in Otx2-null ESCs compared to lose accessibility, why not show examples of these as well? The same is true of EpiLCs. (Or alternatively, provide a good explanation for why not to show these other categories)

We substituted the Tet2 DAR with a more clear example of ESC DAR located in the Hes1 locus that shows low accessibility in Otx2-/- ESCs versus WT ESCs. Examples of ESC DARs and EpiLC DARs that show higher accessibility in Otx2-/- vs WT cells have been added as new panels 2E (DAR in Pebp4 locus) and 2G (DAR in Tdh locus). We also simplified the panels showing only ATAC-seq tracks in WT and OTX2-/- cells, either ESCs (2D-E) or EpiLCs (G-H). Text and figure legends have been modified to accommodate the changes made in Figure 2.

• *

Figure 3: I would be tempted to put Figure S3A and S3B into Figure 3. It would be better to show all 1246 DARs together, either ordered by OTX2 CT&RUN signal, or presented in two pre-defined groups (OTX2-bound vs unbound). I also suggest that the author show OTX2 signals and ATAC-seq signals for the 3028 DARs that gain accessibility in Otx2-null EpiLCs (this could be added to a supplemental figure).

Figures S3A and S3B have been moved to the main figure. Figure S3A is now part of Figure 3C, where all the 1,246 DARs are shown together, separated into two groups (OTX2-bound and -unbound). Figure S3B is now part of Figure 3F. A new heatmap showing the OTX2 and ATAC-seq signals for the 3028 regions that gain accessibility in Otx2-/- EpiLCs has been added as new Figure S3B. Only 28 out of the 3,028 regions overlap an OTX2 peak as shown in the new Figure S3A. These regions appear to be already open in ESCs (Figure S3C) and they do not fully close when OTX2 is absent. This can be explained by either a) the lack of expression of an OTX2 target gene that represses these regions or b) the continuous expression of a gene that is usually repressed by OTX2 in the transition to EpiLCs. In both cases, OTX2 does not directly repress these regions. Figure legends have been amended to incorporate the new panels. The main text will be modified to incorporate these results.

• *

Figure 6: Do the PGCLCs with OTX2 expression have chromatin accessibility profiles similar to somatic cells? Consider adding WT somatic cell data to Figure 6A, which could be an interesting comparison with the Tam d0-d2 samples.

*The heatmap showing the ATAC-seq signal at the additional OTX2-induced regions in somatic cells has been added to Figure 6A. The data show that the regions induced by OTX2 are not open in somatic cells generated in GK15. One possible explanation is the overexpression of OTX2 induces the opening of neural-associated regions, but neural differentiation is not fully supported in GK15 medium (see reviewer 2, point 3). As suggested by reviewer 2, we will perform RT-qPCR of germ layer markers to analyse the identity of somatic cells grown in GK15 (without cytokines) and somatic cells induced by OTX2 overexpression. *

• *

• *

• *

Reviewer 2:

The authors focus solely on the activating role of Otx2 in their data, but given the substantial proportion of DARs that decrease following Otx2 depletion, I presume it is possible that it also has a repressive effect? Either way, this should be discussed.

*As also suggested by reviewer 1 (point 6), we analysed the accessibility level and the OTX2 signal at the 3,028 regions that gain accessibility in Otx2-/- EpiLCs (new Figure S3A-C). These regions show high accessibility in ESCs suggesting that these are ESC regions that do not close properly in the transition to EpiLCs in the absence of OTX2. OTX2 CUT&RUN show a low to absent signal at these regions, with just 28 regions overlapping EpiLCs DARs that show higher accessibility in Otx2-/- cells, suggesting that OTX2 does not have a direct suppressive effect on them. *

• *

The authors state that d2 PGCLCs "show an intermediate position between ESCs and EpiLCs" based on the PCA location. They should be careful to qualify that this is only in the first 2 principal components, because it may well be the case (and is likely) that in other components the PGCLC population is far removed from the pluripotent states.

• The text has been updated as follows: d2 PGCLCs “show an intermediate position between ESCs and EpiLCs on both PC1 and PC2”.*

• *

Reviewer2 Minor Suggestions:

1. Presumably the regions bound by OTX2 in Tet2, Mycn and Fgf5 (Fig1E) are called enhancers because these are known from existing literature. It would be helpful to cite the relevant references to this in the text for those unfamiliar with these. References (Whyte et al, Cell, 2018 – Tet2 and Mycn, Buecker et al, Cell Stem Cell, 2013, Thomas et al, Mol Cell 2021 – Fgf5) have been added to the text and the figure legends.

On page 13, the authors say "To determine whether OTX2 expression is essential to maintain chromatin accessibility in somatic cells..." but this does not seem to be what they test because they are using PGCLC medium. Perhaps I misunderstood, but this could be clarified.

*Expression of OTX2 during the first 2 days of PGCLC differentiation leads to a block of germline differentiation as previously shown in Zhang, Zhang et al, Nature 2018. After 2 days of tamoxifen treatment, cells have acquired somatic fate and cells will undergo somatic differentiation even after tamoxifen is withdrawn after day 2. Nevertheless, we agree with the reviewer that the sentence is of difficult interpretation and we modified the sentence as shown below and as reported in the updated manuscript: “To determine whether OTX2 expression is essential to maintain chromatin accessibility in cells differentiating in the presence of PGC-inducing cytokines after day 2” (page 12). *

On page 14 the authors claim, "These results indicate that...the partner proteins that OTX2 act alongside differ...". While this may be the case, their results do not substantiate this, it is just speculation. Should be toned down.

The text has been modified as follows: "These results suggest that...the partner proteins that OTX2 act alongside differ..."

Page 18, PGCLC differentiation method sections needs to be described as such (ie. Add "For PGCLC differentiation..." before the second paragraph)

*The text “For PGCLC differentiation” has been added at the beginning of the PGCLC differentiation method section. *

It would be helpful to indicate time on the protocol schematics (eg Fig4A, 5A, 5D etc) as I had to keep checking the methods to find out how long the full differentiation time-course was.

*Indication of time has been added to Figures 1, 2, 4, 5 and 6. *

Since the authors compare between the Tam d0-d2 treatments assessed at d2 versus d4 (Figure5B vs 5E) it would be helpful to make the colourbars the same scale, for both ATAC and Cut&Run datasets.

*The heatmap in Figure 5B has been modified. The colourbars of Figure 5B and 5E are now using the same scale. *

• *

Reviewer 3:

1) As a minor point related to this, the second sentence is confusing since it kind of sounds like Otx2-/- and WT cells are compared under the same conditions unless one carefully reads the previous sentence.

The text has been modified to clarify the different medium conditions for WT and OTX2-/- cells, as follows: “In the presence of PGC-inducing cytokines, Otx2-/- cells produce an essentially pure (>90%) CD61+/SSEA1+ population that we refer to as PGCLCs, while wild-type cells differentiated in GK15 medium without cytokines yield a cell population from which PGCLCs are absent” (page 7).

• *

2) "This suggests that OTX2 acts as a pioneer TF to regulate the accessibility of enhancers E1, E2 and E3."

This is from the text corresponding to Fig. 2. That data actually only shows that Otx2-null cells have DARs, so somehow OTX2 affects chromatin accessibility but it could be indirect by controlling transcription of genes that modify chromatin accessibility. It is not until figure 4 that the data suggests that OTX2 directly affects accessibility, perhaps as a pioneer TF.

The authors continue to make many statements about the direct action of OTX2 before the data supporting this is shown, on which I got hung up as a reader. I suggest the authors edit the manuscript to improve this. E.g. "OTX2 may directly control accessibility at these sites (Figure 3E)." and the fact that in 3E and other figure, it says "DARs increased by OTX2 binding" which at that point is not proven, so would better say "Otx2-null vs WT DARs" or something like that.

The sentence "This suggests that OTX2 acts as a pioneer TF to regulate..” has been removed from the text (page 9). The sentence “OTX2 may directly control accessibility at these sites” has been modified with “*suggesting that the presence of OTX2 affects accessibility at these sites” (page 9). The sentence “ Together, these results suggest that OTX2 is required to open these chromatin regions” has been modified to “Together, these results suggest that OTX2 is required for the accessibility of these chromatin regions”. *

The subset of DARs that increase in WT EpiLC and are bound by OTX2 that was called “DARs increased by OTX2 binding” has been renamed as “DARs higher in WT with OTX2 binding”. For consistency, the subset of DARs showing increased accessibility in WT EpiLCs that are not bound by OTX2 are now called “DARs higher in WT without OTX2 binding” (Figure 3, Figure 4, main text and figure legends). We will further revise the manuscript to avoid statements or hypotheses that are not yet supported by data throughout the text.

• *

Reviewer 3 – minor comments:

1) "Comparing wild-type and Otx2-/- ESCs identified 375 differentially accessible regions (DARs) with increased accessibility in wild-type cells, and 743 regions with higher accessibility in Otx2-/- ESCs (Figures 2C). An example of ESC DARs where accessibility is increased in cells expressing OTX2 is the intragenic enhancer of Tet2. Tet2 is expressed at high levels in ESCs but at low levels in EpiLCs."

The authors compare Otx2-null and WT ESCs then proceed to give an example comparing ESCs to EpiLCs, instead of Otx2-null vs WT ESCs, which is confusing.

Furthermore, here and in other places the authors describe ESCs as not expressing OTX2. However, they also show CUT&RUN data for OTX2 in ESCs etc, clearly indicating that it is expressed, just lower (otherwise how could one get anything?).

*We originally chose Tet2 enhancer as an example of the 375 ESC DAR with higher accessibility in WT vs Otx2-/- ESCs as it shows a slightly decreased level of accessibility and OTX2 binding in ESCs. Therefore, the sentence “where accessibility is increased in cells expressing OTX2” refers to WT cells (expressing OTX2) when compared to Otx2-/- cells (OTX2-null). The text has been changed to describe the new panel. The rest of the main text will be checked and modified where appropriate to avoid possible misinterpretations. *

*We also appreciate that the change in accessibility is not clearly visible in the original Figure 2, as also pointed out by Reviewer 1 (point 6). In the updated Figure 2, we show a region in the Hes1 locus as an example of the 375 ESC DARs. Moreover, we simplified the panels showing ATAC-seq tracks of WT and OTX2-/- ESC (Fig. 2D-E) or EpiLCs (Fig. G-H). *

2) "In contrast, in EpiLCs, OTX2 binds almost 40% (446 out of 1,246) of the DARs that are more accessible in wild-type than in Otx2-/- cells (Figure 3B-C). Notably, these regions are mainly located distal to genes (91%, Figure 3D), despite the increased fraction of promoter regions bound by OTX2 in EpiLCs (Figure S1A)."

Are the authors rounding percentages with 2 significant digits, as suggested by the "91%"? If so, 446/1245 ~ 36%, not 40%.

*The text has been modified from “OTX2 binds almost 40%” to “OTX2 binds 36%”. *

3) The results in Figure 4 are nice and the real meat of the paper. One suggestion: It would be helpful is Fig. 4B were split up between the 446 and 800 genes instead of showing all 1246, and if the WT control was shown in the same figure as well.

*Panels with the 446 and 800 regions have been added to Figure 4 instead of the panels with all 1246 regions. WT control has been inserted in Figure 4. The main text and the figure legends have been updated accordingly. *

4) "Enforced OTX2 expression opens additional somatic regulatory regions" - it would be clearer to say "OTX2 overexpression opens additional somatic regulatory regions", since this is really about DARs between EpiLCs that already express OTX2 and those forced to express higher than WT endogenous levels by the OTX2-ER system?

We thank the reviewer for their suggestion. The text has been modified (page 12)

• *

• *

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:

In this manuscript, the authors perform OTX2 CUT&RUN and ATAC-seq in Otx2-null and WT ESCs, EpiLCs and PGCLCs to understand whether the role of OTX2 in restricting mouse germline entry that they previously described (Zhang Nature 2018) mechanistically depends on chromatin remodeling. They identify differentially accessible regions (DARs) between Otx2-null and WT cells at different stages of differentiation and show that many of these are OTX2 bound in WT. They then show using cells expressing OTX2-ER^T2 in Otx2-null Epiblast cells that when OTX2 is moved into the nucleus, the regions that were differentially closed in Otx2-null open within an hour, suggesting chromatin accessibility is directly controlled by OTX2 (rather than indirect effects involving transcription and translation which one would expect to take longer). The scope is narrow, but this is nice work and useful data for the mouse PGC field. However, there are a few places where the data could be strengthened, and the writing is a little confusing in places, for example by stating as fact in early sections what is not proven until later.

Major Comments:

1. "we compared Otx2-/- cells cultured in the presence of PGC-promoting cytokines with wild-type cells cultured in the absence of PGC-promoting cytokines. Under these conditions Otx2-/- cells produce an essentially pure (>90%) CD61+/SSEA1+ population that we refer to as PGCLCs, while wild-type cells yield a cell population from which PGCLCs are absent"

This is not a controlled comparison since one cannot separate the day 2 effect of cytokines from that of the Otx2 knockout. The manuscript would be strengthened if the authors include WT somatic and PGCLCs from the +cytokine conditions, which could be easily sorted out as shown in Supp. Fig. 2. Ideally they would also include Otx2-null somatic cells, although Supp. Fig. 2 shows those are rare under the conditions considered.

As a minor point related to this, the second sentence is confusing since it kind of sounds like Otx2-/- and WT cells are compared under the same conditions unless one carefully reads the previous sentence. 2. "This suggests that OTX2 acts as a pioneer TF to regulate the accessibility of enhancers E1, E2 and E3."

This is from the text corresponding to Fig. 2. That data actually only shows that Otx2-null cells have DARs, so somehow OTX2 affects chromatin accessibility but it could be indirect by controlling transcription of genes that modify chromatin accessibility. It is not until figure 4 that the data suggests that OTX2 directly affects accessibility, perhaps as a pioneer TF.

The authors continue to make many statements about the direct action of OTX2 before the data supporting this is shown, on which I got hung up as a reader. I suggest the authors edit the manuscript to improve this. E.g. "OTX2 may directly control accessibility at these sites (Figure 3E)." and the fact that in 3E and other figure, it says "DARs increased by OTX2 binding" which at that point is not proven, so would better say "Otx2-null vs WT DARs" or something like that. 3. "In ESCs, OTX2 binds <10% (30 out of 375) of DARs that are more accessible in wild-type cells than in Otx2-/- cells (Figure 3A), suggesting that accessibility of ESC DARs is directly due to OTX2 in a small subset of DARs."

When a small number of DARs are OTX2 bound, it does not necessarily suggest that that small set is directly affected by OTX2. It could just mean no DARs are controlled directly by OTX2 and then some are bound by chance by OTX2. Some appropriate statistical null hypotheses about the occurrence of OTX2 motifs might help to see if 10% is more than chance. 4. It would be good if the discussion was broadened to include both human and other transcription factors that are involved. How much of these conclusions could one expect to carry over to human or other mammals? There is some work from the Surani lab considering OTX2 in human. One could even look at published ATAC or OTX2 chip-seq data in hPSCs and potentially learn something interesting. Furthermore, there are studies on other transcription factors modulating chromatin accessibility in the decision between germline and somatic cells, for example PRDM1, PRDM14 (refs in e.g. Tang et al Nat Rev Gen 2016) or TFAP2A (at least in human (Chen et al Cell Rep 2019)). Do these factors affect the same genes? Is a coherent picture emerging of their respective roles in germline entry?

Minor comments:

1. "Comparing wild-type and Otx2-/- ESCs identified 375 differentially accessible regions (DARs) with increased accessibility in wild-type cells, and 743 regions with higher accessibility in Otx2-/- ESCs (Figures 2C). An example of ESC DARs where accessibility is increased in cells expressing OTX2 is the intragenic enhancer of Tet2. Tet2 is expressed at high levels in ESCs but at low levels in EpiLCs."

The authors compare Otx2-null and WT ESCs then proceed to give an example comparing ESCs to EpiLCs, instead of Otx2-null vs WT ESCs, which is confusing. Furthermore, here and in other places the authors describe ESCs as not expressing OTX2. However, they also show CUT&RUN data for OTX2 in ESCs etc, clearly indicating that it is expressed, just lower (otherwise how could one get anything?). 2. "In contrast, in EpiLCs, OTX2 binds almost 40% (446 out of 1,246) of the DARs that are more accessible in wild-type than in Otx2-/- cells (Figure 3B-C). Notably, these regions are mainly located distal to genes (91%, Figure 3D), despite the increased fraction of promoter regions bound by OTX2 in EpiLCs (Figure S1A)."

Are the authors rounding percentages with 2 significant digits, as suggested by the "91%"? If so, 446/1245 ~ 36%, not 40%. 3. The results in Figure 4 are nice and the real meat of the paper. One suggestion: It would be helpful is Fig. 4B were split up between the 446 and 800 genes instead of showing all 1246, and if the WT control was shown in the same figure as well. 4. "Enforced OTX2 expression opens additional somatic regulatory regions" - it would be clearer to say "OTX2 overexpression opens additional somatic regulatory regions", since this is really about DARs between EpiLCs that already express OTX2 and those forced to express higher than WT endogenous levels by the OTX2-ER system?

Significance

Also see summary. Understanding what restricts cells to germline vs somatic lineages is an important question. By providing functional data showing that OTX2 directly controls chromatin accessibility, the authors add an important layer of understanding to their previous finding that OTX2 plays a key role in preventing mouse germline entry. The use of their previously established OTX2-null cells expressing OTX2-ER to rapidly induce nuclear OTX2 in a mutant background or the most part makes their experiments elegant and convincing. In focusing on the role of one gene in one event in one species, it is specialized and narrow in scope and will mostly be of interest to experts in the field, but there is nothing wrong with that.

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

Barbieri and Chambers explore the role of OTX2 on mouse pluripotency and differentiation. To do so, they examine how the chromatin accessibility and OTX2 binding landscape changes across pluripotency, the exit of pluripotency towards formative and primed states, and through to PGCLC/somatic differentiation. The work mostly represents a resource for the community, with possible implications for our understanding of how OTX2 might mediate the germline-soma switch of fates. While the findings of the work are modest, the results seem solid and the manuscript is clear and well-written. I have some comments as indicated below:

• The comparison between Otx2-/- cells in the presence of PGCLC cytokines compared to WT cells in the absence of cytokines seems like it is missing controls to me. I assume the authors wanted to enable homogeneous populations to facilitate their bulk sequencing methods, but it seems to me like they are comparing apples with oranges. It would have been better to have the reciprocal situations (Otx2-/- cells in basal differentiation medium, and WT cells in PGCLC cytokines) with a sorting strategy to better unpick the differences between the presence and absence of Otx2 in the 2 protocols. Having said that, the authors are careful not to draw many comparisons between those populations so I don't think this omission affects their current claims. They should however clarify whether the flow cytometry (Supp Fig2) was used for sorting cells or if all cells were taken for bulk sequencing.
• The authors focus solely on the activating role of Otx2 in their data, but given the substantial proportion of DARs that decrease following Otx2 depletion, I presume it is possible that it also has a repressive effect? Either way, this should be discussed.
• Throughout the text, the authors subject cells (WT / Otx2-/- /Otx2ER ) to different protocols to look at accessibility and Otx2 binding, but with no mention of the cell fate differences that occur in these different conditions. For instance, it is unclear to me to which fate the WT cells without PGCLC cytokines go - I presume this is neural but perhaps this is a mixed fate, given that they are in GK15 rather than N2B27. Likewise, the OTX2ER experiments may promote a mixed population between PGCLC/somatic fates, and this is never described. Ideally transcriptomic data would be collected, but failing that, qPCR data should be obtained to examine this more closely.
• The authors also state that "OTX2 facilitates Fgf5 transcription' (page10) but provide no transcriptional data to substantiate this claim. Again RT-qPCR would help make this point.
• The authors state that d2 PGCLCs "show an intermediate position between ESCs and EpiLCs" based on the PCA location. They should be careful to qualify that this is only in the first 2 principal components, because it may well be the case (and is likely) that in other components the PGCLC population is far removed from the pluripotent states.
• It is unclear to me what the 'increase[d] accessibility' (eg abstract final sentence, Figure 3E) really means at the cellular level. Does this indicate that more cells have this site open, and does this have implications for the heterogeneity of cell fates observed? Since the authors are concerned with fate decisions, this seems like an important consideration that should at least be discussed.

Minor Suggestions:

• Presumably the regions bound by OTX2 in Tet2, Mycn and Fgf5 (Fig1E) are called enhancers because these are known from existing literature. It would be helpful to cite the relevant references to this in the text for those unfamiliar with these.
• On page 13, the authors say "To determine whether OTX2 expression is essential to maintain chromatin accessibility in somatic cells..." but this does not seem to be what they test because they are using PGCLC medium. Perhaps I misunderstood, but this could be clarified.
• On page 14 the authors claim, "These results indicate that...the partner proteins that OTX2 act alongside differ...". While this may be the case, their results do not substantiate this, it is just speculation. Should be toned down.
• Page 18, PGCLC differentiation method sections needs to be described as such (ie. Add "For PGCLC differentiation..." before the second paragraph)
• It would be helpful to indicate time on the protocol schematics (eg Fig4A, 5A, 5D etc) as I had to keep checking the methods to find out how long the full differentiation time-course was.
• Since the authors compare between the Tam d0-d2 treatments assessed at d2 versus d4 (Figure5B vs 5E) it would be helpful to make the colourbars the same scale, for both ATAC and Cut&Run datasets.

Significance

The study examines the binding of OTX2 and subsequent chromatin accessibility in pluripotent, primed and differentiated (PGCLC/Somatic) cell states, including through Otx2-/- cell lines and temporally-controlled exogenous expression of Otx2. As such, it represents a valuable resource into the potential direct targets of Otx2 and their change in accessibility state across cell types. The work is likely to be of interest to those working on understanding the exit of pluripotency, gene regulatory networks, and chromatin remodelling. My expertise is in cell fate decisions, pluripotency regulation and PGC(LC) differentiation.

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

Summary

OTX2 is a pivotal transcription factor that regulates the fate choice between somatic and primordial germ cell (PGC) lineages in early mouse development. In the current study, the authors use in vitro stem cell models to demonstrate that OTX2 mediates this developmental fate decision through controlling chromatin accessibility, whereby OTX2 helps to activate putative enhancers that are associated with somatic fate. By extension, those somatic-associated regulatory regions therefore become inaccessible in cells adopting PGC identity in which Otx2 is downregulated.

Comments

I enjoyed reading this manuscript. The experiments have been carried out well and for the most part the results provide convincing evidence to support the claims and conclusions in the manuscript. I particularly liked the experiments using the inducible Otx2 transgene to examine the acute changes in chromatin accessibility following restoration of OTX2.

I include some suggestions below to the authors for additional analyses that I feel would further strengthen their study.

I also felt that the authors focus almost exclusively on the subset of OTX2-bound sites that lose accessibility in the absence of OTX2. But, as they show in several figure panels, these sites tend to be the minority and that most OTX2-occupied sites do not lose accessibility in Otx2-null cells (actually, more sites tend to gain accessibility). I encourage the authors to modify the text and some of the analyses to give a better balance to their study.

1. Figure 1: The authors report in the methods that they performed OTX2 CUT&RUN in biological duplicates. It would strengthen their results if they showed in Figure S1 some representative data from each replicate separately to show the consistency.
2. Figure 1: The authors write: "...OTX2 binds mostly to putative enhancers." Whether these distal sites are enhancers is not sufficiently evidenced in the manuscript, but it is important information to collect to support their model of OTX2 function. The authors should strengthen their analysis by examining whether OTX2 peaks are enriched at previously defined enhancer regions.
3. Figure 2: I think it would be helpful to remind the reader here that Otx2 is normally downregulated in PGCs, and that Otx2 expression is maintained (at least initially) in somatic cells. This would help explain the logic behind the choice of samples that were profiled. That said, I'm still puzzled why the authors did not examine flow-sorted WT+cyto cells?
4. Figure 2D: I appreciate that the highlighted region at the Tet2 locus is a DAR, but from the genome tracks it looks as though the region still has high accessibility. Are there any other examples to exemplify a more obvious DAR? Additionally, since twice as many DARs gain accessibility in Otx2-null ESCs compared to lose accessibility, why not show examples of these as well? The same is true of EpiLCs. (Or alternatively, provide a good explanation for why not to show these other categories)
5. Figure 2: the authors write: "This suggests that OTX2 acts as a pioneer TF...". However, at this point in the manuscript, there is no evidence to support that OTX2 might have pioneer activity. I think this claim would be better suited to later in the manuscript, or in the discussion, following the finding that reintroduction of OTX2 can induce chromatin accessibility at previously closed sites.
6. Figure 3: I would be tempted to put Figure S3A and S3B into Figure 3. It would be better to show all 1246 DARs together, either ordered by OTX2 CT&RUN signal, or presented in two pre-defined groups (OTX2-bound vs unbound). I also suggest that the author show OTX2 signals and ATAC-seq signals for the 3028 DARs that gain accessibility in Otx2-null EpiLCs (this could be added to a supplemental figure).
7. Figure 3: What is special about the 8% of OTX2-bound site that lose accessibility, versus the 92% of sites that do not?
8. Figure 6: Do the PGCLCs with OTX2 expression have chromatin accessibility profiles similar to somatic cells? Consider adding WT somatic cell data to Figure 6A, which could be an interesting comparison with the Tam d0-d2 samples.
9. Figure 6F: If the 4221 sites are split into those bound by OTX2 versus those that are not (related to Figure 6C) then is there a difference? i.e. are the OTX2-bound sites opening up?
10. Is there any evidence that OTX2 binds and compacts PGCLC enhancers in somatic cells? I appreciate this is different to the main thrust of the authors' model, but being able to show that OTX2 does not compact these sites lends further support to their preferred model of OTX2 opening sites of somatic lineages.
11. Discussion: Have prior studies established a connection between OTX2 and chromatin remodellers that can open chromatin? Or, if not, then perhaps this could be proposed as a line of future research.

Significance

Strengths

The results presented provide a careful dissection of the role of OTX2 in controlling chromatin accessibility in different stages of pluripotent to somatic and PGC fates. The authors do a good job of revealing the stage-specific differences in OTX2 occupancy and chromatin accessibility as well as the different responses following the acute reintroduction of OTX2.

Limitations

I felt that the authors could present/discuss a bit more on alternative possibilities and models, as it would help the reader to better understand why they favour one model over other ones, and presenting these other possibilities could also provide more support for their preferred model.

Whether OTX2 is binding to putative enhancers is inferred but could be evidenced more strongly, as that is important for their model.

Advance

This study provides key information to understand the mechanisms of OTX2 function in cell fate choice. Similar functions have been shown in other contexts for other transcription factors, but this is a nicely done study and adds to our understanding of how transcription factors function in early development to direct cell-fate decisions.

Expertise

My field of expertise lies in the gene regulatory control of early developmental decisions.

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 (Evidence, reproducibility and clarity (Required)):

The study investigates the relationship between replication timing (RT) and transcription. While there is evidence that transcription can influence RT, the underlying mechanisms remain unclear. To address this, the authors examined a single genomic locus that undergoes transcriptional activation during differentiation. They engineered the Pln locus by inserting promoters of varying strengths to modulate transcription levels and assessed the impact on replication timing using Repli-seq. Key Findings: • Figure 1C and 1D: The data show that higher transcription levels correlate with an advanced RT, suggesting that transcriptional activity influences replication timing. • Figure 2: To determine whether transcription alone is sufficient to alter RT, the authors inserted an hPGK reporter at different genomic locations. However, given the findings in Figure 1, which suggest that this is not the primary mechanism, • Figure 3: The authors removed the marker to examine whether the observed effects were due to the promoter-driven Pln locus, which has significantly larger then the marker. • Figure 4: The study explores the effect of increased doxycycline (Dox) treatment at the TRE (tetracycline response element), further supporting the role of transcription in RT modulation. • Figure 5: The findings demonstrate that Dox-induced RT advancement occurs rapidly, is reversible, and correlates with transcription levels, reinforcing the hypothesis that transcription plays a direct role in influencing replication timing. • Figure 6. Shows that during differentiation transcription of Pln is not required for RT advancement.

Overall, the study presents a compelling link between transcription and replication timing, though some experimental choices warrant further clarification. I have no major comments.

__Minor Comments: __Overall, the results are convincing, and the study appears to be well-conducted. In Figure 2, the authors use the hPGK promoter. However, it is unclear why they did not use the constructs from the previous experiments. Given that the hPGK promoter did not advance RT in Figure 1, the results in Figure 2 may not be entirely unexpected.

We took advantage of previously published cell lines using a PiggyBac Vector designed to pepper the reporter gene at random sites throughout the genome; the point of the experiment was to acquire supporting evidence for the hypothesis that any vector with its selectable marker driven by the hPGK promoter will not advance RT no matter where it is inserted. Since there are reports concluding that transcription per se is sufficient to advance RT, it was important to confirm that there was nothing unique about the particular vector or locus into which we inserted our panel of vectors.

ACTION DONE: We have now added the following sentence to the results describing this experiment: “____By analyzing RT in these lines, we could evaluate the effect of a different hPGK vector on RT when integrated at many different chromosomal sites. “

Additionally, the study does not formally exclude the possibility that Pln protein expression itself influences RT. In Figure 1, readthrough transcription at the Pln locus could potentially drive protein expression. It would be useful to know whether the authors address this point in the discussion.

NOT DONE FOR NEED OF CLARIFICATION: It is unclear why a secreted neural growth factor would have a direct effect on replication timing in embryonic stem cells and, in particular, only in cis (remember there is a control allele that is unaffected). We would be happy to address this in the Discussion if we understood the reviewers’ hypothesis. We cannot respond to this comment without understanding the hypothesis being tested as we do not know how a secreted protein could affect the RT of one allele without affecting the other.

Regarding the mechanism, if transcription across longer genomic regions contributes to RT changes, transcription-induced could DNA supercoiling play a role. For instance, could negative supercoiling generated by active transcription influence replication timing?

Yes, many mechanisms are possible.

ACTION DONE: ____We have added the following sentence to the discussion, referencing a seminal paper on that topic by Nick Gilbert: “ ____For example, long transcripts could remodel a large segment of chromatin, possibly by creating domains of DNA supercoiling (Naughton et. al., 2013____).____”

It remains puzzling why Pln transcription does not contribute to replication timing during differentiation. Is there any evidence of chromatin opening during this process? For example, are ATAC-seq profiles available that could provide insights into chromatin accessibility changes during differentiation?

We thank the reviewer for asking this as we should have mentioned something very important here. Lack of necessity for transcription implies that independent mechanisms are functioning to elicit the RT switch. In other work (Turner et. al., bioRxiv, provisionally accepted to EMBO J.), we have shown that specific cis elements (ERCEs) can function to maintain early replication in the absence of transcription.

ACTION DONE: We now explicitly state in the Discussion: “____This is not surprising, given that ERCEs can maintain early RT in the absence of transcription (Turner, bioRxiv).”

ACTION TO BE DONE SOON: We will provide a new Figure 6D showing ATAC-seq changes upon differentiation of mESCs to mNPCs and their location relative to the promoter/enhance deletion. As you will see, there is an ATAC-seq site that appears during differentiation, upstream of the deletion. We will hypothesize in the revised manuscript that these are the elements that drive the RT switch and that future studies need to investigate that hypothesis. We have also added the following sentences to the discussion after the sentence above, stating: “____In fact, new sites of open chromatin, consistent with ERCEs appear outside of the deleted Ptn transcription control elements after differentiation (soon to be revised Figure 6D). The necessity and sufficiency of these sites to advance RT independent of transcription will be important to follow up.”

We also have preliminary data that are part of a separate project in the lab so they are not ready for publication, but are directly relevant to the reviewer’s question. This data shows evidence for a region upstream of the Ptn promoter/enhancer deletion described in Figure 6 that, when deleted, DOES have an effect on the RT switch during differentiation. This deletion overlaps an ATAC-seq site we will show in the new figure 6D.

Reviewer #1 (Significance (Required)):

This is a compelling basic single-locus study that systematically compares replication timing (RT) and transcription dynamics while measuring several key parameters of transcription.

My relevant expertise lies in transcriptional regulation and understanding how noncoding transcription influences local chromatin and gene expression.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

In the manuscript entitled: Transcription can be sufficient, but is not necessary, to advance replication timing", the authors use as they state a "reductionist approach" to address a long-standing question in the replication field on what level the process of transcription within a replication domain can alter the underlying replication timing of this domain. The authors use an elegant hybrid mouse embryonic stem cell line to discriminate the two allelic copies and focus on a specific replication domain harboring the neuronal Ptn gene that is only expressed upon differentiation. The authors first introduce four different promoters in the locus upstream of Ptn gene that drive expression of small transgenes. Only the promoters with highest transcriptional induction could advance RT. If the promoters are placed in such a way that they drive expression of the 96kb Ptn gene, then also some the weaker promoters can drive RT advancement, suggesting that it is a combination of transcriptional strength and size of the transcribed domain important for RT changes. Using a DOX-inducible promoter, the authors show that this happens very fast (3-6h after transcription induction) and is reversible as removal of DOX leads to slower RT again. Finally, deleting the promoter of Ptn gene and driving cells into differentiation still advances RT, allowing the authors to conclude that "transcription can be sufficient but not necessary to advance replication timing."

Major comments: Overall, this is a well designed study that includes all necessary controls to support the author's conclusions. I think it is a very interesting system that the authors developed. The weakness of the manuscript is that there is no mechanistic explanation how such RT changes are achieved on a molecular basis. But I'm confident that the system could be indeed used to further dissect the mechanistic basis for the transcription dependence of RT advancements.

Therefore, I support publication of this manuscript if a few comments below can be addressed.

1) Figure 4 shows a titration of different DOX concentrations and provides clear evidence that the degree of RT advancement tracks well with the level of transcription. As the doses of DOX are quite high in this experiment, have the authors checked on a global scale to what extent transcription might be deregulated in neighbouring genes or genome-wide?

The DOX concentration that we use for all experiments other than the titration is 2 µg/ml, which is quite standard. The high concentrations (up to 16µg/ml) are only used in the titration experiments shown in Figure 4 to demonstrate that we have reached a plateau. In fact, we stated in Materials and Methods that high doses of Dox led to cell toxicity. Looking at the transcription datasets, there are no significant changes in transcription below 8µg/ml, a few dozen significant changes at 8 and more such changes at 16µg/ml of DOX. The tables of genome wide RT and transcription are provided in the manuscript for anyone wishing to investigate the effects of Dox on cellular physiology but at the concentration used in all other experiments (2µg/ml) there are no effects on transcription.

__ACTION DONE: We have now modified the statement in the Materials and Methods to read: “ ____Mild toxicity and changes in genome-wide transcription were observed at 8µg/ml and more so at 16µg/ml”. __

2) One general aspect is that the whole study is only focused on the one single Ptn replication domain. Could the authors extend this rather narrow view a bit and also show RT data in the neighbouring domains. This would be particularly important for the DOX titration experiment that has the potential to induce transcriptional deregulation (see comment above).

__ACTION DONE: We have now added to revised Supplemental Figure 4 a zoom out of 10 Mb surrounding the Ptn gene showing no detectable effects on RT at any of the titration concentrations. __

__ACTION TO BE DONE SOON: To address the generalization of the findings (length and strength matter), we have repeated the ESC to NPC differentiation and performed both Repli-seq and BrU-seq to evaluate RT changes relative to total genomic nascent transcriptional changes. The sequencing reads for this experiment are in our analyst’s hands so we expect this to be ready within a few weeks. We will provide a new Figure 7 comparing genome-wide changes in RT vs. transcription to determine the significance of length and strength of transcription induction to RT advances and the necessity of transcriptional induction for RT advances. We and other laboratories have performed many integrative analyses of RNA-microarray/RNA-seq data vs. RT changes, but not total genomic nascent transcription and not with a focus on the effect of length and strength of transcription. For example, outcomes that would be consistent with our reductionist findings at the Ptn locus would be if we find domains that are advanced for RT with no induction of transcription (transcription not necessary) and little to no regions showing significant induction of transcription without RT advances. __

3) Figure 5 shows that the full capacity to advance RT upon DOX induction of the Ptn gene is achieved after 3h to 6h of DOX induction, so substantially less than a full cell cycle in mEScs (12h). This result suggests that origin licensing/MCM loading cannot be the critical mechanism to drive the RT change because only a small fraction of the cells has undergone M/G1-phase where origins are starting to get loaded. As a large fraction of mESCs (60-70%) are S-phase cells in an asynchronous population, the mechanism is likely taking place directly in S-phase. Could the authors try to synchronize cells in G1/S using double-thymidine block, then induce DOX for 3h before allowing cells to reenter S-phase and then check replication timing of the domain? This can be compared to an alternative experiment where transcription is only induced for 3h upon release into S-phase. This could provide more mechanistic insights as to whether transcription is sufficient to drive RT changes in G1 versus S-phase cells.

We agree that the timing of induction is such that it is very likely that alterations in RT can occur during S phase. The reviewer proposes a reasonable experiment that could be done, but it would require a long delay of this publication to develop and validate those synchronization protocols and we do not have personnel at this time to carry out the experiment. This would be a great initiating experiment for someone to pursue the mechanisms by which transcription can advance RT.

ACTION DONE: We have added the following sentence to the Discussion section on mechanisms: ____The rapid nature of the RT change after induction of transcription suggests that RT changes can occur after the functional loading of inactive MCM helicases onto chromatin in telophase/early G1 (Dimitrova, JCB, 1999; Okuno, EMBO J. 2001; Dimitrova, J. Cell Sci, 2002), and possibly after S phase begins.

Minor comments: • Figure 1B and Figure 6A. Quality of the genome browser snapshots could be improved and certain cryptic labelling such as "only Basic displayed by default" could be removed

ACTION DONE: We have modified these figures.

• The genome browser tracks appear a bit small across the figures and could be visually improved.

ACTION DONE: We have modified the genome browser tracks to improve their presentation

• In figure 1E we see an advancement in RT in Ptn gene caused by nearby enhanced Hyg-TK gene expression induced by mPGK promoter. However, in figure 3D we see mPGK promoter has reduced ability to advance RT of Ptn gene. It would be nice to address this discrepancy in the results.

The reviewer’s point is well taken. We are not sure of the answer. You can see that the transcription is very low in both cases, while the RT shift is greater in one replicate vs. the other.

ACTION DONE: We have, rather unsatisfactorily, added the following sentence to the results section describing Figure 3. “____We do not know why the mPGK promoter was so poor at driving transcription in this context.”

Reviewer #2 (Significance (Required)):

In my point of view, this is an important study that unifies a large amount of literature into a conceptual framework that will be interesting to a broad audience working on the intertwined fields of gene regulation, transcription and DNA replication, as well as cell fate switching and development.

__Reviewer #3 (Evidence, reproducibility and clarity (Required)): __ In their manuscript, "Transcription can be sufficient, but is not necessary, to advance replication timing," Vouzas et al. take a systematic and reductionist approach to investigate a late-replicating domain on chromosome VI. Here, they examine the effect of transcribing a single gene locus, Pleiotrophin, on replication timing. When inserting or manipulating promoters or transcript lengths using CRISPR-Cas9, replication timing was altered in mESCs as judged by a combination of Repli-Seq, Bru-Seq, and RNA-Seq. Importantly, they found that transcription can be sufficient to advance replication timing depending on the length and strength of the expression of an ectopically transcribed gene. Taken together, the manuscript presents a compelling argument that transcription can advance replication timing but is not necessary for it.

Major comments • A schematic or conceptual model summarising the major findings of transcription-dependent and independent mechanisms of RT advancement should be included in the discussion to add to the conceptual framework

NOT DONE: We discussed this at length between the two senior authors and the first author and we do not feel ready to draw a summary model. We do not know what is advancing RT when transcription is induced or not induced, and we are not comfortable choosing one possible model of many. We hope that the added speculations on mechanism in the Discussion will sufficiently convey the future research that we feel needs to be done.

ACTIONS DONE: In addition to the speculation on mechanism that already was in our Discussion section, we have added: On mechanisms of rapid induction of RT change, we have added to the Discussion: “____The rapid nature of the RT change after induction of transcription suggests that RT changes can occur after the functional loading of inactive MCM helicases onto chromatin in telophase/early G1 (Dimitrova, JCB, 1999; Okuno, EMBO J. 2001; Dimitrova, J. Cell Sci, 2002), and possibly after S phase begins.” And “For example, long transcripts could remodel a large segment of chromatin, possibly by creating domains of DNA supercoiling (Naughton et. al., 2013, PMID ____23416946).____ “ On mechanisms of RT advance in the absence of transcription, we have added the following to the Discussion: “____This is not surprising, given that ERCEs can maintain early RT in the absence of transcription (Turner, bioRxiv). In fact, chromatin features with the properties of ERCEs do appear outside of the deleted Ptn transcription control elements after differentiation (soon to be revised Figure 6C). The necessity and sufficiency of these new chromatin features to advance RT independent of transcription will be important to follow up.”

• Vouzas et al. spend a substantial part of the manuscript to delve into the requirements to advance RT and even use a Doxycycline-based titration for temporal advancement of RT. Yet, all conclusions come from the use of hybrid-genome mouse embryonic stem cells (mESCs). Therefore, it remains speculative if and whether findings can be generalized to other cell types or organisms. The authors could include another organism/ cell type to strengthen the relevance of their findings to a broader audience, particular as they identified promoters that drive ectopic gene expression without affecting RT. Showcasing this in other model organisms would be of great interest.

NOT DONE: To set this system up in another cell type or species would take a very long time. We also do not have personnel to carry that approach.

ACTION TO BE DONE SOON: As an alternative approach that partially addresses this reviewer’s concern, we will provide a new Figure 7 with an analysis of RT changes vs. transcriptional changes when mESCs are differentiated to neural precursor cells. As described above in response to Revier #2s criticism #2, we have repeated the ESC to NPC differentiation and performed both Repli-seq and BrU-seq to evaluate RT changes relative to total genomic nascent transcriptional changes. The sequencing reads for this experiment are in our analyst’s hands so we expect this to be ready within a few weeks. We will compare genome-wide changes in RT vs. transcription to determine the significance of length and strength of transcription induction to RT advances and the necessity of transcriptional induction for RT advances. We and other laboratories have performed many integrative analyses of RNA-microarray/RNA-seq data vs. RT changes, but not total genomic nascent transcription and not with a focus on the effect of length and strength of transcription. For example, outcomes that would be consistent with our reductionist findings at the Ptn locus would be if we find domains that are advanced for RT with no induction of transcription (transcription not necessary) and little to no regions showing significant induction of transcription without RT advances.

• OPTIONAL: as with the previous point, the authors went to great depth and length to show how ectopic manipulations affect RT changes on a single locus using genome-wide methods. In addition, the manuscript would benefit from the inclusion of other loci, particularly as transcription of the Ptn locus wasn't needed during differentiation to advance RT at all.

NOT DONE: This rigorous reductionist approach is laborious and to set it up at one gene at a time at additional loci would be a huge effort taking quite a long time.

ACTION TO BE DONE SOON: (same as response above) As an alternative approach that partially addresses this reviewer’s concern, we will provide a new Figure 7 with an analysis of RT changes vs. transcriptional changes when mESCs are differentiated to neural precursor cells. As described above in response to Reviewer #2s criticism #2, we have repeated the ESC to NPC differentiation and performed both Repli-seq and BrU-seq to evaluate RT changes relative to total genomic nascent transcriptional changes. The sequencing reads for this experiment are in our analyst’s hands so we expect this to be ready within a few weeks. We will compare genome-wide changes in RT vs. transcription to determine the significance of length and strength of transcription induction to RT advances and the necessity of transcriptional induction for RT advances. We and other laboratories have performed many integrative analyses of RNA-microarray/RNA-seq data vs. RT changes, but not total genomic nascent transcription and not with a focus on the effect of length and strength of transcription. For example, outcomes that would be consistent with our reductionist findings at the Ptn locus would be if we find domains that are advanced for RT with no induction of transcription (transcription not necessary) and little to no regions showing significant induction of transcription without RT advances.

• The same point of Ptn not needing to be transcribed to advance RT of the respective domain, albeit being a very interesting observation, disturbs the flow of the manuscript, as the whole case was built around transcription and this particular locus-containing domain. Maybe one can adapt the storytelling to fit better within the overall framework.

We would argue that demonstrating induction of Ptn, the only gene in this domain, is sufficient to induce early RT is a logical segway to asking whether, in the natural situation, induction is correlated with advance in RT. Our results show that transcription is sufficient but not necessary, which is expected if there are other mechanisms that regulate RT.

__ACTION DONE: To make this transition more smooth, we have added the following sentence to the beginning of the results section describing Figure 6: “ ____This raises the question as to whether the natural RT advance that accompanies Ptn induction during differentiation requires Ptn transcription, or whether other mechanisms, such as ERCEs (Sima / Turner) can advance RT independent of transcription. “ __

ACTION TO BE DONE SOON:____ To finish the work flow in a way that ties length and strength and sufficiency but not necessity in to the theme of natural cellular differentiation, we will provide a new Figure 7 with an analysis of RT changes vs. transcriptional changes when mESCs are differentiated to neural precursor cells, as described above.

Minor comments • While citations are thorough, some references (e.g., "need to add Wang, Klein, Mol. Cell 2021") are incomplete.

__ACTION TO BE DONE SOON: We apologize that some references seemed to not be incorporated into the reference manager Mendely. Since we are still planning to add one more figure soon and we will need to add some references for the datasets that will be shown in future Figure 6D, after that draft is ready, we will comb the manuscript for any references that were not entered and correct them. __

• The text corresponding to Figure 1C could use more explanation for readers not familiar with the depiction of Repli-Seq data.

ACTION DONE: “____Repli-seq labels nascent DNA with BrdU, followed by flow cytometry to purify cells in early vs. late S phase based on their DNA content, then BrdU-substituted DNA from each of these fractions is immunoprecipitated, sequenced and expressed as a log2 ration of early to late synthesized DNA (log2E/L). BrU-seq labels total nascent RNA, which is then immunoprecipitated an expressed as reads per million per kilobase (RPMK).”

• Figure 1C needs labelling of the x-axes.

ACTION DONE: We have now labeled the X axes.

• Statistical analyses should be used consistently throughout the manuscript and explained in more detail, i.e. significance levels, tests, instead of "Significant differences....calculated using x".

We used the same analysis for all the Repliseq data and the same analysis for all the Bruseq data. We agree that we did not present this consistently in the figure legends and methods.

ACTION DONE:____ To correct the confusion we have clarified the statistical methods in the methods section and referred to methods in the figure legends as follows:

The methods description of statistical significance for RT now reads: “____Statistical significance of RT changes for all windows in each sample, relative to WT, were calculated using RepliPrint (Ryba et al., 2011), with a p-value of 0.01 used as the cut-off for windows with statistically significant differences.”

The methods description of statistical significance for transcription now reads: “____Differential expression analysis, including the calculation of statistically significant differences in expression, was conducted using the R package DESeq2____. In Figure 1, statistical significance was calculated relative to HTK expression in the parental cell line, which is expected to be zero, since the parental line does not have an HTK insertion. In all other Figures significance was calculated relative to Ptn expression in the parental line, which is expected to be zero, since the parental line does not express Ptn.____”

The legend to Figure 1C now reads: The red shading indicates 50kb windows with statistically significant differences in RT between WT casteneus and modified 129 alleles, determined as described in Methods.

The legend to Figure 1E now reads: “The asterisks indicate a significant difference in the levels of HTK expression relative to HTK expression in the parental cell line as described in Methods. ____There are no asterisks for the RT data, as statistical significance was calculated for individual 50kb windows as shown in panel (C).”

Each time significance is measured in the subsequent legends, it is followed by the phrase “, determined as described in Methods” or “presented as in Figure 1C” or “presented as in Figure 1E” as appropriate.

__ __ **Referees cross-commenting** __ Comment on Reviewer#1's review__, comment mentioning ATAC-Seq: Another way to look at this could be to investigate for origin usage changes (BrdU-Seq or GLOE-Seq) of chromosome 6 during differentiation.

NOT DONE: Unfortunately we could not find any studies comparing origin mapping in mESCs and mNPCs.

Comment on Reviewer#2's review, major comment 3: I do agree with their statement that origin loading cannot be the driver of RT change, as MCM2-7 double hexamer loading is strictly uncoupled from origin firing. Hence, any mechanism responsible for RT advance must happen at the G1/S phase transition or during S-phase, most likely due to the regulated activity of DDK/CDK or the limitation and preferred recruitment of firing factors to early origins. This could be tested through overexpression of said factors.

NOT DONE: We agree that manipulating these factors would be a reasonable next approach to sort out mechanism. Due to limited resources and personnel, we will not be able to do this in a short period of time. We also argue that these are experiments for the next chapter of the story, likely requiring an entire PhD thesis (or multiple) to sort out.

ACTION DONE: We have added the following sentence to the Discussion section on mechanisms: ____The rapid nature of the RT change after induction of transcription suggests that RT changes can occur after the functional loading of inactive MCM helicases onto chromatin in telophase/early G1 (Dimitrova, JCB, 1999; Okuno, EMBO J. 2001; Dimitrova, J. Cell Sci, 2002), and possibly after S phase begins.

Reviewer #3 (Significance (Required)):

General: This manuscript presents a compelling study investigating the relationship between transcription and replication timing (RT) using a reductionist approach. The authors systematically manipulated transcriptional activity at the Ptn locus to dissect the elements of transcription that influence RT. The study's strengths lie in its rigorous experimental design, clear results, and the reconciliation of seemingly contradictory findings in the existing literature. However, some aspects could be improved, particularly in exploring the mechanistic details of transcription-independent RT regulation at the investigated domain, the generalisability of the findings to other cells/organisms, and enhancing the presentation of certain data (explanation of e.g. Figure 1c, dense figure arrangement, lack of a summary figure illustrating key findings (e.g., correlation between transcription rate, readthrough effects, and RT advancement)).

Advance: The manuscript directly addresses and reconciles contradictory findings in the literature regarding the effect of ectopic transcription on RT. Previous studies have reported varying effects, with some showing that transcription advances RT (Brueckner et al., 2020; Therizols et al., 2014), while others have shown no effect or only partial effects depending on the insertion site (Gilbert & Cohen, 1990; Goren et al., 2008). The current study conceptually advances the field by systematically testing different promoters and transcript lengths at a single locus (mechanistic insight), demonstrating that the length and strength of transcription, as well as promoter context, influence RT. This presents a unifying concept on how RT can be influenced. The authors also present a tunable system (technical advance) that allows rapid and reversible alterations of RT, which will certainly be useful for future studies and the field.

Audience: The primary audience will be specialised researchers in the fields of replication timing, epigenetics, and gene regulation. This study may be of interest beyond the specific field of replication timing, such as cancer biology, developmental biology, particularly if a more broader applicability of its tools and concepts can be shown.

Expertise: origin licensing, origin activation, MCM2-7, yeast and human cell lines

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 their manuscript, "Transcription can be sufficient, but is not necessary, to advance replication timing," Vouzas et al. take a systematic and reductionist approach to investigate a late-replicating domain on chromosome VI. Here, they examine the effect of transcribing a single gene locus, Pleiotrophin, on replication timing. When inserting or manipulating promoters or transcript lengths using CRISPR-Cas9, replication timing was altered in mESCs as judged by a combination of Repli-Seq, Bru-Seq, and RNA-Seq. Importantly, they found that transcription can be sufficient to advance replication timing depending on the length and strength of the expression of an ectopically transcribed gene. Taken together, the manuscript presents a compelling argument that transcription can advance replication timing but is not necessary for it.

Major comments

• A schematic or conceptual model summarising the major findings of transcription-dependent and independent mechanisms of RT advancement should be included in the discussion to add to the conceptual framework
• Vouzas et al. spend a substantial part of the manuscript to delve into the requirements to advance RT and even use a Doxycycline-based titration for temporal advancement of RT. Yet, all conclusions come from the use of hybrid-genome mouse embryonic stem cells (mESCs). Therefore, it remains speculative if and whether findings can be generalized to other cell types or organisms. The authors could include another organism/ cell type to strengthen the relevance of their findings to a broader audience, particular as they identified promoters that drive ectopic gene expression without affecting RT. Showcasing this in other model organisms would be of great interest.
• OPTIONAL: as with the previous point, the authors went to great depth and length to show how ectopic manipulations affect RT changes on a single locus using genome-wide methods. In addition, the manuscript would benefit from the inclusion of other loci, particularly as transcription of the Ptn locus wasn't needed during differentiation to advance RT at all.
• The same point of Ptn not needing to be transcribed to advance RT of the respective domain, albeit being a very interesting observation, disturbs the flow of the manuscript, as the whole case was built around transcription and this particular locus-containing domain. Maybe one can adapt the storytelling to fit better within the overall framework.

Minor comments

• While citations are thorough, some references (e.g., "need to add Wang, Klein, Mol. Cell 2021") are incomplete.
• The text corresponding to Figure 1C could use more explanation for readers not familiar with the depiction of Repli-Seq data.
• Figure 1C needs labelling of the x-axes.
• Statistical analyses should be used consistently throughout the manuscript and explained in more detail, i.e. significance levels, tests, instead of "Significant differences....calculated using x".

Referees cross-commenting

Comment on Reviewer#1's review, comment mentioning ATAC-Seq: Another way to look at this could be to investigate for origin usage changes (BrdU-Seq or GLOE-Seq) of chromosome 6 during differentiation.

Comment on Reviewer#2's review, major comment 3: I do agree with their statement that origin loading cannot be the driver of RT change, as MCM2-7 double hexamer loading is strictly uncoupled from origin firing. Hence, any mechanism responsible for RT advance must happen at the G1/S phase transition or during S-phase, most likely due to the regulated activity of DDK/CDK or the limitation and preferred recruitment of firing factors to early origins. This could be tested through overexpression of said factors.

Significance

General: This manuscript presents a compelling study investigating the relationship between transcription and replication timing (RT) using a reductionist approach. The authors systematically manipulated transcriptional activity at the Ptn locus to dissect the elements of transcription that influence RT. The study's strengths lie in its rigorous experimental design, clear results, and the reconciliation of seemingly contradictory findings in the existing literature. However, some aspects could be improved, particularly in exploring the mechanistic details of transcription-independent RT regulation at the investigated domain, the generalisability of the findings to other cells/organisms, and enhancing the presentation of certain data (explanation of e.g. Figure 1c, dense figure arrangement, lack of a summary figure illustrating key findings (e.g., correlation between transcription rate, readthrough effects, and RT advancement)).

Advance: The manuscript directly addresses and reconciles contradictory findings in the literature regarding the effect of ectopic transcription on RT. Previous studies have reported varying effects, with some showing that transcription advances RT (Brueckner et al., 2020; Therizols et al., 2014), while others have shown no effect or only partial effects depending on the insertion site (Gilbert & Cohen, 1990; Goren et al., 2008). The current study conceptually advances the field by systematically testing different promoters and transcript lengths at a single locus (mechanistic insight), demonstrating that the length and strength of transcription, as well as promoter context, influence RT. This presents a unifying concept on how RT can be influenced. The authors also present a tunable system (technical advance) that allows rapid and reversible alterations of RT, which will certainly be useful for future studies and the field.

Audience: The primary audience will be specialised researchers in the fields of replication timing, epigenetics, and gene regulation. This study may be of interest beyond the specific field of replication timing, such as cancer biology, developmental biology, particularly if a more broader applicability of its tools and concepts can be shown.

Expertise: origin licensing, origin activation, MCM2-7, yeast and human cell lines

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

In the manuscript entitled: Transcription can be sufficient, but is not necessary, to advance replication timing", the authors use as they state a "reductionist approach" to address a long-standing question in the replication field on what level the process of transcription within a replication domain can alter the underlying replication timing of this domain. The authors use an elegant hybrid mouse embryonic stem cell line to discriminate the two allelic copies and focus on a specific replication domain harboring the neuronal Ptn gene that is only expressed upon differentiation. The authors first introduce four different promoters in the locus upstream of Ptn gene that drive expression of small transgenes. Only the promoters with highest transcriptional induction could advance RT. If the promoters are placed in such a way that they drive expression of the 96kb Ptn gene, then also some the weaker promoters can drive RT advancement, suggesting that it is a combination of transcriptional strength and size of the transcribed domain important for RT changes. Using a DOX-inducible promoter, the authors show that this happens very fast (3-6h after transcription induction) and is reversible as removal of DOX leads to slower RT again. Finally, deleting the promoter of Ptn gene and driving cells into differentiation still advances RT, allowing the authors to conclude that "transcription can be sufficient but not necessary to advance replication timing."

Major comments:

Overall, this is a well designed study that includes all necessary controls to support the author's conclusions. I think it is a very interesting system that the authors developed. The weakness of the manuscript is that there is no mechanistic explanation how such RT changes are achieved on a molecular basis. But I'm confident that the system could be indeed used to further dissect the mechanistic basis for the transcription dependence of RT advancements. Therefore, I support publication of this manuscript if a few comments below can be addressed.

1. Figure 4 shows a titration of different DOX concentrations and provides clear evidence that the degree of RT advancement tracks well with the level of transcription. As the doses of DOX are quite high in this experiment, have the authors checked on a global scale to what extent transcription might be deregulated in neighbouring genes or genome-wide?
2. One general aspect is that the whole study is only focused on the one single Ptn replication domain. Could the authors extend this rather narrow view a bit and also show RT data in the neighbouring domains. This would be particularly important for the DOX titration experiment that has the potential to induce transcriptional deregulation (see comment above).
3. Figure 5 shows that the full capacity to advance RT upon DOX induction of the Ptn gene is achieved after 3h to 6h of DOX induction, so substantially less than a full cell cycle in mEScs (12h). This result suggests that origin licensing/MCM loading cannot be the critical mechanism to drive the RT change because only a small fraction of the cells has undergone M/G1-phase where origins are starting to get loaded. As a large fraction of mESCs (60-70%) are S-phase cells in an asynchronous population, the mechanism is likely taking place directly in S-phase. Could the authors try to synchronize cells in G1/S using double-thymidine block, then induce DOX for 3h before allowing cells to reenter S-phase and then check replication timing of the domain? This can be compared to an alternative experiment where transcription is only induced for 3h upon release into S-phase. This could provide more mechanistic insights as to whether transcription is sufficient to drive RT changes in G1 versus S-phase cells.

Minor comments:

• Figure 1B and Figure 6A. Quality of the genome browser snapshots could be improved and certain cryptic labelling such as "only Basic displayed by default" could be removed
• The genome browser tracks appear a bit small across the figures and could be visually improved.
• In figure 1E we see an advancement in RT in Ptn gene caused by nearby enhanced Hyg-TK gene expression induced by mPGK promoter. However, in figure 3D we see mPGK promoter has reduced ability to advance RT of Ptn gene. It would be nice to address this discrepancy in the results.

Significance

In my point of view, this is an important study that unifies a large amount of literature into a conceptual framework that will be interesting to a broad audience working on the intertwined fields of gene regulation, transcription and DNA replication, as well as cell fate switching and development.

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 study investigates the relationship between replication timing (RT) and transcription. While there is evidence that transcription can influence RT, the underlying mechanisms remain unclear. To address this, the authors examined a single genomic locus that undergoes transcriptional activation during differentiation. They engineered the Pln locus by inserting promoters of varying strengths to modulate transcription levels and assessed the impact on replication timing using Repli-seq.

Key Findings:

• Figure 1C and 1D: The data show that higher transcription levels correlate with an advanced RT, suggesting that transcriptional activity influences replication timing.
• Figure 2: To determine whether transcription alone is sufficient to alter RT, the authors inserted an hPGK reporter at different genomic locations. However, given the findings in Figure 1, which suggest that this is not the primary mechanism,
• Figure 3: The authors removed the marker to examine whether the observed effects were due to the promoter-driven Pln locus, which has significantly larger then the marker.
• Figure 4: The study explores the effect of increased doxycycline (Dox) treatment at the TRE (tetracycline response element), further supporting the role of transcription in RT modulation.
• Figure 5: The findings demonstrate that Dox-induced RT advancement occurs rapidly, is reversible, and correlates with transcription levels, reinforcing the hypothesis that transcription plays a direct role in influencing replication timing.
• Figure 6. Shows that during differentiation transcription of Pln is not required for RT advancement.

Overall, the study presents a compelling link between transcription and replication timing, though some experimental choices warrant further clarification. I have no major comments.

Minor Comments:

Overall, the results are convincing, and the study appears to be well-conducted. In Figure 2, the authors use the hPGK promoter. However, it is unclear why they did not use the constructs from the previous experiments. Given that the hPGK promoter did not advance RT in Figure 1, the results in Figure 2 may not be entirely unexpected.

Additionally, the study does not formally exclude the possibility that Pln protein expression itself influences RT. In Figure 1, readthrough transcription at the Pln locus could potentially drive protein expression. It would be useful to know whether the authors address this point in the discussion.

Regarding the mechanism, if transcription across longer genomic regions contributes to RT changes, transcription-induced could DNA supercoiling play a role. For instance, could negative supercoiling generated by active transcription influence replication timing?

It remains puzzling why Pln transcription does not contribute to replication timing during differentiation. Is there any evidence of chromatin opening during this process? For example, are ATAC-seq profiles available that could provide insights into chromatin accessibility changes during differentiation?

Significance

This is a compelling basic single-locus study that systematically compares replication timing (RT) and transcription dynamics while measuring several key parameters of transcription.

My relevant expertise lies in transcriptional regulation and understanding how noncoding transcription influences local chromatin and gene expression.

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

Manuscript number____: RC-2024-02806R

Corresponding author(s): Hamed Jafar-Nejad, Carmen Paradas

Point-by-point description of the revisions

Reviewer #1 (Evidence, reproducibility and clarity (Required)):

The manuscript by Cho et al uses conditional and inducible conditional mouse models to characterize the function of protein O-glucosyltransferase 1 (POGLUT1), known to cause a type of Limb Girdle Muscular Dystrophy (LGMD-R21), in skeletal muscle satellite cells, differentiation and regeneration. The Authors find that conditional deletion of POGLUT1 in the myogenic progenitors leads to postnatal muscle defects and lethality by postnatal day 30 or so. Muscle progenitors lacking POGLUT1 undergo reduced proliferation and accelerated differentiation, possibly leading to impairment in muscle regeneration. This is supported by an inducible conditional deletion of POGLUT1 in adult satellite cells. Finally, in vitro experiments suggest that POGLUT1 is required for NOTCH pathway activation in myogenic cells and that POGLUT1 could potentially glycosylate specific residues in NOTCH3.

• We thank the reviewer for the concise summary of the major findings of our manuscript and for her/his comments, which have helped us improve the manuscript.

  Major comments

1. What is the control used in Figure 1B and other panels? The genotype should be clearly specified instead of saying controls, in all the figures. It will be easier to interpret the data if densitometry of the western blots is provided, normalized to GAPDH levels. Also, is there some POGLUT1 protein remaining in the satellite cells in the cKOs? Data should be shown for an adult time point also (P30 or later), in addition to P0 and P4, to see whether Poglut1 levels are reduced in adult stages in the muscle and satellite cells in the cKOs.

• As described in the Methods section of the original submission, for all experiments involving Poglut1-cKO, siblings without Poglut1 deletion (Poglut1+/+, Poglut1flox/flox, or Pax7Cre; Poglut1+/+) were used as controls. To address the reviewer's comment, we added this information to the legend of all figures showing Poglut1-cKO

• In the revised version, the densitometry is provided for all western blot images (Figures 1B and 7A). We added western blot with anti-POGLUT1 on whole muscle extract from P21 animals (Figure 1B), which is the latest stage at which we consistently obtain Poglut1-cKO PAX7 staining and injury studies indicate that at P21 there is a significant reduction in the number of satellite cells. Therefore, isolation of sufficient satellite cells from later stages is not feasible.

  1. In Figure 1C, can the tibia weight be normalized to total body weight instead of tibia length and analyzed? Similarly, can the grip strength measurements in Figure 1G be normalized to total body weight and represented? Grip strength measurements in neonates is tricky; the Authors should clearly explain how this was done. The NMJ defects can be characterized better, especially since the Authors discuss about Agrin in detail.

• Normalizing tibialis anterior muscle weight to tibia bone length is commonly used and is especially useful when assessing muscle-specific growth or atrophy, as it provides a standardized measure to compare between different mice. To address the reviewer's comment, in the revised manuscript we have provided the tibialis anterior weight (not normalized), tibia length, and then the normalized TA weight (which was also shown in the first submission). Before normalization, the TA weight in cKO showed ~ 56% reduction compared to control (on average). The tibia length did not show a statistically significant difference between cKO and sibling control. After normalization to tibia length, the TA muscle weight still showed a substantial (55%) reduction in cKO mice compared to controls. These data indicate that a reduction in muscle size is the major if not the sole reason for the reduction observed in cKO body size.

• We reanalyzed the grip strength based on the reviewer's comment and found that when normalized to body weight, the grip strength is not significantly different between cKO and control animals (please see the following figure for the reviewer). Accordingly, we have modified the Results section to avoid any direct suggestion that the muscle function itself is impaired in the cKO animals. Specifically, instead of writing "... suggesting that the muscle function is impaired in the mutant mice" we wrote " ... suggesting reduced muscle strength in the mutant mice". Moreover, at the end of this paragraph, instead of referring to "a severe muscle weakness" we just describe the phenotype as "muscle weakness". Our goal from this paragraph is to suggest that since the animals have much smaller muscles, they are weaker and cannot eat properly; that's why switching to soft food somewhat improves their survival. The observed NMJ defects suggest that there might be functional muscle defects as well. However, to draw conclusions about the muscle function we will need additional experiments, which we believe are beyond the scope of the current manuscript. If the reviewer believes the normalized grip test should be added to the manuscript, we will be happy to do so as a supplementary figure.
• As stated in the original manuscript, grip test was done starting at P12. Following the Boston University Guideline (https://www.bu.edu/research/ethics-compliance/animal-subjects/animal-care/anesthesia/anesthesia-and-analgesia-neonatal-mice-and-rats-iacuc/; recommended by Jacksons Laboratory when working with neonatal mice), we consider P10 to be the end of the neonatal period in mice. This is also supported by the majority of the papers in literature. For example, in a paper entitled "Quantitative analysis of neonatal skeletal muscle functional improvement in the mouse" (PMID 18310108), the authors show that by P28, most anatomical and gene expression parameters of mouse skeletal muscle are close to adults, and that by P21, these parameters are closer to P28 than the early postnatal mice. Also, they show that the expression of neonatal myosin heavy chain is dramatically reduced between P7 and P14, with another 3-4-fold reduction between P14 and P21. Therefore, for muscle development, similar to many other contexts studied in mice, neonatal period ends between P7 and P14 (most likely P8-P10). Therefore, the grip tests have not been performed on neonatal mice. In the revised manuscript, we have clarified this issue in the Methods section.

• The NMJ defects were not the main focus of the manuscript and were performed as a standard characterization of this new mouse model. We agree that additional experiments can be performed to better characterize the NMJ defects in this model, but we believe those experiments are beyond the scope of the current manuscript. Also, given the new data and text added to the revised manuscript, we removed the section related to NMJ and agrin from the Discussion to reduce the manuscript length.

  1. What are the small myofibers seen at the corners of the larger myofibers at P21 in the cKOs in Figure 2E? What is the point that the Authors want to conclude from Figure 2C and D? Clearly, if there are fewer satellite cells, Pax7 transcript levels will decrease. In Figure 2D, how are the satellite cell samples normalized between control and cKOs; did they start with equal number of satellite cells or equal amount of satellite cell RNA between control and cKOs? In Figure 2F, representative images for cKOs should be shown? The conclusion from Figure 2F-G is unclear. Since the Authors claim that the weak laminin staining is resolved in the cKOs by P21, why is α-dystroglycan hypoglycosylation seen in the P21 muscle?

• The structures to which the reviewer is referring are blood vessels. As shown in the following figure generated for the reviewer, staining the muscle sections from control and cKO mice using the endothelial marker CD31 confirms the vascular nature of those structures.

• The decrease in PAX7+ cells was shown based on antibody staining, but this does not necessarily mean that Pax7 mRNA is similarly reduced. Figure 2C and 2D show that loss of Poglut1 in these cells affects Pax7 expression at the mRNA level. We started with equal amounts of RNA for Figure 2D and have added additional text to describe RNA isolation and RT-PCR in the Methods section of the revised manuscript.

• Representative images for cKOs are added to the revised Figure 2, as suggested.
• The conclusion from 2F-G is that a statistically significant reduction is specifically observed in those PAX7+ cells which co-express M-cadherin, which is another well-established satellite cell marker. The data also indicate that some PAX7+ cells persist in the mutant muscle, highlighting the importance of examining the ability of mutant muscle to repair muscle injury, as shown in Figure 4.

• The glycans on α-dystroglycan mediate its binding to laminin and other extracellular matrix proteins, but laminin expression level is not thought to regulate the degree of α-dystroglycan glycosylation. In other words, laminin staining intensity and the degree of α-dystroglycan are not causally related to each other. What we observed in P21 animals mirrors what we have reported for adult patient biopsies: normal level of extracellular matrix protein collagen VI associated with α-dystroglycan hypoglycosylation. It is worth mentioning that based on our previous report using primary LGMD-R21 patient myoblasts and C2C12 cells treated with a Notch inhibitor, reducing Notch signaling leads to α-dystroglycan hypoglycosylation, likely by altering the differentiation dynamics of the myoblasts (PMID: 27807076).

  1. Representative images should be shown as examples for all time points for both genotypes in Figure 3A. Figure 3H should be represented with statistically significant differences marked clearly. Have the Authors checked whether cell death contributes to the decrease in cultured satellite cells and Pax7+ cells in the cKOs in Figure 3F, G? Are the differences between the control and cKOs in fusion index and Myogenin expression (Figure 3J, K) statistically significant?

• Representative images are added in a new supplementary figure (Figure S3A) as requested by the reviewer.

• Thanks for pointing out the lack of statistics in Figure 3H. To address this point and a comment by reviewer 2, we removed those quantifications from the manuscript. Instead, we performed PAX7/MYOD/Ki67 co-staining on cultured cells and quantified the percentage of each cell state based on the expression of these three markers as a more accurate measure of quiescent versus cycling satellite cells, as well as progenitors and precursor cells. The revised quantification and statistical analysis is presented in the revised Figure 3H. This quantification is represented similarly to the data shown in Gattazzo et al 2020.
• Thanks for bringing up the issue of cell death. To address this point, we performed two rounds of anti-Caspase 3 staining on a limited number of myogenic progenitors that we were able to isolate from cKO and control animals during the revisions. Unfortunately, the stainings did not work. To ensure that we do not rule out the possibility of apoptosis without experimental data, we added the following sentence to the Results section: "These data suggest reduced proliferation of myogenic cells upon loss of Poglut1, although we cannot exclude that cell death also contributes to the reduced cell number."

• Thanks for pointing out the lack of statistics for these data. We have added additional independent samples to the data presented in 3J, K and found that the differences between the control and cKOs in fusion index and Myogenin expression are indeed statistically significant.

  1. Figure 4 has multiple problems in my opinion. First, P21 is too early a time point to be talking about regeneration, since satellite cells are just transitioning to become quiescent cells at this time (described in Lepper et al, Nature 2009). It is difficult to distinguish the regenerative and neonatal developmental roles of satellite cells in the experiment that the Authors have carried out. Another issue is that the Authors show a decrease in satellite cells in the cKOs; satellite cell function is clearly known to be required for regeneration. Therefore, there is not much novelty that the cKOs exhibit poor regenerative capability. The data shown in Figure 4B-E is more or less a repetition of what has been shown in Figures 2 and 3.

• After carefully reviewing publications including the one suggested by the reviewer, we found that multiple studies have performed muscle injury experiments at P21 or even younger ages to study regeneration capabilities. Also, please consider that many of our mutant mice are lethal around weaning age which was another reason for selecting P21 as our injury timepoint. To better explain the logic for this choice, we have added the following sentences to the revised manuscript (new text is underlined: "To address this question, we induced muscle injury in mutant and control mice by injecting cardiotoxin (CTX) into TA muscles at P21 (Fig. 4A), an age at which we consistently obtain Poglut1-cKO animals without any dietary changes (Fig. 1E). Importantly, in WT C57BL/6 mice, 51% of PAX7+ cells are reported to be in the quiescent state at P21 (Gattazzo et al., 2020) and the TA muscles of P21 mice exhibit a robust regenerative response to cardiotoxin-induced injury (Lepper et al., 2009)."

• While the reduction in satellite cells (SCs) in cKOs suggests impaired regeneration, it was essential to confirm this experimentally, as remaining SCs could still compensate. Therefore, it was necessary to assess whether the residual satellite cells could proliferate and contribute to muscle repair. Our results demonstrate that our mutant mice fail to repair muscle, providing evidence that the reduction in PAX7+ cells in Poglut1-cKO mice is functionally significant. The goal of these experiments was not to increase novelty; it was to increase experimental rigor and reproducibility so that we don't draw functional conclusions merely based on tissue staining.

• While Figures 2 and 3 examine muscle sections and myoblast cultures, Figure 4B-E presents isolated fibers in an ex vivo culture system. This approach allows us to assess satellite cell activation and proliferation in a different injury model. Figure 4 lets us confirm that the proliferation defect observed in the cKO is consistent across multiple experimental conditions. Additionally, we utilized two forms of muscle injury-in vivo and ex vivo-to comprehensively evaluate the function of PAX7+ cells in WT versus mutant muscles. The ex vivo findings from EDL fibers align with our cell culture experiments in Figure 3, further supporting the observed defect in muscle repair. We believe this approach will increase the likelihood of obtaining reproducible data.

  1. In Figure 5C, D, have the Authors checked the Pax7+ cell numbers between the controls and the i-cKOs? Is the difference seen in 1X TAM due to preexisting reduction in satellite cells in the i-cKOs? What is the explanation by the Authors for the large number of Tomato+ fibers seen in the 2X TAM and 3X TAM uninjured muscle (Figure 3C, E)? Figure 5E is difficult to comprehend and should be represented in a clearer manner.

• New PAX7 staining shows that most of the TOM+ cells (93% in i-cKO, 97% in control) co-express PAX7 (revised Figure 6D and 6E). If preexisting meant "before tamoxifen injection", a preexisting reduction in satellite cells should be highly unlikely, as the Pax7-Cre-ERT2 line has been used by multiple groups to study adult satellite cells, and we have not observed any abnormalities in the Poglut1[flox/flox] strain. Moreover, our control animals harbor both Pax7-Cre-ERT2 and the tdTomato transgenes. The observed difference in 1xTAM most likely reflects spontaneous activation and fusion of these cells following loss of Poglut1, as evidenced by the enhanced appearance of Tomato+ fibers in i-cKO muscles compared to control muscles. At 1xTAM, we can observe a modest increase in Tomato⁺ The reduction becomes more pronounced with repeated tamoxifen, consistent with progressive differentiation of the satellite cell pool and their fusion with myofibers.

• For Figure 5E (now Figure 6C), we made some changes to make the graph clearer and added a sentence to the figure legend to make it easier to follow.

  1. Densitometry for Figure 6A should be included, since N1 ICD levels seem quite variable in the controls also. POGLUT1 expression and shRNA knockdown efficiency should be shown in Figure 6B. What is the correlation between the NOTCH3 glycosylation with reduced NOTCH pathway activation and satellite cell function? This should be clearly discussed. Why was the NOTCH3 glycosylation assay done in HEK293 cells?

• In the revised version, we have included densitometry for Figure 6A (now Figure 7A).

• In the revised version, we show qRT-PCR data indicating that C2C12 cells stably expressing Poglut1 shRNA show ~ 75% decrease in Poglut1 mRNA levels compared to control cells (now Figure 7B), in agreement with our original report of these cells (PMID 21490058). We also performed qRT-PCR for Pax7 and found a significant reduction in Pax7 mRNA levels in Poglut1-knockdown C2C12 cells, further supporting their usage in these signaling assays.
• As discussed in detail in the Results and Discussion sections of the original submission, reports from our lab and others have provided strong evidence that glycosylation of NOTCH1 and NOTCH2 by POGLUT1 promotes signaling mediated by these receptors. Since it was not clear whether this phenomenon is ligand-specific, in the current manuscript we showed that both DLL1-madiated and JAG1-mediated signaling by NOTCH1 and NOTCH2 require the expression of POGLUT1 in the signal-receiving cells. Since NOTCH3 also plays a key role in muscle stem cell biology (albeit a not so well characterized one), we showed in the current manuscript that reducing POGLUT1 in the NOTCH3-expressing cell also leads to a significant reduction in signaling mediated by this receptor. Moreover, we found that NOTCH3 is broadly glycosylated by POGLUT1. In fact, during the revision, we succeeded in confirming four additional POGLUT1 target sites on the mouse NOTCH3 protein (added to the revised Figure 7D and Supplementary Figure S4). These data suggest that similar to NOTCH1 and NOTCH2, POGLUT1-medaited O-glucosylation of NOTCH3 is required for its signaling. To address the reviewer's comment, in the revised manuscript, we have added the following phrase to the Discussion to address the reviewer's comment: "..., strongly suggesting that O-glucosylation of NOTCH3 by POGLUT1 promotes its ligand-mediated activation."
• HEK293 cells express robust levels of POGLUT1 and have been used in previous studies for the characterization of POGLUT1-mediated glycosylation of its target proteins. Analysis of overexpressed target proteins in this cell line is used to establish assay conditions and glycopeptide identification parameters and thereby sets the stage to analyze glycosylation of endogenous protein (in this case, mouse NOTCH3) in specific cell types. We note that in cases examined so far, the glycosylation data obtained from overexpressed proteins in HEK293 or other commonly used mammalian cell lines recapitulate the glycosylation of endogenously expressed POGLUT1 targets very well (both in cell culture and in vivo; for example see PMID 27268051).

  Reviewer #1 (Significance (Required)):

  Based on my expertise as a muscle and stem cell biologist, the manuscript is not clearly thought through, with not many novel inferences that one can draw from the data provided. While the manuscript could be informative to muscle biologists and stem cell investigators, several additional experiments are required to better characterize the phenotypes and provide meaningful conclusions from the study. The role of POGLUT1 in the muscle could be of great interest, especially in light of its role in LGMD-R21, as described by some of the Authors previously. Several pieces of data provided in the current manuscript are disjointed, with few connecting links, such as the NMJ characterization, the NOTCH glycosylation data and the regeneration experiments done on P21 neonates. Better quantitation of data is required as well, as detailed below. Overall, the manuscript may be revised to address the specific comments and reconsidered at a suitable journal.

Response: We hope that our explanations, additional experiments, and changes to the text have helped address the concerns raised by the reviewer.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

Poglut1 mutations have been identified in an autosomal recessive form of muscular dystrophy. Poglut1 encodes an O-glucosyltransferase that modifies Notch receptors and ligands but also has other substrates. In mice, null mutations in Poglut1 are embryonic lethal, which has previously precluded the analysis of Poglut1's role in muscle development and regeneration using murine models. To address this limitation, the authors generated a conditional Poglut1 allele and introduced mutations using Pax7Cre or Pax7CreERT2. They characterize the phenotypic consequences of these mutations and further demonstrate that signaling through the Notch1, Notch2, and Notch3 receptors requires Poglut1 using the C2C12 cell culture model.

Some aspects of the manuscript's description of muscle stem cell behavior and regeneration are not fully up to date. These points should be addressed before publication.

Specific points

1) Evidence Supporting Reduced Notch Signaling as the Cause of the Phenotype The comparison of muscle phenotypes observed in other Notch signaling mutations strongly supports the hypothesis that the phenotype is due to reduced Notch signaling. Relevant studies, such as those addressing development (PMID: 17194759, PMID: 17360543) and those focusing on adult muscle and regeneration (PMID: 21989910, PMID: 22069237, PMID: 22045613, PMID: 30862660) should be cited and discussed in the manuscript. Including these references will strengthen the argument and provide a broader context for the findings.

• We thank the reviewer for this comment. To address this point and a similar comment made by reviewer 3, we have modified the Introduction and Discussion in the revised manuscript. In the Introduction, we have added the following sentences and have cited all 6 references listed by the reviewer: "Loss of function studies for key components of the Notch signaling pathway including Rbpj and Dll1 indicated that disruption of this pathway during muscle development leads to premature differentiation and depletion of muscle progenitor cells (Schuster-Gossler et al., 2007; Vasyutina et al., 2007). In addition, multiple studies have shown that Notch signaling is required in adult mice to prevent the spontaneous or premature differentiation of satellite cells and to maintain a muscle stem cell pool capable of repairing muscle damage (Bjornson et al., 2012; Fukada et al., 2011; Lahmann et al., 2019; Mourikis et al., 2012)." In the third paragraph of the Discussion, we have highlighted the similarities between Poglut1-cKO and i-cKO phenotypes and the phenotypes observed in "animals with germline or conditional loss of various Notch pathway components (Bjornson et al., 2012; Fukada et al., 2011; Lahmann et al., 2019; Mourikis et al., 2012; Schuster-Gossler et al., 2007; Vasyutina et al., 2007)".

  2) Figure 1: NMJ Deficits - Pre- or Postsynaptic? The authors describe the mutant synaptic vesicles as showing a significantly reduced evoked neurotransmitter release (quantal content) compared to controls. This phrasing raises the question: are motor neurons mutated in these animals? It should be clarified why the synaptic vesicles are referred to as "mutant." To my knowledge, Pax7Cre does not recombine in motor neurons, and this discrepancy needs to be addressed. The text should be rephrased to accurately reflect the origins of the observed deficits.

• We are not aware of any reports on recombination in motor neurons by the Pax-Cre line used in our study and had implied this in the initial submission in the following sentence in Discussion: "Therefore, even if agrin is indeed glycosylated by POGLUT1, loss of Poglut1 with Pax7-Cre is not expected to affect the glycosylation of agrin expressed by motoneurons." [please note that given the new data and text added to the revised manuscript, we removed the paragraph related to NMJ and agrin from the Discussion to reduce the manuscript length.] We agree that referring to synaptic vesicles as "mutant" can be misleading and have revised "mutant synaptic vesicles" to "in the NMJs of Poglut1-cKO LAL muscles". To address the discrepancy, we have added the following sentence to the Results section after describing a reduction in quantal content: "Since the Pax7-Cre strain used in our study is not reported to induce recombination in motor neurons (Murdoch et al., 2012), this presynaptic NMJ defect might be secondary to defects in postsynaptic NMJ abnormalities."

  3) Quantification of Myofibers with Internal Nuclei The statement, "We first quantified the ratio of myofibers with internal nuclei, which is an indication for recent fusion of myoblasts to myofibers," is not entirely accurate. Recent studies, such as PMID: 38569550, provide a more nuanced explanation of this phenomenon. The manuscript should reference this study and update the description to ensure it accurately reflects the current understanding of myofiber internal nuclei as markers of muscle pathology or regeneration.

• We thank the reviewer for bringing this paper to our attention. In the revised manuscript, we have changed the above sentence to make it aligned with the observations of the paper mentioned by the reviewer: "We first quantified the ratio of myofibers with internal nuclei, recently reported to specifically result from the fusion of embryonic myogenic cells during limb myogenesis and also driven by myocyte-myocyte fusion in the first phase of postnatal muscle regeneration (Collins et al., 2024)." We also added the following sentence, which we believe makes our interpretation more accurate in light of Collins et al, 2024: "These data suggest enhanced differentiation of Poglut1-deficient progenitors into myocytes followed by continued myocyte-myocyte fusion and/or a delay in the peripheral migration of internal nuclei which normally occurs in the perinatal period (Collins et al., 2024)." This way, in the revised manuscript we do not link the presence of internal nuclei with recent fusion of myoblasts anymore. Finally, we also referred to Collins et al 2024 when describing the appearance of internal nuclei in injured muscles 5 days after injury (Figure 4A).

  4) Figure 3H: Cell Identification in Culture

  The use of PAX7+ MYOD− to identify quiescent cells in culture is not accurate. Instead, PAX7+ Ki67− should be used for this purpose. Similarly, PAX7− MYOD+ does not reliably identify differentiating cells. Instead, staining for MYOG should be used to ensure accuracy. The figure and accompanying text should be adjusted accordingly to reflect these updates.

• To address this issue and a comment by reviewer 1, we performed PAX7/MYOD/Ki67 co-staining on cultured cells and quantified the percentage of each cell state based on the expression of these three markers as a more accurate measure of quiescent versus cycling satellite cells, as well as progenitors and precursor cells. The revised quantification and statistical analysis is presented in the revised Figure 3H. This quantification is represented similarly to the data shown in Gattazzo et al 2020. Based on the reviewer's comment and the results of these quantification, we have modified the Results section as follows (new text is underlined): "These observations indicate that loss of Poglut1 impairs the ability of muscle stem cells to remain in a quiescent state and suggest that the mutant myogenic progenitors might undergo premature differentiation." We then present the fusion index and myogenin staining data to conclude enhanced differentiation.

  5) Description of Satellite Cell Quiescence

  The statement, "About 2-3 weeks after birth, some of the PAX7+ cells generated by active proliferation of embryonic myogenic progenitors enter a quiescent state to generate adult satellite cells," is not entirely correct. The description should be updated based on the findings in PMID: 32763161, which provide a more accurate account of the transition of PAX7+ cells to quiescence and their role in generating adult satellite cells.

• We thank the reviewer for bringing this paper to our attention. To address this issue and one of the concerns raised by Reviewer 1, we have made the following changes to the corresponding paragraph in the revised version: (A) We removed the statement highlighted by the reviewer. (B) We provided better justification for using P21 mice to assess muscle regeneration by quiescent satellite cells by adding the underlined sentences to this section: "To address this question, we induced muscle injury in mutant and control mice by injecting cardiotoxin (CTX) into TA muscles at P21 (Fig. 4A), an age at which we consistently obtain Poglut1-cKO animals without any dietary changes (Fig. 1E). Importantly, 51% of PAX7+ cells are reported to be in the quiescent state at P21 (Gattazzo et al., 2020) and the TA muscles of P21 mice exhibit a robust regenerative response to cardiotoxin-induced injury (Lepper et al., 2009)." We note that Gattazzo et al 2020 is PMID: 32763161, the paper with a more accurate account of satellite cell quiescence mentioned by the reviewer.

  6) Figure 5: Loss of Quiescence in Satellite Cells

  A hallmark phenotype of mutations in Notch signaling genesis the loss of quiescence in satellite cells when the mutation is introduced in the adult,. The authors should include data on Ki67 and MyoD expression in PAX7+ cells of mice with the Poglut1 mutation introduced in the adult by an analysis of the uninjured muscle. This would provide insight into the maintenance of quiescence in the mutant satellite cells.

• Based on the reviewer's recommendation, we aimed to assess Ki67 and MyoD expression in PAX7⁺ cells of adult inducible-cKO (i-cKO) mice in the uninjured muscle. However, due to technical limitations including antibody species incompatibility and the number of available fluorescence channels, we were unable to perform simultaneous triple staining combined with tdTomato visualization on the same tissue sections. Instead, we utilized the Tomato reporter as a marker for PAX7⁺ cells, based on new data indicating that 93-97% of TOM⁺ cells co-express PAX7 (Figure 6E). We performed Ki67 and MyoD double staining and quantified the percentage of Tomato⁺ cells that expressed one or both of these markers. Our analysis showed that TOM⁺/Ki67⁻/MyoD⁻ (quiescent) satellite cells were significantly reduced in the i-cKO mice compared to controls, while the proportions of single-positive cells (Ki67⁺ or MyoD⁺) were increased (Figure 6F-G). These findings are consistent with loss of quiescence and increased activation of satellite cells upon loss of POGLUT1 in the adult mice.

• During the revision, we performed additional experiments (not suggested by the reviewers) to better assess the role of Poglut1 in self-renewal of satellite cells upon injury. First, in a new cohort of i-cKO and control animals, we performed two rounds of tamoxifen injections but skipped the first round of injury and only induced CTX injury after the second round of recombination. Interestingly, analysis of 14 dpi muscles from these animals showed a full repair in both control and i-cKO mice (shown in revised Figure 5B). Second, we inspected the repaired control and i-cKO muscle after one round of recombination + injury for the presence of single tdTomato+ cells next to the repaired myofibers, which would represent the satellite cells formed after the repair. While single tdTomato+ cells were readily seen in the repaired muscle in control mice, we did not see any such cells next to the repaired muscle from i-cKO animals (shown in the revised Fig 5C, n = 4 animals per condition). Together, these new data provide strong evidence that loss of Poglut1 in adult muscle stem cells impairs their ability to return to quiescence and form new satellite cells upon repair of injured muscle.

  7) Figure 6A: Assay of Cleaved NOTCH1 Intracellular Domain

  The text describes that satellite cells isolated from Poglut1-cKO muscles showed a strong reduction in the level of cleaved (active) NOTCH1 intracellular domain compared to control cells. It is unclear whether these satellite cells were freshly isolated and directly assayed or cultured before the assay. Please specify this in the result section.

• These satellite cells were freshly isolated from whole muscle, and the proteins were extracted directly from the isolated satellite cells without culturing them. This is now included in the Results section related to the revised Figure 7A.

  8) Figure 6: Notch Receptor Overexpression in C2C12 Cells The results section should explicitly explain that the Notch 1-3 receptors were overexpressed or transfected in the C2C12 cells. This detail is essential for understanding the experimental design.

• Thank you. The C2C12 cells were transfected with mouse NOTCH receptors 1, 2, or 3 to overexpress each receptor. This is now included in the Results section as suggested.

  Reviewer #2 (Significance (Required)):

  The manuscript by Cho et al. demonstrates that the muscular dystrophy phenotype associated with Poglut1 mutations is caused by a Notch signaling deficit in muscle stem cells. Although previous studies had suggested this connection, alternative mechanisms could not be excluded, as Poglut1 glycosylates a multitude of proteins. Overall, this is a thorough and careful analysis of Poglut1's role in muscle development and regeneration, providing valuable insights into the mechanisms underlying this rare muscle disease.

Response: We sincerely thank the reviewer for her/his positive assessment of our manuscript and for the constructive comments.

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

POGLUT1 is a key enzyme essential for the glycosylation and signaling activity of Notch receptors in C2C12 muscle cells. Elegant work from multiple groups has demonstrated the essential role of Notch signaling in maintaining the quiescence of muscle stem cells known as satellite cells. For instance, genetic deletion of Rbpj in tamoxifen-inducible conditional knockout mouse models causes ectopic expression of MyoD, diminished Pax7 expression, and precocious differentiation and fusion of quiescent satellite cells without entering the cell cycle. This is accompanied by the complete loss of tissue regeneration following repetitive muscle injuries. Hypomorphic mutations of POGLUT1 cause limb-girdle muscular dystrophy (type R21), accompanied by a significant reduction of Notch signaling and a decrease in the number of satellite cells in patient muscles.

Despite the substantial body of evidence establishing the key function of the POGLUT1-Notch-Pax7/MyoD axis in satellite cells, the in vivo function of POGLUT1 in mice has not been studied. In the present study, Cho et al. carefully examined muscle development and regeneration using the POGLUT1 conditional knockout mouse model. The results, at both phenotypic and cellular/molecular levels, align perfectly with the previously established working model of the POGLUT1-Notch-Pax7/MyoD axis. Overall, the experiment is well-designed and executed, and the data are generally of high quality supporting the conclusions. No specific experimental issues was identified.

Major suggestions include adding references in the discussion section: 1) studies of Rbpj cKO in mice (PMID: 22069237; 22045613), as genetic deletion of POGLUT1 using the same Pax7 driver showed almost identical phenotypes across all levels: phenotype, cellular fate changes, and gene expression changes; 2) studies of Notch-Rbpj targets (Fig. S1, PMID: 29795344), as this paper identified multiple genes encoding collagen V and VI as direct targets of Notch signaling in quiescent satellite cells. These findings are consistent with the authors' observations in Fig. 2e.

• We thank the reviewer for the suggestion to add these highly relevant papers to the manuscript. 1) To address this comment and the specific point #1 raised by reviewer 2, we have highlighted the similarities between Poglut1-cKO and i-cKO phenotypes and the phenotypes observed in "animals with germline or conditional loss of various Notch pathway components (Bjornson et al., 2012; Fukada et al., 2011; Lahmann et al., 2019; Mourikis et al., 2012; Schuster-Gossler et al., 2007; Vasyutina et al., 2007)" in the Discussion. Mourikis et al 2012 and Bjornson et al 2012 are the two papers that the reviewer has referred to. 2) To address this point, we have added the following sentences to the Discussion: "Notch signaling has been shown to directly activate the transcription of several genes encoding collagen V and VI in muscle stem/progenitor cells, and collagen V expressed by satellite cells plays a key role in the niche to maintain satellite cell quiescence (Baghdadi et al., 2018). Therefore, ECM abnormalities caused by reduced Notch signaling might contribute to the loss of satellite cell quiescence in i-cKO animals as well." Baghdadi et al, 2018 is PMID: 29795344, to which the reviewer has referred.

  Minor suggestions: 1B: "P0" is an awkward term. Does it refer to newborn day 1 or a near full-term embryo?

• By saying "P0", we are referring to the day the litters are born.

  Fig. 1C: The y-axis label includes an extra space.

• The extra space is removed in the revised figure.

  Fig. 1D: Provide the N number for each genotype/diet group.

• The N number has been added (now Figure 1E).

  Gene nomenclature: Use POGLUT1 exclusively for the human protein and Poglut1 for the mouse protein.

• According to the guidelines of the International Committee for Standardized Genetic Nomenclature for Mice, protein symbols for mice should use all uppercase letters. Therefore, while the gene symbols for mouse and human are different in terms of uppercase versus lowercase usage, the protein symbols for mouse and human are usually identical. Please see the following webpage: https://www.informatics.jax.org/mgihome/nomen/gene.shtml#ps (accessed January 18, 2025).

  Reviewer #3 (Significance (Required)):

  In the present study, Cho et al. carefully examined muscle development and regeneration using the POGLUT1 conditional knockout mouse model. The results, at both phenotypic and cellular/molecular levels, align perfectly with the known working model of the POGLUT1-Notch-Pax7/MyoD axis in satellite cells and muscle regeneration. Overall, the experiment is well-designed and executed, and the data are generally of high quality.

Response: We sincerely thank the reviewer for the positive evaluation of our manuscript as well as the helpful suggestions.