Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Autophagy of the endoplasmic reticulum (ER-phagy) is a fundamental process that is essential for maintaining cellular homeostasis and quality control. We recently identified a novel mechanism regulating ER-phagy in both plants and animals that is based on the ubiquitin-like protein modifiers ATG8 and UFM1, and the ER-associated protein, C53. Here, we use a combination of evolutionary, biochemical, and physiological experiments to investigate the evolution and regulation of this process. We reveal the dynamic evolution of UFM1 and the ubiquity of C53-mediated autophagy across eukaryotes. Leveraging these results, we then identify an ancestral molecular toggle switch, mediated by shuffled ATG8-interacting motifs (sAIMs), that controls C53-mediated autophagy through competitive binding between UFM1 and ATG8. These findings provide new insights into the evolution of UFM1, reveal a conserved mechanism for the regulation of ER-phagy, and raise new and exciting hypotheses about the diversity and function of the UFMylation pathway. We believe that this work will be of interest to those studying autophagy and cellular stress response but will also serve as an interesting example of the benefits of combining evolutionary analyses with biochemical and cellular experiments.

Our manuscript has been reviewed by three reviewers through ReviewCommons, whose comments, and our responses, can be found below. Two of the reviewers (Reviewer 1 and 3) were supportive of our work and its significance whereas Reviewer 2 questioned the novelty of our findings.

Each of the reviewers’ comments can be addressed through a few supporting experiments as well as an improved manuscript which clarifies the novelty and significance of our results. While being supportive of our work, Reviewer 1 requested minor additional experiments to support our mechanistic conclusions and Reviewer 3 suggested that we expand our characterizations of C53 function to additional eukaryotic supergroups. These experiments are straightforward to perform, the materials and protocols to accomplish them are already established, and our overall conclusions are robust to the resulting outcomes.

In contrast, Reviewer 2 did not suggest any additional experiments but rather challenged the novelty of our results as well as some of our interpretations. In particular, Reviewer 2 was uncertain of how our phylogenomic analyses built upon a previous study, published in 2014, which used comparative genomics to identify ubiquitin-related machinery across eukaryotes. Although it was an oversight to not reference this study (we cited a more recent article showing the same results), we were aware of their conclusions that UFMylation was present in the last eukaryotic common ancestor but absent in Fungi. We now clearly outline, both below and within the manuscript, our key phylogenomic results. These were acquired after implementing more advanced and comprehensive comparative genomic searches which allowed us to identify dynamic patterns in UFMylation evolution and permitted co-evolutionary analyses which were not only important for informing our experimental hypotheses but generated new functional questions. Our phylogenomic analyses are also linked to biochemical and physiological data, providing, for the first time, experimental support for our conclusions regarding UFMylation evolution. Similarly, Reviewer 2 suggested that our mechanistic results were an incremental extension of our previous work. Although our current work does of course build on our initial identification of C53-mediated autophagy, this manuscript provides novel insights into the importance and function of this process by revealing its ubiquity across eukaryotes and by characterizing the mechanistic details of its regulation. Ultimately, we disagree with Reviewer 2 but appreciate that this misunderstanding likely resulted from a lack of context and clarity in our manuscript which we have now resolved.

As outlined in detail below, we will address the reviewers concerns through additional experiments, analyses, and improvements to the text.

Thank you for considering our manuscript. We look forward to hearing from you.

Description of the planned revisions

We thank the reviewers for carefully evaluating our manuscript and for providing us with an opportunity to respond to their suggestions and criticisms. As you can see below in our pointby-point response, we address each of the points raised by the reviewers through the addition of supporting experiments, analyses, and an improved text. Altogether, we think these additional experiments and textual changes will significantly improve the manuscript. Therefore, we would like to thank all the reviewers and editors for their time and input.

Referee #1

Evidence, reproducibility and clarity

In this manuscript Picchianti et al. provide novel insights into the interaction of C53 with UFM1 and ATG8. Initially, the authors show that protein modification by UFM1 exists in the unicellular organism Chlamydomonas reinhardtii. To that end they demonstrated that pure Chlamydomonas UBA5, UFC1 and UFM1 proteins, can charge UFC1. Then, they showed that C53 interacts with ATG8 and UFM1. Specifically, they found that the sAIM are essential for the interaction with UFM1, while substituting this motif with canonical AIM prevents the binding of UFM1 but not of ATG8. Since binding of C53 to ATG8 recruits the autophagy machinery, the authors suggest that ufmylation of RPL26 releases UFM1 from C53 which allows the binding of ATG8. Overall, the authors demonstrate that C53 that forms a complex with UFL1 connects between protein ufmylation and autophagy by its ability to bind both UBLs. Here the authors revisited the assumption that only multicellular organisms have the UFM1 system. Using bioinformatic tools they show that it exists also in unicellular organism. Also, they show that in some organisms the E3 complex UFL1, UFBP1 and C53 exist but not UBA5, UFC1 or UFM1. This is a very interesting observation that suggests an additional role for this complex. In Fig 1C the authors show that in Chlamydomonas RPL26 undergoes ufmylation. Please use IP against RPL26 and then a blot with anti UFM1. From the current experiment it is not clear how the authors know that this is indeed RPL26 that undergoes ufmylation

RPL26 is highly conserved across eukaryotes, so by comparing our western blots with previous studies (Walczak et. al., 2019, Wang et al. 2020), we concluded that these bands corresponded to UFMylated RPL26. However, we agree with the reviewer that we need to confirm the identify of RPL26 with additional assays. Since the submission of the manuscript, we tested RPL26 antibodies in Chlamydomonas and showed that they work well. So, we will update our figure with the confirmation westerns.

In the second part of the manuscript the authors characterize the interaction of C53 with ATG8 and UFM1. This is a continuation of their previous published work (Stephani et al, 2020). Here the reviewer thinks that further data on the binding of these proteins to C53 is required. Specifically, defining the Kd of these interactions using ITC or other biophysical method can contribute to the study.

We agree with the reviewer. To obtain the KD values, we will perform ITC experiments with C53 wild type, a C53 sAIM mutant and a C53 cAIM variant titrated with ATG8 and UFM1.

Under normal condition the authors suggest that C53 binds UFM1 and this keeps it inactive. The reviewer thinks that this claim needs further support. Using IP (maybe with crosslinker) the author can show that C53, in normal conditions, bind more UFM1 than ATG8. Also, since the interaction of UFM1 to C53 is noncovalent, it will be nice to show how alternations in UFM1 expression levels can affect the activation of C53.

We thank the reviewer for this suggestion. Since the submission of the manuscript, we have obtained UFM1 overexpression lines. We will pull on C53 using our C53 antibody and check for ATG8 levels in wild type and UFM1 overexpressing lines under normal and stress conditions. We think this will show how alterations in UFM1 levels can affect C53 activation.

Finally, the authors suggest that ufmylation of RPL26 allows binding of ATG8 to C53 and this, in turn, leads to C53 activation. Can the authors show that in cells lacking UBA5, under normal condition or with Tunicamycin treatment, ATG8 does not activate C53 due to the fact that UFM1 does not leave C53.

In Stephani et al., we showed that C53-mediated autophagy requires the UFMylation machinery. In ufl1 and ddrgk1 mutants, C53 becomes insensitive to ER stress. However, to supplement these results, we will perform autophagic flux assays using the native C53 antibody to test autophagic degradation of C53 in a uba5 and ufc1 mutant under normal and tunicamycin stress conditions. The uba5 mutant that we have is a knockdown, so that’s why we will include the ufc1 mutant in our experiments.

Significance

This manuscript advances our understanding of the connection of ufmylation to autophagy which is mediated by C53.

Thank you!

Referee #2

Evidence, reproducibility and clarity

The manuscript from Picchianti et al. seeks to define the role of CDK5RAP3 (hereinafter referred as C53) during autophagy and its interplay with UFMylation. Together with UFL1 and DDRGK1, C53 is a component of a trimeric UFM1 E3 ligase complex that modifies the 60S ribosomal protein RPL26 at the endoplasmic reticulum (ER) surface upon ribosomal stalling (among other proposed functions that are not addressed). Several previous studies have implicated the UFMylation pathway in autophagy or ER-phagy although a non-autophagic fate for UFM1- tagged ribosomal subunits has also been reported. A previous study from the same authors (PMID: 32851973) identified an intrinsically disorder region (IDR) in C53 that is necessary and sufficient for interaction between C53 and autophagy receptor, ATG8. They reported that this IDR comprises four non canonical ATG8 interacting motifs (AIM), named shuffled AIMs (sAIMs) and showed that combinatorial mutagenesis of sAIM1, sAIM2, and sAIM3 abrogates ATG8 binding. A similar effect was observed for plant C53, though an additional canonical AIM (cAIM) in the C53 IDR had to be mutated to completely abolish C53 and ATG8 interaction. The earlier study reported that C53 IDR also interacts with UFM1, and this interaction can be disrupted in vitro by adding increasing concentration of ATG8, suggesting that ATG8 and UFM1 may compete with one another for C53 binding. The present paper attempts to build on this previous work by using phylogenomics to infer a coevolutionary relationship between UFMylation machinery and sAIMs in C53, which the authors argue, constitutes further evidence of the primary importance of a role for UFMylation in ER homeostasis. The manuscript includes a lot of biochemical data using variations of in vitro and in vivo pull-down experiments to define the roles of individual AIMs in mediating the binding of C53 to ATG8 and to UFM1. They also use NMR spectroscopy in an attempt to define the structural basis of the UFM1 and ATG8 binding to C53, concluding that plant C53 interacts with UFM1 mainly through sAIM1, while interaction with ATG8 requires cAIM as well as sAIM1 and sAIM2. Finally, the authors attempt to contextualize these findings by conducting studies on Arabidopsis mutants, showing that replacing sAIMs with cAIMs causes increases sensitivity to ER stress and apparently increases formation of C53 intracellular puncta that may colocalize with ATG8. From these data the authors concluded that the dual-ATG8 and UFM1 binding of C53 IDR regulates C53 recruitment to autophagosomes in response to ER stress. Major Issues: 1) The phylogenomics analysis conclusion that UFM1 is common in unicellular lineages and did not evolve in multicellular eukaryotes is not novel, as another comprehensive analysis of UFM1 phylogeny, published eight years ago - in 2014 - by Grau-Bové et al. (PMID: 25525215), also reported that UFM1, UBA5, UFC1, UFL1 and UFSP2 were likely present in LECA and lost in Fungi. Although the phylogenomic analysis by Picchianti et al. is also extended to DDRGK1 and C53 proteins, and some parasitic and algal lineages, their findings are incremental. Their proposed coevolution of sAIM and UFM1 is based on presence-absence correlation observed within five species (i.e., Albugo candida, Albuco laibachii, Piromyces finnis, Neocallimastix californiae, Anaeromyces robustus). However, this coevolutionary relationship must be further investigated by substantially increasing the taxonomic sampling within the UFM1-lacking group.

We were aware that previous studies had investigated the distribution of UFMylation proteins across eukaryotes and that these analyses had predicted the presence of UFMylation in LECA and subsequent loss in Fungi. We included a more recent citation noting this (Tsaban et al. 2021) but apologise for not citing Grau-Bové et al. (2014), which we have now included. We must emphasize that our results are not incremental. Although we had made a point of emphasizing the presence of UFM1 in LECA, this was to counter a recent and highly cited paper in the field which claimed that UFMylation evolved in plants and animals (Walczak et al. 2019). Below we note the novel and important results from our phylogenomic analyses: 1. We used improved taxonomic sampling and more advanced comparative genomics methods to identify UFMylation components sensitively and specifically across eukaryotes. This involved the inclusion of additional eukaryotic genomes, phylogenetic annotation of orthologs, and genomic searches to complement proteome predictions. These methods are essential for accurately identifying UFMylation components and yield more robust results than using sequence similarity clustering (Tsaban et al. 2021) or un-curated Pfam HMMER search results (Grau-Bové et al. 2014). 2. By placing our UFMylation reconstructions in a modern phylogenetic context we were not only able to support previous observations which noted the presence of UFM1 in LECA and its loss in Fungi (Grau-Bové et al. 2014) and Plasmodium (Tsaban et al. 2021), but also to identify novel patterns in the evolution of UFMylation. This included the observation of recurrent losses in diverse but trophically-related lineages (such as algae and parasites) and revealed the retention of certain UFMylation components in the absence of UFM1. We identified the frequent coretention of UFL1 and DDRGK1 following UFM1 loss in multiple eukaryotic groups, including Fungi, which were previously thought to be devoid of UFMylation machinery. These previously uncharacterized patterns, suggest that these proteins could have alternative functions and may be functionally associated with life history. These results therefore expand on and add complexity to our understanding of the evolution of UFMylation. 3. By conducting a comprehensive and accurate survey of UFMylation components we were able to use our data to examine co-evolutionary trends between C53 and UFM1, which would have been incomplete and inaccurate using previously curated datasets. As the reviewer noted, only five species were identified that encoded C53 but lacked UFM1. This is not a reflection of insufficient taxon sampling, but rather the strong co-evolution between C53 and UFM1 (i.e., when UFM1 is lost, C53 is almost always lost as well). We attempted to identify additional cases by searching hundreds of fungal and oomycete genomes as well as those from other eukaryotes, but no other species were found. We agree with the reviewer that additional taxa would have made our analyses stronger, but importantly, we do not rely on genomic correlations to infer function. Rather, we use these correlations to generate functional hypotheses which we then tested experimentally. In this way, we do not rely on the strength of our correlations. We have now revised the manuscript to include additional context (including citations) and have improved the clarity of the text to better convey the novelty of our findings.

2) The manuscript presents an overwhelming amount of biochemical and structural data obtained from a variety of protein binding techniques (i.e., NMR spectroscopy, in vitro GSTpulldown, fluorescence microscopy-based on-bead binding assays, and native massspectrometry). The results are poorly explained and not organized in a logical manner. Moreover, no attempt was made to explain the rationale behind using one technique over the other or how one method complements another to build a stronger conclusion than any individual approach. Given that none of the methods employed report quantitative measurement of binding affinities between C53 IDR and UFM1 or ATG8, it is not clear how the data presented in this manuscript contribute to our understanding of the proposed competition model for UFM1 and ATG8 binding to C53 IDR. To conclude that an interaction is "stronger" or "weaker" it is necessary to measure equilibrium binding constants. Fortunately, there are suitable techniques, including surface plasmon resonance (SPR), microscale thermophoresis (MST), fluorescence anisotropy, or calorimetry that are available to dissect these complex competitive binding interactions and to build models.

We thank the reviewer for their suggestion. Although we attempted to describe the rationale behind each experiment (please see the line 135-137; on-bead binding assays, line154-157; NMR, 177-181), we agree that the volume of data and variety of techniques warrants additional explanation. We will revise the manuscript to further explain our rationale for using each of the different approaches. As we noted above in our response to reviewer 1, we will also perform relevant ITC binding assays to quantify the interaction between C53, ATG8, and UFM1.

3) The NMR studies have the potential to dissect the types of dynamic binding inherent in unstructured proteins. However, the abundant NMR data presented combined with the aforementioned binding studies, remarkably, do not seem to significantly advance our understanding of how the system is organized or even how UFM1 and ATG8 bind C53, beyond the rather vague and somewhat circular conclusion stated in the abstract: "...we confirmed the interaction of UFM1 with the C53 sAIMs and found that UFM1 and ATG8 bound the sAIMs in a different mode." Or on line 165 "Altogether these results suggested that ATG8 and UFM1 bind the sAIMs withn C54 IDR, albeit in a different manner".

We agree that NMR has the potential to dissect the complex binding interactions between UFM1, ATG8, and C53, but disagree with the reviewer’s interpretation that our NMR data fail to achieve this. To sum up, our NMR data: 1. Revealed the structural basis of the interaction of C53-IDR with ATG8 and UFM1 at atomic resolution by showing that UFM1 binds preferentially to sAIM1 in the fast-intermediate exchange [Fig.4 and Fig. S7B], instead ATG8 binds cAIM in the slow-intermediate exchange, and once cAIM is occupied, it binds sAIM1,2 with lower affinity in the fast-intermediate exchange (Fig.4 and Fig.S7D). 2. Determined conformational changes in C53 IDR upon binding of ATG8, but not UFM1 (Fig.S7E), which lead to increased dynamics in distinct regions in C53 IDR. These data could explain how binding of first ATG8 would trigger C53-dependent recruitment of the tripartite complex to autophagosomes. 3. Identified how UFM1 binds to atypical hydrophobic patch in C53 sAIM, similar to what was shown for the UBA5 LIR/UFIM. To sum up, our results shed light on how both UBLs interact with C53, being sAIM1 the highest affinity binding site for UFM1 while ATG8 binds cAIM preferentially before occupying sAIM1,2. To provide more detailed information on the atomic details of the interaction between C53 and the UBLs, we will perform molecular docking studies by using the restraints obtained from the experimental NMR data.

4) The functional assays performed in Arabidopsis do not support the competitive model between UFM1 and ATG8 for binding to C53 during C53-mediated autophagy. The fluorescence microscopy images do not provide convincing evidence of colocalization between C53 and ATG8. In fact, in contrast to the claims made in the text or the quantification, mCherry-C53 fluorescence does not seem to localize in discrete puncta and its signal does not seem to overlap with ATG8A.

We disagree with the reviewer’s interpretation of these results although we acknowledge that there is some subtlety in interpreting the co-localization data. Importantly, Arabidopsis has 9 ATG8 isoforms and C53 can bind to most of them with varying affinities (see Stephani et al). Because of this, we do not expect C53 puncta to fully colocalize with ATG8A puncta. Additionally, the C53 puncta are smaller and more subtle than ATG8 puncta, which label the entire autophagosome. To reconcile this, we will quantify the effect by performing colocalization analyses under normal and stress conditions. We will also upload all the raw images as supporting material, so that anyone can independently assess our images.

Minor Issues: 1. The authors might choose to avoid teleological arguments such as (line 135): "As the phylogenomic analysis suggested that eh sAIMs have been retained to mediate C53-UFM1 interaction..."

We thank the reviewer for this suggestion and will modify the text accordingly.

1. The authors refer on multiple occasions to C53 "autoactivation" without defining what they mean by this. Do they propose that C53 UFMylates itself?.

We refer to C53 activity as the ability to recruit the autophagy machinery and initiate cargo sequestration and degradation in the vacuole. We attempted to explain this in lines 57-61 but we will reword it more clearly, as suggested by the reviewer.

1. The paper might want to avoid preachy philosophical statements like "Our evolutionary analysis also highlights why we should move beyond yeast and metazoans and instead consider the whole tree of life when using evolutionary arguments to guide biological research." (333- 335). While this is indeed a laudable goal, given the rather limited insights from this study, it is unclear how this paper exemplifies the notion.

We added this statement as we were intrigued by our evolutionary analyses’ ability to link C53 to UFM1 (an association which took years to identify experimentally) and generate useful functional hypotheses about the interaction between C53 sAIMs and UFM1. As we mentioned above, we also wanted to highlight this point in reference to a recent prominent study in the field which drew conclusions after only considering animals, plants, and fungi (Walczak et al., 2019). We believe this point is important and underappreciated by some cell biologists, but we will modify the text to make it more generic: “This work highlights the utility of using evolutionary analyses and eukaryotic diversity to generate mechanistic hypotheses for cellular processes”.

Significance

Overall, while the manuscript contains an abundance of new data, the overall conclusion of the work, stated in the title: "Shuffled ATG8 interacting motifs form an ancestral bridge between UFMylation and C53-mediated autophagy" does not constitute a significant advance beyond other published phylogenomic analysis (below) and the two previous papers by the same authors, including the 2020 paper "A cross-kingdom conserved ER-phagy receptor maintains endoplasmic reticulum homeostasis during stress (PMID: 32851973)" and the 2021 paper "C53 is a cross-kingdom conserved reticulophagy receptor that bridges the gap between selective autophagy and ribosome stalling at the endoplasmic reticulum PMID: 33164651)". While a regulatory interaction between UFMylation and autophagy is of potential importance, the data in this manuscript do not constitute a major advance and fail to provide new mechanistic insight to explain the role of C53 IDR in autophagy and its interplay with UFMylation

We disagree with the reviewer’s suggestion that our work does not constitute a significant advance. We outlined above in detail the novel insights that were obtained from our phylogenomic analysis which involved using improved methods to reveal a much more dynamic and informative picture of UFMylation evolution than has been described previously. Likewise, this manuscript builds substantially on our previous mechanistic work. In our 2020 paper (which is summarized in the mentioned 2021 review article), we identified C53 as an ER-associated protein that binds ATG8 through sAIMs and interacts with the phagophore after RPL26 UFMylation. This work linked C53 activity to ER-phagy and highlighted its importance in plant and animal stress response. However, key questions remained unanswered prior to our current work such as whether this mechanism is conserved across eukaryotes, especially in unicellular species, how C53 activity is regulated, and how UFM1 and ATG8 interact with C53. Our current manuscript builds on this work with the following key results: 1. We use a combination of phylogenomic and experimental analyses to demonstrate that C53 function is conserved across eukaryotes. 2. We reveal a mechanism whereby UFM1 and ATG8 compete for binding at the sAIMs in the C53 IDR and characterize how each of these ubiquitin-like proteins interacts in an alternative way (see the NMR results described above). 3. We show how the sAIMs are required for the regulation of C53-mediated autophagy and reveal the importance of UFM1-ATG8 competition in preventing C53 autoactivation, which causes unnecessary autophagic degradation and impairs cellular stress responses.

These insights are fundamental for understanding the mechanisms regulating C53-mediated autophagy which were unknown before this work. We will therefore adjust our manuscript to more clearly and explicitly explain how our data build on previous observations so that the novelty and significance of our results are clearer.

Referee #3

Evidence, reproducibility and clarity

Picchianti and colleagues have investigated a conserved molecular framework that orchestrates ER homeostasis via autophagy. For this, they have carried out phylogenomics and large-scale gene family analyses across eukaryote diversity as well as a barrage of molecular lab work. The amount of work carried out as well as the overall quality of the study is impressive.

Thank you!

I have only a few comments that should be very easy to tackle. (1) Maybe I missed it, but please upload all alignments used for phylogenetics and phylogenomics for reproducibility to e.g. Zenodo, Figshare or other suitable OA databases.

We included the alignments in the supplementary data, but as suggested, we will upload all the source data including the scripts and the alignments to Zenodo.

(2) "Why these non-canonical motifs were selected during evolution, instead of canonical ATG8 interacting motifs remains unknown" --> Maybe there is no "why" and these were not selected at all. Could be random... drift, non-adaptive constructive neutral evolution. I am not saying that asking "why" in evolutionary biology is wrong. It, however, often does not yield satisfactory answers--or any answer at all.

The reviewer is completely right that “why” is not the right way to frame an evolutionary question. Thank you for pointing this out. We will revise the text and make sure that we remove these kinds of deterministic statements.

(3) The authors make a case for UFMylation in LECA and I am fully sympathetic with this. However, getting rid of misfoled/problematic proteins and subcellular entities is something that prokaryotes also to a certain degree must have (and still do) master. Are inclusion bodies or export their only answers (I don't know)? Of course, in eukaryotes with all their intracellular complexity this is likely more of an issue. Given the scope of this manuscript (i.e. shedding light on that ancient framework, deep evolutionary roots in eukaryote evolution etc. etc.) it would be very interesting to read the authors thoughts on this and also pinpoint the prokaryote/eukaryote divide in light of the machinery discussed here.

Thank you for this suggestion. We did indeed check whether any of the UFMylation machinery were present in prokaryotes and only found homologs of UFSP2. These results are consistent with Grau-Bové et al. (2014) who conducted an equivalent analysis and concluded that UFMylation machinery were derived during eukaryogenesis. We will make reference to this in the revised manuscript.

Significance

This study not only impresses with the volume of experiments and data, but also the courage to show conservation of a molecular framework by working with such a range of distantly-related eukaryotes. The results and conclusions from this study should be interesting to anyone working in the broad fields of cellular stress and/or autophagy--both extremely timely topics.

We thank the reviewer for understanding our take-home message and the advances made. We especially thank the reviewer for understanding the challenge of connecting in silico genomic data with in vivo and in vitro experiments.

CROSS-CONSULTATION COMMENTS

Referee #2 The challenge in providing a fair review of this manuscript is to clearly define what contributions are novel, significant advances. It is difficult to tell the way the manuscript is written, as it is unclear how the new data - which are voluminous- actually advance the model already put forth by the same authors in two previous publications. It is also unfortunate that the authors overlooked the 2004 phylogenomics paper. There clearly are some new pieces of information here, but the overall increment in knowledge is rather minimal. Response from Referee #3 I agree that the authors somehow steamroll the reader with a wealth of data. But I think this can be addressed by the authors by requesting a lot more justification and by giving them the opportunity to put the significant advances into their own words. This is, in my opinion, quite doable in course of a revision. Overall I have to say that I am very sympathetic with the crosseukaryote reactivity approach that the authors have taken. It is quite intriguing.

We thank the reviewers for this useful exchange. We agree that our manuscript was not clear enough to emphasize the novelty of our results which likely resulted from the volume and diversity of the experiments and analyses that were presented. We have now revised the manuscript to improve the context and rationale for the study, the intent and hypotheses behind each experiment, and the novel results and insights obtained in each section.

Response from Referee #2 I agree that the cross-eukaryote approach is intriguing. Shouldn't we be concerned that the 2004 publication already made two of their key points (ie present in LECA, loss in Fungi). What is the incremental insight from this paper? I'd appreciate an opinion from an evolutionary biologist as to how strongly one can conclude functional co-evolution from such correlative data, especially given the rather small number of supporting examples. Is it also necessary to consider counter-examples- ie species that have sAIMs but no UFM1 (I believe that they found a few such cases)?

Importantly, we do not conclude functional co-evolution from our correlative data. Instead, we used these correlations to generate hypotheses that we tested with various experiments in different model systems. For example, the apparent correlation between C53 sAIMs and UFM1 prompted us to test whether or not UFM1 and sAIMs interact. Regardless of sample size or statistical significance, phylogenomic analyses can never demonstrate functional links, only correlations, which is why we combined these two approaches. Although only a few species encoded C53 without UFM1, each of these contained C53 cAIMs and lacked sAIMs (Figure 2c). There are species with UFM1 that lack C53 but this makes sense as UFM1 is used in other processes besides ER-phagy. We have revised the text to make our approach and reliance on certain data clearer.

Response from Referee #3 Well with these deep evolutionary questions this is always a challenge. Where does one stop to sample more homologs for one's analyses (one from each supergroup [which are no longer recognised by the community])? In that sense, the authors are right to make the parsimonious base assumption that if X and Y interact in species A and B (no matter how distant they are related) then X and Y interacted in the last common ancestor of A and B. That being said, if I would have designed this study, I would have sampled more broadly for my in vitro crosseukaryote approach. But also this, I think, could be carried out by the authors in a reasonable timeframe. Specifically, they have now sampled from Amorphea and Archaeplastida, they should add one from TSAR, one Haptista, one Cryptista, and one CRuM. If they synthesised the proteins via a company, they could have the constructs in a few weeks for about 1K Euro - I do not think that this would be an unreasonable request.

We agree that testing C53 function in additional species would strengthen our understanding of the conservation of this pathway across eukaryotes, as it cannot be assumed that orthologous proteins will function in the same way across all species. To our knowledge there is no other work showing experimentally that the UFMylation pathway is working in a single-celled organism. We focussed our efforts on the unicellular green alga, Chlamydomonas due to its relative experimental tractability. However, testing this was not trivial as it required us to establish expression and purification protocols, isolate Chlamydomonas mutants, optimize physiological stress assays, and perform the experiments.

Nevertheless, we agree that we could expand our in vitro assays with C53 orthologs from additional species. As suggested by reviewer 3, we will now synthesize 6 more C53 isoforms from two TSAR representatives (the alveolate, Tetrahymena thermophila, and the stramenopile, Phytophthora sojae), as well as a representative from Haptista (Emiliania), Cryptista (Guillardia), Diplomonada (Trypanosoma), and CRuMs (Rigifila). We will test their interaction with human and plant ATG8 and UFM1 proteins. We have also added two species from CRuMs into our phylogenomic analysis.

The list of experiments that we can do to address the reviewer’s concerns: 1. Repeat experiment in Figure 1C probing with �-RPL26. 2. To calculate KD values, perform ITC experiments with C53 wild-type, C53 sAIM mutant and C53 cAIM variant titrated with ATG8 and UFM1. 3. Perform CoIP experiments using C53 antibody in wild type and UFM1 overexpressing lines and detect for ATG8 association, under normal and stress conditions. 4. We will test autophagic degradation of C53 in uba5 and ufc1 mutants under normal and tunicamycin stress conditions by performing autophagic flux assays using the native C53 antibody 5. Molecular docking studies to see C53’s structural rearrangements leading to ATG8 and UFM1 binding. 6. Figures from co-localization experiments in Figure 5G will be revisited and we will perform additional co-localization analyses such as Pearson coefficient under normal and stress conditions. We will also upload all the raw images as supporting material, so that anyone can independently assess our images. 7. We will upload all the source data for phylogenomic analyses, including scripts and alignments to Zenodo. 8. Test the interaction of 6 newly synthesised C53 isoforms from: (1) an alveolate (tsAr, Ciliate), (2) a stramenopile (tSar, Phaeodactylum), (3) a haptophyte (Emiliania), (4) a cryptophyte (Guillardia), (5) a diplomonad (Trypanosoma) and (6) a CrRuM with human and plant ATG8 and UFM1 proteins.

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

Picchianti and colleagues have investigated a conserved molecular framework that orchestrates ER homeostasis via autophagy. For this, they have carried out phylogenomics and large-scale gene family analyses across eukaryote diversity as well as a barrage of molecular lab work. The amount of work carried out as well as the overall quality of the study is impressive. I have only a few comments that should be very easy to tackle.

1. Maybe I missed it, but please upload all alignments used for phylogenetics and phylogenomics for reproducibility to e.g. Zenodo, Figshare or other suitable OA databases.
2. "Why these non-canonical motifs were selected during evolution, instead of canonical ATG8 interacting motifs remains unknown" --> Maybe there is no "why" and these were not selected at all. Could be random... drift, non-adaptive constructive neutral evolution. I am not saying that asking "why" in evolutionary biology is wrong. It, however, often does not yield satisfactory answers--or any answer at all.
3. The authors make a case for UFMylation in LECA and I am fully sympathetic with this. However, getting rid of misfoled/problematic proteins and subcellular entities is something that prokaryotes also to a certain degree must have (and still do) master. Are inclusion bodies or export their only answers (I don't know)? Of course, in eukaryotes with all their intracellular complexity this is likely more of an issue. Given the scope of this manuscript (i.e. shedding light on that ancient framework, deep evolutionary roots in eukaryote evolution etc. etc.) it would be very interesting to read the authors thoughts on this and also pinpoint the prokaryote/eukaryote divide in light of the machinery discussed here.

Referees cross-commenting

Referee #2

The challenge in providing a fair review of this manuscript is to clearly define what contributions are novel, significant advances. It is difficult to tell the way the manuscript is written, as it is unclear how the new data - which are voluminous- actually advance the model already put forth by the same authors in two previous publications. It is also unfortunate that the authors overlooked the 2004 phylogenomics paper. There clearly are some new pieces of information here, but the overall increment in knowledge is rather minimal.

Response from Referee #3

I agree that the authors somehow steamroll the reader with a wealth of data. But I think this can be addressed by the authors by requesting a lot more justification and by giving them the opportunity to put the significant advances into their own words. This is, in my opinion, quite doable in course of a revision. Overall I have to say that I am very sympathetic with the cross-eukaryote reactivity approach that the authors have taken. It is quite intriguing.

Response from Referee #2

I agree that the cross-eukaryote approach is intriguing. Shouldn't we be concerned that the 2004 publication already made two of their key points (ie present in LECA, loss in Fungi). What is the incremental insight from this paper?

I'd appreciate an opinion from an evolutionary biologist as to how strongly one can conclude functional co-evolution from such correlative data, especially given the rather small number of supporting examples. Is it also necessary to consider counter-examples- ie species that have sAIMs but no UFM1 (I believe that they found a few such cases)?

Response from Referee #3

Well with these deep evolutionary questions this is always a challenge. Where does one stop to sample more homologs for one's analyses (one from each supergroup [which are no longer recognised by the community])? In that sense, the authors are right to make the parsimonious base assumption that if X and Y interact in species A and B (no matter how distant they are related) then X and Y interacted in the last common ancestor of A and B. That being said, if I would have designed this study, I would have sampled more broadly for my in vitro cross-eukaryote approach. But also this, I think, could be carried out by the authors in a reasonable timeframe. Specifically, they have now sampled from Amorphea and Archaeplastida, they should add one from TSAR, one Haptista, one Cryptista, and one CRuM. If they synthesised the proteins via a company, they could have the constructs in a few weeks for about 1K Euro - I do not think that this would be an unreasonable request.

Significance

This study not only impresses with the volume of experiments and data, but also the courage to show conservation of a molecular framework by working with such a range of distantly-related eukaryotes. The results and conclusions from this study should be interesting to anyone working in the broad fields of cellular stress and/or autophagy--both extremely timely topics.

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 from Picchianti et al. seeks to define the role of CDK5RAP3 (hereinafter referred as C53) during autophagy and its interplay with UFMylation. Together with UFL1 and DDRGK1, C53 is a component of a trimeric UFM1 E3 ligase complex that modifies the 60S ribosomal protein RPL26 at the endoplasmic reticulum (ER) surface upon ribosomal stalling (among other proposed functions that are not addressed). Several previous studies have implicated the UFMylation pathway in autophagy or ER-phagy although a non-autophagic fate for UFM1-tagged ribosomal subunits has also been reported.

A previous study from the same authors (PMID: 32851973) identified an intrinsically disorder region (IDR) in C53 that is necessary and sufficient for interaction between C53 and autophagy receptor, ATG8. They reported that this IDR comprises four non canonical ATG8 interacting motifs (AIM), named shuffled AIMs (sAIMs) and showed that combinatorial mutagenesis of sAIM1, sAIM2, and sAIM3 abrogates ATG8 binding. A similar effect was observed for plant C53, though an additional canonical AIM (cAIM) in the C53 IDR had to be mutated to completely abolish C53 and ATG8 interaction. The earlier study reported that C53 IDR also interacts with UFM1, and this interaction can be disrupted in vitro by adding increasing concentration of ATG8, suggesting that ATG8 and UFM1 may compete with one another for C53 binding.

The present paper attempts to build on this previous work by using phylogenomics to infer a co-evolutionary relationship between UFMylation machinery and sAIMs in C53, which the authors argue, constitutes further evidence of the primary importance of a role for UFMylation in ER homeostasis. The manuscript includes a lot of biochemical data using variations of in vitro and in vivo pull-down experiments to define the roles of individual AIMs in mediating the binding of C53 to ATG8 and to UFM1. They also use NMR spectroscopy in an attempt to define the structural basis of the UFM1 and ATG8 binding to C53, concluding that plant C53 interacts with UFM1 mainly through sAIM1, while interaction with ATG8 requires cAIM as well as sAIM1 and sAIM2. Finally, the authors attempt to contextualize these findings by conducting studies on Arabidopsis mutants, showing that replacing sAIMs with cAIMs causes increases sensitivity to ER stress and apparently increases formation of C53 intracellular puncta that may colocalize with ATG8.

From these data the authors concluded that the dual-ATG8 and UFM1 binding of C53 IDR regulates C53 recruitment to autophagosomes in response to ER stress.

Major Issues:

1. The phylogenomics analysis conclusion that UFM1 is common in unicellular lineages and did not evolve in multicellular eukaryotes is not novel, as another comprehensive analysis of UFM1 phylogeny, published eight years ago - in 2014 - by Grau-Bové et al. (PMID: 25525215), also reported that UFM1, UBA5, UFC1, UFL1 and UFSP2 were likely present in LECA and lost in Fungi. Although the phylogenomic analysis by Picchianti et al. is also extended to DDRGK1 and C53 proteins, and some parasitic and algal lineages, their findings are incremental. Their proposed coevolution of sAIM and UFM1 is based on presence-absence correlation observed within five species (i.e., Albugo candida, Albuco laibachii, Piromyces finnis, Neocallimastix californiae, Anaeromyces robustus). However, this coevolutionary relationship must be further investigated by substantially increasing the taxonomic sampling within the UFM1-lacking group.
2. The manuscript presents an overwhelming amount of biochemical and structural data obtained from a variety of protein binding techniques (i.e., NMR spectroscopy, in vitro GST-pulldown, fluorescence microscopy-based on-bead binding assays, and native mass-spectrometry). The results are poorly explained and not organized in a logical manner. Moreover, no attempt was made to explain the rationale behind using one technique over the other or how one method complements another to build a stronger conclusion than any individual approach. Given that none of the methods employed report quantitative measurement of binding affinities between C53 IDR and UFM1 or ATG8, it is not clear how the data presented in this manuscript contribute to our understanding of the proposed competition model for UFM1 and ATG8 binding to C53 IDR. To conclude that an interaction is "stronger" or "weaker" it is necessary to measure equilibrium binding constants. Fortunately, there are suitable techniques, including surface plasmon resonance (SPR), microscale thermophoresis (MST), fluorescence anisotropy, or calorimetry that are available to dissect these complex competitive binding interactions and to build models.
3. The NMR studies have the potential to dissect the types of dynamic binding inherent in unstructured proteins. However, the abundant NMR data presented combined with the aforementioned binding studies, remarkably, do not seem to significantly advance our understanding of how the system is organized or even how UFM1 and ATG8 bind C53, beyond the rather vague and somewhat circular conclusion stated in the abstract: "...we confirmed the interaction of UFM1 with the C53 sAIMs and found that UFM1 and ATG8 bound the sAIMs in a different mode." Or on line 165 "Altogether these results suggested that ATG8 and UFM1 bbind the sAIMs withn C54 IDR, albeit in a different manner".
4. The functional assays performed in Arabidopsis do not support the competitive model between UFM1 and ATG8 for binding to C53 during C53-mediated autophagy. The fluorescence microscopy images do not provide convincing evidence of colocalization between C53 and ATG8. In fact, in contrast to the claims made in the text or the quantification, mCherry-C53 fluorescence does not seem to localize in discrete puncta and its signal does not seem to overlap with ATG8A.

Minor Issues:

1. The authors might choose to avoid teleological arguments such as (line 135): "As the phylogenomic analysis suggested that eh sAIMs have been retained to mediate C53-UFM1 interaction..."
2. The authors refer on multiple occasions to C53 "autoactivation" without defining what they mean by this. Do they propose that C53 UFMylates itself?.
3. The paper might want to avoid preachy philosophical statements like "Our evolutionary analysis also highlights why we should move beyond yeast and metazoans and instead consider the whole tree of life when using evolutionary arguments to guide biological research." (333-335). While this is indeed a laudable goal, given the rather limited insights from this study, it is unclear how this paper exemplifies the notion.

Referees cross-commenting

Referee #2

The challenge in providing a fair review of this manuscript is to clearly define what contributions are novel, significant advances. It is difficult to tell the way the manuscript is written, as it is unclear how the new data - which are voluminous- actually advance the model already put forth by the same authors in two previous publications. It is also unfortunate that the authors overlooked the 2004 phylogenomics paper. There clearly are some new pieces of information here, but the overall increment in knowledge is rather minimal.

Response from Referee #3

I agree that the authors somehow steamroll the reader with a wealth of data. But I think this can be addressed by the authors by requesting a lot more justification and by giving them the opportunity to put the significant advances into their own words. This is, in my opinion, quite doable in course of a revision. Overall I have to say that I am very sympathetic with the cross-eukaryote reactivity approach that the authors have taken. It is quite intriguing.

Response from Referee #2

I agree that the cross-eukaryote approach is intriguing. Shouldn't we be concerned that the 2004 publication already made two of their key points (ie present in LECA, loss in Fungi). What is the incremental insight from this paper?

I'd appreciate an opinion from an evolutionary biologist as to how strongly one can conclude functional co-evolution from such correlative data, especially given the rather small number of supporting examples. Is it also necessary to consider counter-examples- ie species that have sAIMs but no UFM1 (I believe that they found a few such cases)?

Response from Referee #3

Well with these deep evolutionary questions this is always a challenge. Where does one stop to sample more homologs for one's analyses (one from each supergroup [which are no longer recognised by the community])? In that sense, the authors are right to make the parsimonious base assumption that if X and Y interact in species A and B (no matter how distant they are related) then X and Y interacted in the last common ancestor of A and B. That being said, if I would have designed this study, I would have sampled more broadly for my in vitro cross-eukaryote approach. But also this, I think, could be carried out by the authors in a reasonable timeframe. Specifically, they have now sampled from Amorphea and Archaeplastida, they should add one from TSAR, one Haptista, one Cryptista, and one CRuM. If they synthesised the proteins via a company, they could have the constructs in a few weeks for about 1K Euro - I do not think that this would be an unreasonable request.

Significance

Overall, while the manuscript contains an abundance of new data, the overall conclusion of the work, stated in the title: "Shuffled ATG8 interacting motifs form an ancestral bridge between UFMylation and C53-mediated autophagy" does not constitute a significant advance beyond other published phylogenomic analysis (below) and the two previous papers by the same authors, including the 2020 paper "A cross-kingdom conserved ER-phagy receptor maintains endoplasmic reticulum homeostasis during stress (PMID: 32851973)" and the 2021 paper "C53 is a cross-kingdom conserved reticulophagy receptor that bridges the gap between selective autophagy and ribosome stalling at the endoplasmic reticulum PMID: 33164651)". While a regulatory interaction between UFMylation and autophagy is of potential importance, the data in this manuscript do not constitute a major advance and fail to provide new mechanistic insight to explain the role of C53 IDR in autophagy and its interplay with UFMylation

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 manuscript Picchianti et al. provide novel insights into the interaction of C53 with UFM1 and ATG8. Initially, the authors show that protein modification by UFM1 exists in the unicellular organism Chlamydomonas reinhardtii. To that end they demonstrated that pure Chlamydomonas UBA5, UFC1 and UFM1 proteins, can charge UFC1. Then, they showed that C53 interacts with ATG8 and UFM1. Specifically, they found that the sAIM are essential for the interaction with UFM1, while substituting this motif with canonical AIM prevents the binding of UFM1 but not of ATG8. Since binding of C53 to ATG8 recruits the autophagy machinery, the authors suggest that ufmylation of RPL26 releases UFM1 from C53 which allows the binding of ATG8. Overall, the authors demonstrate that C53 that forms a complex with UFL1 connects between protein ufmylation and autophagy by its ability to bind both UBLs.

Here the authors revisited the assumption that only multicellular organisms have the UFM1 system. Using bioinformatic tools they show that it exists also in unicellular organism. Also, they show that in some organisms the E3 complex UFL1, UFBP1 and C53 exist but not UBA5, UFC1 or UFM1. This is a very interesting observation that suggests an additional role for this complex. In Fig 1C the authors show that in Chlamydomonas RPL26 undergoes ufmylation. Please use IP against RPL26 and then a blot with anti UFM1. From the current experiment it is not clear how the authors know that this is indeed RPL26 that undergoes ufmylation

In the second part of the manuscript the authors characterize the interaction of C53 with ATG8 and UFM1. This is a continuation of their previous published work (Stephani et al, 2020) . Here the reviewer thinks that further data on the binding of these proteins to C53 is required. Specifically, defining the Kd of these interactions using ITC or other biophysical method can contribute to the study.

Under normal condition the authors suggest that C53 binds UFM1 and this keeps it inactive. The reviewer thinks that this claim needs further support. Using IP (maybe with crosslinker) the author can show that C53, in normal conditions, bind more UFM1 than ATG8. Also, since the interaction of UFM1 to C53 is noncovalent, it will be nice to show how alternations in UFM1 expression levels can affect the activation of C53. Finally, the authors suggest that ufmylation of RPL26 allows binding of ATG8 to C53 and this, in turn, leads to C53 activation. Can the authors show that in cells lacking UBA5, under normal condition or with Tunicamycin treatment, ATG8 does not activate C53 due to the fact that UFM1 does not leave C53.

Significance

This manuscript advances our understanding of the connection of ufmylation to autophagy which is mediated by C53.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Reply to the reviewers

Referee #1

Evidence, reproducibility and clarity

1. This manuscript constructs a gene expression model with various factors. Specifically, the effect of cell size on gene expression is considered, which is often ignored by previous studies. One interesting finding is that the absolute number of the gene products and the concentration can have different distributions. Some predictions of the models are validated by experimental data on E. coli and yeast. This manuscript uses the mean-field approximation for cell volume, which has good accuracy when the number of stages is large. The usage of the power spectrum has a satisfactory effect on studying the concentration oscillation.

Response: Thank you for the positive comments.

1. Overall the paper was very difficult to follow and digest easily because of all the different factors and mechanisms invoked. It is mainly an issue of providing sufficient details for each of the factors and organizing them in a systematic and logical way. Although there is a supplementary appendix, it was hard to keep track of all the elements in the main manuscript. Perhaps something like Fig 1 of the Appendix can be presented in the main body to outline all the ingredients and how they affect each other.

Response: In the revised manuscript, we moved Supplementary Fig. S1 in the previous version into the main text to outline all the ingredients and how they affect each other (see page 8, Fig. 2). Moreover, we provided many details for each of the biological factors and tried to organize them in a more systematic and logical way (see pages 3-7).

1. It might be good to provide a more detailed description of the goal (studying gene product number and concentration under different parameters) after introducing the full and the reduced models. A table of symbols would also be helpful.

Response: In the revised manuscript, we added a table explaning the meaning of all model parameters (see page 4, Table 1). Moreover, we provided a detailed description of the goal of the present paper after introducing the full and reduced models (see page 7).

1. Some technical details in the Methods section are in fact helpful in understanding the conclusions. They can be moved to the Results section.

Response: In the revised manuscript, we moved many technical details in Methods and Supplementary Notes to the main text to help the readers better understand the conclusions (see pages 5-10).

1. One concern is that the central concept of this manuscript, “stage”, is not thoroughly discussed. This concept should have some significant biological meaning, not just be coined for mathematical convenience.

Response: In the revised manuscript, we explained in detail the biological meaning of the effective cell cycle stages (see page 4). Specifically, recent studies have revealed that in many cell types, the accumulation of some activator to a critical threshold is used to promote mitotic entry and trigger cell division, a strategy known as activator accumulation mechanism. In E. coli, the activator was shown to be FtsZ; in fission yeast, it is believed to be a protein upstream of Cdk1, the central mitotic regulator, such as Cdr2, Cdc25, and Cdc13. Biophysically, the N effective stages can be understood as different levels of the key activator. Moreover, we pointed out that the power law form for the rate of cell cycle progression may come from cooperativity of the key activator that triggers cell division.

1. Fig. 1(b) is a little strange. For the left panel, the x-axis (stage) is discrete, then the volume (y-axis) should be a step function, not a straight red line.

Response: In the revised manuscript, we added some red dots in the stage-volume plot to show the dependence of the mean cell volume vk on cell cycle stage k for the mean-field model (see page 3, Fig. 1). Moreover, we emphasized that the joining of these dots by a straight red line is simply a guide to the eye.

Significance

1. The main advance is a more complete model of gene expression under more realistic organism growth conditions.

Response: Thank you for acknowledging the results of the manuscript.

Referee #2

Evidence, reproducibility and clarity

1. Jia et al. introduce a modeling framework to represent stochastic gene expression, with an explicit representation of cell volume growth, cell cycle progression (and its dependency on cell volume) and gene dosage compensation. The model is very elegant and general in that it can represent a variety of situations, simply as a matter of parametrization. Under a simplifying assumption, the authors derive a number of metrics (include stationary distribution of gene product and power spectrum of gene product fluctuation dynamics), for both absolute number and concentration of gene product molecules. They use their model and derivations to examine under which conditions cell can achieve homeostasis in the concentration of the expressed gene product, despite changes in cell volume and gene copy number following replication. They also present and discuss the conditions giving rise to specific features (i.e. bimodality in stationary distribution, peak in power spectrum) and examine these features in experimental data to conclude to infer the underlying homeostasis strategies. The model is rather general and powerful. The simplifying assumption seems reasonable (and the authors investigate to some extent its limitations, i.e. Fig. 2). The conclusions are overall convincing.

Response: Thank you for the positive comments.

Major comments 1. My main concern is that the metrics that the authors use to assess concentration homeostasis (i.e. the γ parameter and the presence / absence of peak in power spectrum) do not seem quite appropriate to describe how much variability / fluctuations in concentration are driven by cell cycle effects. Indeed, the γ parameter measures how much the *average* concentration in each cell cycle stage varies throughout the cell cycle. However, this variability should be compared to the total variability due to both cell cycle effects and stochastic bursting dynamics. A given level of cell-cycle dependency (say γ = 0.2) could be very visible if gene expression is weakly noisy (e.g. B low and hni high) and completely invisible is gene expression is highly bursty (large B and small hni). In the latter situation, cell-cycle effects would be meaningless for the cell to minimize. In essence, reusing the authors notations, I think γ/φ1/2 , would be a more relevant metric to observe.

Response: In the revised manuscript, we showed that the total concentration noise φ can be decomposed as φ = φext + φint, where φext is the extrinsic noise which characterizes the fluctuations between different stages due to cell cycle effects and φint is the intrinsic noise which characterizes the fluctuations within each stage due to stochastic bursty synthesis and degradation of the gene product (see page 11). Based on the above decomposition, we introduced a new metric γ = φext/φ, which characterizes the accuracy of concentration homeostasis. Clearly, the new metric γ reflects the relation contribution of cell cycle effects in the total concentration variability. All discussions about concentration homeostasis are based on the new metric γ in the revised manuscript. Moreover, all figures have been updated by using this new metric.

1. Similarly, when inspecting the peak in the power spectrum, the weight of the Lorentzian function(s) creating the peak, should be compared to the stationary component (λN , uN in the authors’ notations).

Response: We cannot quite understand why the weights uk of the Lorentzian functions should be compared to the stationary component uN . In fact, all the weights uk except uN are actually complex numbers and we are not so sure about the meaning of uk/uN . However in the revised manuscript, we emphasized that the power spectrum G(ξ) is normalized so that G(0) = 1 throughout the paper (see page 13). To better understand concentration oscillations and its relation to homeostasis, we depicted both γ and H as a function of B and hni (see Supplementary Fig. S5). As expected, the off-zero peak becomes lower as B increases and as hni decreases since both of them correspond to an increase in concentration fluctuations which counteracts the regularity of oscillations; noise above a certain threshold can even completely destroy oscillations. Furthermore, we found that γ and H have similar dependence on B and hni. This again shows that the occurrence of concentration oscillations is intimately related to the visibility of cell cycle effects in concentration fluctuations.

A complementary analysis including these two points and a discussion the relative contribution of cell-cycle effects and bursting dynamics in the total variability/fluctuation of concentrations would be important to include.

Response: In the revised manuscript, we made some complementary analysis and discussion about the relative contribution of cell cycle effects and stochastic birth-death dynamics in the total variability of concentrations (see pages 11-14).

Minor comments 3. The dashed line on Fig. 3a is defined as κ = √ 2 1−β . First is this empirical or does it come from a derivation? Second, it seems incomplete since it should depend on w. Intuitively, this line should correspond to the value of κ that would best mimic balanced biosynthesis in the case where β 6= 1. In other words, κ should be so that hρB0 /V (t)iprereplication = hκρB0 /V (t)ipostreplication, which yields κ = 2w(1−β) ∗ (w − 1)/w ∗ [2w(β−1) − 1]/[2(1−w)(β−1) − 1]. This indeed simplifies into κ = √ 2 1−β when w = 0.5.

Response: Thank you for providing such a beautiful derivation. In the revised manuscript, we added this derivation into the main text (see pages 12-13). Moreover, we also made it clear that this relation can also be obtained from the perspective of power spectrum (see page 14).

1. η is used in the caption of Fig. 2, which is cited on page 4. But it is defined only 2 sections later, on page 6.

Response: In the revised manuscript, we gave the definition of η in both Table 1 and the caption of Fig. 3 (Fig. 2 in the old version). Please see page 4, Table 1 and page 9, Fig. 3.

1. w is used in the main text, but only defined in the caption of Fig. 3.

Response: In the revised manuscript, we gave the definition of w in both Table 1 (see page 4) and on page 7.

1. w is defined as “the proportion of cell cycle before replication”. Is this in terms of cell cycle stages (i.e. w = N0/N) or actual time?

Response: In the revised manuscript, we made it clear that w represents the proportion of cell cycle duration before replication, which should be distinguished from the proportion N0/N of cell cycle stages before replication (see page 7). This is because the transition rate between cell cycle stages is an increasing function of cell size, which means that earlier (later) stages have longer (shorter) durations.

1. Fig. 3 indicates that power spectra are normalized so that G(0) = 1, but G(0) = 10 on the first two graphs.

Response: Corrected as suggested (see page 12, Fig. 4). Thank you.

1. Page 11: “bimodality in the concentration distribution is significantly less apparent”. I would suggest rephrasing “bimodality in the concentration distribution is absent” since there should be no reference to “significance” and bimodality is either present or absent (binary), not less apparent.

Response: Corrected as suggested. Thank you.

Referees cross-commenting

1. Regarding the comment from reviewer 3 that ”a direct validity test should use data sets of at least two types (total, nascent RNA, etc)”. I almost made a related comment in my review, but then I held it off: This issue with using nascent RNA data is that their model does not allow an ON state. They assume that gene products are produced in instantaneous bursts, which is a fair assumption if the lifetime of gene products is large compared to the time the gene stays ON. This is ok if the considered ”gene products” are mRNA or proteins, but not nascent RNAs (for which the lifetime is the time to transcribe the gene). I did not make this comment in the end because I think the model is useful regardless. To comply with reviewer 3’s request, maybe the authors could use distributions of mRNA and protein products, but I’m not sure that such data exists (since they need cell-cycle-resolved data).

Response: It is not possible to validate our model with nascent mRNA data because the model in its present form cannot predict nascent mRNA fluctuations. This is because unlike mature mRNA, nascent mRNA cannot be assumed to decay via first-order kinetics. A detailed response is provided below to the original comment made by Referee 3. Regarding the comment on the use of cell-cycle-resolved data measuring mRNA and protein expression – while we agree it would make an excellent test of our model, we could not find such a dataset in the literature. We point out that our model, in its present form, is interesting as it is, as a detailed biological model of mature mRNA and protein number / concentration fluctuations in growing cells. Its predictions are yet to be fully confirmed and hence may stimulate the development of further experimental single-cell studies.

Significance

1. The advance of this paper is essentially technical. The authors present a model that incorporates and unifies previously studied effects (cell volume homeostasis, concentration homeostasis, bursting transcription). There is no major conceptual novelty, but the combination of these different aspects and the derivations that authors present are very valuable and might be applicable to interpret data in various species.

Response: Thank you for acknowledging the results of the manuscript.

Referee #3

Evidence, reproducibility and clarity

1. The manuscript analyses a phenomenological model of stochastic gene expression. The model couples bursty transcription with cell growth, division and DNA replication. The cell cycle is divided into a large number of stages whose exponential lifetimes depend on the cell volume. It is argued that concentrations of gene products are distributed according to mixed Gamma distributions, whereas the copy numbers follow mixed negative binomial distributions. The number of modes can be different for concentrations and copy numbers, for instance the copy numbers can be unimodal while concentrations are bimodal. The case when the mean concentration does not depend on the cell cycle stage is called perfect homeostasis. It is argued that perfect homeostasis leads to Gamma distribution of the gene product concentration and that deviations from a Gamma distributions result mainly from deviations of the concentration from perfect homeostasis. It is also proposed that concentration homeostasis is difficult to obtain. These qualitative predictions of the model are tested using two data sets, one for E.coli and another for fission yeast.

Response: Thank you for acknowledging the results of the manuscript.

Major comments 1. A huge number of states called “cell cycle stages” have exponential life times. On my opinion, this sequence of stages is just a technicality for keeping the model within a discrete Markovian framework. More natural choices are possible, such as piecewise deterministic Markov processes, age structured diffusions, etc. The biological significance (if there is any) of such states should be explained.

Response: In the revised manuscript, we explained in detail the biological meaning of the effective cell cycle stages (see page 4). Specifically, recent studies have revealed that in many cell types, the accumulation of some activator to a critical threshold is used to promote mitotic entry and trigger cell division, a strategy known as activator accumulation mechanism. In E. coli, the activator was shown to be FtsZ; in fission yeast, it is believed to be a protein upstream of Cdk1, the central mitotic regulator, such as Cdr2, Cdc25, and Cdc13. Biophysically, the N effective stages can be understood as different levels of the key activator. Moreover, we pointed out that the power law form for the rate of cell cycle progression may come from cooperativity of the key activator that triggers cell division.

1. The timescales of stochastic gene expression are not correctly taken into account. It is considered that during an exponential stage the bursting approximation describes gene expression in terms of Gamma distributions for concentrations and in terms of negative binomial distributions for copy numbers. This approximation is only valid if the lifetime of a stage is much larger than the time needed to generate a burst. For RNA, this condition cannot be fulfilled for a large number of states N and/or for two states promoters with a relatively long ON state. For the protein and/or in the case of translational bursting, the condition is even more difficult to fulfil. I agree with the Reviewer 2 that once the master equation accepted the results make sense. But my criticism is different and concerns the master equation itself. In this equation the burst is considered instantaneous, whereas it needs finite time in reality. Concerning nascent mRNA, ON/OFF etc. I disagree. The notion of instantaneous burst with well defined burst size and burst frequency on a stage has a meaning if the lifetime of this stage (which is not mRNA or protein lifetime) is short. The model validity should be clearly stated.

Response: Thank you for pointing out this important issue. When we talk about the validity of the model, we should stick to the full model, instead of the mean-field model. This is because once the full model makes sense, the mean-field model must work well when N ? 15, as we have shown in Fig. 3 and Supplementary Fig. S3. Hence our reply is based on the validity of the full model. We will reply to the above comments from the following three aspects. First, we agree with the referee that in our model, we assume that the gene product is produced in instantaneous bursts with the reaction scheme G ρpk (1−p) −−−−−−→ G + kM, k ≥ 1, M d −→ ∅, (1) where the mean burst size scales as V (t) β . Of course, in reality there is a finite time for the bursts to occur. A more general assumption is that within each cell cycle, the gene expression dynamics is characterized by the following three-stage model: G ρ −→ G ∗ , G∗ r −→ G, G∗ sV (t) β −−−−→ G ∗ + M, M u−→ M + P, M v −→ ∅, P d −→ ∅, (2) where the first two reactions describe the switching of the gene between an inactive state G and an active state G∗ the middle two reactions describe transcription and translation, and the last two reactions describe the degradation of the mRNA M and the protein P. Here the synthesis rate of mRNA depends on cell volume via a power law form with power β ∈ [0, 1]. Dosage compensation can be modeled by a decrease in the gene activation rate (for each gene copy) from ρ to κρ/2 upon replication. Previous studies have revealed that the bursting of mRNA and protein has different biophysical origins: transcriptional bursting is due to a gene that is mostly inactive, but transcribes a large number of mRNA when it is active (r ? ρ and s/r is finite), whereas translational bursting is due to rapid synthesis of protein from a single short-lived mRNA molecule (v ? d and u/v is finite). Under the above timescale separation assumptions, both mRNA and protein are produced in a bursty manner with the reaction scheme described by Eq. (1). The burst frequency for mRNA and protein are both ρ before replication and κρ after replication. The mean burst size for mRNA is (s/r)V (t) β and the mean burst size for protein is (su/rv)V (t) β , both of which have a power law dependence on cell volume (see pages 5-6). In Supplementary Figs. S1 and S2, we compare the mRNA and protein distributions for the bursty model with the reaction scheme given by Eq. (1) and the three-stage model with the reaction scheme given by Eq. (2), where both models under consideration have a cell cycle and cell volume description. It can be seen that the distributions for the two models are very close to each other under the above timescale separation assumptions with the bursty model being more accurate as r/ρ and v/d increase. Moreover, we find that the accuracy of the bursty model is insensitive to the value of the number of stages N. Here the values of N are chosen so that the ratio of the average time spent in each stage (T /N, where T ≈ (log 2)/g is the mean cell cycle duration) and the mean burst duration time (1/ρ) ranges from ∼ 0.5 − 2. This shows that the effectiveness of the bursty model does not require that the lifetime of a cell cycle stage is sufficient long. Due to mathematical complexity, we only focus on the bursty model in the present paper. The consistency between the gene product distributions for the two models justifies our bursty assumption. Second, while we assume bursty expression here, our model naturally covers non-bursty expression since the latter can be regarded as a limit of the former. Hence all the conclusions in the present paper are applicable to both bursty and non-bursty expression. In the revised manuscript, we emphasized this point (see page 4 for a detailed explanation). Last but not least, if the lifetime of the gene product is much shorter compared to the lifetime of each cell cycle stage, then the gene expression dynamics will rapidly relax to a quasi-steady state for each stage. In this case, the gene product fluctuations at each stage can be characterized by a gamma distribution in terms of concentrations and by a negative binomial distribution in terms of copy numbers, and hence the distribution of concentrations (copy numbers) for a population of cells is naturally a mixture of N gamma (negative binomial) distributions. However, the powerfulness of our analytical distribution (see page 10, Eq. (8)) is that it serves an accurate approximation when N ? 1 without making any timescale assumptions. The effectiveness of our analytical distributions is validated in Supplementary Fig. S3 for three different cases: (i) the degradation rate d of the gene product is much smaller than the cell cycle frequency f; (ii) d and f are comparable; (iii) d is much larger than f. In the revised manuscript, we also emphasized these points (see page 10).

1. DNA replication is a stochastic event and does not occur after a fixed number of exponential stages as it is considered in this model. Concerning replication: in the model this occurs after exactly N0 steps. In reality, replication occurs somewhere between the start of S and G2/M. N0 is in fact a random variable. Probably a new mean field assumption is needed here with some justification, but I have seen nothing in the paper.

Response: We agree with the referee that replication of the whole genome occurs in the S phase, which occupies a considerable portion of the cell cycle and thus cannot be assumed to occur after a fixed number of exponential stages. However, our model is for a single gene and since the replication time of a particular gene is much shorter than the total duration of the S phase, it is reasonable to consider it to be instantaneous. In addition, recent experiments have shown that the time elapsed from birth to replication for a particular gene occupies an approximately proportion of the cell cycle, which is called the stretched cell cycle model. This is also consistent with our assumption that replication of the gene of interest occurs after exactly N0 stages. While replication occurs after a fixed number of stages, nevertheless the time of replication is stochastic since each stage has a random lifetime. In the revised manuscript, we emphasized these points (see pages 4-5).

1. The results in the Methods were derived heuristically and their relation to the master equation (12) is not explicit (except for the part concerning moments and their power spectrum). Furthermore, one would like to have some estimates of the biases introduced by the mean field approximation. Concerning biases introduced by the mean field approximation: Figure 2 is a numerical simulation, some analytical estimates could be better. As Figure 2 looks rather convincing, I reclassify this as minor comment.

Response: We agree with the referee that the derivation of moments is rigorous, but the derivation of the analytical distribution given in Methods is not rigorous and cannot be directly obtained from the master equation. In the revised manuscript, we emphasized that the analytical distribution is not exact but it serves as a very good approximation (see pages 10 and 22). We showed that the analytical distribution agrees well with stochastic simulations when the number of cell cycle stages N ≥ 15 (see page 9, Fig. 3 and Supplementary Fig. S3). The logic behind our approximate distribution is that while the gene product may produce complex distribution of concentrations (copy numbers), when the number of cell cycle stages is large, the distribution must be relatively simple within each stage and thus can be well approximated by a simple gamma (negative binomial) distribution (see page 22). Due to the complexity of our model, it is very difficult to provide any analytical estimates on the bias introduced by the mean-field approximation. Often the bias of an approximation can be estimated when the approximation emerges from a systematic method such as van Kampen’s system-size expansion (see Ref. [21]). However, our mean-field model cannot be seen as the zero order term of some expansion and hence it is not possible to calculate the next-order correction which would be needed to estimate the error. However, we have tested very large swathes of parameter space and found that the mean-field approximation always works well when N ≥ 15 which is the physiologically relevant regime for most types of cells (see discussion on P. 7).

1. The model is not minimal and depends on a huge number of parameters. It is not clear how these parameters were found and if overfitting was avoided. One may have doubts about the identifiability of the parameter N. What difference is between N = 59 and N = 60 (the value of N for the cyanobacterium)?

Response: In the revised manuscript, we used synthetic data to show that all the model parameters involved in our model (except d and β which can be determined based on a priori knowledge) can be accurately estimated from cell-cycle resolved lineage data of cell volume and gene expression (see Supplementary Note 7). We provided details of the parameter inference method, compared the input parameters with the estimated ones and verify that they are identifiable (see Supplementary Table 1). We did not use real data to test our inference method because we could not find cell-cycle resolved lineage data for mRNA or proteins. As we noted, this is in principle possible via cell-cycle fluorescent markers. We also note that parameter inference for less detailed but similar models have been made in our previous papers — the parameters related to cell volume dynamics have been inferred in E. coli (see Ref. [51]) and fission yeast (see Ref. [52]) using the method of distribution matching, and the parameters related to gene expression dynamics have be estimated in E. coli (see Ref. [40]) using the method of power spectrum matching. Moreover, for our purpose, i.e. to investigate the effect of cell cycle and cell volume on gene expression, we do believe that our model is minimal. We captured cell growth with only one parameter g, the degree of balanced biosynthesis with one parameter β (β = 0 corresponds to the case where the synthesis rate is independent of cell volume and β = 1 corresponds to the case where the synthesis rate scales linearly with cell volume), the variability in cell cycle duration with only one parameter N, gene replication with only one parameter N0, gene dosage compensation with only one parameter κ (κ = 1 corresponds to perfect dosage compensation and κ = 2 corresponds to no dosage compensation), and the variation of size control strategy across the cell cycle with two parameters α0 and α1 (αi → 0 corresponds to timer, αi = 1 corresponds to adder, and αi → ∞ corresponds to sizer). The biological meaning of the cell cycle stages were clarified in the revised manuscript (see page 4). For our purpose, we believe that our model cannot be simpler.

1. The authors should make clear which cell biology aspects are important, which are less important, and which were neglected in the context of their problem. Thus, in their model, cell cycle acts on gene expression mainly by duplication of burst sources and thus by increase of burst frequency after replication. Another important source of gene expression variability during the cycle, the mitotic transcription repression, is neglected.

Response: In the revised manuscript, we clarified which cell biology aspects are important for gene expression dynamics (see page 17). Specifically, in our model, cell cycle and cell volume act on gene expression mainly by (i) the dependence of the burst size on cell volume; (ii) the increase in the burst frequency upon replication; (iii) the change in size control strategy upon replication; (iv) the partitioning of molecules at division. Point (iv) strongly affects copy number fluctuations, while it has little influence on concentration fluctuations. In addition, in the revised manuscript, we also elucidated the limitations of our model including mitotic transcription repression and others (see pages 19-20).

1. The validity test of the model is indirect. It was tested that the concentration distribution deviates from Gamma and that the deviation correlates positively to the lack of accuracy of the concentration homeostasis. However, many models can have this behaviour. A direct validity test should use data sets of at least two types (total, nascent RNA, etc.) allowing direct estimates of some model parameters (such as burst size and frequency using nascent RNA). Concerning parsimony, I think that the authors should test it. Are all the parameters identifiable? Is there any overfitting? They could use parameter uncertainty, comparison of training /testing errors, etc. Some details about the parameter fitting method should be provided.

Response: Regarding the parameter fitting and identifiability we have provided a detailed response to a previous comment above. However we emphasize that for the generation of Fig. 7, we did not need to estimate all model parameters from data. Hence in the previous version of the manuscript, no such estimation was done — we simply extracted the homeostasis accuracy γ, the height H of the off-zero peak of the power spectrum, and the Hellinger distance D of the concentration distribution from its gamma approximation directly from data. Finally, we point out that our model can be used to predict the dynamics of mature mRNAs, but it cannot be used to describe the dynamics of nascent mRNAs. This is because nascent mRNAs do not decay via a first-order reaction but their removal, i.e. their detachment from the gene which leads to mature mRNA, is better approximated by a reaction with a fixed decay time. This models the elongation time of nascent transcripts which does not suffer from much noise because the RNAP velocity is to a good approximation constant along the gene. See e.g. the following two papers for details: H. Xu, S. O. Skinner, A. M. Sokac, I. Golding, Stochastic kinetics of nascent RNA. Phys. Rev. Lett. 117, 128101 (2016). S. Braichenko, J. Holehouse, R. Grima. Distinguishing between models of mammalian gene expression: telegraph-like models versus mechanistic models. J. R. Soc. Interface 18, 20210510 (2021). Because of the fixed delay, the delay telegraph model (the telegraph model with a delayed degradation reaction) is non-Markovian and very different from the usual Markovian telegraph model which describes the dynamics of mature mRNA within each cell cycle. See e.g. the Supplementary Information of the following paper: X. Fu, et al. Accurate inference of stochastic gene expression from nascent transcript heterogeneity. bioRxiv (2021). Given the mathematical complexity introduced by a fixed delay, using it to describe the dynamics of nascent mRNA within each cell cycle leads to a non-Markovian model that is even more analytically intractable than the present one for mature mRNA. While an interesting research question, this is clearly far removed from the scope of our current manuscript.

Minor comments 8. The introduction could be more pedagogical. Right now it is just an accumulation of loosely related and sometimes abruptly introduced statements. For instance, we understand that the authors want to oppose their approach to other extant approaches. However, extant approaches should be better reviewed, some of them are aged structured and perfectly suited for analysing cell cycle data. It would be useful for the reader that an example of observation explained by their model and not explained by other models (age structured or not) is discussed in detail. The model of this work does not explain size control, it just assumes that this holds, and does not discuss cell population aspects. A more nuanced positioning of this approach with respect to the literature would be useful for judging its value.

Response: In the revised manuscript, we rewrote the introduction part to make it more pedagogical (see pages 1-2). In particular, we compared three popular models describing the cell size dynamics and the associated size homeostasis. The advantages and disadvantages of the three models were discussed.

1. The meaning of N should be discussed from the very start when the model is introduced.

Response: In the revised manuscript, we explained in detail the biological meaning of the effective cell cycle stages (see page 4). Specifically, recent studies have revealed that in many cell types, the accumulation of some activator to a critical threshold is used to promote mitotic entry and trigger cell division, a strategy known as activator accumulation mechanism. In E. coli, the activator was shown to be FtsZ; in fission yeast, it was believed to be a protein upstream of Cdk1, the central mitotic regulator, such as Cdr2, Cdc25, and Cdc13. Biophysically, the N effective stages can be understood as different levels of the key activator. Moreover, we pointed out that the power law form for the rate of cell cycle progression may come from cooperativity of the key activator that triggers cell division.

1. The authors call constitutive expression the situation when the mean copy number does not depend on the volume. This choice should be clarified as in general constitutive as opposed to specific, localised or transitory expression refers to non-regulated gene expression. It seems to me that in this context, expression is only partially constitutive (independent on the volume).

Response: In the present paper, constitutive expression means that the gene product is produced one at a time and is not produced in a bursty manner. It does not mean that the mean copy number does not depend on the volume. In the revised manuscript, we provided a more detailed discussion about how constitutive expression can be viewed as a limit of bursty expression (see page 4).

1. In figure 1b and for exponential growth the y axis should be log(volume) instead of volume. The mean field approximation is called both “of novel type” (Discussion) and “which has a long history of successful use in statistical physics” (p4). If something is novel, then one should clearly explain why.

Response: In fact, the y-axis in Fig. 1(b) should be volume instead of log(volume). This is because the x-axis represents the cell cycle stage instead of the real time. Note that for the adder strategy (α0 = α1 = 1), it follows from Eq. (3) on page 7 that the mean cell volume at stage k is vk = v1 + (k − 1)M0/N0, which linearly depends on k. This explains why the red curves in Fig. 1(b) are straight lines instead of exponential curves. In the revised manuscript, we also explained why the mean-field approximation used is novel (see page 7). Specifically, we pointed out that the mean-field approximation is not made for the whole cell cycle, rather we make the approximation for each stage and thus different stages have different mean cell volumes. This type of piecewise mean-field approximation, as far as we know, is novel and has not been used in the study of concentrating fluctuations before.

1. The word “cyclo-stationarity” is used with not much definition. If this means just stationary distribution of the gene products why not use just “stationarity” instead. What means “cyclo”? A number of properties were called “rare” but it is not clear on what grounds.

Response: In the revised manuscript, we removed the term “cyclo-stationarity” and simply assumed that the copy number and concentration distributions of the gene product at each cell cycle stage have reached the steady state (see page 8). In addition, for each property that was called “rare”, we explained the reasons in detail (see pages 14 and 17).

1. I did not find a proof that the copy number distribution has less modes than the concentration distribution.

Response: In fact, it is very difficult to prove that the concentration distribution has less modes than the copy number distribution. However, we have tested very large swathes of parameter space and found that the number of modes of the concentration distribution is always less than or equal to that of the copy number distribution. In the revised manuscript, we emphasized this point (see page 16).

Significance

1. The strength of this work is that it incorporates in a stochastic gene expression model a number of ideas on size control and dosage compensation that were discussed elsewhere from a cell population point of view. However, the proposed model is based on a number artificial choices that are difficult to justify biologically: a huge number of cell cycle discrete states and inappropriate handling of the timescales characterizing stochastic gene expression. Furthermore, the model is not minimal but depends instead on a huge number of parameters. I found the paper difficult to read and in the results presentation is not suitable for biologists that would need more details on the justification of the modelling choices and on the experimental validation of the model.

Response: All these points have been addressed in previous replies.

1. For mathematicians, the calculations are rather standard and may seem trivial.

Response: Our model is complex due to the coupling between gene expression dynamics, cell volume dynamics, and cell cycle events. It is far more complex than standard models of gene expression (see e.g. Refs. [2,84,85]) because of the large amount of biology encapsulated in it and we presented a first analytical- and simulation-based analysis of concentration fluctuations when concentration homeostasis is broken.

The computations of many quantities in the present paper are non-trivial. First, we showed that the generalized added volumes before and after replication both have an Erlang distribution. Using this property, we computed the mean cell volume in each cell cycle stage which is needed in the mean-field approximation. Furthermore, the computations of the power spectrum of concentration fluctuations are also highly non-trivial. The analytical expression of the power spectrum allows us to precisely determine the onset of concentration homeostasis. While the computations of moments of concentration fluctuations are standard, we used to the moments to construct an analytical concentration distribution which serves as an accurate approximation when N is large. Our concentration distribution is generally valid when concentration homeostasis is broken and goes far beyond recent models for growing cells which require concentration homeostasis and which do not take into account DNA replication, dosage compensation and size control mechanisms that vary with the cell cycle phase (e.g. Ref. [26] ).

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

The manuscript analyses a phenomenological model of stochastic gene expression. The model couples bursty transcription with cell growth, division and DNA replication. The cell cycle is divided into a large number of stages whose exponential lifetimes depend on the cell volume. It is argued that concentrations of gene products are distributed according to mixed Gamma distributions, whereas the copy numbers follow mixed negative binomial distributions. The number of modes can be different for concentrations and copy numbers, for instance the copy numbers can be unimodal while concentrations are bimodal. The case when the mean concentration does not depend on the cell cycle stage is called perfect homeostasis. It is argued that perfect homeostasis leads to Gamma distribution of the gene product concentration and that deviations from a Gamma distributions result mainly from deviations of the concentration from perfect homeostasis. It is also proposed that concentration homeostasis is difficult to obtain. These qualitative predictions of the model are tested using two datasets, one for E.coli and another for fission yeast.

Major comments:

The model encompasses a number of artificial choices:

• A huge number of states called "cell cycle stages" have exponential life times. On my opinion, this sequence of stages is just a technicality for keeping the model within a discrete Markovian framework. More natural choices are possible, such as piecewise deterministic Markov processes, age structured diffusions, etc. The biological significance (if there is any) of such states should be explained.
• The timescales of stochastic gene expression are not correctly taken into account. It is considered that during an exponential stage the bursting approximation describes gene expression in terms of Gamma distributions for concentrations and in terms of negative binomial distributions for copy numbers. This approximation is only valid if the lifetime of a stage is much larger than the time needed to generate a burst. For RNA, this condition cannot be fulfilled for a large number of states N and/or for two states promoters with a relatively long ON state. For the protein and/or in the case of translational bursting, the condition is even more difficult to fulfil.
• DNA replication is a stochastic event and does not occur after a fixed number of exponential stages as it is considered in this model. The results in the Methods were derived heuristically and their relation to the master equation (12) is not explicit (except for the part concerning moments and their power spectrum). Furthermore, one would like to have some estimates of the biases introduced by the mean field approximation. The model is not minimal and depends on a huge number of parameters. It is not clear how these parameters were found and if overfitting was avoided. One may have doubts about the identifiability of the parameter N. What difference is between N=59 and N=60 (the value of N for the cyanobacterium)? The authors should make clear which cell biology aspects are important, which are less important, and which were neglected in the context of their problem. Thus, in their model, cell cycle acts on gene expression mainly by duplication of burst sources and thus by increase of burst frequency after replication. Another important source of gene expression variability during the cycle, the mitotic transcription repression, is neglected.<br /> The validity test of the model is indirect. It was tested that the concentration distribution deviates from Gamma and that the deviation correlates positively to the lack of accuracy of the concentration homeostasis. However, many models can have this behaviour. A direct validity test should use datasets of at least two types (total, nascent RNA, etc.) allowing direct estimates of some model parameters (such as burst size and frequency using nascent RNA).

Minor comments:

The introduction could be more pedagogical. Right now it is just an accumulation of loosely related and sometimes abruptly introduced statements. For instance, we understand that the authors want to oppose their approach to other extant approaches. However, extant approaches should be better reviewed, some of them are aged structured and perfectly suited for analysing cell cycle data. It would be useful for the reader that an example of observation explained by their model and not explained by other models (age structured or not) is discussed in detail. The model of this work does not explain size control, it just assumes that this holds, and does not discuss cell population aspects. A more nuanced positioning of this approach with respect to the literature would be useful for judging its value.

The meaning of N should be discussed from the very start when the model is introduced.

The authors call constitutive expression the situation when the mean copy number does not depend on the volume. This choice should be clarified as in general constitutive as opposed to specific, localised or transitory expression refers to non-regulated gene expression. It seems to me that in this context, expression is only partially constitutive (independent on the volume).

In figure 1b and for exponential growth the y axis should be log(volume) instead of volume.

The mean field approximation is called both "of novel type" (Discussion) and "which has a long history of successful use in statistical physics" (p4). If something is novel, then one should clearly explain why.<br /> The word "cyclo-stationarity" is used with not much definition. If this means just stationary distribution of the gene products why not use just "stationarity" instead. What means "cyclo"?

A number of properties were called "rare" but it is not clear on what grounds.

I did not find a proof that the copy number distribution has less modes than the concentration distribution.

Referees cross-commenting

Part 1

I agree with the Reviewer 2 that once the master equation accepted the results make sense. But my criticism is different and concerns the master equation itself. In this equation the burst is considered instantaneous, whereas it needs finite time in reality.

Part 2 (response to Part 2 of Rev2)

• concerning replication: in the model this occurs after exactly N_o steps. In reality, replication occurs somewhere between the start of S and G2/M. N_o is in fact a random variable. Probably a new mean field assumption is needed here with some justification, but I have seen nothing in the paper
• concerning biases introduced by the mean field approximation: Figure 2 is a numerical simulation, some analytical estimates could be better. As Figure 2 looks rather convincing, I reclassify this as minor comment.
• concerning nascent mRNA, ON/OFF etc. I disagree. The notion of instantaneous burst with well defined burst size and burst frequency on a stage has a meaning if the lifetime of this stage (which is not mRNA or protein lifetime) is short. The model validity should be clearly stated.
• concerning parsimony, I think that the authors should test it. Are all the parameters identifiable? Is there any overfitting? They could use parameter uncertainty, comparison of training /testing errors, etc. Some details about the parameter fitting method should be provided.

Significance

The strength of this work is that it incorporates in a stochastic gene expression model a number of ideas on size control and dosage compensation that were discussed elsewhere from a cell population point of view. However, the proposed model is based on a number artificial choices that are difficult to justify biologically: a huge number of cell cycle discrete states and inappropriate handling of the timescales characterizing stochastic gene expression. Furthermore, the model is not minimal but depends instead on a huge number of parameters.

I found the paper difficult to read and in the results presentation is not suitable for biologists that would need more details on the justification of the modelling choices and on the experimental validation of the model. For mathematicians, the calculations are rather standard and may seem trivial. I am a systems biologist with a background in mathematics and theoretical physics.

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:

Jia et al. introduce a modeling framework to represent stochastic gene expression, with an explicit representation of cell volume growth, cell cycle progression (and its dependency on cell volume) and gene dosage compensation. The model is very elegant and general in that it can represent a variety of situations, simply as a matter of paramterization. Under a simplifying assumption, the authors derive a number of metrics (include stationary distribution of gene product and power spectrum of gene product fluctuation dynamics), for both absolute number and concentration of gene product molecules. They use their model and derivations to examine under which conditions cell can achieve homeostasis in the concentration of the expressed gene product, despite changes in cell volume and gene copy number following replication. They also present and discuss the conditions giving rise to specific features (i.e. bimodality in stationary distribution, peak in power spectrum) and examine these features in experimental data to conclude to infer the underlying homeostasis strategies.

Major comments:

The model is rather general and powerful. The simplifying assumption seems reasonable (and the authors investigate to some extent its limitations, i.e. Fig. 2). The conclusions are overall convincing.

1. My main concern is that the metrics that the authors use to assess concentration homeostasis (i.e. the γ parameter and the presence/absence of peak in power spectrum) do not seem quite appropriate to describe how much variability/fluctuations in concentration are driven by cell cycle effects. Indeed, the γ parameter measures how much the average concentration in each cell cycle stage varies throughout the cell cycle. However, this variability should be compared to the total variability due to both cell cycle effects and stochastic bursting dynamics. A given level of cell-cycle dependency (say γ=0.2) could be very visible if gene expression is weakly noisy (e.g. B low and <n> high) and completely invisible is gene expression is highly bursty (large B and small <n>). In the latter situation, cell-cycle effects would be meaningless for the cell to minimize. In essence, re-using the authors notations, I think γ / ϕ^1/2, would be a more relevant metric to observe.
2. Similarly, when inspecting the peak in the power spectrum, the weight of the Lorenztian function(s) creating the peak, should be compared to the stationary component (λ_N, u_N in thhe authors' notations).

A complementary analysis including these two points and a discussion the relative contribution of cell-cycle effects and bursting dynamics in the total variability/fluctuation of concentrations would be important to include.

Minor comments:

1. The dashed line on Fig. 3a is defined as κ = sqrt(2)^(1-β). First is this empirical or does it come from a derivation? Second, it seems incomplete since it should depend on ω. Intuitively, this line should correspond to the value of κ that would best mimic balanced biosynthesis in the case where β≠1. In other words, κ should be so that <ρB' / V(t)>_prereplication = <κρB' / V(t)>_postreplication which yields κ = 2^(ω(1-β)) * (ω-1)/ω * [2^(ω(β-1))-1]/[2^((1-ω)(β-1))-1] This indeed simplifies into κ = sqrt(2)^(1-β) when ω=0.5.
2. η is used in the caption of Fig. 2, which is cited on page 4. But it is defined only 2 sections later, on page 6.
3. ω is used in the main text, but only defined in the caption of Fig. 3.
4. ω is defined as "the proportion of cell cycle before replication". Is this in terms of cell cycle stages (i.e. ω=N_0/N) or actual time?
5. Fig. 3 indicates that power spectra are normalized so that G(0)=1, but G(0)=10 on the first two graphs.
6. Page 11: "bimodality in the concentration distribution is significantly less apparent". I would suggest rephrasing "bimodality in the concentration distribution is absent" since there should be no reference to "significance" and bimodality is either present or absent (binary), not less apparent.

Referees cross-commenting

Part 1.

I agree with reviewer 1 that a table of symbols would be helpful. On reviewer 3's second Major Comment, I don't think that the "the lifetime of a stage [has to be] much larger than the time needed to generate a burst". From how the authors write and solve the master equation, I don't think that such a separation of timescale is necessary. The authors should indeed clarify this and if reviewer 3 is correct, then that's indeed a major limitation. On reviewer 3's second Major Comment, I don't think that the "the lifetime of a stage [has to be] much larger than the time needed to generate a burst". From how the authors write and solve the master equation, I don't think that such a separation of timescale is necessary. The authors should indeed clarify this and if reviewer 3 is correct, then that's indeed a major limitation. On reviewer 3's comment "DNA replication [...] does not occur after a fixed number of exponential stages", I don't think I agree with this statement. Cell cycle progression relies on an ensemble of biochemical reactions. Representing this as a set of exponential waiting-time distributions with different means is probably amongst the most general and agnostic ways of representing this. Whether these exponential waiting-times only depend on cell volume is another question. This actually links back to reviewer 3's first Major comment and reviewer 1's comment that the concept of "stage" should be better discussed.

Regarding the need for "estimates of the biases introduced by the mean field approximation" (reviewer 3), I guess that's the goal of figure 2. Maybe reviewer 3 should make more explicit what she/he would like to see.

Regarding the comment from reviewer 3 that "a direct validity test should use datasets of at least two types (total, nascent RNA, etc)". I almost made a related comment in my review, but then I held it off: This issue with using nascent RNA data is that their model does not allow an ON state. They assume that gene products are produced in instantaneous bursts, which is a fair assumption if the lifetime of gene products is large compared to the time the gene stays ON. This is ok if the considered "gene products" are mRNA or proteins, but not nascent RNAs (for which the lifetime is the time to transcribe the gene). I did not make this comment in the end because I think the model is useful regardless. To comply with reviewer 3's request, maybe the authors could use distributions of mRNA and protein products, but I'm not sure that such data exists (since they need cell-cycle-resolved data).

I disagree with the statements that "the proposed model is based on a number artificial choices that are difficult to justify biologically" and that "the model is not minimal but depends instead on a huge number of parameters." In my opinion, the model is elegantly simple to capture the mechanisms under study (i.e. the effect of cell cycle and cell volume on stochastic gene expression). It is expressed so that the model captures a broad range of situations (i.e. it reduces to simpler models as a matter of choosing parameter values, e.g. \Beta=0 => transcription independent of cell cycle; \alpha => \infty cell cycle depends only on size ...). I do not think that a series of exponential distributions for cell cycle progression is inappropriate, it is the most agnostic and general way of representing an ensemble of biochemical reactions that would be meaningless to describe explicitly. Instead, only their dependency on cell volume is taken into account (and in a very general way, i.e. parameters 'a' and \alpha). It is fair to ask the authors to clarify the concept of "stage", but I see this model as being as simple as possible, but not simpler, for the authors' purpose.

Finally, I agree that the paper is probably "not suitable for biologists" but disagree that "for mathematicians, the calculations are rather standard and may seem trivial."

Part 2. Resp. to reviewer 3 on the master equation (Part 1 of Rev3):

Ok, I understand better your comment. What you mean by "the time needed to generate a burst" is the time that the gene produces RNAs, not the lifetime of the gene product (which is 1/d). That's true. It is essentially the same ifdea as what I write in my previous comment about nascent RNA data not being well captured by the model. Again, I think this is fine for "gene products" that are somewhat stable (not the case for nascent RNAs, but ok for mRNAs and proteins). This is fine by me as long as the authors explicit better this limitation of their model.

Part 3. Response to Reviewer 3 (Part 2 of Rev 3)

• concerning replication: Note that the mean field approximation is on cell volume, not on stage progression ("To simplify this model, [...] we ignore volume fluctuations at each stage but retain fluctuations in the time elapsed between two stages", p3). So the time at which replication occurs is already a random variable in the model. It is the sum of all the exponentially distributed random variables corresponding to stages 1 to N_0. The resulting distribution of replication time from the start of cell cycle is a random variable, which can be anything from very deterministic (N_0 very high) to very variable (N_0 very low).
• concerning nascent mRNA, ON/OFF etc. : I'm not sure I get your objection, but the best is probably to let the authors respond to your original comment.
• concerning parsimony: Ok, you're right. The authors should test it.

Significance

The advance of this paper is essentially technical. The authors present a model that incorporates and unifies previously studied effects (cell volume homeostasis, concentration homeostasis, bursting transcription). There is no major conceptual novelty, but the combination of these different aspects and the derivations that authors present are very valuable and might be applicable to interpret data in various species.

The paper is suitable for a physics/mathematics/computational audience. It is rather technical and would not be understood by readers with only a biology background.

Field of expertise of the reviewer: Gene regulation, single-molecule imaging, stochastic modeling.

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 a report on "Concentration fluctuations due to size-dependent gene expression and cell-size control mechanisms," by jia, Singh, and Grima. This manuscript constructs a gene expression model with various factors. Specifically, the effect of cell size on gene expression is considered, which is often ignored by previous studies.

One interesting finding is that the absolute number of the gene products and the concentration can have different distributions. Some predictions of the models are validated by experimental data on E. coli and yeast.

This manuscript uses the mean-field approximation for cell volume, which has good accuracy when the number of stages is large. The usage of the power spectrum has a satisfactory effect on studying the concentration oscillation. Overall the paper was very difficult to follow and digest easily because of all the different factors and mechanisms invoked. It is mainly an issue of providing sufficient details for each of the factors and organizing them in a systematic and logical way. Although there is a supplementary appendix, it was hard to keep track of all the elements in the main mauscript. Perhaps something like Fig 1 of the Appendix can be presented in the main body to outline all the ingredients and how they affect each other.

It might be good to provide a more detailed description of the goal (studying gene product number and concentration under different parameters) after introducing the full and the reduced models. A table of symbols would also be helpful.

Some technical details in the Methods section are in fact helpful in understanding the conclusions. They can be moved to the Results section.

One concern is that the central concept of this manuscript, "stage", is not thoroughly discussed. This concept should have some significant biological meaning, not just be coined for mathematical convenience.

Fig. 1(b) is a little strange. For the left panel, the x-axis (stage) is discrete, then the volume (y-axis) should be a step function, not a straight red line.

Significance

The main advance is a more complete model of gene expression under more realistic organism growth conditions.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Reply to the reviewers

We wish to thank all three reviewers for their thorough examination of our manuscript and their constructive criticism that allowed us to increase its quality. You will see that, following their recommendations, we have included a good amount of new data in the manuscript. Specifically, we added a new figure with experiments proposed by the reviewers (now Fig. 4), as well as Figs. S3 and S4. In addition, we expanded one paragraph of our Discussion to comment on a very recent article published by Huang et al in Nature Structural and Molecular Biology with conclusions pertaining the interplay of Rpd3 and Gcn5 in PHO5 gene regulation. Below we include the point-by-point response (in blue) with the changes we have implemented to address their specific points. All the additions and changes in the manuscript are made in red.

Referee #1

Evidence, reproducibility and clarity

In this manuscript, Novačić et al., investigate into a mechanisms of the non-coding transcriptiondriven regulation of the phosphate-responsive PHO5 gene. The authors employ CRSPRi system to discern direct contribution of the antisense non-coding transcription (CUT025) expressed during phosphate -rich conditions to transcriptional repression of the yeast PHO5 gene and therefore challenging previous study from the Svejstrup's lab that proposed a positive role for non-coding transcription in control of PHO5 gene. They propose a model where non-coding transcription represses PHO5 by mediating recruitment of Rpd3 histone deacetylase leading to altered chromatin structure at PHO5 promoter due to reduced recruitment of the RSC chromatin remodelling complex. Overall, the data presented in the manuscript are of a good quality, experiments are well controlled and nicely presented. Manuscript is well written. My specific comments are below: 1. I am somewhat confused by the data presented in Figure 5. While there is similar impact on the chromatin structure seen in rrp6D and air1Dair2D strains (Fig 5C) that corresponds to more "closed" configuration of chromatin , it is not consistent with H3 ChIP data that show higher nucleosome occupancy across PHO5 UAS in rrp6D but loss of nucleosomes in the double mutant (or there is a mistake perhaps while plotting the data?)

We now realize that the data was plotted confusingly, and we apologize for it. While doing the H3 ChIP experiment we only prepared the +Pi samples for the air1Δ air2Δ double mutant. In the figure we only included this one data point for the double mutant, which could lead to the false conclusion that at other timepoints there are no histones at its PHO5 promoter region. We decide to remove this data point from the figure to avoid the confusion and only keep the air1Δ air2Δ data for the ClaI assay. We believe that this should not be an issue as this data point is not critical for the conclusions we are making.

1. To further explore direct link between nc transcription, Rpd3 and rrp6 mediated effect, I suggest to test the effect on PHO5 induction upon rpd3 and rrp6 deletions in CRISPRi CUT025 background.

We performed this experiment and now include it as Fig. S3 in the manuscript. As expected, expressing the CRISPRi system only made difference when Rpd3 was present.

1. It seems that most noticeable effect of blocking nc transcription by an elegant approach that utilizes CRISPRi system on the phosphatase activity is seen between 0-1.5h of induction. I suggest taking additional time points at 30-45 min.

We took additional timepoints and the results were incorporated as the new Fig. 5E. The CRISPRi effect resulting in higher acid phosphatase activity was still most noticeable after 1,5 h of induction. This was mostly in line with the fact that the difference in PHO5 mRNA levels was most pronounced after 30 min of induction (Fig. 5D), as the time needed to achieve measurable protein level after induction can lag significantly for secretory proteins, such as acid phosphatase. Secretory proteins are cotranslationally translocated into the ER, after which they traverse the secretory pathway and undergo modifications before being finally exported to the periplasm where their activity can be measured. Consequently, the increase in acid phosphatase activity upon induction is only measurable after at least an hour.

1. How do authors explain that the effect of the exosome mutations are reversed and phosphatase activity is increased at later time point (20 h, Fig 2A)? I suggest using more distinct colour for dis3 mutants.

That effect is indeed somewhat surprising. We hypothesize that the effects we are seeing after 20 h reflect the specific conditions of prolonged induction, i.e. keeping the chromatin open or semi-open for a very long period of time, which do not necessarily reflect the early gene induction period that we are using as a read-out of the effect of different mutations on acid phosphatase expression kinetics. We previously noticed a similar effect with chromatin remodeler-related mutants (e.g. rsc2Δ, unpublished result from S. Barbarić group), which speak in favour of the prolonged induction conditions resulting in a chromatin state with its own specialized cofactor requirements. We therefore consider the chromatin state after prolonged induction a topic for another study, however, we now comment on this effect in the manuscript. The dis3 mutants are now shown in more distinct colours.

1. Figure 5A -label "H3 ChIP"

The label was added.

1. Error bars are quite high in Fig 1C, perhaps it is worth repeating the experiment

Since significant differences in PHO5 mRNA levels can be seen between wt and rrp6Δ mutant cells at 0,75 and 3 h of induction, we feel that the higher error bars at 5 h of induction are not worth repeating the experiment – especially since the values are bound to converge to a similar one after a longer induction period, as demonstrated in Fig. 1D.

Significance

significant of interest for general audience

Referee #2

Evidence, reproducibility and clarity

The authors study the PHO5 locus, which is known to a have antisense transcript and that has previously been shown the be important for activation of Pho5 sense transcription. The authors challenge the idea by an extensive analyses. They show the Pho5-AS represses sense transcription, and thus fits in the category as AS repressors instead of activators. They show a correlative data that when antisense goes down and sense goes up. They show that increase antisense levels leads to decrease sense levels. They use mutants of decay pathways to increase the levels antisense transcription. Moreover, they used crispri to repress the antisense transcript. Lastly, they show that histone deacetylation represses Pho5 sense. The data in the manuscript is convincing, and well presented. One thing that needs further clarification is the strategy to increase anti-sense levels by deletion mutants of decay or depletion of decay pathways. While it is clear that this stabilizes the pho5-AS and decrease pho5-sense, it is not clear that this causes an increase in transcription. Perhaps, it is possible that antisense transcript itself has a repressive effect. If one really wanted to increase antisense transcription than the antisense promoter should be increased in strength. On the other the CriprI experiment is very convincing. I am surprised how well the crisprI system works, it is thought to be not so efficient at blocking elongating polymerase and good at blocking initiation.

We thank the reviewer for this feedback. We performed additional experiments which you will find described below. Based on the results, we would like to keep the point about AS transcription causing the effect.

Major comments: - Are the key conclusions convincing? Perhaps, the conclusion that increased transcription leads to repression is not completely convincing. The authors use mutants in rrp6, exosome, and nrd1 to increase Pho5-AS transcription elongation. However, I am always under impression that these mutants stabilize the transcript. And the authors acknowledge this in their manuscript. So how do you discriminate between increased stability versus increased elongation? I support the conclusion that inhibition of Pho5-AS leads to increase Pho5-S. However, increase in elongation is not directly demonstrated. While still possible, it is equally possible that a more stable pho5-AS transcript has a repressive an effect on Pho5-AS. - Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether? See above. Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation. If the authors want to keep the message that increased transcription of Pho5-AS leads to more repression that may need to consider additional experiments. For example, increasing transcription from the antisense promoter.

We performed the proposed experiment and now include it in the manuscript as Fig. 4AB. Briefly, we inserted the strong constitutive TEF1 promoter in the antisense configuration downstream of the PHO5 gene ORF, so that it drives AS transcription. The results of this experiment very clearly show the inverse relationship between PHO5 mRNA and AS transcripts levels at +Pi conditions. Importantly, this strong constitutive AS transcription had an even more pronounced effect on PHO5 gene expression than deletion mutant backgrounds (in which, like in wt cells, the AS promoter is presumably weak), and did not allow for full level of PHO5 gene expression to be reached. To verify that the AS RNA itself does not have a regulatory role, but rather the act of its transcription represses the corresponding gene, we performed an additional experiment with appropriate diploid strains. The design of this experiment is standardly used to test whether an AS transcript can work in trans (for example see Nevers et al. 2018 NAR Fig. 6). This experiment is now included as Fig. 4C. Together, the results of these experiments paint a clear picture of AS transcription, and not AS level/stability itself, driving the repression of the PHO5 gene.

• Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments. To me this is an optional experiment, but it would benefit the manuscript
• Are the data and the methods presented in such a way that they can be reproduced? yes - Are the experiments adequately replicated and statistical analysis adequate? yes

Minor comments: - Specific experimental issues that are easily addressable. - Are prior studies referenced appropriately? yes - Are the text and figures clear and accurate? Yes - Do you have suggestions that would help the authors improve the presentation of their data and conclusions? no

Significance

• Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field. The manuscript challenges previous work where it was claimed that Pho5-AS is important for activation of Pho5-S. As such, it is important work. In the field of noncoding the transcription the Pho5-AS fits in a class of AS transcript that has been well described.
• Place the work in the context of the existing literature (provide references, where appropriate). See above.
• State what audience might be interested in and influenced by the reported findings. In researchers in field of transcription, chromatin, and more specifically in yeast gene regulation.
• Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate. Chromatin, transcription, yeast.

Referee #3

Evidence, reproducibility and clarity

Novačić et al present a manuscript entitled "Antisense non-coding transcription represses the PHO5 model gene via remodeling of promoter chromatin structure" which is a locus-specific follow up to previous studies from Soudet and Stutz groups on genome-wide analysis of transcription interference mediated by antisense transcripts in S cerevisiae. Critically, the authors here employ a CRISPRi approach to reduce antisense transcription from reaching the PHO5 promoter and in doing so show that kinetics of PHO5 induction are increased as would be predicted from their previous model. Additionally, they show predicted epistasis between rpd3 and rrp6 on PHO5 expression and gcn5 and rrp6 that are consistent with their model. Comments are relatively minor but should be addressed. Introduction p3. "This mechanism was subsequently explored genome-wide in yeast, which revealed a group of genes that in the absence of Rrp6 accumulate AS RNAs and are silenced in an HDACdependent manner (14)." This sentence appears awkward- perhaps move "in the absence of Rrp6" to after "AS RNAs"?

Corrected as proposed.

p3 "Under a high phosphate concentration Pho4 undergoes phosphorylation by the cyclindependent-kinase (Pho80-Pho85)" Since "the" is used, don't use parentheses around Pho80-Pho85

Corrected as proposed.

Methods Give amount/concentration of glycine used in quenching formaldehyde for ChIP. Give the exact wash conditions and buffers not "extensively"

All of those details are now provided in the manuscript. Figure 4C.

Describe schematic in legend

It is now described.

Figure 4D. Indicate time of induction in legend.

This was lacking for Figs. 4B-C (now 5B-C) so we added it there.

Figure 5A. air∆ data are missing from later time points?

Please see our first response to Reviewer 1. We removed the air1Δ air2Δ double mutant data, as we only had one data point for it in this assay.

Figure 6. Legend needs to indicate what Pi conditions are. Since PHO5 expressed, appears to be low Pi. An issue that needs to be discussed is that rpd3∆ appears to decrease expression of PHO5 AS. Is this simply because of increased PHO5 expression? Does rpd3∆ have any effects on AS in high Pi? This is important to interpret if effects of rrp6 and rpd3 are epistatic or additive.

We thank the Reviewer for bringing this to our attention. To explore the effect of rpd3Δ on PHO5 AS level, we quantified the PHO5 AS transcript by RT-qPCR with cells grown in (chemically defined) high Pi medium, which we now include in Fig. 7A. We find that rpd3Δ mutation has practically no effect on PHO5 AS transcript level both in the wt and the rrp6Δ mutant background. This result speaks in favor of rrp6Δ and rpd3Δ being epistatic rather than additive.

Figure 7. Sth1-CHEC data are hard to interpret. Some sort of quantification might be required as effects are not clear from the browser track nor is it clear from browser track that the results are reproducible. Examination of Sth1-AA effects in gcn5∆ background might be more compelling that the effect on RSC is via acetylation. Otherwise it is a bit hard to say as RSC could be functioning in parallel to the acetylation-dependent pathways implicated.

We agree that the presumption that histone acetylation recruits RSC to the PHO5 gene promoter had to be tested. We therefore include the experiment involving Sth1-AA depletion in the gcn5Δ background as Fig. 8A. This experiment was complicated by the fact that RSC is highly abundant (and at the same time essential for cell viability), but we resolved this by starting to deplete RSC two hours before gene induction. These results position RSC and Gcn5 in the same pathway. In contrast, more complete Sth1 depletion severely impaired viability of the rrp6Δ mutant, making it hard to interpret the effect, so we now include this result as Fig. S4.

To show the effect of AS transcription on RSC recruitment to the PHO5 promoter more quantitatively, we re-analyzed the Sth1-CHEC data (for two independent biological replicates) and now include the log2 values for the changes in Sth1 binding in the text of the manuscript.

Significance

The work is focused and narrower in impact but important because direct tests of locus-specific effects are performed, validating models from previous genomic analyses. **Referees cross-commenting**

I think the other reviews are very reasonable. I would just suggest to the authors that they think carefully about the reviews and decide what they think is most valuable to improving the work/presentation

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

Novačić et al present a manuscript entitled "Antisense non-coding transcription represses the PHO5 model gene via remodeling of promoter chromatin structure" which is a locus-specific follow up to previous studies from Soudet and Stutz groups on genome-wide analysis of transcription interference mediated by antisense transcripts in S cerevisiae. Critically, the authors here employ a CRISPRi approach to reduce antisense transcription from reaching the PHO5 promoter and in doing so show that kinetics of PHO5 induction are increased as would be predicted from their previous model. Additionally, they show predicted epistasis between rpd3 and rrp6 on PHO5 expression and gcn5 and rrp6 that are consistent with their model. Comments are relatively minor but should be addressed.

Introduction

p3. "This mechanism was subsequently explored genome-wide in yeast, which revealed a group of genes that in the absence of Rrp6 accumulate AS RNAs and are silenced in an HDAC-dependent manner (14)."

This sentence appears awkward- perhaps move "in the absence of Rrp6" to after "AS RNAs"?

p3 "Under a high phosphate concentration Pho4 undergoes phosphorylation by the cyclin-dependent-kinase (Pho80-Pho85)"

Since "the" is used, don't use parentheses around Pho80-Pho85

Methods

Give amount/concentration of glycine used in quenching formaldehyde for ChIP. Give the exact wash conditions and buffers not "extensively"

Figure 4C. Describe schematic in legend

Figure 4D. Indicate time of induction in legend.

Figure 5A. air∆ data are missing from later time points?

Figure 6. Legend needs to indicate what Pi conditions are. Since PHO5 expressed, appears to be low Pi. An issue that needs to be discussed is that rpd3∆ appears to decrease expression of PHO5 AS. Is this simply because of increased PHO5 expression? Does rpd3∆ have any effects on AS in high Pi? This is important to interpret if effects of rrp6 and rpd3 are epistatic or additive.

Figure 7. Sth1-CHEC data are hard to interpret. Some sort of quantification might be required as effects are not clear from the browser track nor is it clear from browser track that the results are reproducible. Examination of Sth1-AA effects in gcn5∆ background might be more compelling that the effect on RSC is via acetylation. Otherwise it is a bit hard to say as RSC could be functioning in parallel to the acetylation-dependent pathways implicated.

Significance

The work is focused and narrower in impact but important because direct tests of locus-specific effects are performed, validating models from previous genomic analyses.

Referees cross-commenting

I think the other reviews are very reasonable. I would just suggest to the authors that they think carefully about the reviews and decide what they think is most valuable to improving the work/presentation

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

Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate).

The authors study the PHO5 locus, which is known to a have antisense transcript and that has previously been shown the be important for activation of Pho5 sense transcription. The authors challenge the idea by an extensive analyses. They show the Pho5-AS represses sense transcription, and thus fits in the category as AS repressors instead of activators. They show a correlative data that when antisense goes down and sense goes up. They show that increase antisense levels leads to decrease sense levels. They use mutants of decay pathways to increase the levels antisense transcription. Moreover, they used crispri to repress the antisense transcript. Lastly, they show that histone deacetylation represses Pho5 sense.

The data in the manuscript is convincing, and well presented. One thing that needs further clarification is the strategy to increase anti-sense levels by deletion mutants of decay or depletion of decay pathways. While it is clear that this stabilizes the pho5-AS and decrease pho5-sense, it is not clear that this causes an increase in transcription. Perhaps, it is possible that antisense transcript itself has a repressive effect. If one really wanted to increase antisense transcription than the antisense promoter should be increased in strength. On the other the CriprI experiment is very convincing. I am surprised how well the crisprI system works, it is thought to be not so efficient at blocking elongating polymerase and good at blocking initiation.

Major comments:

• Are the key conclusions convincing?

Perhaps, the conclusion that increased transcription leads to repression is not completely convincing. The authors use mutants in rrp6, exosome, and nrd1 to increase Pho5-AS transcription elongation. However, I am always under impression that these mutants stabilize the transcript. And the authors acknowledge this in their manuscript. So how do you discriminate between increased stability versus increased elongation? I support the conclusion that inhibition of Pho5-AS leads to increase Pho5-S. However, increase in elongation is not directly demonstrated. While still possible, it is equally possible that a more stable pho5-AS transcript has a repressive an effect on Pho5-AS.<br /> - Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

See above. - Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

If the authors want to keep the message that increased transcription of Pho5-AS leads to more repression that may need to consider additional experiments. For example, increasing transcription from the antisense promoter. - Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

To me this is an optional experiment, but it would benefit the manuscript - Are the data and the methods presented in such a way that they can be reproduced?

yes - Are the experiments adequately replicated and statistical analysis adequate?

yes

Minor comments:

• Specific experimental issues that are easily addressable.
• Are prior studies referenced appropriately?

yes - Are the text and figures clear and accurate?

Yes - Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

no

Significance

• Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field.

The manuscript challenges previous work where it was claimed that Pho5-AS is important for activation of Pho5-S. As such, it is important work. In the field of noncoding the transcription the Pho5-AS fits in a class of AS transcript that has been well described. - Place the work in the context of the existing literature (provide references, where appropriate).

See above. - State what audience might be interested in and influenced by the reported findings.

In researchers in field of transcription, chromatin, and more specifically in yeast gene regulation. - Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate.

Chromatin, transcription, yeast.

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 manuscript, Novačić et al., investigate into a mechanisms of the non-coding transcription-driven regulation of the phosphate-responsive PHO5 gene. The authors employ CRSPRi system to discern direct contribution of the antisense non-coding transcription (CUT025) expressed during phosphate -rich conditions to transcriptional repression of the yeast PHO5 gene and therefore challenging previous study from the Svejstrup's lab that proposed a positive role for non-coding transcription in control of PHO5 gene. They propose a model where non-coding transcription represses PHO5 by mediating recruitment of Rpd3 histone deacetylase leading to altered chromatin structure at PHO5 promoter due to reduced recruitment of the RSC chromatin remodelling complex.

Overall, the data presented in the manuscript are of a good quality, experiments are well controlled and nicely presented. Manuscript is well written. My specific comments are below:

1. I am somewhat confused by the data presented in Figure 5. While there is similar impact on the chromatin structure seen in rrp6D and air1Dair2D strains (Fig 5C) that corresponds to more "closed" configuration of chromatin , it is not consistent with H3 ChIP data that show higher nucleosome occupancy across PHO5 UAS in rrp6D but loss of nucleosomes in the double mutant (or there is a mistake perhaps while plotting the data?)
2. To further explore direct link between nc transcription, Rpd3 and rrp6 mediated effect, I suggest to test the effect on PHO5 induction upon rpd3 and rrp6 deletions in CRISPRi CUT025 background.
3. It seems that most noticeable effect of blocking nc transcription by an elegant approach that utilizes CRISPRi system on the phosphatase activity is seen between 0-1.5h of induction. I suggest taking additional time points at 30-45 min.
4. How do authors explain that the effect of the exosome mutations are reversed and phosphatase activity is increased at later time point (20 h, Fig 2A)? I suggest using more distinct colour for dis3 mutants.
5. Figure 5A -label "H3 ChIP"
6. Error bars are quite high in Fig 1C, perhaps it is worth repeating the experiment

Significance

significant

of interest for general audience

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Revision Plan

1. General Statements

We really appreciate the positive comments and suggestions of the reviewers on our submitted manuscript. We think we will be able to solve the issues inquired by reviewers by adding new data and revising the phrases as detailed below.

2. Description of the planned revisions

Reviewer #1:

Major comments

Localization analysis of a transiently expressed MAP70 transgene with inactivating phosphosite mutations would be important to see whether the identified conserved phosphosites are relevant for MAP70 interaction with MTs. This experiment could be performed rapidly using transient expression in BY-2 cells.

We agree on the importance of this analysis. Therefore we are currently preparing fluorescent markers of Nt-MAP70-2-like and its phospho-blocked (Ala) version to coexpress with MT and nuclear markers in BY-2 cells. We estimate that we need three more months to complete this experimsnt.

The authors propose that PP2 blocks phragmoplast formation by preventing phosphorylation of class II Kinesin-12 proteins. In support, authors show that PP2 treatment correlates with a decrease in KIN12A phosphopeptide count (not fully abolished) and its failure to localize to emerging phragmoplasts in BY-2 cells and Physcomitrium. As class II Kinesis-12 proteins have been previously implicated in phragmoplast assembly this is a fairly reasonable hypothesis, but would benefit from the analysis of transgenic KIN12A variants carrying inactivating (A) or potentially activating (D/E) phosphosite mutations. Is loss of phosphorylation sufficient to prevent phragmoplast localization? Can an activated variant rescue PP2-induced KIN12A localization and cell division defects? As above, using transient expression in BY-2 cells would be a fast approach to tackle these questions.

We are currently preparing fluorescent markers of phospho-blocked (Ala) and phospho-mimic (Asp) versions of KIN12A (PAKRP1) to coexpress with MT and nuclear markers in BY-2 cells. We will check whether they localize to phragmoplast and also test PP2 effects. We would need three more months to complete these analyses.

Reviewer #2:

Major comments

• The manuscript would strongly benefit from being revised by a native english speaker. There are many unusual or awkward formulation, in particular in the abstract.

We apologize for unnatural sentences. After adding new data and correcting the manuscript, we will ask a native english speaker to revise it.

Reviewer #3:

Major comments

The major concern is lack of evidence to connect MAP70 and MT disruption upon treatment with PD-180970, in contrast to PP2, which was shown to affect localization of Kinesin-12. I wonder if authors could use taxol to stabilize MTs, then observe the localization of MAP70 with application of PD-180970?

As we responded to reviewer 1, we are preparing the fluorescent marker of Nt-MAP70-2-like to coexpress with MT and nuclear markers in BY-2 cells. By using this multi-color marker, we will test whether PD-180970 affects the localization of MAP70 on MTs, also using taxol. However, in our experiene, taxol is not a very effective inhibitor and may not work in our transient expression system in BY-2 cells. In that case, we will analyze whether phospho-mimic (Asp) version can prevent MT disruption in the presence of PD-180970 to assess the relation of PD-180970, MAP70 and MT disruption.

I have another concern on the action of PD-180970. PD-180970 appears to affect ubiquitously indispensable proteins for MTs. If PD-180970 disrupt MT by inhibiting phosphorylation of some MAPs, it must need time for turnover of proteins phosphorylated before PD-180970 was applied. In the proteomics experiment, author treated the cells with the compounds for 8-9 hr. On the other hand, in BY-2 cells, PD-18970 disrupted MTs only 30 min after application of PD-180970. I wonder if proteins were replaced during the 30 min. Could authors examine how long it takes to affect interphase MTs? If PD-180970 disrupts MTs in a 5-10 min like oryzalin, it is unlikely that inhibition of phosphorylation of proteins like MAP70 caused MT disruption. Rather, it may inhibit some proteins that have activity to disrupt microtubules but are usually inactivated by phosphorylation or inhibit something directly without phosphorylation.

We agree that there is no evidence that PD-180970 disrupts MTs by inhibiting phosphorylation of MAP70. In our live-imaging system, in which reagents are added to liquid cultivation medium, the time from the reagent application to the arrival to each cell varies. Therefore, in order to accurately measure the time required for the inhibitor to take effect, it is necessary to design a new assay system, such as using fluorescent dyes to monitor the reagent's diffusion. In addition, since some reactions mediated by protein phosphorylation occur rapidly, minute-order observations might not be sufficient. Therefore, as an alternative strategy to assess the direct involvement of MAP70 phosphorylation on MT stabilization, we will examine whether PD-180970 induces MT disruption using strains expressing the phospho-blocked (Ala) and phospho-mimic (Asp) versions of MAP70 described above.

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

Reviewer #1:

Minor comments

The authors identified the analogs PD-166326 and PP1 as potent inhibitors of cell division. For completeness, it would be interesting to include a description of these arrest phenotypes and how they compare with that of PD180870 or PP2.

We have added the effects of all tested compounds on Arabidopsis embryos in Fig. S3C and Table S1. Based on this data and the results of tobacco BY-2 cells, we have compared the effects of PD-166326 and PD180870, and PP1 and PP2 in Results.

Although there are two more obvious candidates in the phosphoproteome datasets on which the authors focus on, there is very little discussion on whether the other top hits and whether they might be involved in cell division. On a related note, there is no discussion on the specificity of these compounds and the likelihood of phenotypes unrelated to cell division.

We have added the information of “Similar proteins in Arabidopsis” and “Description and putative functions” for all identified candidates for PD-180970 and PP2 in Table S2 and S3, respectively. With referring this information, we have added the sections to describe the possible contributions of these candidates on MT organization and phragmoplast formation in Results. In addition, we have described the specificity of these compounds and the phenotypes unrelated to cell division in the section for the results of Arabidopsis roots (Fig. S2A).

1st results section:

"...developed into the globular stage without causing morphological defects..."

Should omit the word "causing" or replace with "any/detectable"

We have omitted the word "causing".

Reviewer #2:

Even if the identification of the kinase(s) targeted by these two compounds is missing, the characterisation of at least two downstream effectors of these elusive kinase(s) inhibited by PD-180970 and PP2 is an important step forward. I would recommend to this point make very clear in the writing (e.g. already in the abstract). Upon a superficial reading, the reader could assume that MAP70s and PAKRP1s are the direct molecular targets of these compounds.

We appreciate the very positive comments. To clarify this point, in addition to the following responses to each suggestion, we have changed the last sentense of the abstract to “These properties make PD-180970 and PP2 useful tools for transiently controlling plant cell division at key manipulation nodes that are conserved in diverse plant species”.

Major comments

• I would modify the title to shift the emphasis from the methodology to the biological targets identified.

We have changed the title to “Identification of novel compounds inhibiting microtubule organization and phragmoplast formation in diverse plant species”.

• Concerning MAP70s the authors claim that there is little functional data about this family. Yet, a recent paper (https://www.science.org/doi/10.1126/sciadv.abm4974) identifies MAP70-5 as necessary for the proper organisation of CMTs in the endodermis and its ability to actively remodel to accommodate emergence of the lateral root primordium in Arabidopsis thaliana. This could provide a functional context to test several of the predictions that the authors list in the discussion.

We have referred this paper in Results and Discussion, as “MAP70-5 was reported to increase MT length in vitro and to reorganize cortical MTs to alter the endodermal cell shape for lateral root initiation, suggesting that MAP70-5 mediates dynamic change of MT arrays”.

Minor comments

• The narrative would be improved by moving the section "PD-180970 and PP2 do not irreversibly damage viability" before the phosphoproteomic section.

We have moved the “irreversibly” section to before the “phosphoproteomics” section.

Reviewer #3:

Minor comments

In supplemental data, authors show only 12 or 14 candidates of the target. It is interesting how other MAPs including homologues of MAP70 and Kiesnin-12 in BY-2 cells were scored in the phospho-proteomics assay. I suggest authors show longer lists of proteomics including other MAPs. It would be valuable information for the research community.

We apologize for not providing the complete dataset. We have added Dataset S1 of total protein sequences that we predicted from published RNA-sea data of BY-2 cells, and all identified proteins of phosphoproteomics assay for PD-180970 and PP2 in Datasets S2 and S3, respectively. We have moved the lists of top candidates to Tables S2 and S3.

In Abstract, authors should mention that the two compounds reduced phosphorylation level of diverse proteins including MAP70 and Kinesin-12. This is very important results and, otherwise, it may cause misunderstanding of the activity of the compounds. In addition to this, it is better to rephrase the following sentence. "presumably by inhibiting MT-associated proteins (MAP70)" with "presumably by inhibiting phosphorylation of MT-associated proteins (MAP70)."

To avoid such a misunderstanding, we have changed the descriptions in Abstract to “Phosphoproteomic analysis showed that these compounds reduced phosphorylation level of diverse proteins. In particular, PD-180970 inhibited phosphorylation of the conserved serine residues in MT-associated proteins (MAP70). PP2 significantly reduced the phosphorylation of class II Kinesin-12, and impaired its localization at the phragmoplast emerging site”. Due to this change, the suggested sentence was eliminated. Also in Discussion, we have mentioned the reduction of phosphorylation of various proteins by stating, "we found that PD-180970 and PP2 reduced the phosphorylation levels of diverse proteins. These parts may be further modified depending on the results of the phospho-blocked (Ala) and phospho-mimic (Asp) analyses.

Page7 line 1st. it would be better to insert "of MAP70 family" after "in the conserved MT-binding domain" because the MT binding domains are unique to the MAP70 family. I could not understand why this is " (2nd line) consistent with PD-18970 severely disrupting all the tested MT structure". At current stage, there is no evidence that dephosphorylation of MAP70 caused the microtubule disruption. I suggest authors remove the sentence (", which was~MT structures").

We agreed on both points and have corrected them as the reviewer suggested.

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 by Kimata and colleagues describes identification of compounds that inhibit microtubule organization and cell division in plants. In this manuscript, authors screened two chemical libraries and successfully found two compounds, PD-180970 and PP2", as potent inhibitors of cell division. Because the two compounds act as kinase inhibitor in animal cells, authors analyzed their effects on phosphorylation by using BY-2 suspension cells. Among the affected proteins, authors focused on two microtubule-related proteins, MAP70 and Kinesin-12. Authors showed that PD-180970 disrupt all structures of microtubules in BY-2 cells. All members of Arabidopsis MAP70 shared the target residues, although Arabidopsis map70 mutants grew normally. PP2 abolished the localization of Kinesin-12 to phragmoplasts, and inhibit the cell plate formation. Both compounds worked in other plant species including moss. I think highly of the unique screening system using Arabidopsis zygote and BY-2 cells developed by the authors. Although the direct targets and specificity of the compounds are still to be determined, I think the two compounds should become powerful tools in plant cell biology in future. Generally, the manuscript is well written, and the data are of high quality. I have, however, several suggestions as below.

Major

The major concern is lack of evidence to connect MAP70 and MT disruption upon treatment with PD-180970, in contrast to PP2, which was shown to affect localization of Kinesin-12. I wonder if authors could use taxol to stabilize MTs, then observe the localization of MAP70 with application of PD-180970?

I have another concern on the action of PD-180970. PD-180970 appears to affect ubiquitously indispensable proteins for MTs. If PD-180970 disrupt MT by inhibiting phosphorylation of some MAPs, it must need time for turnover of proteins phosphorylated before PD-180970 was applied. In the proteomics experiment, author treated the cells with the compounds for 8-9 hr. On the other hand, in BY-2 cells, PD-18970 disrupted MTs only 30 min after application of PD-180970. I wonder if proteins were replaced during the 30 min. Could authors examine how long it takes to affect interphase MTs? If PD-180970 disrupts MTs in a 5-10 min like oryzalin, it is unlikely that inhibition of phosphorylation of proteins like MAP70 caused MT disruption. Rather, it may inhibit some proteins that have activity to disrupt microtubules but are usually inactivated by phosphorylation or inhibit something directly without phosphorylation.

Minor

In supplemental data, authors show only 12 or 14 candidates of the target. It is interesting how other MAPs including homologues of MAP70 and Kiesnin-12 in BY-2 cells were scored in the phospho-proteomics assay. I suggest authors show longer lists of proteomics including other MAPs. It would be valuable information for the research community.

In Abstract, authors should mention that the two compounds reduced phosphorylation level of diverse proteins including MAP70 and Kinesin-12. This is very important results and, otherwise, it may cause misunderstanding of the activity of the compounds. In addition to this, it is better to rephrase the following sentence. "presumably by inhibiting MT-associated proteins (MAP70)" with "presumably by inhibiting phosphorylation of MT-associated proteins (MAP70)."

Page7 line 1st. it would be better to insert "of MAP70 family" after "in the conserved MT-binding domain" because the MT binding domains are unique to the MAP70 family. I could not understand why this is " (2nd line) consistent with PD-18970 severely disrupting all the tested MT structure". At current stage, there is no evidence that dephosphorylation of MAP70 caused the microtubule disruption. I suggest authors remove the sentence (", which was~MT structures").

Significance

Redundancy of genes prevent researchers from exploring the genetic mechanisms of cell division. Time-specific manipulation of plant cell division by optogenetics or pharmacology has not been established. Identification of compounds that can specifically affect cell division is desired for further investigation of plat cell division. Although the direct targets and specificity of the compounds are still to be determined, I think the screening system and the two compounds identified by the authors should become powerful tools in plant cell biology in future. This work will influence not only plant biologists but also broad readership including cell/developmental biologists and chemical biologists.

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:

Kimata et al. report on the identification and characterisation of two compounds inhibiting cell division in plants. Motivated by the need to circumvent genes function redundancy to study cell division in plants, the authors screened 170 biologically active compounds for inhibitors of the first highly stereotypical division of the Arabidopsis zygote. They identify two compounds PD-180970 and PP2 that very potently block this division. Monitoring the effect of these compounds on microtubules dynamics and using high resolution imaging on BY-2 cell cultures, they conclude that PD-180970 inhibits MT organization while PP2 inhibits phragmoplast formation. These two compounds have reversible effects and are active in several plants species ( Arabidopsis, Tobacco, Cucumber and Physcomitrella). Both compounds targets kinases in animal cells. To shed light on the molecular mechanism perturbed by these compounds, the authors performed a phospho-proteomic profiling of synchronised BY-2 cells upon treatment by either of this compounds, controlled by inactive analogs. They identify two proteins which phosphorylation is severely reduced by the compounds. PD-180970 reduces the phosphorylation of members of the MAP70 at three conserved S residues, two mapping in the microtubule binding domains. PP2 reduces phosphorylation of PAKRP1/KIN12A and PAKRP1L//KIN12B a pair of phragmoplasts-associated kinesins. The authors show that PP2 disrupts the phragmoplast-localisation of the both kinesins, phenocopying the effects of the double mutant pakrp1/pakrp1l and thus providing a likely molecular mechanisms for the effects of PP2.

Overall the manuscript is solid, the experiments well executed and controlled and the results precisely. The conclusions are supported by the data and the manuscript is clearly structured.

Even if the identification of the kinase(s) targeted by these two compounds is missing, the characterisation of at least two downstream effectors of these elusive kinase(s) inhibited by PD-180970 and PP2 is an important step forward. I would recommend to this point make very clear in the writing (e.g. already in the abstract). Upon a superficial reading, the reader could assume that MAP70s and PAKRP1s are the direct molecular targets of these compounds.

Major comments:

• I would modify the title to shift the emphasis from the methodology to the biological targets identified.
• Concerning MAP70s the authors claim that there is little functional data about this family. Yet, a recent paper (https://www.science.org/doi/10.1126/sciadv.abm4974) identifies MAP70-5 as necessary for the proper organisation of CMTs in the endodermis and its ability to actively remodel to accommodate emergence of the lateral root primordium in Arabidopsis thaliana. This could provide a functional context to test several of the predictions that the authors list in the discussion.
• The manuscript would strongly benefit from being revised by a native english speaker. There are many unusual or awkward formulation, in particular in the abstract.

Minor comments:

• The narrative would be improved by moving the section "PD-180970 and PP2 do not irreversibly damage viability" before the phosphoproteomic section.

Significance

Plant cell biologists interested in cell division and microtubules will find this pre-print enticing. The compounds identified will reveal useful tools to analyse cell division in plants and the manuscript provides a significant technical advance. Although my expertise does not lay in the field of chemical inhibitors of cell division, there are to my knowledge, no compounds that selectively inhibit phragmoplast growth like PP2. The manuscript paves the way for further studies such as genetic suppressor screens to identify the plant kinase(s) targeted by these compounds.

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 manuscript, Kimata and colleagues describe the identification of two reversibly acting compounds that affect specific stages of cell division through a chemical screen in in vitro cultured Arabidopsis zygotes. They further characterize the effects of these compounds on cell division using advanced imaging techniques, and demonstrate that they perturb cell cycle progression in multiple plant species and systems, thus acting on conserved pathways. Finally, using phosphoproteomics and transgenic approaches in cultured plant cells the authors identified potential indirect targets of these compounds. The work described is very thorough and well presented, with conclusions supported by the data. In addition it provides information on two compounds that can be broadly applied in plant molecular research and demonstrate the feasibility of an in vitro ovule culture system for chemical screening in plants.

Major comments

Localization analysis of a transiently expressed MAP70 transgene with inactivating phosphosite mutations would be important to see whether the identified conserved phosphosites are relevant for MAP70 interaction with MTs. This experiment could be performed rapidly using transient expression in BY-2 cells.

The authors propose that PP2 blocks phragmoplast formation by preventing phosphorylation of class II Kinesin-12 proteins. In support, authors show that PP2 treatment correlates with a decrease in KIN12A phosphopeptide count (not fully abolished) and its failure to localize to emerging phragmoplasts in BY-2 cells and Physcomitrium. As class II Kinesis-12 proteins have been previously implicated in phragmoplast assembly this is a fairly reasonable hypothesis, but would benefit from the analysis of transgenic KIN12A variants carrying inactivating (A) or potentially activating (D/E) phosphosite mutations. Is loss of phosphorylation sufficient to prevent phragmoplast localization? Can an activated variant rescue PP2-induced KIN12A localization and cell division defects? As above, using transient expression in BY-2 cells would be a fast approach to tackle these questions.

Minor comments

The authors identified the analogs PD-166326 and PP1 as potent inhibitors of cell division. For completeness, it would be interesting to include a description of these arrest phenotypes and how they compare with that of PD180870 or PP2.

Although there are two more obvious candidates in the phosphoproteome datasets on which the authors focus on, there is very little discussion on whether the other top hits and whether they might be involved in cell division. On a related note, there is no discussion on the specificity of these compounds and the likelihood of phenotypes unrelated to cell division.

1st results section: "...developed into the globular stage without causing morphological defects..." Should omit the word "causing" or replace with "any/detectable"

Significance

The work described in this manuscript identified and characterized two compounds that affect specific processes important for cell division in plant cells. Furthermore, the authors demonstrate the feasibility of an in vitro ovule culture system for chemical screening of specific processes in early plant embryos. The identified compounds are effective in multiple tissues in phylogenetically distant land plants and their effects are reversible. These compounds can be useful to, for example, manipulate the microtubule cytoskeleton, cell cycle, or ploidy in different plant models and different contexts, and more specifically to cell biologists studying the molecular mechanisms of cell division.

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

REVIEWER #1

__Summary: __Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate).

In virtue of the classical cancer stem cells (CSC) marker ALDH1A1 and SMAD Response Element (SRE) promoter, the authors engineered CSC-derived extracellular vesicles (EVs). By performing the second sortase (SrtA)-based proximity labeling, the authors detected the immune cells that specifically interacted with the CSC-EVs and demonstrated that CSC-EVs preferentially target MHC-II- macrophages and PD-1+ T cells.

---

Major comments: - Are the key conclusions convincing? No. CD63 is accepted as the exosome marker, but cannot represent the whole population of EVs. Especially, we do have the information on the percentage of CD63+ EVs among the total population derived by CSCs. However, it seems impossible to estimate the total population derived by CSCs. It is the inherent flaw of the strategy, which limits the accuracy of the labeling. One possible method is to label CD81+ and CD9+ EVs, together with CD63+ EVs, to study the immune cells interacting with CSC-EVs in vitro and in vivo.

We would like to thank this Reviewer for the time spent reading our manuscript and for highlighting the fact that other EVs markers could have been used to track EVs. We would like to point out that the Sortase-A experimental strategy is independent from any assumptions on EV markers since SrtA is fused to a commonly-used transmembrane domain (from PDGFR, see Hamilton et al. Adv Biosys 2020). Nonetheless, we followed up on the Reviewer suggestion and performed additional experiments to assess the overlap between this generic membrane marker, CD63 and CD81. To this end, we have performed multicolor nano-flow cytometry and stained EVs for SrtA (via its flag peptide) and CD63 or CD81 (new suppl. Fig. S2B-C). We used Flag staining to detect the PDGFR transmembrane domain as a generic membrane marker (__new suppl. Fig. S1C __and ref. 50). We observed that Flag staining colocalized with both CD63+ and CD81+ EVs, indicating not only that the use of a general-purpose transmembrane domain transcends classical EV biomarkers, but also that CD63 and CD81 label largely overlapping EV subpopulations, as previously reported (Jeppesen DK et al. Cell 2019). Accordingly, SrtA- and CD63-GFP-based strategies yielded very similar results (Fig. 2 and 4).

Compared with the normal cancer cells, cancer stem cells are a very small population. It is reasonable to consider that the CSC-EVs is also a small population among total EVs. Therefore, it is quite questionable to compare the interaction of normal cancer cells-derived EVs and CSC-EVs with immune cells.

We fully agree with this reasoning, and it is exactly because of this contrast that our observations of the specific behavior of CSC-EVs are very relevant for CSC biology. Our experimental design includes proper control groups and is based on validated approaches (Hamilton et al. Adv Biosys 2020).

• Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether? Yes. The authors stated "such EV-mediated intercellular communication between CSC and these immune cells contributed to the observed spatial interactions and niche sharing." Not enough evidence supported the statement.

We have removed these claims from the text.

• Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation. As mentioned before, if the authors could perform the labeling CD81+ and CD9+ CSC-EVs and study the interaction with immune cells, the conclusion may be more convincing.

• Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments. The suggested experiments are time-consuming.

We appreciate the Reviewer acknowledging that repeating the CD63-GFP experiments in the manuscript using CD81-GFP and CD9-GFP fusion reporters is very time-consuming. Nonetheless, we have provided proof that the SrtA approach labels both CD63+ and CD81+ EVs (new suppl. Fig. S2B-C), which, together with the fact that both CD63-GFP- and SrtA-based approaches yielded very similar results (Fig. 2 and 4), strongly indicates that repeating these experiments with additional reporters will add limited value to already sound conclusions.

• Are the data and the methods presented in such a way that they can be reproduced? Yes.

• Are the experiments adequately replicated and statistical analysis adequate? Yes.

Minor comments: - Specific experimental issues that are easily addressable. Yes

• Are prior studies referenced appropriately? The references related SrtA-mediated labeling were not sufficiently referenced.

The full characterization of SrtA-based strategy was cited in reference 50 (Hamilton et al. Adv Biosys 2020). We would be happy to include any reference this Reviewer thinks is missing.

• Are the text and figures clear and accurate? Yes

• Do you have suggestions that would help the authors improve the presentation of their data and conclusions? No

Significance: - Describe the nature and significance of the advance (e.g., conceptual, technical, clinical) for the field.

Cancer cells are heterogeneous. It is natural to believe that EVs are heterogeneous due to their different origin. Considering the important role of cancer stem cells during tumor development and treatment resistance acquisition, it is important to understand the function of CSC-EVs in the tumor microenvironment. However, considering the methodology is questionable, I am not sure the conclusions are convincing. For Figure 3, there are many pieces of literature on this topic and showing the data that macrophages in CSCs niches are good for the maintenance of CSC. So, it is not novel.

We thank the Reviewer to point out the importance of understanding the function of CSC-EVs in the tumor microenvironment. We hope we have addressed the methodology issues raised by this Reviewer. Although recent students outline the relationship between CSC and macrophages biology, very little is known about the role of EVs in this interaction. The novelty of our work stems from the use of advanced genetic engineering approaches that allow us to demonstrate directly in vivo, without any in vitro manipulation of CSC-EVs, that CSC-EVs come in contact with macrophages (and other specific immune subsets).

• Place the work in the context of the existing literature (provide references, where appropriate).

• State what audience might be interested in and influenced by the reported findings. Cancer stem cells or extracellular vesicles are timely topics and would be interesting to people in the cancer and EV fields.

• Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate., EVs biology, with no special focus on CSCs.

REVIEWER #2

__Summary: __In this work Dr. Pucci and colleagues use flowcytometry and in vivo approaches to define the potential role of EVs originating from cancer stem cells to mediate intercellular communication with cells of immune origin in the cancer microenvironment. The work is interesting, the team used specific promoters to drive the expression of EV markers specifically in cancer stem cells. Another interesting approach is the use of sortase to label neighbour cells in the cancer microenvironment.

Overall the reviewer has the impression the work is quite superficial and the conclusions cannot be claimed by the results presented in the paper for the following reasons:

1. Cancer stem cells are a relatively small fraction when compared to the entire cancer cell population, therefore it is possible that the EV released tend to accumulate in macrophages because those are cells competent for specialized internalization and clearance of EV (e.g. PMID: 30745143). Second accumulation in macrophages does not mean any kind of signaling, it may just be that the EVs are degraded.

We thank the Reviewer for understanding and appreciating our work. We would like to point out that the novelty of our work is the identification of a specific macrophage subset (MHC-II– macrophages) that is mainly targeted by CSC-EVs (Fig.2). We observed a selective enrichment of these macrophage subset when tracking CSC-EVs, which argues against passive uptake, as the Reviewer seems to suggest. Moreover, these class-II negative macrophages were also found at increased frequency within the CSC niche (Fig.4), further suggesting an active process. This information is not trivial and cannot be inferred from the literature.

We are not claiming any signaling mechanism, which will be the focus of future work, including the role of CSC-EVs on maintaining MHC-II– immunosuppressive macrophage populations. We have amended the main text to clarify these points.

The sortase experiment is very interesting, however key controls are missing. For example, a thorough in vitro characterization of the system is needed: a. No clear description of the vectors used is provided (how is the labelling fluorescent protein released by the cells? How far can the protein diffuse?); b. The sortase's labelling efficiency is not characterized. c. Which proteins are targeted by sortase in the acceptor cells? There is any protein that can specifically be labelled by sortase on the cell surface of acceptor cells? This is not explained or validated. d. The authors claim that the labelling is provided by EVs harboring sortase on their surface, however also the plasma membrane of the cells may efficiently label cells. This should be explored and discussed. Which is the enzymatic sortase activity present on the EVs? How the authors can exclude that the red fluorescent protein is simply internalized by the neighbor cells? This should also be evaluated.

We thank the Reviewer for the interest in the SrtA-based approach, which we have thoroughly described and validated in Hamilton et al. Adv Biosys 2020 (ref. 50). We apologize if we did not reference that work properly. In that publication, we characterized SrtA labelling efficiency under various conditions and with different substrates, we mentioned which proteins can be targeted by SrtA on the surface of cells (that is, any protein with an N-terminal glycine, such as MHC-I, VE-Cadherin, CD19, integrins, …). We have clarified the above details in the new version of the manuscript.

We apologize for not including a schematic of the lentiviral vectors used, that we have now added (new suppl. Fig. S1). These schematics show that the red fluorescent protein is released from cells because it is fused to a signal sequence. In order to control for internalization of the red fluorescent protein by neighboring cells, we have used a control group in which CSC do not express SrtA while the bulk of tumor cells (including CSC) still secrete the red fluorescent protein. The values for SrtA activity are calculated by subtracting baseline internalization of red fluorescent protein by each individual immune subset. We have amended the Methods section to clarify this.

We are glad to hear that the Reviewer fully understood how the SrtA-based approach works. As this Reviewer mentions, it is not possible to discriminate between CSC and CSC-EVs since SrtA is present on both. This is a limitation of current EV technology in general. Although we were careful in wording our results and conclusions, we have revised the manuscript to take this into further consideration. The manuscript now claims that the SrtA approach unveils short-range interactions between CSC, CSC-EVs and immune cells due to their proximity.

Method section should be expanded, map of vectors provided and possibly deposited.

We apologize for not including a schematic of the lentiviral vectors used, that we have now added (new suppl. Fig. 1). We have expanded description of the Methods. We will promptly deposit the lentiviral transfer plasmids employed in this work with Addgene upon publication.

__Minors: __The paper should be expanded, and experiments better described.

We have expanded the paper and the description of the experiments, as requested.

CROSS-CONSULTATION COMMENTS I find the comments of Reviewer 1 important to be addressed. I might have under-estimated the amount of time necessary to revise the work. I still believe the material and method section is insufficient.

We thank the Reviewer for acknowledging that major revision of our work is very time-consuming. We hope to have addressed both Reviewer 1 and 2 comments.

Significance: The paper present technological innovation that can result of interest for the large audience of EV enthusiasts. The scientific advancement is limited since the conclusion: These results suggest that combination therapies targeting CSC, tumor macrophages and PD1 may synergize, is known and the work presented does not really support it since there is no evidence the EVs have any signaling role. Perhaps the authors should work more on tightening their result to a cell biology perspective of cancer niche interaction.

We thank the Reviewer for understanding the technological innovation. We will remove the statement on combination therapy and refocus on a cell biology perspective.

REVIEWER #3

In this paper, the authors used genetically engineered CSC-derived EVs to perform sortase-mediated in vivo proximity labeling and interrogate interactions of these vesicles with immune cells in the TME. The authors show that these EVs mediate intra-tumoral recruitment of immune cells, MHC class II(-) macrophages and PD1+ T cells, to the CSC niche and define EV-mediated special interactions of these immune cells within this niche. The manuscript is timely and novel, as it introduces a new experimental platform for identification, characterization and monitoring of CSC-derived EVs within the TME. Much has been recently learned about tumor cell-derived "tEVs", while almost nothing is known about CSC-derived tEVsCSC. Here, using genetic engineering, the authors have created specifically labeled fluorescent (GFP) tEVsCSC and studied interactions of these vesicles with immune cells in the TME. Two different HNSCC mouse models, MOC2 (carcinogenesis-dependent) and mEER (Ras dependent), were used. CSC populations were identified as cells with the brightest GFP fluorescence (~5%). These cells also expressed the known stem cell markers and formed oospheres in vitro. The authors then show that tEVcsc preferentially targeted MHC II (-) macrophages, which avidly uptake these EVs. tEVcsc also showed preferential tropism towards PD1+ T cells. Further, the authors demonstrate that location-dependent labeling indicates the presence and "clustering" of MHC II (-) macrophages and PD1+ T cells in the same niche within the TME. The generation of genetically modified labeled fluorescent tEVcsc and tEVs and in vitro as well as in vivo analyses of their interactions with immune cells in the TME were technically demanding. These studies were expertly performed, and the results are convincing. The data presentation is adequate, but the figure legends are sparce, and the text is densely narrated and somewhat difficult to read. Some more clarity in Results and more explicitly documented correlative data would clarify and enhance the message the authors convey.

We thank the Reviewer for fully understanding the novelty of our work. We agree that the description of results and figures can be improved. We have now clarified the narration of results and figure legends so they can be better understood.

The Discussion is rationally written, but the comments in Abstract and in conclusions about combination therapies and targeting CSC, tumor macrophages and PD1 to lower HNSCC recurrence are not appropriate. There is nothing in this manuscript about immunotherapy and these comments should be deleted.

We agree with the Reviewer and we have removed these statements from the text.

Overall, this is an interesting, timely and novel manuscript using genetically modified, fluorescently labeled EVs to explore their interactions with immune cells in the TME of HNSS. Technical and experimental approaches are complex, but appear to be well done, providing an experimental model for probing cellular interactions in the TME at a single-cell level. I recommend acceptance after modifications as suggested above

Significance: Significance is high, as it advances our understanding of the interactive role of CSC-derived EVs with macrophages and T cells in the tumor microenvironment.

We thank the Reviewer for appreciating the significance of our 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 #3

Evidence, reproducibility and clarity

In this paper, the authors used genetically engineered CSC-derived EVs to perform sortase-mediated in vivo proximity labeling and interrogate interactions of these vesicles with immune cells in the TME. The authors show that these EVs mediate intra-tumoral recruitment of immune cells, MHC class II(-) macrophages and PD1+ T cells, to the CSC niche and define EV-mediated special interactions of these immune cells within this niche.

The manuscript is timely and novel, as it introduces a new experimental platform for identification, characterization and monitoring of CSC-derived EVs within the TME. Much has been recently learned about tumor cell-derived "tEVs", while almost nothing is known about CSC-derived tEVsCSC. Here, using genetic engineering, the authors have created specifically labeled fluorescent (GFP) tEVsCSC and studied interactions of these vesicles with immune cells in the TME. Two different HNSCC mouse models, MOC2 (carcinogenesis-dependent) and mEER (Ras dependent), were used. CSC populations were identified as cells with the brightest GFP fluorescence (~5%). These cells also expressed the known stem cell markers and formed oospheres in vitro. The authors then show that tEVcsc preferentially targeted MHC II (-) macrophages, which avidly uptake these EVs. tEVcsc also showed preferential tropism towards PD1+ T cells. Further, the authors demonstrate that location-dependent labeling indicates the presence and "clustering" of MHC II (-) macrophages and PD1+ T cells in the same niche within the TME.

The generation of genetically modified labeled fluorescent tEVcsc and tEVs and in vitro as well as in vivo analyses of their interactions with immune cells in the TME were technically demanding. These studies were expertly performed, and the results are convincing. The data presentation is adequate, but the figure legends are sparce, and the text is densely narrated and somewhat difficult to read. Some more clarity in Results and more explicitly documented correlative data would clarify and enhance the message the authors convey. The Discussion is rationally written, but the comments in Abstract and in conclusions about combination therapies and targeting CSC, tumor macrophages and PD1 to lower HNSCC recurrence are not appropriate. There is nothing in this manuscript about immunotherapy and these comments should be deleted.

Overall, This is an interesting, timely and novel manuscript using genetically modified, fluorescently labeled EVs to explore their interactions with immune cells in the TME of HNSS. Technical and experimental approaches are complex, but appear to be well done, providing an experimental model for probing cellular interactions in the TME at a single-cell level. I recommend acceptance after modifications as suggested above

Significance

Significance is high, as it advances our understanding of the interactive role of CSC-derived EVs with macrophages and T cells in the tumor microenvironment.

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 work Dr. Pucci and colleagues use flowcytometry and in vivo approaches to define the potential role of EVs originating from cancer stem cells to mediate intercellular communication with cells of immune origin in the cancer microenvironment. The work is interesting, the team used specific promoters to drive the expression of EV markers specifically in cancer stem cells. Another interesting approach is the use of sortase to label neighbour cells in the cancer microenvironment.

Overall the reviewer has the impression the work is quite superficial and the conclusions cannot be claimed by the results presented in the paper for the following reasons: 1. Cancer stem cells are a relatively small fraction when compared to the entire cancer cell population, therefore it is possible that the EV released tend to accumulate in macrophages because those are cells competent for specialized internalization and clearance of EV (e.g. PMID: 30745143). Second accumulation in macrophages does not mean any kind of signaling, it may just be that the EVs are degraded.

1. The sortase experiment is very interesting, however key controls are missing. For example, a thorough in vitro characterization of the system is needed:
   • a. No clear description of the vectors used is provided (how is the labelling fluorescent protein released by the cells? How far can the protein diffuse?);
   • b. The sortase's labelling efficiency is not characterized.
   • c. Which proteins are targetd by sortase in the acceptor cells? There is any protein that can specifically be labelled by sortase on the cell surface of acceptor cells? This is not explained or validated.
   • d. The authors claim that the labelling is provided by EVs harboring sortase on their surface, however also the plasma membrane of the cells may efficiently label cells. This should be explored and discussed. Which is the enzymatic sortase activity present on the EVs? How the authors can exclude that the red fluorescent protein is simply internalized by the neighbour cells? This should also be evaluated.
2. Method section should be expanded, map of vectors provided and possibly deposited.

Minors:

The paper should be expanded, and experiments better described.

Referees cross-commenting

I find the comments of Reviewer 1 important to be addressed. I might have under-estimated the amount of time necessary to revise the work.

I still believe the material and method section is insufficient.

Significance

The paper present technological innovation that can result of interest for the large audience of EV enthusiasts. The scientific advancement is limited since the conclusion: These results suggest that combination therapies targeting CSC, tumor macrophages and PD1 may synergize, is known and the work presented does not really support it since there is no evidence the EVs have any signaling role. Perhaps the authors should work more on tightening their result to a cell biology perspective of cancer niche interaction.

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:

Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate).

In virtue of the classical cancer stem cells (CSC) marker ALDH1A1 and SMAD Response Element (SRE) promoter, the authors engineered CSC-derived extracellular vesicles (EVs). By performing the second sortase (SrtA)-based proximity labeling, the authors detected the immune cells that specifically interacted with the CSC-EVs and demonstrated that CSC-EVs preferentially target MHC-II- macrophages and PD-1+ T cells.

Major comments:

• Are the key conclusions convincing?

No.

CD63 is accepted as the exosome marker, but cannot represent the whole population of EVs. Especially, we do have the information on the percentage of CD63+ EVs among the total population derived by CSCs. However, it seems impossible to estimate the total population derived by CSCs. It is the inherent flaw of the strategy, which limits the accuracy of the labeling. One possible method is to label CD81+ and CD9+ EVs, together with CD63+ EVs, to study the immune cells interacting with CSC-EVs in vitro and in vivo. Compared with the normal cancer cells, cancer stem cells are a very small population. It is reasonable to consider that the CSC-EVs is also a small population among total EVs. Therefore, it is quite questionable to compare the interaction of normal cancer cells-derived EVs and CSC-EVs with immune cells. - Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

Yes. The authors stated "such EV-mediated intercellular communication between CSC and these immune cells contributed to the observed spatial interactions and niche sharing." Not enough evidence supported the statement. - Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

As mentioned before, if the authors could perform the labeling CD81+ and CD9+ CSC-EVs and study the interaction with immune cells, the conclusion may be more convincing. - Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

The suggested experiments are time-consuming. - Are the data and the methods presented in such a way that they can be reproduced?

Yes. - Are the experiments adequately replicated and statistical analysis adequate?

Yes.

Minor comments:

• Specific experimental issues that are easily addressable.

Yes - Are prior studies referenced appropriately?

The references related SrtA-mediated labeling were not sufficiently referenced. - Are the text and figures clear and accurate?

Yes - Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

No

Significance

• Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field.

Cancer cells are heterogeneous. It is natural to believe that EVs are heterogeneous due to their different origin. Considering the important role of cancer stem cells during tumor development and treatment resistance acquisition, it is important to understand the function of CSC-EVs in the tumor microenvironment. However, considering the methodology is questionable, I am not sure the conclusions are convincing.

For Figure 3, there are many pieces of literature on this topic and showing the data that macrophages in CSCs niches are good for the maintenance of CSC. So, it is not novel.

• Place the work in the context of the existing literature (provide references, where appropriate).
• State what audience might be interested in and influenced by the reported findings.

Cancer stem cells or extracellular vesicles are timely topics and would be interesting to people in the cancer and EV fields. - Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate.,

EVs biology, with no special focus on CSCs.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Answers to reviewers’ comments

(Reviewers comments are in italics. Text modifications in the manuscript file are in blue.)

Overall, we acknowledge referee’s careful reading of the paper and comments that we think have helped further improvement of the manuscript.

On the attached pages are our detailed point by point responses to the referees’ comments along with a description of how the manuscript was modified in accordance.

New data included:

In response to the comments and suggestions of both reviewers 1 and 3, we conducted new experiments to test genetic interactions between different actors of the BMP and activin pathways. These new results confirm and complement the analyses described in the original manuscript. Furthermore, as suggested by reviewer 2, we have further studied the phenotypes of hiPSC-CM, by analyzing gene expression profiles and by analyzing the morphological changes induced as a result of PAX9 knockdown.

NB: The title has been slightly modified, to highlight the conserved features of the genetic architecture of cardiac performance revealed in the study

__Former title: __Genetic architecture of natural variation of cardiac performance in flies.

__Novel title: __Genetic architecture of natural variation of cardiac performance: From flies to humans.

Reviewer 1

1. 1. The authors utilized the RNAi-mediated knockdown approach in their functional validation studies. It is not clear how each genetic variation (SNP) affects its associated genes. Could some of the SNPs activate the candidate gene expression? For the 4 candidate genes that failed to show cardiac defects, could the overexpression of these 4 genes alter cardiac performance? Answer 1- Of course, we cannot predict direction of the effect of the variants on the function of the genes. In this context, loss-of-function experiments are subjected to a risk of false negatives. It is indeed possible that in the case of a lack of effect of the loss of function, a gain of function could reveal an effect. But gain-of-function experiments are difficult to control, and often subjected to non-specific effects because it is complicated to control the level of over-expression compared to endogenous expression. This did not seem suitable for an extensive analysis of a large number of genes. We therefore chose to test only for loss of function.

In addition, our approach to testing heart-specific RNAi aims to assess the quality of the association results by comparing RNAi for genes identified by GWAS to randomly selected genes. It is not intended to describe precisely the involvement of each gene individually.

(See also answer to reviewer 2 comment n°2 and the modifications to the manuscript that have been made and which address these criticism)

* 2. babo is the type I activin receptor, not type 2. *

Answer 2- Thank you, we have corrected this error.

• The authors show BMP and activin pathway genetically interacts to affect cardiac performance. But it is interesting to find that these interactions are in a trait-dependent manner. For example, it seems that babo and dpp epistatically interact to regulate FS, while they additively regulate HP and DI. The authors need to discuss the complex genetic interaction further. *

Answer 3- See reply to reviewer 3, comment N°2 below.

4*. Both snoo and sog are identified from GWAS. How about babo and dpp? Are there any identified SNPs associated with babo and dpp? *

Answer 4- Considering GWAS for mean phenotypes, there is no variant in dpp that are within the 100 best ranked SNPs nor within the variants identified using fast epistasis. But given the size of the DGRP population we are far from being exhaustive, as we do not reach saturation. It is therefore difficult to comment on these ‘negative’ results. However, we do identify one variant in babo using fast epistasis (see figure 2B and Table S3).

5. It is unclear why the mad KD behaves oppositely to dpp mutant, although both proteins are involved in BMP pathway. In Figure S5, the mad KD shows reduced FS and HP, but dpp LOF mutant shows increased FS and HP (Figure S4). Can the authors perform RNAi to knockdown dpp specific in the heart to reexamine the role of dpp in the regulation of cardiac function. The whole body LOF mutant dpp-d14 might not target cardiac tissue directly to control heart performance like mad KD.

Answer 5- (see also answer to reviewer 3 comment n°2) We did perform heart specific dpp RNAi experiments together with other tests for interactions using new allelic combinations of activin and BMP pathways and therefore can compare heart specific knock down to heterozygotes for amorphic mutations for both dpp and mad.

Regarding dpp, congruent effects on HP, DI, SI, ESD and EDD were observed between mutant and RNAi, while RNAi had opposite effects on FS compared to heterozygotes dppd14 mutants (decreased and increased FS compared to control, respectively). In the case of mad, heterozygous mutants had no effect on FS, EDD and ESD, but similarly to dpp mutants it increased SI, DI and HP. mad RNAi uniquely decreased HP, DI and SI and increased AI. However, similarly to dpp RNAi, it induced a decrease of FS.

Thus, systemic versus heart specific knockdown of genes induce specific effects, suggesting cardiac non-autonomous interactions. This complex picture of TGFb involvement is now discussed in the result section (see below, Reviewer 3, major comment 2).

6*. The authors selected two novel genes to study the conversed regulation in both flies and human iPSC cells. Besides testing these novel genes, the authors should also verify whether the conserved pathways, like TGF-beta, regulate heart performance in human iPSC cells similar to the flies. *

Answer 6- We focused on poxm/Pax9 and sr/Egr2 because none of these TFs were known to have cardiac function in fly nor in mammals. Our paralleled analyses in fly and hiPS-CM illustrates how the description of the genetic architecture of cardiac traits in flies can accelerate discovery in mammals.

There is extensive literature describing the involvement of TGF B /BMP and Activin pathways in heart development and diseases in humans, hence the choice not to focus on these pathways in iPS-CM.

Reviewer 2:

1. • It will be interesting to compare this fly GWAS to human heart disease GWAS data (for example, cardiomyopathy, arrhythmia, heart failure) from patients. Such cross comparison could make the data set more valuable. * Answer 1- We actually did make this comparison (Table 2, Table S11) and we agree it significantly validates our approach. This identified a set of orthologous genes associated with cardiac traits both in Drosophila and humans, supporting the conservation of the genetic architecture of cardiac performance traits, from arthropods to mammals.

2. RNAi is the only experimental approach in this manuscript to validate the functional significance from data analyses. Authors may consider using genetic mutations such as deficiency lines or P-element lines to offer an alternative approach. This is simply a suggestion to improve the rigor and reproducibility, not absolutely required. *

Answer 2- In an attempt to provide a consistent analysis of loss of gene function, our strategy was to concentrate our analysis on the effects of heart specific knock down. This allows us to compare -in a global way- the effects of the knock down of genes identified by GWAS to those of randomly selected genes.

Our objective was to provide a global view of the heart specific effects of the identified genes, and not to characterize precisely the involvement of each of them, using a combination of mutant alleles, RNAi and gain of function. Given the experimental burden of analyzing cardiac function, such a strategy would have indeed required us to concentrate only a very small number of genes.

We however recognize that this strategy has limitations:

• Some variants may lead to gain-of-function effects of genes, and our strategy is not able to test for these effects.

• Some variants may come from non-cell-autonomous effects, which would not be replicated by our targeted RNAi strategy in the heart.

Therefore, the false negative rate of our experiments is difficult to estimate.

We have tried to put this into perspective and to highlight the limitations of our analysis in the results section describing RNAi validation of GWAS results.

“To assess in an extensive way whether mutations in genes harboring SNPs associated with variation in cardiac traits contributed to these phenotypes ….. (…)

…… These results therefore supported our association results. It is important to emphasize that our approach is limited to testing the effect of tissue-specific gene knock down. Since some of the variants may lead to increased gene function and/or expression, this can lead to a false negative rate that is difficult to estimate. In addition, some of the associated variants may influence heart function by non cell-autonomous mechanisms, which would not be replicated by cardiac specific RNAi knock down.”

*In order to validate the roles of predicted TF binding sites, the best approach would be introducing point mutations using CRISPR/Cas9 within the binding motif then testing out molecular and physiological outcomes. Rather authors chose to test indirectly to knock down those TFs. If so, authors need to at least acknowledge the potential caveats of such approach and the limitation in related data interpretation. *

Answer 3- The reviewer is right, the definitive proof of the involvement of a potential TF binding site on the regulation of a gene located in cis requires to mutate the binding site and to analyze the effect on the expression of the corresponding gene. But this may not be sufficient to definitely demonstrate that the potential TF is indeed a regulator of that gene (the binding motif may be target of yet another TF): definitive proof may require motifs/TF DNA binding domain swaps. This would have been out of the scope of the present study. In addition, the effects on heart performance of mutating one TFBS at a time (among several dozens) may be too weak to allow their characterization with available tools and approaches.

We acknowledge however that our approach provides an indirect validation of transcription factors binding sites predictions. This was, in our opinion, the most efficient way to evaluate the potential effect of predicted transcription factors.

We clarify this in the result section:

“We did not test individually the effects on cardiac performance of mutations in predicted TFBSs located near the SNPs because any individual effect would probably be too small to be detectable by the available methods. Rather, we tested the potential involvement of their cognate TFs by cardiac specific RNAi mediated KD”

• hiPSC-CM data is somewhat limited by only showing the HR and AP duration data. It is recommended to include some immunocytochemistry data to show the morphology, sarcomere structure of these hiPSC-CMs. Gene expression data generated by qPCR or RNA-seq in particular focusing CM structure and function genes would be helpful too.*

Answer 4- As suggested by referee 2, we have now performed gene expression analysis and immunostaining of PAX9 KD which gave the strongest phenotype in iPSC-CM (Figure 4 J-M). This unraveled increased expression of Na+ and K+ channels, which is in line with APD shortening phenotype, as well as down regulation of CASQ2, consistent with calcium transient shortening. Expression analysis also revealed increased sarcomeric genes and NPPA/B expression, which was consistent with increased CM size as quantified by the area of TNNT2 staining per nuclei.

These new data are described at the end of the result section:

“APD shortening for PAX9 KD was coincident with increased expression of Na+ and K+ ion channels (SCN5A, KCNH2 and KNCQ1) (Figure 4J), supporting the APD shortening phenotype. In this context, the AP kinetics also correlated with shorter calcium transient duration (Figure S8A-D and H-K), including faster upstroke and downstroke calcium kinetics and increased beat rate (peak frequency) (Figure S8E-G and L, M), consistent with decreased expression of Calsequestrin 2 isoform (CASQ2) associated with PAX9 KD (Figure 4J). Finally, assessment of the PAX9 KD effect on sarcomeric content revealed an increase in sarcomeric gene expression (Figure 4K), and an upregulation of genes associated with an hypertrophic response (NPPA, NPPB and NPR1 (Battistoni Et al Circulating biomarkers with preventive, diagnostic and prognostic implications in cardiovascular diseases, Int J Cardiol, 2012, vol. 157) which was coincident with increased CM size as quantified by the area of TNNT2 staining per cardiac nuclei (Figure 4 L, M).

Collectively, these data illustrate conserved functions for poxm/PAX9 and sr/EGR2 in setting the cardiac rhythm and identify PAX9 as a novel and key regulator of cardiac performance at the cellular level, via the integrated regulation of expression of genes controlling electrophysiology, calcium handling and sarcomeric functions in hiPSC-CMs.”

Reviewer 3

Major Comments:

1- There is an assumption in the use of RNAi knockdown to validate the genes identified in the quantitative analysis, and that is that natural variants are themselves hypomorphic. It is possible that among the variants identified some are hypermorphic, or among the transcription factor binding sites that variants lead to increased factor binding. While RNAi knockdown is an excellent choice to begin validation, I do not think the authors can rule out that a gene not functionally validated by their RNAi tests does not have a role in cardiac function.

Answer 1. Please see our answers to reviewer 1 comment n°1 and reviewer 2 comment n°2.

* 2- After performing RNAi knockdown to validate genes identified by GWAS the authors focus on the TGFbeta signaling pathway for downstream analysis. To do so they examine heterozygotes for sog, a repressor of BMP signaling, and snoo, an activator of Activin pathway. The data from the snoo/sog heterozygote is compelling in its disruption of heart phenotypes, and the authors conclude a "coordinated action of activin and BMP." snoo, however, also works as a transcriptional repressor in the BMP pathway, so it's possible that the effects the authors are seeing here could be confined to an increase in BMP signaling. Unlike snoo and sog, mutations in babo and dpp are both expected to have negative effects on Activin and BMP signaling, respectively. The babo/dpp interaction is not as quantitatively convincing as the snoo/sog data, despite the integral roles both babo and dpp play in their respective pathways. If both pathways are connected, why do snoo/sog heterozygotes affect SI phenotypes, while babo/dpp heterozygotes affect fractional shortening? I think the authors data suggest an interesting potential interaction between these pathways, which could be confirmed by examining further mutant combinations, knockdowns or increased expression transgenes, but falls short of a "confirmed synergystic genetic interaction." It does, however, underscore the value of the data in the paper for opening up new avenues for future study. *

Answer 2 (and reviewer 1 comments 3 and 5).

These comments led us to reconsider the analysis of the phenotypes associated with loss of function of the TGFb pathway, and to analyze other pathway components combinations.

We acknowledge reviewer 3 criticisms on snoo/sog experiments, which are difficult to interpret given the broad action snoo may have on both BMP and activin pathways. We have addressed this in the result section.

We have also analyzed other allelic combinations of BMP and activin pathways components, which strengthen the analysis performed on dpp/babo. Indeed, we tested babo/tkv heterozygotes (respectively specific activin and BMP receptors) and found significant genetic interactions for ESD and EDD. Albeit non-significant, babo/tkv double heterozygotes display a tendency to non-additive effects on FS (p= 0,054). mad/smox heterozygotes (respectively specific downstream TFs of BMP and activin pathways) display interactions (non-additive effects) on HP, SI, DI, ESD and EDD. These new results (Supplemental Figure 4) are thus supporting the hypothesis of genetic interactions between the pathways, but also reveal, as suggested by reviewer 3, a complex relationship between both pathways since interactions are revealed for specific traits in each of the mutant combinations analyzed.

The phenotypes related to the individual loss of function of each of the actors of these pathways (dpp, tkv and mad for BMP; babo and smox for activin) are however very similar. When they have an effect, heterozygous amorphic alleles of these genes display increased phenotypes related to rhythmicity (HP, DI, SI, AI) and FS, but decreased cardiac diameters (ESD and EDD).

Finally, as pointed out by reviewer 1, the picture is certainly even more complex since the phenotypes of RNAi mediated heart specific loss of function are not always similar to those of systemic loss of function. Indeed, mad RNAi causes a reduction of HP, DI, SI and FS (Figure S5) whereas heterozygotes for mad12 have either no or opposite effect on these phenotypes, and mad RNAi causes a significative increase in AI whereas mad12 has no effect (Figure S4). The discrepancy between tissue specific RNAi and heterozygous background was also found in the case of dpp, but specifically for the FS. Indeed, as suggested by reviewer 1 we have analyzed the loss of function of dpp by heart-specific RNAi. dpp RNAi results in a reduction of the FS (like mad RNAi) whereas the loss of function in the whole-body results in an increase of the FS.

We therefore re-wrote the whole corresponding section of the results and modified Figure S4 to include babo/tkv; smox/mad and dppRNAi data.

“We further focused on the TGFb pathway, since members of both BMP and activin pathways were identified in our analyses. We tested different members of the TGFb pathway for cardiac phenotypes using cardiac specific RNAi knockdown (Figure 2C), and confirmed the involvement of the activin agonist snoo (Ski orthologue) and the BMP antagonist sog (chordin orthologue). Notably, Activin and BMP pathways are usually antagonistic (Figure 2D). Their joint identification in our GWAS suggest that they act in a coordinated fashion to regulate heart function. Alternatively, it may simply reflect their involvement in different aspects of cardiac development and/or functional maturation. In order to discriminate between these two hypotheses, we tested if different components of these pathways interacted genetically. Single heterozygotes for loss of function alleles show dosage-dependent effects of snoo and sog on several phenotypes, providing an independent confirmation of their involvement in several cardiac traits (Figure S4). Importantly, compared to each single heterozygotes, snooBSC234/ sogU2 double heterozygotes flies showed non additive SI phenotypes (two-way ANOVA p val: 2,1 10-7) suggesting a genetic interaction (Figure 2E and Figure S4A). It is worth noting however that snoo is also a transcriptional repressor of the BMP pathway (PMID: 16951053). The effect observed in snooBSC234/ sogU2 double heterozygotes can therefore alternatively arise as a consequence of an increased BMP signaling without affecting the activin pathway. We thus tested other allelic combinations for loss of function alleles of BMP and activin pathways. babo/tkv heterozygotes (respectively activin and BMP type 1 receptors) displayed non additive ESD and EDD phenotypes (Figure S4C). Synergistic interaction of BMP and activin pathways was also suggested by the analysis of fractional shortening in loss of function mutants for babo and dpp, the BMP ligand (Figure S4B). Of note, babo/tkv double heterozygotes also displayed a tendency to non-additive effects on FS albeit non-significant (two-way anova p= 0,054). In addition, mad/smox heterozygotes (specifc downstream TFs of BMP and activin pathways) displayed non-additive effects on several traits, including phenotypes related to rhythmicity (HP, SI, DI) and contractility (ESD and EDD) (Figure S4D). Altogether, cardiac performance in response to allelic combinations of activin and BMP supported a coordinated action of both pathways in the establishment and/or maintenance of cardiac activity. This was further supported by the observation that simple heterozygotes for the tested loss of function alleles displayed similar trends with respect to cardiac performance, irrespective of the pathway considered (dpp, tkv and mad for BMP; babo and smox for activin). Indeed, they displayed either no effect or increased fractional shortening and rhythmicity phenotypes (HP, DI, SI, AI), and decreased cardiac diameters (ESD and EDD). This suggests coordinated activity of both pathways. Importantly, the genetic interactions were tested using amorphic alleles that lead to systemic loss of function. The observed phenotypes may thus not unravel cardiac specific effects of the pathways. In support of this, mad cardiac specific RNAi knock down was tested (see below, Figure S5) and lead to a decreased HP, DI, SI and FS whereas heterozygotes for mad12 have either no (FS) or opposite (HP, DI, SI) effect on these phenotypes (Figure S4D). Inversely, mad RNAi caused a significant increase in AI whereas mad12 had no effect. However, heart specific dpp RNAi knock down (Figure S4E) lead to similar phenotypic trends compared to dppd14 (increased HP, DI, SI, decreased EDD and ESD) with the notable exception of FS which was reduced following cardiac specific KD (Figure S4E), but increased in dppd14heterozygotes (Figure S4B). Taken together, these data point to a complex picture of TGFb pathway activity in regulating cardiac performance, involving both the activin and the BMP pathways as well as gene specific effects with both systemic and tissue-specific contributions.”

*Minor Comments: *

* There is an enormous amount of data in this paper, but there are places where things are summarized a little too briefly. For example, there are no definitions given at the beginning of the Results section for traits like "Heart Period" or "Systolic Interval," which would make this work significantly more accessible for other Drosophila researchers. (They do touch on this when they explain later in the paper that certain variants are "associated with quantitative traits linked to heart size and contractility" but more background earlier would be helpful.) When we consider heart performance traits, what is the baseline from known mutants? In other words, where is the line between variation and defect? *

Answers:

• We have detailed the description of the traits analyzed at the beginning of the result section. We hope this improves the ease of reading in the direction suggested by the reviewer. “7 cardiac traits were analyzed across the whole population (Dataset S1 and Table 1). As illustrated in Figure 1A, we analyzed phenotypes related to the rhythmicity of cardiac function: the systolic interval (SI) is the time elapsed between the beginning and the end of one contraction, the diastolic interval (DI) is the time elapsed between two contractions and the heart period (HP) is the duration of a total cycle (contraction + relaxation (DI+SI)). The arrhythmia index (AI, std-dev(HP)/mean (HP)) is used to evaluate the variability of the cardiac rhythm. In addition, 3 traits related to contractility were measured. The diameters of the heart in diastole (End Diastolic Diameter, EDD), in systole (End Systolic Diameter, ESD), and the Fractional Shortening (FS), which measures the contraction efficacy (EDD-ESD/EDD).“

• With respect to the baseline of cardiac performance, there is no simple answer. The baseline is influenced by the genetic background and the experimental conditions. This is the reason why any analysis of mutants or RNAi is conducted in comparison with its own control, analyzed at the same time. Concerning the DGRP lines, no baseline can be defined, since the objective is to measure the diversity of cardiac performance traits within a natural population.

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

Saha et al. have conducted a robust analysis of genes involved in cardiac function in the fruit fly by analyzing heart traits amongst the sequenced inbred lines of the Drosophila Genetic Reference panel. Using seven quantitative traits they identified hundreds of genes associated with natural variation of heart function in the young adult fly. Among the most highly represented groups of genes in their study are transcription factors, and subsequent analysis of SNPs demonstrated that natural variations were frequently found in the vicinity of transcription factor binding sites. Moreover, these transcription factors had already been shown to be associated with variations in heart function. This analysis underscored the importance of transcriptional regulatory networks in heart function. The authors used a heart-specific Gal4 line to drive RNAi knockdown of multiple candidate genes from their quantitative analysis, and showed that knockdown frequently led to cardiac defects. This analysis revealed an interaction between the Activin and BMP signaling pathways in heart activity, a surprising finding given that previous data had shown these pathways to be antagonistic. The authors go on to identify additional genes involved with within-line variation, these genes were also enriched for transcriptional regulators. Finally, the authors identify the human orthologs of their GWAS-associated genes and demonstrate that the genes Stripe and pox meso are associated with increased heart rate.

Major Comments:

There is an assumption in the use of RNAi knockdown to validate the genes identified in the quantitative analysis, and that is that natural variants are themselves hypomorphic. It is possible that among the variants identified some are hypermorphic, or among the transcription factor binding sites that variants lead to increased factor binding. While RNAi knockdown is an excellent choice to begin validation, I do not think the authors can rule out that a gene not functionally validated by their RNAi tests does not have a role in cardiac function.

After performing RNAi knockdown to validate genes identified by GWAS the authors focus on the TGFbeta signaling pathway for downstream analysis. To do so they examine heterozygotes for sog, a repressor of BMP signaling, and snoo, an activator of Activin pathway. The data from the snoo/sog heterozygote is compelling in its disruption of heart phenotypes, and the authors conclude a "coordinated action of activin and BMP." snoo, however, also works as a transcriptional repressor in the BMP pathway, so it's possible that the effects the authors are seeing here could be confined to an increase in BMP signaling. Unlike snoo and sog, mutations in babo and dpp are both expected to have negative effects on Activin and BMP signaling, respectively. The babo/dpp interaction is not as quantitatively convincing as the snoo/sog data, despite the integral roles both babo and dpp play in their respective pathways. If both pathways are connected, why do snoo/sog heterozygotes affect SI phenotypes, while babo/dpp heterozygotes affect fractional shortening? I think the authors data suggest an interesting potential interaction between these pathways, which could be confirmed by examining further mutant combinations, knockdowns or increased expression transgenes, but falls short of a "confirmed synergystic genetic interaction." It does, however, underscore the value of the data in the paper for opening up new avenues for future study.

Minor Comments:

There is an enormous amount of data in this paper, but there are places where things are summarized a little too briefly. For example, there are no definitions given at the beginning of the Results section for traits like "Heart Period" or "Systolic Interval," which would make this work significantly more accessible for other Drosophila researchers. (They do touch on this when they explain later in the paper that certain variants are "associated with quantitative traits linked to heart size and contractility" but more background earlier would be helpful.) When we consider heart performance traits, what is the baseline from known mutants? In other words, where is the line between variation and defect?

Significance

This novel and thorough study is characterized by meticulous data analysis and demonstrated functional significance. The lists of genes represent a wealth of information to begin to understand the etiology of cardiac performance variations, to understand the networks regulating cardiac function and to identify potential human-disease related alleles. It complements GWAS studies done on human populations (such as Arking et al., 2014). It is a fantastic example of the strength of the DGRP and GWAS as tools to understand complex traits. This data will be an important resource for researchers studying cardiac performance in any organism.

The reviewer is a Drosophila geneticist with expertise in mesodermal development and gene regulatory networks.

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 well-written, very comprehensive manuscript that aims to take advantage of the Drosophila Genetic Reference Panel (DGRP) to identify GWAS associated with natural variation of cardiac traits. The authors carefully analyzed the genetic architecture of these natural variation of cardiac performance in the fly, validated a few in both the live fly heart and human iPSC-CMs. Through these efforts, the authors suggested the value of this fly "cardiac performance GWAS" resource and future impact on identifying new gene regulators and pathways critical for heart development and function. This study is very unique and valuable, representing the first effort of such kind in the field. The authors fully leveraged the powerfulness of model organism such as fly to offer a refreshing view on how genetic variants and their interactions potentially contribute to differences in heart performance. Quantitative genetics, functional annotations, and network analyses are thorough and rigorous. The results could be valuable to the community and trigger many follow-up studies. Below are some suggestions for the authors to consider to further strengthen their manuscript:

1. It will be interesting to compare this fly GWAS to human heart disease GWAS data (for example, cardiomyopathy, arrhythmia, heart failure) from patients. Such cross comparison could make the data set more valuable.
2. RNAi is the only experimental approach in this manuscript to validate the functional significance from data analyses. Authors may consider using genetic mutations such as deficiency lines or P-element lines to offer an alternative approach. This is simply a suggestion to improve the rigor and reproducibility, not absolutely required.
3. In order to validate the roles of predicted TF binding sites, the best approach would be introducing point mutations using CRISPR/Cas9 within the binding motif then testing out molecular and physiological outcomes. Rather authors chose to test indirectly to knock down those TFs. If so, authors need to at least acknowledge the potential caveats of such approach and the limitation in related data interpretation.
4. hiPSC-CM data is somewhat limited by only showing the HR and AP duration data. It is recommended to include some immunocytochemistry data to show the morphology, sarcomere structure of these hiPSC-CMs. Gene expression data generated by qPCR or RNA-seq in particular focusing CM structure and function genes would be helpful too.

Significance

This study is very unique and valuable, representing the first effort of such kind in the field. The authors fully leveraged the powerfulness of model organism such as fly to offer a refreshing view on how genetic variants and their interactions potentially contribute to differences in heart performance. The results could be valuable to the community and trigger many follow-up studies.

This reviewer was trained as a fly geneticist during PhD, then a stem cell biologist using hiPSC-CM during postdoc. As a PI, this reviewer has ample experience dealing with genomics, genetics data related to CV biology or 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 #1

Evidence, reproducibility and clarity

In this manuscript, Saha et al. investigated the natural variation and new genetic mechanisms underlying cardiac performance using a collection of inbred Drosophila Genetic Reference Panel (DGRP). Through GWAS analysis, the authors identified more than 500 unique variants associated with 7 cardiac performance traits. These variants are mapped to 332 genes, and located mostly in the 1Kb upstream regions of the TSS. The authors have also functionally verified 42 candidate genes and the knockdown of 38 of them results in cardiac performance defects, including genes in TGF-beta pathway. Finally, the authors examined two novel genes, poxm/PAX9 and sr/EGR2, in both flies and human iPSC-derived cardiomyocytes, which revealed conserved mechanisms for cardiac function across species. I only have minor concerns, as listed below.

1. The authors utilized the RNAi-mediated knockdown approach in their functional validation studies. It is not clear how each genetic variation (SNP) affects its associated genes. Could some of the SNPs activate the candidate gene expression? For the 4 candidate genes that failed to show cardiac defects, could the overexpression of these 4 genes alter cardiac performance?
2. babo is the type I activin receptor, not type 2.
3. The authors show BMP and activin pathway genetically interacts to affect cardiac performance. But it is interesting to find that these interactions are in a trait-dependent manner. For example, it seems that babo and dpp epistatically interact to regulate FS, while they additively regulate HP and DI. The authors need to discuss the complex genetic interaction further.
4. Both snoo and sog are identified from GWAS. How about babo and dpp? Are there any identified SNPs associated with babo and dpp?
5. It is unclear why the mad KD behaves oppositely to dpp mutant, although both proteins are involved in BMP pathway. In Figure S5, the mad KD shows reduced FS and HP, but dpp LOF mutant shows increased FS and HP (Figure S4). Can the authors perform RNAi to knockdown dpp specific in the heart to reexamine the role of dpp in the regulation of cardiac function. The whole body LOF mutant dpp-d14 might not target cardiac tissue directly to control heart performance like mad KD.
6. The authors selected two novel genes to study the conversed regulation in both flies and human iPSC cells. Besides testing these novel genes, the authors should also verify whether the conserved pathways, like TGF-beta, regulate heart performance in human iPSC cells similar to the flies.

Significance

Overall, the manuscript is well written and the experiments are well-thought-out. The newly identified mechanisms in this study will provide important insights into the genetic architecture of the complex cardiac performance traits.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

**Note:** This preprint has been reviewed by subject experts for *Review Commons*. Content has not been altered except for formatting.

---

Reply to the reviewers

**General Statements [optional]**

Please find bellow the preliminary revision plan for our manuscript entitled “Recognition of copyback defective interfering rabies virus genomes by RIG-I triggers the antiviral response against vaccine strains” by Wahiba Aouadi et al. (RC-2022-01386). Reviewer’s’ comments/questions and suggestions are represented in blue in the text.

**Description of the planned revisions**

We thank the Reviewer#1 for underlining that identification of rabies virus 5’ copy-back DI genomes as “presumably bound to RIG-I is a useful advancement”, her/his interest in “observed difference between the responses to the two strains of virus” (THA strain and the vaccine SAD strain), and for emphasizing that “identification of the rabies viral RNAs that activate RIG-I is a significant finding for the rabies specialists”.

Reviewer#1: In more details for the Evidence, reproducibility and clarity (Required) Much of the studies relied on weak methodologies. For example, in Fig 1, reporter assays were used, instead of measuring IFN mRNA levels; it is also not clear what is the nature of the promoter driving the reporter. Is it ISRE, which responds to IFN or is it the IFNb promoter, which responds to transcription factors activated by RIG-I? It is also not clear what is the nature of the RNA that was transfected. Is it total RNA from infected cells or is it purified viral RNA? No matter what, these results are quite predictable from the literature.

Regarding the Reviewer #1 comment on type-I IFN cell report results “relying “on weak methodologies”, we would like to recall that to provide pieces of evidence that RIG-I-specific RNA ligands are produced during the infection with rabies virus we used several previously validated technics: - i) Fig.1A: transfected into ISRE-reporter cell line (ISRE, which responds to IFN) that is a classical validated tool to efficiently detect ISRE-activation even upon transfection of low quantities of immunoactive RNA ligands (PMID: 28768856, PMID: 27011352, PMID: 24098125, PMID: 29996094, PMID: 31761719, PMID: 23595062). - ii) Fig1B: Cell overexpressing LGP2 approach that has been previously developed and validated (Sanchez et al., 2019). LGP2 overexpressing cells provide a possibility to functionally distinguish between RIG-I and MDA5-driven activation of type-I IFN signaling. As noted in the corresponding figure legend in this experiment, the IFN-b promoter-reporter assay was used (“which responds to transcription factors activated by RIG-I”). - iii) Fig.1C -similar to Fig1A experiments performed in ISRE-reported cell lines partially depleted in either RIG-I or MDA5 (siRNA-based approach) to complement the Fig. 1B with additional functional validation using siRNAs.

We apologize that we haven’t provided a detailed explanation for the origin of transfected RNA used all through Fig. 1. In the revised Fig1 we will correct the figure legend to explain the origin of total RNA used in experiments: Total RNA purified from SK.N.SH cells infected with THA or SAD for all experimental approaches presented in the figure. Moreover, as suggested by Reviewer#1 for Fig.1 we will add experiments measuring IFN-b mRNA by RT-qPCR.

Referee #1

Evidence, reproducibility and clarity

A lot of effort was devoted to distinguish between RIG-I and MDA5 as the receptor of rabies viral RNA producing conflicting results from the binding assays and the reporter assays.

This comment of Reviewer#1 is not clear to us. We have the feeling that our results do not show any conflict when analyzing the results represented in Fig. 2-3. They demonstrate that RIG-I and not MDA5 works as the key cytosolic sensor upon infection with rabies virus. Further, the apparent conflict observed by the Reviewer#1 about the fact that we failed to detect any specific RABV RNA ligands upon infection with THA strain (Fig.3A) while significant enrichment of immunoactive RNA ligands on RIG-I (Fig.2C) were observed can be easily commented and explained. We proposed in the revised version of our manuscript to discuss the possibility and to provide the results showing that enrichment in 5’PPP endogenous RNA ligands on RIG-I upon infection with THA RABV could explain the results observed on Fig.2C /Fig.3A. Indeed, in our recently accepted for publication study, we observed that a large spectrum of RNA virus infections leads to the mobilization of endogenous RNA ligands (transcripts of RNA Polymerase III) on RIG-I (https://www.cell.com/iscience/fulltext/S2589-0042(22)00871-9). Furthermore, we observed that upon infection Polymerase III transcripts can activate RIG-I signaling pathways even in the absence of RIG-I-specific viral RNA ligands. To address this possibility in the revised manuscript, we propose to perform additional analysis of our RNAseq results to demonstrate enrichment of endogenous RNA ligands on RIG-I in rabies virus-infected cells.

Significance

Conceptually, the paper does not add much to the literature. As pointed out by the authors, RIG-I-specific partners had been identified before for many RNA viruses including other rhabdoviruses.

We additionally underline that although there is a slowly growing number of studies characterizing RLR-specific RNA ligands directly from infected cells with a slowly growing number of characterized viruses, to our knowledge our study provides the first characterization of RLRspecific RNA ligands in Rabies virus-infected cells and that the amount of these ligands differs between wild type viruses and vaccine strains. Furthermore, none of the previously published studies on Rabies virus used similar experimental approaches. We believe that only stepwise characterization of RLR-specific RNA ligands for different RNA virus families is fully original regarding rabies virus and will further provide a wider and more fundamental vision on the distribution of RIG-I and MDA5 specificities for sensing RNA viruses.

Referee #3

Evidence, reproducibility and clarity

We thank Reviewer#3 for stressing that our “study is highly significant for understanding virus sensing mechanisms and to inform understanding of vaccine actions.” For the Reviewer#3 specific comments:

The signaling analyses is focused on ISRE/promoter induction, which is several steps downstream from RIG-I. An more comprehensive signaling analysis is required to define the RLR pathway engagement, including examination of RIG-I binding to MAVS, IRF3 activation induced by viral RNA and recovered RIG-I or MDA5 ligands, and induction of IRF3-target gene expression (such as RSAD, IFI44, IFIT1, IFIT2) and interferon-stimulated gene (ISG) expression such as Mx1, Mx1, OAS, etc.

We thank Reviewer#3 for his comments and also appreciate that additional characterization of type-I IFN signaling pathway activation by RABV RNA will deeper our research results. We will add additional experimental results to answer the comments suggested by the Reviewer#3 for each Figure, as presented below:

Figure 1. RLR activation readout here relies exclusively on promoter/reporter assay. Assessment of endogenous IRF3, IRF3-target gene expression, and ISG expression needs to be included. Also, what are the dynamics of RLR signaling activation during infection over a time course? This is important to know and to associate with the accumulation of the cb RNAs.

We will perform additional transfection of total RNA purified from SK.N.SH cells infected with THA or SAD to HEK293T (or other relevant cells) to detect by WB analysis the phosphorylation of IRF3. As suggested by the Reviewer#3 we will also perform gene expression analysis targeting RSAD, IFI44, IFIT1, IFIT2. Additionally, kinetics of the SK.N.SH cells infection with THA or SAD strains of RABV will be studied to detect the accumulation of 5’cbDI genomes during the infection as suggested at the second part of the comment by the Reviewer#3.

Figure 2. The RLR-bound RNA signaling analysis is incomplete. The authors need to include analysis of IRF3 and gene expression as noted above. Also, the authors should assess RLRbound RNAs collected over a time course of infection, thus enabling an understanding of the temporal dynamics of RLR ligand and biological activity of this virus-host interaction.

In order to reply to this comment we will provide additional characterization of type-I IFN signaling in ST-RLR cells infected with THA and SAD, comparing to the mock-infected cells. For this, we will perform western blot analysis of IRF3P in total protein lysates and carry additional analysis of our NGS data to visualize ISG expression profiles in the same conditions (THA, SAD, and mock). Unfortunately, it will be experimentally difficult to assess RLR-bound RNAs collected over a time course of infection. However, as our NGS analysis demonstrated accumulation of 5’cb DI RNA as specific RNA ligands of RIG-I, we can follow the kinetics of accumulation of these 5’cb DI RNAs in SK.N.SH and ST-RLR cells as described above in response to the Fig.1 comment of the Reviewer#3.

Figure 3. These are strong data sets and are convincing. For panel C, one can see several RIGI-bound peaks. The authors should provide more information on the length of these peaks, please include in Table 1. Also for MDA5 there also are peaks but the histogram is saturated. The peaks and valleys need to be delineated, ideally in a large table. The needs to be confirmation of these motifs or RNAs as actually binding to RIG-I and MDA5. This binding activity needs to be shown in gel-shift assay or other suitable approach of direct RIG-I binding of specific RNAs produced in vitro corresponding to mapped regions shown in the figure 3. Also, a more careful analysis of MDA5-assocaited RNA needs to be conducted to ascertain if it has immune stimulatory/signaling activity. By assess IRF3 activation this activity might be identified.

Based on the Reviewer#3 suggestions for the Fig.3C we will additionally summarize in Supplementary Table 4 RNA reads that are represented as enriched on RIG-I for the 5’ part of the RABV genome. Indeed, the full-length genome binding to MDA5 was observed for RNA- reads importantly in SAD-infected cells. However, we believe that how encapsidated full-length viral genome can still be detected by MDA5 in virus-infected cells needs to be addressed in a separate study. Additional experiments for detecting the IRF3 activation in ST-RLR cells will be performed as described above.

Figure 4: VERY important: Do these RNAs bind to RIG-I in vitro, and do they activate IRF3 when transfected into cells, what is the role of 5'ppp in this activity?? These data are needed to make the strong conclusions stated by the authors.

We are grateful to Reviewer#3 suggestions for Fig.4. We will address whether the detected RABV 5’cb DI RNAs are specific RIG-I ligands. We will synthetize and transfect these RNA molecules and study how efficiently they activate type-I IFN signaling (by IFN-b and ISRE reporter approaches as well as by gene expression assay analysis as suggested in Fig.1 by Reviewer#3). We will also address IRF3P efficiency upon cell transfection with DI-2170 and DI-1668. As controls, we will use previously described RIG-I/MDA5-specific RNA ligands and treat RNA transcripts with calf intestine alkaline phosphatase (CIP) to remove 5’ppp groups.

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

No revisions have already been incorporated in the transferred manuscript.

**Description of analyses that authors prefer not to carry out**

As described above to answer to the Reviewer#3 suggestion, how encapsidated full-length viral genome can still be detected by MDA5 in virus-infected cells needs to be addressed in a separate study.

Referee #2

Evidence, reproducibility and clarity

We thank the Reviewer#2 for underlining that our study “shed light on the RLR recognition of RABV RNAs upon infection” and that our study “clarify the mechanism of cellular immunity differences between RABV pathogenic strain and vaccine attenuated strain. Reviewer#2 suggested to “verify whether the difference in this mechanism is caused by the difference in the viral genome, whether the N gene or L gene of the two can be exchanged by reverse genetics, and then infect the cells to verify whether the 5'cb DI genomes can be generated just as this paper.”

We agree with Reviewer#2 that applying reverse genetics for RABV genome by exchanging N and L genes could provide a more in-depth characterization of 5’cb DI generation and pathogenicity of RABV. However, these additional experiments cannot be provided within the scope of this paper and will take time for the revision process. We believe, that this question needs to be addressed in a separate study by exchanging either N and L genes using reverse genetics.

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

Bourhy and colleagues present their study focused on defining how rabies virus (RV) vaccine strains trigger RIG-I innate immune signaling. Triggering RIG-I or MDA5 leads to innate immune activation, and in vaccinology this is an important component of immune adjuvant actions that serve to overall enhance vaccine immunity. The group applied in vitro infection and RNA analyses including assessment of RLR-dependent signaling by RNAs recovered from infected cells, and RNAseq of RLR-associated RNA from virus-infected cells, showing that RIG-I binds to copyback (cb) RNAs of defective interfering (DI) genomes produced during replication by the RV vaccine strain SAD but not by THA strain which likely does not produce the cb RNAs. The study extends previous work showing that RIG-I senses RV RNA to now show That RIG-I binds to the cb RNAs. Data on MDA5 is included to show that MDa5 is bound across the RV negative strand but the RNA recovered from MDA5 in infected cells does not stimulate innate immune signaling.

Specific comments:

The signaling analyses is focused on ISRE/promoter induction, which is several steps downstream from RIG-I. An more comprehensive signaling analysis is required to define the RLR pathway engagement, including examination of RIG-I binding to MAVS, IRF3 activation induced by viral RNA and recovered RIG-I or MDA5 ligands, and induction of IRF3-target gene expression (such as RSAD, IFI44, IFIT1, IFIT2) and interferon-stimulated gene (ISG) expression such as Mx1, Mx1, OAS, etc.

Figure 1. RLR activation readout here relies exclusively on promoter/reporter assay. Assessment of endogenous IRF3 , IRF3-target gene expression, and ISG expression needs to be included. Also what are the dynamics of RLR signaling activation during infection over a time course? This is important to know and to associate with the accumulation of the cb RNAs.

Figure 2. The RLR-bound RNA signaling analysis is incomplete. The authors need to include analysis of IRF3 and gene expression as noted above. Also, The authors should assess RLR-bound RNAs collected over a time course of infection, thus enabling an understanding of the temporal dynamics of RLR ligand and biological activity of this virus-host interaction.

Figure 3. These are strong data sets and are convincing. For panel C, one can see several RIG-I-bound peaks. The authors should provide more information on the length of these peaks, please include in Table 1. Also for MDA5 there also are peaks but the histogram is saturated. The peaks and valleys need to be delineated, ideally in a large table. The needs to be confirmation of these motifs or RNAs as actually binding to RIG-I and MDA5. This binding activity needs to be shown in gel-shift assay or other suitable approach of direct RIG-I binding of specific RNAs produced in vitro corresponding to mapped regions shown in the figure 3. Also, a more careful analysis of MDA5-assocaited RNA needs to be conducted to ascertain if it has immune stimulatory/signaling activity. By assess IRF3 activation this activity might be identified.

Figure 4: VERY important: Do these RNAs bind to RIG-I in vitro, and do they activate IRF3 when transfected into cells, what is role of 5'ppp in this activity?? These data are needed to make the strong conclusions stated by the authors.

Review cross-commenting:

I agree completely with the comments provided by other two reviewers.

Significance

The study is highly significant for understanding virus sensing mechanisms and to inform understanding of vaccine actions.

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 entitled by"Recognition of copy-back defective interfering rabies virus genomes by RIGI triggers the antiviral response against vaccine strains"，which clarify the mechanism of cellular immunity differences between RABV pathogenic strain and vaccine attenuated strain in vitro. This paper using next-generation sequencing (NGS) combined with bioinformatics tools, it was found that the RABV vaccine attenuated strain replication in vitro induces a high release of 5' copy-back defective interfering genomes, which enhances a strong antiviral response. However, RABV pathogenic strain replication in vitro is characterized by the absence of defective interfering genomes thus induces a weak RLR-mediated innate immunity antiviral response. This paper demonstrated that IFN response induced by RLR RABV RNA recognition was principally mediated by RIG-I. 5'cb DI viral genomes that enhance RIG-I detection and therefore strongly stimulate the IFN response were exclusively produced by the RABV vaccine strain. To verify whether the difference in this mechanism is caused by the difference in the viral genome, whether the N gene or L gene of the two can be exchanged by reverse genetics, and then infect the cells to verify whether the 5'cb DI genomes can be generated just as this paper.

Review cross-commenting:

I agree completely with the comments provided by other two reviewers.

Significance

This paper shed light on the RLR recognition of RABV RNAs upon infection.

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

Rabies virus, a rhabdovirus, is a major human pathogen and mammalian cells respond to infection by this virus by triggering type I IFN synthesis. Here, the authors report that the cytoplasmic antiviral sensor RIG-I recognizes viral RNA to initiate signaling. Moreover, the vaccine strain SAD activates RIG-I more effectively than the pathogenic strain, THA. During SAD replication, two major 5' copy-back defective interfering genomes were generated and they bound to RIG-I to activate it.

Much of the studies relied on weak methodologies. For example, in Fig 1, reporter assays were used, instead of measuring IFN mRNA levels; it is also not clear what is the nature of the promoter driving the reporter. Is it ISRE, which responds to IFN or is it the IFNb promoter, which responds to transcription factors activated by RIG-I? It is also not clear what is the nature of the RNA that was transfected. Is it total RNA from infected cells or is it purified viral RNA? No matter what, these results are quite predictable from the literature. A lot of effort was devoted to distinguish between RIG-I and MDA5 as the receptor of rabies viral RNA producing conflicting results from the binding assays and the reporter assays. A major weakness of these experiments is in the use of convoluted cell lines which added to the weakness of the reporter assays as outlined above. Identification of the DI viral sequences that presumably bound to RIG-I is a useful advancement.

Significance

Conceptually, the paper does not add much to the literature. As pointed out by the authors, RIG-I-specific partners had been identified before for many RNA viruses including other rhabdoviruses. The observed difference between the responses to the two strains of virus is interesting but multiple strains need to be tested to make a meaningful interpretation of the data. Nonetheless, identification of the rabies viral RNAs that activate RIG-I is a significant finding for the rabies specialists.

This reviewer's expertise is in antiviral innate immune response and the type I IFN system.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

This reviewer did not leave any comments

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

Major points:

1. The authors claim the Alyref and Gabpb1 were repressed by H3K9me3 in SCNT embryos, however, they didn't provide the direct evidence of H3K9me3 modification in Alyref and Gabpb1 promoter or enhancer.
2. The authors should provide direct evidence that how Alyref and Gabpb1 regulate pluripotency, cell viability and apoptotic related genes which are essential for morula arrest in knock out embryos.

Minor points:

1. Figure1A, which developmental stage of SCNT or IVF embryos are used for RNA-seq?
2. In Figure S2B, FPKM values of Alyref in SCNT embryos with TSA+VC-1 is inconsistent low. More repetitions are recommended.
3. The samples of single embryo RNA-seq are less. More repetitions are recommended. 4.They lack the data of embryos transfer and offsprings experiment of SCNT embryos by the addition of Alyref and Gabpb1 through mRNA injection.

Significance

The manuscript by Ihashi et al aims to find specific genes responsible for the arrest of SCNT embryos. The authors identified Alyref and Gabpb1 by siRNA screening and verified morulae arrest phnotype in Alyref and Gabpb1 KO IVF embryos, and single embryo RNA-seq revealed that Alyref is needed for the formation of inner cell mass. The preimplantation development of cloned embryos was aided by the addition of Alyref and Gabpb1 by mRNA injection. Overall, this is an interesting study that demonstrate incomplete activation of Alyref1 and Gabpb1 will lead to preimplantation arrest of SCNT embryos. However, this is a very preliminary study that some important issues are need to address, thus, this manuscript is far from a publishable form.

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:

The authors picked up candidate genes which might be important for SCNT embryo development based on RNA-seq data (basically genes downregulated in SCNT embryos compared to normal ones) and performed siRNA-knockdown on the candidate 15 genes. Two genes, Alyref and Gabpb1, were required for normal development. KO of each gene resulted in embryonic lethality, confirming the KD data. RNA-seq showed that Alyref KO affected lineage specification while Gabpb1 KO resulted in apoptosis. Finally the authors injected mRNA for Alyref and Gabpb1 into SCNT embryos and observed improved development.

Major comments:

1. The data provided are overall convincing and support the authors' conclusion. However, as the authors may understand, the paper lacks mechanistic insights into how Alyref and Gabpb1 work in early embryos. Furthermore, it is still not very clear why Alyref and Gabpb1 are downregulated in SCNT embryos. It seems that the authors speculate that somatic H3K9me3 might regulate the expression of those genes, but it is still possible that H3K9me3 just inhibits the expression of upstream regulator of Alyref and Gabpb1. The paper lacks experiments to address these possibilities.

2. Fig. 7: Given the current status of the SCNT field, it would be important to show the birth rate of SCNT embryos upon mRNA injection.

3. Fig. 2C-D: It would be important to know when the differential expression of Alyref or Gabpb1 protein can be seen to understand the relationship between SCNT RNA-seq data and KO embryo phenotype.

Minor comments:

1. State clearly in the manuscript which stage of embryos was used for RNA-seq analysis or other assays.

2. Fig 1A: "DEGs" might be confusing. Do the authors refer to downregulated genes as DEGs?

3. L113: State clearly what the "transcriptional activators" in this study are. Also, "Repression mix" and "Activation mix" in Fig1C are not easily understandable.

4. Fig6A: Are the pathways indicated top significant ones? If not, the IPA result should be indicated in an unbiased manner. Also, is it meaningful to show -log10(q-value) = ~1?

5. S6A: The stage at which the Pou5f1 signal was measured is not clearly indicated; Pou5f1 expression becomes high in blastocysts, so comparison between 72 hpi morula would be appropriate.

6. L217, Fig 6F: Is this indeed based on unsupervised hierarchical clustering?

7. "heterozygous mutant mice" should be used rather than "hetero mice".

Significance

The study is constructed based on the previous finding that Kdm4d overexpression significantly improved SCNT embryo development. However, it is still important to know why the removal of H3K9me3 can exert such an effect. The authors tackled this question and suggested that two genes (Alyref and Gabpb1) expressed upon H3K9me3 removal play important roles for SCNT embryo development. Unfortunately, while it is now widely accepted that Dux expression is important for SCNT development in the field, the authors did not test nor discuss how Dux is involved in the authors' findings. In addition, there would be many other things to be done in this study (described in major comments) to contribute to the SCNT field.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Response to previous review

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

The authors use newly available probes to show that the phosphoinositide PI(3,4)P2 plays a previously undescribed role in FcgammaR-mediated phagocytosis. Using RAW macrophages, they show that PI(3,4)P2 is enriched at the plasma membrane and also at phagocytic cups internalizing IgG-opsonized sheep red blood cells. Pharmacological inhibition using wortmannin, and also expression of membrane-targeted INPP4B phosphatase showed that PI(3,4)P2 production depends on PI3 kinase activity. Further experiments using selective inhibitors showed that PI(3,4)5P2 is mainly derived by dephosphorylation of PI(3,4,5)P3, likely by multiple phosphatases such as SHIP1 or OCRL. Depletion of PI(3,4)P2 at the plasma membrane by INPP4B also resulted in strongly decreased internalization of red blood cells, although their attachment to macrophages seemed unaltered, pointing to defects in particle engulfment. The authors then tested the potential role of lamellipodin, one of the few known PI(3,4)P2 specific effectors. Lamellipodin was found to be enriched at phagocytic cups, and this enrichment was shown to be dependent on the presence of PI(3,4)P2, by targeting of INPP4B to the plasma membrane. Macrophages depleted of lamellipodin by shRNA treatment showed reduced phagocytic efficiency and also aberrant phagocytic cup formation. As VASP is a known binding partner of lamellipodin and involved in actin polymerization, the authors next tested its potential involvement. Overexpression experiments showed that VASP colocalizes with lamellipodin at phagocytic cups. Sequestering of VASP at mitochondria through a respective construct containing the VASP binding site of ActA, together with a mitochondrial targeting sequence, showed that this also results in incompletely formed phagocytic cups and reduced phagocytic efficiency. Similar effects were observed upon expression of a lamellipodin construct with mutated binding sites for VASP. Collectively, the authors propose that PI(3,4)P2 is localized produced at phagocytic cups through the sequential activity of PI3 kinase and PI5 phosphatase, that it recruits lamellipodin and its binding partner VASP, and that this cascade is necessary for proper phagocytic cup formation and closure and thus phagocytic capacity of cells. This is an interesting study that uncovers a novel role for PI(3,4)P2 in phagocytic cup formation and closure. It is very well controlled, and the claims of the study are supported by the presented data. Statistical analysis is sound.

Major comments: 1) The localization of VASP at phagocytic cups is only shown by overexpression of constructs. Endogenous staining of VASP should support this finding.

We agree with the reviewer that localization of the endogenous VASP would strengthen our conclusions. We have therefore performed the suggested experiments and in the revised manuscript include a new panel (E) in the revised Figure 6 showing immunostaining of endogenous VASP during phagocytosis. The result confirms the localization of the GFP-chimeric protein.

2) It is unclear whether the roles of PI(3,4)P2, lamellipodin, and VASP are restricted to FcgammaR-mediated phagocytosis. Their potential involvement in CR3-mediated phagocytosis should be discussed or addressed in a basic set of experiments.

In the revised manuscript we have extended our original observations to analyze also CR3-mediated phagocytosis, as recommended by the reviewer. A new supplemental figure (Supplemental Figure 10) now documents that PtdIns(3,4)P2 is also accumulated and Lamellipodin and VASP are recruited to the phagocytic cup during CR3-mediated phagocytosis. These results imply that the role of this lipid and its effectors extend to other modes of phagocytosis. These new observations are discussed on Page 14 of the revised manuscript.

Minor comments: 1) A very recent study (Körber and Faix, EJCB, 2022) describes the role of VASP in macroendocytosis in Dictyostelium. Specifically, VASP is found to be important for proper cup closure. The results are of direct importance to the current study and should be cited accordingly.

Thank you for bringing this study to our attention. We now discuss the findings of Körber & Faix on Page 9 of the revised text.

2) direct labelling of the figures would be helpful in assessing the manuscript

To facilitate re-assessment of the paper, we have added the Figure numbers directly to the individual figures in the manuscript as suggested.

Reviewer #1 (Significance (Required)): This study highlights the role of an underappreciated phospholipid in phagocytosis. It also describes for the first time a role for lamellipodin in formation of phagocytic cups and confirms the recent finding that also VASP is necessary for phagocytic cup closure. The paper should be of interest to researchers working on host-pathogen interaction, regulation of the actin cytoskeleton, and also to the general cell biological community

Reviewer´s expertise: Actin regulation Microtubule-based transport Adhesion, migration, invasion Phagocytosis

We thank the reviewers for his/her comments and suggestions that have clearly improved the manuscript.

Reviewer____ #2 (Evidence, reproducibility and clarity (Required)):

Review of Montano-Rendon et al: 'PI(3,4)P2, Lamellipodin and VASP coordinate cytoskeletal remodeling during phagocytic cup formation in macrophages'

The authors employ biosensors for PI(3,4)P2 in RAW 264.7 macrophages to identify localized pools of PIP2 that were sensitive to INPP4B and wortmannin (Fig1). The biosensors for PIP2 are enriched on the forming phagocytic cup (Fig2, movies) in these macrophage cells. Inhibitors for PI3K blocked the recruitment of this biosensor to the membrane.

Overall, the data are clear with the exceptions noted below. Krause et al (Dev Cell 2004) published a manuscript looking at PIP2, Lpd, and VASP in non-macrophage cells (fibroblasts, HeLa, etc...) where the influence of PI(3,4)P2 and these proteins was found to regulate actin and lamellipodial membrane extensions. This study also implicated Lpd protein coordinated actin networks in the docking of pathogens such as Vaccinia virus and EPEC bacteria. Given the additional reports of these proteins participating in dorsal ruffling (Michael et al Curr Bio 2010) and invasion (Carmona et al Oncogene 2016), it comes as no surprise that they participate in phagophore formation and phagocytosis. These studies are referenced, but having this in mind does diminish the novelty of implicating Lpd and VASP in the phagocytic process, though it seems to be the first time this machinery was directly implicated in macrophage cells.

We would like to point out that docking of viruses or dorsal ruffling are very different biological processes from phagocytosis and that the common involvement of Lamellipodin in these very disparate processes does not, in our view, detract from the novelty of our studies.

Specific Comments Although the images and movies graphically demonstrate a PI(3,4)P2 enrichment on phagocytic structures, the authors could provide some additional images that include fluorescently tagged phagocytic cargo such as the erythrocytes used. The addition of a fluorescent marker or phase image would be especially beneficial in the experiments where a lack of cPHx-biosensor recruitment is seen to the docked phagocytic cargo.

We thank the reviewer for this suggestion. In the revised manuscript figures now include micrographs of the fluorescently labelled particles or phase-contrast images where appropriate.

Otherwise, readers are left with the impression that perturbations such as INPP4B compromise docking and phagocytic cup formation altogether (Fig 2C)- which is perhaps the authors point? Make this clear?

We apologize for the ambiguity of the former version of the manuscript. Initially, we noticed that particle engulfment -which is what we believe the reviewer means by “cup formation”- was the main defect in INPP4B-CaaX expressing cells. However, since the reviewer raised the possibility, we have gone back and re-analyzed the data and found that cells expressing the INPP4B-CaaX also have a small (~35%) decrease in particle engagement/binding (Reviewer uses the term “docking”). This suggests that the plasmalemmal pool of PtdIns(3,4)P2 in resting cells supports the actin dynamics at the cell surface which allows the RAW cells to survey their immediate environment and thereby contact more potential prey. This new finding is included in and discussed the revised manuscript. We thank the reviewer for prompting us to consider this alternative mechanism.

There has already been an implication for PI3K in the phagocytic process, perhaps verifying that initial formation/membrane extension stages of phagocytosis are impacted by targeting the D-4 position of PIP2 would be of interest?

PtdIns(4,5)P2 is well known to be essential for actin polymerization and is increased transiently at the sites of phagocytosis (Botelho et al., 2000 J Cell Biol; Scott et al., 2005 J Cell Biol; Fairn et al., 2009 J Cell Biol.). It is not clear whether the reviewer is curious about the possible consequences of converting PtdIns(4,5)P2 to PtdIns5P prior to activation of PI3K. Whether PtdIns5P itself has biological activity is a subject of debate and, to our knowledge, its existence has not been documented at sites of phagocytosis. It is also unclear whether PtdIns5P would serve as an effective substrate for PI3K and, if so whether the putative product, PtdIns(3,5)P2 that is normally found in endomembranes, would be functionally relevant.

Depletion of PI(3,4)P2 through the expression of the INPP4B phosphatase demonstrated a reduction in phagocytic uptake of red blood cells (Fig4). The readout for this assay relied upon what appears to be differential labeling of phagocytosed red blood cells, though there are examples of cargo that is supposedly inside the macrophages labeled in green? Perhaps the authors can reconcile this and make the methods more clear for this approach?

Thank you for this comment; we apologize if the original text was unclear in this regard. In the revised manuscript a detailed description of the staining protocol we used to distinguish inside from outside particles is now included in the Methods section. It is also worth pointing out that the green-only SRBCs in the INPP4B-CaaX panel in Figure 3 indicate targets that were fully internalized by those RAW macrophages not expressing BFP-INPP4B-CAAX (see image below)

Fig4 demonstrates the PI(3,4)P2 dependent recruitment of Lamellipodin (Lpd) to the phagocytic cup, which is clear. Lpd is found to be necessary for effective phagocytic uptake in Fig 5. There is no blotting/qPCR data for the verification of Lpd knockdown shown?

RAW macrophages and other macrophage cell lines are rather refractory to transfection, resulting in only a minor (10-20%) fraction of the cells expressing transfected constructs. For this reason, immunoblotting or qPCR analyses of the entire population yield misleading results, not reflective of the comparatively small transfected sub-population of cells. To overcome this limitation, we co-transfected the shRNA-containing plasmid with a smaller amount of a plasmid containing a fluorescent protein used to identify transfectants visually (a 5:1 ratio of shRNA:EGFP). By using a 5:1 ratio of the plasmids we ensured that cells expressing the fluorescent protein had a high likelihood of also expressing the shRNA. In this manner, the Lpd-depleted cells could be scored separately from the untransfected, wild-type cells following immunostaining (Supplemental Figure 5). Note that some immunostaining persisted in the Lpd-silenced cells, in all likelihood because some of the antibody binding is nonspecific, as is commonly seen in immunostaining. Nevertheless, the data indicate that substantial silencing of Lpd is achieved when transfecting the shRNA.

The authors demonstrate a co-localization of Lpd/VASP proteins at the phagocytic cup of these macrophages in Fig 6 and sequester VASP protein to the mitochondria with some ActA derived fusion proteins to functionally block phagocytosis. The functional interaction of Lpd/VASP is further explored with experiments utilizing Ena/VASP mutants in Fig7, demonstrating a dependence on this interaction to promote phagocytic uptake.

Reviewer #2 (Significance (Required)): see above

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

In this study, Montano-Rendon and colleagues address the role of phosphatidylinositol (3,4)-bisphosphate in phagocytosis by RAW macrophages. Using small molecule inhibitors, they show that dephosphorylation of PI(3,4,5)P3 is the main source of PI(3,4)P2 in phagocytosis. Using an elegant approach based on overexpression of a PI(3,4)P2-specific phosphatase, they show that the selective depletion of PI(3,4)P2 impairs phagosome formation. Moreover, they identify two PI(3,4)P2 interacting proteins involved in phagocytosis: lamellipodin and VASP. They show that shRNA silencing of lamellipodin arrests phagocytosis, as well as mistargeting of VASP to mitochondria by a fusion protein. Overall, this is a high-quality study, well designed and written. I hence support publication, and only have a few relatively minor comments that the authors should consider as I believe it would improve the quality of the manuscript.

The role of PI(3,4)P2 in the actin organisation in phagocytosis has been shown previously in various studies, see for example PMID: 16418223, 27806292 and review 32296634. In these studies, different mechanisms have been proposed of how PI(3,4)P2 affects the cytoskeleton and phagocytic process. It would be good to discuss how the findings with lamellipodin and VASP relate to these previously described mechanisms.

We now include and discuss the references recommended by the reviewer to highlight that the importance of PtdIns(3,4)P2 extends to dendritic cells and HL60 neutrophils.

In figure 6, a role for VASP in phagocytosis is shown by mistargeting it to mitochondria using a fusion protein consisting of a VASP binding region and a mitochondrial targeting motif. While this is an elegant approach, I wonder why not simply shRNA is used, similar to lamellipodin?

We decided to use this approach because macrophages (including RAW cells) express other members of the Ena/VASP family of proteins such as EVL (Coppolino et al., 2001 J Cell Sci) that could potentially substitute for VASP; simultaneously silencing multiple, distinct members of the Ena/VASP family poses an experimental challenge. Moreover, in our experience introducing siRNA into RAW cells, even when using electroporation, is often insufficient to generate robust silencing of certain genes (e.g. Levin-Konigsberg, et al., 2019 Nature Cell Biology). Thus, we took advantage of the robust, more globally effective ActA-based molecular tool. To demonstrate its effectiveness, we now include a new Supplemental figure (Supplemental Figure 6, reproduced below) using immunostaining that shows how virtually all of the endogenous VASP is sequestered to the surface of mitochondria when the MITO-FP4 is expressed.

Supplemental Figure 6. MITO-FP4 targets endogenous VASP to the Mitochondria

In figure 3A: How was the inside-outside staining performed? I cannot find this information in the Methods.

We apologize for the omission. The inside/outside staining protocol is now detailed in the Methods section of the manuscript.

Figures are overall good quality. However, in figure 1,2, and 4 individual cells are shown in the graphs, whereas figures 3, 5, 6 and 7, and the supplementary figures only show averages with bar graphs. Please change these graphs to all show individual cells, as this will allow to see the variation among cells.

Thank you for the suggestion. The graphs have been modified to violin plots to show the variation and distribution of results amongst the individual cells and experiments.

Reviewer #3 (Significance (Required)):

see above

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

In this study, Montano-Rendon and colleagues address the role of phosphatidylinositol (3,4)-bisphosphate in phagocytosis by RAW macrophages. Using small molecule inhibitors, they show that dephosphorylation of PI(3,4,5)P3 is the main source of PI(3,4)P2 in phagocytosis. Using an elegant approach based on overexpression of a PI(3,4)P2-specific phosphatase, they show that the selective depletion of PI(3,4)P2 impairs phagosome formation. Moreover, they identify two PI(3,4)P2 interacting proteins involved in phagocytosis: lamellipodin and VASP. They show that shRNA silencing of lamellipodin arrests phagocytosis, as well as mistargeting of VASP to mitochondria by a fusion protein. Overall, this is a high-quality study, well designed and written. I hence support publication, and only have a few relatively minor comments that the authors should consider as I believe it would improve the quality of the manuscript.

The role of PI(3,4)P2 in the actin organisation in phagocytosis has been shown previously in various studies, see for example PMID: 16418223, 27806292 and review 32296634. In these studies, different mechanisms have been proposed of how PI(3,4)P2 affects the cytoskeleton and phagocytic process. It would be good to discuss how the findings with lamellipodin and VASP relate to these previously described mechanisms.

In figure 6, a role for VASP in phagocytosis is shown by mistargeting it to mitochondria using a fusion protein consisting of a VASP binding region and a mitochondrial targeting motif. While this is an elegant approach, I wonder why not simply shRNA is used, similar to lamellipodin? In figure 3A: How was the inside-outside staining performed? I cannot find this information in the Methods.

Figures are overall good quality. However, in figure 1,2, and 4 individual cells are shown in the graphs, whereas figures 3, 5, 6 and 7, and the supplementary figures only show averages with bar graphs. Please change these graphs to all show individual cells, as this will allow to see the variation among cells.

Significance

see above

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

Review of Montano-Rendon et al: 'PI(3,4)P2, Lamellipodin and VASP coordinate cytoskeletal remodeling during phagocytic cup formation in macrophages'

The authors employ biosensors for PI(3,4)P2 in RAW 264.7 macrophages to identify localized pools of PIP2 that were sensitive to INPP4B and wortmannin (Fig1). The biosensors for PIP2 are enriched on the forming phagocytic cup (Fig2, movies) in these macrophage cells. Inhibitors for PI3K blocked the recruitment of this biosensor to the membrane.

Overall, the data are clear with the exceptions noted below. Krause et al (Dev Cell 2004) published a manuscript looking at PIP2, Lpd, and VASP in non-macrophage cells (fibroblasts, HeLa, etc...) where the influence of PI(3,4)P2 and these proteins was found to regulate actin and lamellipodial membrane extensions. This study also implicated Lpd protein coordinated actin networks in the docking of pathogens such as Vaccinia virus and EPEC bacteria. Given the additional reports of these proteins participating in dorsal ruffling (Michael et al Curr Bio 2010) and invasion (Carmona et al Oncogene 2016), it comes as no surprise that they participate in phagophore formation and phagocytosis. These studies are referenced, but having this in mind does diminish the novelty of implicating Lpd and VASP in the phagocytic process, though it seems to be the first time this machinery was directly implicated in macrophage cells.

Specific Comments

Although the images and movies graphically demonstrate a PI(3,4)P2 enrichment on phagocytic structures , the authors could provide some additional images that include fluorescently tagged phagocytic cargo such as the erythrocytes used. The addition of a fluorescent marker or phase image would be especially beneficial in the experiments where a lack of cPHx-biosensor recruitment is seen to the docked phagocytic cargo. Otherwise, readers are left with the impression that perturbations such as INPP4B compromise docking and phagocytic cup formation altogether (Fig 2C)- which is perhaps the authors point? Make this clear? There has already been an implication for PI3K in the phagocytic process, perhaps verifying that initial formation/membrane extension stages of phagocytosis are impacted by targeting the D-4 position of PIP2 would be of interest?

Depletion of PI(3,4)P2 through the expression of the INPP4B phosphatase demonstrated a reduction in phagocytic uptake of red blood cells (Fig4). The readout for this assay relied upon what appears to be differential labeling of phagocytosed red blood cells, though there are examples of cargo that is supposedly inside the macrophages labeled in green? Perhaps the authors can reconcile this and make the methods more clear for this approach?

Fig4 demonstrates the PI(3,4)P2 dependent recruitment of Lamellipodin (Lpd) to the phagocytic cup, which is clear. Lpd is found to be necessary for effective phagocytic uptake in Fig 5. There is no blotting/qPCR data for the verification of Lpd knockdown shown? The authors demonstrate a co-localization of Lpd/VASP proteins at the phagocytic cup of these macrophages in Fig 6 and sequester VASP protein to the mitochondria with some ActA derived fusion proteins to functionally block phagocytosis. The functional interaction of Lpd/VASP is further explored with experiments utilizing Ena/VASP mutants in Fig7, demonstrating a dependence on this interaction to promote phagocytic uptake.

Significance

see above

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #1

Evidence, reproducibility and clarity

The authors use newly available probes to show that the phosphoinositide PI(3,4)P2 plays a previously undescribed role in FcgammaR-mediated phagocytosis. Using RAW macrophages, they show that PI(3,4)P2 is enriched at the plasma membrane and also at phagocytic cups internalizing IgG-opsonized sheep red blood cells. Pharmacological inhibition using wortmannin, and also expression of membrane-targeted INPP4B phosphatase showed that PI(3,4)P2 production depends on PI3 kinase activity. Further experiments using selective inhibitors showed that PI(3,4)5P2 is mainly derived by dephosphorylation of PI(3,4,5)P3, likely by multiple phosphatases such as SHIP1 or OCRL. Depletion of PI(3,4)P2 at the plasma membrane by INPP4B also resulted in strongly decreased internalization of red blood cells, although their attachment to macrophages seemed unaltered, pointing to defects in particle engulfment.

The authors then tested the potential role of lamellipodin, one of the few known PI(3,4)P2 specific effectors. Lamellipodin was found to be enriched at phagocytic cups, and this enrichment was shown to be dependent on the presence of PI(3,4)P2, by targeting of INPP4B to the plasma membrane. Macrophages depleted of lamellipodin by shRNA treatment showed reduced phagocytic efficiency and also aberrant phagocytic cup formation. As VASP is a known binding partner of lamellipodin and involved in actin polymerization, the authors next tested its potential involvement. Overexpression experiments showed that VASP colocalizes with lamellipodin at phagocytic cups. Sequestering of VASP at mitochondria through a respective construct containing the VASP binding site of ActA, together with a mitochondrial targeting sequence, showed that this also results in incompletely formed phagocytic cups and reduced phagocytic efficiency. Similar effects were observed upon expression of a lamellipodin construct with mutated binding sites for VASP.

Collectively, the authors propose that PI(3,4)P2 is localized produced at phagocytic cups through the sequential activity of PI3 kinase and PI5 phosphatase, that it recruits lamellipodin and its binding partner VASP, and that this cascade is necessary for proper phagocytic cup formation and closure and thus phagocytic capacity of cells. This is an interesting study that uncovers a novel role for PI(3,4)P2 in phagocytic cup formation and closure. It is very well controlled, and the claims of the study are supported by the presented data. Statistical analysis is sound.

Major comments:

1. The localization of VASP at phagocytic cups is only shown by overexpression of constructs. Endogenous staining of VASP should support this finding.
2. It is unclear whether the roles of PI(3,4)P2, lamellipodin, and VASP are restricted to FcgammaR-mediated phagocytosis. Their potential involvement in CR3-mediated phagocytosis should be discussed or addressed in a basic set of experiments.

Minor comments:

1. A very recent study (Körber and Faix, EJCB, 2022) describes the role of VASP in macroendocytosis in Dictyostelium. Specifically, VASP is found to be important for proper cup closure. The results are of direct importance to the current study and should be cited accordingly.
2. direct labelling of the figures would be helpful in assessing the manuscript

Significance

This study highlights the role of an underappreciated phospholipid in phagocytosis. It also describes for the first time a role for lamellipodin in formation of phagocytic cups and confirms the recent finding that also VASP is necessary for phagocytic cup closure. The paper should be of interest to researchers working on host-pathogen interaction, regulation of the actin cytoskeleton, and also to the general cell biological community

Reviewer´s expertise: Actin regulation Microtubule-based transport Adhesion, migration, invasion Phagocytosis

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

1. General Statements

We would like to thank the reviewers for their insightful and useful comments about our manuscript. Based on these comments and as outlined in our revision plan, we plan to strengthen our findings by performing new experiments and quantitative analyses. This particularly applies to our nanoscale (dSTORM) imaging dataset which was discussed by multiple reviewers.

We also appreciate the reviewers’ overall positive evaluation of the significance of our labeling method for the axon initial segment studies. With regards to this, we would like to highlight that this manuscript particularly addresses the labeling of “difficult-to-label” neuronal proteins, such as large ion channels and transmembrane proteins. Although we and another group have recently reported click labeling of neurofilament light chain (PMID: 35031604) and AMPAR regulatory proteins (PMID: 34795271) in primary neurons, both of these proteins have a small size between ~30-68 kDa and compared to larger ion channels/transmembrane proteins are “easier” to express in primary neurons. The novelty in the current manuscript is that we successfully applied this method for the labeling of large and spatially restricted AIS components, such as NF186 and Nav1.6 (186 and 260 kDa, respectively). As some of the reviewers also pointed out, the size and complexity of these proteins makes labeling of the AIS rather challenging. We also used our approach to study the localization of epilepsy-causing Nav1.6 variants and could exclude the retention in the cytoplasm as a possible cause of their loss of function. Finally, we improved the efficiency of genetic code expansion in primary neurons by developing AAV-based viral vectors. Although AAVs are routinely used for gene delivery to neurons, AAVs for click-based labeling need to encode multiple components of the orthogonal translational machinery for genetic code expansion. By trying different promoters and gene combinations, we developed several variants that enable high efficiency of the genetic code expansion in neurons. On their own, these findings will facilitate further genetic code expansion and click chemistry studies, beyond the labeling of the axon initial segment.

2. Description of the planned revisions

Reviewer #2

• On lines 107 and 108, the sentence "The C-terminal HA-tag allowed us to detect the full-length NF-186 protein by immunostaining it with an anti-HA antibody" would have a better place just after lines 104-105 " [...] we modified the previously described plasmid (Zhang et al., 1998) by moving the hemagglutinin (HA) tag from the N terminus to the C terminus".

We will modify the text as the reviewer suggested.

• Fig.2b: the AnkG staining looks substantially longer than that showed in c. However, the results on AIS length show no significant changes in between the groups. This is visually misleading, the authors should choose a picture for the WT construct that is representative of the data.

We thank the reviewer for bringing this up. We will replace the panel in Fig.2b with a more representative image of NF186 WT construct in the revised version of the manuscript.

• Line 238: what is the rationale behind choosing these cells? For example, have they been used in other studies for similar purposes? If so, please provide the reference.

We initially probed neuroblastoma ND7/23 which are commonly used for the electrophysiological recordings of recombinant Nav1.6 (PMID: 30615093, 22623668, 25874799, 27375106). Although we were able to record Na+ currents in those cells, only a small portion of channels was detected on the cell surface by microscopy (Suppl. Fig. 5a). As we discuss in the manuscript (lines 237-240), we then switched to N1E-115-1 cells in which we obtained a higher level of expression of the recombinant NaV1.6 channels on the cell surface (Suppl. Fig. 5b). These cells have also been previously used for the electrophysiological studies of voltage-gated sodium channels, including Nav1.6 (PMID: 8822380, 24077057). We will modify the text and include these references in the revised manuscript.

• Figure 3c, the authors omitted the comparison with the WT construct this time, as opposed to the neurofascin experiments. What is the reason?

As shown by others (PMID: 31900387) and us in this manuscript, one of the main issues with the expression of the recombinant NF186 in neurons was that overexpression led to mislocalization of NF186 in neuronal soma and processes. This was particularly true for WT construct and certain amber mutants (e.g. K809TAG). Based on previous reports (PMID: 31900387), we then tested a weak human neuron-specific enolase promoter. This reduced expression level and improved localization of NF186. However, since we still observed some neurons with mislocalized NF186 WT even with the enolase promoter, we found it important to quantitively compare the AIS length of WT construct and amber mutants to surrounding untransfected cells. On the other hand, since we did not have overexpression and mislocalization problem with Nav1.6 WT construct (all observed neurons have signal localizing in the AIS), we measured only the AIS length of the amber mutants. However, to avoid any confusion, we will also measure the AIS size of the neurons expressing Nav1.6 WT construct and compare it to surrounding cells and amber mutants. For this, we will need to perform new experiments and acquire new images. We will include the data in the revised manuscript.

• Fig. 4: why did the authors chose these cells for electrophysiology experiments and not neurons? Explain the rationale in the text or, alternatively, cite similar studies using the same tool.

Due to the branched neuronal processes which cause the space clamp problem in voltage clamp experiments with neurons, round and none-branching cells are frequently used to examine the biophysical properties of ion channels, including Nav1.6. By far, most of studies investigating the biophysical properties of NaV1.6 channels were performed in neuroblastoma cells e.g. ND7/23 and N1E-115-1 cells (PMID: 25874799; 25242737). We tested these two types of cells and found that N1E-115-1 cells supported higher expression level of the recombinant NaV1.6 channels on the cell surface than the ND7/23 cells (Suppl. Fig 5). Hence, N1E-115-1 were more suitable to get robust and reliable recordings (as we also discuss above in the response to reviewer’s comment). We will clarify this in the revised manuscript.

• Fig.4, biophysical properties: did the authors find differences in passive properties? Measures of resting potential, membrane resistance and cell capacitance should be reported.

Passive properties such as resting membrane potential and membrane resistance are important functional features in neurons measured in current clamp experiments, but not applicable for ND7/23 and N1E-115-1 cells used in our voltage clamp experiments. To measure the Na+ current mediated by WT or mutant NaV1.6 channels expressed in N1E-115-1 cells, the endogenous Na+ channels were blocked by tetrodotoxin and the endogenous K+ channels were blocked by tetraethylammonium chloride, CsCl and CsF in extracellular and intracellular solutions. Under these conditions, resting potential and membrane resistance are not relevant for experiments. Cell capacitance reflects the size of the cell surface area, which can affect the number of channels expressed on the cell surface. To eliminate the effect of different cell sizes, Na+ current densities normalized by cell capacitances were used in our experiments. We will report on these values in the revised manuscript.

• Fig 4, STORM images. The periodic distribution of the dots should be enhanced with some sort of arrows or lines, for the non-specialist audience.

Based on the comments from multiple reviewers, we plan to obtain additional dSTORM images of the neurons expressing recombinant Nav1.6 WT or amber mutants. We also intend to improve the visualization of these results by updating/modifying existing figures and including quantitative data.

• Line 374: rat or mouse primary neurons?

We are here referring to both, rat and mouse neurons. The images shown in Fig. 06 and Suppl. Fig. 08 were obtained from rat cortical neurons expressing Nav1.6 or fluorescent reporter. However, we were also able to successfully transduce mouse neurons with AAV92A carrying orthogonal translational machinery (data not shown). We will clarify this in the revised manuscript.

**Referees cross-commenting**

I fully agree with the following remarks from Reviewers #3, #4 and #5. This is a point that I have raised in my report too. The authors need better images to show the periodicity we visualization, and a quantification would be of great benefit to support the claim with numbers (and how these compare to similar studies in the literature):

R3: 2. For the dSTORM analysis of the tagged Nav1.6 protein, I also cannot tell there is periodic organization from the image directly. Some analysis is needed there. R4: 2."As there was no obvious difference in the nanoscale organization of the NaV1.6WT 317 -HA or NaV1.6TAG 318 -HA channels (Fig. 4. e-g), these experiments confirmed that the NaV1.6 overexpression, TCO*A319 Lys incorporation, and click labeling did not affect the nanoscale periodic organization of the sodium channels in the AIS." It is clearly noticeable that for WT, the spot density is more compared to the other two mutants. Why is that so? Using cluster analysis, one can quantify spot density and discuss nanoscale organization quantitatively. The author should quantify the periodicity and compare it among different variants and with previous reports. R5: 3. The authors claim that there was no obvious difference in the nanoscale organization of the NaV1.6WT 317 -HA or NaV1.6TAG 318 -HA channels (Fig. 4. e-g), but it is hard to conclude this without any quantification and statistical analysis. Sodium channels have been shown to be associated with the membrane-associated periodic skeleton structures in neurons and average autocorrelation analysis has been developed to quantify the degree of periodicity of such structural organizations (Han et al. PNAS 114(32)E6678-E6685, 2017). The authors should use this approach to quantify and compare the average autocorrelation amplitudes.

As we outlined in our responses to the individual reviewers’ comments below, we will address these questions by performing new experiments and quantifications.

I also agree with these comments from Reviewers #3 and #5:

R3: 4. It is unclear, for all the presented data, whether all the cells are collected from a single biological replicate or from multiple replicates. At least 2-3 replicates are needed to see the reproducibility in terms of labeling efficiency, and other related conclusions. R5: 1. The authors should indicate how many replicates were performed and how many cells were analyzed for each experiment.

We thank the reviewers for bringing this up. By mistake, we omitted this important information. We will include it in the revised manuscript, but we would like to highlight here that each experiment was repeated at least 3 times.

Reviewer #3

1. There is some patch-like background from the 488 channel from the click reaction, some of which have very as strong signal as the staining on the neurons. What is the potential cause for this? With immunostaining on HA, the background doesn't affect too much on the image data interpretation. However, the major goal of this method development is to use it in live-cell without immunostaining. Without another reference, the high background might cause issues in data interpretation. Can the author also suggest way to avoid or lower this in the discussion?

We thank the reviewer for bringing this up. We have occasionally observed patch-like background in what appears to be the cell debris. Such dead cells do not have an intact cell membrane and therefore can absorb cell-impermeable ATTO488-tetrazine dye during click labeling. This kind of background is also present in the control neurons transfected with the WT Nav1.6, which suggests that it originates from the UAA and tetrazine-dye accumulations. Additionally, since these patches are not visible with the immunostaining, they do not contain our protein of interest, which further confirms that they contain only dye and UAA accumulations. Depending on the neuron prep/quality before and after transfections, the presence of these patches is more or less obvious. However, despite the background we did not have problems identifying AIS during live cell imaging. Especially when overall neuronal health is optimal after transfections, AIS can easily be distinguished from patches that are positioned outside of labeled neurons. We will investigate this further and discuss it in the revised manuscript.

1. For the dSTORM analysis of the tagged Nav1.6 protein, I also cannot tell there is periodic organization from the image directly. Some analysis is needed there.

We will address this in the revised manuscript by performing additional experiments and quantifications. We also wrote a detailed answer below, in the response to the other reviewers.

1. The authors use the AIS length as a parameter to evaluate the function of the clickable mutant of NF186, and using patch clamp for functional validation of the clickable mutant of Nav1.6. In both cases, the comparison is done between the mutant and the WT construct, but both in transfected cell and exogenously expressed. It's also worth comparing with untransfected cells as the true native situation.

We agree with the reviewer that it is important to compare transfected cells with untransfected cells. As the reviewer points out, we have already performed some of these comparisons. When it comes to the NF186, we used the AIS length as a parameter to estimate if the expression of clickable mutant affected the AIS structure. As we show in the Fig. 02, we co-immunostained neurons transfected with NF186-HA WT or TAG constructs. We used HA antibody to detect neurons expressing NF186, while the ankG was used as a marker of the AIS length. To check if the AIS length of transfected cells is affected, we compared the length of transfected cells (expressing NF186, HA+) to surrounding untransfected cells (HA-). When it comes to the Nav1.6, we also compared the AIS length of cells expressing Nav1.6 (HA+) to surrounding untransfected cells (HA-). Similarly to the experiments with NF186, this allowed us to check if the expression of the recombinant Nav1.6 affect the AIS structure. What is missing is the comparison with untransfected conditions (i.e. neurons that are simply stained with ankG). We assume that is what the reviewer is referring to? We will also include these data in the revised manuscript. Furthermore, since we introduced a labeling modification in NaV1.6, we wanted to check if such modification would affect its function. To do so, as routinely done in the field (PMID: 25874799), we rendered the WT and TAG channels TTX-resistant and recorded only recombinant Na+ currents in neuroblastoma cells in the presence of TTX. Perhaps we misunderstand the reviewer’s comment, but in this regard measurements of untransfected cells are not relevant since they would not allow us to compare WT and TAG mutants.

1. It is unclear, for all the presented data, whether all the cells are collected from a single biological replicate or from multiple replicates. At least 2-3 replicates are needed to see the reproducibility in terms of labeling efficiency, and other related conclusions.

We thank the reviewer for the observation. By mistake, we omitted this important information. We will include in the revised version of the manuscript. We would like to highlight here that each experiment was repeated at least 3 times.

Reviewer #4

1."Confocal microscopy revealed that the hNSE promoter lowered the WT and clickable NF186-HA expression levels and consequently improved the localization of these proteins." Is the lower expression level a measure of localization improvement? How does the author conclude that the localization has improved?

Previous report (PMID: 31900387) suggested that the overexpression of the recombinant WT NF186 can affect its trafficking, leading to the NF186 mislocalization. We observed the same in our experiments with CMV NF186 (in particular for NF186 WT). Hence, based on the PMID: 31900387 we probed weak neuron specific enolase promoter. Since the WT was the most problematic in terms of the ectopic expression, we checked if AIS localization was improved with enolase promoter for this construct. To this aim, we counted number of neurons that with mislocalized signal or with the signal in the AIS for both, CMV and enolase promoter. We could observe that number of neurons with mislocalized signal was lower for enolase promoter. Since there were more neurons with the AIS-specific signal when NF186 was expressed from enolase promoter compared to CMV, we concluded that enolase promoter lowered expression and improved localization of the NF186. Therefore, we used enolase promoter for click labeling of NF186 amber mutants. We will include the results of this analysis in the revised version of the manuscript.

2."As there was no obvious difference in the nanoscale organization of the NaV1.6WT 317 -HA or NaV1.6TAG 318 -HA channels (Fig. 4. e-g), these experiments confirmed that the NaV1.6 overexpression, TCO*A319 Lys incorporation, and click labeling did not affect the nanoscale periodic organization of the sodium channels in the AIS." It is clearly noticeable that for WT, the spot density is more compared to the other two mutants. Why is that so? Using cluster analysis, one can quantify spot density and discuss nanoscale organization quantitatively. The author should quantify the periodicity and compare it among different variants and with previous reports.

We thank the reviewer for these suggestions. We will address these remarks by performing additional new experiments and quantifications. The difference in the level of the expression of the recombinant Nav1.6 might explain differences in the spot density for WT vs. TAG clickable mutants. However, as the reviewer suggested quantitative analysis is needed to address these concerns. We also intend to quantify the periodicity and compare it among different variants and with previous reports. It is just important to note that in the current version of the manuscript we looked at the nanoscale organization of the subset of Nav1.6 channels. The reason being that we used anti-HA antibody which will only detect our recombinant protein which got incorporated into the AIS and not the endogenous Nav1.6.

Minor comments

1."Although NF186K809TAG 158 -HA (Supplementary Fig. 4) showed bright click labeling, we excluded it from the analysis due to its frequent ectopic expression along the distal axon." How frequently is this bright click labeling observed for this mutation? Is it not observed for other mutations at all? The authors should state this point clearly with some statistics.

We are not sure what is the exact question from the reviewer. If we understand it correctly, the reviewer is asking us to quantify how frequent was the ectopic expression of this amber mutant compared to other mutants? And not the click labeling (as written in their original comment), since click labeling was observed for all the mutants independently of their ectopic expression?

2."Immunostaining with anti-HA antibody revealed that the expression of NaV1.6WT 239 -HA on the membrane of the N1E-115-1 cells was higher than on the ND7/23 cells (Supplementary Fig. 5a-c). However, click labeling of both NaV1.6K1425TAG 240 -HA and NaV1.6K1546TAG 241 -HA with ATTO488-tz was not successful (Supplementary fig. 5d) indicating insufficient expression of the clickable constructs." Is this due to insufficient expression level or accessibility? The author should make this statement clear.

We thank the reviewer for bringing this up. We will clarify this in the revised version of the manuscript. We believe that the click labeling of the K1546TAG mutant in N1E-115-1 cells is absent due to the insufficient expression of the channels on the membrane, since this mutant was successfully labeled in the primary neurons that represent more native environment and where Nav1.6 form high-density clusters. K1425TAG mutant is not labeled due to the insufficient expression on the membrane in N1E-115-1 cells as well. However, since this mutant is also poorly labeled in primary neurons, we can speculate that K1425TAG position might be less accessible for the tetrazine-dye compared to K1546TAG. To further support our claim that due to the insufficient expression click labeling is low/absent in neuronal cells, we can use NF186 as an additional example. When NF186 was expressed from strong CMV promoter, we observed click labeling for all the mutants in ND7/23 cells (Suppl. Fig.01). However, when CMV was replaced with neuron specific enolase promoter, the expression was of NF186 was substantially lower in ND7/23 cells and click labeling was absent (data not shown). We will clarify this in the revised manuscript.

1. Authors should clearly state the drift correction procedure of 3D STORM data. What are the localization precision and photon count for 3D STORM experiments?

We processed 3D dSTORM data in NIS-elements AR software. We used the automatic drift correction from the NIS-elements software that is based on the autocorrelation. We will provide further and updated information in the revised manuscript, including the localization precision and photon count for the new dSTORM images.

1. "Click labeling of NaV1.6 channels in living primary neurons" What kind of primary neurons have been used for click labeling of NaV1.6 channels? Is there any specific reason why authors have chosen cortical neurons for labeling NF186? Does this labeling strategy depend on primary neuron type?

For the establishment and click labeling of Nav1.6 we used primary rat cortical neurons (Fig. 03, Fig. 06). The same neuronal type has been used for click labeling of NF186 (Fig. 02). We established labeling of the AIS components in cortical neurons because we use those routinely in the laboratory. However, this labeling strategy does not depend on the neuronal type. As we show in Fig. 05, to study localization of the loss-of-function pathogenic Nav1.6 variants we used mouse hippocampal neurons. The reason for this is that in previous study the same neuronal type was used to characterize these two mutations (lines 361-362). This demonstrates nicely that method can be easily transferred to any neuronal type. Furthermore, we were also able to label Nav1.6 and NF186 in mouse cortical neurons (data are not shown in the manuscript). We will clarify this in the revised manuscript.

Reviewer #5

1.Throughout the manuscript, only one representative image containing one AIG is shown for each condition without statistics and quantifications, so the conclusions are not sufficiently convincing. For example, in Fig. 1b, c, e; Fig. 2b,c,d,e; Fig. 3b,c,d,e ; Fig. 5c; and Supplementary Fig.1-6, the authors should quantify the average fluorescence intensities both for HA immunostaining and ATTO488-tz labeling in different conditions, as well as the labeling ratios (fluorescence intensity ratios between ATTO488 and AF647/AF555) . Without statistics and quantifications, it is unclear whether there is any significant difference between the constructs with different TAG positions, or between different transfection methods (e.g., lipofectamine 2000 vs 3000).

We agree with the reviewer that the quantitative analysis is important and we will provide more quantitative data in the revised manuscript. At the same time, we are a bit confused by this comment which seems to refer to missing quantifications in one of the schemes (Fig. 1) and overlooks existing quantifications (e.g. quantitative analysis of the data set from Fig. 5c is shown in Fig. 5d). However, as suggested by the reviewer and to strengthen our data, in addition to the quantifications already provided in the manuscript (e.g. Fig. 2d: AIS length of NF186TAG constructs; Fig. 3f: AIS length of Nav1.6 TAG constructs; Fig. 5d: click-labeling intensity of LOF mutants), we intend to quantify the differences between labeling ratios of different mutants and transfection methods. When it comes to the different transfection methods, some data is already provided in the manuscript (e.g. we counted number of transfected versus transduced neurons) but we intend to clarify and expand on this in the revised manuscript.

1. The only quantification done was for the average AIS length, but the statistical tests should be performed between different conditions and the corresponding P values should be provided. It seems that the transfected neurons generally have a longer AIS length than the transfected neurons (Fig. 2d and 3f). Could the authors provide an explanation for this?

We are a bit confused by the first part of this comment. We measured the AIS lengths of NF186 WT or NF186 TAG as well as Nav1.6 TAG and compared it to the AIS lengths of surrounding untransfected cells (Fig. 2d and Fig.03f). In addition, we compared the AIS lengths of the NF186 WT and TAG to each other, and Nav1.6 TAG to each other. To analyze the differences, we performed statistical tests and provided the corresponding p values in the figure legends (Fig. 02 and 03). Further details on the statistical analysis are provided in supplementary tables (Suppl. table 01 and 02). Regarding the 2nd question, we have also noticed that the AIS lengths of transfected neurons appear longer than those of untransfected cells. This seems to be more pronounced in the case of NF186 which is expressed at the higher level compared to the Nav1.6. The appearance of slightly longer AIS is most likely the consequence of the fact that recombinant constructs are overexpressed in the neurons that express endogenous NF186 and Nav1.6. However, this difference in the AIS length is not significant to the controls. We will discuss this further in the revised manuscript.

1. The authors claim that there was no obvious difference in the nanoscale organization of the NaV1.6WT 317 -HA or NaV1.6TAG 318 -HA channels (Fig. 4. e-g), but it is hard to conclude this without any quantification and statistical analysis. Sodium channels have been shown to be associated with the membrane-associated periodic skeleton structures in neurons and average autocorrelation analysis has been developed to quantify the degree of periodicity of such structural organizations (Han et al. PNAS 114(32)E6678-E6685, 2017). The authors should use this approach to quantify and compare the average autocorrelation amplitudes.

We are thankful to the reviewer for suggestions on how to quantify the periodicity of recombinant sodium channels and how to more accurately compare WT and TAG mutants at the nanoscale level. We will perform additional experiments and analysis in order to address the concerns of this and other reviewers.

1. The authors should also obtain dSTORM images for the click labeled neurons to demonstrate if the click labeling method would provide sufficient labeling efficiency for dSTORM, compared to immunostaining (HA and Ankyrin G immunostaining).

We would like to thank the reviewer for this suggestion. We have already shown in our previous work that STED can be performed with click labeled neurons (PMID: 35031604). When it comes to this manuscript and AIS labeling, we have already obtained preliminary dSTORM images of click-labeled NF186. Since the expression of Nav1.6 is lower compared to NF186, the labeling is also less bright and dSTORM is a bit more challenging. To try to overcome this issue, in addition to dSTORM of click-labeled Nav1.6, we are planning to try click-PAINT (PMID: 27804198). Click-PAINT has been used for super-resolution imaging of less abundant targets in cells and could possibly allow super-resolution imaging of Nav1.6. We will report on these new experiments in the revised version of the manuscript.

1. It seems that the click labeling has a off-target/background labeling in the soma of the neuron (see Fig. 3c,d. Could the authors quantify and determine the sources of such off-target labeling?

We thank the reviewer for pointing this out. We will clarify this in the revised manuscript, but by looking at the other examples from our dataset it appears to us that this background is present in WT constructs as well. In the current version of the manuscript, this is not clear since the WT image that is shown in the Fig. 03b is a single plane confocal image. Therefore, we will replace it in the revised manuscript with a z-stack in which the presence of the background is more obvious (due to the maximum intensity projection). In addition, we will conduct additional control experiments to clarify this.

Minor comments:

1. The authors should indicate how many replicates were performed and how many cells were analyzed for each experiment.

We thank the reviewer for bringing this up. By mistake, we omitted this important information. We will include this information in the revised manuscript, but we would like to highlight here that each experiment was repeated at least 3 times.

1. The display range (i.e., intensity scale bar) was indicated only for a small portion of the fluorescence images. It is better to be consistent and show the display range for all images presented.

We will include intensity scale bars in all the images in the revised version of the manuscript.

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

Not applicable.

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

Reviewer #3, comment #5. One application presented in this manuscript is to evaluate the effect of epilepsy-causing mutations of Nav1.6. By comparing the intensity of ATTO488, the result suggests that there is no significant impact of these mutations on membrane tracking. I am wondering if the author should study the membrane tracking by also looking at the diffusion in live-cell with the labeling method. The comparison of the intensity only can be achieved by just immunostaining. It doesn't really demonstrate the benefit of live-cell labeling and imaging with the presented method.

Generally speaking, one of the advantages of click labeling is its compatibility with live cell labeling. As the reviewer also points out, this is especially useful for live-cell imaging but is not limited to it. In addition, click labeling allows selective labeling of membrane population of Nav1.6 in living neurons. We took advantage of this and used cell-impermeable dyes to label unnatural amino acids incorporated into extracellular part of Nav1.6 (Scheme 03a). On the contrary, HA tag that allows immunodetection of recombinant Nav1.6 is added to the intracellular C terminus. Hence, by anti-HA immunostaining total (intra- and extracellular) epilepsy-causing Nav1.6 channel population will be detected. That is why in this case live-cell click labeling was advantageous compared to the conventional immunostaining. We will clarify this in the revised manuscript. In addition, we would like to note that when we started the experiments with the epilepsy-causing mutations, we wanted to a) check if they are present on the membrane and b) depending on the outcome of those experiments follow the trafficking of these LOF Nav1.6 mutants. Since patch clamp recordings of pathogenic Nav1.6 showed loss of Na+ currents, we at first assumed that they are not properly expressed on the membrane. However, our click labeling showed that the pathogenic channels were detected at the AIS membrane despite the loss of Na+ currents. This was also somewhat surprising to us and we would love to investigate this further. We also appreciate the reviewer’s suggestion in this regard and we hope to be able to use all the advantages of our labeling approach in our follow-up studies. However, keeping in mind time and resources limitations, live-cell trafficking study might be beyond the scope of this revision.

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 #5

Evidence, reproducibility and clarity

Summary:

In this manuscript, Nevena Stajković et al. present a method for live labeling of the proteins localized at the axon initial segment (AIS) of cultured neurons using unnatural amino acids (UAAs) carrying strained alkenes and click chemistry. Using this method, the authors showed the successful labeling of two AIS-localized proteins, the 186 kDa isoform of neurofascin (NF186) and the 260 kDa voltage-gated sodium channel (NaV1.6). The authors also showed the transduction of neurons using adeno-associated viruses (AAVs) had higher efficiency than transfection by lipofectamine in delivering the vectors expressing required components for the click labeling.

Major comments:

1. Throughout the manuscript, only one representative image containing one AIG is shown for each condition without statistics and quantifications, so the conclusions are not sufficiently convincing. For example, in Fig. 1b, c, e; Fig. 2b,c,d,e; Fig. 3b,c,d,e ; Fig. 5c; and Supplementary Fig.1-6, the authors should quantify the average fluorescence intensities both for HA immunostaining and ATTO488-tz labeling in different conditions, as well as the labeling ratios (fluorescence intensity ratios between ATTO488 and AF647/AF555) . Without statistics and quantifications, it is unclear whether there is any significant difference between the constructs with different TAG positions, or between different transfection methods (e.g., lipofectamine 2000 vs 3000).
2. The only quantification done was for the average AIS length, but the statistical tests should be preformed between different conditions and the corresponding P values should be provided. It seems that the transfected neurons generally have a longer AIS length than the transfected neurons (Fig. 2d and 3f). Could the authors provide an explanation for this?
3. The authors claim that there was no obvious difference in the nanoscale organization of the NaV1.6WT 317 -HA or NaV1.6TAG 318 -HA channels (Fig. 4. e-g), but it is hard to conclude this without any quantification and statistical analysis. Sodium channels have been shown to be associated with the membrane-associated periodic skeleton structures in neurons and average autocorrelation analysis has been developed to quantify the degree of periodicity of such structural organizations (Han et al. PNAS 114(32)E6678-E6685, 2017). The authors should use this approach to quantify and compare the average autocorrelation amplitudes.
4. The authors should also obtain dSTORM images for the click labeled neurons to demonstrate if the click labeling method would provide sufficient labeling efficiency for dSTORM, compared to immunostaining (HA and Ankyrin G immunostaining).
5. It seems that the click labeling has a off-target/background labeling in the soma of the neuron ( see Fig. 3c,d. Could the authors quantify and determine the sources of such off-target labeling?

Minor comments:

1. The authors should indicate how many replicates were performed and how many cells were analyzed for each experiment.
2. The display range (i.e., intensity scale bar) was indicated only for a small portion of the fluorescence images. It is better to be consistent and show the display range for all images presented.

Significance

Unnatural amino acid (UAA)-based minimal tags for live-cell protein labeling in mammalian cells were invented about ten years ago (Lang et al., 2012b, Lang et al., 2012a, Nikic et al., 2014, Plass et al., 2012, Uttamapinant et al., 2015), and these authors recently introduced this labeling method to label live cultured neurons (Arsić et al., 2022). Therefore, it is unclear whether the method present in this manuscript has any significant advance compared to the Arsić et al. paper, given that the major difference between the two papers is that in the current manuscript, AIS localized proteins were labeled, whereas in the Arsić et al. paper, neurofilaments were labeled in the neurons. Therefore, the method presented in the current manuscript does not provide much novelty or technical advance compared to what has been described in the Arsić et al. paper.

My expertis is super-resolution flurescence imaging, cell labeling methods, and neurobiology.

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 #4

Evidence, reproducibility and clarity

The manuscript demonstrates a novel method of labeling two large components of the initial axon segment, neurofascin (NF186) and Nav1.6 using unnatural amino acids and click chemistry in live cells. They have applied their method for epilepsy causing two Nav1.6 variants without affecting their functionality. Since these proteins are larger in size, selecting the labeling sites and transfection efficiency become critical factors. They have targeted different lysine sites and shown the best performing labeling site. Also, they have developed a viral vector to improve transfection efficiency.

The experiments are well designed, and the manuscript is nicely written. In my opinion, the manuscript can be accepted, but the author should address the following comments.

Major comments

1. "Confocal microscopy revealed that the hNSE promoter lowered the WT and clickable NF186-HA expression levels and consequently improved the localization of these proteins." Is the lower expression level a measure of localization improvement? How does the author conclude that the localization has improved?
2. "As there was no obvious difference in the nanoscale organization of the NaV1.6WT 317 -HA or NaV1.6TAG 318 -HA channels (Fig. 4. e-g), these experiments confirmed that the NaV1.6 overexpression, TCO*A319 Lys incorporation, and click labeling did not affect the nanoscale periodic organization of the sodium channels in the AIS." It is clearly noticeable that for WT, the spot density is more compared to the other two mutants. Why is that so? Using cluster analysis, one can quantify spot density and discuss nanoscale organization quantitatively. The author should quantify the periodicity and compare it among different variants and with previous reports.

Minor comments

1. "Although NF186K809TAG 158 -HA (Supplementary Fig. 4) showed bright click labeling, we excluded it from the analysis due to its frequent ectopic expression along the distal axon." How frequently is this bright click labeling observed for this mutation? Is it not observed for other mutations at all? The authors should state this point clearly with some statistics.
2. "Immunostaining with anti-HA antibody revealed that the expression of NaV1.6WT 239 -HA on the membrane of the N1E-115-1 cells was higher than on the ND7/23 cells (Supplementary Fig. 5a-c). However, click labeling of both NaV1.6K1425TAG 240 -HA and NaV1.6K1546TAG 241 -HA with ATTO488-tz was not successful (Supplementary fig. 5d) indicating insufficient expression of the clickable constructs." Is this due to insufficient expression level or accessibility? The author should make this statement clear.
3. Authors should clearly state the drift correction procedure of 3D STORM data. What are the localization precision and photon count for 3D STORM experiments?
4. "Click labeling of NaV1.6 channels in living primary neurons" What kind of primary neurons have been used for click labeling of NaV1.6 channels? Is there any specific reason why authors have chosen cortical neurons for labeling NF186? Does this labeling strategy depend on primary neuron type?

Significance

Although the use of unnatural amino acids and click chemistry for labelling has been shown before from the same group, labelling large proteins, especially ion channels, without affecting their function is always challenging because of the accessibility of the labelling site as well as poor transfection efficiency. Here, they have selected two such large essential proteins: NF186 and Nav1.6, which are associated with epilepsy, and developed a method for fluorophore labelling with minimal perturbation. Other approaches namely using fluorescent proteins, biotin-streptavidin chemistry and halo-tag have been reported before to label these proteins, but these have a strong impact on their mislocalisation and perturbing their functionality. Therefore, this method will be of great importance in the field of studying these proteins.

Expertise: Live-cell confocal and multi-photon microscopy imaging, Super-resolution microscopy imaging, Live-cell labelling, and Amyloid aggregations

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

The manuscript by Stajkovic et al the describes step-wise generation and validation of the fluorescent labeling of NF186 and Nav1.6 in primary neurons by non-natural amino acid and click chemistry. For each protein of interest, the authors started by generating constructs carrying amber codon at different positions, and then selected for the best construct(s) by judging (1) the labeling efficiency, (2) whether the particular labeling position affect the function of the protein, and (3) whether the labeled protein shows any mislocalization. During the trouble shooting process, the authors also introduced adeno-associated viral (AAV) vectors for more efficiently delivering constructs into the cells. The method described in the manuscript could become a reference for researchers who aim to label similar neuronal proteins.

Specific comments:

1. There is some patch-like background from the 488 channel from the click reaction, some of which have very as strong signal as the staining on the neurons. What is the potential cause for this? With immunostaining on HA, the background doesn't affect too much on the image data interpretation. However, the major goal of this method development is to use it in live-cell without immunostaining. Without another reference, the high background might cause issues in data interpretation. Can the author also suggest way to avoid or lower this in the discussion?
2. For the dSTORM analysis of the tagged Nav1.6 protein, I also cannot tell there is periodic organization from the image directly. Some analysis is needed there.
3. The authors use the AIS length as a parameter to evaluate the function of the clickable mutant of NF186, and using patch clamp for functional validation of the clickable mutant of Nav1.6. In both cases, the comparison is done between the mutant and the WT construct, but both in transfected cell and exogenously expressed. It's also worth comparing with untransfected cells as the true native situation.
4. It is unclear, for all the presented data, whether all the cells are collected from a single biological replicate or from multiple replicates. At least 2-3 replicates are needed to see the reproducibility in terms of labeling efficiency, and other related conclusions.
5. One application presented in this manuscript is to evaluate the effect of epilepsy-causing mutations of Nav1.6. By comparing the intensity of ATTO488, the result suggests that there is no significant impact of these mutations on membrane tracking. I am wondering if the author should study the membrane tracking by also looking at the diffusion in live-cell with the labeling method. The comparison of the intensity only can be achieved by just immunostaining. It doesn't really demonstrate the benefit of live-cell labeling and imaging with the presented method.

Significance

The data itself is mostly convincing, however, I do not see much novelty from this manuscript. Both the labeling method using non-natural amino acid and click chemistry and AAV delivery are established. However, I can see that for research groups who specifically interested in studying these two proteins or proteins closed related, the results from this manuscript could be of direct help.

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:

This study proposes a novel tool for AIS live and fixed labelling based on biorthogonal click chemistry. Stajkovic and colleagues used this method to specifically label the AIS proteins NF186 and Nav1.6 of mouse and rat neurons, and did a thorough process of optimization to get convincing results. The authors considered different promoters, transfection strategies and the use of AAVs to get the most efficient labelling strategy for both proteins. They have also gone through a strong validation process based on transfection efficiency, quality of staining, potential effects on AIS length and nanostructure, and electrophysiological properties. Finally, Stajkovic and colleagues used this tool to study how two epilepsy-causing Nav1.6 mutant variants affect AIS function, providing interesting data to the understanding of this pathology. In summary, this method convincingly overcomes some well-described issues associated with pre-existing AIS live cell labelling tools by being minimally "invasive" to the proteins of interest. Besides the scientific content, another strong point of this article is the clarity of the manuscript and the figures: the presence of schematics (i.e. Fig. 1) and the detailed description of experiments and results will help non-specialist readers to follow the study. I strongly recommend this article for journal publication.

Major comments:

I have no major comments

Minor comments:

I have some minor comments:

• On lines 107 and 108, the sentence "The C-terminal HA-tag allowed us to detect the full-length NF-186 protein by immunostaining it with an anti-HA antibody" would have a better place just after lines 104-105 " [...] we modified the previously described plasmid (Zhang et al., 1998) by moving the hemagglutinin (HA) tag from the N terminus to the C terminus".
• Fig.2b: the AnkG staining looks substantially longer than that showed in c. However, the results on AIS length show no significant changes in between the groups. This is visually misleading, the authors should choose a picture for the WT construct that is representative of the data.
• Line 238: what is the rationale behind choosing these cells? For example, have they been used in other studies for similar purposes? If so, please provide the reference.
• Figure 3c, the authors omitted the comparison with the WT construct this time, as opposed to the neurofascin experiments. What is the reason?
• Fig. 4: why did the authors chose these cells for electrophysiology experiments and not neurons? Explain the rationale in the text or, alternatively, cite similar studies using the same tool.
• Fig.4, biophysical properties: did the authors find differences in passive properties? Measures of resting potential, membrane resistance and cell capacitance should be reported.
• Fig 4, STORM images. The periodic distribution of the dots should be enhanced with some sort of arrows or lines, for the non-specialist audience.
• Line 374: rat or mouse primary neurons?

Referees cross-commenting

I fully agree with the following remarks from Reviewers #3, #4 and #5. This is a point that I have raised in my report too. The authors need better images to show the periodicity visualization, and a quantification would be of great benefit to support the claim with numbers (and how these compare to similar studies in the literature):

R3: 2. For the dSTORM analysis of the tagged Nav1.6 protein, I also cannot tell there is periodic organization from the image directly. Some analysis is needed there. R4: 2."As there was no obvious difference in the nanoscale organization of the NaV1.6WT 317 -HA or NaV1.6TAG 318 -HA channels (Fig. 4. e-g), these experiments confirmed that the NaV1.6 overexpression, TCO*A319 Lys incorporation, and click labeling did not affect the nanoscale periodic organization of the sodium channels in the AIS." It is clearly noticeable that for WT, the spot density is more compared to the other two mutants. Why is that so? Using cluster analysis, one can quantify spot density and discuss nanoscale organization quantitatively. The author should quantify the periodicity and compare it among different variants and with previous reports. R5: 3. The authors claim that there was no obvious difference in the nanoscale organization of the NaV1.6WT 317 -HA or NaV1.6TAG 318 -HA channels (Fig. 4. e-g), but it is hard to conclude this without any quantification and statistical analysis. Sodium channels have been shown to be associated with the membrane-associated periodic skeleton structures in neurons and average autocorrelation analysis has been developed to quantify the degree of periodicity of such structural organizations (Han et al. PNAS 114(32)E6678-E6685, 2017). The authors should use this approach to quantify and compare the average autocorrelation amplitudes.

I also agree with these comments from Reviewers #3 and #5:

R3: 4. It is unclear, for all the presented data, whether all the cells are collected from a single biological replicate or from multiple replicates. At least 2-3 replicates are needed to see the reproducibility in terms of labeling efficiency, and other related conclusions. R5: 1. The authors should indicate how many replicates were performed and how many cells were analyzed for each experiment.

Significance

The proposed tool in this article represents a big step forward in the field of AIS live cell imaging. As stated by the authors in the introduction, previous studies have described methods based on tagging fluorescent proteins to the protein of interest or labelling the extracellular part of proteins with antibodies. The same studies reported several issues: the interference with important domains of the protein due to the size and the position of the tag in the case of fluorescent proteins (Dumitrescu et al., 2016, PMID: 27932952; Dzhashiashvili et al., 2007, PMID: 17548513), or the failure to report plasticity changes in the AIS in the case of antibodies (Dumitrescu et al., 2016, PMID: 27932952). This tool can be useful for research teams aiming to understand, for example, the live development of the AIS or understanding the trafficking of its proteins. The authors have applied this method to two transmembrane proteins (NF186 and Nav1.6), but as they state in their discussion, it will be useful to tag other candidates, including cytoplasmic proteins. One of the main problems of immunocytochemistry is to find the right antibody to detect your protein. Sometimes, absence of proof is not proof of absence: just because the protein is not detected via immunostaining does not mean that the protein is not expressed there. This tool offers an alternative to these challenging scenarios.

My expertise keywords: axon initial segment, neuronal polarity, axon biology, super resolution microscopy.

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

AIS organization is highly complex and unique and AIS labeling in living cells has been problematic. In this study, authors comprehensively searched and optimized live cell markers for AIS.The main advantage of the labeling approach is that the small fluorescent dye is directly attached to the proteins of interest. Therefore, the protein of interest is modified in a minimally invasive way.

Significance

I was first wondering whether tool development is enough important to be published as such but here, the development and testing of the constructs, whether they affect cell functionality, is very comprehensive. This all makes this tool very useful for other researchers. I can take the method directly to use and I don't need to do all testing by myself. I feel that this tool development takes field forward.

Own expertise: I have personally gone through the AIS live cell labeling problems and therefore I can easily appreciate the work done here.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

We are sincerely grateful to the reviewers for several key comments that led us to correct some mistakes and better appreciate how to put our findings in the context of recently published data. These changes undoubtedly improved the manuscript.

Many other reviewer comments seem to equate chaperone binding with a functional chaperone role in de novo folding. These are not the same. Cytosolic chaperones presumably “sample” nearly every protein that is synthesized by cytoplasmic ribosomes. This does not mean that every such protein would misfold if even one of those chaperones failed to bind it. If we want to understand what chaperone mutations might cause human disease due to septin misfolding, for example, it will not be enough to catalog all the chaperones that bind septins. We have already done that. What will help is to understand which chaperones make functional contributions to septin folding and complex assembly. Our study is the first to experimentally address chaperone roles in de novo septin folding, period. We take responsibility for not being sufficiently clear about the goals of our work, and, to emphasize these points, we added one sentence to the Introduction and revised another.

Another consistent criticism was that the use of the E. coli system, both in vivo and in vitro, limited our ability to gain insight into the folding of septins in eukaryotic cells and led to a “tessellated view”. For example, reviewers claimed that our model about translation elongation rates for Cdc12 were “based mainly on the E. coli system and bioinformatics analysis”. We disagree with this interpretation. Key evidence in support of our model come from published data in yeast, specifically the much higher density of ribosomes on Cdc12 and the accumulation of ribosomes on the Pro-rich cluster near the Cdc12 N terminus. These are precisely the kinds of “more stringent analysis” in “authentic yeast” (to use Reviewers’ language) that we would have wanted to do to test our model, had they not already been done by others. Without specific suggestions, we struggle to imagine what other kinds of experiments the Reviewers have in mind, apart from a eukaryotic version of a reconstituted cell-free translation system, which Reviewer #1 admits “would be substantially difficult” and “time consuming”. While we are intrigued by the reconstituted eukaryotic cell-free translation system that was published last year (which we mentioned on lines 994-995) and look forward to exploring it in future studies, it is not commercially available and we agree that the amount of effort required to prepare it ourselves is unrealistic for the current study. Most importantly, we do not find in the critiques provided any specific reason why our E. coli-based systems experiments are intrinsically less “stringent” or “rigorous”.

Accordingly, we think that, together with the results of multiple new experiments (detailed below), the extensive re-writing and re-ordering that we have done in the revised manuscript will be enough to better emphasize the importance and rigor of our findings and thus to address all of the Reviewers’ specific concerns.

Reviewer 1 thought that our manuscript “does not even provide new information, since the involvement of CCT and the Hsp70 system is not novel” and thought that “the key finding of this manuscript is how chaperones are involved in the de novo folding of septins, which is not conceptually new because of previous findings, including those of the authors”. Reviewer #3 also stated that “the function of Tric/CCT in septin folding and assembly is well documented”.

We were quite surprised at this reaction, since we dedicated a significant portion of the original manuscript (lines 68-76 and 319-322) to explicitly discussing the only other paper in the literature that specifically addresses the question of whether or not CCT is required for de novo septin folding. As a reminder, that paper explicitly stated that “it is unlikely that CCT is required to fold septins de novo” and “septins probably do not need CCT for biogenesis or folding”. With regard to involvement of the Hsp70 system, the only existing evidence in the literature on this subject is the aggregation of some septins in ssb1∆ ssb2∆ cells. Like the CCT study, that study did not distinguish whether this was a result of problems during septin synthesis and before septin complex assembly, or, alternatively, whether pre-folded and assembled septins were subject to disassembly, misfolding, and aggregation. Our experiments specifically test the fate of newly-synthesized septins prior to assembly in living cells. Our previous findings documented physical interactions between wild-type septins and multiple chaperones but did not address whether these interactions had any functional relevance. We previously reported functional effects of interactions between chaperones and MUTANT septins but, again, these studies did not address functional chaperone requirements for WILD-TYPE septins. While we did our best to highlight these points in the original document without devoting excessive amounts of text, we accept responsibility for not making these points sufficiently clear and to address this issue we added additional text, including the text quoted above, to the Introduction.

While Reviewer #3 commented that the manuscript “is overall well presented”, Reviewer 1 thought that the manuscript was “complicated to read” with “no logical connections, just a list of many results” and mentioned that part of the difficulty was “that it contains many negative results”.

In addition to reorganizing the manuscript, as suggested by the reviewers, we added more text at the beginning and end of nearly every section to even more explicitly state the logical connections between results. In our opinion, negative results of properly controlled experiments are valuable to the research community, and we do not understand what it is about negative results that makes them difficult to read about. Many of the extra experiments we performed were in anticipation of being asked to perform them by reviewers, some of which generated negative results. We are reluctant to remove negative results unless there is a more compelling reason. For example, to address another reviewer concern, we did remove the negative results with the Ydj1–Ssa2 compensatory mutants.

Reviewer #2: “4) Figure 2: The labeling on the protein structure makes it seem like the exact region for Ydj1 and Hsp70 was experimentally identified, when it hasn’t.”

We acknowledge that the first sentence of the figure legend (“the colored ribbon follows the color scheme in the sequences at right for overlapping β-aggregation, Ydj1 and Hsp70-binding sites”) could be misinterpreted, since only in the second sentence does it say “Sequence alignments show predicted binding sites”. We corrected this mistake, and added the text “Predicted chaperone binding sites” as the first words in the legend to this figure.

Reviewer #2: “8) The authors confusingly jump back and forth between different Septins and different chaperone (Ssa1-4, Ydj1, Sis1, Hsp104). We would ask the authors to re-arrange the manuscript, collating all the yeast work in one section and bacterial work in another.”

We re-arranged the manuscript and put all the yeast work in one section and all the bacterial work in another, with the exception of the studies of individually purified Cdc3 and Cdc12, which we put in between the yeast studies of the kinetics of de novo assembly and the yeast studies of post-translational assembly. Our reasoning is that the studies with the purified proteins demonstrate challenges with maintaining native conformations in the absence of chaperones and other septins, which flows naturally into the yeast studies asking about the ability of “excess” septins to maintain oligomerization-competent conformations in the absence of other septins and when we experimentally eliminate specific chaperones. All of the work actually manipulating E. coli genes/proteins is now together.

Reviewer #3: “1. The co-translational binding of CCT to nascent polypeptide chains has been studied (Stein et al., Mol Cell 2019). While the authors indicate that septin subunits are engaged co-translationally, they do not comment which ones are interacting with CCT and at which state of translation. This information is crucial and should also be mentioned in the discussion section.”

We are grateful to the Reviewer for bringing up this point, which we had overlooked. We hadn’t noticed that, in the end, only Cdc3 met the CCT confidence threshold to be included in the supplemental data of the Stein et al. paper. All septins co-purified with CCT in an earlier Dekker et al proteomic study, so we strongly suspect that the failure of the other septins to meet the confidence threshold in the Stein et al paper reflects the sensitivity of that assay, rather than a significant difference in how septin GTPase domains interact with CCT. We also hadn’t appreciated that according to that study, the main sites in the Cdc3 GTPase domain bound by CCT and Ssb are the same. Hence our statement that Ssb bound to septins “earlier” during translation, and CCT bound “later” was wrong. Instead, the overlapping Ssb and CCT site in Cdc3 turns out to be remarkably consistent with a conclusion from Stein et al paper, that CCT binds Rossmann-fold proteins like septins at sites where “early” beta strands have been translated and expose a chaperone-binding surface that later becomes buried by an alpha helix. We corrected our mistake in the text and in our model figure and added: (1) a new supplemental figure with predicted septin structures and a sequence alignment indicating where CCT and Ssb bound; and (2) text discussing the confidence thresholds for “calling” septin-CCT interaction, the Rossmann-fold binding, and how we interpret Ssb and CCT binding to the same site.

Reviewer #3 “3. Figure 3: It is recommended to also follow Cdc10-GFP and Cdc12-GFP fluorescence. This will on the one hand generalize the presented findings and provide a direct link to other parts of the study (e.g. crosslinking analysis of Cdc10).

We carried out the requested experiment for Cdc12, using Cdc12-mCherry rather than Cdc12-GFP because of the formation of non-native foci that we observed with Cdc12-GFP. We also attempted to analyze Cdc10, using an existing GAL1/10-promoter-driven Cdc10-mCherry plasmid that we’d made a few years ago, but it did not behave as expected, with high expression even in the absence of galactose (not shown), which prevented us from performing the requested experiment. We have a Cdc10-GFP plasmid with the inducible MET15 promoter, but this promoter does not provide sufficiently low levels of expression in repressive conditions, so there would be too much expression at the beginning of the experiment for us to accurately follow accumulation thereafter. Instead, we tried the only other plasmid we had with the GAL1/10-promoter controlling a tagged septin: Cdc11-GFP. Above a certain threshold of expression, Cdc11-GFP formed unexpected cortical foci, but we were still able to perform the analysis and found a clear delay in septin ring signal in cct4 cells, providing the requested generalization to other septins, if not Cdc10.

Reviewer #3 “5. Figure 4C: The finding that only ssb1 but not ssb2 knockouts have an effect on joining of free Cdc12-mCherry subunits into septin rings is puzzling. Similarly, Ssb1 largely acts co-translationally, while in this assay post-translational septin ring assembly is monitored. The authors need to comment on these two points.”

We did not examine ssb2 knockouts, so we do not know to what the Reviewer is referring in the first point. If the Reviewer means that they are puzzled by the fact that we saw a phenotype in cells in which only SSB1 was deleted and SSB2 remained, we offer two explanations. As can be seen in the Saccharomyces Genome Database entry for SSB1 (https://yeastgenome.org/locus/S000002388/phenotype), there are at least a dozen known phenotypes associated with deletion of SSB1 in cells with wild-type SSB2. We even showed a very clear septin misfolding/mislocalization phenotype in Supplemental Figure 4D. Thus while our findings are new and provide novel insights into Ssb function, they are not unprecedented. The Reviewer is correct that most Ssb is ribosome-bound and thus Ssb1 “largely acts co-translationally” but ~25% of Ssb is not ribosome-associated (PMID: 1394434). Furthermore, the lack of a strong phenotype for ssb1∆ cells in our new kinetics-of-folding experiment (see below), plus the realization that Ssb and CCT both bind the same site in Cdc3, leads us to a new model: Ssb acts both co- and post-translationally in septin folding, but only the post-translational function is associated with a phenotype in ssb1∆ cells, because in that assay we drastically overexpress a tagged septin and thereby exceed the Ssb chaperone capacity that remains when we delete SSB1. This logic also explains the first ssb1∆ phenotype we saw, when overexpressing Cdc10(D182N)-GFP. In the kinetics-of-folding assay, on the other hand, tagged septin expression is much lower and reducing the amount of total Ssb by ~50% (via SSB1 deletion) likely does not compromise Ssb function in folding the tagged septin. We therefore removed our statement that “Ssb dysfunction leaves nascent septins in non-native conformations that are aggregation-prone and unrecognizable to CCT”, revised our model figure accordingly, and added new text and citations to explain our new model.

Reviewer #3 “Additionally, they should test whether the appearance of septin ring fluorescence is slowed down in ssb1 mutants (as shown for cct4-1 mutant cells in Figure 3B).”

We agree that slower septin folding in ssb1∆ cells is a prediction of our model, and we performed the requested experiment and include the results in our revised manuscript. The new data show that the appearance of septin ring fluorescence is not delayed in ssb1∆ mutants, which is easily explained by the ability of Ssb2 to chaperone the folding of the low levels of tagged septin that we express in these kinds of experiments (see above).

Reviewer #3: “7. Figure 5G: The data is not convincing. This reviewer cannot detect a specific Cdc12 band accumulating in presence of GroEL/ES.”

We re-ran the reactions again with fresh reagents and this time ran the gel longer to reduce excess signal from free fluorescent puromycin and the bright Cdc10 bands. We now see a very clear band for full-length Cdc12 in the reaction with added GroEL/ES, fully consistent with our mass spectrometry results. We updated the figure with the new results.

Reviewer #3: “Furthermore, the activity tests done for the chaperonin system are confusing (Supplemental Figure 7). The ATPase rate (slope!) of GroEL/GroES seems higher as compared to GroEL but according to the authors it should be opposite.”

In our assays, the ATPase activity is so fast that for our “time 0” timepoint, much of it has already occurred by the time the reaction can be physically stopped and measured. In other words, the handling time is such that we can’t visualize what happened in the earliest stages of the reaction, where the rates could accurately be estimated as slopes. This is obvious from the fact that at time 0, the absorbance for the “GroEL alone” reaction is already more than twice the absorbance for GroEL+ES. We added clarifying text to the figure legend.

Reviewer #3: “The refolding assay using Rhodanese as substrate is also confusing: What is the activity of native Rhodanese? The aggregated Rhodanese sample seems to have substantial activity that is not too different from a GroEL/ES-treated one. From the presented data it is not clear to the reviewer to which extend GroEL/ES prevents aggregation and supports folding of denatured Rhodanese.”

We thank the Reviewer for bringing this to our attention, because made we mistakenly left out the values for native Rhodanese with the reporter. With regard to the aggregated Rhodanese, we failed to note that this sample contains urea. When the urea absorbance is subtracted, it is clear that the GroEL/ES-treated sample has higher activity. Furthermore, some native enzyme is likely still active within the aggregated sample, explaining the “substantial activity” that the Reviewer correctly notes. We corrected the figure and added clarifying text to the figure legend.

Reviewer #3: “the study goes astray following aspects that does not seem relevant to this reviewer (e.g. the role of N-terminal proline residues for Cdc12 translation, Fig. 5E/F).”

We acknowledge that we did a poor job of introducing the N-terminal Pro-rich cluster in Cdc12 with relation to our model of slow Cdc12 translation. Instead, we have revised and reorganized the manuscript to set up these experiments as a direct test of our model: if ribosome collisions on the body of the ORF drive mRNA decay, then decreasing the spacing of those ribosomes should exacerbate the problem, and eliminating the Pro-rich cluster (where published yeast data already show ribosomes accumulate) is the most logical way to test the prediction. Far from being irrelevant, the results fit the prediction perfectly and thus support the model. We expect that this change will highlight the importance of these experiments for the reader.

Reviewer #2: “1) Fig. 1 Is the folding of Cdc3 being measured in cells lacking chaperones mentioned towards the end of the paper or are the authors referring to the lack of yeast proteins?”

We are unclear as to what the Reviewer is asking here. The title of Figure 1 states that these are “purified yeast septins” and the figure legend further emphasizes this fact. Additionally, the Coomassie-stained gel in Figure 1A shows a single band, corresponding to purified 6xHis-Cdc3. The proteins were purified from wild-type E. coli cells, so all E. coli chaperones were present when Cdc3 initially folded, but chaperones and all other proteins were removed during the purification and prior to the analysis. We do not know what change to make.

Reviewer #2 asked “How do the authors account for the septin defect in Ssa4 delete cells in unstressed conditions where Ssa4 would be very low already? According to the authors previous work, Ssa2 and 3 should be able to compensate.”

We explicitly addressed this point in the original manuscript (lines 893-898). Again, we think here the Reviewer is equating chaperone binding with chaperone function. According to our previous work, Ssa2 and Ssa3 are able to bind septins, but this does not mean that they can fold septins the same way as Ssa4. We cite several papers that discuss the distinct functional roles for the different Ssa proteins. We do not think that additional clarification of this point would strengthen the manuscript.

Reviewer #3: “6. Figure 5B: It is unclear why Cdc3 is observed in the pulldown of His-tagged Cdc12 (37˚C), although no Cdc12 was isolated under these conditions. How is that possible?”

That is not possible. As we indicate in the figure legend and with the red asterisk, the only band appearing in that lane is a non-specific band that cross-reacts with the anti-Cdc3 and/or anti-Cdc11 antibodies. This is why it is also present in the “No septins” control lanes. We made the asterisk larger to help accentuate this point.

Reviewer #3: “Furthermore, the authors observe a specific effect on Cdc12-Cdc11 assembly in the E. coli groEL mutant. How do they rationalize this specific effect as Cdc12-Cdc3 assembly remained unchanged? This observation also seems in conflict with the suggestion of the authors that Cdc12 preferentially recruits Cdc11 before interacting with Cdc3 (page 45, lane 1024).”

Cdc11 was not expressed in the groEL mutants because no Cdc11 gene was present in those cells, as explained in the body text and indicated in the labeling above the lanes in Figure 5A. The band near the size of Cdc11 is a non-septin protein that bound to the beads in the groEL-mutant cells, as is shown in the immunoblot using anti-Cdc11 antibodies in Figure 5B. Thus there is no conflict to rationalize.

Reviewer #1: “The only evidence that CCT binds to septin is the list of LC-MS/MS. Western blotting would provide more solid data.” and “2) The cross-linking experiments appears not to have been successful. Why are the Ssas, Ydjs etc not detected here? “

First, CCT subunits are relatively low-abundance, expressed at 5- to 50-fold lower levels than other chaperone families in the yeast cytosol (see PMID: 23420633). To the Reviewer’s second point, we did in fact detect other chaperones in our crosslinking mass spectrometry experiments, including Ydj1, multiple Ssa and Ssb chaperones, Hsp104, etc., as can be seen in Table S1. However, they were also detected in negative control experiments. This is not surprising, given that these chaperones are among the most common “contaminants” of affinity-based purification schemes (see the CRAPome database at https://reprint-apms.org/). It was for this reason we had to perform so many negative control experiments, which likely produced some false negative results, as some “real” interactions were likely discarded when the same chaperone showed up in our controls. We added a figure panel with a Venn diagram of overlap between experimental and control samples, and text pointing out this caveat of our approach.

Second, in this experiment we attempted to identify proteins that transiently interact with a specific region of Cdc10 that will later become buried in a septin-septin oligomerization interface. Due to the transient nature of the interaction, we do not expect to detect high levels of crosslinked chaperones. Mass spectrometry is significantly more sensitive than immunoblotting, so there is no guarantee that we would be able to detect a band even if the crosslinking works as desired. Indeed, the crosslinked bands we saw by immunoblot for GroEL were quite faint (see Figure 2F), despite the fact that GroEL and the T7-promoter-driven Cdc10 were among the most abundant proteins in those E. coli cells.

Third, there is no commercially available, verified antibody recognizing yeast Cct3 for which to perform the requested immunoblot experiment. Since both the N and C termini of CCT subunits project into the folding chamber, it is unwise to use a standard epitope tagging approach, as the tags may compromise function. Indeed, for purification purposes others inserted an affinity tag in an internal loop in Cct3 (PMID: 16762366). We have a yeast strain with Cct6 tagged in an analogous way, but to perform the requested immunoblot experiment with Cct3 would require creating or obtaining the Cct3-tagged strain, deleting NAM1/UPF1, and introducing our Bpa tRNA/synthetase and GST-6xHis-Cdc10 plasmids. Given the sensitivity of detection concerns stated above, we doubt this would help.

In summary, we prefer not to attempt the requested immunoblot experiments.

Reviewer #1: “-Fig. 3B ant related Figures: The experiment to see if GFP-tagged septin accumulates in the bud neck is important, but only the graphs after the analysis are shown. The authors should provide the readers with representative examples from imaging data.”

We are confused, because the images at the bottom of Figure 3A already show what the Reviewer requests. As stated in the figure legend, these are representative examples of the imaging data from a middle timepoint of one of the experiments. It would be nearly impossible (for space reasons) to provide representative images for all of the timepoints for all of the genotypes for all of the experiments. Since in our new experiments we introduce new tagged septins (Cdc11-GFP and Cdc12-mCherry), we also now include representative images of cells expressing these proteins, as well.

Reviewer #2: “3) If the authors had evidence of chaperone interaction from their previous study, why did they not simply do IPs with fragments of the septins/chaperones?”

We are unclear why the Reviewer is suggesting IPs after referring to our previous study. IPs are a poor choice for transient interactions, which is why we mostly avoided them in previous studies, and instead used a novel approach (BiFC) to “trap” chaperone–septin interactions. Moreover, we seek to identify chaperones that bind wild-type septins at future septin-septin interfaces on the path towards the native conformation. Fragments of septin proteins would likely misfold and would therefore likely attract chaperones that wouldn’t normally bind the full-length septin. Indeed, our previous studies demonstrated that even a single non-conservative amino acid substitution was sufficient to alter chaperone-septin binding. Thus IPs with fragments of septins or chaperones would be highly unlikely to yield informative results for the questions we seek to answer. We strongly prefer not to attempt these suggested experiments.

Reviewer #2: “5) While differences between Ssa paralogs are highly interesting, using deletions of Ssas is not useful, given that yeast compensate by overexpressing other paralogs. The yeast GFP Septin assays should be repeated in yeast lacking all Ssas and expressing one paralog on a constitutive promoter (See numerous papers by Sharma and Masison).”

We disagree that ssa deletions are “not useful”, since if the overexpressed paralogs cannot fulfill the same function as the deleted SSA, then we will see a phenotype. Which we do. Furthermore, we had already obtained and thoroughly tested a strain like the ones mentioned by the reviewer (ECY487, a.k.a. JN516, from Betty Craig’s lab, with ssa2∆ ssa3∆ ssa4∆ and SSA1, which is constitutively expressed, PMID: 8754838), but we found that, as published, it divides slightly more slowly even under the most permissive of conditions. The requested strain cannot be analyzed using our method, because slow accumulation of ring fluorescence could be attributed to other defects unrelated to septin folding. Thus we strongly prefer not to attempt the suggested experiments.

Reviewer #2: “7) The authors need to clarify the experiment with the Ydj1 D36N and Ssa2 R169H. In Reidy et al, they never fully biochemically test this system and it was never examined for Ssa2-Ydj1. The authors would need to do some fundamental experiments to demonstrate the validity and functionality of this double mutant in yeast.”

Given that this experiment was unable to generate meaningful data, since the mutations affected the kinetics of induction of the GAL1/10 promoter, we do not think the requested biochemical experiments would add any value to the study. Instead, we removed these studies from the manuscript.

Reviewer #3: “4. Figure 3B: The difference between wt and cct4-1 cells in appearance of septin ring fluorescence is observed at one timepoint. Since this experiment is considered highly relevant, the authors are asked to include another timepoint to bolster the conclusion that Cdc3-GFP folding and thus septin ring assembly is delayed in the CCT mutant.”

We carried out new experiments with cct4-1 cells using Cdc12-mCherry and Cdc11-GFP with more timepoints than in our original cct4-1 experiments with Cdc3-GFP. Since these experiments provide the same kinds of results, but at multiple timepoints, we do not see the value in repeating the Cdc3-GFP experiment.

Reviewer #3: “If Ssb1 functions to maintain Cdc12 in an assembly competent state preventing misfolding, one would expect either enhanced degradation or aggregation of Cdc12-mCherry in ssb1 mutant cells. Did the authors check for such scenario? Septin aggregation has been shown in a ssb1 ssb2 double deletion strain (Willmund et al., 2013), yet the data shown here predict that aggregation might already occur in single ssb1 mutants.”

We already examined septin aggregation in single ssb1 mutants and showed these data (Supplementary Figure 4D). Indeed, this phenotype was the rationale for testing post-translational septin assembly in ssb1 single mutants. We have seen no evidence of septin degradation in any context (as we mentioned on line 889), so we would not expect it here. While we added new text and a very new citation showing that many “misfolded” conformations of wild-type E. coli proteins avoid aggregation and degradation, we do not think that the suggested experiments would add enough value to the current study to justify the effort, time and expense.

Reviewer #3: “Fig. 3C: The figure showing septin ring fluorescence does not include error bars. This is crucial, also because the difference between wt and ssa4 mutant cells is not large.”

There are, in fact, error bars included in the figure, as can be most clearly seen for the final timepoint for the ssa4∆ cells. For most of the other timepoints the error bars are smaller than the data point symbols (the circles and squares). We do not think that adjusting the size or opacity of the symbols to better show the error bars will be sufficiently valuable to justify the effort.

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 presented work the authors studied the folding and assembly of septin subunits and the role of molecular chaperones in this process. They used a variety of diverse in vitro and in vivo assays to study the interaction between septin subunits and chaperones and to characterize their conformational states. The manuscript includes a huge amount of work that is overall well presented. Yet, the diverse approaches also lead to a tessellated view. While the combined study of septin folding in E. coli and S. cerevisiae cells has advantages, a more stringent analysis in one organism (authentic yeast) would increase rigor. The manuscript also suffers from overinterpretation of data in parts (e.g. translation rate of Cdc12 and how this might be affected by folding and chaperones) and occasionally the study goes astray following aspects that does not seem relevant to this reviewer (e.g. the role of N-terminal proline residues for Cdc12 translation, Fig. 5E/F). The paper will therefore substantially benefit from streamlining and major rewriting. Concerning the roles of chaperones: the function of Tric/CCT in septin folding and assembly is well documented, the involvement of other chaperones (e.g. Hsp70: Ssb2 or Ssa4) remains less clear and was not fully explored. Considering these limitations, a major revision of the study will be necessary.

Major points:

1. The co-translational binding of CCT to nascent polypeptide chains has been studied (Stein et al., Mol Cell 2019). While the authors indicate that septin subunits are engaged co-translationally, they do not comment which ones are interacting with CCT and at which state of translation. This information is crucial and should also be mentioned in the discussion section.
2. The evidence for direct interaction between Cdc10 and CCT is rather weak as it is based on CCT absence in a mass spec list of proteins from a control experiment.
3. Figure 3: It is recommended to also follow Cdc10-GFP and Cdc12-GFP fluorescence. This will on the one hand generalize the presented findings and provide a direct link to other parts of the study (e.g. crosslinking analysis of Cdc10).
4. Figure 3B: The difference between wt and cct4-1 cells in appearance of septin ring fluorescence is observed at one timepoint. Since this experiment is considered highly relevant, the authors are asked to include another timepoint to bolster the conclusion that Cdc3-GFP folding and thus septin ring assembly is delayed in the CCT mutant.
5. Figure 4C: The finding that only ssb1 but not ssb2 knockouts have an effect on joining of free Cdc12-mCherry subunits into septin rings is puzzling. Similarly, Ssb1 largely acts co-translationally, while in this assay post-translational septin ring assembly is monitored. The authors need to comment on these two points. Additionally, they should test whether the appearance of septin ring fluorescence is slowed down in ssb1 mutants (as shown for cct4-1 mutant cells in Figure 3B). If Ssb1 functions to maintain Cdc12 in an assembly competent state preventing misfolding, one would expect either enhanced degradation or aggregation of Cdc12-mCherry in ssb1 mutant cells. Did the authors check for such scenario? Septin aggregation has been shown in a ssb1 ssb2 double deletion strain (Willmund et al., 2013), yet the data shown here predict that aggregation might already occur in single ssb1 mutants.
6. Figure 5B: It is unclear why Cdc3 is observed in the pulldown of His-tagged Cdc12 (37{degree sign}C), although no Cdc12 was isolated under these conditions. How is that possible? Is the appearance of Cdc3 reflecting non-specific binding to the used resin? Furthermore, the authors observe a specific effect on Cdc12-Cdc11 assembly in the E. coli groEL mutant. How do they rationalize this specific effect as Cdc12-Cdc3 assembly remained unchanged? This observation also seems in conflict with the suggestion of the authors that Cdc12 preferentially recruits Cdc11 before interacting with Cdc3 (page 45, lane 1024).
7. Figure 5G: The data is not convincing. This reviewer cannot detect a specific Cdc12 band accumulating in presence of GroEL/ES. Furthermore, the activity tests done for the chaperonin system are confusing (Supplemental Figure 7). The ATPase rate (slope!) of GroEL/GroES seems higher as compared to GroEL but according to the authors it should be opposite. The refolding assay using Rhodanese as substrate is also confusing: What is the activity of native Rhodanese? The aggregated Rhodanese sample seems to have substantial activity that is not too different from a GroEL/ES-treated one. From the presented data it is not clear to the reviewer to which extend GroEL/ES prevents aggregation and supports folding of denatured Rhodanese.

Minor points:

Fig. 3C: The figure showing septin ring fluorescence does not include error bars. This is crucial, also because the difference between wt and ssa4 mutant cells is not large.

Review Cross-commenting:

I also concur with the comments of the other reviewers. The manuscript is in need of extensive revision.

Significance

see statement above

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 study, the authors set out to understand the requirements for yeast Septin folding. They purify Cdc3 from bacteria homo-oligomerized. This was followed by very nice HDX-MS analysis demonstrating that the NTE is flexible and may be intrinsically disordered. Follow up experiments using cross-linking mass spectrometry identified interaction of Cdc10 with Cct3. Finally the authors queried the impact of both yeast and e.coli chaperones on septin folding and demonstrated a dependence on yeast Ssa4 and e. coli GroEL.

Major comments:

1. Fig. 1 Is the folding of Cdc3 being measured in cells lacking chaperones mentioned towards the end of the paper or are the authors referring to the lack of yeast proteins?
2. The cross-linking experiments appears not to have been successful. Why are the Ssas, Ydjs etc not detected here? The authors only pull out one interactor which is not validated going forward.
3. If the authors had evidence of chaperone interaction from their previous study, why did they not simply do IPs with fragments of the septins/chaperones?
4. Figure 2: The labeling on the protein structure makes it seem like the exact region for Ydj1 and Hsp70 was experimentally identified, when it hasn't.
5. While differences between Ssa paralogs are highly interesting, using deletions of Ssas is not useful, given that yeast compensate by overexpressing other paralogs. The yeast GFP Septin assays should be repeated in yeast lacking all Ssas and expressing one paralog on a constitutive promoter (See numerous papers by Sharma and Masison).
6. How do the authors account for the septin defect in Ssa4 delete cells in unstressed conditions where Ssa4 would be very low already? According to the authors previous work, Ssa2 and 3 should be able to compensate.
7. The authors need to clarify the experiment with the Ydj1 D36N and Ssa2 R169H. In Reidy et al, they never fully biochemically test this system and it was never examined for Ssa2-Ydj1. The authors would need to do some fundamental experiments to demonstrate the validity and functionality of this double mutant in yeast.
8. The authors confusingly jump back and forth between different Septins and different chaperone (Ssa1-4, Ydj1, Sis1, Hsp104). We would ask the authors to re-arrange the manuscript, collating all the yeast work in one section and bacterial work in another.

Review Cross-commenting:

I agree with the comments made by the other two reviewers.

Significance

While the final figure 6 is nice, the data only hint at this model, much more work in vivo and in vitro to solidify this mechanism is needed. Overall, while this work has some merit, the experimental path seems a bit disorganized is highly preliminary-it is unclear what the impactful findings are.

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

The cellular folding of polymer-forming septins is an important but poorly understood issue. The authors and other groups previously found that cytoplasmic chaperones interact with septins and are involved in oligomerization, but the details have not been analyzed. Therefore, the authors conducted a series of experiments using the budding yeast septin system as a model using many methodologies. Specifically, the experiments include purified yeast septins, in vivo cross-linking, live-cell imaging of yeast, chaperones-deficient strains of E. coli, and a reconstituted cell-free translation system in E. coli. As a result of such experiments, the authors not only gained much insight into the role of chaperones in the de novo folding of septins but also found that chaperonins are co-translationally involved in the folding of septins. Furthermore, they even argue that the translation elongation rates are associated with the co-translational folding, based mainly on the E. coli system and bioinformatics analysis.

Major comments:

First, I should say that this paper is complicated to read. There are no logical connections, just a list of many results, making it difficult to grasp what is important. It is a complex combination of budding yeast and E. coli systems, but the E. coli results should not be immediately extended to cases in eukaryotes. Another reason for the difficulty in reading the manuscript is that it contains many negative results.

The key conclusion in this manuscript seems to be the effect and the role of chaperones in the folding coupled with translation, as shown in the model diagram in Figure 6. However, the experiments related to translation are not justified because all the translation-related results are derived from the data in the E. coli system. Since the goal of this manuscript is to gain insight into the folding of septins in eukaryotic cells, I do not agree with drawing conclusions about septin folding based on the E. coli experiments.

If the E. coli experiments are excluded, the key results in this manuscript, in my view, are very few, only Fig. 3B and partly Fig. 4D, in which the formation of the septin ring is slowed using a CCT mutant strain. If the point that translation speed is involved is not convincing, then this manuscript does not even provide new information, since the involvement of CCT and the Hsp70 system is not novel. If the model depicted in Fig. 6 is to be justified, the authors need to carefully study the eukaryotic translation system. However, it would be practically difficult because it would take long time and a reconstructed cell-free translation system used in this manuscript is not easy. However, it would be substantially difficult because it would be time consuming and not easy to use a eukaryotic version of a reconstituted cell-free translation system used in this manuscript.

Minor comments:

• The only evidence that CCT binds to septin is the list of LC-MS/MS. Western blotting would provide more solid data.
• Fig. 3B ant related Figures: The experiment to see if GFP-tagged septin accumulates in the bud neck is important, but only the graphs after the analysis are shown. The authors should provide the readers with representative examples from imaging data.

Review Cross-Commenting:

I totally agree with both reviewers. I believe the manuscript needs a major revision with significant restructuring. Furthermore, I think it would be better to at least consider the possibility of splitting it up.

Significance

The key finding of this manuscript is how chaperones are involved in the de novo folding of septins, which is not conceptually new because of previous findings, including those of the authors. Many methodologies are used, but each is not a novel methodology and is not new. Finally, I am working on the mechanisms of action of chaperones, especially chaperonins, the intracellular dynamics of proteins, and proteomics from both biochemical and cell biology perspectives.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

General response to the reviewer

We thank all reviewers for their constructive comments on our manuscript. We were very pleased to see that the reviewers found our study ‘…represent new insight in the field’ (rev#1) and ‘…contains important and exciting novel findings’ (rev#2), and ‘…gives a more detailed perspective on how Src proteins (Src42A in Drosophila) control epithelial stability and the contraction of specific surfaces of epithelial cells’ (rev#3). The reviewers raised a number of specific points that we partially addressed already in a preliminary revision of the manuscript. Some more points will require some additional experiments that we will incorporate in a fully revised version of the manuscript.

Reviewer #1

(Evidence, reproducibility and clarity (Required)): Highest priority: 1) The Src42A knockdown and germline clone experiments both cause defects in cellularization (Fig. 2B and 9A), which could result in differences in the state of the blastoderm epithelium (cell size, cell number, structural integrity, organization, etc.) between the experimental and control conditions. In addition, Src42A knockdown appears to affect the size and shape of the egg (Fig. 9A and 9C). The manuscript would be strengthened if the authors included data to demonstrate that the initial structure of the epithelium is mostly normal (quantifications of cell size, number, etc.) in the Src42A RNAi condition, as this would bolster the argument that germband extension, rather than due to indirect effects resulting from the cellularization defects. The authors may have relevant data to do this on-hand, for example using data associated with figures 1, 3, 6, and 9.

Response:

The cellularization phenotype of src42A knockdown embryos has a penetrance of about 50% and exhibits a variable expressivity. We attempted to characterize this phenotype in detail, but failed to identify any dramatic differences in cellularization of the src42A knockdown embryos compared to wild type. The localization of E-cadherin, in turn is not affected, but occasionally, nuclei are dropping out of the blastoderm before cellularization is accomplished. This can result in patches of irregular cellularization, but the blastoderm epithelium in stage 6 embryos did not display major defects in overall structure. We will present additional data on the cellularization phenotypes in the fully revised manuscript. As the referee suggested, we will analyze our data to determine potential effects on the cell size, cell number and overall organization of the blastoderm before germband extension. We plan to present these data as an additional Suppl. Mat. Figure in the full revision.

Lower priority:

5) Figure 8 - in my opinion, using a FRAP or photoconversion approach would be a more convincing demonstration of differences in E-cadherin residency times / turnover rate than time-lapse imaging of E-cadherin:GFP alone. Authors should decide whether this improvement is worth the investment.

Response:

We thank the reviewer for this comment. While we believe that the data presented in Fig. 8 demonstrates a significant difference in the E-cadherin residence time based on E-cadherin-GFP fluorescence intensity, we agree with the referee that FRAP analyses would provide additional evidence to support our conclusion. For the full revision, we will therefore attempt to perform FRAP-experiments on src42A knockdown embryos expressing E-cadherin-GFP and compare the recovery time to the wild type.

Reviewer #1 (Significance (Required)):

The manuscript by Backer et al. examines the function of Src42A in germband extension during Drosophila gastrulation. Prior studies in the field have shown that Src family kinases play an important role in the early embryo, including cellularization (Thomas and Wieschaus 2004), anterior midgut differentiation (Desprat et al. 2008), and germband extension (Sun et al. 2017; Tamada et al. 2021). In this study, the authors showed that Src42A was enriched at adherens junctions and was moderately enriched along junctions with myosin-II. They then showed that maternal Src42A depletion exhibits phenotypes, starting with cellularization and including a defect in germband extension. The authors focus on defects in germband extension and found that Src42A was required for timely rearrangement of junctions and that the Src42A RNAi phenotype is enhanced by Abl RNAi. Finally the authors show that E-cadherin turnover is affect by Src42A depletion.

Overall, this study provided a higher resolution description of how Src42A regulates the behavior of junctions during germband extension. I thought the authors conclusions were well supported by the data and represent new insight in the field.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

Summary: Chandran et al. investigate the role of Src42A in axis elongation during Drosophila gastrulation. Using maternal RNAi and CRISPR/Cas9-induced germline mosaics, they revealed that Src42A is required to contract junctions at anterior/posterior cell interfaces during cell intercalations. Using time-lapse imaging and image analysis, they further revealed the role of Src42A in E-Cad dynamics at cell junctions during this process.

By analyzing double knockdown embryos for Src42A and Abl, they further showed that Src42A might act in parallel to Abl kinase in regulating cell intercalations. The authors proposed that Src42A is involved in two processes, one affecting tension generated by myosin II and the other acting as a signaling factor at tricellular junctions in controlling E-Cad residence time. Overall, the data are clear and nicely quantified. However, some data do not convincingly support the conclusion, and statistical analyses are missing for an experiment or two. Methods for several quantifications also need improvement in writing. Also, several figures (Figures 6-8) do not match the citation in the text and need to be corrected.

Page and line numbers were not indicated in the manuscript. For my comments, I numbered pages starting from the title page (Title, page 1; Abstract, page 2, Introduction, pages 3-6; Results, pages 7-14; Discussion, pages 15-18; M&M, 19-23; Figure legends, 28-30) and restarted line numbers for each page. For Figures 6-8 that do not match the citation in the text, I still managed to look at the potentially right panels. All the figure numbers I mention here are as cited in the text. My detailed comments are listed below.

Response:

We apologize for the lack of organization of the manuscript and the figure numbering. In the revised version we have added page numbers, line numbers and we corrected the figure numbers.

Major comments: 1. b-Cat/E-Cad signals at the D/V and A/P junctions in Src42Ai (Figs. 5-6). These data are critical for their major conclusion and should be demonstrated more convincingly.

In Fig. 5A, the authors said, "When the AP border was cut, the detached tAJs moved slower in Src42Ai embryos compared to control (Fig. 5A)". However, even control tAJs do not seem to move that much in the top panels, and I found the images not very convincing.

Response:

We thank the referee for commenting on the lack of clarity in the presentation of the data. The overall movement within the first 10 seconds after the laser cut (determined by movement of adjacent D/V tAJs from each other) was about 2 µm in the wildtype, while in the mutant it was 1 µm. Despite this 50% difference, it may be difficult to appreciate this difference from looking at Fig. 5A in our original submission. The yellow lines in Fig 5A only showed the region of the cut, but did not indicate the movement of the tAJ from each other, which may have led to a distraction from the actual movement. We will change the annotation and the marks within the figure to visualize the movement much more clearly in the full revision. In the fully revised manuscript, we will also add movies from the experiments including marks of the tricellular junctions to follow the displacement as part of the Supplemental Material.

Based on the genetic interaction between Src42A and Abl using RNAi (Fig. 7), the authors argue that Src42A and Abl may act in parallel. However, the efficiency of Abl RNAi has not been tested. It can be done by RT-PCR or Abl antibody staining. Also, the effect of Abl RNAi alone on germband extension should be tested and compared with Src42A & Abl double RNAi embryos. I expect the experiments can be done within a few weeks without difficulty.

Response:

We agree with the referee that it is important to determine the level of depletion in Abl RNAi embryos in order to interpret the genetic relationship between Abl and Src42A. In the full revision of the manuscript, we will follow the advice of the referee and analyze the knockdown, preferably by antibody labeling with an anti-Abl antibody. We will also generate single knockdowns of abl in embryos and determine their effect on germband extension compared to wildtype and src42/abl double knockdown.

Minor comments:

Fig. 2 - Fig. 2B: Higher magnification images of the defective cytoplasm can be shown as insets.

Response:

We will add some higher magnification images of the cellularization phenotype in the full revision of the manuscript. In addition, as mentioned in the response to reviewer #1, we will provide a more detailed analysis of the cellularization in src42Ai embryos in the fully revised manuscript.

• Fig. 2E: A simple quantification of the penetrance of cuticle defects in Src42A mutants and RNAi will be helpful, as shown in Fig. S3.

Response:

In the full revision, we will add the quantification of the occurrence of the different classes of cuticle phenotypes.

Fig. 9 - Fig. 9A: Magnified views of the cytoplasmic clearing can be added as insets.

Response: As described in our response to the comments made by referee #1, we will add a more detailed analysis of the cellularization phenotype in the full revision.

Page 14, lines 9-10: More explicit description of the phenotype rather than just "stronger compared to Src42Ai" will be helpful.

Response:

In the full revision, we will add a more detailed description of the phenotype and re-analyze and present data on the hatching rate, stage of lethality and cuticle phenotypes.

Reviewer #2 (Significance (Required)): This work revealed the role of Src42A in regulating germband extension. A previous study suggested the roles of Src42A and Src64 in this developmental process using a partial loss of both proteins (Tamada et al., 2021). Using different approaches, the authors demonstrated a role of Src42A in regulating E-Cad dynamic at cell junctions during Drosophila axis elongation. Most of the analyses were done with maternal knockdown using RNAi, but they successfully generated germline clones for the first time and confirmed the RNAi phenotypes. Overall, this work contains important and exciting novel findings. This work will be of general interest to cell and developmental biologists, particularly researchers studying epithelial morphogenesis and junctional dynamics. I have expertise in Drosophila genetics, epithelial morphogenesis, imaging, and quantitative image analysis.

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

Chandran et al. report on the function of Src42A during cell intercalation in the early Drosophila gastrula. They create a Src42A-specific antibody (there are two Src genes in the fly genome) and examine the localization of Src42A and observe a planar-polarized distribution at cell interfaces. They then measure cell-contractile dynamics and show that T1 contraction is slower after Src42A disruption. The authors then argue that Src42A functions in a parallel pathway to the Abl protein, and that E-cadherin dynamics (turnover) is altered in Src42A disrupted embryos. Src function at these stages has been studied previously (though not to the degree that this study does), and in some respects the manuscript feels a little preliminary (please label figures with figure number!), but after editing this should be a polished study that merits publication in a developmentally-focused journal.

1) Does the argument that Src42A has two functions fully make sense? Myosin II function is known to affect E-cadherin stability (and vice versa), so it seems that Src42A could affect both MyoII and Ecad by either decreasing Myosin II function/engagement at junctions or by destabilizing Ecad.

Response:

We thank the referee for raising an important point that we may not have discussed appropriately in our initial submission. We agree that the reciprocal relationship between actomyosin and E-cadherin might not be reflected equivocally in our manuscript. As the referee points out, Src42A could affect both MyoII planar localization and E-cadherin dynamics through the same pathway. Previous studies showed that Src is involved in translating the planar polarized distribution of the Toll-2 receptor by recruiting Pi3-Kinase activity to the Toll-2 receptor complex resulting in planar polarized distribution of MyoII at the A/P interfaces. These data, however do not address the possibility that a well-known Src target, the E-cadherin/ß-Catenin complex, which is extensively remodeled in germband extension contributes to the delay in germband extension. The observed defects in both studies can be attributed to both a defect in abnormal planar polarization of MyoII and the abnormal dynamics of the E-cadherin/ß-catenin complex. In either of these cases, we suggest that Src42A phosphorylates distinct substrates, the Toll-2 intracellular domain in the MyoII planar polarity pathway and the E-cad/ß-Cat complex controlling E-cad dynamics. Given the relationship between MyoII and E-cadherin, however, it is not possible to decide whether these two effects are independent functions of Src42A or are consequences of each other. Since we cannot resolve a possible epistatic relationship between these potential two activities of Src42A, we decided to extend the discussion on this topic by taking both possible scenarios into account and discussing them appropriately. We will add this discussion in the full revision of the manuscript.

) One obvious question that arises is the nature of cleavage defects that are mentioned that happen previously to intercalation. For example, is E-cad normal prior to intercalation initiating? How specific are the observed defects to GBE?

Response:

please see response to referee #1

3) Pg. 10, "the shrinking junction along the AP axis strongly reduces its length with an average of 1.25 minute" - what is this measurement? How much is "strongly"?

Response:

We thank the referee for pointing out our inappropriate qualitative statement of the experimental data, which was indeed misleading. The measurement of the shrinking junction was based upon the time it takes for the AP interface junction between two adjacent vertices on the DV axis to shrink into a single 4-cell vertex. The time for this contraction was on average 1 minute 25 seconds. The data in Fig.4 A’,C show that after 2 minutes in the control embryo 100% of the observed AP junctions have collapsed and the extension of the new DV junction along AP axis has begun. At the same timepoint of 2 minutes in the src42A knockdown, we show in Fig. 4B’,D that the shrinking of the AP junction interface has still not been completed in 60% of the cases.

In the full revision, we will remove the qualitative statement and replace it with a correct description of the measurements taken and will refer to the data described in Fig. 4 A-D.

4) Also pg. 10, "the AP junction was not markedly reduced after 1 minute" - what is the criteria for this statement? X%? 1 minute is very specific, it feels like how much of a reduction/non-reduction should also be specific.

Response:

please see response to point 3.

Reviewer #3 (Significance (Required)):

This study gives a more detailed perspective on how Src proteins (Src42A in Drosophila) control epithelial stability and the contraction of specific surfaces of epithelial cells.

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

Reviewer #2 and #3 noted that the manuscript was somewhat unorganized with regard to lacking the numbering of pages, lines and figures. We also noted that in the submission process the figures were not presented in the correct order. In the preliminary revision of the manuscript, we fixed these problems to facilitate the evaluation of our transferred manuscript by editorial boards.

In addition, we also addressed issues that the referees mentioned by editing the text according to their comments. We also addressed problems regarding the presentation of the figures and statistical analyses of the data. The following changes were made:

1. We added page numbers and line numbers.
2. We added figure numbers to the figure panels.
3. We corrected ordering of figures in the transferred manuscript.
4. We addressed the following comments by statistical analyses, editing the text and the figures:

   Regarding comments from Reviewer #1:

Highest Priority:

2) There is a discrepancy in the staging of embryos used between some of the analyses, which make it hard to interpret some of the data. For example, characterization of the knockdowns in Fig. 1A and B are based on stages 10 and 15, whereas the majority of the paper is focused on earlier stages 6 - 8 during germband extension (e.g., Fig. 1D). The analysis for Fig. 1B would be more meaningful if it was done on the same stages used for subsequent phenotypic analysis so they can be directly compared.

Response:

We thank the referee for pointing out an apparent misunderstanding caused by the description of Fig. 1A,B. The data presented in Fig.1A and 1B do not show RNAi knockdown experiments, but show a comparison between embryos that are heterozygous or homozygous for the loss-of-function allele src42A26-1. These data were intended to demonstrate that zygotic mutants still maintain levels of maternal Src42A protein up until late stages of development. Data for embryos at an earlier stage (stage 5) were shown in the Supplementary Fig. S1E, where no difference in protein levels of Src42A can be observed between heterozygous and homozygous zygotic src42A26-1 embryos.

At the beginning of the results sections 1 and 2 of the preliminary revised manuscript, we added a sentence to address the referee’s concern that earlier stages exhibit no difference in protein levels and will refer to Fig. S1E. We also more explicitly spelled that out that the experiment (referring to Fig.1A,B and S1) was intended to look at zygotic mutants and to demonstrate that our novel Src42A antibody was able to detect the reduction of maternal Src42A protein in mid- to late-stage homozygous zygotic embryos.

3) There is incongruence between figures in terms of which junctional pools (bAJs vs. tAJs) of beta-catenin and E-cadherin are quantified that makes it difficult to draw comparisons between analyses. For example, pTyr levels are examined for both bAJs and tAJs in Figure 3, however, only tAJs are considered in Fig. 8. Similarly, in some cases planar cell polarity is considered (e.g., comparison of levels at AP vs DV bAJs in Fig. 6 and 9), and in other cases (e.g. Fig. 8) it is not.

Response:

We thank the referee for commenting on the different readouts for different pools of cell junctions in our experiments. In our study we considered effects on src42A on both, bAJs and tAJs by RNAi knockdown of src42A. We decided to present the data for bAJ and tAJ in separate figures for clarity and structure. For example, the data for the effect of src42A knockdown on the planar polarized distribution on bAJs of E-cadherin were presented in Fig.6, while the effect on E-cadherin residence time in tAJs were presented in Fig.8. The analysis pTyr levels considered both pools in order to determine whether src42A knockdown leads to an overall reduction of pTyr levels or to a reduction in a specific junctional pool. From our data we conclude that pTyr levels show a similar reduction in both, the bAJ and the tAJ junctions.

In order to address the reviewer’s comment, we have linked the figures more stringently with the results text of the preliminary revision. We only referred to the reduction in PTyr levels in Fig. 3 to point out that both junctional pools are affected by reduced PTyr in src42i embryos. Furthermore, we referred to the individual figure panels when addressing junctional pools and explain the rationale to focus on particular pools (bAJs or tAJ) in the experiments in detail. For Fig. 6 we point out in the preliminary revised manuscript that we focus the analyses on the known planar polarized distribution of beta-catenin and E-Cadherin.

Lower priority: 1) Introduction, 2nd paragraph - The modes of cell behaviors described to drive cell intercalation leaves out another clear example in the literature - Sun et al., 2017 - which describes a basolateral cell protrusion-based mechanism. While the authors cite this paper later, leaving it out when summarizing the state of the field misrepresents the current knowledge of the range of mechanisms responsible.

Response:

We thank the referee for this remark. In the preliminary revision, we have added to the introduction that the cell behaviors associated with germband elongation include apical and basolateral rearrangements of the cells indicating that basolateral protrusions also contribute to the set of mechanisms that drive germ band elongation.

2) 'defective cytoplasm' - this term is confusing, and could perhaps be replaced with 'cellularization defect', or something similar.

Response:

We agree that the term we applied for the cellularization defect may be misleading. The observation, we intended to describe with the term was a defect in the cytoplasmic clearing which occurs in the last syncytial division and the beginning of the cell formation process. We changed the description of this observation according now refer to the defect in the preliminary revised manuscript as ‘cytoplasmic clearing defect’.

3) Tests of statistical significance are not uniformly applied across the figures. For instance, Figures 3G + H indicate statistical significance, but Fig. 3D + E do not. Performing statistical tests throughout the paper, or clearly articulating a rationale when they are not used, would strengthen the manuscript. Specifically, the authors should consider this for Fig. 3D + E, and Fig. 7D + E, to support their arguments that rates of germband extension are different between conditions.

Response:

We agree with the reviewer and have provided statistical analysis for the data displayed in Fig. 3D,E and Fig. 7D,E in the preliminary revision of the manuscript.

4) Page 12 - "We found that Src42A showed a distinct localization at the tAJs (Fig. 1B)": Figure 1B shows a quantification of levels at bAJs, not tAJs.

Response:

In the preliminary version of the revised manuscript, we added a quantification of the localization of Src42A at the tAJs as a part of Suppl Fig. S4. In Fig. S4A-C we show that Src42A is enriched in comparison to the bAJs.

Regarding comments from reviewer #2:

Major Comments:

In Fig. 6A, b-Cat signals look fuzzier and dispersed and have more background signals in the control, compared to the Src42Ai background. Also, b-Cat signals in the control image do not seem to show enrichment at the D/V border, as shown in Tamada et al., 2012.

Response:

We agree with the referee that the image in Fig. 6A for the control is fuzzier and looks dispersed. This is due to the fixation method that we used. In this experiment we did not apply heat fixation, but used formaldehyde fixation in which b-catenin protein, in addition to the junctional pool, is also maintained in the cytoplasm creating the fuzzy cytoplasmic staining. We chose to do this in order to be able to co-immunolabel the embryos with b-catenin and E-cadherin antibodies; the latter staining is not working with the heat fixation applied in the Tamada et al. 2012 paper. Despite the slightly lower quality of the staining, the quantification of the data clearly indicated an effect of src42A knockdown on the planar polarized distribution of E-cad/b-cat complex does show an enrichment. In the preliminary revision added a note to the figure legend to indicate the fact that the fixation procedure was not optimized for b-catenin junctional staining. In the preliminary revision we also added a quantification of live imaging data recording E-cadherin-GFP in wild-type and src42Ai embryos. We present these additional data in Fig. S5 in the preliminary revision of the manuscript. These data are consistent with the results in Fig. 6 from the immunolabeling and support our conclusion that E-cadherin AP/DV ratio is increased in Src42A knockdown embryos.

In Fig. 6B, C, it is not clear how the intensity was measured and how normalization was done. Was the same method used for these quantifications as "Protein levels at bicellular and tricellular AJs" on pages 21-22? Methods should be written more explicitly with enough details.

Response:

We thank the referee for pointing out the lack of detail in explaining how the quantification was done. In the preliminary revision of the manuscript, we extended a paragraph entitled ‘Protein levels at bicellular and tricellular junctions’ in the methods section that will serve this purpose and describe the methods that were applied for each quantification and the method as to how the data were normalized.

Does each sample (experimental repeat) for the D/V border in Fig. 6B match the one right below for the A/P border in Fig. 6C? It should be clearly mentioned in the figure legend. The ratio of the DV intensity to AP intensity will better show the compromised planar polarity of the b-Cat/E-Cad complex.

Response:

We thank the reviewer for pointing out a lack of clarity in our presentation. The experimental repeats for each measurement do indeed match, i.e. the measurement of the DV border matches the same adjacent 4-cell pair in the same embryo and in total 5 distinct embryos were analyzed for each experiment. In the preliminary revision of the manuscript, we explain this detail of the experimental design in the figure legend. In the preliminary revision, we also determined the ratios of DV/AP cell interfaces for b-Cat and E-Cad and added this quantification as panel 6C and 6E for a clearer presentation of the data.

Minor notes: Page 4, missing comma after "For example"

Response: The text was edited accordingly.

Page 4, "inevitable" does not make sense in this context

Response: We eliminated ‘inevitable’ and replaced it with ‘critical’ to better indicate the importance of Canoe protein for germband elongation.

Page 7, lines 6-7 - The localization of Src42A in control should be described in more detail and more clearly here.

Response: In the preliminary revised manuscript, we extended our description of the distribution of Src42A in more detail pointing out its dynamics and differential distribution at distinct plasma membrane domains.

Supplemental Fig S1 - Fig. S1D: Based on the head structure and the segmental grooves, the embryo shown here is close to late stage 13/early stage 14, not stage 15. - Fig S1E: It will be helpful if the predicted protein band and non-specific bands are indicated by arrows/arrowheads in the figure.

Response:

We thank the referee for their careful observation of the embryonic stage. We agree that the embryo was actually a younger stage. In the preliminary revision, we replaced the images with an example of an older stage. We will also add clear annotations as arrows to clearly mark the specific protein bands in Fig. S1E.

Page 7, lines 21-22 - "Src42A was slightly enriched at the AP interface" - To argue that, quantification should be provided.

Response:

We thank the referee for pointing out a qualitative statement that we made with regard to the distribution of Src42A at the AP cell interfaces. In the preliminary revision of the manuscript, we present an additional quantification of the imaging data of Src42A immunolabeling. In Figure S4A-C, we now present a quantification of the enrichment of Src42A at the tricellular junctions. In addition, the new Fig. S4D,E shows a quantification of the planar polarized distribution of Src42A at the AP cell interfaces.

Figure 1 - Fig. 1B: Src42A levels should be compared between control (Src42A/+) and Src42A/Src42A for each stage. It currently shows a comparison between Src42A/Src42A of stages 10 and 15.

Response:

We thank the referee for the comment. As indicated in our response to referee #1, the point of this analysis was to (1) provide evidence for the specificity of our new anti-Src42A antibody and (2) to demonstrate the presence of substantial material contribution of Src42A protein in zygotic mutant. We do not see the advantage to provide a detailed developmental Western-blot analysis, but we provide data in Suppl. Mat Fig S1E showing that the level of Src42A is unimpaired in stage 6 zygotic src42A[26-1] homozygous mutant embryos.

• Fig. 1B: The figure legend says, "dotted line represents mean value and error bars," but there are no dotted lines shown in the figure. Also, what p-value is for ****? It should be mentioned in the figure legend. It also says Src42A levels were normalized against E-Cad intensity here (stages 10 and 15). They have shown that E-Cad levels are affected in Src42A RNAi during gastrulation (Fig. 6). Is E-Cad not affected in Src42A26-1 zygotic mutants at stages 10 and 15?

Response:

We thank the referee for pointing out inaccuracies in the presentation and the description of Fig.1B. In the preliminary revision, we emphasized the marks on the graph and provide p-values throughout. Regarding the E-Cadherin levels: E-cadherin levels were altered in src42A RNAi knockdown embryos, but not in zygotic mutants, even at later developmental stages.

Page 8, line 14 - "Embryos expressing TRiP04138 showed reduced hatching rates with variable penetrance and expressivity depending on the maternal Gal4 driver used (Fig. 2B)" - Fig. 2B doesn't seem to be a right citation for this sentence.

Response:

We agree with the referee and in the preliminary revised manuscript we corrected the reference to the conclusion drawn from Figure 2A’, which does show the relationship of hatching rate to the various maternal Gal4 drivers.

• Fig. 2C: It will be helpful to indicate two other non-specific bands in the figure with arrows/arrowheads with a description in the figure legend.

Response:

In the preliminary revision, we added an arrow to mark the band specific for Src42A and asterisks to mark unspecific bands in Fig 2C.

Page 9, line 9 - This is the first time that the fast and the slow phases of germband extension are mentioned. As these two phases are used to compare the Src42A and Src42A Abl double RNAi phenotypes, they should be introduced and explained better earlier, perhaps in Introduction.

Response:

We thank the referee for pointing out that the two phases of germband extension were not introduced. We added a sentence to introduce and define the distinct phases of extension movements in the preliminary revision.

Fig. 3 - Fig. 3A: It will be helpful to mark the starting and the ending points of germband elongation with different markers (arrows vs. arrowheads or filled vs. empty arrowheads).

Response:

In the preliminary revision, we added distinct markers to indicate the start and endpoints of germband elongation to make this figure easier to read.

• Fig. 3C figure legend: R2 is wrongly mentioned in Fig. 3D, E. Also, R2 (coefficient of determination) needs to be defined either in the figure legend or Materials & Methods.

Response:

We thank the referee for pointing this misleading reference to us. In the preliminary revision we corrected the reference to R2 in Fig,3D,E and will describe the definition of R2 in the figure legend.

• Fig. 3D, E: statistical analysis is missing.

Response:

In the preliminary revision, we included a statistical analysis of the data (see ref #1). We changed the figure to indicate the data sets that were analyzed and added the p-values to the figure legend.

• Fig. 3G and H should be cited in the text.

Response:

In the preliminary revision, we added references to Fig 3G,H in the result section to the annotation of Fig.3F).

• Fig. 3F: It should be mentioned that the heat map is shown for pY20 signals in the figure legend, with an intensity scale bar in the figure.

Response:

In the preliminary revision, we added an intensity scale bar to the figure panel and mentioned the relationship to the PY20 signal.

Fig. 7A: Arrows can be added to mark the delayed germband extension.

Response:

In the preliminary revision, we added arrows to mark the anterior and posterior extent of the germband.

Fig. 8A: It should be mentioned that the heat map is shown for E-Cad signals in the figure legend, with an intensity scale bar in the figure.

Response:

In the preliminary revision, we added an intensity scale to the heat map and mention the relationship to the E-cadherin signal in the figure legend.

Fig. S3G: An arrowhead can be added to the gel image to indicate the band described in the legend.

Response:

In the preliminary revision, we added an arrow to help annotating the Src42A-specific bands on the Western blot.

• Fig. 9B: Arrow/arrowheads can be added to show the absence of the signals in the nurse cells.

Response:

In the preliminary revision, we added markers to help recognizing the reduced signal in the nurse cells and the oocyte.

• Fig. 9C: Indicate the ending point of the germband extension by arrows.

Response: In the preliminary revision, we added arrows to mark the anterior and posterior extent of the germband.

Regarding comments from reviewer #3:

Minor notes: Page 4, missing comma after "For example"

Response: The text was edited accordingly.

Page 4, "inevitable" does not make sense in this context Response:

In the preliminary revision, we eliminated ‘inevitable’ and replaced it with ‘critical’ to better indicate the importance of Canoe protein for germband elongation.

Description of analyses that authors prefer not to carry out

Referee #1 point2 and Referee#2 minor comment figure 1. Both referees suggest that figure 1 AB should include earlier developmental stages according to the stages looked at in the RNAi knockdown experiment.

Response:

The referees’ comments are likely based on a misunderstanding. The data that the reviewer are referring to present analyses of the zygotic phenotype of embryos homozygous for the src42A26-1 loss of function allele. They are not related to the maternal RNAi knockdown experiments, but were meant to demonstrate the existence and extent of a maternal pool of Src42A protein, that persists even to late stages in development. The maternal knockdown mutants are analyzed in detail at the appropriate stages in Fig. 2.

As described in our response above, we don’t feel that a detailed developmental stage Western analysis of wildtype and src42A26-1 embryos would provide significant additional insights. As mentioned in our response above, data for an earlier developmental stage (before germband elongation, as requested by the referees, were provided in Suppl. Fig. S1E.

Referee #1 Point 6) Figure 8E - showing images of multiple tAJs, rather than z-slices of a single vertex, would better support the claim here, as the assertion is that Src42a levels are different between control and sdk RNAi conditions, and not that it varies in the z-dimension.

Response:

The image series of Fig. 8E shows one representative example of multiple tAJs that have been imaged for this experiment (n=6 for wild type and n=10 for sdk RNAi). We think that the presentation of Z-slices for this experiment is important as the protein distribution needs to be considered for a larger area along the apical-lateral cell interface. In addition the quantification of the data for multiple tAJs was presented in Fig. 8F,G as a graph. We would therefore rather not change this figure in the revised manuscript.

Referee #3 suggests that anti MyoII staining should accompany the analysis of tension measurements in the germband.

As this analysis has already been performed by Tamada et al. 2021, we decided not to reproduce these data, but rather extend the analysis towards tension measurements, which support the findings by Tamada et al. 2021 on a functional level. We do not see the added value of adding MyoII labeling.

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

Chandran et al. report on the function of Src42A during cell intercalation in the early Drosophila gastrula. They create a Src42A-specific antibody (there are two Src genes in the fly genome) and examine the localization of Src42A and observe a planar-polarized distribution at cell interfaces. They then measure cell-contractile dynamics and show that T1 contraction is slower after Src42A disruption. The authors then argue that Src42A functions in a parallel pathway to the Abl protein, and that E-cadherin dynamics (turnover) is altered in Src42A disrupted embryos. Src function at these stages has been studied previously (though not to the degree that this study does), and in some respects the manuscript feels a little preliminary (please label figures with figure number!), but after editing this should be a polished study that merits publication in a developmentally-focused journal.

1. Does the argument that Src42A has two functions fully make sense? Myosin II function is known to affect E-cadherin stability (and vice versa), so it seems that Src42A could affect both MyoII and Ecad by either decreasing Myosin II function/engagement at junctions or by destabilizing Ecad.
2. One obvious question that arises is the nature of cleavage defects that are mentioned that happen previously to intercalation. For example, is E-cad normal prior to intercalation initiating? How specific are the observed defects to GBE?
3. Pg. 10, "the shrinking junction along the AP axis strongly reduces its length with an average of 1.25 minute" - what is this measurement? How much is "strongly"?
4. Also pg. 10, "the AP junction was not markedly reduced after 1 minute" - what is the criteria for this statement? X%? 1 minute is very specific, it feels like how much of a reduction/non-reduction should also be specific.
5. It seemed odd to mention altered myosin levels but then skip over a measurement of myosin in favor of an indirect measurement such as interface recoil. Again (point 1), it seems that changes in Myosin II recruitment could cause changes in Ecad turnover.

Minor notes:

Page 4, missing comma after "For example"

Page 4, "inevitable" does not make sense in this context

Significance

This study gives a more detailed perspective on how Src proteins (Src42A in Drosophila) control epithelial stability and the contraction of specific surfaces of epithelial 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 #2

Evidence, reproducibility and clarity

Summary:

Chandran et al. investigate the role of Src42A in axis elongation during Drosophila gastrulation. Using maternal RNAi and CRISPR/Cas9-induced germline mosaics, they revealed that Src42A is required to contract junctions at anterior/posterior cell interfaces during cell intercalations. Using time-lapse imaging and image analysis, they further revealed the role of Src42A in E-Cad dynamics at cell junctions during this process.

By analyzing double knockdown embryos for Src42A and Abl, they further showed that Src42A might act in parallel to Abl kinase in regulating cell intercalations. The authors proposed that Src42A is involved in two processes, one affecting tension generated by myosin II and the other acting as a signaling factor at tricellular junctions in controlling E-Cad residence time. Overall, the data are clear and nicely quantified. However, some data do not convincingly support the conclusion, and statistical analyses are missing for an experiment or two. Methods for several quantifications also need improvement in writing. Also, several figures (Figures 6-8) do not match the citation in the text and need to be corrected.

Page and line numbers were not indicated in the manuscript. For my comments, I numbered pages starting from the title page (Title, page 1; Abstract, page 2, Introduction, pages 3-6; Results, pages 7-14; Discussion, pages 15-18; M&M, 19-23; Figure legends, 28-30) and restarted line numbers for each page. For Figures 6-8 that do not match the citation in the text, I still managed to look at the potentially right panels. All the figure numbers I mention here are as cited in the text. My detailed comments are listed below.

Major comments:

1. b-Cat/E-Cad signals at the D/V and A/P junctions in Src42Ai (Figs. 5-6). These data are critical for their major conclusion and should be demonstrated more convincingly.

In Fig. 5A, the authors said, "When the AP border was cut, the detached tAJs moved slower in Src42Ai embryos compared to control (Fig. 5A)". However, even control tAJs do not seem to move that much in the top panels, and I found the images not very convincing.

In Fig. 6A, b-Cat signals look fuzzier and dispersed and have more background signals in the control, compared to the Src42Ai background. Also, b-Cat signals in the control image do not seem to show enrichment at the D/V border, as shown in Tamada et al., 2012.

In Fig. 6B, C, it is not clear how the intensity was measured and how normalization was done. Was the same method used for these quantifications as "Protein levels at bicellular and tricellular AJs" on pages 21-22? Methods should be written more explicitly with enough details.

Does each sample (experimental repeat) for the D/V border in Fig. 6B match the one right below for the A/P border in Fig. 6C? It should be clearly mentioned in the figure legend. The ratio of the DV intensity to AP intensity will better show the compromised planar polarity of the b-Cat/E-Cad complex. 2. Based on the genetic interaction between Src42A and Abl using RNAi (Fig. 7), the authors argue that Src42A and Abl may act in parallel. However, the efficiency of Abl RNAi has not been tested. It can be done by RT-PCR or Abl antibody staining. Also, the effect of Abl RNAi alone on germband extension should be tested and compared with Src42A & Abl double RNAi embryos. I expect the experiments can be done within a few weeks without difficulty.

Minor comments:

Page 2, line 14 - The abbreviation for tAJs was not introduced before.

Page 7, line 6 - A reference should be cited for the Src42A26-1 allele.

Figure 1 - Fig. 1B: Src42A levels should be compared between control (Src42A/+) and Src42A/Src42A for each stage. It currently shows a comparison between Src42A/Src42A of stages 10 and 15. - Fig. 1B: The figure legend says, "dotted line represents mean value and error bars," but there are no dotted lines shown in the figure. Also, what p-value is for ****? It should be mentioned in the figure legend. It also says Src42A levels were normalized against E-Cad intensity here (stages 10 and 15). They have shown that E-Cad levels are affected in Src42A RNAi during gastrulation (Fig. 6). Is E-Cad not affected in Src42A26-1 zygotic mutants at stages 10 and 15?

Page 7, lines 6-7 - The localization of Src42A in control should be described in more detail and more clearly here.

Supplemental Fig S1

• Fig. S1D: Based on the head structure and the segmental grooves, the embryo shown here is close to late stage 13/early stage 14, not stage 15.
• Fig S1E: It will be helpful if the predicted protein band and non-specific bands are indicated by arrows/arrowheads in the figure.

Page 7, lines 21-22

• "Src42A was slightly enriched at the AP interface" - To argue that, quantification should be provided.

Page 8, line 14

• "Embryos expressing TRiP04138 showed reduced hatching rates with variable penetrance and expressivity depending on the maternal Gal4 driver used (Fig. 2B)" - Fig. 2B doesn't seem to be a right citation for this sentence.

Fig. 2

• Fig. 2B: Higher magnification images of the defective cytoplasm can be shown as insets.
• Fig. 2C: It will be helpful to indicate two other non-specific bands in the figure with arrows/arrowheads with a description in the figure legend.
• Fig. 2E: A simple quantification of the penetrance of cuticle defects in Src42A mutants and RNAi will be helpful, as shown in Fig. S3.

Page 9, line 9

• This is the first time that the fast and the slow phases of germband extension are mentioned. As these two phases are used to compare the Src42A and Src42A Abl double RNAi phenotypes, they should be introduced and explained better earlier, perhaps in Introduction.

Fig. 3

• Fig. 3A: It will be helpful to mark the starting and the ending points of germband elongation with different markers (arrows vs. arrowheads or filled vs. empty arrowheads).
• Fig. 3G and H should be cited in the text.
• Fig. 3C figure legend: R2 is wrongly mentioned in Fig. 3D, E. Also, R2 (coefficient of determination) needs to be defined either in the figure legend or Materials & Methods.
• Fig. 3D, E: statistical analysis is missing.
• Fig. 3F: It should be mentioned that the heat map is shown for pY20 signals in the figure legend, with an intensity scale bar in the figure.

Fig. 7A: Arrows can be added to mark the delayed germband extension.

Fig. 8A: It should be mentioned that the heat map is shown for E-Cad signals in the figure legend, with an intensity scale bar in the figure.

Fig. S3G: An arrowhead can be added to the gel image to indicate the band described in the legend.

Fig. 9

• Fig. 9A: Magnified views of the cytoplasmic clearing can be added as insets.
• Fig. 9B: Arrow/arrowheads can be added to show the absence of the signals in the nurse cells.
• Fig. 9C: Indicate the ending point of the germband extension by arrows.

Page 14, lines 9-10: More explicit description of the phenotype rather than just "stronger compared to Src42Ai" will be helpful.

Significance

This work revealed the role of Src42A in regulating germband extension. A previous study suggested the roles of Src42A and Src64 in this developmental process using a partial loss of both proteins (Tamada et al., 2021). Using different approaches, the authors demonstrated a role of Src42A in regulating E-Cad dynamic at cell junctions during Drosophila axis elongation. Most of the analyses were done with maternal knockdown using RNAi, but they successfully generated germline clones for the first time and confirmed the RNAi phenotypes. Overall, this work contains important and exciting novel findings.

This work will be of general interest to cell and developmental biologists, particularly researchers studying epithelial morphogenesis and junctional dynamics.

I have expertise in Drosophila genetics, epithelial morphogenesis, imaging, and quantitative image analysis.

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

Highest priority:

1. The Src42A knockdown and germline clone experiments both cause defects in cellularization (Fig. 2B and 9A), which could result in differences in the state of the blastoderm epithelium (cell size, cell number, structural integrity, organization, etc.) between the experimental and control conditions. In addition, Src42A knockdown appears to affect the size and shape of the egg (Fig. 9A and 9C). The manuscript would be strengthened if the authors included data to demonstrate that the initial structure of the epithelium is mostly normal (quantifications of cell size, number, etc.) in the Src42A RNAi condition, as this would bolster the argument that germband extension, rather than due to indirect effects resulting from the cellularization defects. The authors may have relevant data to do this on-hand, for example using data associated with figures 1, 3, 6, and 9.
2. There is a discrepancy in the staging of embryos used between some of the analyses, which make it hard to interpret some of the data. For example, characterization of the knockdowns in Fig. 1A and B are based on stages 10 and 15, whereas the majority of the paper is focused on earlier stages 6 - 8 during germband extension (e.g., Fig. 1D). The analysis for Fig. 1B would be more meaningful if it was done on the same stages used for subsequent phenotypic analysis so they can be directly compared.
3. There is incongruence between figures in terms of which junctional pools (bAJs vs. tAJs) of beta-catenin and E-cadherin are quantified that makes it difficult to draw comparisons between analyses. For example pTyr levels are examined for both bAJs and tAJs in Figure 3, however, only tAJs are considered in Fig. 8. Similarly, in some cases planar cell polarity is considered (e.g., comparison of levels at AP vs DV bAJs in Fig. 6 and 9), and in other cases (e.g. Fig. 8) it is not.

Lower priority:

1. Introduction, 2nd paragraph - The modes of cell behaviours described to drive cell intercalation leaves out another clear example in the literature - Sun et al., 2017 - which describes a basolateral cell protrusion-based mechanism. While the authors cite this paper later, leaving it out when summarizing the state of the field misrepresents the current knowledge of the range of mechanisms responsible.
2. 'defective cytoplasm' - this term is confusing, and could perhaps be replaced with 'cellularization defect', or something similar.
3. Tests of statistical significance are not uniformly applied across the figures. For instance, Figures 3G + H indicate statistical significance, but Fig. 3D + E do not. Performing statistical tests throughout the paper, or clearly articulating a rationale when they are not used, would strengthen the manuscript. Specifically, the authors should consider this for Fig. 3D + E, and Fig. 7D + E, to support their arguments that rates of germband extension are different between conditions.
4. Page 12 - "We found that Src42A showed a distinct localization at the tAJs (Fig. 1B)": Figure 1B shows a quantification of levels at bAJs, not tAJs.
5. Figure 8 - in my opinion, using a FRAP or photoconversion approach would be a more convincing demonstration of differences in E-cadherin residency times / turnover rate than time-lapse imaging of E-cadherin:GFP alone. Authors should decide whether this improvement is worth the investment.
6. Figure 8E - showing images of multiple tAJs, rather than z-slices of a single vertex, would better support the claim here, as the assertion is that Src42a levels are different between control and sdk RNAi conditions, and not that it varies in the z-dimension.

Significance

The manuscript by Backer et al. examines the function of Src42A in germband extension during Drosophila gastrulation. Prior studies in the field have shown that Src family kinases play an important role in the early embryo, including cellularization (Thomas and Wieschaus 2004), anterior midgut differentiation (Desprat et al. 2008), and germband extension (Sun et al. 2017; Tamada et al. 2021). In this study, the authors showed that Src42A was enriched at adherens junctions and was moderately enriched along junctions with myosin-II. They then showed that maternal Src42A depletion exhibits phenotypes, starting with cellularization and including a defect in germband extension. The authors focus on defects in germband extension and found that Src42A was required for timely rearrangement of junctions and that the Src42A RNAi phenotype is enhanced by Abl RNAi. Finally the authors show that E-cadherin turnover is affect by Src42A depletion.

Overall, this study provided a higher resolution description of how Src42A regulates the behavior of junctions during germband extension. I thought the authors conclusions were well supported by the data and represent new insight in the field.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

Uveal melanoma is the most common primary intraocular malignancy in adults. About 50% of patients develop metastases, being the liver the most common place for them. Despite over 50 years of study, little progress has been done in efficacious treatments. In this report, the authors aimed at a better understanding of the mechanistic drivers for cancer aggressiveness and poor overall survival at the metastatic stage. To accomplish this, the authors performed a series of elegant genomics and transcriptomics analyses and identified molecular aberrations in miR16, which has been previously associated with other malignancies. The authors demonstrated that high level of miR16 sponges inversely correlate with poor overall survival. As a reference and validation, they are using the TCGA data analysis. Lastly, the authors generated a signature for survival prediction based on 4 genes, which was confirmed using an independent study.

The methodology was very elegant. The appropriate analyses were done.

Significance

This manuscript provides incremental knowledge to the field. In the last 5 years there are many manuscripts addressing different transcription factors or miRNA molecules and their role in different cancers. Uveal melanoma is an orphan disease with high unmet need. Prognostication is highly valuable, however; it is the treatment where we need the most attention.

The authors did elegant studies to demonstrate the relevance of miR16. This is not part of the standard of care, but the prognostic tool of the selected 4 genes, could be very helpful. I wish they could have included a sentence on the impact in the field.

The response to the following questions can make the manuscript more robust: Description of the TCGA - how many of the primary tumors had clinically detectable metastases? This is important as you are describing a potential companion diagnostic testing to predict OS. We need additional information on the UM cells, which can be found in literature. It is necessary for the audience to understand which of these cell lines come from patients that die from metastases, which ones had additional malignancies. There are UM cell lines - commercially available ? for which the primary and metastatic line were developed from the same patient. Those are very helpful, especially if one of the objectives is demonstrating poor OS due to metastatic disease. That was not clear from the manuscript. Levels of miR16 - What were the copy numbers in healthy patients? Do we need them relative to 501 Mel? or should we compare to a system that is not dysregulated? Why were the DROSHA KO be lung cancer cells? HCT116? Why choosing this cell line?

Overall, I consider this a good manuscript and with some tweaking it can be better.

Referees cross-commenting

Based on the different reviews and the added note, we all agree the manuscript needs work and is not ready yet to be accepted. We all agreed that needs to be re-structured as we all mentioned about key missing pieces in the writing. It will be of help to the authors to go back to the guidelines to know the max words and extend their manuscript.

The timing needed is might not be too relevant, as we all agreed it needs work.

Thank you to my reviewers/colleagues as they pointed out things very comprehensively.

Agreed with their comments.

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 an extremely short report that identified a potentially new mechanism as to how miR-16 may be involved in uveal melanoma (UM). In this report the authors used previous data, that identified that miR-16 is involved in UM, to gain a more comprehensive understanding of the mechanism involved. miR-16 was selected as a candidate due to its location in chromosome 3, which UM patients of the have chromosome 3 monosomy. The group evaluated the transcriptome following miR-16 overexpression in a single cell line, and as others often find with miRNAs, there was a cohort of up- and down-regulated RNAs. Figure 1 is a summary of the workflow, indicating the number of up- and down-regulated RNAs with validations performed. In figure 2 the authors identified 2 cohorts based on the previous expression data, one of which defines a more at-risk group. While the date presented in Figures 1 and 2 is interesting overall, the conclusions are mostly drawn from a supplemental figure. The data in this manuscript needs to be revaluated and the most critical included in the main figures. If done soe, and rewritten appropriately this study could have a bigger impact. As is, this study would be of low to moderate interest. There are also some instances, early on, where conclusions are overstated. Overall the text is lacking in description to conduct a thorough review, the authors fail to provide an introduction to put the study in the context of the field, and they provide a one-sentence discussion. Some of these issues are defined below, but due to the lack of description of the studies, and overstatements made, this is not a thorough review.

Major issues:

The main text and Figure 1 inappropriately referred to upregulated RNAs as miR-16 sponges. At this point in the manuscript, these are nothing more that upregulated RNAs.

The text is often vague and lacks discretion that is essential for the Reviewer to understand the study. Some sentences are fragments and are not clear. Some (but not all) examples of difficulties encountered while attempting the review are indicated below as well.

Figure 1E legend. "in function of the experimental workflow detailed". "In function" does not make sense in this sentence. It is not clear what the authors are referring to. Similarly, "In function of their expression...." In function aging is inappropriate and the sentence is not clear. Also "MRE for miRNA response element". Perhaps the authors mean "MRE = miRNA response element." Also, the text "10arbouring" is included int this legend. In brief, Figure 1E is extremely difficult to evaluate with the poor text in the legend.

Multiple figures have odd wording to indicate biological replicates. This needs to be clarified in better, complete sentences.

For Figure S2, Fold induction is indicated as a %. This is not appropriate. It is either a fold change or a change in %, but not both.

How is the experiment done in Figure S3A a "kinetic" experiment? There is no kinetic analysis here.

Details of the critical biotinylating study are completely lacking. And since this is the most critical experiment that gets to the main point of the study, this should be well defined and part of the main figures in the manuscript. It should not be in the supplemental information. All of Figure S2 should be in the main part of the manuscript.

All axis in Figure S2B are not labeled appropriately. Enrichment is used, however not all RNAs are enriched. Perhaps fold change would be a better and more accurate name. Those on the left side (in blue) are depleted, not enriched.

For Figure S3C, the authors should change "blue ones" to "solid blue bars indicate". Same for S3E

For Figure S3D, "logo" is inappropriate and not the correct term. This is a consensus sequence, not a logo.

How did the authors make the conclusion that the upregulated RNAs are targets of miR-16 if they do not have a canonical miR-16 binding sites? They could easily be indirect RNAs that are elevated post miR-16 exposure. The authors do not validate that the cohort of RNAs upregulated are indeed miR-16 targets. Thus, the overstatement of "sponge" RNAs (Figure 1H) or even "target" RNAs (Figure 1F) without appropriate validation is overstated. Simply doing the biotinylating study is still not enough to conclude direct interactions of these RNAs with miR-16. These can be false positives, that are not well controlled for due to poor selection of a control RNA.

Figure 1G is not large enough to see, nor is the inclusion of it clear in the text of the manuscript.

Figure 1H, referring to upregulated RNAs, post miR-16 expression as sponges is inappropriate unless they have all been validated as miR-16 sponges. These could merely be RNAs that are indirectly upregulated following miR-16 transfection, and their upregulating following miR-16 overexpression has been validated. However their miR-16 sponging activity has not been validated. Similarly for Figure 2A.

Figure 2 is poorly defined. This needs clarification and the font should be increased. There are also "..." in the figure legend which is inappropriate. Many things are not defined such as "CN" which the reviewer is assuming means copy number. Also the colors and description for "Yes/Dead/Male" are not clear. What are these? How are they relevant?

For Figure 2B legend, what is meant by miR-16 "in function"?

The authors should show the level of upregulation of miR-16 following transfection for all experiments where miR-16 is transfected.

For all figures where qRT-PCR was conducted, what are RNAs normalized to. This should be indicated on the axis and/or in the figure legend. While in the methods section, this should also be present in the main body of text (ie. figures).

For Sup Figure 2A the authors indicate that RNA levels were compared to 501Mel. They should show the 501mel levels in the same graph. They also state that the absolutely copy number was determined from Norther Blot. As the authors likely know, quantification using qRT-PCR is much more quantitative than Northern. They should conduct qRT-PCR for the main cell line they are comparing to. The Northern is also not shown and the reference provided for it is for a Nature Review article, not for a study that shows a Northern blot.

It is not clear what the control RNA was for all the studies. Specifically for the biotinylated studies, the authors should use another miRNA, not a non-specific control. Because a control miRNA will also binds AGO and other miRNA-associated factors, non-specific binding due to these factors could be better controlled for. The non-specific RNA will not account for these factors.

Sup Fig 4 is missing details. What orientation is the consensus sequence shown in relative to the miRNA (5'-3' or 3'-5')? Other details are missing as well, this is just one example of many issues.

For Sup Fig 5A the CT values should be included. That gives a better direct comparison than a graph of something that is indicated as not determined. You cannot graph something that is not determined.

For Sup Fig 5C the font cannot be read it is too small.

For Sup Fig 5D, again, how much miR-16 is present when overexpressed. Would this amount be physiologically achievable?

The title is poor and not descriptive enough for the study. It reads more like the title for a review article.

Methods for siRNAs indicated kinetic as well. Not clear what kinetic data were acquired during this study.

Significance

Referees cross-commenting

I am in full agreement with the additional comments made by Reviewer #1. I however disagree with Reviewer #3 that the study was well conducted and "elegant". Based on multiple issues (many cooperated) between R#1 and R#2 I do not feel that this study is acceptable and will take an extraordinary amount of time to be acceptable for publication.

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:

The article presents interesting data regarding the role of miR-16 in the development and progression of uveal melanoma. The authors propose the analysis of miR-16 activity as a marker for uveal melanoma progression instead of miR-16 expression. To analyze the activity of miR-16 they propose a risk model developed around a 4 genes signature that was shown to be able to stratify uveal melanoma patients into low and high-risk groups. Unfortunately, the current version of the manuscript is hard to understand and to carefully evaluate due to a lack of structure and clear presentation of the experimental design, methods, results, and discussions.

Major Comments:

• The manuscript needs to be restructured into clear sections including a detailed introduction, materials and methods, results, discussions, and conclusion. The present format is hard to follow with information and results being spread across multiple documents and parts of the manuscript.
• The authors mention that the level of miR-16 reached after transfection is higher than that in the physio-pathological level, but there is no data to support this. I recommend adding a quantification of miR-16 levels achieved after transfection as part of the S2.
• The way of using references is confusing as it is not clear what was done in the results section and what work is cited from the literature. The results section should focus strictly on presenting the results achieved by the experiments while delineating clearly the work that was done in other articles.
• The discussion section needs to be extended to better present the role of specific investigated genes and proteins like PYGB and PTP4A3 in the development and progression of uveal melanoma.
• The experiments analyzing the sequestration of miR-16 at non-canonical sites are performed using the cell line HCT116 WT and DROSHA KO. HCT116 is a human colon adenocarcinoma cell line, a tumor with a completely different histology. These experiments should be performed also on a human uveal melanoma cell line in order to ensure consistency of the results.

Minor Comments:

• The manuscript is lacking consistency regarding the usage of abbreviations. These should be defined the first time when used in the text.
• The figure S2 needs to be adjusted. It is hard to understand in S2B how the statistical analysis was performed. I recommend representing each line with the two conditions side by side to increase clarity.
• When presenting the miR-16 interactome the data are spread in three different sources Figures S2, 1D-E, and Table S1 which makes it difficult to follow the images. I recommend presenting these data in the same figure.
• The manuscript requires English grammar and style editing. There are several words misspelled and phrases with a complicated syntax that makes it difficult to understand.
• The method of presenting the data in Figures 1 F and H needs to be reorganized. The current version makes it very hard to understand how the gene expression changed after miR-16 exposure.

Significance

The article presents important results regarding the role of miR-16 in uveal melanoma by an innovative approach analyzing the activity of miR-16 instead of its expression. The authors focus on the relationship between miR-16 sponges and targets and through a set of elegantly designed experiments they identify a set of 4 genes whose expression can be used as a risk predictor model in uveal melanoma. The audience of this article can be represented by both clinicians and researchers that could take advantage of the results presented. Also, the approach of the article can open new directions for other cancers and miRNAs.

Referees cross-commenting

Would go with a revision of the manuscript according to the comments

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Manuscript number: RC-2021-01016

Corresponding author(s): Dennis Klug

1. General Statements [optional]

Dear editor, dear reviewers,

thank you very much for the quick review of our manuscript as well as for the constructive criticism and the interesting discussion of our results. Reading the comments, we realized that we may have put too much emphasis on the in vivo microscopy of sporozoites and their interaction with the salivary gland. We believe that the generated mosquito lines can be used to address different scientific questions, the in vivo microscopy of host-pathogen interactions being only one of them. Because of this imbalance, and to address some of the reviewers' comments, we have partially rewritten the manuscript (particularly the introduction). At the same time, we have implemented additional data on the inducibility of the promoters used, as well as on the functionality of hGrx1-roGFP2 in the salivary glands. Furthermore, we created an additional figure to better present the expression patterns of trio and saglin promoters within the median lobe, and we expanded the section on in vivo microscopy of sporozoites. We hope that these results further highlight the significance of our study. Accordingly, we have also changed the title of the manuscript to „A toolbox of engineered mosquito lines to study salivary gland biology and malaria transmission” to indicate the broad applicability of the generated mosquito lines and we have included an additional co-author, Raquel Mela-Lopez, who conducted the redox analysis. We hope that these changes will adequately answer the questions of the reviewers and address any concerns they may have had. We look forward to hearing from you.

With our kind regards,

Dennis Klug

Katharina Arnold

Raquel Mela-Lopez

Eric Marois

Stéphanie Blandin

2. Point-by-point description of the revisions

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

**Summary**

This manuscript reports the generation and characterization of transgenic lines in the African malaria mosquito Anopheles coluzzii that express fluorescent proteins in the salivary glands, and their potential use for in vivo imaging of Plasmodium sporozoites. The authors tested three salivary gland-specific promoters from the genes encoding anopheline antiplatelet protein (AAPP), the triple functional domain protein (TRIO) and saglin (SAG), to drive expression of DsRed and roGFP2 fluorescent reporters. The authors also generated a SAG knockout line where SAG open reading frame was replaced by GFP. The reporter expression pattern revealed lobe-specific activity of the promoters within the salivary glands, restricted either to the distal lobes (aapp) or the middle lobe (trio and sag). One of the lines, expressing hGrx1-roGFP2 under control of aapp promoter, displayed abnormal morphology of the salivary glands, while other lines looked normal. The data show that expression of fluorescent reporters does not impair Plasmodium berghei development in the mosquito, with oocyst densities and salivary gland sporozoite numbers not different from wild type mosquitoes. Salivary gland reporter lines were crossed with a pigmentation deficient yellow(-) mosquito line to provide proof of concept of in vivo imaging of GFP-expressing P. berghei sporozoites in live infected mosquitoes.

**Major comments**

Overall the manuscript is very well written with a clear narrative. The data are very well presented. The generation of the transgenic mosquito lines is elegant and state-of-the art, and the new reporter lines are thoroughly characterized.

This is a nice piece of work that is suitable for publication, although the in vivo imaging of sporozoites is somewhat preliminary and would benefit from additional experiments to increase the study impact.

We would like to thank the reviewer for his/her appreciation of our manuscript. In the revised version, we have included additional experiments on in vivo imaging of sporozoites, which allowed us to quantify moving and non-moving sporozoites imaged under the cuticle of live mosquitoes. Although this is still a proof of concept, we believe that these new data provide novel interesting data and will better illustrate potential applications.

The reporter mosquito lines express fluorescent salivary gland lobes, yet the authors only provide imaging of parasites outside the glands. It would be relevant to provide images of the parasite inside the fluorescent glands.

We have now included images showing sporozoites inside the salivary glands in vivo in Figure 8C and discuss possible ways to further improve resolution and efficiency of the imaging procedure in lines 563-586.

The advantage of the pigmentation-deficient line over simple reporter lines is not clear, essentially due to the background GFP fluorescent in figure 5C. Imaging of GFP-expressing parasites should be performed in mosquitoes after excision of the GFP cassette under control of the 3xP3 promoter. This would probably allow to document the value of the reporter lines more convincingly.

Indeed, by incorporating two Lox sites in the transgenesis cassette, we designed the yellow(-)KI line to permit removal of the fluorescent cassette and completely exclude expression of the transgenesis reporter EGFP. Still, EGFP expression in the yellow(-)KI adults is restricted to the eye and ovary, as we show now in Figure 7 supplement 1D. In contrast, no EGFP fluorescence was observed in the thorax area (Figure 7 supplement 1D). Therefore, we believe that the benefit of removing the fluorescence cassette for this study is limited. Moreover, the generation of such a line would take at least 3-4 months before experiments could be performed. Nevertheless, we agree with the reviewer that removal of the fluorescence cassette would be instrumental for follow-up studies. To draw the reader's attention to this issue, we now discuss background fluorescence in lines 378-387.

Along the same line, it is unclear if the DsRed spillover signal in the GFP channel is inherent to the high expression level or to a non-optimal microscope setting. This is a limitation for the use of the reporter lines to image GFP-expressing parasites.

We have discussed this problem with the head of the imaging platform at our institute, and we believe that it is not a problem that occurs due to incorrect settings. Rather, it seems to be due to the significant expression differences of the two fluorescence reporters used. We agree with the reviewer that this is a limitation and discuss the problem now in lines 416-412 and 565-567.

The authors should fully exploit the SAG(-) line, which is knockout for saglin and provides a unique opportunity to determine the role of this protein during invasion of the salivary glands. This would considerably augment the impact of the study. In this regard, line 131 and Fig S3E: why is there persistence of a PCR band for non-excised in the sag(-)EX DNA?

We definitely share the reviewer's enthusiasm about saglin and its role in parasite development in mosquitoes. We have thoroughly characterized the phenotype of sag(-) lines with respect to fitness and Plasmodium infection. These results are described in a spearate manuscript currently in peer review and available as a preprint on bioRxiv (https://doi.org/10.1101/2022.04.25.489337). Furthermore, in the revised manuscript, we have included additional data on the transcriptional activity of the saglin promoter with respect to the onset of expression and blood meal inducibility (Figure 2). In addition, we have included a completely new Figure 3 to highlight the spatial differences in transcriptional activity of the saglin promoter compared with the trio promoter. These new data are commented in lines 206-276.

There might be a misunderstanding in the interpretation of the genotyping PCR. The PCR shown in Figure 1 – figure supplement 3, displays PCR products for different genomic DNAs (sag(-)EX, sag(-)KI and wild type) using the same primer pair. „Excised“ refers to sag(-)EX while „non excised“ refers to sag(-)KI and „control“ to wild type. Primers were chosen in a way to yield a PCR product as long as the transgene has integrated, only the shift in size between „excised“ and „non excised“ indicates the loss of the 3xP3-lox fragment. We have now changed the labeling of the respective gel in Figure 1 – figure supplement 3 to make this clearer.

Did the authors search for alternative integration of the construct to explain the trioDsRed variability?

We validated trio-DsRed cassette insertion in the X1 locus by PCR. The only way to rule out an additional integration of the transgene would be whole genome sequencing, which we did not perform. Still, we believe that the observed expression patterns are due to locus-specific effects of the X1 locus. Indeed, several lines of evidence point in this direction: (1) transgenesis was realized using the phage Φ31 integrase that promotes site-specific integration (attP is 38bp long and very unlikely to occur as such in the mosquito genome) and for which we never detected insertion in other sites in the genome for other constructs inserted in X1 and other docking lines; (2) additional unlinked insertions would have been easily detected during the first backcrosses to WT mosquitoes we perform in order to isolate the transgenic line and homozygotise it; (3) we have often observed variegated expression patterns for other transgenes located in the X1 locus in the past, leading us to believe that this locus is subjected to variegation influencing the expression of the inserted promoters. Usually, the variation we observe is simpler (e.g. strong and weak expression of the fluorescent reporter placed under the control of the 3xP3 promoter in the same tissues where it is normally expressed), but some promoters are more sensitive to nearby genomic environment than others, which we believe is the case for trio. Finally, should there be additional insertions of the transgenesis cassette in the genome, they should all be linked to the X1 locus as we would otherwise have detected them in the first crosses as mentioned above, which is unlikely. Thus, although very unlikely, we cannot exclude a single additional and linked insertion possibly explaining the high/low DsRed patterns, but variegation would still be required to explain other patterns. We have mentioned this alternative explanation in the manuscript in lines 522-524.

Line 254-255. Does the abnormal morphology of SG from aapp-hGrx1-roGFP2 result in reduced sporozoite transmission?

This is an interesting question. For future experiments, it could indeed be important to test if the transmission of sporozoites by the generated salivary gland reporter lines is not impaired. However, the quantification of the number of sporozoites in aapp-hGrx1-roGFP2 expressing salivary glands did not reveal any significant differences from the wild type (Figure 5 – figure supplement 1B) and would definitely be sufficient to infect mice. As we have no evidence for reduced invasion of sporozoites in the salivary glands of aapp-hGrx1-roGFP2 and of the DsRed reporter lines, no good reason to believe that the expression of fluorescent proteins would interfere with parasite transmission, and as we produced these lines as tools to follow sporozoite interaction with salivary glands, we have not performed transmission experiments.

Of note, we have now included images of highly infected salivary glands of all reporter lines in Figure 5 – figure supplement 1D to confirm that expression of the respective fluorescence reporter does not interfere with sporozoite invasion. Also we have not observed that sporozoites do not invade salivary gland areas displaying high levels of hGrx1-roGFP2.

**Minor comments**

-Line 51: sporogony rather than schizogony

Schizogony was replaced with sporogony.

-Line 56: sporozoites are not really deformable as they keep their shape during motility

This sentence was removed.

-In the result section, it is not clearly explained where constructs were integrated.

We have now included the sentence „...with an attP site on chromosome 2L...“ (line 173) and the respective reference (PMID: 25869647) to give more information about the integration site.

Line 106 and 434-435: for the non-expert reader, it is not clear what X1 refers to, strain or locus for integration?

X1 refers to both, the locus and the docking line. We have rephrased the beginning of the result section (previously line 106) to give more information about the integration site as mentioned above.

-Line 112-115: the rational of integrating GFP instead of SAG is not clearly explained here, but become clearer in the discussion (line

We have slightly rephrased the sentence to better explain the reasoning for this procedure (lines 182-184).

-Line 140: FigS2A instead of S3A

This mistake was corrected in the revised manuscript.

-Perhaps mention that GFP reporters (SG) might be useful to image RFP-expressing parasites.

We have now included an image of the aapp-hGrx1-roGFP2 line infected with a mCherry expressing P. berghei strain in Fig. 7D.

-Line 236: the authors cannot exclude integration of an additional copy (as mentioned in the discussion line 367-368).

As discussed above, we removed „..as a single copy...“ and introduced the possibility of an additional integration linked to X1 (lines 522-524).

-Line 257-258. The title of this section should be modified as SG invasion was not captured.

The title was rephrased. It reads now „Salivary gland reporter lines as a tool to investigate sporozoite interactions with salivary glands” (line 356-357).

-Line 287: remove "considerable number" since there is no quantification.

This was removed. In addition, we included new data in this section of the manuscript and rephrased the results accordingly (lines 406-427).

-Line 400-402: Klug and Frischknecht have shown that motility precedes egress from oocysts (PMID 28115054), so the statement should be modified.

Thank you for this suggestion. The passage was modified accordingly.

-Line 404: remove "significant number" since there is no quantification.

This section was rephrased and the phrase "significant number" was removed (lines 406-427).

-Line 497: typo "transgenesis"

The typo was correct in the revised manuscript.

-FigS1: add sag-DsRed in the title

Thank you for spotting this inconsistency, we corrected this mistake (line 1134).

-Stats: Mann Whitney is adequate for analysis in fig 2C but not 2B, where ANOVA should be used (more than 2 groups).

We have performed now an one-way-ANOVA test and adapted figure and figure legend accordingly.

Reviewer #1 (Significance (Required)):

This work describes a technical advance that will mainly benefit researchers interested in vector-Plasmodium interactions. Invasion of salivary glands by Plasmodium sporozoites is an essential step for transmission of the malaria parasite, yet remains poorly understood as it is not easily accessible to experimentation. The development of transgenic mosquitoes expressing fluorescent salivary glands and with decreased pigmentation provides novel tools to allow for the first time in vivo imaging in live mosquitos of the interactions between sporozoites and salivary glands.

Reviewer's expertise: malaria, Plasmodium berghei, genetic manipulation, host-parasite interactions

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

The first achievements of the Klug et al. study are the (i) genetical engineering of the Anopheles coluzzii mosquitoes reared in insectarium, that stably express distinct fluorescent reporters (DsRed and hGrx1-roGFP2 and EGFP) under the putative "promoters" of genes reported to encode proteins expressed differentially in the pluri-lobal salivary glands(Sg) of anthropophilic blood-feeding adult females, (ii) the analysis of the promoter activity - based on the selected fluorescent reporter - with a primary focus on the salivary gland/Sg (including at the Sg lobe level) of the adult female but also considering the preimaginal developmental time with larvae and pupa samples. Of note, some data confirm the already reported time-dependent and blood meal-dependent promoter activity for the related Anopheles species. The last part presents preliminary dataset on live imaging of Plasmodium berghei sporozoites with the aim of highlighting the usefulness of these A. coluzzii transgenic

lines to better understand how the rodent Plasmodium sporozoites first colonize and then settle as packed cells in Sg acinar host cells.

**Major comments**

The two first objectives presented by the authors have been convincingly achieved with (i) the challenging production of four different lines expressing different single or double reporters chosen by the authors (and appropriately presented in the result text and figure sections), (ii) the careful analysis of the spatiotemporal expression of the DsRed reporter under two "promoters" studied and with regards to the blood feeding event parameter. However, if the reason why the authors have put so much effort in the production of their transgenic mosquitoes is (and as mentioned) to provide a significant improved setting enabling the behavioral analysis of sporozoites upon colonization and survival in the Sg, it seems this part is kind of limited. Likely in relation with this perception is the fact I found the introductory section often confusing and not enough direct to the points: in particular distinguishing the rationale from the necessity to produce appropriate models, and clarifying what is/are the added value(s) offered by these new transgenic lines models when compared to what exist (in Anopheles stephensi) with specific evidence that argue for this knowledge gain. At this stage, it is unfortunately not clear to me, what is the bonus of imaging the Plasmodium fluorescent sporozoites in hosts with fluorescent salivary gland lobes if one can not monitor key events of the Sg-sporozoite interaction that were not reachable without the fluorescent mosquito lines. Furthermore, it should be better explained why the rodent Plasmodium species has been chosen rather Plasmodium falciparum (or other human species) for which A. coluzzii is a natural host; may be just mentioning that this study would serve as a proof of concept but bringing real biological insights would be fine.

We would like to thank the reviewer for his/her evaluation of our manuscript, which has helped us clarify our manuscript on several points. Our goal here was a proof of concept demonstrating potential applications for the fluorescent salivary gland reporter lines and for the low pigmented yellow(-) line we generated. In vivo imaging of sporozoites in salivary glands is one possible application that we intended to use as proof-of-concept, but we tailored the manuscript too restrictively with this aim in mind and neglected other applications as well as characterization of the biology of salivary glands in general. To improve this, we have included further data on the blood inducibility of the promoters tested (Figure 2), the functionality of roGFP2 in the salivary glands (Figure 5), and the use of the generated lines in the examination and definition of expression patterns of salivary gland proteins in vivo (Figure 6). Accordingly, we have adjusted the entire manuscript to adequately describe all the results presented. We have also rephrased major parts of the abstract and the introduction to better describe the impact of salivary gland biology on the transmission of pathogens, and to explain the anatomy of salivary glands in more detail.

We agree with the reviewer that it would be desirable to show direct salivary gland-sporozoite interactions in vivo. Still we believe that having mosquito lines expressing a fluorescent marker in the salivary gland as well as weakly pigmented mosquitoes are a first step to make this visualization possible, although we cannot provide a lot of quantitative data about this interaction yet.

1- The three genes and gene products selected by the authors should definitively be more systematically explained, which means for example the authors need to introduce the different mosquito species and the parasite-mosquito host pairs they are then referring to for the promoter/encoded proteins of their interest. In the same vein, I did not find any information as to the choice of the mosquito species (A. Coluzzii) for the current work. I was curious to know what is the advantage since better knowledge was available with Anopheles stephensi with respect to (i) Saglin and its promotor activity, (ii) aap driven dsRed expression (lines already existing) and (iii) sporozoite-gland interaction.

We have largely reworded the introduction to clarify the rationale for selecting these three promoters while providing a better understanding of salivary gland biology in general.

The choice of the mosquito species depends, in our opinion, strongly on the perspective and on the experiments to be performed. We agree with the reviewer that the malaria mosquito A. stephensi is a widely used model, based on its robustness in breeding and its high susceptibility to P. berghei and P. falciparum infections. However, in these cases, both vector-parasite pairs are to some extend artificial. Indeed, although it is also a vector of P. falciparum in some regions, A. stephensi mostly transmits P. vivax that cannot be cultured in vitro. Thus research efforts on this vector-parasite pair is limited. Also, due to the emerging number of observed differences between Anopheles species and their susceptibility to Plasmodium infection and transmission, more research has recently been conducted on African mosquito species. This effect is also reinforced by the fact that P. falciparum, unlike all other Plasmodium species infecting humans, causes the most deaths, making control strategies for species from the A. gambiae complex such as A. coluzzii particularly important. As a result, the number of available genetic tools in A. coluzzi/gambiae has overpaced A. stephensi. These include mosquito lines with germline-specific expression of Cas9 for site-directed transgenesis, lines expressing Cre for lox-mediated recombination, and several docking lines. Such tools are, as far as we know, not available in A. stephensi and were essential in reaching our objectives. Docking lines are of particular interest because they allow reliable integration into a characterized locus, which is an advantage over random transposon-mediated integration. Random insertion sites have generally not been characterized in the past, which can cause problems since integrations regularly occur in coding sequences. Docking lines also enable comparison of different transgenes as they are all integrated in the same genetic environment, which does not ensure some expression variation as illustrated in our manuscript. For all these reasons, we have thus chosen to work with A. coluzzii.

Concerning the use of the murine malaria parasite P. berghei instead of the human one P. falciparum, there are two reasons that motivated our choice. (1) For in vivo imaging of sporozoites, we needed a parasite line that is strongly fluorescent at this stage, and there is no such line existing for P. falciparum. Actually, there is no fluorescent P. falciparum line able to efficiently infect A. coluzzii reported thus far, as reporter genes have all been inserted in the Pfs47 locus that is required by P. falciparum for A. coluzzii colonization. (2) Imaging P. falciparum infected mosquitoes, especially with sporozoites in their salivary glands, requires to have access to a confocal microscope in a biosafety level 3 laboratory. Hence our objective here was indeed to provide a proof of principle of in vivo imaging of sporozoites in the vicinity or inside salivary glands using our engineered mosquitoes, and to provide a first analysis of this process using P. berghei as a model of infection. Nevertheless, we agree with the reviewer that the goal should be to work as close as possible to the human pathogen.

Despite the wide range of topics that this study touches on, we want to try and keep the manuscript as concise as possible. Therefore, we have not discussed the advantages and disadvantages of the different vector-parasite pairs and ask the reviewer to indulge us in this.

2- To help clarifying the added value of the present study, introducing the species names of the mosquito and the Plasmodium that serve as a model would be appreciated.

We have included now the name of the used Plasmodium species in line 361. At this position we also give now more details about the transgene this line is carrying. We mention the used mosquito species A. coluzzii now at different positions in the manuscript (e.g. lines 52, 162 and 177).

3- Since a focus is the salivary gland of the blood feeding female Anopheles sp., a rapid description of the glands with different lobes and subdomains the results and figure 1 nicely refer to, would help in the introduction.

We explain now the anatomy of female and male mosquito salivary glands in the introduction (lines 119-123). The different lobes are now also indicated in the salivary gland images shown in several figures including Figure 1.

4- That description could logically introduce the few proteins actually identified with lobe specific or cell domain specific expression (apical versus basal side, intracellular or surface expose, vacuole, duct...) profiles. The context with regards to sporozoite biology would then easily validate the "promoter choice". As a minor remark, I miss the reason why the authors wrote " the astonishing degree of order of the structures (referring to the packing of sporozoites within the Sg acinars) raise the question whether sporozoite can recognize each other". Please clarify since packing/accumulation can be passive due to cell mechanical constraints and explain what this point has to see with the question and experimental work proposed here?)

We thank you for this suggestion. We have reworded key parts of the introduction to make the reasons for using the three selected promoters clearer. We also mention now other proteins expressed in the salivary glands which have been characterized in more detail because of their effect on blood homeostasis (e.g. anticoagulants) (lines 136-139).

The mention of stack formation of salivary gland sporozoites served only to clarify that almost nothing is known about the behavior of sporozoites within the salivary glands in vivo to explain why new methods are needed to make these processes visible. We have now reworded this passage to make this clearer, and we also mention that stack formation could also occur due to mechanical constraints, as suggested by the reviewer (lines 101-102, 106-110).

5- The selection of hGrx1-roGFP2 is quite interesting and justified but there is then no use of this reporter property in the preliminary characterization of the Sg and Sg-sporozoite interaction. Could the authors provide such characterization?

We have now implemented data testing the functionality of hGrx1-roGFP2 in the salivary glands. We also show qualitatively that the redox state of glutathione does not change upon infection with P. berghei sporozoites (Figure 6). We now describe and discuss these new data in lines 337-354.

6- Figure 1: it would be nice to add in the legend at what time the dissection/imaging has been made (age, blood feeding timing?). I would also omit the double mutant trio-Dsred/aapDsred in the main figure (may be supplemental) since the two single mutants Dsred separately together with the double mutant (with different fluorescence) already provide the information. I would suggest to regroup the phenotypic presentation of the transgenic line made in the KI mosquitoes (current figure 5) in the main figure 1.

We have now added the missing information about the age of dissected mosquitoes and their feeding status in the legend of Figure 1. We also thank the reviewer for the suggestion to replace one image displaying aapp and trio promoter activity in trans-heterozygous mosquitoes with an image of the pigment deficient mutant yellow(-)KI. Still, due to the changes made to the manuscript based on the reviewers comments in general, we have now implemented new data highlighting the functionality of the generated salivary gland reporter lines investigating the redox state of glutathione as well as the expression pattern of the saglin and trio promoters at the single cell level (see Figure 3 and 6). Therefore it would no longer seem logical to introduce the yellow(-)KI mutant in Figure 1 while further data on this mutant are provided in the last two figures of the manuscript and discussed later in the manuscript (Figure 7 and 8). In addition we believe that co-expression of different transgenes (carrying fluorescent reporters) in the median and the distal lobes could potentially be interesting for certain applications. We believe that readers who might actually be interested in combining both transgenes in a cross would like to see the outcome to better evaluate the usefulness before experiments are planned and performed. This is especially true because localization as well as expression strength may differ between different fluorescence reporters while using the same promoter (e.g. the hGrx1-roGFP2 construct appears less bright and more localized to the apex of the distal-lateral lobes than dsRed, while expression of both reporters is driven by the aapp promoter in aapp-hGrx1-roGFP2 and aapp-DsRed, respectively).

7- Figure 2:

1. a) Is there anything known on the Sgs' size change overtime. It seems that between day 1 and 2 there is an increase of size and volume as much as I can evaluate the volume (Fig S4). Could that mean that there is increase in cell number in the lobes and therefore more cells expressing the transgene which would account for the signal intensity increase rather than more transcripts per cell? Thank you for this interesting question. The changes in the morphology of the salivary glands in Anopheles gambiae following eclosion have been studied in detail by Wells et al., 2017 (PMID: 28377572) which we cite now in the introduction (line 122-123). According to this reference, cell counts of the salivary gland are not changing upon emergence of the adult mosquito. However, we agree with the reviewer that the glands appear smaller and differ in morphology directly after eclosion. We noted that glands of freshly emerged females are more „fragile“ during dissections and lack secretory cavities, as reported by Wells et al., 2017. We believe that the increase in size occurs through the formation and filling of the secretory cavities which has been reported to take place within the first 4 days after emergence (Wells et al., 2017). This observation is in accordance with our observations that the promoters of the saliva proteins AAPP and Saglin display only weak activity after hatching, or, in the case of TRIO are not yet active directly after emergence. The timing of the formation of the secretory cavities is also in agreement with our time course experiment (Figure 2) which shows a strong increase in fluorescence intensity in dissected glands within the first 4 days after emergence.

2. b) why choosing 24h after the blood meal to assess promoter activity in the Sgs? Do we have any information on how the blood meal impact on the Sgs'development. At this time anyway the sporozoites are far from being made. Yosshida and Watanabe 2006 mentioned at significant decrease of Sg proteins post-blood feeding. Could the authors detail their rationale based on what the questions they wish to address Thank you for this question. Unfortunately, the data available in the literature on this topic are very sparse, so we could only refer to few previous publications. The decision to quantify the fluorescence signals as early as 24 hours after blood feeding was based on Yoshida et al, Insect Mol. Biol, 2006, PMID: 16907827. The authors of this study generated the first salivary gland reporter line in A. stephensi by using the aapp promoter sequence to drive DsRed expression, and showed by qRT-PCR that DsRed transcripts increase 1-2 days after blood feeding compared to controls. Consistent with this observation and because we were concerned that putative changes in protein levels would only be visible for a short period of time, we began quantification one day after feeding. Since we observed significant changes in fluorescence intensity for the aapp-DsRed and sag(-)KI lines 24 hours after blood feeding, we retained the experimental setup and did not change it further. Nevertheless, we agree with the reviewer that different time points could help determine how long the effect lasts, and whether trio expression might also be regulated by blood feeding, but at a later time point. Still, our main objective here was to validate that the ectopic expression of DsRed driven by the aapp promoter in the aapp-DsRed line was indeed induced upon blood feeding as previously reported (PMID: 16907827). This experiment allowed us to confirm the inducibility of aapp in a different way and to show for the first time that saglin, but not trio, is induced one day after blood feeding. Our transgenic lines could be used for follow-up studies investigating the inducibility of salivary gland-specific promoters by different stimuli, or after infection with Plasmodium sporozoites. For example, for trio, transcription has been shown to increase after infection of the salivary gland by Plasmodium (PMID: 29649443).

8- Figure 3: The figure is quite informative in terms of subcellular localization. Concerning the section "Natural variation of DsRed expression in trio-DsRed mosquitoes", I think it could be shortened because because it is a bit out of the focus the study.

We agree with the reviewer that this part of the manuscript sticks a bit out and is not perfectly in line with the remaining results because it doesn’t deal with the salivary gland. Still, we would like to emphasise that in this work, we particularly want to show possible applications of the generated mosquito lines to address unanswered questions in host-parasite interactions and salivary gland biology. As a result, this manuscript establishes potentially important tools. For this reason, we feel it is important to mention the natural variation in DsRed expression, as this natural variation can have a significant impact on crossing schemes (especially with lines inheriting other DsRed-marked transgenes) and experiments (e.g. visualizing DsRed expression by western blot in larval and pupal stages). Furthermore, it is important for the use of the line to show that the transgene is inserted only once, at the expected location, which we try to emphasize with figure 4 – figure supplement 1 and figure 4 – figure supplement 2.

We would also like to note that transgenesis in Anopheles is a relatively young field of research and altered expression patterns of ectopically used promoters have rarely been described so far, although this could have major implications e.g. in the case of gene drives. Therefore, we hope that the data shown will bring this previously neglected observation more into focus and highlight the importance of accurate characterization of generated transgenic mosquito lines.

9- In contrast the last section of live imaging of P. berghei sporozoites in the vicinity and within salivary gland should be expanded. The 2 sentences summarizing the data are quite frustrating "We also observed single sporozoites moving actively through tissues in a back and forth gliding manner (Fig. 6B, Movie 3) or making contact with the salivary gland although no invasion event could be monitored"

We have now implemented new data and extended Figure 8 showing the results of the in vivo imaging in a qualitative manner. We have rephrased the result and discussion section accordingly.

10- I am aware of the technical difficulties to perform live imaging of sporozoite on whole mosquitoes, even when the salivary gland lobe under observation is closely apposed to the cuticle but that seems to be the final aim of the authors. I looked very carefully to the three movies and I am sorry but at this stage I could not make meaningful analysis out of them, and could not agree with the conclusions: for instances, the authors specify that sporozoites were undergoing back and forth movements (movie 3) but I do not see that and do not see the Sg contours in the available movies? The authors should also add bar and time scales to their movies. Having an in-depth description with regards to the sub-domain marked by a relevant reporter would strengthen the study, even if images are not collected in the whole mosquito to get higher resolution.

We thank the reviewer for this comment. We have to admit that parasite imaging in fluorescent salivary glands in vivo is an ambitious goal given the complex biological system we are working with. We believe that the system presented in our manuscript is a first and important step to enable the analysis of the interaction of sporozoites with salivary glands, although in-depth analysis will require further optimization and considerable time, especially to generate quantitative data. Therefore, we now downstate the significance of our results in this respect and changed the title accordingly. Still, we also provide a more detailed analysis of the data we have already collected (Figure 8 and lines 406-427). Because we focus on the analysis of sporozoites in the thorax area in the revised manuscript, the outlines of the salivary gland are not necessarily visible in the images.

I am not sure I understand the relevance of this quite condensed sentence in the text. Could the authors rephrase and expand if they wish to keep the issues they refer to. "The sporozoites' distinctive cell polarization and crescent shape, in combination with high motility, allows them to „drill" through tissues". I would stress more on the main unknown in terms of sporozoite-Sg interactions and the need to get right models for applying informative approaches (i.e. here, imaging).

We thank you for this suggestion. The sentence mentioned has been removed in its entirety. We have also adjusted the text accordingly and reworded most of the introduction to make the narrative clearer (lines 91-119).

Of note, it could help to point that the "Sgs is a niche in which the sporozoites which egress from the oocyst could mature and be fully competent when co-deposited with the saliva into the dermis of their intermediary hosts"

We have now implemented a similar sentence in the introduction (lines 93-98).

Reviewer #2 (Significance (Required)):

1- Clear technical significance with the challenging molecular genetics achieved in the mosquito A. coluzzii.

2- More limited biological significance: fair analysis and gain of knowledge of spatio-temporal of reporter expression under the selected promoter but limited significance of the final goal analysis which concerns the Plasmodium sporozoite biology once egressed from oocysts

As stated above, we changed the title to place the focus on the engineered mosquito lines.

3- Previous reports cited by the authors have used the DsRed reporter and the aap promoter in another Anopheles (i.e. A. stephensi, Yoshida and Watanabe, Insect Mol Biol, 2006; Wells and Andrew, 2019) which is also a natural host and vector for human Plasmodium spp.) with significantly more resolutive 3D visualization of GFP-fluorescent P. berghei but in dissected salivary glands and not in whole mosquitoes. The Wells and Andrew publication entitled "Salivary gland cellular architecture in the Asian malaria vector mosquito Anopheles stephensi" in Parasite Vectors, 2015 would deserve to be reference and described.

Thank you very much for this suggestion. We considered citing Wells and Andrews (PMID: 26627194). However, this reference focuses very specifically on the subcellular localization of AAPP and shows only highly magnified sections of immunostained dissected and fixed salivary glands. Working only with the AAPP promoter, we felt it important to refer to the previously observed expression pattern along the entire salivary gland, as shown in Yoshida and Watanabe (PMID: 16907827). Nevertheless, we have cited two other publications by Wells and Andrews (PMID: 31387905 and 28377572) at various points in the manuscript.

4- Audience: I would say that this work should be of interest of mostly scientists investigating Plasmodium biology (basic and field research) or in entomology of Diptera.

5- To describe my fields of expertise, I can refer to my extensive initial training in entomology including at one point in the genetic basis of mosquito-virus interaction. I have also been working for more than 20 years in the field of Apicomplexa biology (Plasmodium and Toxoplasma) and I have long-standing interest in live and static high-resolution imaging.

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

Klug et al. generated salivary gland reporter lines in the African malaria mosquito Anopheles coluzzii using salivary gland-specific promoters of three genes. Lobe-specific reporter activity from these promoters was observed within the salivary glands, restricted either to the distal lobes or the medial lobe. They characterized localization, expression strength and onset of expression in four mosquito lines. They also investigated the possibility of influences of the expressed fluorescent reporters on infection with Plasmodium berghei and salivary gland morphology. Using crosses with a pigmentation deficient mosquito line, they demonstrated that their salivary gland reporter lines represent a valuable tool to study the process of salivary gland colonization by Plasmodium parasites in live mosquitoes. SG positioning close to the cuticle in 20% of females in this strain is another key finding of this study.

The key findings from this study are largely quite convincing. The authors have created a suite of SG reporter strains using modern genetic techniques that aid in vivo imaging of Plasmodium sporozoites.

Vesicular staining within salivary acinar cells should be stated as "vesicle-like" staining unless a co-stain experiment in fixed SGs is conducted using antisera against the marker protein(s) and antisera against a known vesicular marker (e.g. Rab11). It may also be possible to achieve this in vivo using perfusion of a lipid dye (e.g. Nile Red), but this is not necessary. As is, in Fig. 3A, there are images in which it appears that the vesicle-like staining is located both within acinar cells' cytoplasm and in the secretory cavities (e.g. Fig. 3A: aapp-DsRed bottom and middle), and this is fine, but should be more inclusively stated. Fixed staining of the reporter strain SGs would allow for clarification of this point. In previous work, other groups have observed vesicle-like structures in both locations (e.g. PMID: 33305876).

Thank you very much for this suggestion. Indeed, when we observed the vesicle-like localization, we had similar ideas and considered investigating the identity of the observed particles in more detail. Ultimately, however, we concluded that the localization of DsRed does not play a critical role in the use of the lines as such and believe that a more detailed investigation of the trafficking of the fluorescent protein DsRed is beyond the scope of this study.

We have thus followed the suggestion of the reviewer and now use the phrase „vesicle-like“ throughout the manuscript. In addition, we extended the discussion on the different localizations observed and presented some explanations that might have led to this observation. We also included a new reference that investigated the localization of AAPP using immunofluorescence (PMID: 28377572).

Morphological variation is extensive among individual mosquito SGs, thought to impact infectivity, and well documented in the literature. The manuscript should be edited to make it much clearer (e.g. n = ?) exactly how many SGs, especially in microscopy experiments, were imaged before a "representative" image was selected from each data point and in any additional experiment types where this information is not already presented. Figure S8 is an example where this was done well. Figure 3A-B is an example where this was not well done. All substantial variation (e.g. "we detected a strangulation..." - line 189) across individual SGs within a data point should be noted in the Results. Because of the genetics and labor involved, acceptable sample sizes for minor conclusions may be small (5-10), but should be larger for major conclusions when possible.

Thank you for this comment. We have improved this point by specifying precisely the number of samples and of repetitions in the respective figure legends. For example, we have now quantified the proportion of moving sporozoites and report both the number of sporozoites evaluated and the number of microscopy sessions required (see Figure 8).

Thank you for this comment. We have improved this point by specifying precisely the number of samples and of repetitions in the respective figure legends. For example, we have now quantified the proportion of moving sporozoites and report both the number of sporozoites evaluated and the number of microscopy sessions required (see Figure 8). Regarding Figure 3, fluorescence expression and localization in salivary gland reporter lines was actually very uniform in each line. We added the following sentence in the legend of revised figures 3 and 5: “Between 54 and 71 images were acquired for each line in ≥3 independent preparation and imaging sessions. Representative images presented here were all acquired in the same session”.

Sporozoite number within SGs has been shown to be quite variable across the infection timeline, by mosquito species, by parasite strain, in the wild vs. in the lab, and according to additional study conditions. The authors mention that the levels they observed are consistent with their prior studies and experience, but they did not utilize the reporter strains and in vivo imaging to support these conclusions, instead relying on dissected glands and a cell counter. It is important for these researchers to attempt to leverage their in vivo imaging of SG sporozoites for direct quantification, likely using the "Analyze Particles" function in Fiji. The added time investment for this additional analysis would be around two weeks for one person experienced in the use of the imaging software.

Thank you for this interesting suggestion. Indeed, it would be beneficial to use an imaging based approach to quantify the sporozoite load inside the salivary glands. We already used „watershed segmentation“ in combination with the „Analyze Particles“ function in Fiji on images of infected midguts to determine oocyst numbers. Still, we believe this analysis cannot be applied to images of infected salivary glands mainly because of differences in shape and location of the oocyst and sporozoite stages. Sporozoites inside salivary glands form dense, often multi-layered stacks. Because of this close proximity, watershedding cannot resolve them as single particles which could subsequently be counted. This creates an unnecessary error by counting accumulations of sporozoites as one, likely leading to an underestimation of actual parasite numbers. Furthermore, given that the proximity issue could be resolved e.g. by performing infections yielding lower sporozoite densities, another problem would be that infected salivary glands prepared for imaging are often slightly damaged leading to a leak of sporozoites from the gland into the surrounding. These leaked sporozoites are likely not included on images which would then be used for analysis, potentially leading again to an underestimation of counts. Since these issues are circumvented by the use of a cell counter, we believe that this method is still the method of choice in acquiring sporozoite numbers.

Nevertheless, we can understand the reviewer's concern that counts performed with a hemocytometer do not reflect the variability in the sporozoite load of individual mosquitoes. To highlight that all generated reporter lines can have high sporozoite counts, we have now included images of highly infected salivary glands for each line in Figure 7D.

This manuscript is presented thoughtfully and such that the data and methods could likely be well-replicated, if desired, by other researchers with similar expertise.

The statistical analysis is appropriate for the experiments conducted. It is currently unclear if some experiments were adequately replicated. That information should be added to the paper throughout where it is missing.

We do appreciate your comments on our efforts to give all required information for other laboratories to replicate our experiments. We have added the missing information about the number of independent experiments in the respective figure legends wherever appropriate.

Studies from multiple groups should be more thoroughly referenced when the authors are describing the "vesicle-like" staining patterns observed in SGs from reporter strains (e.g. Fig. 3A). Is this similar to the SG vesicle-like structures observed previously (e.g. PMIDs: 28377572, 33305876, and others)?

Thank you for this comment. We did not discuss this observation in detail in the first version of our manuscript because the observed localization was rather unexpected, as DsRed was not fused to the AAPP leader/signal peptide. The observed localization is therefore difficult to explain, however, we have expanded the discussion on this (lines 465-482) and now cite one of the proposed references (PMID: 28377572, lines 468-469).

There are minor grammar issues in the manuscript text (e.g. "Up to date" should be "To date"). The figures are primarily presented very clearly and accurately. One minor suggestion: In cases such as Fig. S2A images 3 and 6, where some of the staining labels are very difficult to read, please move all labels for the figure to boxes located directly above the image.

We are sorry for the grammatical errors we have missed in the first version of our manuscript. We have now performed a grammar check over the whole manuscript. We have also increased the font size of the captions in the above figures and tried to make them better readable by moving the captions over the images.

The data and conclusions are presented well.

Reviewer #3 (Significance (Required)):

This report represents a significant technical advance (improved in vivo reporter strain and sporozoite imaging), and a minor conceptual advance (active sporozoite active motility), for the field.

This work builds off of previous SG live imaging studies involving Plasmodium-infected mosquitoes (e.g. Sinnis lab, Frischneckt lab, etc.), addressing one of the major challenges from these studies (reliable in vivo imaging inside mosquito SGs).

This work will appeal to a relatively small audience of vector biology researchers with an interest in SGs. Many in the field still see the SGs as intractable, instead choosing to focus on the midgut due to ease of manipulation. Perhaps work like this will spark new interest in tangential research areas.

I have sufficient expertise to evaluate the entirety of this manuscript. Some descriptors of my perspective include: bioinformatics, SG molecular biology, mosquito salivary glands, microscopy, RNA interference, SG infection, and SG cell biology.

Reviewer #4 (Evidence, reproducibility and clarity (Required)):

Klug et al generated transgenic mosquito lines expressing fluorescent reporters regulated by salivary gland specific promoters and characterized fluorescent reporter expression level over the time, subcellular localization of fluorescent reporters, and impact on P. berghei oocyst and salivary gland sporozoite generation. In addition, by crossing one of the lines (aapp-DsRed) with yellow(-) KI mosquitoes, they open up the possibility to perform in vivo visualization of salivary glands and sporozoites.

Overall the generation and characterization of these transgenic lines is well-done and will be helpful to the field. However, there are several concerns with the in vivo imaging data shown in Figure 6, which does not convincingly show fluorescent sporozoites in the lobe or secretory cavity of a fluorescent salivary gland lobe. This needs to be addressed. Points related to this concern are outlined below:

(1) Although the authors mention that the DsRed signal was strong enough to see with GFP channel, it would be more appropriate to show that the DsRed signal from salivary glands and GFP channel image co-localize.

We now show a merge of the GFP and DsRed signal in Figure 7 – figure supplement 2 The yellow appearance of the salivary gland in the merge likely indicates the spillover of the DsRed signal into the GFP channel. In addition we discuss the issue in lines 416-412 and 565-567.

(2) Mosquitoes were pre-sorted using the GFP fluorescence of the sporozoites on day 17-21. From figure 4B, median salivary gland sporozoite number was about 10,000 sporozoites/mosquito on day 17-18. However, in Figure 6A there are no sporozoites in the secretory cavities. They should be able to see sporozoites in the cavities at this time. Can the authors confirm that they can visualize sporozoites in secretory cavities in vivo and perhaps show a picture of this.

This is entirely correct. We also examined mosquitoes for the presence of sporozoites in the salivary glands and wing joints prior to imaging, as shown in Figure 7B and Figure 7 – figure supplement 2A, to increase the probability that sporozoites could be observed. Nevertheless, the area of the salivary gland that comes to the surface is often small and limited to a few cells that can be imaged with good resolution. Unfortunately, these same cells were often not infected although other regions of the salivary glands must have been very well infected based on the previously observed GFP screening (Figure 7B). In addition, with the confocal microscope available to us, we struggled to achieve the necessary depth to image sporozoites in the cavities of the salivary gland cells. For this reason, we were often able to detect a strong GFP signal in the background, but not always to resolve the sporozoites sufficiently well. Still, we have now included an image showing sporozoites in salivary glands (Figure 8C). However, we believe that the method can be further improved to be more efficient and provide better resolution. We discuss possible ways to further improve the imaging in lines 563-586.

(3) There is no mention of the number of experiments performed (reproducibility) and no quantification of the imaging data. In the results (line 287-288), the authors state that sporozoites are present in tissue close to the gland and sometimes perform active movement. How can this be? Do they believe these sporozoites are on route to entering? More relevant to this study would be a demonstration that they can see sporozoites in the secretory cavities of the salivary gland epithelial cells, this should shown. If they have already performed a number of experiments, I would suggest to do quantification of the number of sporozoites observed in defined regions . The mention that sporozoites are moving is confounded by the flow of hemolymph. How do they know that the sporozoites are motile versus being carried by the hemolymph. Perhaps it's premature to jump to sporozoite motility in the mosquito when they haven't even shown sporozoite presence in the salivary glands.

Thank you very much for this comment. We have followed the suggestions of the reviewer and have now quantified the behavior of sporozoites in the thorax area of the mosquito. For the analysis, we only considered sporozoites that could be observed for at least 5 minutes. This analysis revealed that 26% of persistent sporozoites performed active movements, which in most cases resembled patch gliding previously described in vitro. We adjusted the results section accordingly. In addition, we have changed the figure legend to accurately indicate the number of experiments performed. Likewise, we now also provide an image of sporozoites that we assume are located in the salivary gland (Figure 8C). Although we have not yet been able to image and quantify vector-sporozoite interactions extensively (further improvements would be required, as mentioned previously), we believe these results illustrate the potential of the transgenic lines.

(4) In vivo imaging has been performed with the mosquito' sideways. Was this the best orientation? Have you tried other orientations like from the front (Figure 5B orientation).

It is true that in the abdominal view as shown in Figure 7B the fluorescence in the salivary glands is very well visible. This is mainly due to the fact that in this area the cuticle is almost transparent and therefore serves as a kind of "window". Nevertheless, the salivary glands are not close to the cuticle in this position, which makes good confocal imaging impossible. Imaging always worked best where the salivary gland was very close to the cuticle, and this was always laterally. However, there were differences in the position of the salivary glands in individual mosquitoes, which also led to slight differences in the imaging angle.

Overall, the text is easy to follow and I have only few suggestions.

Thank you for this comment.

In the result section, the authors describe the DsRed expression during development of mosquito (line 194-236) after they describe subcellular localization of fluorescent reporters. I felt the flow was disrupted. Thus, this part (line 194-236) could summarize and move to line 135. In this way, the result section flow according to the main figures.

Thank you very much for this suggestion. We have considered your idea, but based on the changes we have made in response to reviewer comments and new data implemented in the form of two new figures, we believe the current order in the results section is more appropriate. The rationale was primarily to first characterize the expression of fluorescent reporters in the salivary glands of all lines before going into more detail on expression in other tissues of a single line. We then finish with potential applications like in vivo imaging of sporozoite interactions with salivary glands.

Also, and as mentioned previously (reviewer 2, point 8), we believe it is important to describe the variability of ectopic promoter expression at a given locus with sufficient details, as this has not been characterized thus far despite its importance.

In the result section, text line 186-190, the authors describe the morphological alternation of salivary gland in aapp-hGrx1-roGFP2. I would suggest to mention that this observation was only in one of lateral lobe. (I saw that it was mentioned in the figure legend but not in the main text.)

We believe there has been a misunderstanding. The morphological alteration in salivary glands expressing aapp-hGrx1-roGFP2 was observed in all distal-lateral lobes to varying degrees (quantification in Figure 6E). To include as many salivary glands as possible in the quantification and because in some images only one distal-lateral lobe was in focus, only the diameter of one lobe per salivary gland was measured and evaluated. We have now revised the legend to prevent further misunderstandings.

In the discussion section, author discuss localization of fluorescent reporters (line 322-331). When I looked at aapp-DsRed localization pattern (Figure 3A), the pattern looked similar to the previous publication by Wells et al 2017 (https://www.nature.com/articles/s41598-017-00672-0). This publication used AAPP antibody and stain together with other markers (Figure 4-7). This publication could be worth referring in the discussion section.

Thank you for this suggestion. According to the information available through Vectorbase, we did not fuse DsRed with any coding sequence of AAPP that could potentially encode a trafficking signal. Therefore, it is rather unlikely that the observed DsRed localization in our aapp-DsRed line and the localization observed by AAPP immunofluorescence staining in WT mosquitoes match. This is further exemplified by the cytoplasmic localization of hGrx1-roGFP2 in the aapp-hGrx1-roGFP2 line, where the reporter gene was cloned under the control of the same promoter. For this reason, we had not mentioned this reference in the first version of the manuscript. In the revised manuscript, we have included now the suggested reference (lines: 475-476) and extended the discussion on possible reasons which led to the observed localization pattern.

In the text, authors describe salivary gland lobes as distal lobes and middle lobe. It would be more accurate to refer to the lobes as the lateral and medial lobes. The lateral lobes can then be sub-divided into proximal and distal portions. I would suggest to use distal lateral lobes, proximal lateral lobes and median lobe as other references use (Wells M.B and Andrew D.J, 2019).

Thank you for this suggestion. We have corrected the nomenclature for the description of the salivary gland anatomy as suggested throughout the manuscript.

Overall, the figures are easy to understand and I have following suggestions and questions.

Figure 1C) It is hard to see WT salivary gland median lobe. If authors have better image, please replace it so that it would be easier to compare WT and transgenic lines.

We have replaced the wild-type images of salivary glands in this figure and labeled the median and distal-lateral lobes accordingly (see Figure 1).

Figure 2) While it was interesting to observe the significant expression differences between day 3 and day 4, have you checked if this expression maintained over time or declines or increases (especially on day 17-21 when author perform in vivo imaging)?

Thank you for this interesting question. We have not quantified fluorescence intensities in mosquitoes of higher age. Nevertheless, we regularly observed spillover of DsRed signaling to the GFP channel during sporozoite imaging, suggesting that expression levels, at least in aapp-DsRed expressing mosquitoes, remain high even in mosquitoes >20 days of age (see Figure 8A). We also confirmed this observation by dissecting salivary glands from old mosquitoes, whose distal lateral lobes always showed a strong pink coloration even in normal transmission light (data not shown).

Figure 3A) There is no description of "Nuc" in figure legend. If "nuc" refers to nucleus, have you stained with nucleus staining dye (example, DAPI)?

Thank you for spotting this missing information in the legend. Initial images shown in this figure were not stained with a nuclear dye. To test whether the observed GFP expression pattern really colocalizes with DNA, we performed further experiments in which salivary glands from both aapp-hGrx1-roGFP2 and sag(-)KI mosquitoes were stained with Hoechst. We have now included these new data in Figure 3 - figure supplement 1. It appears that GFP is concentrated around the nuclei of the acinar cells, which makes the nuclei clearly visible even without DNA staining.

Figure 4B) The number of biological replicates in the figure and the legend do not match (In the figure, there are 3-5 data points and, in the legend, text says 3 biological replicates.)

Thank you for spotting this inconsistency. The number of biological replicates refers to the number of mosquito generations used for experiments. The difference is due to the fact that sometimes two experiments were performed with the same generation of mosquitoes using two different infected mice. We have clarified the legend accordingly to avoid misunderstandings.

Figure 4C) The number of data points from (B) is 5. However, in (C) only 4 data points are presented.

We have corrected this mistake. In the previous version, the results of two technical replicates were inadvertently plotted separately in (B) instead of the mean.

Figure 5) I would suggest to have thorax image of P. berghei infected mosquito to show both salivary glands and parasites.

Thank you for this suggestion. Images in Figure 7B (previously Figure 5) were replaced with an infected specimen to show salivary glands (DsRed) and sporozoites (GFP) together.

Reviewer #4 (Significance (Required)):

The transgenic lines that authors created have potential for in vivo imaging of salivary gland and sporozoite interactions. Since the aapp and trio lines have distinct fluorescence expression, they could help elucidate why sporozoites are more likely to invade distal lateral lobes compare to median lobe.

My areas of expertise are confocal microscope imaging, mosquito salivary gland and Plasmodium infection and sporozoite motility.

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 #4

Evidence, reproducibility and clarity

Klug et al generated transgenic mosquito lines expressing fluorescent reporters regulated by salivary gland specific promoters and characterized fluorescent reporter expression level over the time, subcellular localization of fluorescent reporters, and impact on P. berghei oocyst and salivary gland sporozoite generation. In addition, by crossing one of the lines (aapp-DsRed) with yellow(-) KI mosquitoes, they open up the possibility to perform in vivo visualization of salivary glands and sporozoites.

Overall the generation and characterization of these transgenic lines is well-done and will be helpful to the field. However, there are several concerns with the in vivo imaging data shown in Figure 6, which does not convincingly show fluorescent sporozoites in the lobe or secretory cavity of a fluorescent salivary gland lobe. This needs to be addressed. Points related to this concern are outlined below:

(1) Although the authors mention that the DsRed signal was strong enough to see with GFP channel, it would be more appropriate to show that the DsRed signal from salivary glands and GFP channel image co-localize.

(2) Mosquitoes were pre-sorted using the GFP fluorescence of the sporozoites on day 17-21. From figure 4B, median salivary gland sporozoite number was about 10,000 sporozoites/mosquito on day 17-18. However, in Figure 6A there are no sporozoites in the secretory cavities. They should be able to see sporozoites in the cavities at this time. Can the authors confirm that they can visualize sporozoites in secretory cavities in vivo and perhaps show a picture of this.

(3) There is no mention of the number of experiments performed (reproducibility) and no quantification of the imaging data. In the results (line 287-288), the authors state that sporozoites are present in tissue close to the gland and sometimes perform active movement. How can this be? Do they believe these sporozoites are on route to entering? More relevant to this study would be a demonstration that they can see sporozoites in the secretory cavities of the salivary gland epithelial cells, this should shown. If they have already performed a number of experiments, I would suggest to do quantification of the number of sporozoites observed in defined regions . The mention that sporozoites are moving is confounded by the flow of hemolymph. How do they know that the sporozoites are motile versus being carried by the hemolymph. Perhaps it's premature to jump to sporozoite motility in the mosquito when they haven't even shown sporozoite presence in the salivary glands.

(4) In vivo imaging has been performed with the mosquito' sideways. Was this the best orientation? Have you tried other orientations like from the front (Figure 5B orientation).

Overall, the text is easy to follow and I have only few suggestions.

In the result section, the authors describe the DsRed expression during development of mosquito (line 194-236) after they describe subcellular localization of fluorescent reporters. I felt the flow was disrupted. Thus, this part (line 194-236) could summarize and move to line 135. In this way, the result section flow according to the main figures.

In the result section, text line 186-190, the authors describe the morphological alternation of salivary gland in aapp-hGrx1-roGFP2. I would suggest to mention that this observation was only in one of lateral lobe. (I saw that it was mentioned in the figure legend but not in the main text.)

In the discussion section, author discuss localization of fluorescent reporters (line 322-331). When I looked at aapp-DsRed localization pattern (Figure 3A), the pattern looked similar to the previous publication by Wells et al 2017 (https://www.nature.com/articles/s41598-017-00672-0).

This publication used AAPP antibody and stain together with other markers (Figure 4-7). This publication could be worth referring in the discussion section.

In the text, authors describe salivary gland lobes as distal lobes and middle lobe. It would be more accurate to refer to the lobes as the lateral and medial lobes. The lateral lobes can then be sub-divided into proximal and distal portions. I would suggest to use distal lateral lobes, proximal lateral lobes and median lobe as other references use (Wells M.B and Andrew D.J, 2019).

Overall, the figures are easy to understand and I have following suggestions and questions. Figure 1C) It is hard to see WT salivary gland median lobe. If authors have better image, please replace it so that it would be easier to compare WT and transgenic lines.

Figure 2) While it was interesting to observe the significant expression differences between day 3 and day 4, have you checked if this expression maintained over time or declines or increases (especially on day 17-21 when author perform in vivo imaging)? Figure 3A) There is no description of "Nuc" in figure legend. If "nuc" refers to nucleus, have you stained with nucleus staining dye (example, DAPI)? <br /> Figure 4B) The number of biological replicates in the figure and the legend do not match (In the figure, there are 3-5 data points and, in the legend, text says 3 biological replicates.) Figure 4C) The number of data points from (B) is 5. However, in (C) only 4 data points are presented. Figure 5) I would suggest to have thorax image of P. berghei infected mosquito to show both salivary glands and parasites.

Significance

The transgenic lines that authors created have potential for in vivo imaging of salivary gland and sporozoite interactions. Since the aapp and trio lines have distinct fluorescence expression, they could help elucidate why sporozoites are more likely to invade distal lateral lobes compare to median lobe.

My areas of expertise are confocal microscope imaging, mosquito salivary gland and Plasmodium infection and sporozoite motility.

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

Klug et al. generated salivary gland reporter lines in the African malaria mosquito Anopheles coluzzii using salivary gland-specific promoters of three genes. Lobe-specific reporter activity from these promoters was observed within the salivary glands, restricted either to the distal lobes or the medial lobe. They characterized localization, expression strength and onset of expression in four mosquito lines. They also investigated the possibility of influences of the expressed fluorescent reporters on infection with Plasmodium berghei and salivary gland morphology. Using crosses with a pigmentation deficient mosquito line, they demonstrated that their salivary gland reporter lines represent a valuable tool to study the process of salivary gland colonization by Plasmodium parasites in live mosquitoes. SG positioning close to the cuticle in 20% of females in this strain is another key finding of this study.

The key findings from this study are largely quite convincing. The authors have created a suite of SG reporter strains using modern genetic techniques that aid in vivo imaging of Plasmodium sporozoites.

Vesicular staining within salivary acinar cells should be stated as "vesicle-like" staining unless a co-stain experiment in fixed SGs is conducted using antisera against the marker protein(s) and antisera against a known vesicular marker (e.g. Rab11). It may also be possible to achieve this in vivo using perfusion of a lipid dye (e.g. Nile Red), but this is not necessary. As is, in Fig. 3A, there are images in which it appears that the vesicle-like staining is located both within acinar cells' cytoplasm and in the secretory cavities (e.g. Fig. 3A: aapp-DsRed bottom and middle), and this is fine, but should be more inclusively stated. Fixed staining of the reporter strain SGs would allow for clarification of this point. In previous work, other groups have observed vesicle-like structures in both locations (e.g. PMID: 33305876).

Morphological variation is extensive among individual mosquito SGs, thought to impact infectivity, and well documented in the literature. The manuscript should be edited to make it much clearer (e.g. n = ?) exactly how many SGs, especially in microscopy experiments, were imaged before a "representative" image was selected from each data point and in any additional experiment types where this information is not already presented. Figure S8 is an example where this was done well. Figure 3A-B is an example where this was not well done. All substantial variation (e.g. "we detected a strangulation..." - line 189) across individual SGs within a data point should be noted in the Results. Because of the genetics and labor involved, acceptable sample sizes for minor conclusions may be small (5-10), but should be larger for major conclusions when possible.

Sporozoite number within SGs has been shown to be quite variable across the infection timeline, by mosquito species, by parasite strain, in the wild vs. in the lab, and according to additional study conditions. The authors mention that the levels they observed are consistent with their prior studies and experience, but they did not utilize the reporter strains and in vivo imaging to support these conclusions, instead relying on dissected glands and a cell counter. It is important for these researchers to attempt to leverage their in vivo imaging of SG sporozoites for direct quantification, likely using the "Analyze Particles" function in Fiji.

The added time investment for this additional analysis would be around two weeks for one person experienced in the use of the imaging software.

This manuscript is presented thoughtfully and such that the data and methods could likely be well-replicated, if desired, by other researchers with similar expertise.

The statistical analysis is appropriate for the experiments conducted. It is currently unclear if some experiments were adequately replicated. That information should be added to the paper throughout where it is missing.

Studies from multiple groups should be more thoroughly referenced when the authors are describing the "vesicle-like" staining patterns observed in SGs from reporter strains (e.g. Fig. 3A). Is this similar to the SG vesicle-like structures observed previously (e.g. PMIDs: 28377572, 33305876, and others)?

There are minor grammar issues in the manuscript text (e.g. "Up to date" should be "To date"). The figures are primarily presented very clearly and accurately. One minor suggestion: In cases such as Fig. S2A images 3 and 6, where some of the staining labels are very difficult to read, please move all labels for the figure to boxes located directly above the image.

The data and conclusions are presented well.

Significance

This report represents a significant technical advance (improved in vivo reporter strain and sporozoite imaging), and a minor conceptual advance (active sporozoite active motility), for the field.

This work builds off of previous SG live imaging studies involving Plasmodium-infected mosquitoes (e.g. Sinnis lab, Frischneckt lab, etc.), addressing one of the major challenges from these studies (reliable in vivo imaging inside mosquito SGs).

This work will appeal to a relatively small audience of vector biology researchers with an interest in SGs. Many in the field still see the SGs as intractable, instead choosing to focus on the midgut due to ease of manipulation. Perhaps work like this will spark new interest in tangential research areas.

I have sufficient expertise to evaluate the entirety of this manuscript. Some descriptors of my perspective include: bioinformatics, SG molecular biology, mosquito salivary glands, microscopy, RNA interference, SG infection, and SG cell 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 #2

Evidence, reproducibility and clarity

The first achievements of the Klug et al. study are the (i) genetical engineering of the Anopheles coluzzii mosquitoes reared in insectarium, that stably express distinct fluorescent reporters (DsRed and hGrx1-roGFP2 and EGFP) under the putative "promoters" of genes reported to encode proteins expressed differentially in the pluri-lobal salivary glands(Sg) of anthropophilic blood-feeding adult females, (ii) the analysis of the promoter activity - based on the selected fluorescent reporter - with a primary focus on the salivary gland/Sg (including at the Sg lobe level) of the adult female but also considering the preimaginal developmental time with larvae and pupa samples. Of note, some data confirm the already reported time-dependent and blood meal-dependent promoter activity for the related Anopheles species. The last part presents preliminary dataset on live imaging of Plasmodium berghei sporozoites with the aim of highlighting the usefulness of these A. coluzzii transgenic lines to better understand how the rodent Plasmodium sporozoites first colonize and then settle as packed cells in Sg acinar host cells.

Major comments

The two first objectives presented by the authors have been convincingly achieved with (i) the challenging production of four different lines expressing different single or double reporters chosen by the authors (and appropriately presented in the result text and figure sections), (ii) the careful analysis of the spatiotemporal expression of the DsRed reporter under two "promoters" studied and with regards to the blood feeding event parameter. However, if the reason why the authors have put so much effort in the production of their transgenic mosquitoes is (and as mentioned) to provide a significant improved setting enabling the behavioral analysis of sporozoites upon colonization and survival in the Sg, it seems this part is kind of limited. Likely in relation with this perception is the fact I found the introductory section often confusing and not enough direct to the points: in particular distinguishing the rationale from the necessity to produce appropriate models, and clarifying what is/are the added value(s) offered by these new transgenic lines models when compared to what exist (in Anopheles stephensi) with specific evidence that argue for this knowledge gain. At this stage, it is unfortunately not clear to me, what is the bonus of imaging the Plasmodium fluorescent sporozoites in hosts with fluorescent salivary gland lobes if one can not monitor key events of the Sg-sporozoite interaction that were not reachable without the fluorescent mosquito lines. Furthermore, it should be better explained why the rodent Plasmodium species has been chosen rather Plasmodium falciparum (or other human species) for which A. coluzzii is a natural host; may be just mentioning that this study would serve as a proof of concept but bringing real biological insights would be fine.

1- The three genes and gene products selected by the authors should definitively be more systematically explained, which means for example the authors need to introduce the different mosquito species and the parasite-mosquito host pairs they are then referring to for the promoter/encoded proteins of their interest. In the same vein, I did not find any information as to the choice of the mosquito specie (A. Coluzzii) for the current work. I was curious to know what is the advantage since better knowledge was available with Anopheles stephensi with respect to (i) Saglin and its promotor activity, (ii) aap driven dsRed expression (lines already existing) and (iii) sporozoite-gland interaction.

2- To help clarifying the added value of the present study, introducing the species names of the mosquito and the Plasmodium that serve as a model would be appreciated.

3- Since a focus is the salivary gland of the blood feeding female Anopheles sp., a rapid description of the glands with different lobes and subdomains the results and figure 1 nicely refer to, would help in the introduction.

4- That description could logically introduce the few proteins actually identified with lobe specific or cell domain specific expression (apical versus basal side, intracellular or surface expose, vacuole, duct...) profiles. The context with regards to sporozoite biology would then easily validate the "promoter choice". As a minor remark, I miss the reason why the authors wrote " the astonishing degree of order of the structures (referring to the packing of sporozoites within the Sg acinars) raise the question whether sporozoite can recognize each other". Please clarify since packing/accumulation can be passive due to cell mechanical constraints and explain what this point has to see with the question and experimental work proposed here?)

5- The selection of hGrx1-roGFP2 is quite interesting and justified but there is then no use of this reporter property in the preliminary characterization of the Sg and Sg-sporozoite interaction. Could the authors provide such characterization?

6- Figure 1: it would be nice to add in the legend at what time the dissection/imaging has been made (age, blood feeding timing?). I would also omit the double mutant trio-Dsred/aapDsred in the main figure (may be supplemental) since the two single mutants Dsred separately together with the double mutant (with different fluorescence) already provide the information. I would suggest to regroup the phenotypic presentation of the transgenic line made in the KI mosquitoes (current figure 5) in the main figure 1.

7- Figure 2:

a) Is there anything known on the Sgs' size change overtime. It seems that between day 1 and 2 there is an increase of size and volume as much as I can evaluate the volume (Fig S4). Could that mean that there is increase in cell number in the lobes and therefore more cells expressing the transgene which would account for the signal intensity increase rather than more transcripts per cell?

b) why choosing 24h after the blood meal to assess promoter activity in the Sgs? Do we have any information on how the blood meal impact on the Sgs'development. At this time anyway the sporozoites are far from being made. Yosshida and Watanabe 2006 mentioned at significant decrease of Sg proteins post-blood feeding. Could the authors detail their rationale based on what the questions they wish to address

8- Figure 3: The figure is quite informative in terms of subcellular localization. Concerning the section "Natural variation of DsRed expression in trio-DsRed mosquitoes", I think it could be shortened because because it is a bit out of the focus the study.

9- In contrast the last section of live imaging of P. berghei sporozoites in the vicinity and within salivary gland should be expanded. The 2 sentences summarizing the data are quite frustrating "We also observed single sporozoites moving actively through tissues in a back and forth gliding manner (Fig. 6B, Movie 3) or making contact with the salivary gland although no invasion event could be monitored"

10- I am aware of the technical difficulties to perform live imaging of sporozoite on whole mosquitoes, even when the salivary gland lobe under observation is closely apposed to the cuticle but that seems to be the final aim of the authors. I looked very carefully to the three movies and I am sorry but at this stage I could not make meaningful analysis out of them, and could not agree with the conclusions: for instances, the authors specify that sporozoites were undergoing back and forth movements (movie 3) but I do not see that and do not see the Sg contours in the available movies? The authors should also add bar and time scales to their movies. Having an in-depth description with regards to the sub-domain marked by a relevant reporter would strengthen the study, even if images are not collected in the whole mosquito to get higher resolution.

I am not sure I understand the relevance of this quite condensed sentence in the text. Could the authors rephrase and expand if they wish to keep the issues they refer to. "The sporozoites' distinctive cell polarization and crescent shape, in combination with high motility, allows them to „drill" through tissues". I would stress more on the main unknown in terms of sporozoite-Sg interactions and the need to get right models for applying informative approaches (i.e. here, imaging).

Of note, it could help to point that the "Sgs is a niche in which the sporozoites which egress from the oocyst could mature and be fully competent when co-deposited with the saliva into the dermis of their intermediary hosts"

Significance

1- Clear technical significance with the challenging molecular genetics achieved in the mosquito A. coluzzii.

2- More limited biological significance: fair analysis and gain of knowledge of spatio-temporal of reporter expression under the selected promoter but limited significance of the final goal analysis which concerns the Plasmodium sporozoite biology once egressed from oocysts

3- Previous reports cited by the authors have used the DsRed reporter and the aap promoter in another Anopheles (i.e. A. stephensi, Yoshida and Watanabe, Insect Mol Biol, 2006; Wells and Andrew, 2019) which is also a natural host and vector for human Plasmodium spp.) with significantly more resolutive 3D visualization of GFP-fluorescent P. berghei but in dissected salivary glands and not in whole mosquitoes. The Wells and Andrew publication entitled "Salivary gland cellular architecture in the Asian malaria vector mosquito Anopheles stephensi" in Parasite Vectors, 2015 would deserve to be reference and described.

4- Audience: I would say that this work should be of interest of mostly scientists investigating Plasmodium biology (basic and field research) or in entomology of Diptera.

5- To describe my fields of expertise, I can refer to my extensive initial training in entomology including at one point in the genetic basis of mosquito-virus interaction. I have also been working for more than 20 years in the field of Apicomplexa biology (Plasmodium and Toxoplasma) and I have long-standing interest in live and static high-resolution imaging.

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 reports the generation and characterization of transgenic lines in the African malaria mosquito Anopheles coluzzii that express fluorescent proteins in the salivary glands, and their potential use for in vivo imaging of Plasmodium sporozoites. The authors tested three salivary gland-specific promoters from the genes encoding anopheline antiplatelet protein (AAPP), the triple functional domain protein (TRIO) and saglin (SAG), to drive expression of DsRed and roGFP2 fluorescent reporters. The authors also generated a SAG knockout line where SAG open reading frame was replaced by GFP. The reporter expression pattern revealed lobe-specific activity of the promoters within the salivary glands, restricted either to the distal lobes (aapp) or the middle lobe (trio and sag). One of the line, expressing hGrx1-roGFP2 under control of aapp promoter, displayed abnormal morphology of the salivary glands, while other lines looked normal. The data show that expression of fluorescent reporters does not impair Plasmodium berghei development in the mosquito, with oocyst densities and salivary gland sporozoite numbers not different from wild type mosquitoes. Salivary gland reporter lines were crossed with a pigmentation deficient yellow(-) mosquito line to provide proof of concept of in vivo imaging of GFP-expressing P. berghei sporozoites in live infected mosquitoes.

Major comments

Overall the manuscript is very well written with a clear narrative. The data are very well presented. The generation of the transgenic mosquito lines is elegant and state-of-the art, and the new reporter lines are thoroughly characterized.

This is a nice piece of work that is suitable for publication, although the in vivo imaging of sporozoites is somewhat preliminary and would benefit from additional experiments to increase the study impact.

The reporter mosquito lines express fluorescent salivary gland lobes, yet the authors only provide imaging of parasites outside the glands. It would be relevant to provide images of the parasite inside the fluorescent glands.

The advantage of the pigmentation-deficient line over simple reporter lines is not clear, essentially due to the background GFP fluorescent in figure 5C. Imaging of GFP-expressing parasites should be performed in mosquitoes after excision of the GFP cassette under control of the 3xP3 promoter. This would probably allow to document the value of the reporter lines more convincingly.

Along the same line, it is unclear if the DsRed spillover signal in the GFP channel is inherent to the high expression level or to a non-optimal microscope setting. This is a limitation for the use of the reporter lines to image GFP-expressing parasites.

The authors should fully exploit the SAG(-) line, which is knockout for saglin and provides a unique opportunity to determine the role of this protein during invasion of the salivary glands. This would considerably augment the impact of the study. In this regard, line 131 and Fig S3E: why is there persistence of a PCR band for non-excised in the sag(-)EX DNA?

Did the authors search for alternative integration of the construct to explain the trioDsRed variability?

Line 254-255. Does the abnormal morphology of SG from aapp-hGrx1-roGFP2 result in reduced sporozoite transmission?

Minor comments

-Line 51: sporogony rather than schizogony

-Line 56: sporozoites are not really deformable as they keep their shape during motility

-In the result section, it is not clearly explained where constructs were integrated. Line 106 and 434-435: for the non-expert reader, it is not clear what X1 refers to, strain or locus for integration?

-Line 112-115: the rational of integrating GFP instead of SAG is not clearly explained here, but become clearer in the discussion (line

-Line 140: FigS2A instead of S3A

-Perhaps mention that GFP reporters (SG) might be useful to image RFP-expressing parasites.

-Line 236: the authors cannot exclude integration of an additional copy (as mentioned in the discussion line 367-368).

-Line 257-258. The title of this section should be modified as SG invasion was not captured.

-Line 287: remove "considerable number" since there is no quantification.

-Line 400-402: Klug and Frischknecht have shown that motility precedes egress from oocysts (PMID 28115054), so the statement should be modified.

-Line 404: remove "significant number" since there is no quantification.

-Line 497: typo "transgenesis"

-FigS1: add sag-DsRed in the title

-Stats: Mann Whitney is adequate for analysis in fig 2C but not 2B, where ANOVA should be used (more than 2 groups).

Significance

This work describes a technical advance that will mainly benefit researchers interested in vector-Plasmodium interactions. Invasion of salivary glands by Plasmodium sporozoites is an essential step for transmission of the malaria parasite, yet remains poorly understood as it is not easily accessible to experimentation. The development of transgenic mosquitoes expressing fluorescent salivary glands and with decreased pigmentation provides novel tools to allow for the first time in vivo imaging in live mosquitos of the interactions between sporozoites and salivary glands.

Reviewer's expertise: malaria, Plasmodium berghei, genetic manipulation, host-parasite interactions

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

We thank the two reviewers for the precious time devoted in the evaluation of our original manuscript and for the useful feedback that guided its revision. We addressed all points raised by the reviewers as detailed below. The changes in the revised manuscript are reported using green characters so that they can be more easily identified in the next evaluation. We would like to comment on the sentence of Reviewer 2 that

“The bulk of this paper was either demonstrated or predicted by another paper Hanemaaijer et al. in 2020 using very similar methods.”

We disagree with this statement and here is why. In Hanemaaijer et al. (2020) the authors demonstrate Nav1.2 permeability to calcium by experiments in HEK-293 cells expressing the channel. Then, they show that a calcium transient associated with the action potential (AP) in the AIS is mediated by sodium channels, but the pharmacology they used (TTX) could not distinguish between Nav1.2 and Nav1.6 (both channels being TTX-sensitive) and produced the blockade of the action potential. Thus, they inferred that at least part of the calcium transient associated with the AP is mediated by Nav1.2, but they could not exclude the involvement of Nav1.6. This is extremely important since we clearly demonstrate that it is only Nav1.2 that carries a calcium current and not Nav1.6. Regarding the interaction between Nav1.2 and BK channels, this possibility was only suggested in the Discussion on the basis of the channel distribution and its biophysical properties. However, here again, there was no formal demonstration. This seminal study is of course important nevertheless and hence was cited in our manuscript six times (reference [12]). However, this study did not reach formal conclusions. The reason why the authors could not further advance from this hypothesis relies on the lack of methods. In our study we fulfilled this limitation by adding two technical advances that were not available before. First, we achieved a selective partial block of Nav1.2 thanks to the newly optimized G1G4Huwentoxin-IV peptide. Second, we directly estimated the effect of Nav1.2 block on the sodium current underlying the AP generation using an imaging technique that we recently developed (Filipis & Canepari 2021, reference [15] in the manuscript). Thanks to these significant methodological differences with respect to Hanemaaijer et al. (2020), we unambiguously demonstrated a Nav1.2 calcium influx component associated with the AP and we assessed the question of the target of this calcium signal which turned out to be the BK channel. In line with that, we argue reasonably that when a research study reports interesting results and from there on suggests important working hypotheses that are not formally demonstrated, and then a second study demonstrates experimentally that these hypotheses are correct (thanks to innovative methodologies), then this second study is also of major importance and cannot be considered a confirmation of the first study.

Rebuttal to Reviewer 1

However, statistical analysis should be performed with non-parametric tests such as Mann-Whitney or Wilcoxon tests and not t-test because of the small samples (t-test can be used only for large samples).

We strongly agree that a “two-population” t-test, being parametric, is not adequate when comparing two or more small sets of independent samples. However, in this study, we never compared sets of independent samples, but we compared two sets of correlated samples where the first sample is the measurement in control condition and the second sample is the same measurement after addition of the channel blocker. Thus, for this type of datasets, we used a “paired” t-test. While the standard Mann-Whitney or Wilcoxon are not appropriate for sets of correlated samples, we followed the reviewer recommendation to perform a non-parametric test and we applied the “Wilcoxon ranked sign test” (non-parametric equivalent to the paired t-test). The results of this additional test were consistent with the paired t-test. As reported in the manuscript, all data and metadata used in this study will be available in the public repository Zenodo (doi: 10.5281/zenodo.5835995) after the manuscript will be accepted by a journal. Thus, while in the manuscript we only report p

1) abstract, line 5, "...by a recent peptide,...". I guess the authors mean "...by a peptide recently identified..."

We replaced with “recently modified peptide” since the wild-type Huwentoxin-IV was identified some time ago.

Rebuttal to Reviewer 2

1. It would be fair to point out somewhere in the introduction or discussion that the role of Nav1.2 and its interaction with BK channels was already predicted by Hanemaaijer et al. (2020).

We pointed out this prediction in the Discussion of the revised manuscript. It must be said that the important work of Hanemaaijer et al. (2020) (reference [12]) is cited six times in the manuscript. As stated above, the interaction between Nav1.2 and BK channels was only suggested in Hanemaaijer et al. (2020) and not demonstrated.

I would recommend citing Huang & Rasband (2018) for a relatively up to date review of channels in the AIS.

This review is now cited in the Discussion of the revised manuscript (reference [38]).

The authors should state whether the L5 cells are L5a or L5b (or whether they didn't distinguish).

Although L5a and L5b could be in principle recognised also by their morphology, no attempt was done to distinguish between two functionally different groups of cells. This is now stated in the Introduction of the revised manuscript.

The choice of colors in the figures made it hard to see which line was which. The authors use blue vs light blue and green vs light green, which were indistinguishable on my screen and in print. The orange vs yellow figs could just be made out.

In control conditions, we use green for sodium, red for voltage and blue for calcium, which also makes it easy to follow in the simulations. We followed the reviewer’s recommendation and used always grey traces for all traces after addition of the channel blocker. It should be easier now in the revised manuscript to visually discriminate the traces.

Are the plots on the RHS of Fig. 2b averages?

The plots are from the cell in panel a. This is now stated in the figure legend of the revised manuscript.

I don't understand the sentence starting, "This signal translated in a substantial delay..."

We replaced this sentence, in the revised manuscript, with “In our experiments, the peak of the Ca2+ current in the distal axon preceded both the peak of the somatic AP and the peak of the Ca2+ in the proximal part of the AIS”. We thank the reviewer for finding this ambiguous sentence.

I also didn't understand the point being made in the sentence: "The evident anticipation of the AP peak, also with respect to the somatic AP peak, suggests that Ca2+ influx associated with the AP is mainly mediated by VGNCs in the distal axon, whereas it is mainly mediated by VGCCs in the proximal axon." The block of calcium with VGCC-blockers is convincing in itself but this argument based on timing isn't convincing. Isn't the point that VGCCs are to be found closer to the soma, despite the fact that Nav1.2 is also found closer to the soma, so that the presence of Ca-conducting Na+ channels near the cell body could be masked?

We erased this sentence in the revised manuscript.

I think it would make more sense to write, "the effects produced by 1 µM IK CAKC inhibitor tram-34 were not significant".

We replaced “variable” with “not significant” in the revised manuscript.

There is a mismatch in Fig. S5 between the 400 nM, 800 nM, 1600 nM labelling in the figure and the "40 nM, 80 nM or 160 nm" in the legend.

We thank the reviewer for having found this mismatch. The values are those in the figure labels, so we corrected the values in the figure legend in the revised manuscript.

In Fig. S7c-d, the authors write "... the widening of the distal axonal AP was observed in 4 cells, whereas the AP waveforms did not change in 3 cells". I think this is not appropriate if the results are statistically insignificant. It implies that the authors know somehow that the outliers are not just noise.

We changed this inappropriate sentence 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 #2

Evidence, reproducibility and clarity

Filipis et al. present a study of the underlying mechanisms for the generation of action potentials in the axon initial segment (AIS) of layer 5 somatosensory pyramidal neurons. The experiments employed three different imaging approaches for following sodium, voltage and calcium changes at two different locations in the axon near the cell body. This was combined with pharmacology for comparing the contribution of two different types of sodium channels: Nav1.6 and Nav1.2, the latter having a concomitant Ca2+ conductance. They confirm the previously established findings in terms of the location of Nav1.6 lying more distally than Nav1.2 on the AIS, and show that while Nav1.6 contributes mostly to the initiation of the AP, Nav1.2 plays an important role in controlling the shape of the AP via Ca2+ activation of BK (potassium) channels. The study represents a careful attempt to investigate a difficult subject owing to the small size of the axon initial segment and the difficulty of disentangling the different channels and influences. I find the paper well carried out and well presented and a useful contribution to understanding the firing properties of these important neurons.

I only have a few minor points:

1. It would be fair to point out somewhere in the introduction or discussion that the role of Nav1.2 and its interaction with BK channels was already predicted by Hanemaaijer et al. (2020).
2. I would recommend citing Huang & Rasband (2018) for a relatively up to date review of channels in the AIS.
3. The authors should state whether the L5 cells are L5a or L5b (or whether they didn't distinguish).
4. The choice of colors in the figures made it hard to see which line was which. The authors use blue vs light blue and green vs light green, which were indistinguishable on my screen and in print. The orange vs yellow figs could just be made out.
5. Are the plots on the RHS of Fig. 2b averages?
6. I don't understand the sentence starting, "This signal translated in a substantial delay..."
7. I also didn't understand the point being made in the sentence: "The evident anticipation of the AP peak, also with respect to the somatic AP peak, suggests that Ca2+ influx associated with the AP is mainly mediated by VGNCs in the distal axon, whereas it is mainly mediated by VGCCs in the proximal axon." The block of calcium with VGCC-blockers is convincing in itself but this argument based on timing isn't convincing. Isn't the point that VGCCs are to be found closer to the soma, despite the fact that Nav1.2 is also found closer to the soma, so that the presence of Ca-conducting Na+ channels near the cell body could be masked?
8. I think it would make more sense to write, "the effects produced by 1 µM IK CAKC inhibitor tram-34 were not significant".
9. There is a mismatch in Fig. S5 between the 400 nM, 800 nM, 1600 nM labelling in the figure and the "40 nM, 80 nM or 160 nm" in the legend.
10. In Fig. S7c-d, the authors write "... the widening of the distal axonal AP was observed in 4 cells, whereas the AP waveforms did not change in 3 cells". I think this is not appropriate if the results are statistically insignificant. It implies that the authors know somehow that the outliers are not just noise.

Hanemaaijer, N. A., Popovic, M. A., Wilders, X., Grasman, S., Arocas, O. P., & Kole, M. H. (2020). Ca2+ entry through NaV channels generates submillisecond axonal Ca2+ signaling. Elife, 9, e54566. Huang, C. Y. M., & Rasband, M. N. (2018). Axon initial segments: structure, function, and disease. Annals of the New York Academy of Sciences, 1420(1), 46-61.

Significance

• The results are useful and important (but perhaps not major).
• The bulk of this paper was either demonstrated or predicted by another paper Hanemaaijer et al. in 2020 using very similar methods.
• The audience for this paper will be a relatively specialized group focusing on biophysical explanations for axonal excitability and/or modellers.
• My expertise is a specialization on the electrical properties of these neurons (L5 pyramidal neurons).

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 paper investigates the mechanisms of BK potassium channels activation in the axon initial segment (AIS) of cortical neurons. Using whole-cell patch-clamp recording, voltage imaging of the AIS, pharmacological tools, use of peptides and computer simulations, the authors show that calcium influx via Nav1.2 channels activates BK channels, thus shaping the action potential waveform.

The manuscript is well written and the data are clear. The conclusions are convincing as they are in agreement with a recent study showing that Nav1.2 are permeable to calcium ions (Hanemaaijer et al., eLife 2020). The results and methods are presented in a way that they can be reproduced. However, statistical analysis should be performed with non-parametric tests such as Mann-Whitney or Wilcoxon tests and not t-test because of the small samples (t-test can be used only for large samples).

Minor points:

abstract, line 5, "...by a recent peptide,...". I guess the authors mean "...by a peptide recently identified..."

Significance

The study is interesting as it provides a target (activation of BK channels) for the Nav1.2-mediated calcium influx. Audience: specialized journal in neurobiology.

My field of expertise is the cellular neurophysiology (axon, ion channels, synapse and plasticity). I have sufficient expertise to evaluate all the aspects of this paper.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Response to the reviewers

Manuscript number: RC-2022-01407

Corresponding author(s): Ivana, Nikić-Spiegel

1. General Statements

We would like to thank the reviewers for careful reading of our manuscript and for their insightful and useful comments. We are happy to see that the reviewers find these results to be of interest and significance. The way we understand reviewers’ reports, their main concerns can be roughly divided in following categories: 1) providing more quantitative data 2) interpretation of the Annexin V/PI assay 3) additional evidence for calpain involvement. We intend to address these experimentally or by modifying the text, as outlined below.

2. Description of the planned revisions

Reviewer #1

Fig1A/B o SYTO 16 staining suggests slight reshaping of nucleus upon spermine NONOate, showing less blurry punctae. From the SYTO 16 profile, this should be quantifiable.

By looking at the shown examples and the entire dataset, it appears to us as if neuronal nuclei are shrinking upon spermine NONOate treatment resulting in their less blurry appearance. We are not sure if this is what the reviewer is referring to, but this can also be quantified by measuring changes in neuronal nuclear size. We already have this data from the measurements shown in Fig4 and we intend to show it in the revised version of the manuscript. Line profile measurements are also possible, but the nuclear size quantification might be more suitable for this purpose.

o There is a subset of neuron nuclei that are SYTO 16 positive. Please quantify the ratio

We will use our existing dataset to quantify the ratio of NFL positive and SYTO16 positive nuclei.

FigS1A o Show NeuN with Anti-NFL merged figures

We will show merged NeuN and anti-NFL images, which might require rearrangement of the existing figures and figure panels. We will do this in the revised manuscript.

FigS1C o Show quantification and timeline. I want to know whether there is also a plateau reached here.

As the data shown in the FigS1C do not include NeuN staining, we will do additional experiments and perform proposed quantifications.

FigS2A-F o Though the statements might be true, selecting one nucleus for a line profile as a statement for the whole dataset seems problematic. Average a larger number of unbiased selected nuclei profiles across multiple cultures to make a stronger statement, or a percentage of positive nuclei as in FigS1b.

Corresponding images and line profiles are representative of the entire dataset. However, we agree with the reviewer that this is not obvious from the current manuscript version. Thus, to strengthen our findings, we intend to quantify the percentage of positive nuclei as in FigS1b. The only difference will be that instead of NeuN, we will use SYTO16 as a nuclear marker. The reason being that the existing datasets contain images of NFL and SYTO16 and not NeuN.

FigS3 • There are no fluorescence profiles, no quantification

As the reviewer suggests, we will quantify the ratio of NFL positive and SYTO16 positive nuclei, and include the quantifications in the revised manuscript.

General statement: There do seem to be punctated patterns of non-nucleus accumulating NFL fragments. Can they be localized to any specific structure?

We assume that the reviewer is referring to neuronal/axonal debris. They are present after injury but they do not colocalize with nuclear stains. We will address this in the revised manuscript.

Fig1C-F • I find it too simplistic to categorize c+f and d+e together. There is a huge difference in the examples of nuclear localization between d and e. To not comment on their distinction (if that is consistent) is problematic. Also, since we don't see a merge with either NeuN or SYTO 16, reader quantification is difficult.

We thank the reviewer for bringing this up. We will carefully check our entire dataset and we will update the figures and the text accordingly. We will also show the corresponding SYTO16 images, as the reviewer suggested.

Would the microfluidic device construction allow for time to transport any axonally damaged fragments to the soma?

Yes, the construction of the microfluidic devices allows the transport of axonal proteins back to the soma. Based on our experiments, it seems that damaged NFL from the axonal compartment could be contributing to the accumulation of NFL fragments in the nuclei. However, this contribution seems to be minimal as we cannot detect nuclear NFL upon the injury of axons alone. Alternatively, it could be that the processing of axonal NFL fragments proceeds differently if neuronal bodies are not injured and that this is the reason we don’t detect the NFL nuclear accumulation upon injury of axons alone. We will discuss this in the revised manuscript.

Fig2C+D • The statement ".... no annexin V was detected on the cell membrane" needs to be shown more clearly

We will modify figures to address this comment.

• Please provide merged AnnexinV/PI images

We will modify figures to address this comment.

• The conclusion about 2D, that nuclear accumulated NFL overlaps with PI is not supported by the example image shown. There are plenty of PI positive spots that are not NFL positive and even several NFL positive ones that do not have a clear PI staining. Please quantify and then show a very clear result in order to be able to suggest necrosis as the underlying process.

We are not sure if we understand the reviewer’s concern correctly. We will try to clarify it here and in the revised text. If necessary, we will tone down our conclusion, but the reason why not all of PI positive spots are NFL positive is most likely due to the fact that not all injured nuclei are NFL positive. We quantified in FigS1 that up to 60% of nuclei under injury conditions show NFL accumulations. That is why we are not surprised to see some PI positive/NFL negative nuclei. And the fact that there are some NFL positive nuclei which appear to be PI negative is most likely related to the fact that the PI binding is affected. In addition, upon closer inspection of NFL and PI panels in Fig2d it can be observed that NFL positive nuclei are also PI positive, albeit with a lower PI fluorescence intensity. We will modify the figure to show this clearly in the revised manuscript.

FigS5 C+D • If the case is made that nitric oxide damage induces necrosis, then why is it that the AnnexinV example of Staurosporine exposure (which induces apoptosis) looks similar to that of nitric oxide damage in Fig2d and necrosis induction with Saponin looks very different?

We thank the reviewer for bringing this up. We will try to clarify this in the revised manuscript. Regarding the specific questions, the most likely explanation why staurosporine treated neurons look similar to the ones treated with spermine NONOate is that in the late stages of apoptosis cell membrane ruptures and allows for the PI to label nuclei. This is probably the case here as illustrated by the nucleus in the middle of the image (FigS5c) that shows the fragmentation characteristic for the apoptosis. This is not happening in early apoptotic cells due to the presence of an intact plasma membrane. On the other hand, the reason why saponin treated cultures look different compared to spermine NONOate is that membranes are destroyed by saponin so that the PI can enter the cell. For that reason, there could have not been any AnnexinV binding to the membrane which would correspond to the AnnexinV signal of spermine NONOate treated neurons. As we will discuss below, we did not try to mimic spermine NONOate-induced injury with saponin treatment. Instead this was a control condition for PI labeling and imaging. We also used a rather high concentration of saponin which probably destroyed all the membranes which was not the case with spermine NONOate treatment. We intend to do additional control experiments to address this.

• Additionally, does necrosis induction with Saponin also cause NFL fragment accumulation in the nucleus? Please show a co-staining of them. Also, the authors want to make a claim about reduce PI binding in NFL accumulated necrotic cells. In these examples, the intensity of the nuclear stain of PI with Saponin looks dimmer than with Staurosporine. Are the color scalings similar? It might be that the necrotic process itself causes reducing binding of PI and is not related to the presence of NFL.

With regards to this question, it is important to note that Annexin V and PI imaging was done in living cells. To obtain the corresponding anti-NFL signal as shown in Fig 2c,d we had to fix the neurons, perform immunocytochemistry and identify the same field of view. We tried to do the same procedure after saponin treatment (Supplementary Figure 5d) but the correlative imaging was very difficult due to the detachment of neurons from the coverslip after the saponin treatment. For this reason, we could not identify the same field of view co-stained with NFL. However, other fields of view did not show NFL fragment accumulation. This could also be the consequence of the high saponin concentration that we used as we discuss above. We have also noticed the reduced intensity of PI binding in the nuclei of saponin-treated neurons. However, if the necrotic process itself reduces the binding of PI to the DNA, then all of the neurons treated with spermine NONOate would have an equally low PI signal. In our experiments, only the nuclei which contained NFL accumulations had a low PI signal, while the signal of NFL-negative nuclei was higher (as shown in Fig2d). We would also like to point out again that the saponin treatment was our control of the PI’s ability to penetrate cells and bind the DNA, as well as our imaging conditions, and not the control of the necrotic process itself. This is the reason why we didn’t go into details about neuronal morphology and NFL localization upon saponin treatment. We thank the reviewer for pointing this out since it prompted us to reevaluate what we wrote in the corresponding paragraph of the manuscript. We realized that the confusion might stem from our explanation of the AnnexinV/PI assay controls in the lines 196-198 (“Additional control experiments in which neurons were treated with 10 μM staurosporine (a positive control for induction of apoptosis) or with 0.1% saponin (a positive control for induction of necrosis) confirmed the efficiency of the annexin V/PI assay (Supplementary Fig. 5c,d).”). We will modify this portion of the text to clearly state that staurosporine and saponin treatments were controls of the AnnexinV and PI binding to their respective targets and not of the apoptosis/necrosis process. When it comes to the saponin treatment, our intention was only to permeabilize the membranes in order to allow PI penetration and DNA binding and not to induce necrosis or to mimic the effect of the spermine NONOate. We also intend to perform experiments with lower concentration of saponin to try to address this experimentally in addition to the text modifications.

Fig3d • Please show similarly scaled images from controls for proper comparison

We will show similarly scaled images of the control neurons so that they can be properly compared. They were initially not scaled the same for visualization purposes, but we will modify this in the revised manuscript.

• How do the authors scale the degree and kinetics of induced damage between application of hydrogen peroxide/CCCP and glutamate toxicity? Does glutamate toxicity take longer to affect the cell, not allowing enough time to accumulate NFL fragments in the nucleus?

It is challenging to scale the degree and kinetics of induced damage with different stressors. That is why we did not intend to do this. Instead we set different injury conditions based on the published literature. That is why can only speculate when it comes to this. In this regard, it can be that the glutamate toxicity takes “longer” to affect the cells even though it is very difficult to compare them on a timescale, especially when considering different mechanisms of action. We will discuss this limitation in the revised manuscript.

Fig4B • Some groups (like NO and NO + emricasan) have much larger numbers of close to 0 intensity, compared to the control group. Why?

We were wondering the same when we analyzed the data. The fact that our nuclear fluorescence intensity analysis picked up NFL signal in control neurons which had no nuclear NFL accumulation made us realize that the intensity measured in the nuclei of control group comes entirely from the out of focus fluorescence – from neurofilaments in cell bodies, dendrites and axons (an example can be seen in the FigS6). That is why we presented the corresponding data with a cut-off value based on the control signal (as mentioned in lines 238-240). Since the oxidative injury causes NFL degradation (not only in neuronal soma, but also neuronal processes), the overall fluorescence intensity of the NFL immunocytochemical staining is reduced in injured neurons. We can see that in all of our images. Consequently, there is no contribution of out of focus fluorescent signal to the measured fluorescence intensity in the majority of nuclei. Due to that, the nuclei without NFL accumulation (at least 40% of injured nuclei) will appear to have a close to 0 intensity of the fluorescent signal. We will discuss and clarify this additionally in the revised manuscript.

• Please add the ratio of above/below threshold (50/50 obviously in controls)

We will update the figure in the revised manuscript.

• The description of the CTCF value calculation seems a little... muddled? Several parameters are described whereas "integrated density" is not even used. Why not simply mean intensity of nuclear ROI-mean intensity of background ROI?

We included the integrated density in the description since it is measured together with the raw integrated density and can also be used for the CTCF value calculation. However, since we didn’t use it for the CTCF calculation, we will remove it from the corresponding section of the manuscript. We calculated the CTCF value instead of calculating mean intensity of the nuclear ROI - mean intensity of the background ROI, since the CTCF value also takes into account the area of the ROI and not just the mean intensity.

• Also, please tell me if the areas for nuclear ROIs change, as I noted for Fig1A/B

We will include this information in the revised manuscript.

• To make sure that one of the 3 experimental repeats didn't skew the results, please show the median fluorescence intensity for each individual experiment to clarify that the supposed effect is repeated across experiments.

We have already noticed that in the earliest of the three experiments overall fluorescence intensity was higher, but this was consistent across all the experimental groups and did not skew the results or affect the overall conclusion. However, we will double-check this and revise the figure.

• From the text "...and due to the NFL degradation during injury...": this seems to contradict the process? Either the NFL fragment accumulates in the nucleus or it is degraded during injury. And isn't the degradation through calpain what supposedly allows this fragment of NFL to go to the nucleus in the first place? I reckon that the authors are possibly trying to reconcile why there are many close-to-0 intensity nuclei in the NO and NO + emricasan groups, but I don't feel the explanation given here fits.

As we tried to explain in our response above, we think that the overall degradation of neurofilaments in neurons affects the fluorescence intensity originating from the out of focus neurofilaments. Therefore, the nuclei without NFL accumulation in injured conditions have a close to 0 fluorescence intensity. Additionally, we think that this is not an either/or situation, but that both degradation and nuclear accumulation of NFL happen simultaneously. We also think that degradation of axonal NFL and the transport of its tail domain to the soma will at least partially contribute to the accumulation in the nucleus. In any case, degradation and nuclear accumulation seem to be differentially regulated in individual neurons, as some of them show nuclear NFL accumulation and some not. Furthermore, calpain and other mechanisms could also cause NFL degradation up to the point at which these fragments can no longer be recognized by the anti-NFL antibody leading to the loss of signal. We will try to clarify this in the revised version of the manuscript.

Fig5 • Does the distribution of this GFP in B match any of the various antibody stainings of different NFL fragments? Perhaps this is still a valid fragment of NFL, just not picked up by any AB?

The GFP signal in B appears rather homogenous and it does not match any of the various antibody stainings of different NFL fragments. As the reviewer points out, this could also be a valid fragment of NFL fused to GFP that none of our antibodies is recognizing. We will clarify this in the revised manuscript.

• "... and was indistinguishable from the full277 length NFL-GFP." Based on what parameters?

We will clarify this in the revised text, but we meant in terms of overall neurofilament network and cell appearance, which is commonly used to test the effect of NFL mutations.

• The authors claim that b is different from d, but I am not convinced. I would like to see a time dependent curve from multiple cells showing a differential change in nuclear and cytosolic GFP signal.

As we also wrote in the manuscript, in the majority of neurons that were monitored during injury we were not able to detect an increase in the GFP fluorescence intensity in the nucleus. This is what prompted further experiments with NFL(ΔA461–D543)-FLAG. We will clarify this additionally in the revised manuscript and perform line profile intensity measurements to show the difference in nuclear and cytosolic GFP signal.

• Secondly, the somatic GFP intensity for NFL increases for full length NFL-GFP. How is this explained, if it is only a separation of NFL and GFP? If anything, GFP should float away. And if the answer is that NFL is recruited to the nucleus, you showed that inhibition of calpain activity partially prevents that. So, if calpain activity is necessary for the transport of NFL to the nucleus, then wouldn't it also cut the GFP from NFL before it reaches the nucleus?

We thank the reviewer for bringing this up and we apologize for the confusion. This can be explained by the fact that the images were scaled in a way that the GFP signal over time could still be seen easily (i.e. differently across different time points which we unfortunately forgot to mention in the figure legend). In the revised manuscript, we will either scale the images the same or we will alternatively show the displayed grey values in individual panels.

Fig6 • It is recommended to overlap the transfected cells with a stain for endogenous NFL to show that despite the absence of the FLAG-tag, there is still NFL.

We did not overlap the anti-NFL with anti-FLAG and SYTO16 staining, due to the space constraint and the intent to clearly show the overlap of FLAG and SYTO16 signals in the merged images above the graphs. However, the line profile intensity measurements were done in all three channels and show that despite the absence of FLAG, there is still NFL in the nucleus (Fig6b), or that both FLAG and NFL are present in the nucleus (Fig6d, NFL signal shown in gray). However, as this is not obvious and can easily be overlooked, we will show the endogenous NFL staining overlap in the revised version of the manuscript.

Fig7 • „ ...all disrupted neurofilament assembly...": this sounds like the staining for native NFL supposedly shows a distortion due to a dominant negative effect of the expression of these constructs? Please clarify.

Yes, we were referring to the disruption of neurofilament assembly due to a dominant negative effect of the expression of NFL domains. We will clarify this in the revised version of the manuscript.

Discussion: • The authors show that after overepression of the head domain only, it possibly passively diffuses into the nucleus even in the absence of oxidative injury. However, it seems to be suggested as well that the head domain would not be freely floating around if it wouldn't be for increased calpain activity as a result of oxidative injury in the first place. Therefore, a head domain fragment localized in the nucleus would still more prominently happen upon oxidative injury and interact with DNA through prior identified putative DNA interaction sites from Wang et al. Please comment.

That is correct. Upon injury and calpain cleavage, it is conceivable that a fragment containing the NFL head domain would also be present in the cell and could potentially diffuse to the nucleus and interact with the DNA. However, by staining injured neurons with an antibody that recognizes amino acids 6-25 of the NFL head domain, we were not able to detect an NFL signal in the nucleus (FigS2a,b). It could be that either the NFL head domain does not localize in the nuclei upon injury, or that the fragment localizing in the nucleus does not contain amino acids 6-25 of the NFL head domain. As the putative DNA-binding sites described by Wang et al involve 7 amino acids located in the first 25 residues of the NFL head domain, we would expect to detect it with the aforementioned antibody. However, as that was not the case we speculated that the interaction of NFL and DNA occurs differently in living cells, as opposed to the test tube conditions utilized by Wang et al. We will comment and clarify this in the revised version of the manuscript.

• Reviewer #2*

• Major Comments:

• The initial data presented in the paper is good, does response of oxidative damage with proper controls, testing the antibodies to NF-L and etc. (Fig. 1-Fig. 4). *

We thank the reviewer for their positive feedback.

1. The evidence for calpain involvement in NF-L cleavage during oxidative damage is missing. Provide the evidence for full length NF-L construct and deletion mutants transfected into cells by immunoblot for cleavage of NF-L, perform nuclear and cytoplasmic extract preparations and show that enrichment of the tagged cleaved NF-L fragment in nuclear fraction.

We thank the reviewer for their comments and suggestions. Since we saw in our microscopy experiments that calpain inhibition reduced the accumulation of NFL in the nucleus, and since it is known that NFL is a calpain substrate (Schlaepfer et al., 1985; Kunz et al., 2004 and others), we did not perform additional experiments to confirm the involvement of calpain in NFL degradation during injury. However, to strengthen our findings, we intend to perform the suggested experiments and include the results in the revised manuscript.

1. Show calpain activation during oxidative damage by performing alpha-Spectrin immunoblots identify calpain specific 150-kda Spectrin and caspase specific 120-kDa fragment generation in these cells. Also, calpain activation can be measured by MAP2 level alteration and p35 to p25 conversion. Without this evidence it's very hard to believe if the calpain activity is increased or decreased during oxidative damage and these markers are altered by using calpain inhibitors.

To confirm the calpain activation, we intend to perform anti-alpha spectrin and/or anti-MAP2 blots in lysates of control and injured neurons and include the results in the revised manuscript.

1. The premise that NF proteins are absent in cell bodies and present only in axons is not correct. It has been demonstrated by multiple investigators that NFs are present in the perikaryon and dendrites of many types of neurons (Dahl, 1983, Experimental Cell Research)., Dr. Ron Liem's group showed NF protein expression in cell bodies of dorsal root ganglion cells (Adebola et ., 2015, Human Mol Genetics) and also showed N-terminal antibodies for NF-L, NF-M and NF-H stain rat cerebellar neuronal cell bodies and dendrites (Kaplan et al., 1991, Journal of Neuroscience Research) when NFs are less phosphorylated. (Schlaepfer et al., 1981, Brain Research) show staining of cell bodies of cortex and dorsal root ganglion cell bodies with NF antibody Ab150, and Yuan et al., 2009 in mouse cortical neurons with GFP tagged NF-L.

We are not sure what the reviewer is referring to since we cannot find a corresponding section in which we claim that NF proteins are absent in cell bodies. We wrote the following “Anti-NFL antibody staining of neurons treated with the control compound showed the expected neurofilament morphology, that is, a strong fluorescence intensity in axons and lower intensity in cell bodies and dendrites (Fig. 1a)” in our results section (lines 119-121), but the claim we were trying to make there was that NF proteins are particularly abundant in axons. We will clarify this in the revised manuscript.

1. Quantifying NF-L signal or tagged NF-L fragment signals in the cell body by ICC has many problems and making conclusions. It's extremely difficult to have control over levels of proteins in transfected overexpression models and comparing two or three different constructs with each other by ICC. Not every cell expresses same levels of protein in transfected cells and quantifying it by ICC again has a major problem. This can be addressed if there are stable lines that express equal levels of protein in all cells that comparisons can be made. Under thesese circumstances validation of the hypothesis presented in the study has no strong direct evidence to demonstrate that calpain is activated and NF-L fragment translocate to the nucleus.

We agree that the results from overexpression-based experiments should be interpreted with caution as levels of expression vary between the cells. We intend to discuss this in the revised manuscript. However, we find it difficult to experimentally address this comment since we are not sure which specific experiments the reviewer is referring to. With regards to this, we would like to emphasize that most of the initial experiments in which we observed NFL accumulation in the nuclei of injured neurons were based on the ICC labeling of endogenous NFL and didn’t involve its overexpression. This includes labeling of endogenous NFL in various types of neurons, comparing the effects of different types of oxidative injury, as well as testing the effects of calpain inhibition on the observed nuclear accumulation (Figures 1-4; Supplementary Figures 1-6). We later resorted to the overexpression experiments in primary neurons (Figures 5-7; Supplementary Figure 7, 10) to gain more information about the identity of NFL fragment which was detected in the nucleus. Due to the low transfection efficiency of primary neurons, we performed an additional set of overexpression experiments in neuroblastoma ND7/23 cells (Figure 8; Supplementary Figures 8,9) and obtained similar results in a higher number of cells. We agree that having stable cell lines which e.g. express same levels of NFL domains would be a more elegant approach and we intend to make them for our follow-up studies, however the generation of said stable cell lines might be beyond the scope of this revision. Furthermore, looking at our data with overexpression of NFL domains in ND7/23 cells (Supplementary Figure 8,9), it appears to us as if different domains are rather homogenously expressed in different cells. While the expression levels might vary, it seems that they all show the same trend when it comes to their localization (which was the main point of those experiments).

1. The interpretation that NF-L preventing DNA labeling cells is misinterpretation. NFs have very long half-life compared to other proteins. Due to oxidative damage, DNA is degraded in the cells but NFs that have very long half-life you see as NFs rings in the dead cells. So, NFs do not prevent DNA labeling, but DNA or chromatin is degraded in dead cells.

We thank the reviewer for their useful insight. DNA degradation could certainly be the reason why we observe a lower fluorescence intensity of the propidium iodide fluorescence in the nuclei of injured neurons. We intend to discuss this in the revised manuscript. However, if the DNA degradation is the only reason for the lower PI fluorescence intensity, then the PI fluorescence intensity would be the same in all injured nuclei. In our experiments, we saw the reduced PI fluorescence intensity in nuclei that contained NFL accumulations and not in other nuclei. Additionally, we observed a reduction of SYTO16 fluorescent labeling of nuclei which contained accumulations of the NFL tail domain, even in the absence of oxidative injury. Due to these reasons we speculated that NFL accumulation in the nucleus might hinder nuclear dyes from interacting with the DNA. But this is only a speculation and we will try to clarify this further in the revised manuscript including alternative explanations.

Minor comments: 1. In the introduction on page 4 reference is missing for NF transport, aggregation and perikaryal accumulation (on line 93).

We will add a reference to the revised manuscript.

1. The statement in discussion on page 14 line 454 for Zhu et al., 1997 study is not accurate. It should be modified to sciatic nerve crush not spinal cord injury.

We will correct this mistake in the revised manuscript.

1. What is the size of the calpain cleaved NF-L tail domain? If you perform immunoblots on cell extracts treated with oxidative agents one would know it.

We will perform immunoblots on cell lysates and incorporate the corresponding results in the revised manuscript.

1. Authors could make their conclusions clear. This is particularly true for the experiments in Figure 4 panels c and d. It is very difficult to understand the conclusions of the experiments. First state the expectation and then described whether the expectation is true or different.

We will do as the reviewer suggested in the revised manuscript.

1. The ICC images are at extremely low magnification. They should be shown at 100x or 120x so that details of the cell body and the nucleus can be seen.

Our intention was to show larger fields of view and wherever appropriate insets, but we will try to improve this in the revised manuscript by either zooming in, cropping or adding additional insets with individual cell bodies and nuclei. In general, images were taken with an optimal resolution/pixel size in mind for any of the used objectives (60x/1.4 NA or 100x/1.49 NA) and we can easily modify our figure panels to show more details.

1. Oxidative damage leads to beaded accumulation of NF-L in neurites and axons. Authors should address this issue.

We will discuss this in the revised manuscript.

1. The combination treatment of the inhibitors (last 3 sets of the Fig. 4 b) has no statistical significance should be removed.

Actually, these differences were statistically significant (Supplementary Table 1). For clarity and as described in the figure legend (line 516: “The most relevant significant differences are indicated with an asterisk”) we showed only a subset of them on the graph, but we will change this in the revised manuscript.

1. Why only two antibodies recognize cleaved NF-L? If the antibodies at directed at tail region, they should recognize it unless the phosphorylated tail at Ser473 may inibit the antibody binding. In that case NF-L Ser473 specific antibody (EMD Millipore: MABN2431) may be used to test this idea.

This is a very good point that we also wonder about. Even if all antibodies are directed at tail region, exact epitopes are not described for all of them. That makes it also difficult for us to understand and speculate on this. However, we have already ordered the new antibody as suggested by the reviewer and we will experimentally test it.

**Referees cross-commenting**

I agree with the reviewer#1 about presenting the quantification data for the indicated figures to make conclusions strong and see how much of variation is there among sampled cells.

As discussed in our response to reviewer #1, we will provide additional quantifications.

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

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

Reviewer #2, major comment 7. Authors could do chromatin immunoprecipitation (chip) analysis to identify NF-L binding sites on chromatin and perform gel shift assays to show NF-L tail domain binding to specific consensus DNA sequences.

We thank the reviewer for their suggestion. We are very interested in performing additional experiments and identifying the NFL binding sites on the DNA (either by chromatin immunoprecipitation or DamID-seq) and we intend to perform these experiments as soon as possible. Unfortunately, at the moment we do not have the expertise to perform such experiments in our lab. Instead, this type of follow-up project requires establishing a collaboration which is beyond the scope of this revision.

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 authors hypothesize that Neurofilament-L subunit of NFs participates in oxidative medicated damage by calpain activation and cleavage of its C-terminal region, translocation to nucleus and activation of toxicity driven gene expression. Authors used cell culture system, NF-L gene transfections and immunocytochemistry to make their conclusions.

Major Comments:

1. The initial data presented in the paper is good, does response of oxidative damage with proper controls, testing the antibodies to NF-L and etc. (Fig. 1-Fig. 4).
2. The evidence for calpain involvement in NF-L cleavage during oxidative damage is missing. Provide the evidence for full length NF-L construct and deletion mutants transfected into cells by immunoblot for cleavage of NF-L, perform nuclear and cytoplasmic extract preparations and show that enrichment of the tagged cleaved NF-L fragment in nuclear fraction.
3. Show calpain activation during oxidative damage by performing alpha-Spectrin immunoblots identify calpain specific 150-kda Spectrin and caspase specific 120-kDa fragment generation in these cells. Also, calpain activation can be measured by MAP2 level alteration and p35 to p25 conversion. Without this evidence it's very hard to believe if the calpain activity is increased or decreased during oxidative damage and these markers are altered by using calpain inhibitors.
4. The premise that NF proteins are absent in cell bodies and present only in axons is not correct. It has been demonstrated by multiple investigators that NFs are present in the perikaryon and dendrites of many types of neurons (Dahl, 1983, Experimental Cell Research)., Dr. Ron Liem's group showed NF protein expression in cell bodies of dorsal root ganglion cells (Adebola et ., 2015, Human Mol Genetics) and also showed N-terminal antibodies for NF-L, NF-M and NF-H stain rat cerebellar neuronal cell bodies and dendrites (Kaplan et al., 1991, Journal of Neuroscience Research) when NFs are less phosphorylated. (Schlaepfer et al., 1981, Brain Research) show staining of cell bodies of cortex and dorsal root ganglion cell bodies with NF antibody Ab150, and Yuan et al., 2009 in mouse cortical neurons with GFP tagged NF-L.
5. Quantifying NF-L signal or tagged NF-L fragment signals in the cell body by ICC has many problems and making conclusions. It's extremely difficult to have control over levels of proteins in transfected overexpression models and comparing two or three different constructs with each other by ICC. Not every cell expresses same levels of protein in transfected cells and quantifying it by ICC again has a major problem. This can be addressed if there are stable lines that express equal levels of protein in all cells that comparisons can be made. Under thesese circumstances validation of the hypothesis presented in the study has no strong direct evidence to demonstrate that calpain is activated and NF-L fragment translocate to the nucleus.
6. The interpretation that NF-L preventing DNA labeling cells is misinterpretation. NFs have very long half-life compared to other proteins. Due to oxidative damage, DNA is degraded in the cells but NFs that have very long half-life you see as NFs rings in the dead cells. So, NFs do not prevent DNA labeling, but DNA or chromatin is degraded in dead cells.
7. Authors could do chromatin immunoprecipitation (chip) analysis to identify NF-L binding sites on chromatin and perform gel shift assays to show NF-L tail domain binding to specific consensus DNA sequences.

Minor comments:

1. In the introduction on page 4 reference is missing for NF transport, aggregation and perikaryal accumulation (on line 93).
2. The statement in discussion on page 14 line 454 for Zhu et al., 1997 study is not accurate. It should be modified to sciatic nerve crush not spinal cord injury.
3. What is the size of the calpain cleaved NF-L tail domain? If you perform immunoblots on cell extracts treated with oxidative agents one would know it.
4. Authors could make their conclusions clear. This is particularly true for the experiments in Figure 4 panels c and d. It is very difficult to understand the conclusions of the experiments. First state the expectation and then described whether the expectation is true or different.
5. The ICC images are at extremely low magnification. They should be shown at 100x or 120x so that details of the cell body and the nucleus can be seen.
6. Oxidative damage leads to beaded accumulation of NF-L in neurites and axons. Authors should address this issue.
7. The combination treatment of the inhibitors (last 3 sets of the Fig. 4 b) has no statistical significance should be removed.
8. Why only two antibodies recognize cleaved NF-L? If the antibodies at directed at tail region, they should recognize it unless the phosphorylated tail at Ser473 may inibit the antibody binding. In that case NF-L Ser473 specific antibody (EMD Millipore: MABN2431) may be used to test this idea.

Significance

The study has very high significance. The results obtained with proper experimentation great implications in understanding how NF proteins are degraded in the cells and how these degraded fragments would alter neurodegeneration during oxidative stress and other conditions.

Referees cross-commenting

I agree with the reviewer#1 about presenting the quantification data for the indicated figures to make conclusions strong and see how much of variation is there among sampled 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

Summary:

The manuscript presented by Arsić and Nikić-Spiegel investigates a physiological consequence when neurons in vitro are exposed to oxidative stress injury, specifically a supposed interaction of the tail subdomain of the neurofilament light chain (NFL), after cleavage of the full NFL protein by calpain.

General comments:

The conclusions the authors draw from individual non-quantified example images are sometimes seen to be too simplistic when the shown examples ask for a more thorough investigation, especially when specific merged images are not available. It is highly recommended that the authors use the available data to come to more comprehensive answers across the entire acquired dataset. This for instance happens only in figures 4 and 8 and should be extended to other figures as well. There is not necessarily doubt about the author's general claims, but convincing the reader requires showing the variability and effect size of the entire group beyond a single selected example.

If these more thorough quantifications continue to support the author's claims, then I find no objections for publication of this data.

Specific comments:

Fig1A/B - SYTO 16 staining suggests slight reshaping of nucleus upon spermine NONOate, showing less blurry punctae. From the SYTO 16 profile, this should be quantifiable. - There is a subset of neuron nuclei that are SYTO 16 positive. Please quantify the ratio

FigS1A - Show NeuN with Anti-NFL merged figures

FigS1C - Show quantification and timeline. I want to know whether there is also a plateau reached here.

FigS2A-F - Though the statements might be true, selecting one nucleus for a line profile as a statement for the whole dataset seems problematic. Average a larger number of unbiased selected nuclei profiles across multiple cultures to make a stronger statement, or a percentage of positive nuclei as in FigS1b.

FigS3 - There are no fluorescence profiles, no quantification

General statement:

There do seem to be punctated patterns of non-nucleus accumulating NFL fragments. Can they be localized to any specific structure?

Fig1C-F - I find it too simplistic to categorize c+f and d+e together. There is a huge difference in the examples of nuclear localization between d and e. To not comment on their distinction (if that is consistent) is problematic. Also, since we don't see a merge with either NeuN or SYTO 16, reader quantification is difficult. - Would the microfluidic device construction allow for time to transport any axonally damaged fragments to the soma?

Fig2C+D - The statement ".... no annexin V was detected on the cell membrane" needs to be shown more clearly - Please provide merged AnnexinV/PI images - The conclusion about 2D, that nuclear accumulated NFL overlaps with PI is not supported by the example image shown. There are plenty of PI positive spots that are not NFL positive and even several NFL positive ones that do not have a clear PI staining. Please quantify and then show a very clear result in order to be able to suggest necrosis as the underlying process.

FigS5 C+D - If the case is made that nitric oxide damage induces necrosis, then why is it that the AnnexinV example of Staurosporine exposure (which induces apoptosis) looks similar to that of nitric oxide damage in Fig2d and necrosis induction with Saponin looks very different? - Additionally, does necrosis induction with Saponin also cause NFL fragment accumulation in the nucleus? Please show a co-staining of them. Also, the authors want to make a claim about reduce PI binding in NFL accumulated necrotic cells. In these examples, the intensity of the nuclear stain of PI with Saponin looks dimmer than with Staurosporine. Are the color scalings similar? It might be that the necrotic process itself causes reducing binding of PI and is not related to the presence of NFL.

Fig3d - Please show similarly scaled images from controls for proper comparison - How do the authors scale the degree and kinetics of induced damage between application of hydrogen peroxide/CCCP and glutamate toxicity? Does glutamate toxicity take longer to affect the cell, not allowing enough time to accumulate NFL fragments in the nucleus?

Fig4B - Some groups (like NO and NO + emricasan) have much larger numbers of close to 0 intensity, compared to the control group. Why? - Please add the ratio of above/below threshold (50/50 obviously in controls) - The description of the CTCF value calculation seems a little... muddled? Several parameters are described whereas "integrated density" is not even used. Why not simply mean intensity of nuclear ROI-mean intensity of background ROI? - Also, please tell me if the areas for nuclear ROIs change, as I noted for Fig1A/B - To make sure that one of the 3 experimental repeats didn't skew the results, please show the median fluorescence intensity for each individual experiment to clarify that the supposed effect is repeated across experiments. - From the text "...and due to the NFL degradation during injury...": this seems to contradict the process? Either the NFL fragment accumulates in the nucleus or it is degraded during injury. And isn't the degradation through calpain what supposedly allows this fragment of NFL to go to the nucleus in the first place? I reckon that the authors are possibly trying to reconcile why there are many close-to-0 intensity nuclei in the NO and NO + emricasan groups, but I don't feel the explanation given here fits.

Fig5 - Does the distribution of this GFP in B match any of the various antibody stainings of different NFL fragments? Perhaps this is still a valid fragment of NFL, just not picked up by any AB? - "... and was indistinguishable from the full277 length NFL-GFP." Based on what parameters? - The authors claim that b is different from d, but I am not convinced. I would like to see a time dependent curve from multiple cells showing a differential change in nuclear and cytosolic GFP signal. - Secondly, the somatic GFP intensity for NFL increases for full length NFL-GFP. How is this explained, if it is only a separation of NFL and GFP? If anything, GFP should float away. And if the answer is that NFL is recruited to the nucleus, you showed that inhibition of calpain activity partially prevents that. So, if calpain activity is necessary for the transport of NFL to the nucleus, then wouldn't it also cut the GFP from NFL before it reaches the nucleus?

Fig6 - It is recommended to overlap the transfected cells with a stain for endogenous NFL to show that despite the absence of the FLAG-tag, there is still NFL.

Fig7 - „ ...all disrupted neurofilament assembly...": this sounds like the staining for native NFL supposedly shows a distortion due to a dominant negative effect of the expression of these constructs? Please clarify.

Discussion:

• The authors show that after overepression of the head domain only, it possibly passively diffuses into the nucleus even in the absence of oxidative injury. However, it seems to be suggested as well that the head domain would not be freely floating around if it wouldn't be for increased calpain activity as a result of oxidative injury in the first place. Therefore, a head domain fragment localized in the nucleus would still more prominently happen upon oxidative injury and interact with DNA through prior identified putative DNA interaction sites from Wang et al. Please comment.

Significance

This in vitro study, despite its acknowledged caveats, can provide novel support for the claim that calpain induced cleavage of the NFL may play a role in downstream gene expression in order to regulate a neural response upon oxidative injury. Further investigation into this topic may provide further understand of physiological gene expression through interaction with cleavage products as well as yield possible therapeutic targets for pathological conditions. This study therefore may be of interest to a broad audience.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

In this manuscript Woglar et al. use several light and electron microscopy techniques combined with averaging/registration methodologies to produce a comprehensive molecular map of the centriole in the C. elegans gonad. The images produced are very impressive and potentially very informative, allowing the authors to draw several important conclusions (e.g. about the chirality of the structure, and the potential organisation of Sas-6 in the cartwheel, the latter of which has been controversial in this species). Thus, although the manuscript is largely descriptive, there is a lot here that will be of great interest to the centriole field. The manuscript is generally well written and well presented, and, although I am not a great expert in all of these techniques, the data seems to solidly support the main conclusions. I therefore have only a small number of relatively minor suggestions for improvements.

Minor Comments:

1. It should be clarified whether the centrioles being examined here are organising genuine PCM and MTs. I know that in the embryo SPD-2 and SPD-5 are considered the main organisers of the mitotic PCM, and these centrioles are in S-phase or G2 (so I'm not sure if they are organising any PCM). SPD-5 is located internally to SPD-2, perhaps suggesting that these centrioles are not organising a bona fide PCM? On the other hand, TBG-1 and MZT-1 are located at the periphery, so I assume these centrioles are organising MTs?
2. I think the labels (A, B, C) in Figure S1 are probably in the wrong order and are not referred to correctly in the main text.
3. In Figure S1A two centrioles are shown that seem to be touching at their proximal ends, which I initially interpreted as meaning the centrioles were engaged. If so, there seems to be a long tail of Sas-6 connecting the two centrioles that extends well below the centriole MTs. However, reading the legend, I think this interpretation is incorrect, and the images are showing two separate centrioles that just happen to be touching? Perhaps swap in another image that won't lead to this potential confusion?

Significance

Although several papers have reported high resolution molecular mapping of centrioles, this one is perhaps the most detailed and does a nice job of superimposing the molecular structures on high quality EM images. Not all of these C. elegans proteins are obviously conserved, but C. elegans is a 'poster-child' model organism for centriole research, and this broad architecture will be of great interest to the entire centriole/centrosome (and also cilia) fields. In addition, the observation of chirality that is intrinsic to the inner centriole structure, and that Sas-6 is likely organised into rings rather than a steep helix, are important conclusions.

I am an expert in centrioles and high resolution imaging, but not EM.

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, Woglar et al describe molecular features of the C. elegans centriole with unprecedented detail. By adapting U-ExM to extracted gonads and combining it with EM and TEM data, the authors precisely mapped the location of 12 components. They uncovered that these centrioles are shorter than in the embryo, have the same structural elements, and show an offset of centriolar proteins distribution relative to microtubules which results in chirality. Their detailed analysis also identified two novel electron-dense regions: the Inter Paddlewheel Density (IPD); and the SAS-6/4/1 Containing Density (SCD). This manuscript is a very nice description of C. elegans centrioles and we have mostly minor comments to improve it.

1. Regarding the duplication and maturation section, the authors state in the abstract: "We uncovered that the procentriole assembles from a location on the centriole margin characterized by SPD-2 and ZYG-1 accumulation.". The data collected by the authors do not provide evidence of enrichment of ZYG-1 and SDP-2 prior to procentriole assembly (in the main text the authors clearly say they are speculating). This statement in the abstract should be corrected to more accurately match what is described in the main text and supported by the results.
2. It is stated in the main text that the procentrioles can emanate from the middle of the centriole but no representative image is shown (only shown for off-centered procentrioles or very short templates). It is also referred that this may have implications on chirality- it would be important to explain better those implications, as well as offer an example of this configuration.
3. The authors mention "core PCM" throughout the manuscript without explaining or referencing its definition. Would be useful to the reader if more information is provided.
4. FigS1.A looks strange because procentrioles seem much longer than centrioles and their relative orientation does not seem to be orthogonal. If this image is representative, it would be helpful to have a diagram explaining the image.
5. In the main text it is said: "Four components were found to localize to the paddlewheel: HYLS-1[N], SPD-2, SPD-5 and PCMD-1." and SDP-5 is represented in the final scheme (Fig. 7). However, an overlay of SPD5 and EM data is never shown. The authors may extrapolate that SPD-5 localizes there because it is interior to SPD-2 with no offset compared to α-tubulin, but if this is the case it should be clearer in the text.
6. A statistics section is missing in which the program used is detailed and whether the {plus minus} values in the figures depict SD or SEM. The number of independent experiments should also be mentioned.
7. Although symmetrization has been increasingly adopted by the field, it would still be useful to reference previous examples of its application in centriole structure analysis.
8. S1B and S1C figure labels are swapped.
9. The authors claim that "the procentriole likewise harbors little SAS-4 initially and that more protein is recruited at prometaphase, resulting in similar levels of SAS-4 in the centriole and the procentriole by then (Fig. 2D)". Can the authors provide some sort of semi-quantitative readout?
10. In Figure 5A side view, the presence of an inner tube is not very clear. Given that diameter quantifications were done using the mostly side views, it would be beneficial if the authors could provide a clearer image.

Significance

Overall, these observations contribute toward a better understanding of centriole structure, molecular composition and diversity, with a particular focus on C. elegans. The precision of the approach developed by the authors (U-ExM and EM overlay) is a valuable tool and will be of interest to the centriole biology field and to cell biologists in general.

Reviewer expertise: Cellular and molecular biologists working in the field of centrioles.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #1

Evidence, reproducibility and clarity

The authors analyze here the organization of centrioles in C. elegans, by combining the physical expansion of the specimens (by about 5-fold) with stimulated emission depletion (STED) microscopy. They analyze a large number of centriole components in different experiments, and they combine the data into a convincing model of the centriole, which is presented in conjunction with electron microscopy images of this structure. The work is solid, well-performed and technically sound. While this reviewer is not a centriole expert, the work also appears to be sufficiently novel, simply due to its precision, to warrant publication.

Significance

I only have one suggestion, which the authors may consider. Most of their work involves analyzing the symmetry of the structures, as presented, for example, in Fig. 4. However, symmetry problems, observable in individual structures, may also be informative. Are specific proteins more prone to variable localization, as, for example, SPD-2-C or SPD-5, while others are more stereotypically organized? Could an analysis of the variability of the stainings provide information on flexibility in the centriole organization?

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

The team explores a previously developed "centriole stability assay" to monitor the disappearance of centrioles after RNAi-dependent depletion of various centrosome components. Important roles in centriole stability are found for the PCM and for cartwheel proteins, in addition to proteins of the centriole wall. The remainder of the study focuses on the centriole wall protein ANA1: induced degradation of ANA1 during Drosophila oogenesis strongly reduces the PCM and other centriole markers, and ANA1-dependent defects cannot be prevented by GFP-Polo-PACT, which is otherwise known to protect from the loss of PCM. In complementary experiments, forced targeting of ANA1 to the PCM, or overexpression of AN1 protects centriole integrity.

Significance

The study shows that ANA1 is important for the integrity of centrosomes. Generally, this work is well executed and correctly controlled. The novelty of the results is somewhat limited, since a role of ANA1 in centrosome assembly has already been reported by others. The present work emphasizes aspects of centrosome protein maintenance, but doesn't provide mechanistic details of protein turnover. The manuscript should be of interest to the scientific community working on the centrosome.

Other comments:

I wonder whether the results from the centrosome maintenance experiment with GFP-Polo-PACT (Figure 3) are really very telling: since PCM and other centriole markers are lost upon ANA1-depletion, GFP-Polo-PACT cannot target to the PCM, and it is therefore unsurprising that GFP-Polo-PACT fails to provide its protective effect. Would expression of GFP-Polo-PACT prior to addition of ANA1-RNAi have a protective effect?

Minor point:

Figure 1H: it is unclear to me how centrioles are identified with the BLD10 marker in samples that have been treated with BLD10 RNAi.

Referees cross-commenting

I agree very much with reviewers 1 and 2 that a role of ANA1 "downstream" of the PCM is not really supported by the data.

I also think that all other points raised in the reviews merit attention.

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 paper, the authors show that the turnover of centriole components is necessary for proper centriole maintenance within Drosophila cultured cells (during prologued cell cycle arrest) and within Drosophila oocytes, where centrioles are normally degraded prior to fertilisation. They highlight Ana1 as an important player in centriole maintenance. The authors begin with a candidate screen to identify core centriole proteins that are required to properly maintain centrioles. They then focus on Ana1, given that its depletion had the strongest effect, and show that its depletion leads to a reduction in the levels of centriole components in Drosophila oocytes. They show that the previously observed ability of centriole-targeted Polo to counteract centriole loss depends at least in part on Ana1 and that targeting Ana1 to centrioles also counteracts centriole loss. The authors conclude that Ana1 is a component of the PCM-promoted centriole integrity pathway.

Major comments

1. The authors say that Plk4 depletion does not lead to centriole loss, but there are significant differences in centriole number between the control and Plk4 depletion cells in Fig 1F and S1D. Please comment.
2. One of the main results is that depletion of centriole components leads to a reduction in centrosome numbers when measured 8 days after S-phase arrest. I wonder whether a restriction of centriole duplication could add to this effect? Any cells that were in G2 or M phase when the drugs were added would presumably progress into the following S-phase and duplicate their inherited centrioles, but not if centriole duplication proteins had been depleted. It's true that Plk4 depletion leads to a relatively mild centriole loss phenotype, but can the authors be sure that this is not due to variations in the efficiency of different RNAi constructs? Perhaps the authors can show that Plk4 depletion efficiently prevents centriole duplication under otherwise normal conditions.
3. The authors show that Ana1 depletion has the strongest effect, but this could in theory be due to differences in RNAi efficiency. I don't expect the authors to show the efficiency of all RNAi constructs, but they could state in the text that this is a caveat e.g. "...although we cannot rule out the possibility that differences in RNAi efficiency lead to the observed differences in severity of phenotype..."
4. A key conclusion is that core centriole components turnover to some extent and that the incorporation of new molecules is necessary for centriole maintenance. This is a very interesting and important point and so it would be nice to have more direct data to support it. This could be done in different ways, including transfecting fluorescently tagged centriole components after S-phase arrest and showing that some molecules become incorporated into the centrioles, or by performing FRAP experiments. Of course, it is possible that the turnover is so low that the incorporated fluorescent molecules cannot be detected...
5. The authors show that depletion of Ana1 from oocytes leads to a reduction in the intensity of centriole markers. They do not measure centrosome numbers, as the centrosomes cluster too tightly. The authors therefore can't be certain that Ana1 depletion leads to a reduction in centrosome numbers. The authors could show this by inhibiting centrosome clustering while depleting Ana1. There is a recent BioRxiv paper showing that centrosome clustering can be inhibited by depletion of Kinesin-1.
6. In Figure 3B the authors show that expression of GFP-Polo-PACT partially rescues the effect of "all PCM" depletion, but this seems strange given that Polo's role is presumably to recruit PCM (which has been depleted). Can the authors comment? Also, it would make sense to test whether GFP-Polo-PACT can rescue centriole loss after the depletion of Ana1 alone (not Ana1 and all PCM). If Ana1 has a role in recruiting Polo (either directly or indirectly), which has been shown previously in mitotic cells, then there should be a rescue to some extent.
7. In Fig4A,C, the authors say that γ-tubulin levels at centrosomes increase when GFP-Polo is forced onto the centrosomes - the graph seems to show a big increase, but the pictures do not...? Are the authors measuring total levels at all centrosomes? If so, I think they should be measuring the average at individual centrosomes. Also, why is the level of GFP alone not much higher when expressed with GFPnanoPACT (Fig 1B)? Presumably GFP should be recruited to the centrosomes by GFPnanoPACT.
8. The authors show that tethering Ana1-GFP to the centrioles counteracts centriole loss in oocytes (Fig4G). They say that the centrosomes are most likely inactive because they don't recruit PCM, but they have only looked at γ-tubulin, which is a downstream component of the PCM. I think it is important to check whether Polo is recruited, given that tethering Polo to centrioles also counteracts centriole loss and that a recent paper showed that Ana1 has a role in recruiting Polo to centrosomes (Alvarez-Rodigo et al., 2021). The authors also say that these centrosomes do not organise microtubules but do not show the data.
9. The authors propose that Ana1 is downstream of the PCM, and so over-expressing Ana1 should at least partially rescue centriole loss after PCM depletion. But I don't really agree with this. If Ana1 relies on the PCM then how would its overexpression manage to rescue the phenotype in the absence of the PCM? The finding that over-expressing Ana1 partially rescues centriole loss may instead suggest that Ana1 is either upstream of the PCM or part of an independent pathway. Indeed, the authors show that depletion of both the PCM and Ana1 has a stronger effect than either depletions individually - this is indicative of two independent pathways.

Minor comments

1. When the authors say that the centriole wall and cartwheel components are "dynamic" I think that they need to make it clear that this "dynamicity" is not very fast. Using the term dynamic tends to suggest rapid turnover (like in the PCM). Perhaps the authors could use the term "slow exchange" or something similar.
2. The authors currently use a 0 or 1 centriole categorisation - it would be nice to see the breakdown of what percentage of cells have 0, 1, 2, or >2 centrioles, perhaps in a supplementary excel file.

Significance

How centrioles are eliminated in certain cells is an interesting question and the data presented is also relevant to understanding centriole biology in general, because it seems that some apparently very stable structural proteins actually turnover. It is widely known that PCM proteins turnover relatively quickly, but core centriole proteins are considered to be stably incorporated. The data will therefore raise interest in the centrosome field. I do, however, feel that for the authors to make this point more strongly it would be good to show this more directly. Overall, this is a very interesting paper that is well written. The data is well presented and supports the conclusions that centriole components turnover and that Ana1 is involved in maintaining centriole integrity.

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 manuscript Pimenta-Marques build on their previous work addressing how centrioles are stabilized and maintained or destabilized and disassembled, depending on the cell type and developmental context. Using Drosophila cell culture and oogenesis as an in vivo model for centriole destabilization, they identify the centriole wall protein Ana1 as a central player in centriole stability. Its presence is required for the maintenance even of mature centrioles, suggesting that there is continued turnover of centriole structural components.

Major comments:

1. The experiments and results are very well described and most of the conclusions are supported by the data. One aspect needs clarification though. It is not clear to this reviewer how the authors envision the regulation and mechanism by which Ana1 functions in centriole stability. The data suggest that it can stabilize centrioles independent of PCM (Fig. 3B, 5B), yet the authors claim in the results and discussion that it functions downstream of PCM. As presented, this does not make sense. I would argue the opposite, it may function upstream or in parallel to the PCM. Related to the above, the last sentence of the intro states: "Finally, we found that both Polo and the PCM require ANA1 to promote centriole structural integrity." This is shown for Polo, but where is the data showing that PCM requires ANA1 for promoting centriole stability?
2. I have a concern regarding the number n used for statistics in the quantifications. In many cases it seems that the number n of cells etc. was used (e.g. n>100 cells) rather than the number of experiments (e.g. n=3). The statistics should measure variability between experimental repetitions, not between cells etc. If statistics were indeed not done on experiments and would have to be changed, some of the observed effects may not be statistically significant and would require additional experimental replicates, which would increase the time needed for revision.

Minor comments:

1. I would advice the authors to improve the presentation of the figures. In particular the labels are in many cases very small and difficult to read. Readability is also reduced by the use of bold font in the labels and a mix of various font sizes within single figure panels.
2. The result section could be shortened/become more readable by moving several paragraphs to the intro or discussion.
3. The introduction is quite long and some parts read more like an introduction of a review on the topic.

Significance

This is a nice, focused study on the requirements underlying centriole stability and maintenance. The first part identifies the cartwheel, the centriole wall, and the PCM as important for centriole maintenance. The remaining parts identify and focus on the essential role of ANA1 in this process. This is an important finding, since the mechanisms underlying centriole stability and maintenance are poorly understood, yet highly relevant. Some cell types inactivate and/or disassemble centrioles during differentiation and this is likely important to their function. Providing more mechanistic insight, for example, regarding the relationship between ANA1 and PCM recruitment or the regulation of ANA1's centriole function by Polo, would have further strengthened the study. The audience interested in this work will be cell and developmental biologists. My expertise is in centrosome biology and microtubule organization.

Referees cross-commenting

I agree with the additional points raised by the other reviewers. I still think that overall the paper is fine and most things could be addressed in a reasonable time frame. The work does not provide much mechanism though. In this regard, the confusing placement of ANA1 downstream of PCM, would be the only mechanistic aspect, and it seems the authors got it wrong, at least based on the provided data. Here, additional experiments could elucidate these relationships further, but if this is not the goal, text changes could also address this and it would remain a smaller, more focused study.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

Summary:

The manuscript describes how a rhythmically active transcription factor is important for molting cycles. The first part of the manuscript focuses on oscillating genes, and the authors nicely show a rhythmic transcription of these genes. Indeed, using RNAPolII Chip-Seq experiments, they show a rhythmic recruitment of the RNAPolII to the promoters of the oscillatory genes they have previously described. They then demonstrate that GFPs driven by the promoter of these oscillating genes, and inserted as a single copy, can very accurately recapitulate the rhythmic transcription of oscillating genes. It is interesting to see the weak impact of introns and 3'UTR on the rhythmic expression. In the second part of the manuscript, the authors perform an RNAi screen, looking for oscillating transcription factors (TF) important for molting. The goal of this approach is to identify core oscillators that could control molting cycles. Focusing their screen on oscillating TF allows them to exclude TF that would be required for embryonic or larval viability unrelated to molting. Among the 6 candidates they have identified, 5 have already been linked to molting. They focus their study on grh-1, which has never before been described during larval development. They characterize the molting phenotype of grh-1 defective worms using the AID degron system. They monitor the molting cycles using a luciferase assay in a liquid culture where transgenic worms for luciferase are grown in the presence of bacteria supplemented with luciferin. Using this approach (which is quantitative and allows high-throughput analysis), they show that grh-1 is required for each molt in a dose-dependent manner. The GRH-1 protein oscillates and peaks shortly before each molt entry, reinforcing the idea that GRH-1 is an important core TF for molting. The authors finally show that the oscillating activity of GRH-1 is crucial for the molting.

Major comments:

Overall, the data are clear and convincing, and the results are quantified with care. The first part of the manuscript represents a significant amount of work with the RNAPolII Chip-Seq, the different single-copy integrants and RT-PCRs. Then, the authors provide a quantitative assessment of the molting process by combining their grh-1-AID construct with the luciferase system. This strengthens the quality of the manuscript.

My substantial suggestion is that the authors could consider extending the scope of the study in two alternative ways:

• One question that immediately springs to mind is: what are the targets of grh-1? By applying their luciferase assay to further possible downstream targets of grh-1, they could attempt to phenocopy the grh-1 molting defect, and then look if the oscillating expression of these targets is eliminated in grh-1 defective animals. The binding site of grh-1 is apparently known (Venkatesan, 2003), so is it possible to reduce the potential target list among the oscillating genes using a bioinformatics approach? This requires a substantial amount of work (2 to 3 months) but it would help tell a more complete story.
• It is striking that myrf-1 and nhr-23 RNAi display the same molting defects as grh-1 RNAi as show in figure 2B. Have the authors considered testing the genetic interactions of grh-1 with these two other candidates? Do they belong to the same GRN? Does grh-1 depletion impact the expression of the nhr-23 and myrf-1, or vice versa? Do they have the same target genes (Chip-seq data for nhr-23 are available)? This, again, would significantly strengthen the paper and would make the second part more complete. This alternative piece of work would require less experiments than the first suggestion but would be also of great interest. These two points are only suggestions as it represents a significant amount of work, and the paper could very well be published in its current form.

About the luciferase assay for grh-1, nhr-23 and myrf-1 RNAi, the authors observe "an apparent arrest in development or death following atypical molts". What do they mean by "atypical molt" at this stage of the paper? Indeed, for these candidates, the luminescence traces are highly perturbed after the second molt (for grh-1 and nrh-23 RNAi) or the third molt (for myrf-1), but these abnormal traces seem to reflect an arrest in development or larval lethality rather than an atypical molting. Can the authors clarify this point?

In the part on the phenotypic analysis of GRH-1 depleted animals, the authors conclude the paragraph with "GRH-1 is required for viability at least in part through its role in proper cuticle formation". This role in proper cuticle formation refers to the cuticle break in the head region as observed in time lapse. It would be useful to have a visual test of the cuticle permeability using an Hoechst staining.

The authors show GFP::GRH-1 pictures at different stages to describe a rhythmic protein accumulation (see also my minor comment on GFP picture quality). From the perspective of whether all tissues are oscillating, it would be interesting to see if all the cells they mention in the text are showing the same rhythmic fluorescence.

In relation to the previous comment, which tissue is responsible for the defects observed by the degradation of GRH-1? Is it possible to use a tissue-specific depletion of AID-tagged GRH-1 using Seam-cell specific, rectal cell specific, vulval precursors specific promoters, etc...?

In the last part of the results, the authors show that molting requires oscillatory GRH-1 activity by depleting GRH-1 at variable times in L2. It would be interesting to know what happens if a stable (non-oscillating) amount of GRH-1 protein is maintained over time in the worms (using a non-oscillating promoter).

Minor comments:

In figure 5 B, C, D, it seems that right before entering the M2, M3 and M4 respectively, there is a peak of luminescence (a thin bright line) and a strong luminescent signal is detected at the molt exit. Can the authors comment on that?

If I understand correctly, for the GRH-1 GFP CRISPR reporter (Fig S7, S8 and S9), the authors have imaged single worms in microchambers on a spinning disk microscope. I fail to see why they used such a sophisticated approach to describe the expression pattern of GRH-1. This imaging setup is ideal for timelapse. However, in the context of which cell express GRH-1, the resolution is not good enough to fully assess cell identity. Indeed, the GFP images are a bit blurry, and it is difficult to make out the difference between real GFP fluorescence and gut autofluorescence. It would be helpful to have better quality pictures with a more regular setup, i.e., a 2% agarose pad mounted on a regular microscope or confocal. For non-specialists of the C. elegans anatomy, small insets for each category of cells mentioned (seam cells, vulval precursors etc.) would be appreciated.

It would be easier to assess the GRH-1 expression decrease in adults if the pictures were shown in parallel with the larval pictures, with the same brightness/contrast correction (if any). Make insets to compare the same cells between different stages.

How can the authors quantify the duration of molts 3 and 4 in fig S4 when these molts are not seen in the luciferase assay in fig 2B? Can the authors clarify this point?

Writing/clarity: For non-specialists, mention why the authors used a PEST sequence in their constructs and explain what the eft-3 promoter is (they mention it in the Luciferase assay, but it is not clear enough).

In Fig S4, make clearer that EV = MOCK.

In the methods, the authors refer to the we146 mutant strain, but they neither use it nor mention it in the body of the text. Producing such a mutant strain is great and it should be mentioned in the results, with an explanation as to why they are dying. Otherwise, it should be removed from methods.

In the Methods, the genotype of the strains is misleading. For example HW1372 : EG6699; xeSi... is not the regular way to write a C. elegans genotype. It should be written as: HW1360 xeSi131[F58H1.2p::GFP::H2B::Pest::unc-54 3', unc-119+] II; unc-119(ed3) III as the strain used to generate the MosSci insertion is described in the paragraph on Transgenic reporter strain generation.

For the GFP CRISPR strain, the authors write either GFP::GRH-1 in Fig S7, S8 and S9, grh-1::gfp::3xflag in the methods or GRH-1-GFP fusion in the results. The authors should homogenize the way they write this reporter strain. Whether it is an Nterminal or a Cterminal fusion will determine how they should label it.

Enlarge the font size for Fig S8 and 9 for the scale bar.

Significance

The author's prior analysis (Meeuse, et al 2020) showed that mRNA oscillations are coupled with developmental processes, including molting. The present paper extends that finding by showing that oscillating transcript levels are directly linked to a rhythmic recruitment of the RNAPolII on their promoter. Then Meeuse and her colleagues use the molting as a model system to access the importance of oscillating TF for rhythmic processes. Through an RNAi screen, they have identified 6 candidates involved in the molting process. One of the candidates, grh-1, is characterized further. They combine a quantitative-based analysis (luciferase assay) with a time and dose-controlled degradation of GRH-1 to clearly describe the impact of grh-1 depletion on molting. This time and dose control is very smart and key to their study. Overall, the paper adds some interesting piece of information to the field of rhythmic control of molting cycles, as it shows that oscillating transcription factors provide a developmental clock in this process. But this notion is not completely new, as it has been shown in other developmental processes like the circadian clock. Moreover, how molting cycles are controlled by GRH-1 remains to be elucidated.

My field of expertise is GNRs studies, the genetic of C. elegans, embryonic and larval development in C.elegans, timelapse and confocal imaging. I do not have expertise in Chip-Seq analysis.

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 paper, Meeuse et al. analyze the determinants of rhythmical molting in C. elegans using a combination of RNA-polymerase ChIP-seq, RNA-seq, RNAi and imaging coupled to targeted degradation. The main finding is the discovery of the importance for the molting process of GRH-1, a transcription factor homolog to Vertebrate Grainyhead, as well as the identification of 5 additional transcription factors for which molting is defective.

Major comments:

The experiments are generally well performed, and the results convincingly demonstrate the function of GRH-1 in molting.

The coupled RNA-seq and ChIP-seq experiment (Fig. 1A/B) was done only once.

One of the claims (p3) is that rhythmic transcription of oscillating genes is driven by rhythmic RNA pol II occupancy. If this is suggested visually by Figure 1A/B, I think this claim merits further statistical analysis. How is ChIP-seq correlated with RNA-seq globally and at the individual gene level? How good is the correlation? Can the authors provide a supplementary table with this correlation at the gene level? In the same paragraph, a couple examples of genes for which RNA polII ChIP does not correlate with RNA-seq would be helpful to the reader (they are currently cited as "instances where oscillating mRNA levels were not accompanied by rhythmic RNAPII promoter binding"). Please provide gene names.

The oscillatory transcription of several promoters is tested using GFP fusions coupled with q-RT PCR. It is not clear to me whether these experiments were repeated. Additionally, the authors state that each qPCR was repeated (only once), hence both data points should be shown in the graphs on Fig. 1C and S2.

The authors perform then RNAi knock-down of 92 transcription factors involved in molting using bioluminescence and identify 6 genes involved in molts, three of which have been characterized previously. They focus on one of the other three, grh-1, as its orthologs are involved in epidermal biology in other organisms. Targeted degradation of GRH-1 using the auxin degron confirms the function of GRH-1 in each molt, in an auxin-concentration dependent manner. GFP tagging of GRH-1 shows an accumulation prior to each molt, suggesting the transcription factor is necessary for the onset of molting. As the GRH-1 target site is known (PMID 12888489), the paper would be greatly strengthened if the authors could loop back to the RNA-seq/ChIP-seq dataset and highlight which cycling genes have indeed a GRH-1 binding site in their promoter sequence and whether this correlates with one specific phase of the molting cycle.

Similar to the ChIP/RNA-seq experiments, it is not clear to me whether the different bioluminescence experiments were performed once or twice.

As stated above, the results are very convincing and the conclusions quite clear. Additionally, all wet lab experiments are described in very many details. However, I feel that the description of the data analysis is too succinct to allow reproducing the experiments, even using the GEO data. I would expect the authors to provide a github public link with the scripts to perform the RNA-seq and ChIP-seq analyses, the Matlab scripts to analyze bioluminescence experiments (Fig. 2,5,7) and the CNN used to analyze single worm molting, even if the method is to be described elsewhere.

Minor comments:

The authors should provide descriptive statistics for their high-throughput sequencing (read number, mapping statistics etc...).

In figure 1C, it would be helpful to the reader to write peak phase and amplitude of the tested genes on the graph.

In the same figures, for gene F58H1.2, for which the correlation between the reporter and the endogenous gene is not perfect, the authors "suspect [...] that the reporter may lack relevant promoter or intronic enhancer elements". An alternative explanation is that post-transcriptional regulation occurs for this mRNA, a hypothesis which should be added to the text.

Significance

As stated above, this manuscript highlights convincingly that GRH-1 is involved in the molting cycle in the nematode C. elegans, a conserved function of the gene between evolutionary distant species for skin biology.

Field of expertise: C. elegans, high-throughput sequencing methods, imaging.

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 Meeuse and colleagues describes a series of detailed and elegant experiments addressing the molecular mechanisms underlying oscillatory gene expression patterns in the nematode Caenorhabditis elegans and how these are required for molting between larval stages. By performing ChIP-seq on synchronised populations at 12 different timepoints (each separated by 1h), the authors find that RNA pol II (RNAPII) occupancy at >2000 promoters shows an oscillating behaviour that is paralleled by similar changes in mRNA abundance. To identify transcription factors (TFs) involved in generating these cycles, the authors conducted a candidate RNAi screen on 92 TFs annotated as been expressed in an oscillating manner themselves. Monitoring developmental progression of individual animals by a luminescence-based assay, the authors identify 6 TFs that caused either death (or arrest) after aberrant molts (3 TFs) or prolonged duration on molts (3 TFs). One of these, the Grainyhead/LSF transcription factor GRH-1 is then investigated in further details. Temporal control of GRH-1 depletion is achieved by TIR1/auxin-mediated protein degradation. This revealed that GRH-1 is required during each larval molt. In agreement with the criteria for including grh-1 among the candidate RNAi clones, GRH-1 protein levels are shown to peak immediately before entry to each molt.

I am very enthusiastic about the manuscript at several levels: conceptualisation, experimental design, quality of data and clarity of text and discussion. I have only two comments in the category of "major comments":

1. I do not see how the experiments presented in Figure 7 can be used as argument to conclude that "Molting requires oscillatory GRH-1 activity" (title of last Results section and also reflected in the title of the manuscript). I think the experiment nicely shows that there is a timepoint beyond which removal of GRH-1 no longer interferes with the upcoming molt but that doesn't imply that GRH-1 necessarily needs to oscillate. Stronger evidence could be provided by inducible, non-oscillating GRH-1 expression.
2. After reading the manuscript, one is left with the obvious question: which of the many oscillating genes are direct targets of the oscillating TF GRH-1? Experiments to answer this are not strictly needed for the current manuscript, which stands perfectly on its own, but it would make a significant increase in the overall understanding of oscillating genes and GRH-1 in particular.

Minor comments:

The sampling used to generate the data represented in Figure 1 correspond to from 22 hours until 33 hours of post-embryonic development at 25{degree sign}C. It would be useful to indicate how these time points relate to larval development L1-L4.

The authors state that co-oscillation of RNAPII and mRNA is not observed for all genes. Although this might indeed be due to technical limitations are suggested by the authors, it would be relevant to provide more information (e.g. number or percentage of genes).

Are the images of GRH-1::GFP expression in larvae (Fig S8) and adults (Fig S9) acquired with identical settings? I assume so, but it is not obvious, particularly because the images are in separate figures.

Have the authors examined if GRH-1 activity is not only controlled by oscillating transcription, but also post-translationally (e.g. phosphorylation, nuclear import, etc.) as is the case for many TFs. The authors could potentially "overlay" the GRH-1::GFP data with transcriptional data (available at least for 22-33h) to see if the shapes coincide.

Significance

Oscillating genes are found in many biological contexts and are fascinating examples of tightly controlled gene expression. The Großhans laboratory has previously identified close to 4,000 oscillating transcripts and is a leader in this field. The current manuscript incorporates a variety of sophisticated techniques that together enable the authors to identify six genes that are required for rhythmic molting in C. elegans. The protein most deeply studied in the manuscript, GRH-1, is homologous to Grainyhead, which is involved in ECM remodeling and other cyclic processes. The findings in this manuscript are therefore of potential relevance across a broad evolutionary scale.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

Summary:

Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate).

The manuscript is dedicated to a study of functional roles of a panel of cell migration regulatory factors, and notably the highly homologous family of ARF GTPases. The chosen model is a prostate cancer cell line, used in a number of assays in culture, as well as in a study of primary tumor formation and metastasis in mice. The authors apply a rigorous quantitative approach to their assays of 2D and 3D migration in culture, and use artificial intelligence for the analysis of results, thus upgrading their work from mere phenotypic observations, and gaining statistically significant results. The main finding of the study is the discovery of a unique role of ARF3, a regulatory protein that is shown to control a switch between individual and collective cell migration depending on its abundance. In fact, a depletion of ARF3 leads to an increased individual cell migration and invasion, and to increased metastasis formation in mice, whereas an overexpression of ARF3 favors a sheet-like collective cell migration, which is also more efficient than control in culture, but does not induce metastasis in vivo. This phenomenon appears to depend on the levels of cellular N-cadherin, that is shown to be positively regulated by ARF3 on the protein level, by a mechanism that remains unclear. Finally, the authors analyze the expression of ARF3 and N-cad in a variety of tumors of different origins and grades, and attempt to show the prognostic value of these factors for progression-free survival.

Major comments:

• Are the key conclusions convincing?

The majority of conclusions are convincing, with the exception of the observations I will address in the next paragraph. - Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

In Fig. 2B, the phenotypes 1 (Spread, pink line) and 5 (Spindle, yellow line) are described in the text as showing a "modest but robust increase". No such increase is evident by eye, and if it is statistically robust, the information should be presented in the Figure, or the statement should be modified/removed from the article.

In Fig. 2E, the lower middle panel shows a dramatically increased quantity of acini, whereas the authors specifically say that proliferation is not impacted by the KD of ARF3, and indeed, the KD2 looks very much like the control in this respect. It is misleading, and a more typical panel should be presented for ARF3_KD1.

In Fig. 3, the authors study the effects of simple versus double KD of ARF1 and ARF3, and conclude that a double KD leads to a phenotype "midway" between the two simple KDs. However, with regard to the 3D invasion assays (Figs. 3DEJ), it looks like a double KD is less efficient than either of the simple ones, as if ARF1 and 3 were partially mutually dependent in this regulation. It is not clear what "midway phenotype" the authors are talking about. - Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

The authors make a strong case for physical interaction and mutual stabilization of ARF3 and N-cad, though the negative regulation of N-cad following ARF3 depletion is not obvious from Fig. 5B (positive regulation is very clear). Moreover, it is difficult to understand why a total disappearance of ARF3 has such a discreet effect on N-cad, whereas a very modest overexpression of ARF3 leads to such a dramatic increase of N-cad. Perhaps, some experiments with proteasome inhibition (using MG132, for example) could substantiate the authors' claim about the mutual stabilization of the two proteins. - Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

Yes, the experiments are realistic and should not take more than a month. - Are the data and the methods presented in such a way that they can be reproduced?

Yes. - Are the experiments adequately replicated and statistical analysis adequate?

I am not sufficiently qualified in artificial intelligence algorithms to judge this part of the study. In general, as mentioned above, all differences characterized as "significant" or "robust" should have a statistical basis for this statement, which was not always the case in the manuscript.

Minor comments:

• Specific experimental issues that are easily addressable.

Please see above. - Are prior studies referenced appropriately?

To the best of my knowledge, yes. - Are the text and figures clear and accurate?

Yes, with the exceptions described before. - Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

Please see above.

Referees cross-commenting

I fully agree with the detailed and careful analysis made by the reviewers 1 and 2. I do not have any additional comments.

Significance

• Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field.

The study shows, for the first time and in a very clear way, that the small GTPase ARF3 has a unique function in determining the pattern of human cancer cell migration, that this function depends on the C-terminal domain of ARF3 and on N-cadherin (by mechanisms that remain to be elucidated), and that this phenomenon is important for metastases formation in vivo. - State what audience might be interested in and influenced by the reported findings.

The study will be interesting for the field of cell migration, but also for specialists in cancer and metastasis formation. - Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate.

Cell migration 2D 3D, acini assay, RAC-WAVE-ARP2/3 pathway. I am not sufficiently qualified to evaluate the robustness of the artificial intelligence algorithms, nor to judge the relevance of the analysis presented in Fig. 7.

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, Sandilands et al. analyze the role of ARF GTPases ARF1 and ARF3 in human prostate cancer cell setting with specific focus on 3D versus 2D growth and invasion. The study connects to previous work where the authors conducted similar studies with ARF GTPase exchange factor IQSEC1 acting as an invasion promoting factor. Now, the authors interrogate the "ARFome" in prostate cancer cell line PC3 by use of a lentiviral shRNA library. The authors conduct detailed 3D and 2D cell culture analyses to identify specific differences in cell morphology between the different single knockout clones, showing loss of ARF1 and ARF3 as key switches of a spindle like morphology that is associated with enhanced migration and invasion. Interestingly, the authors find ARF3 functioning significantly dependent on the C-terminal region, and, further, that ARF3 is a direct companion of N-cadherin levels, whose downregulation leads to enhanced migratory capability. The main findings of the manuscript are: (1) ARF1 and ARF3 knockdown elicits key-differences in 3D morphology and migratory capacity, (2) specifically ARF3 is associated with maintenance of N-cadherin levels via PSD and RAB11FIP4 effectors, (3) whose downregulation leads to enhanced metastatic spread in a orthotopic xenograft mouse model, and (4) lowered N-cadherin protein / ARF3 mRNA levels identify more aggressive human prostate tumors. Analyses and experiments conducted in the manuscript are highly extensive, they follow robust methods and are well controlled. Results described are highly detailed and are impressively visualized and presented. Key data are highlighted, and the manuscript is clearly structured. However, some data could be described and argued more concisely, which would strongly support the results shown. I recommend publication after some minor but important changes.

Major comments:

1. The entire results are based on studies of a single cell line PC3 derived from a highly aggressive metastasic lesion. To infer such an essential principle of tumor invasion and migration from this may be a bit precarious, and perhaps this principle should also be demonstrated in another cell line. The choice of PC3 cells and its implications should at least be discussed.
2. Results showing cell morphology page 6 / Fig.2: end of paragraph: stating "normally suppressing invasion": seems too far at this state of the manuscript, as these experiments are shown in the next section. maybe better "involved in preservation of a rounded phenotype". Results Suppl.Fig.3f: please use the same colors for the morphologies as in Fig. 2 etc. (round - red, spindle - green, spread - blue).
3. Conclusive sentences should not be put at the beginning of an experiment, before one can know its outcome: Results page 8, Fig.4g, bottom: "This revealed that the C-terminus of ARF3 is required for sheet type invasive activity" maybe put that rather as a conclusion of the whole section.
4. Results showing 2D migration and 3D invasion: In the illustrations of migration and invasion assays shown throughout Figs 3-5 and Suppl. Figs 3 and 4, please clearly state and indicate for each case whether this is 2D migration or 3D invasion, as these two assays are very similar, which is a little confusing throughout the manuscript. Suppl.Fig.3e,k,l: is this 2D or 3D? Results Fig.4b+d and Fig.5g: please amend "3D invasion".
5. Summarizing sketch Results Fig.5k: in the scheme on the right side indication of respective presence / absence of ARF3/N-cadherin is missing. Which state induces which condition? Please amend.
6. ARF3 suppresses PC3 xenograft metastasis shown in Fig.6: N-cadherin stainings of mouse xenografts and metastases are missing.
7. Patient data ARF3 mRNA correlations Fig.7 and Suppl.Fig.6ab / Results page 11: the whole section describing ARF3 mRNA levels in diverse tumor types is too long and a little bit confusing: maybe shorten the text, put Fig.7f+g supplemental, and please indicate the combined GENT2 database in Fig. S6b.
8. Patient data CDH2 mRNA/protein correlations: Fig.7 and Suppl.Fig.6cd / Results page 12: one should also shorten and sharpen this section. Please also exchange Suppl. Fig. panels 6 dc to have the same order as Suppl. Fig.6ab. Regarding the detailed analyses of CDH2 mRNA levels, make a too long story short, essential are N-cadherin protein levels, and these results shown in Fig. 7 hik and Fig. 7rst should be enlarged and highlighted. All other data (Fig. 7jl) and the right bars of panels Fig. 7mno, as well as Fig. 7pq are rather supplemental.
9. The finding of elevated N-cadherin levels correlated with reduced invasion/migration and according better tumor outcome is surprising, particularly with regard to the quite established "EMT" dogma of N-cadherin driven single cell migration. Could you go into more detail about this property of N-cadherin driven mode of reduced tumor spread in the discussion?

Minor comments:

1. Manuscript title: maybe rephrase and reverse order of events: first invasion, then metastasis?
2. Results Suppl.Fig.1ij: which cells are shown? Please amend RWPE-1 and PC3.
3. Results page 5: although already described in Nacke et al 2021, please explain the term "acinus".
4. Results page 6: maybe introduce an additional subchapter? Some subchapter titles could be more explicative.
5. Results page 6, 2nd paragraph: please describe what was done: "revealed that knockdown of ..."
6. Results page 7: please explain more detailed class I ARFs, which ARFs are included in this class?
7. Results page 7, Fig.3f-j and 2nd paragraph: maybe better switch: first 3D, then 2D? Results Fig.3j: maybe amend indication of knockdown "KD" as indicated in Figure 3b-e.
8. The conclusion on page 8 top "this suggests that a function of ClassI ARFs may be to regulate molecules that control collective bahaviours" is quite broad, please be more specific.
9. Suppl.Fig.4n: should be named "l"?
10. Results page 8, introducing sentence: please formulate more clearly and following the previous results.

Significance

Contents/Level of interest/merit:

This study by Sandilands et al. analyzes cellular models of ARF-GTPase linked changes of cell morphology and migration to detect altered prostate cancer cell metastatic behaviour. Understanding the contribution of specific ARF GTPases in cancer cell shape and movement might help to identify markers of disease progression and metastasis.

Strengths/Conclusions:

The authors perform whole ARF-compendium knockdown and conduct detailed data analysis and visualization. The authors perform morphological analyses and conduct migration and invasion studies. Mechanistically, they confirm expression changes of N-cadherin, the key adhesive protein that is regulated. This study underlines the importance of analyzing subtle GTPase pathway differences by detailed morphological observations and methods.

Comment/Weakness:

The authors show extensive and detailed data that have been thoroughly analyzed, and results are presented and described fluently. There is a clear sequence of results description that is presented detailed. Form and contents of the paper is sound. The experiments are highly connected to previous experiments and data and this is also the major drawback of this manuscript: there is a lack of clear description of what is shown because it is already presupposed. Therefore, some sections should be worded and presented in more detail to present results more explicitly. The manuscript can be accepted with minor but essential revisions.

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 study represents a tour de force in advancing our understanding of the Arf family and it's associated regulators/effectors in controlling the morphology of individual cells and collective behaviours in 3D ECM. The authors use shRNA-mediated knockdown in high throughput imaging and AI approaches to define the influence of the ARFome on the morphology of prostate cancer PC3 cells growing as acini structures in 3D ECM. This allowed classification of ARFs and effectors/regulators on the basis of associated phenotypes, and this in itself is a useful resource for the community. It also drove the authors to focus on ARF3 and its regulator ESD and effector Rab11-FIP4. Analysing cell motility in elegant 2D and 3D assays showed that ARF3 levels control the collective migration of PC3 cells in 3D invasion, and the authors relate these findings to an interaction between Arf3 and N-cadherin. Using a mouse model of intra-prostatic injection of PC3 cells they demonstrate clear links between ARF3 levels and metastasis in vivo, and patient data further supports the link between Arf3, N-cadherin and metastasis. Experiments are well controlled and complex data are beautifully presented. In general data support the conclusions, where this is not as clear is highlighted below.

Major comments:

Figure 4: The use of the Arf3/4 chimeras is an interesting approach, used to show that the ARF3 C-terminus is important for its function related to migration/invasion. However the effect this has is not clear- it is not GTP loaded efficiently and may therefore act as a dominant negative. Furthermore the authors do not indicate which intracellular compartment ARF3 associates with, or if this is altered when the ARF3 C-term is replaced by that of ARF4 (ARF3N/1C).

Figure 5: Links to N-cadherin are clearly interesting, but the model proposed in Figure 5K is a little speculative. Clearly it is possible that ARF3/Rab11-FIP4 regulate N-cadherin trafficking such that loss of the pathway leads to degradation and gain promotes stability, but it is also possible that expression levels are controlled at the level of transcription. This could be assessed by a simple surface labelling experiment in wt, overexpressing and knockdown cells, and/or by analysing localisation of N-cad with respect to ARF3 and late endosomes/lysosomes when ARF3 levels are manipulated. Does ESD knockdown similarly impact N-cad?

Figure 6: The metastasis experiments are highly relevant to metastatic prostate cancer. Essentially the overall conclusion that high Arf3 (overexpression) suppresses and low Arf3 (knockdown) supports metastasis are well supported by the data, and the wildtype Arf3 levels sit in between (hence trends are observed but aren't statistically significant). Here it would be interesting to compare ARF3 levels in patient tumours with those in wt, overexpressing and knockdown PC3 cells in the mouse model (if sections are available) to give confidence that the overexpression is within the physiological range. If it were also possible to analyse N-cadherin levels in tumours or metastases that would provide an even stronger case for the mechanism proposed.

Minor comments:

Figure 2: The classification of phenotypes into groups is interesting, but the trends in some groups (eg Group4, 6 and 7) seem very similar. It wasn't clear to me if these are AI generated? Also, knockdown of individual ARFs is often in different groups- is this a reflection of knockdown level?

Significance

The manuscript provides a very significant advance in our understanding of the function of the ARFome with respect to cell morphodynamics. The first two figures and supporting data represent a fantastic resource for the field, and the remaining figures provide new insight into the function of ARF3 in collective cell movement and metastasis in mouse models and patients. Whilst ARF6 and its function in cell migration/invasion/metastasis is well studied, ARF3 has received relatively little attention. This study is therefore of broad interest to the trafficking community, and the new links between ARF3 and invasion/metastasis are broadly of interest to the cell biology and cancer communities. The mechanistic link between N-cadherin and ARF3 is fairly well defined and the fact that high/low levels of both correspond to improved/poor outcomes is a major strength of the study. Expertise: Vesicle trafficking/cell migration/invasion/cancer

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

The work of Sako and co-workers tries to employ a single (two) molecule technique in nanodisks to learn about EGFR TM-JM dimerization versus oligomerization. The authors vary the lipid composition in the disk as well as the phosphorylation state of Thr654 which resides in close proximity to several basic amino acids. The outcome suggested by the authors is that Thr654 phosphorylation switches EGFR from a signaling state to a scaffold state. This a tough problem to investigate and single molecule techniques such as FRET might be a successful way of finding answers. I should mention that I am not a single molecule specialist.

The authors also performed live cell experiments in CHO cells that do not express endogenous EGFR. The results in figure 9 are interesting. However, receptor internalization has not been mentioned.

In general, the work is premature for publication. My major critique points are 1) that there is a lack of statistical analysis in most of the figures and 2) that the lipid composition lacks highly negatively charged phosphoinositides, which are known to bind EGFR at the membrane interface. Likewise, what do we learn from artificial lipid compositions that lack cholesterol? Even non-phase-separated membrane areas contain cholesterol (maybe less). 3) Receptor internalization is not discussed.

Major critique points:

Figure 2. FRET changes after acceptor photobleaching. The excitation and emission wavelengths need to be stated. It is not clear from the traces that there is only a single donor peptide in the disk. Only the traces in the two left panels and the one in the upper right support FRET between Cy3 and Cy5. The "FRET efficiency" is misleading because it suggests that all traces show an increase in donor fluorescence after acceptor bleaching which is clearly not the case. Why is the donor fluorescence dropping? Donor bleaching? Figure 2a supports this. If so, this gives a nonsense FRET readout. There should be an accompanying bar graph with full statistical analysis. There should be control experiments in which two Cy3-tagged EGFR TM-JM molecules are used and photobleaching is still aiming for Cy5.

Figure 3. There is no statistical analysis. As said for Figure 2: FRET efficiency seems only partially reflecting the single fluorophore traces. A change from 0.9 to 0.95 FRET efficiency seems small and statistical relevance is not determined.

Figure 4. The result that cholesterol overwrote the Thr-P effect means that phosphorylation is not relevant in the cellular environment. Does this not contradict the main hypothesis?

Figure 5. It is unclear how fluorescence intensity of CY3-labelled EGFR TM-JM peptides change. Is this due to homo-FRET? Again, a lack of cholesterol is not meaningful.

Figure 9. I don't understand the results from the EGFR expression control. 30 min after EGF stimulation, the EGFR should have been internalized and destroyed in the lysosome. Why is there more EGFR? Do CHO cells lack the machinery to internalize RTKs? Again, no statistics in Figure 10.

Significance

As the test system is very artificial, I don't see major impact. The role of Ser/Thr phosphorylation sites is investigated quite substantially.

I think the study should be re-designed. The cell work is quite interesting but major factors such as receptor internalization have not been considered.

Audience: cell biologists

My expertise: we work on EGFR signaling and EGFR phosphorylation.

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 main goal of this paper is to investigate the assembly of the transmembrane (TM)-juxtamembrane (JM) region of EGFR, using single-pair FRET imaging and a nanodisc technique. In particular, the authors address the role of a threonine residue (Thr654) that is phosphorylated after ligand association and is located in the JM domain of EGFR. The authors showed that anionic lipids, cholesterols, and EGFR Thr654 phosphorylation (pT654) regulate the dimerization and/or oligomerization of EGFR.

Previously, these authors reported that anionic lipids caused the dimerization of JM domains, and that pT654 together with acidic lipids induced the dissociation of the EGFR dimer. In this manuscript, authors show that EGFR dimers assembled in cholesterols vs. anionic lipids display significant conformational changes mainly concerning the positioning of JM and TM. Furthermore, they state that phosphorylation of Thr654 (pT654) promoted oligomerization of TM-JM peptides in cholesterol containing membranes. Finally, the authors show distinct pT654-dependent functional roles and oligomerization states for EGFR in living cells. Authors propose a model in which membrane cholesterol and pT654 work cooperatively to switch the EGFR function from the kinase dimer to the scaffold oligomer

Major comments:

•Not clear what Figure 2 is supposed to indicate? Not clear what peptides are being tested. There is an indication in figure legend that photobleaching was performed but it is not clear how, when or why? If the idea is to validate FRET signal and analysis, then maybe photobleaching was performed to obliterate the acceptor dye and unquench the donor dye? If that is the case the authors must explain the experiment and what they are achieving with it. Maybe showing unquenching of the donor would be advised. Also, there is no mention in the text of the photobleaching experiment. If the goal of figure 2 is only to validate FRET analysis, then maybe it should be moved to suppl data.

•Figure 3-4 address the role of cholesterol and pThr654 on the positioning of TM or JM. TM domains come closer in the presence of pT654 and membrane cholesterol. However, cholesterol but not pTh654 increased the proximity between the C-terminus of JM domains in PC 152 and PC/PS membranes. Although interesting results, the connection between the TM and JM conformational changes and oligomerization results is not direct or even interpretable into a model. The experiment that may be lacking is the testing of the role of cholesterol and pThr654 on the positioning of JM vs TM peptides. These experiments were suggested in Figure 2 but Figure 3-4 test the conformational changes of TM or JM but not of TM vs JM. Actually, Table 1 includes the values for Fig 3-4 FRET measurements which are TM-TM and JM-JM not TM vs JM as suggested in the title. The authors need to clarify this issue.

•Figure 5: The authors suggest that a cooperative effect of cholesterol and Thr654 phosphorylation induces higher-order assembly of the TM-JM peptides. The relationship between these experiments and that of Figure 3-4 needs to be better described.

• igure 6-7: Authors state that in all conditions other than non-phosphorylated peptides in the PC/PS membrane, Cy3 distributions after Cy5 photobleaching (red) peaked at the fluorescence intensity of ~100. However, the absence of statistical analysis of the comparison between different distributions makes this figure difficult to interpret and analyze. If not possible then the authors should stress that the results interpretation is based solely on a qualitative analysis of the curves.

•Figure 8 model in which PS and cholesterol are proposed to exert competitive effects on dimerization and oligomerization, while pT654 disrupts the PS-induced JM dimer, and promotes oligomerization of the peptides, is not readily supported by the presented data.

•Regarding the experiments of EGFR activation/phosphorylation using pT654 and pY1068 in Figure 9.

-Blots should be clearly labeled (mainly pEGFR versus total EGFR and pT654 versus pY1068). Can authors clarify why in panel a only is showed 0 and 30 min (in pY1068 there is 3 time points) and why two main bands (one about 160 kDa and another about 210 kDa) are observed in all phospho or total EGFR immunoblots?

-In panel a, it would be expected a stronger pT654 signal upon EGF in wt PMA(-)? Additionally, here the total EGFR blot is missing.

-In panel b, it is important to clarify what band(s) are being quantified and statistical analyses (t-test or ANOVA for example) should be provided. Also, the graph indicates an about 2-fold increase in pY1068 signal at time 0 when wt PMA(-) is compared with pT654A PMA(-). This is apparently given by low total EGFR in pT654A condition, perhaps lower levels of transfection? it would be also a plus if it is possible to have beta actin or other loading control for blots or mention lower levels of transfection if it is the case.

-Figure 9 presentation and analysis should be improved since it is an important part of main conclusions of the present work. Did authors considered to analyze also the effect a phosphomimetic in T654?

•Figure 10b:

-this reviewer cannot detect any difference between conditions/experiments, if there is something going on, authors need to find a better way to visualize the results and use statistical analysis to determine significance. Considering the lack of statistical analysis, Figure 10d overinterprets results since 10a-c do not provide strong supporting claims.

-In methods authors mention that fluorescence is measured at the basal plasma membrane. However, it is not clear why at the condition pma- egf- the signal of egfr-gfp is so weak. Also, it not very clear why convexness indicates high immobilization.

The following experiments and claims are over-interpreted or speculative. The reviewer would suggest limiting the conclusions extracted from these experiments. Moreover, Discussion is too long, convoluted, and speculative.

•Oligomerization experiments are not convincing, and they are overinterpreted

•homo-FRET (self-quenching) between two N-terminal labeled Cy3 peptides are speculative and not clear how to validate and demonstrate the presence of home-FRET. Maybe fluorescence-anisotropy-based homo-FRET detection could be included.

Minor comments:

•This sentence "Fractions of PS 97 (inner leaflet) and cholesterol (inner and outer leaflets) mimicked those in the plasma membrane" should be referenced

•Figure 2a: the authors should explain what is the meaning of 5 seconds between the two images?

•Figure 3 legend should include what type of peptide was used. According to the text the N-terminal regions of the TM domains were used in Fig. 3. However, that is not included in Fig 3 figure legend.

•Figure 3-7 need the distribution and the 95% percentile section as in Figure 1 for statistical analysis of FRET measurements

•Figures 9 and 10 should include statistical analysis to demonstrate significance

Significance

It should be mentioned that a significant amount of work included in this manuscript was already extensively covered in previous publications of the group. In particular, this paper follows the previous one "Lipid-Protein Interplay in Dimerization of Juxtamembrane Domains of Epidermal Growth Factor Receptor" (Biophys J. 2018 Feb 27; 114(4): 893-903), where authors evaluate the effects of anionic lipids on EGFR dimerization. The authors cite this previous work in the manuscript, but nevertheless the first figures of the present manuscript are very similar to the previous work mentioned. The only difference being that the authors now are adding cholesterol as a new variable. Moreover, text throughout the main part of the manuscript (especially description of Results and Introduction) are very similar to the previous work published. Overall, this similarity between results and approach, decrease the significance and novelty of the present work.

Maeda et al., describes an advance regarding the role of cholesterol in EGFR molecular assembly. This is compared here to the previous findings about effect of phosphorylation in EGFR Thr654 residue and anionic lipids in EGFR molecular assembly. Since EGFR is a widely studied receptor both in health and disease, especially in cancer, the mechanisms of EGFR activation and assembly are of high interest both in basic and clinical research. It would be interesting if the authors of the present work could contextualize the physiological relevance of the findings and how for instance depletion or increase in cholesterol physiologically could impact in EGFR related functions.

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 the manuscript titled "Threonine phosphorylation regulates the molecular assembly and signaling of EGFR in cooperation with membrane lipids", Maeda et al. investigate the associations of the EGF receptor, in particular the short TM-JM linker peptide that connects the extracellular to the intracellular domain and couples ligand binding to kinase activity. A particular focus lies on the influence of cholesterol and the phosphorylation of T654 on this process. Based on single-molecule FRET and intensity measurements in lipid nanodiscs, biochemical activity assays and analysis of EGFR cluster formation in live cell, they conclude that phosphorylation of T654 promotes clustering through dissociation of anti-parallel JM dimers and cholesterol supports the clustering process. In particular, they challenge the view that clusters contain kinase-active EGFR dimers, but rather are tight assemblies of EGFR in form with reduces kinase activity. This is a very interesting story and could explain observations made by the authors and others, e.g. decreased activity of EGFR after initial clustering. The experiments are convincing, but the authors should revisit the interpretation of their results; in particular, they should consider the possibilities (i) that nanodiscs might have different sizes depending on the lipid composition, (ii) that a bias towards repeated interactions is introduced when selecting nanodiscs containing three peptide molecules. Also, the normalization to basal levels of EGFR expression and activity in the Western blot and luminescence assay should be treated more carefully. Nevertheless, it is a good story that would be a good contribution to the EGFR field.

Major comments:

1.Line 93: the authors write that with cholesterol, there were two peaks. But there are three peaks (Fig. 1e, red line). I understand the following sentence that they used the peak at 7-8mL for their experiments and the analysis in Fig. 1f. They then write that this fraction had a similar size without cholesterol. But I see only one peak without cholesterol (at 11-13mL), and it shows a bigger size, as they also wrote before. Or do they refer to the very small bump in the black line at 7-8mL?

2.Figure 6: If the density of proteins in the lipid is low (see previous comment), the selection of nanodiscs with two Cy3 and one Cy5 introduces a bias, in particular in the conditions where there is no higher-order assembly occurring (i.e. with non-phosphorylated T654). Since the three proteins are then restricted to a very small membrane area, they are experiencing a high apparent density, and the equilibrium between assembled and dissociated 3-protein clusters will be shifted to the trimer.

3.Figure 6: In connection to the previous comment, the previous comment #1 becomes very critical. If I understand the red line (PC/PS+chol) and black line (PC/PS) of Fig. 1 correctly, the size of the nanodiscs is bigger for PC/PS than for PC/PS+cholesterol. If this is the case, it will have an impact on the apparent density of the three protein molecules in the nanodisc, since a larger nanodisc means lower membrane density or lower encounter probability. Also, a quick MC simulation shows that the average distance of two randomly positioned proteins in a disc is 0.9*radius, which would be 5nm for the 11nm nanodiscs, and therefore within the Forster radius of Cy3 and Cy5 (5.3nm). With larger nanodiscs, this would change and less FRET would occur in the dissociated state. Therefore, the size of the PC only and PC/PS nanodiscs should be measured similar to the PC/PS+cholesterol. The effects of nanodisc size on apparent density and on FRET in the dissociated state should be considered.

4.Fig. 9b: It seems that the T654A mutation increases EGFR expression since the EGFR-normalized levels are higher in wt, but lower in T654A. Was this effect directly quantified, e.g. by comparing to a standard protein? How can this be explained? The authors claim that pY1068 levels were higher for T654A than for WT. Although this is correct, the levels normalized to total EGFR were lower for T654A than for WT. This is somehow contradictory and should be resolved.

5.Is it possible that basal Grb2 recruitment is different for WT and T654A, and therefore the baseline for the luminescence measurement is different? This should be carefully investigated, since Fig. 9b suggests that T654A has a critical impact on EGFR expression in its inactive state, and therefore can be assumed to influence basal EGFR activity.

6.Line 401: The claim that pT654 promotes downstream signal transduction is based on the apparently increased EGF-induced Grb2 recruitment in the presence of PMA (which leads to T654 phosphorylation) in the EGFR wt but not in the T654A mutant. This claim might not hold when considering that maybe basal levels of Grb2 interaction were different under the different conditions. Then the apparent increase would only be caused by the normalization to the basal luminescence level (see comment #5).

Minor comments:

1.Line 66: the first work should probably be "although" instead "also".

2.Line 101: it is unclear what "16 types of nanodiscs" mean. Were there more lipid compositions than the four compositions in Fig. 1d investigated? Or was it always nanodiscs prepared with the same lipid composition (and same fraction) but 16 different proteins or combinations of proteins?

3.Fig. 3-7 (all FRET and intensity distribution diagrams): An error for a each bin should be calculated or estimated so it is easier to assess if a bump in the line is more than random, and if the difference between two distributions is significant.

4.Line 128: "PS competed with cholesterol when T654 was non-phosphorylated (Fig. 3d)." What does compete exactly mean here and how is this deduced from the data?

5.Line 139: "These results confirmed the results of our previous study". The authors could clarify this by stating what the conclusion was/is (T654 phosphorylation dissociates dimers when acidic lipids are present).

6.Figure 5: adding the conditions into the panels would clarify the figure.

7.Line 167: In how far can be assumed that co-localization of multiple peptides in the same nanodisc is caused by interaction? Could they not be incorporated by coincidence, because of a high density? What is the probability to obtain two non-interacting proteins in the same nanodisc?

8.Figure 6: I cannot find the details to determine pre- and post-bleaching intensities. The authors should state in the methods the time they include to determine the average value on each side of the bleaching event. The same for all values where intensities or FRET efficiencies were calculated by averaging over multiple points of the time course.

9.Line 177: If the low intensity for N-terminally labeled peptides was caused by homo-FRET, there should be an increase of intensity after the first Cy3 bleached. Did the authors observe that? Also, when the trajectories are getting complicated by homo-FRET during dissociation/association, the selection of nanodiscs with two Cy3 might become ambiguous. What were the exact selection criteria? There is no information in the text or methods.

10.Figures 6a/7a: The authors should double check the origin of the data. The panels in the bottom left of Fig. 6a and the bottom left of Fig. 7a seem to contain identical data.

11.Figure 7b-e: Although there are differences in the intensity histograms, I find them to be quite subtle, compared to the effect of pT654. Therefore, I find the claims made in lines 224-229 (induction of oligomerization by cholesterol, and oligomerization instead of dimer by pT654) too strong. Based only on the visual assessment of the intensity distributions, I would rather suggest a shift towards one or the other condition. A model calculation with assumptions for FRET values and distances for certain conditions would strengthen their claims.

12.Figure 8: The different configurations and why they are induced in certain condition needs to better explained. In the main text, there is only one sentence (lines 224-229), and in the figure legend, there is no explanation at all.

13.Line 230: "...whether the peptides in the oligomers directly interacted or not." Since there are no other proteins involved except the peptides, I am wondering how an indirect interaction in the oligomer would work?

14.Fig. 10b: It is surprising that the authors observe a large fraction of oligomers for EGFR WT without EGF or PMA, since EGFR WT is thought to be primarily monomeric in its inactive state. This should be discussed.

15.Fig. 10b: The distributions might better reflect the process of oligomerization if the fraction is multiplied by the size of the oligomer. This would show reflect how many receptors are clustered, instead of how many clusters are present (one large cluster should make the same fraction as the sum of two smaller clusters half the size).

16.Fig. 10a: The authors should present high quality movies of the observations to allow the reader to appreciate the clustering/immobilization better and also to facilitate the comparison of raw data for other researchers in the field.

17.Many experiments in this work assume that the intensity of a fluorophore remains constant while it remains in a specific monomer or dimer state. This should be confirmed with nanodiscs containing positive and negative control proteins for dimerization. E.g. a monomeric protein should give a constant intensity, and a constitutively dimeric protein should give two intensity levels, one before and one after photobleaching of the first dye molecule. Also, FRET should remain quite constant over time in a constitutive dimer labeled with Cy3 and Cy5.

18.It is questionable in how far the postulated trimer formation will happen if the extracellular and kinase domains are present in the full EGFR, since these domains are quite bulky and might not allow certain trimer configurations, e.g. the close trimer of the JM domain. This should be discussed.

Significance

The prevalent view that EGFR clusters contain kinase-active dimers is challenged. The authors claim that instead, tight assemblies of EGFR in form with reduces kinase activity after the initial activation phase. Based on the novel lipid nanodisc FRET experiments, different configurations of the TM-JM domain can be discriminated in the membrane environment. This study is a good contribution to the EGFR field and will be interesting also for people in the single-molecule imaging field. The reviewer has experience in single-molecule imaging and has also worked on the EGFR.

Referess cross-commenting

I want to add two comments:

1. For reviewer #2 's first major comment on the photobleaching: From my point of view as a single-molecule researcher, this experiment is quite clear. Photobleaching is a process that inevitably occurs when using high intensity, as required for single-molecule imaging; therefore, it was not "performed", but (unfortunately) occurs. Only the data before the photobleaching event are useful. But I agree with the reviewer that Fig. 2 could be moved to the supplement, or, in a condensed form, could be merged, e.g. with Fig. 3 and Fig. 4.

2. Concerning reviewer #2 's significance statement: I was not aware of the fact that the nanodisc approach has been used and published by the same authors before. They should have clearly indicated this in a written form, e.g. by saying that they used a previously developed nanodisc FRET assay. Also there is no reference to the previous work in the first results section or the nanodisc section of the methods. This diminishes the significance and novelty of the work considerably in my opinion. Also, as the reviewer correctly observes, in the text of this manuscript there are a number of similarities to the mentioned publication which also reduces the originality of this work.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

Summary:

Lemonidis et al perform a structural and biochemical investigation of FANCI-Ub/FANCD2 complex. The key findings are that FANCD2-Ub is more rapidly deubiquitinated that FANCI-Ub by USP1:UAF1, particularly in the context of the full FANCD2-Ub:FANCI-Ub complex. They present a structure of FANCI-Ub/FANCD2 and show key features that distinguish it from the other FA core complex structures. Several regions are unstructured in their model, which the authors propose is because of the absence of ubiquitin on FANCD2. The exposure of Ubiquitinated lysine of FANCD2 is suggested to provide a mechanism for the rapid reubiquitination of this site when bound to FANCI-Ub, and they also show this biochemically as a faster ubiquitination of FANCD2 when bound to FANCI-Ub instead of FANCI-Ub. A very neat conclusion is made about the dual role of FANCI-Ub in maintaining FANCD2-Ub in a DNA clamped state.

Major comments:

No major issues identified. The data supports the conclusions. Some mutagenesis studies that investigate the hypotheses generated by the structure would increase the overall impact of the study.

Minor comments:

The role of FANCI-Ub and DNA binding in protecting both FANCI and FANCD2 from deubiquitination was first shown in van Twest et al 2017 Molecular Cell, however this is not referred to in the manuscript.

USP1:UAF1 has been shown to have DNA binding activity (eg Liang et al 2016 Cell Reports). Could this DNA binding activity potentially reduce its activity on I-Ub-D2-Ub in the presence of DNA (ie through sequestration away from the substrate by binding to other DNA?). This could be tested by showing that titrating the amount of DNA in the reaction does not inhibit deubiquitination. Also, one previous study showed that DNA stimulates deubiquitination of FANCD2. This study used complexes where FANCI was not ubiquitinated, but is a different result to what is seen in Figure 4A (but perhaps because deubiquitination is already 100% in the absence of DNA in these experiments). Some investigation/discussion of this discrepancy would be good to include. Some further discussions speculating why a FANCI-Ub is necessary in the FA pathway (ie why not just have FANCD2-Ub as the clamping mechanism given that it is sufficient to lock the protein onto DNA?) would increase the interpretation of the work by non-specialists.

Significance

The work provides a significant advance in our understanding of the mechanism by which di-monoubiquitinated FANCD2:FANCI clamps the complex onto DNA. The first structures of I-UbD2 are presented, and some evidence for why they exist in cells (as an intermediate state of deubiquitination of the complex). Previous studies already showed that FANCI-Ub and DNA binding prevent the deubiquitination activity (eg van Twest et al 2017, Arkinson et al 2018, Rennie et al 2020, Wang et al 2020) of USP1:UAF1.

The work has clinical relevance - USP1:UAF1 inhibitors and FA core complex inhibitors are being developed by several biotech for cancer therapy (this is not referenced in the manuscript, and could perhaps be included in the introduction?). Researchers in several fields will be interested in the work: those working on FA pathway and mechanistic understanding of FANCD2 monoubiquitination, as well as those working more generally in ubiquitination and deubiquitination pathways. The structures included "complete the set" for all states of FANCD2:FANCI ubiquitination states, and this is a highly cited field. My expertise is in understanding the mechanistic basis of FANCD2 and FANCI ubiquitination and deubiquitination so I am well suited to evaluate all aspects of the proposal.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #2

Evidence, reproducibility and clarity

The manuscript by Lemonidis et al describes a structural and biochemical basis for the interdependency of FANCI-FANCD2 monoubiquitylation. Continuing their landmark work in the FANCI-FANCD2-Ub Cryo-EM structure, this study shows that the homeostasis of FANCI-Ub-FANCD2-Ub (ID2-diUb) can be partly explained by accessibility differences of the ubiquitylated forms of FANCI and FANCD2 complex. Using structure and enzymatic assays, the authors show that FANCI-Ub is not as susceptible for deubiquitylation by USP1-UAF1 deubiquitinase (DUB) when in complex with FANCD2 and DNA. In contrast, FANCD2 deubiquitiylation is more efficient when FANCI is unmodified, and more protected when it is in complex with FANCI-Ub and DNA. Interestingly, FANCI ubiquitylation progresses at a faster rate when in complex with FANCD2-Ub and DNA. Even though this is somewhat an incremental study, nevertheless, the mechanisms of USP1 can or cannot remove ubiquitylated FANCI-FANCD2-DNA complex is intriguing and opens new areas of study to understand whether several different states of modified and unmodified FANCI-FANCD2 exists physiologically or not and what functional differences they may have.

The major limitions of this study is the lack of functional connection to the described mechanisms of deubiquitylation or reubiquitylation of FANCI-FANCD2 complexes. In the future, hopefully, this group or others will generate point mutants that can selectively separate the functions of FANCI or FANCD2 deubiquitylation or FANCI or FANCD2 reubiquitylation to show the importance of dynamic regulation of these processes in a cellular or in vivo context.

A major comment is whether the authors can by structural inference figure out what part of USP1 N-terminal extension is required for interaction and recognition of FANCI-FANCD2-Ub-DNA complexes. This is the only part of the study that wasn't very satisfying. Perhaps the authors can address this with some modeling and/or with additional mutational analyses on USP1.

One minor comment is that the authors only cite the Smorgorzewska et al, Cell, 2007 paper for observing the first interdependency of FANCI-FANCD2 ubiquitylation in cells (Citations in the Intro and Discussion sections). The authors should also cite Sims et al NSMB, 2007 for the exact same observation shown in the two papers just to be consistent.

Significance

Significance of the study is as indicated above.

The audience is the Fanconi Anemia and ubiquitin field.

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 manuscript, Lemonidis et al concentrate on the role of FANCI ubiquitination in the ubiquitination and deubiquitination of the FANCD2 in the FANCI-FANCD2 complex. They work with purified FANCI and FANCD2 proteins that were in vitro ubiquitinated and with purified USP1-UAF1, allowing them to test the kinetics of the ubiquitination and deubiquitination reactions under different conditions. They also show a cryo EM structure of IUbD2 at a resolution of 4.1Å.

Using biochemical assays, the authors show that ubiquitination of one of the subunits (FANCI or FANCD2) enhances the ubiquitination of the other. D2Ub can be de-ubiquitinated with high efficiency, with no protection by DNA and FANCI. However, when FANCI is ubiquitinated, FANCD2 de-ubiquitination is decreased. It is also more easily re-ubiquinated. The efficiency of de-ubiquitination of FANCIUb is strongly decreases when complexed to DNA and FANCD2, and it is independent of the FANCD2 ubiquitination. The structure shows that the IUbD2, like IUbD2Ub and ID2 Ub, is clamped on the DNA, but the FANCD2K561 site in exposed explaining how re-ubiquitination of the FANCD2 is facilitated.

Major comments:

The key conclusions are overall convincing. The IUbD2 structure shows a conformation in which the FANCD2 lysine is more easily accessible than in ID2 (3B). Biochemical assays support the claim that the ubiquitination of FANCI and FANCD2 promote the ubiquitination of their opposing subunit in ID2 (FANCD2, and FANCI, respectively). Furthermore, robust biochemical assays and structures in Figure 4 support the claim that the ubiquitin on FANCI is relatively more protected than the ubiquitin on FANCD2, of which the latter can be de-ubiquitinated with high efficiency.

1. The major caveat is that these structures were generated using "engineered Ube2T and SpyCatcher-SpyTag", and not via the FA core complex. Recent structural work by the Passmore and Pavletich labs shows the intricate process of ID2 complex ubiquitination, and propose a sequential ubiquitination model that is tied closely to the unique structural properties of the FA core and the ID2 interface. Similarly, the biochemistry is done in the absence of the core complex which may change the kinetics of the ubiquitination and deubiquitination. So, although the biochemical experiments performed are robust, these caveats need to be explicitly noted in the manuscript.
2. Ideally, the authors should also solve cryo EM structure of IUbD2Ub and compare this to 6VAE to show that their in vitro ID2 assembly leads to structures aligning with the FA-core complex ubiquitinated form of ID2.
3. I would suggest including ubiquitinated FANCD2 in the PIFE experiments in Figure 2, to assess differential binding affinity of I and IUb. This would help serve as a control that FANCD2 ubiquitination also strengthens the ID2 complex binding the DNA.
4. The authors claim that "ubiquitination of either of the two subunits of ID2 (FANCI or FANCD2) actually favors ubiquitination of the other subunit" (supported by biochemical and structural data in Figure 3). In Figure 3A-B, it is difficult to draw a conclusion on the "accessibility of the lysine residues" from these figures. Authors should include a measurement of the distances between the Lysines and the NTD of the opposite subunit. This would add a numeric measure to "accessibility".
5. Wang et al, 2020 show that the FANCI K523 is only accessible for ubiquitination upon prior FANCD2 ubiquitination. However, the IUbD2 state could be formed by FANCD2-deubiquitination of IUbD2Ub. Figure 5 may need to changed to remove the arrow ID2 state to IUbD2.
6. In Figure 3B, the structure of IUbD2Ub should be included to allow a proper comparison of the Lysine accessibility of IUbD2 and ID2Ub versus IUbD2Ub.

Minor comments:

1. Please include specific method of expressing, purifying and ubiquitinating FANCI and FANCD2. This method doesn't become clear by referencing previous papers, and these methods might have changed.
2. The final paragraph needs to be reworded. The following sentence is very confusing "Of these two interfaces, one is absent in equivalent ubiquitin-FANCI interactions in ID2Ub-DNA structure (Wang et al, 2020)(Fig. 4D-E)." It does not come across that ubiquitin of FANCIUb lacks an interface with FANCD2.
3. "Our structure indicates that this ubiquitin-FANCD2 interface ..." probably should be "Our structure indicates that the interface between FANCD2 and the ubiquitin of FANCIUb ..."
4. I would suggest adding additional supplemental figures of Figure 3B including 1 extra angle for each structure. This would help gauge "Lysine accessibility".
5. Please add PDB images in Figure 4D from the IUbD2Ub structure of both the interface between FANCD2 and ubiquitin of FANCIUb, and FANCI and the ubiquitin of FANCD2Ub. This is important to allow for the comparison of the different ubiquitin interfaces amongst the IUbD2Ub, ID2Ub, and IUbD2 structures

References:

Previous in vitro and in vivo data well referenced in introduction and discussion.

Statistics:

Satisfactory

Comments on time and resources:

Solving the cryo EM structure of IUbD2Ub is not a trivial ask, but the authors may already have it or may be inclined to do it since they already have access to purified IUb D2Ub and have optimized cryo EM conditions for IUbD2 in this paper. The remaining suggestions seems to be minor (repeat of biochemical experiment, measurement of pre-existing data, and minor changes to figures).

Significance

This work fits very well with the recent structural advances describing ubiquitination and de-ubiquitination of the Fanconi ID2 complex: Wang et al. 2020 and 2021, Alcon et al. 2020. These papers were groundbreaking in describing the mechanism in which FANCD2 and FANCI are ubiquitinated by the FA core complex. The same group (Rennie et al. 2021) solved a structure of ID2 bound to USP1-UAF1 which was also important for understanding of the Fanconi anemia pathway function. The current manuscript sheds light on the role of the Ubiquitinated FANCI in the function of the FANCID2 dimer.

The manuscript will be of interest to the DNA repair, genome integrity, hereditary diseases and ubiquitin fields. This review was written by an investigator in the genome integrity and Fanconi anemia field.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

To reviewer #1

Reviewer #1 (Evidence, reproducibility and clarity (Required)): Please see combined review below in the next section, Reviewer #1 (Significance (Required)):

This is a descriptive manuscript providing a few new insights into a well-recognized and biologically important phenomenon - the lymphatic endothelial cells have heterogeneous origins in different organs. Overall, the idea of Islet1 lineage contributes to regional lymphatic vessel formation during a particular developmental stage is exciting and proven with detailed and careful lineage tracing. The first observation that Islet1 lineage gives rise to cardiac lymphatic vessels was published by the same group in Dev Bio in 2019 so the novelty here is dampened, although the pharyngeal lymphatics and the exact time of these non-venous origin lymphatic vessels arise were not previously characterized - so the current manuscript does provide new important insights. Both the data quality and manuscript layout need improvements, especially when it comes to defining where Islet1 is expressed at all the stages and statistics. The following suggestions will deepen the scope of the manuscript:

First of all, we would like to express our appreciation to the reviewer for all the constructive comments. We carefully read the reviewer’s comments and discussed it. We agree with the reviewer that our manuscript needs improvements with changes in layout several additional experiments. We have also included several description and new immunostaining data (e.g., Isl1,VEGFR3 and LYVE1 co-staining), to confirm our new findings and highlight the importance of the current manuscript beyond our previous one in Dev Biol in 2019. We also have included detailed quantification methods, single-channel images with improved data resolution, and improved clarity of the manuscript.

Specific points were addressed as follows:

COMMENTS BY THE REVIEWER

1. Whether Isl1 lineage is independent of venous-derived endothelial cells.

2. This point is very important: the manuscript does not actually show Isl1 expression through the stages they are inducing. I would want to be sure that lymphatic endothelial cells at this stage don't express Isl1. Another way to get at this is to maybe use other second-heart fields or even broader mesoderm drivers that are known to be never expressed in endothelial cells to confirm the findings.

Response:

In our previous work (Maruyama et al, Dev Biol 452:134–143, 2019, Figure 2C), we demonstrated that Isl1+ lineages using Isl1-Cre mice did not contribute to endothelial cells in the cardinal vein and its branches (intersomitic vessels: ISVs), which had been thought to the primary and biggest sources of lymphatic endothelial cells (LECs). In this paper, we confirmed this finding using Isl1-MerCreMer mice with tamoxifen treatment at E8.5 (Figure 4J). We have scanned whole embryos and detected no eYFP+ cells in the cardinal vein or ISVs (the detailed quantification methods have been added in Methods section). Consistently, another group (Lioux et al., Dev Cell 52:350-363, 2020) re-evaluated this point using Isl1-Cre mice that the Isl1+ lineage contribute to endothelial cells of the cardinal vein only by less than 2%, which neither explains the abundant contribution of the Isl1+ lineage to coronary lymphatics (>50%) nor its restriction to the ventral heart. Based on these reports, we supposed that the Isl1+ lineage was independent of LECs derived from the cardinal vein and ISVs.

In the revised manuscript, we added new data showing thorough expression patterns of Isl1, Prox1, Flk1, and PECAM in the E9.0 to E11.5 pharyngeal arches and cardinal veins by immunostaining and presented them as Supplemental Figure 4. In these sections, we detected Isl1 and Prox1 expression with partial overlapping within the pharyngeal mesodermal core, whereas Isl1 was co-expressed with Flk1, or PECAM neither in vessel-like structures around the mesodermal core nor in the cardinal vein and their surrounding Prox1+/PECAM+ LECs (Supplemental Figure 4H’ and J’) confirming the independency. These findings have been described in the manuscript as follows:

To identify possible Isl1+ LEC progenitors, we investigated the expression patterns of Isl1, Prox1, and vascular endothelial markers (Flk1 and PECAM) by immunostaining sections of E9.0 to E11.5 pharyngeal arches and cardinal veins. Consistent with the previous report (Cai et al., 2003), Isl1 was abundantly expressed in the core mesoderm of the first and second pharyngeal arches corresponding to the CPM from E9.0 to E11.5 (Nathan et al., 2008), where Prox1+ cells also aggregated and partially overlapped with Isl1 signals (Supplemental Figure 4A, A’ C, C’ E, E’ G, G’ I, I’). By contrast, Flk1+ or PECAM+ cells were distributed mainly around the CPM and not expressed Isl1 (Supplemental Figure 4A, A’ C, C’ E, E’ G, G’ I, I’). Furthermore, Isl1 was expressed neither in the endothelial layer of the cardinal vein nor in surrounding Prox1+/PECAM+ LECs (Supplemental Figure 4B, B’ D, D’, F, F’, H, H’, J, and J’). Taken together with the result from Myf5-CreERT2 mice, these results indicate that Isl1+ non-myogenic CPM cells may serve as LEC progenitors independent of venous-derived LECs and the commitment to LEC differentiation occurs before E9.5 in the pharyngeal arch region. (Page 6-7, lines 187-200)

Regarding the use of other second-heart field drivers as the reviewer recommended,

Lioux et al., have already shown the contribution of Mef2c-AHF+ , which marked CPM-derived cells including second heart field, cranial musculatures and connective tissues(Adachi et al., 2020), to ventral cardiac lymphatics. We are also trying to introduce Mef2c-AHF-Cre mice, but it is unfortunately delayed due to the pandemic of COVID-19.

The author stated that Islet1 lineage gives rise to lymphatic endothelial cells via the Tie2 mechanism but did not elaborate on this part. What is the potential relationship between Islet1 and Tie2? Or Tie2 just serves as a pan-endothelial lineage marker here?

Response:

To clearly demonstrate the relationship between Isl1+ and Tie2+ lineages in facial lymphatics, we added schematic representation in Figure 6G, which showed the differential Tie2 expression in lymphatic vessels in the tongue and facial skin.

Related to point 1 and 2, it has been thought that almost all LECs are formed from cardinal vein-derived Tie2+ endothelial cells. However, we identified the presence of Isl1+/Tie2+ LECs in the tongue, which are apparently not originated from the cardinal vein. In previous reports using Tie2-GFP mice or in situ hybridization of Tie2, Tie2 was not detected in the developing LECs at E9.5, 11.5, 13.5, and E15.5(Motoike et al., 2000; Srinivasan et al., 2007). In adult mice, Tie2 expression in lymphatics was only observed in restricted regions (Morisada et al., 2005; Tammela et al., 2005). Taken together with our present data that the differentiation fate of Isl1+ CPM-derived LECs was determined between E6.5 and E9.5 (Figure 2-4, Supplemental Figure 3), Tie2 is supposed to be transiently expressed during LEC differentiation in the tongue from early Isl1+ CPM cells, although it remains difficult to identify the Tie2-expressing stage during non-venous LEC differentiation.

It will be an important future subject to identify the stage and implication of transient Tie2 expression in the lineage and, in this paper, we want to just note that the Tie2+ lineage does not always mean the derivation from cardinal vein endothelial cells.

This point has already been included in the manuscript as follows:

The present study further indicates that the LECs in the tongue are derived from Tie2-expressing cells among the Isl1+ lineage. Although it is unclear whether Isl1+-derived cells at the Tie2-expressing stage represent a venous endothelial identity, this result means that Tie2+ LECs are not equivalent to cardinal vein-derived LECs. (Page 10, 298-301)

The effect of lineage-specific Prox1 knockout is very descriptive, without any discussion of the potential biological function of such cellular origin heterogeneity. This part may be worth a few follow-up experiments in later embryonic stages or even in postnatal stages. The authors demonstrated that loss of Prox1 in Islet1 lineage decreases the number of lymphatic vessels and leads to lymphangiectasia, but whether this phenotype can be later compensated or shows any clinical impact was not proven. Therefore, the statement made in line 206 is questionable, and whether Islet1 lineage-derived lymphatic endothelial cells are dispensable/indispensable remains unclear.

Response:

We agree with the reviewer in that additional follow up experiments using later embryonic or postnatal stages will give an insight into the potential biological function of cellular origin heterogeneity. We are generating lineage-specific Prox1 knockout mice by treating Isl1-MerCreMer; Prox1fl/fl mice with tamoxifen at E8.5 to analyze phenotypes in the facial lymphatic vessels.

The layout of the manuscript needs to be reorganized: 1. Details in statistical methods and quantification logic were completely missing from the manuscript. For example, definitions of "a sample" (how many sections are taken from one biological sample and how many fields take from one section, etc.), "number of vessels per field", "diameters", and of what parameters the numbers were normalized to, etc. need to be described in the materials and methods section. For instance, it is not clear how "tomato+ lymphatic vessels per field/Vegfr3+ lymphatic vessels" was defined. First, what proportion of tomato+ cells need to colocalize with Vegfr3 expression cells in a specific vessel to make this vessel being determined as a "tomato+ lymphatic vessel"? Most data provided here are section immunostaining where "multiple vessels" are very likely coming from different cross-sections of one same vessel in the same field. Second, Vegfr3 can stain venous endothelial cells in earlier stages so the specificity of this marker can be controversial. These are some important technical aspects to include in the revised version. Figures needing more description in quantification methods include but are not limited to Fig 1H, 2H, 2P, 5K-R.

We have revised the statistical methods from the ratio of the count of the number to ratio of the area of lymphatic vessels in Figure 1H, 2H, P, and Supplemental Figure 3I to represent more precisely the contribution of Tomato+ cells to lymphatic vessels. We also added more detailed description of the quantification methods in ‘Materials and Methods’ section, as follows:

Quantification of the section and whole mount images

For the quantification of section immunostaining at E16.5 embryos, the average of two 16-μm-thick sections taken every 50 μm and 10x power field of views (0.42 mm2/field) for each anatomical part (the larynx, the skin of the lower jaw, the tongue, and the cardiac outflow tracts) were subjected to the analyses. In the facial skin, lymphatic vessels in superficial layers of dermis were subjected to the analyses. The middle sagittal sections, including the aorta, larynx, and tongue, which were selected as hall marks of midline, was chosen from created sections. The coronal sections, including both eyes, tongue, and olfactory lobes with left and right symmetrical features, was selected. For E12.5 embryos (Figure 4O), two 16-μm-thick sagittal sections taken every 50 μm, including the 1st and 2nd pharyngeal arches and outflow tracts, were subjected to analyses. The area and the number of cells were measured manually using ImageJ software. For the whole mount immunostaining of embryos and the heart, the whole samples were scanned every 20 μm and confirmed eYFP contribution to LECs (Figure 3) and cardinal veins (Figure 4J, and Supplemental Figure 4B, D, F, H, J). (Page 13, lines 402-416)

We also have tested expression patterns of VEGFR3 with Prox1 or LYVE1 as Supplemental Figure 1. At E14.5, VEGFR3 was widely co-expressed with Prox1 in the tongue, facial skin, and around the pulmonary artery (Supplemental Figure 1A-C’). At E16.5, VEGFR3 was co-expressed with LYVE1 in the tongue, facial skin, and around the pulmonary artery. Thus, we thought that VEGFR3 could be used as a marker of LECs in these cardiopharyngeal region.

This point has been included in the manuscript as follows:

Co-immunostaining of platelet endothelial cell adhesion molecule (PECAM) and vascular endothelial growth factor receptor 3 (VEGFR3), which we confirmed its co-localization with lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1) at E14.5 and E16.5 (Supplemental Figure 1), revealed tdTomato+ LECs in and around the larynx, the skin of the lower jaw, the tongue, and the cardiac outflow tracts, at various frequencies, whereas no such cells were found on the dorsal side of the ventricles, which agrees with our previous study (Maruyama et al., 2019). (Page 4, lines 112-119)

Data resolution needs to be improved. The magnification of the figures in Fig 1-4 is not sufficient to demonstrate the marker colocalization as described in the texts. Single-channel images (such as the ones shown in Fig 5-6 but in higher magnifications) are also necessary to show the co-expression of markers.

Response:

There was a limit on the data capacity when submitting the manuscript. We were therefore obliged to reduce the quality of images and file size. We have revised the figures to add several higher magnification and single-channel images with improved data resolution throughout Figure 1-4.

The experimental design is not well-elaborated in the context. For example, the scientific logic of choosing a particular time point/stage for lineage-knockout induction or sample collection needs to be justified. Also, it seems that the authors are using fl/+ as control littermates in most of the experiments. Any specific reason favors using fl/+ heterozygous instead of fl/fl littermates without cre exposure, which is the more commonly used control sample in this kind of comparison, should be addressed.

Response:

Knockdown of Prox1 in the Tie2+ lineage has shown to cause an initial failure in specification of LECs at E14.5 with no appearance of lymphatics even at E17.5(Klotz et al., 2015; Lioux et al., 2020; Maruyama et al., 2019), indicating that the effect on lymphatic vessels would not be compensated even at E16.5. In addition, the systemic lymphatic network formation is almost completed at E16.5(Srinivasan et al., 2007), and the lineage trace was also evaluated at this stage. Thus, it was reasonable to compare the phenotype at E16.5.

This point has been addressed in the text as follows:

When Prox1 is knocked down in the Tie2+ lineage, an initial failure in specification of LECs was confirmed at E14.5 with a lack of LECs even at E17.5(Klotz et al., 2015; Lioux et al., 2020; Maruyama et al., 2019). Therefore, we compared lymphatic vessel phenotypes at E16.5, by which systemic lymphatics formation is normally completed(Srinivasan et al., 2007). (Page 7, lines 208-212)

In Prox1-flox(Prox1fl/+) mice, recombinant cells were labeled with EGFP(Iwano et al., 2012), as already described in the manuscript (Page 7, lines206-208). Therefore, the recombined cells can be visualized by EGFP expression in both heterozygous (fl/+) and homozygous (fl/fl) mice, which enables phenotype analysis referring to the recombined (knocked-out in fl/fl) cells. Importantly, these mice showed no specific phenotypes(Klotz et al., 2015; Maruyama et al., 2019). It is therefore reasonable to use heterozygous mice as controls to compare the phenotype appropriately. Although fl/fl littermates without cre exposure could usually serve as controls, they do not express EGFP in the Prox1 lineage, detracting from their utility(Klotz et al., 2015; Maruyama et al., 2019).

Some of the phrases are not clear in the text- either because of the writing style or because the corresponding figures failed to support the statements. These include but are not limited to lines 104-106, 122, 206, 226, and 228-233.

104-106: we crossed Isl1-Cre mice, which express Cre recombinase under the control of the Isl1 promoter and in which second heart field-derivatives are effectively labeled, with the transgenic reporter line R26R-tdTomato at E16.5.

Response:

We have re-phrased this sentence as follows:

we crossed Isl1-Cre mice, which express Cre recombinase under the control of the Isl1 promoter and in which second heart field-derivatives are effectively labeled, with the transgenic reporter line R26R-tdTomato and analyzed at E16.5, when lymphatic networks are distributed throughout the whole body. (Pages 4, lines 109-112)

122: After tamoxifen was administered at E8.5, tdTomato+ cells were broadly detected in the muscle in the head and neck regions at E16.5, indicating effective Cre-mediated recombination of the target gene.

Response;

We have re-phrased this sentence as follows:

After tamoxifen was administered at E8.5, tdTomato+ cells were broadly detected in the skeletal muscle in the head and neck regions at E16.5, indicating effective Cre recombination in CPM-derived musculatures. (Page 4, lines 130-132)

We have also included red arrowheads, indicating CPM-derived musculatures in Supplemental Figure 1.

206: These results suggested that defects in LEC differentiation and/or maintenance due to Prox1 deletion in the Isl1+ lineage were compensated for by other cell sources, probably of venous origin, in facial skin, but not in the tongue, resulting in impaired lymphatic vessel formation in the tongue.

Response:

We have re-phrased this sentence as follows:

These results suggested that defects in LEC differentiation and/or maintenance due to Prox1 deletion in the Isl1+ lineage were compensated for by LECs from other cell sources, probably of venous origin, in facial skin, but not in the tongue. (Page 7-8, lines 231-233)

226:　 Almost all of the LYVE1+/PECAM+ lymphatic vessels in the tongue were positive for eGFP in the Tie2-Cre;Prox1fl/+ heterozygous mice (Figure 6A and Supplemental Figure 3D), indicating that the majority of LECs derived from Isl1+ CPM cells developed through Tie2 expression in the tongue.

Response:

We have added new cartoon in Figure 6G to more clearly show the relation of Tie2 expression in Isl1+ lineages. Previous reports have used Tie2-Cre mice to show the vein-derived LECs (Klotz et al., 2015; Srinivasan et al., 2007) , because most of cardinal vein endothelium were composed of Tie2+ lineages. In our present study, in the tongue, most of the LECs were derived from Isl1+/Tie2+ lineages (Figure 1D, H, Figure 2D, Figure 4G, Figure 5B, N, Figure 6A and Supplemental Figure 3F, I). These data suggested that there was a group of Tie2+ lineages even though they are derived from non-venous Isl1+ lineages.

Reference needed for Myf5-Cre as a driver for Myogenic CPM in the results section. Response:

We have included several reference, as shown below:

(Harel et al., 2012, 2009; Heude et al., 2018)

1. Harel et al., Dev cell, 2009
2. Harel et al., PNAS, 2012
3. Heude et al., eLife, 2018

4. In discussion the reference to Pitx2-driven mesenteric lymphatic heterogeneity (Mahadevan et al 2014) is missing yet Islet1 has been shown downstream of Pitx2 (Davis et al 2008). The authors should discuss their findings of gut lymphatic heterogeneity in this context, considering that mediastinum is mesentery-derived.

Response:

Isl1+ CPM-derived LECs have been distributed to the anterior mediastinum and their relationship to mesenteric lymphatic vessels, which continuous with the thoracic duct in the posterior mediastinum, is currently unclear. However, since this paper is valuable for understanding the heterogeneity of the origins of LECs, we have included the indicated paper (Mahadevan et al., 2014) in ‘Introduction’ section to show gut lymphatic heterogeneity. (Page 3, line 67)

To reviewer #2

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

The manuscript entitled "The cardiopharyngeal mesoderm contributes to lymphatic vessel development" identified a novel non-venous origin of craniofacial and cardiac LECs using genetic lineage tracing. Their results also revealed the spatiotemporal difference between CPM- and venous-derived LECs. Overall, the paper is well-organized and has certain implications for understanding lymphatic development. However, some issues still need to be improved:

First of all, we would like to express our appreciation to the reviewer for all the constructive comments. We carefully read the reviewer’s comments and discussed it. We agree with the reviewer’s comments to make the text easier to understand and emphasize what we really want to say.

Specific points were addressed as follows:

(1). Clearly, the introduction needs to be more concise and focused on the main questions you propose to answer and why these questions are important.

Response:

We have revised introduction section to be more concise and focus on the developmental process of lymphatic vessels and its relation to CPM. (Page 2-4, lines 41-103)

(2). In the discussion section, you should focus on how the questions have been answered and what they mean. And it would be rash to infer the role of LECs in lymphatic malformation. It would be helpful to validate the changes of CPM-derived LECs in LM patient samples.

Response:

We have revised the discussion section to be more concise. To demonstrate our findings more clearly, we have also revised and added some cartoons in Figure 6G and Figure 7.

(3). For the statistical analysis, all the quantitative data should be tested for statistical significance. There are several bar charts lacking P values.

Response:

We have included P values in the Figure legends.

Reviewer #2 (Significance (Required)):

This study enriched the contribution of CPMs to broader regions of the facial, cardiac and laryngeal lymphatic network and revealed the spatiotemporally difference between CPM- and venous-derived LECs, which provided some basic reference for understanding lymphatic vessel development.

To reviewer #3

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

Short summary of the findings and key conclusions:

The work from Murayama and colleagues traces the ontogenetic origin of the endothelial cells of the lymphatic vessels in the head and neck region. Using the Cre-lox-based mouse genetics approach, they conclude that the lymphatic endothelial cells (LECs) in this region have mixed origin, with contributions from both the cardiopharyngeal mesoderm (CPM) as well as from cardinal vein. The lineage tracing study is buttressed by assaying LEC formation following selective deletion of the key LEC regulator Prox1 in CPM lineage.

First of all, we would like to express our appreciation to the reviewer for all the constructive comments. We carefully read the reviewer’s comments and discussed it.

Specific points were addressed as follows:

Major comments: 1) The key conclusions: LECs in the head and neck region derive from CPM. LECs in this region have mixed developmental origins. Both these conclusions are convincingly supported by the study. However, the work would greatly be strengthened by Pax3-Cre lineage tracing. This would complement the Isl1-Cre lineage tracing. As the authors observe, the LEC descendants of Isl1+ cells also appear to go through Tie2+ state. Therefore, Tie2-Cre study has not helped to delineate the LECs of CPM and cardinal vein origins. In this context, tracing with Pax3-Cre is likely to give a very clear picture of LEC origins.

Response:

We agree with the reviewer in that the data using Pax3-Cre mice will strengthen our manuscripts. Unfortunately, we could not find out researchers who had this line in our society in Japan. For using this line, we need to get cryo-recovered mice from Jaxon laboratory. It will take at least several months. Therefore it is not realistic for us to use Pax3-Cre mice in this work because of time limitation. Instead, we addressed this issue by rewriting the discussion on the possible complementation with the Pax3-Cre lineage by citing (Lupu et al., 2022; Stone and Stainier, 2019).

This point has been addressed in the text as follows:

A recent study has suggested that Pax3+ paraxial mesoderm-derived cells contribute to the cardinal vein and therefore venous-derived LECs originate from the Pax3+ lineage (Stone and Stainier, 2019). The same group has further argued that the Pax3+ lineage gives rise to lymphatic vessels on the trunk side through lymphangiogenesis(Lupu et al., 2022). Therefore, the Isl1+ and Pax3+ lineages may complement each other to form systemic lymphatic vessels. (Page 10, lines 314-319)

2) In addition, the article should be revised to include the number of sections and the number of cells counted per embryo in the Figure legend in each case. This will help assess how robust and reliable are the measurements.

Response:

We have revised the statistical methods from the ratio of the count of the number to the area in Figure 1H, 2H, P, and Supplemental Figure 3I to demonstrate more precisely the contribution of Tomato+ cells in lymphatic vessels. We also added more detailed description of the quantification methods in ‘Materials and Methods’ section, as follows:

Quantification of the section and whole mount images

For the quantification of section immunostaining at E16.5 embryos, the average of two 16-μm-thick sections taken every 50 μm and 10x power field of views (0.42 mm2) for each anatomical part (the larynx, the skin of the lower jaw, the tongue, and the cardiac* outflow tracts) were subjected to the analyses. In the facial skin, lymphatic vessels in superficial layers of dermis were subjected to the analyses. The middle sagittal sections, including the aorta, larynx, and tongue, which were hall marks of midline, was selected from created sections. The coronal sections, including both eyes, tongue, and olfactory lobes with left and right symmetrical features, was selected. For E12.5 embryos (Figure 4O), two 16-μm-thick sagittal sections taken every 50 μm, including the 1st and 2nd pharyngeal arches and outflow tracts, were subjected to analyses. The area and the number of Prox1+ cells were measured manually using ImageJ software. For the whole mount immunostaining of embryos and the heart, the whole samples were scanned every 20 μm and confirmed eYFP contribution to LECs (Figure 3) and cardinal veins (Figure 4J, and Supplemental Figure 4B, D, F, H, J). * (Pages 13, lines 402-416)

We have also included the number of eYFP+/Prox1+ cells among Prox1+ cells in the first and second pharyngeal in the Figure 4O legends as follows;

• (the number of eYFP+/Prox1+ cells (10.83 (mean) ± 1.249 (SEM)): Prox1+ cells (30.83 ± 4.549)) or E9.5 (the number of eYFP+/Prox1+ cells (2.833 ± 1.108): Prox1+ cells (35.50　±　5.847)). (Page 23, lines 684-686)*

Minor comments: 1) Several groups have contributed to the CPM literature. The citation of seminal works from Tzahor and Kelly groups is good, however, work from other groups has not been cited. For example, reports such as Heude et al and Grimaldi et al from Tajbakhsh group are very relevant to this work.

Response:

According to reviewer’s suggestion, we have included following references in the introduction section for the explanation of CPM derivatives. (P3, line 70)

1. (Heude et al., 2018)
2. (Grimaldi et al., 2022)

   2) It would help the reader if the authors explain the reasons for selecting specific regions, such as the tongue, and the skin of the lower jaw, for the study.

Response:

This is because many lymphatic vessels are distributed in these cardiopharyngeal area and these area is well known as anatomical parts where lymphatic malformation most often occurs. This has been mentioned in the manuscript as follows:

From a clinical viewpoint, head and neck regions contributed by the CPM are the most common sites of lymphatic malformations (LMs) (Page 3-4, lines 99-100)

3) The authors should consider presenting the wholemount images, such as those in Figures 3A and 3E for Figures 5 and 6. This would help assess the lymphatic vessel development in a holistic manner.

Response:

Although we tried to do the whole mount images of facial and tongue lymphatics, we could not succeed. Antibodies did not penetrate well on the tongue and, as for lymphatics of facial skin, their complicated morphology prevented clear visualization. Whole-mount imaging of the entire head was difficult for the same reason. In our experience, the antibody was useful for immunostaining of the early-stage embryos (up to E11.5) and the surface area of the heart, where lymphatic vessels were distributed on the epicardium. Even in the whole-mount heart, we have not succeeded in clear and estimable imaging of the vascular structure in the myocardium. Instead, we improved the quality of images and statistical comparisons in the revised manuscript, which we believe makes it more convincing.

Reviewer #3 (Significance (Required)):

The nature and significance of the advance for the field & the work in the context of the existing literature: Groups working in the domain of cardiopharyngeal mesoderm (CPM) have focussed on skeletal muscle and heart development. This pool is also known to give rise to skeletal tissues as well as blood vessel endothelium. A recent work Nomaru et al. (Morrow group, Nat Commun 2021) has identified a multi-lineage primed population in the cardiopharyngeal field. In this context, the work from Maruyama and colleagues highlights the versatility of CPM by providing evidence for the emergence of LEC from this multipotent pool. This complex developmental potential of CPM has implications to understand the evolutionary origin of CPM itself.

The connective tissues in the head/neck have mixed origins (Heude et al, 2018 and Grimaldi et al 2022 from Tajbakhsh group)- from CPM as well as neural crest. This work shows mixed origin for LECs. These works begin to put together the pieces of the puzzle of vertebrate head evolution. Jacob proposed evolution is tinkering. This appears to be true both at the molecular level as well as the cellular level. Head tissues appear to have been put together by exploiting varied sources.

The study is of broad interest to developmental biologists.

Reviewer: A developmental biologist with an interest in understanding the axial patterning of mesoderm early during mammalian development. Not an expert in lymphatic vasculature development.

References for the revision

Adachi N, Bilio M, Baldini A, Kelly RG. 2020. Cardiopharyngeal mesoderm origins of musculoskeletal and connective tissues in the mammalian pharynx. Development 147:dev185256. doi:10.1242/dev.185256

Cai C-L, Liang X, Shi Y, Chu P-H, Pfaff SL, Chen J, Evans S. 2003. Isl1 Identifies a Cardiac Progenitor Population that Proliferates Prior to Differentiation and Contributes a Majority of Cells to the Heart. Dev Cell 5:877–889. doi:10.1016/s1534-5807(03)00363-0

Grimaldi A, Comai G, Mella S, Tajbakhsh S. 2022. Identification of bipotent progenitors that give rise to myogenic and connective tissues in mouse. Elife 11:e70235. doi:10.7554/elife.70235

Harel I, Maezawa Y, Avraham R, Rinon A, Ma H-Y, Cross JW, Leviatan N, Hegesh J, Roy A, Jacob-Hirsch J, Rechavi G, Carvajal J, Tole S, Kioussi C, Quaggin S, Tzahor E. 2012. Pharyngeal mesoderm regulatory network controls cardiac and head muscle morphogenesis. Proc National Acad Sci 109:18839–18844. doi:10.1073/pnas.1208690109

Harel I, Nathan E, Tirosh-Finkel L, Zigdon H, Guimarães-Camboa N, Evans SM, Tzahor E. 2009. Distinct Origins and Genetic Programs of Head Muscle Satellite Cells. Dev Cell 16:822–832. doi:10.1016/j.devcel.2009.05.007

Heude E, Tesarova M, Sefton EM, Jullian E, Adachi N, Grimaldi A, Zikmund T, Kaiser J, Kardon G, Kelly RG, Tajbakhsh S. 2018. Unique morphogenetic signatures define mammalian neck muscles and associated connective tissues. Elife 7:e40179. doi:10.7554/elife.40179

Klotz L, Norman S, Vieira JM, Masters M, Rohling M, Dubé KN, Bollini S, Matsuzaki F, Carr CA, Riley PR. 2015. Cardiac lymphatics are heterogeneous in origin and respond to injury. Nature 522:62–67. doi:10.1038/nature14483

Lioux G, Liu X, Temiño S, Oxendine M, Ayala E, Ortega S, Kelly RG, Oliver G, Torres M. 2020. A Second Heart Field-Derived Vasculogenic Niche Contributes to Cardiac Lymphatics. Dev Cell 52:350–363. doi:10.1016/j.devcel.2019.12.006

Lupu I-E, Kirschnick N, Weischer S, Martinez-Corral I, Forrow A, Lahmann I, Riley PR, Zobel T, Makinen T, Kiefer F, Stone OA. 2022. Direct specification of lymphatic endothelium from non-venous angioblasts. Biorxiv 2022.05.11.491403. doi:10.1101/2022.05.11.491403

Mahadevan A, Welsh IC, Sivakumar A, Gludish DW, Shilvock AR, Noden DM, Huss D, Lansford R, Kurpios NA. 2014. The Left-Right Pitx2 Pathway Drives Organ-Specific Arterial and Lymphatic Development in the Intestine. Dev Cell 31:690–706. doi:10.1016/j.devcel.2014.11.002

Maruyama K, Miyagawa-Tomita S, Mizukami K, Matsuzaki F, Kurihara H. 2019. Isl1-expressing non-venous cell lineage contributes to cardiac lymphatic vessel development. Dev Biol 452:134–143. doi:10.1016/j.ydbio.2019.05.002

Morisada T, Oike Y, Yamada Y, Urano T, Akao M, Kubota Y, Maekawa H, Kimura Y, Ohmura M, Miyamoto T, Nozawa S, Koh GY, Alitalo K, Suda T. 2005. Angiopoietin-1 promotes LYVE-1-positive lymphatic vessel formation. Blood 105:4649–4656. doi:10.1182/blood-2004-08-3382

Motoike T, Loughna S, Perens E, Roman BL, Liao W, Chau TC, Richardson CD, Kawate T, Kuno J, Weinstein BM, Stainier DYR, Sato TN. 2000. Universal GFP reporter for the study of vascular development. Genesis 28:75–81. doi:10.1002/1526-968x(200010)28:23.0.co;2-s

Nathan E, Monovich A, Tirosh-Finkel L, Harrelson Z, Rousso T, Rinon A, Harel I, Evans SM, Tzahor E. 2008. The contribution of Islet1-expressing splanchnic mesoderm cells to distinct branchiomeric muscles reveals significant heterogeneity in head muscle development. Development 135:647–57. doi:10.1242/dev.007989

Srinivasan RS, Dillard ME, Lagutin OV, Lin F-J, Tsai S, Tsai M-J, Samokhvalov IM, Oliver G. 2007. Lineage tracing demonstrates the venous origin of the mammalian lymphatic vasculature. Gene Dev 21:2422–2432. doi:10.1101/gad.1588407

Stone OA, Stainier DYR. 2019. Paraxial Mesoderm Is the Major Source of Lymphatic Endothelium. Dev Cell 50:247-255.e3. doi:10.1016/j.devcel.2019.04.034

Tammela T, Saaristo A, Lohela M, Morisada T, Tornberg J, Norrmén C, Oike Y, Pajusola K, Thurston G, Suda T, Yla-Herttuala S, Alitalo K. 2005. Angiopoietin-1 promotes lymphatic sprouting and hyperplasia. Blood 105:4642–4648. doi:10.1182/blood-2004-08-3327

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

Short summary of the findings and key conclusions:

The work from Murayama and colleagues traces the ontogenetic origin of the endothelial cells of the lymphatic vessels in the head and neck region. Using the Cre-lox-based mouse genetics approach, they conclude that the lymphatic endothelial cells (LECs) in this region have mixed origin, with contributions from both the cardiopharyngeal mesoderm (CPM) as well as from cardinal vein. The lineage tracing study is buttressed by assaying LEC formation following selective deletion of the key LEC regulator Prox1 in CPM lineage.

Major comments:

1. The key conclusions: LECs in the head and neck region derive from CPM. LECs in this region have mixed developmental origins. Both these conclusions are convincingly supported by the study. However, the work would greatly be strengthened by Pax3-Cre lineage tracing. This would complement the Isl1-Cre lineage tracing. As the authors observe, the LEC descendants of Isl1+ cells also appear to go through Tie2+ state. Therefore, Tie2-Cre study has not helped to delineate the LECs of CPM and cardinal vein origins. In this context, tracing with Pax3-Cre is likely to give a very clear picture of LEC origins.
2. In addition, the article should be revised to include the number of sections and the number of cells counted per embryo in the Figure legend in each case. This will help assess how robust and reliable are the measurements.

Minor comments:

1. Several groups have contributed to the CPM literature. The citation of seminal works from Tzahor and Kelly groups is good, however, work from other groups has not been cited. For example, reports such as Heude et al and Grimaldi et al from Tajbakhsh group are very relevant to this work.
2. It would help the reader if the authors explain the reasons for selecting specific regions, such as the tongue, and the skin of the lower jaw, for the study.
3. The authors should consider presenting the wholemount images, such as those in Figures 3A and 3E for Figures 5 and 6. This would help assess the lymphatic vessel development in a holistic manner.

Significance

The nature and significance of the advance for the field & the work in the context of the existing literature:

Groups working in the domain of cardiopharyngeal mesoderm (CPM) have focussed on skeletal muscle and heart development. This pool is also known to give rise to skeletal tissues as well as blood vessel endothelium. A recent work Nomaru et al. (Morrow group, Nat Commun 2021) has identified a multi-lineage primed population in the cardiopharyngeal field. In this context, the work from Maruyama and colleagues highlights the versatility of CPM by providing evidence for the emergence of LEC from this multipotent pool. This complex developmental potential of CPM has implications to understand the evolutionary origin of CPM itself.

The connective tissues in the head/neck have mixed origins (Heude et al, 2018 and Grimaldi et al 2022 from Tajbakhsh group)- from CPM as well as neural crest. This work shows mixed origin for LECs. These works begin to put together the pieces of the puzzle of vertebrate head evolution. Jacob proposed evolution is tinkering. This appears to be true both at the molecular level as well as the cellular level. Head tissues appear to have been put together by exploiting varied sources.

The study is of broad interest to developmental biologists.

Reviewer: A developmental biologist with an interest in understanding the axial patterning of mesoderm early during mammalian development. Not an expert in lymphatic vasculature 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 #2

Evidence, reproducibility and clarity

The manuscript entitled "The cardiopharyngeal mesoderm contributes to lymphatic vessel development" identified a novel non-venous origin of craniofacial and cardiac LECs using genetic lineage tracing. Their results also revealed the spatiotemporal difference between CPM- and venous-derived LECs. Overall, the paper is well-organized and has certain implications for understanding lymphatic development. However, some issues still need to be improved:

1. Clearly, the introduction needs to be more concise and focused on the main questions you propose to answer and why these questions are important.
2. In the discussion section, you should focus on how the questions have been answered and what they mean. And it would be rash to infer the role of LECs in lymphatic malformation. It would be helpful to validate the changes of CPM-derived LECs in LM patient samples.
3. For the statistical analysis, all the quantitative data should be tested for statistical significance. There are several bar charts lacking P values.

Significance

This study enriched the contribution of CPMs to broader regions of the facial, cardiac and laryngeal lymphatic network and revealed the spatiotemporally difference between CPM- and venous-derived LECs, which provided some basic reference for understanding lymphatic vessel 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

Please see combined review below in the next section,

Significance

This is a descriptive manuscript providing a few new insights into a well-recognized and biologically important phenomenon - the lymphatic endothelial cells have heterogeneous origins in different organs. Overall, the idea of Islet1 lineage contributes to regional lymphatic vessel formation during a particular developmental stage is exciting and proven with detailed and careful lineage tracing. The first observation that Islet1 lineage gives rise to cardiac lymphatic vessels was published by the same group in Dev Bio in 2019 so the novelty here is dampened, although the pharyngeal lymphatics and the exact time of these non-venous origin lymphatic vessels arise were not previously characterized - so the current manuscript does provide new important insights. Both the data quality and manuscript layout need improvements, especially when it comes to defining where Islet1 is expressed at all the stages and statistics. The following suggestions will deepen the scope of the manuscript:

1. Whether Isl1 lineage is independent of venous-derived endothelial cells.
2. This point is very important: the manuscript does not actually show Isl1 expression through the stages they are inducing. I would want to be sure that lymphatic endothelial cells at this stage don't express Isl1. Another way to get at this is to maybe use other second-heart fields or even broader mesoderm drivers that are known to be never expressed in endothelial cells to confirm the findings.
3. The author stated that Islet1 lineage gives rise to lymphatic endothelial cells via the Tie2 mechanism but did not elaborate on this part. What is the potential relationship between Islet1 and Tie2? Or Tie2 just serves as a pan-endothelial lineage marker here?
4. The effect of lineage-specific Prox1 knockout is very descriptive, without any discussion of the potential biological function of such cellular origin heterogeneity. This part may be worth a few follow-up experiments in later embryonic stages or even in postnatal stages. The authors demonstrated that loss of Prox1 in Islet1 lineage decreases the number of lymphatic vessels and leads to lymphangiectasia, but whether this phenotype can be later compensated or shows any clinical impact was not proven. Therefore, the statement made in line 206 is questionable, and whether Islet1 lineage-derived lymphatic endothelial cells are dispensable/indispensable remains unclear.

The layout of the manuscript needs to be reorganized:

1. Details in statistical methods and quantification logic were completely missing from the manuscript. For example, definitions of "a sample" (how many sections are taken from one biological sample and how many fields take from one section, etc.), "number of vessels per field", "diameters", and of what parameters the numbers were normalized to, etc. need to be described in the materials and methods section. For instance, it is not clear how "tomato+ lymphatic vessels per field/Vegfr3+ lymphatic vessels" was defined. First, what proportion of tomato+ cells need to colocalize with Vegfr3 expression cells in a specific vessel to make this vessel being determined as a "tomato+ lymphatic vessel"? Most data provided here are section immunostaining where "multiple vessels" are very likely coming from different cross-sections of one same vessel in the same field. Second, Vegfr3 can stain venous endothelial cells in earlier stages so the specificity of this marker can be controversial. These are some important technical aspects to include in the revised version. Figures needing more description in quantification methods include but are not limited to Fig 1H, 2H, 2P, 5K-R.
2. Data resolution needs to be improved. The magnification of the figures in Fig 1-4 is not sufficient to demonstrate the marker colocalization as described in the texts. Single-channel images (such as the ones shown in Fig 5-6 but in higher magnifications) are also necessary to show the co-expression of markers. Figure numbers are missing from the main figures making it a bit challenging to read.
3. The experimental design is not well-elaborated in the context. For example, the scientific logic of choosing a particular time point/stage for lineage-knockout induction or sample collection needs to be justified. Also, it seems that the authors are using fl/+ as control littermates in most of the experiments. Any specific reason favors using fl/+ heterozygous instead of fl/fl littermates without cre exposure, which is the more commonly used control sample in this kind of comparison, should be addressed.
4. Some of the phrases are not clear in the text- either because of the writing style or because the corresponding figures failed to support the statements. These include but are not limited to lines 104-106, 122, 206, 226, and 228-233.
5. Reference needed for Myf5-Cre as a driver for Myogenic CPM in the results section.
6. In discussion the reference to Pitx2-driven mesenteric lymphatic heterogeneity (Mahadevan et al 2014) is missing yet Islet1 has been shown downstream of Pitx2 (Davis et al 2008). The authors should discuss their findings of gut lymphatic heterogeneity in this context, considering that mediastinum is mesentery-derived.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

The authors do not wish to provide a response at this time.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #2

Evidence, reproducibility and clarity

Summary:

In this study, Dobson and colleagues use the well-established Drosophila mitonuclear model to ask how diversity in mitochondrial and nuclear background impacts reproductive response to various diets (DMN, diet-by-mito-by-nuclear interactions). For this, they generated populations from three different geographical locations by introgressions to produce different mitonuclear combinations, which they produced in triplicates, and after preliminary analyses they excluded one geographical population (Canada), which brought down the total number of populations from 27 to 12. Then, they studied at the effects of 3 diets: a standardized control diet, a diet enriched in essential amino acids, and one enriched in lipids on various reproductive traits in two feeding regimes: one chronic (parents and offspring fed the diets), or parental (parents fed the experimental diets, eggs developed on a standard medium). They found that the magnitude of the dietary effects and the effect of parental nutrition on offspring fitness were dependent on mitonuclear genotype, with some diets unexpectedly deleterious in specific cases. They also show that this DMN variation is repeatable among independent replicates, and that mitonuclear epistasis plays a major role in the response to nutrition. Finally, they find an association between DMN interactions and a polymorphism in the mitochondrial gene long ribosomal RNA (mt:lrRNA), showing that a mitochondrial regulatory factor plays a role in DMN effects.

Major comments:

The authors have collected an impressive and very useful amount of data, with rigorous breeding regimes and statistical analysis. Their findings, interpretations and conclusions are in my opinion strongly supported by their experimental data. I have no major concerns with the study, and I believe it deserves to be in a wide-audience and high-impact journal. I only have some minor corrections and suggestions listed as follows.

Minor comments:

• Have the authors thought about looking at rates of food consumption, and could some populations be consuming less (or none) of the experimental diets, possibly explaining lethality?

• Line 136: this study's data does indeed show diet-dependent effects of mitonuclear interactions, but it not the first one, maybe rephrase? See refs (a couple are cited in the supplementary text, but it would be useful to add to the main text): Aw, W. C. et al (2018). PLoS Genetics doi:10.1371/journal.pgen.1007735 Camus, M. F., Moore, J., & Reuter, M. (2020). Biology Letters, 16(2), 20190891. doi:10.1098/rsbl.2019.0891 Camus, M. F., Kotiadis, V., Carter, H., Rodriguez, E., & Lane, N. (2022). bioRxiv, doi:10.1101/2022.02.10.479862 Cormier, R. P. J. et al. (2019). Scientific Reports. doi:10.1038/s41598-018-36060-5 Rodríguez, E. et al. (2021). Frontiers in Genetics. doi:10.3389/fgene.2021.734255 Towarnicki, S. G., & Ballard, J. W. O. (2018). Frontiers in Genetics. doi:10.3389/fgene.2018.00593

• Figure 1D and results section lines 156-178: why is the term "eggs" included in development? I assume it refers to the measure of the amount of eggs, could the authors explain and mention it in the main text and/or figure legend? And why is it "eggs+1" in fig S2D, I could not find an explanation in the methods for this difference in the label.

• Line 221, last sentence of the paragraph: I would make it clearer that this was in the "chronic" paradigm, I found it difficult to follow here.

• Line 446 typo in the figure legend, should read "24h on..."

• Figure 2A: "Genome origin" section of the legend could be placed on the top of the legend, I would find it easier to see as we read the figure from left to right.

• Figure 2B: some kind of separation between T and C (dashed lines for eg.) would be appreciated to be able to compare better at a glance.

• Figure S2D: boxplot lines seem very thin, could the outline be made clearer? It was especially difficult to see on a printed version.

• Line 506, S2B legend: should "Benin" be used instead of "Dahomey"? Keeping the Dahomey mentioned only in Supporting text line 153.

• Figure S3B and its legend text: the diets now appear completely different, except for EAA. Why is it "sya", should it be the same as "rearing diet" in S2B, or is it the "control" diet? "Mar", margarine, is to my understanding the lipid diet mentioned throughout the paper? I assume this is a matter of fixing the legend labels, unless I misunderstood this part. In the legend text lines 532-536, I could not find any explanation for this difference in labels.

• Supplementary text line 43: "wild-type" meaning Benin flies?

• Supp. text line 46: "development" medium is sya? See comment above about confusing diet names.

• Supp. text line 51: I think the figures referenced are not the correct ones (S2B and S2C)? Should they be S2D and S2E?

• Supp. text line 64: "AA1" is repeated twice in the parentheses, should it read "AA3" instead?

CROSS-CONSULTATION COMMENTS

I have read reviewer #1's comments and I see that we both agree on the quality and relevance of the paper, and the very minor corrections needed. Looking forward to seeing this published.

Significance

This study appears to me as very relevant in the context of predicting outcomes of dietary regimes (and more) given a certain mitonuclear background, which is very much needed in the field and in the context of personalized medicine. Repeatability of the effects gives confidence to the type of breeding regime and analyses done in the field of mitonuclear interactions (for example, in the research done by the authors cited previously). Pinpointing a synonymous SNP in a non-coding region of mtDNA is important in the context of deciphering the mechanism(s) behind DMN. As I come from the mitochondrial physiology and aging field, and currently work on mitonuclear interactions, these findings appear to me of great importance, but they will also appeal to a wider audience in the evolutionary biology and biomedicine fields; these results will contribute to advancing our knowledge of mitochondria-nucleus interactions in different environments, up to the human personalized medicine perspective.

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:

Here, the authors investigate mitochondrial x nuclear x diet interactions in Drosophila melanogaster. They start by constructing D. melanogaster populations comprising fully-factorial combinations of mitochondrial and nuclear genomes from Australia, Benin, and Canada. This creates discrete populations with distinct combinations of mitochondrial and nuclear genomes. These populations are then exposed to various diets, notably a control diet, a diet high in essential amino acids (which promotes fecundity), and a diet high in plant-based lipids (which represses fecundity). They first screened for evidence of repeatable mitonuclear effects on fecundity, which they found. They then screened for additional fitness traits looking for influence of mito-nucleotype on response to chronic vs parental dietary changes. Once again, they were able to find evidence of phenotypic variation dependent on mito-nucleotype; however, the effect of mito-nucleotype on traits was variable. Certain mito-nucleotypes exhibited lower to near lethality progeny counts after amino acid feeding (which is thought to promote fecundity), while others exhibited normal counts. When quantifying the size of various effects, they found that mitonucleotype interactions often had comparable effect size to that of diet:mitotype interactions, diet:nucleotype interactions, and diet on its own. Finally, they were able to associate this mito-nucleotype interaction with a mt:lrRNA C/T polymorphism that had nucleotype-dependent effects on fertility.

Major Comments:

I find the key conclusions convincing, and I feel as if though the authors sufficiently show that there are mitonucleotype x diet interactions on fecundity/fertility traits. Furthermore, their ability to associate the phenotypic variation with a mt:lrRNA polymorphism is also of interest. I do not feel as if though the authors make claims that lack support in the paper. I do not feel as if though additional experiments would be essential to support their claims; however, I would be remiss not to acknowledge that the bulk of the experimental results came down to a 2 x 2 population exploration (Australia-Australia, Australia-Benin, Benin-Australia, Benin-Benin). This study would have been aided by more mitonucleotypes; however, I understand that to generate additional mitonucleotypes would have taken additional time and resources. As most of my comments are minor and should be easily addressed by the authors, I place them in the following section.

Minor Comments:

Below are minor comments and questions I had while reading the manuscript. I apologize in advance if any of my questions are answered in the main text, and I missed them while reviewing.

• Page 3, line 35 - you state that variation in mitochondrial function can contribute to variation in dietary optima. I would add a citation here.

• Page 4, line 74 - If possible, I would add a little more clarifying detail regarding the crossing scheme to the main text. I would just be clear that the generation of these mitonucleotypes takes advantage of the maternal transmission of the mitochondrial genome by crossing virgin females from population 1 to males from population 2, followed by continual backcrossing of virgin females each generation to males from population 2. You go into specific details in the materials & methods section, but I was surprised not to see it earlier. I know this is probably more detail, but I think it's better to be very clear about the crossing scheme used.

• Page 5, line 90 - I would move the citations for the fact that enriching essential amino acids promotes fecundity from the supplement to the main text. You say that this is an established manipulation, and it would be best to point to examples of this manipulation, especially as some of your results find that fertility is negatively affected.

• Page 8, line 129 - In this paragraph, you speak about the impacts of chronic EAA feeding in the AA, BA, and AB mitonucleotypes, but you do not coment on BB? Is there a reason for this? (This connects to one of my comments later)

General Questions - This may be outside the scope of your study, but I was surprised that there was no commentary on co-adaptation (or the potential lack of) between the mitochondrial and nuclear genome in the discussion. It's not immediately clear to me how isolated and for how long the Australian and Benin populations are, but I would expect there to be co-evolution/co-adaptation between the mitochondrial and nuclear genomes. Consequently, I would have expected AA and BB populations to show elevated fitness values; however, it actually seems as if though the AA mitonucleotype performs the worst when given the chronic EAA diet. I wonder if you could comment on the co-evolutionary potential between these two genomes.

Significance

• It is becoming increasingly clear when studying mitochondrial x nuclear interactions that the environmental context of the study can significantly influence the results (see Camus et al. 2019, Rodriguez et al. 2021, and Towarnicki and Ballard 2018). This study furthers our understanding of mitochondrial x nuclear x environment interactions by exploring fully-factorial combinations of mitonucleotypes on two distinct diets (enriched for essential amino acids or plant-based lipids) and evaluating the fertility/fecundity of the different mitonucleotypes. They were able to identify a signle mt:lrRNA SNP that was associated with the measured phenotypes, raising the question of whether and how mitochondria-derived regulatory factors influence phenotypic variation. I also find the comparison to the omnigenic model compelling, particularly their commentary that mitochodnrial genes could contribute to the "core gene" set for several notable phenotypes.

• I believe that this work would be of interest to those obviously studying the evolution and importance of mitochondrial x nuclear x environmental interactions, and I also believe that this work would be of interest to those interested in the effects of various nutrients/diets in both Drosophila and humans.

• My expertise is as a theoretical population and evolutionary geneticist. I am primarily interested in genetic conflict, including between the mitochondrial and nuclear genome. I am interested in how these two genomes evolve to both cause and resolve genetic and sexual conflict.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Reviewer #1

In this article, Amico et al. explore how Spindly self-regulates its interaction with Dynein-Dynactin. They propose that Spindly adopts an auto-inhibited, closed conformation that blocks the CC1 box and Spindly motif, preventing its interaction with dynein-dynactin. The authors used a combination of X-ray crystallography, biochemistry, and structure predictions to detail the intramolecular interactions in Spindly that mediate this closed state. They then use analytical SEC to test their proposed auto-inhibition mechanism by monitoring Spindly binding to the pointed end complex. They suggest that auto-inhibited Spindly is unable to bind Dynein-Dynactin regardless of the presence or absence of Spindly's cargo, the RZZ complex. In contrast, by using mutagenesis to prevent this auto-inhibition, the authors show that uninhibited Spindly can interact with members of the Dynein-Dynactin complex. Finally, they use cellular experiments to show that relieving autoinhibition prevents the proper localization of Spindly and Dynein-Dynactin to kinetochores during mitosis, likely due to the formation of ectopic Spindly-Dynein-Dynactin complexes in these cells.

This is an interesting paper that provides important insights into the mechanism of Spindly regulation and its associations with its interacting partners. However, additional work is necessary to support some of their conclusions. In addition, the text is at times quite dense and harder to follow, which prevents their findings as being impactful as could be possible for the bigger picture paradigms of kinetochore function.

We thank the reviewer for a supportive assessment and for raising some concerns that we have now fully addressed in our revision.

Major Points:

The crosslinking and mass photometry experiments are done at highly differing concentrations (5 μM vs. 10 nM). The mass photometry should be performed at the same concentration as the crosslinking experiments to determine if Spindly forms a higher order oligomer at the higher concentration. These results will aid in the interpretation of the crosslinking mass spectrometry experiments, as the observed interactions could be intermolecular contacts rather than intramolecular contacts if Spindly is tetrameric at these concentrations, as is suggested in figure 4E for specific Spindly constructs.

We thank the reviewer for raising this point. Mass photometry (MP) requires very low sample concentrations as it is essentially a single molecule technique, and therefore the particle density cannot be increased arbitrarily. To assess whether the Spindly construct is prevalently tetrameric at the concentration of the crosslinking experiment, we performed the crosslinking experiments at the standard concentration, and only then diluted the samples and performed MP measurements. The results, displayed in two new panels (Figure 1 – Supplement 1M-N), show that crosslinked samples are primarily dimeric, providing further evidence that we are looking at bona fide intra-dimer contacts.

In figure 2, more conclusive evidence is needed to show that full length Spindly does not form a complex with Dynein-Dynactin. My interpretation of the gels in figure 2D suggests that full length Spindly does form a complex with Dynein-Dynactin, as in the final gel (red outline) it looks as if full length Spindly is indeed peaking with the rest of the Dynein-Dynactin proteins, albeit with excess Spindly eluting later. Figure legends containing protein concentrations used in SEC assays would aid in the interpretation of this data.To conclusively show that full length Spindly doesn't form a complex with Dynein-Dynactin, additional assays will be necessary, such as pull-down assays, or mass photometry.

We have now added concentrations of binding species at the relevant points of the figure legends.

The essence of the reviewer’s concern is that full length Spindly, like BicD2, binds the DD, which would invalidate our model that Spindly is auto-inhibited in absence of a second trigger (other than DD), or alternatively showing that auto-inhibition can be easily overcome. Our conclusion that Spindly remains auto-inhibited, however, is strongly supported by the gels in Figures 2D-G. There, the peak containing DD and BicD2 and eluting around 6.2 ml (panel D) is not visible when BicD2 is replaced with Spindly (panel E), and RZZ does not change this (panels F-G). Note that the peak at 6.6 ml appears to be a contaminant, possibly DNA, and it is visible also with individual Dynein and Dynactin samples. These experiments strongly support our point and we have tried to improve the presentation of the results by boxing relevant fractions of the displayed SDS-PAGEs.

We have now also repeated these experiments with recombinant human Dynactin. The new results are displayed in Figure 2 – Supplement 2. Also in this case, we see minimal complex formation with Spindly and complex formation with BicD2, even if the trailing of Dynein, Dynactin, and Spindly in the earlier elution fractions (already in the absence of complex formation) makes the gels harder to interpret. We also note that these experiments are consistent with those with the isolated PE complex.

Regretfully, we cannot gather additional information by mass photometry because even our positive control dissociates at the extremely low concentrations required to image this very large complex.

In figure 3C, 3E, and figure 5C, there is a shift in the PE peaks in the presence of Spindly, but it isn't clear why doesn't the complex doesn't elute earlier than Spindly alone. If the complex is dissociating on the column, additional assays are necessary to confirm that these Spindly constructs stably interact with PE. If this shift is also accompanied by a major change in shape, thus allowing Spindly to elute later than it does alone, this needs to be explored or explained further.

Elution from a size exclusion chromatography column is dominated by the hydrodynamic radius of the macromolecule. In this particular case, Spindly is highly elongated and essentially sets an upper limit for the elution volume of both the un-complexed and complexed protein. We have described this behavior in many other cases of highly elongated proteins (e.g. Huis in ‘t Veld et al. eLife 2019). We are aware that the binding affinity for the interaction of Spindly and the PE complex is low, and therefore are not surprised to observe dissociation of the complex during the SEC run, i.e. upon dilution of the sample after incubation. In these experiments, we have tried to focus on the shift in elution volume of the PE complex from its elution position in isolation.

The authors should provide better a rationale for why the pointed-end complex is used in figure 3 in lieu of the complex used in figure 2.

We now write that the Spindly motif of adaptors binds the pointed end complex with measurable affinity also in absence of Dynein (near line 294). We then clarify that “As the Spindly motif is predicted to sit within the autoinhibited portion of the protein, we hypothesized that the PE-Spindly motif interaction could be used as a proxy to measure the autoinhibition status of Spindly, bypassing the need to form the entire Dynein-Dynactin-Spindly complex.”

In Figure 5I, WT Spindly also binds to LIC, although less WT Spindly is bound to LIC than Spindly CC2* or Spindly deltaRV. This should be addressed in the text.

Thank you for pointing this out. We have now clarified this in the text near line 440.

The authors claim that the mechanism they describe may be a paradigm for dynein activation by other adaptors at various cellular locations, but they aren't able to identify a mechanism for how Spindly converts from its auto-inhibited state to its permissive state. A more thorough examination of this mechanism is necessary to claim that this mechanism could be paradigmatic, or a revision of the text is needed.

Following an additional concern by reviewer 3, we have now revised the text to meet this concern. So, both in the last sentence of the abstract, and in the last paragraph of the discussion, we do not any longer discuss our results as paradigmatic, although we have reasons to believe that they might be eventually recognized as such, after additional examples will have been analyzed.

Minor Points:

1) The manuscript could benefit from careful review of the text, captions, and figures, as a few minor typos and inconsistencies in the figures and text were present.

We have now re-reviewed the text and figures to try eliminate residual inconsistencies.

2) The list of common structural and functional features of Dynein-Dynactin adaptors could be indicated more clearly.

We have re-written this part of the Introduction, where we now indicate more clearly the features of the DD complex

3) Several times the authors use alpha fold predictions to confirm their data. Although the predictions support several of their conclusions, saying that predictions can confirm the data is an overstatement.

We thank the reviewer for pointing this out. We now replaced “confirmed” with “also supported” on line 190, where we explicitly referred to AF2 predictions as “confirmatory”. We also re-wrote a statement in the Discussion where we had commented on the power of AF2 and indicate that it “became available in the late phases of our work as a guiding and validation tool” (line 524).

4) Figure 1H would be improved by the addition of the amino acid numbers in the domain diagram.

Fixed – we also added amino acid numbers in 1G for consistency.

5) Concentrations used for each protein for the analytical SEC experiments should be listed in the figure or caption.

Thank you for suggesting this. We have now added the protein concentrations for these experiments directly in the legends.

6) In addition to the caption, it would be helpful to the reader to indicate which experiments use farnesylated Spindly.

Done in legends wherever applicable.

7) Error bars are missing from the WT sample in figure 5J. This figure would benefit from statistical analysis.

Done – see also point 4, Reviewer 2.

Significance:

This paper builds on recent work from the Mussachio lab and others exploring the nature of the fibrous corona at kinetochores and the molecular basis for dynein recruitment. This paper is focused on the structural nature of the interactions that underlie Spindly recruitment to kinetochores and its interactions with dynein and other factors. Although reductionist in its approach, this paper has the potential to have broad implications for thinking about the control of corona assembly and dynein recruitment with an elegant auto-regulation of Spindly. Researchers interested in cell division, chromosome segregation, kinetochore function, dynein regulation, and the structural basis for core cellular processes should be interested in this paper.

Reviewer #2

The study by d'Amico et al. presents an in-depth analysis of how intramolecular folding of the coiled-coil adaptor Spindly regulates its interaction with the motor dynein and its obligatory co-factor dynactin. Using biochemical reconstitution and diverse biophysical approaches (including cross-linking mass spectrometry, X-ray crystallography, AF2-based structure prediction, size exclusion chromatography, and analytical ultracentrifugation), the authors uncover and dissect an intricate Spindly autoinhibition mechanism. At kinetochores Spindly is known to co-oligomerize into filaments with the RZZ complex (its kinetochore receptor/cargo), which drives expansion of the outermost kinetochore region (the corona). Here the authors show that Spindly is a dimer in solution and that successive coiled-coil segments interact with each other in an asymmetric 'closed' conformation that is unable to form a complex with dynein and dynactin. Specifically, a 2-residue insertion in the middle of Spindly's first coiled-coil (CC1) creates a kink that allows CC1 to fold back on itself, which has two important structural consequences: it brings a key segment in CC2 (residues 276-309) in contact with a CC1 region called the CC1 box (previously shown to bind dynein light intermediate chain), and it blocks a motif at the beginning of CC2, called the Spindly motif, from accessing the pointed end complex that caps dynactin's minifilament. Mutations in either the CC1 box, the CC1 2-residue insertion, or the CC2(276-309) segment, 'open up' full-length Spindly and promote its interaction with the dynactin pointed end complex and, in case of the latter two types of mutants, with dynein light intermediate chain. CC1 box-deficient Spindly and the CC2 segment mutant (which corresponds to two charge-inverting point mutations) also support complex formation of Spindly and intact dynein-dynactin. Interestingly, while the CC2 mutant can bind to RZZ, the interaction between RZZ and wild-type Spindly is insufficient to make Spindly competent for dynein-dynactin binding (even when RZZ-Spindly are phosphorylated by mitotic kinases). The authors therefore propose that releasing Spindly from autoinhibition requires an additional trigger at the kinetochore, which likely involves an interaction between the Spindly CC2(276-309) segment and an as yet unidentified kinetochore component. The CC2 mutant is also shown to be defective in kinetochore recruitment and in Spindly-RZZ filament formation in vitro, suggesting kinetochore recruitment of Spindly is coupled to kinetochore expansion through a mechanism involving CC2(276-309).

The experiments are of excellent technical quality and the results are presented in a logical and concise manner. There is clarity in the writing (the introduction deserves particular praise), and the authors' conclusions are fully supported by the data. Although there is no direct structural evidence for Spindly's closed conformation, as the authors themselves are careful to point out, the numerous Spindly mutants that are characterized (only some of which are mentioned in the summary above) in aggregate make a convincing case for the proposed autoinhibition mechanism.

We are very grateful to the reviewer for supporting our work

Minor comments:

• Page 5: "605-residue adaptor Spindly". State that "605-residue" refers to the human protein.

We have added this clarification

• Page 8: "The region of Spindly downstream of the Spindly box (residues 281-322) is very conserved among Spindly orthologues, but not among other members of the BICD adaptor family (Figure 1 - Supplement 1L)." This is not very obvious from the alignment shown in the figure.

We agree with the reviewer that the text, as written, was confusing. We have now rephrased it and write “Downstream of the Spindly box, sequences of Spindly orthologues and BICD family adaptors diverge”

• Page 13: "...(A23V-A24V) mutant, which has been previously shown to inhibit the interaction with the LIC2 in a similar assay (Gama et al., 2017)." The LIC isoform used in the referenced study was LIC1.

Thank you for identifying this error. We have corrected the text accordingly.

-Figure 5J: Information about statistical significance should be added.

Done. See also Minor point 7, Reviewer 1.

-Figure 7B - D: Red on black is not an ideal color choice for these graphs.

We now replaced red with yellow

-Page 15: When discussing the recently discovered interphase functions of Spindly, also cite Clemente et al. (2018; doi:10.3390/jdb6020009) and Conte et al. (2018; doi:10.1242/bio.033233).

We apologize for the involuntary omission of these two references, which have now been included in the revised manuscript.

-Page 17: "Evidence supporting this idea is that mutations in the 276-306 region, including the deletion of this entire fragment or the introduction of charge-inverting point mutations at residues 295 and 297 respectively abolish or largely decrease the kinetochore recruitment of Spindly ((Raisch et al., 2021) and this study),...". Sacristan et al. (2018) should also be cited in this context, as this study established the importance of residues 274-287 for Spindly recruitment to kinetochores.

We agree and apologize for the inadvertent omission. We have now included the Sacristan et al. reference in this context.

• Page 17: "In vitro, the 276-306 region is also required for the assembly of RZZ-Spindly filaments (this study and (Raisch et al., 2021))." It could also be mentioned here that residues 274-287 of Spindly are necessary for RZZ-Spindly filament formation in cells, as shown by Sacristan et al. (2018).

We have now reported this fact on lines 560-561.

• Page 17: "Plausibly, the solution to this conundrum will require biochemical reconstitutions addressing the spectrum of interactions that this protein establishes at the kinetochore." Presumably, "this protein" refers to Spindly, but this is not clear since the subject of the preceding sentence is RZZ.

Done – line 565

Significance

Cargo transport by cytoskeletal motors must be tightly regulated to establish and maintain intracellular organization and for faithful execution of development, including cell division. Much of this regulation occurs at the motor-cargo interface but remains poorly understood at the molecular level. In recent years it has become clear that adaptor proteins not only provide a physical link between motors and their cargo but also participate in motor activation. Adaptor-coupled activation is particularly important for dynein, because adaptors promote dynein's interaction with its essential co-factor dynactin.

BICD2 (along with other Bicaudal D proteins) is the most intensely studied dynein adaptor and has long been known to be subject to autoinhibition with regard to dynein-dynactin binding, which is relieved by cargo binding to the BICD2 C-terminal region. A important question has been whether the same regulatory logic applies to other dynein adaptors. The study by d'Amico et al. presents the first evidence that conformational inhibition extends to adaptors other than Bicaudal D proteins. The study also reveals that Spindly's autoinhibition mechanism is more complex than that of BICD2. This likely reflects Spindly's dual function in dynein-dynactin recruitment and kinetochore expansion. The results of d'Amico et al. suggest that the Spindly autoinhibition mechanism has evolved to coordinate the two processes, and this idea is further supported by a recent study on the RZZ-Spindly interaction from the same group (Raisch et al. 2021; doi:10.1101/2021.12.03.471119). One of the most important insights from d'Amico et al. is that there must be another binding partner of Spindly at kinetochores besides the RZZ complex that participates in the relief of Spindly autoinhibition. The study has therefore identified an important future research direction. It will be interesting to investigate whether additional adaptors follow the multi-step activation model proposed here for Spindly.

Regarding the technical aspects, the study illustrates that AF2-based structure prediction is a powerful tool for investigating conformational regulation, and it introduces an important innovation: the ability to generate recombinant human dynactin opens the door to the engineering of dynactin mutants, which promises to accelerate mechanistic dissection of this essential dynein co-factor. In conclusion, the study represents a significant step forward in our understanding of how dynein-cargo interactions are regulated by adaptor proteins and is therefore of general interest for researchers studying the molecular mechanisms of chromosome segregation as well as intracellular transport.

Reviewer #3

The Dynein-Dynactin (DD) complex interacts with different activating adaptors to assemble functional motor complexes capable of moving along microtubules while transporting various cargoes. However, it remains poorly understood how DD activation is precisely controlled so that Dynein-mediated transport is only stimulated at the appropriate time and place. DD adaptor regulation is likely a crucial piece of this puzzle. In this manuscript, the authors show that Spindly, a mitotic adaptor of DD complex, undergoes a series of conformational rearrangements that result in efficient Spindly autoinhibition and affect its ability to bind DD. The work from d'Amico et al includes an impressive amount of biochemical and biophysical data, supported by well-designed experiments that are carefully documented. Resorting to crosslinking experiments and protein structural modelling, the authors find that several intramolecular contacts occur between specialized domains within Spindly N-terminus. The resulting compact conformation occludes important DD-binding motifs in Spindly and, thus, limits the access of DD to the adaptor. By utilizing different Spindly mutants predicted to render the adaptor more elongated, the authors bypass Spindly autoinhibition and rescue binding to DD in vitro. Surprisingly, unlike other DD adaptors, Spindly autoinhibition is not relieve upon binding to its cargo (the RZZ complex) arguing that the interaction with an additional binding partner is require to fully unleash the potential of Spindly to bind DD. In line with this, the authors identify a Spindly mutant that is unable to localize to kinetochores from human cells, despite its open conformation. Collectively, this work provides significant advances in the understanding of Spindly regulation and brings a new perspective to the mechanism of DD adaptor activation and therefore should be of interest for a wide audience.

We are very grateful to the reviewer for the support and for the thorough and constructive evaluation of our work.

Major concerns:

• The authors show that Spindly 33-605 is able to form a complex with DD which eventually enables the recruitment of Dynactin to kinetochores from Spindly 33-605-expressing cells. This result is unexpected since this Spindly mutant lacks CC1 box, which has been previously shown to be required for the kinetochore localization of Dynactin (Sacristan et al 2018). A more comprehensive discussion about this discrepancy would enrich the article and benefit the audience.

We thank the reviewer for pointing this out. We now write (line 492): “This result was unexpected, because the CC1 box has been previously shown to be required for kinetochore localization of Dynactin (Sacristan et al., 2018)”

• In Fig.7, the authors show that two Spindly mutants (Spindly CC2* and Spindly chimera) are unable to fully decorate the kinetochores from human cells. The same is true for Spindly AA/VV mutant. Do the authors know whether these mutants are expressed as stable proteins in cells for example by performing a western blot analysis?

In this revised version of our manuscript, we have explained more clearly that in this experiment we electroporate recombinant proteins. These are essentially the same proteins that we use for the experiments in vitro. This provides an internal test in these experiments, because we can verify, through their successful expression and purification, that the proteins are stable. We cannot exclude, however, that the proteins are “treated differently” in cells, for instance because they interact differently with certain binding partners in ways that modifies their stability. As the proteins are not expressed continuously, but rather introduced in the cells in a single electroporation event several hours before imaging, the overall levels of these proteins may differ. We have now included a representative western blot (Figure 7 – Supplement 1C) that demonstrates the levels of electroporated proteins in the experiments in Figure 7. SpindlyCC2*appears to be present at somewhat lower levels than the other constructs. mChSpindly33-605 and Spindlychimera, on the other hand, were present at very similar levels, supporting our conclusion that a kinetochore-binding region is impaired in the latter. We now refer in the main text to the uncertainty created by the comparatively lower cellular levels of SpindlyCC2*. We have also chosen more representative kinetochores for the insets of CC2* and Chimera in Figure 7A.

• In line with the previous point, could the authors tether each Spindly mutant to the kinetochore for example by fusing the construct to known kinetochores proteinssuch as Mis12 and test whether these fusion constructs are now able to recruit Dynactin to kinetochores?

This would be a potentially interesting experiment. However, reasoning that Spindly is a strong dimer that needs to interact with another strong hexamer like the RZZ complex, discouraged us as these stoichiometries would almost certainly complicate the interpretation of these experiments. It is clear that further work will be required to define the complete picture for this complex system.

• The authors conclude that the 2-step or multistep mechanism involved in the regulation of Spindly activation may be a common mechanism to different DD adaptors. However, the authors point out to existing differences between the conformational arrangement of Spindly and another DD adaptor, BICD2, arguing against a common mode of regulation for all adaptors. This needs to be clarified.

The reviewer has a good point and we have indeed tuned this down. We have re-written the last sentence of the abstract and replaced it with “Thus, our work illustrates how Dynein can be specifically activated at a defined cellular locale.” We also write (line 592): Whether a similar 2-step or multistep mechanism applies to additional cargo-adaptor systems is an important question for future studies.

Minor concerns:

• In Fig.2D, full length Spindly does not bind DD in vitro. This is most likely to occur because Spindly N-terminus adopts a compacted conformation and hinders the access to DD-binding motifs. In Fig.2B, the authors show a structural prediction for Spindly 1-275 which should adopt a more elongated shape. According to prevailing model, this construct should now be able to bind DD in a similar biochemical assay.

We agree with the reviewer that Spindly1-275 (and Spindly∆276-306) might be expected to be strong DD binders based on our model. Indeed, these proteins bind to the PE, albeit apparently weakly. Nevertheless, as explained in lines 350 and following, these mutants appear to form higher oligomers and we have not been able to show convincingly that they are fully open and available to bind DD.

• In Gama et al 2017, LIC1 was able to pull down a wild-type N-terminal Spindly construct. How do the authors reconcile this with the data presented in this manuecript?

We have expanded the discussion, also to answer major point 5, reviewer 1, on line 446 and following, where we also refer to the observation of Gama et al. 2017.

• The section where the authors test point mutations to open Spindly ("Opening up Spindly with point mutations") should be better contextualized. The transition is difficult to follow as it is.

We have now rephrased this part of the text to make are thoughts clearer.

• In the text, it is not clear whether Mps1 kinase is required to promote RZZ oligomerization in the presence of Spindly chimera, an uninhibited Spindly mutant. According to the model, this mutant construct should drive oligomerization independently of Mps1 (as the N-terminal deletion construct from Sacristan et al 2018).

The reviewer is correct and we have rephrased this part of the text to clarify

• The nomenclature the authors adopt for the CC1 second conserved motif (SCM) and for the Spindly motif (SM) can be confusing at some point when identifying each mutant in the text and figures. Nomenclature should be standardized.

We agree with the reviewer and have now adopted a different nomenclature for the CC2 box or second conserved motif, namely HBS1, for Heavy Chain Binding Site 1. This functional annotation derives from work of one of our laboratories (Carter) and has been discussed with, and approved by, Geert Kops, whose laboratory had originally proposed the name “CC2 box”, as well as Reto Gassmann, Erika Holzbaur, Roberto Dominguez, Sam Reck Peterson, Rick McKenney, and Ahmet Yildiz.

• In Fig.6A, mCh-Spindly 33-605 and mCh-Spindly chimera lines have the same color.

Thank you for spotting this subtle mistake. We have corrected the color line.

Significance

The work represents a significant advance in the field and it would be of interest for a wide range of audiences.

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

The Dynein-Dynactin (DD) complex interacts with different activating adaptors to assemble functional motor complexes capable of moving along microtubules while transporting various cargoes. However, it remains poorly understood how DD activation is precisely controlled so that Dynein-mediated transport is only stimulated at the appropriate time and place. DD adaptor regulation is likely a crucial piece of this puzzle. In this manuscript, the authors show that Spindly, a mitotic adaptor of DD complex, undergoes a series of conformational rearrangements that result in efficient Spindly autoinhibition and affect its ability to bind DD. The work from d'Amico et al includes an impressive amount of biochemical and biophysical data, supported by well-designed experiments that are carefully documented. Resorting to crosslinking experiments and protein structural modelling, the authors find that several intramolecular contacts occur between specialized domains within Spindly N-terminus. The resulting compact conformation occludes important DD-binding motifs in Spindly and, thus, limits the access of DD to the adaptor. By utilizing different Spindly mutants predicted to render the adaptor more elongated, the authors bypass Spindly autoinhibition and rescue binding to DD in vitro. Surprisingly, unlike other DD adaptors, Spindly autoinhibition is not relieve upon binding to its cargo (the RZZ complex) arguing that the interaction with an additional binding partner is require to fully unleash the potential of Spindly to bind DD. In line with this, the authors identify a Spindly mutant that is unable to localize to kinetochores from human cells, despite its open conformation. Collectively, this work provides significant advances in the understanding of Spindly regulation and brings a new perspective to the mechanism of DD adaptor activation and therefore should be of interest for a wide audience.

Major concerns:

• The authors show that Spindly 33-605 is able to form a complex with DD which eventually enables the recruitment of Dynactin to kinetochores from Spindly 33-605-expressing cells. This result is unexpected since this Spindly mutant lacks CC1 box, which has been previously shown to be required for the kinetochore localization of Dynactin (Sacristan et al 2018). A more comprehensive discussion about this discrepancy would enrich the article and benefit the audience.
• In Fig.7, the authors show that two Spindly mutants (Spindly CC2* and Spindly chimera) are unable to fully decorate the kinetochores from human cells. The same is true for Spindly AA/VV mutant. Do the authors know whether these mutants are expressed as stable proteins in cells for example by performing a western blot analysis?
• In line with the previous point, could the authors tether each Spindly mutant to the kinetochore for example by fusing the construct to known kinetochores proteinssuch as Mis12 and test whether these fusion constructs are now able to recruit Dynactin to kinetochores?
• The authors conclude that the 2-step or multistep mechanism involved in the regulation of Spindly activation may be a common mechanism to different DD adaptors. However, the authors point out to existing differences between the conformational arrangement of Spindly and another DD adaptor, BICD2, arguing against a common mode of regulation for all adaptors. This needs to be clarified.

Minor concerns:

• In Fig.2D, full length Spindly does not bind DD in vitro. This is most likely to occur because Spindly N-terminus adopts a compacted conformation and hinders the access to DD-binding motifs. In Fig.2B, the authors show a structural prediction for Spindly 1-275 which should adopt a more elongated shape. According to prevailing model, this construct should now be able to bind DD in a similar biochemical assay.
• In Gama et al 2017, LIC1 was able to pull down a wild-type N-terminal Spindly construct. How do the authors reconcile this with the data presented in this manuecript?
• The section where the authors test point mutations to open Spindly ("Opening up Spindly with point mutations") should be better contextualized. The transition is difficult to follow as it is.
• In the text, it is not clear whether Mps1 kinase is required to promote RZZ oligomerization in the presence of Spindly chimera, an uninhibited Spindly mutant. According to the model, this mutant construct should drive oligomerization independently of Mps1 (as the N-terminal deletion construct from Sacristan et al 2018).
• The nomenclature the authors adopt for the CC1 second conserved motif (SCM) and for the Spindly motif (SM) can be confusing at some point when identifying each mutant in the text and figures. Nomenclature should be standardized.
• In Fig.6A, mCh-Spindly 33-605 and mCh-Spindly chimera lines have the same color.

Significance

The work represente a significant advance in the field and it would be of interest for a wide range of audiences.

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 d'Amico et al. presents an in-depth analysis of how intramolecular folding of the coiled-coil adaptor Spindly regulates its interaction with the motor dynein and its obligatory co-factor dynactin. Using biochemical reconstitution and diverse biophysical approaches (including cross-linking mass spectrometry, X-ray crystallography, AF2-based structure prediction, size exclusion chromatography, and analytical ultracentrifugation), the authors uncover and dissect an intricate Spindly autoinhibition mechanism. At kinetochores Spindly is known to co-oligomerize into filaments with the RZZ complex (its kinetochore receptor/cargo), which drives expansion of the outermost kinetochore region (the corona). Here the authors show that Spindly is a dimer in solution and that successive coiled-coil segments interact with each other in an asymmetric 'closed' conformation that is unable to form a complex with dynein and dynactin. Specifically, a 2-residue insertion in the middle of Spindly's first coiled-coil (CC1) creates a kink that allows CC1 to fold back on itself, which has two important structural consequences: it brings a key segment in CC2 (residues 276-309) in contact with a CC1 region called the CC1 box (previously shown to bind dynein light intermediate chain), and it blocks a motif at the beginning of CC2, called the Spindly motif, from accessing the pointed end complex that caps dynactin's minifilament. Mutations in either the CC1 box, the CC1 2-residue insertion, or the CC2(276-309) segment, 'open up' full-length Spindly and promote its interaction with the dynactin pointed end complex and, in case of the latter two types of mutants, with dynein light intermediate chain. CC1 box-deficient Spindly and the CC2 segment mutant (which corresponds to two charge-inverting point mutations) also support complex formation of Spindly and intact dynein-dynactin. Interestingly, while the CC2 mutant can bind to RZZ, the interaction between RZZ and wild-type Spindly is insufficient to make Spindly competent for dynein-dynactin binding (even when RZZ-Spindly are phosphorylated by mitotic kinases). The authors therefore propose that releasing Spindly from autoinhibition requires an additional trigger at the kinetochore, which likely involves an interaction between the Spindly CC2(276-309) segment and an as yet unidentified kinetochore component. The CC2 mutant is also shown to be defective in kinetochore recruitment and in Spindly-RZZ filament formation in vitro, suggesting kinetochore recruitment of Spindly is coupled to kinetochore expansion through a mechanism involving CC2(276-309).

The experiments are of excellent technical quality and the results are presented in a logical and concise manner. There is clarity in the writing (the introduction deserves particular praise), and the authors' conclusions are fully supported by the data. Although there is no direct structural evidence for Spindly's closed conformation, as the authors themselves are careful to point out, the numerous Spindly mutants that are characterized (only some of which are mentioned in the summary above) in aggregate make a convincing case for the proposed autoinhibition mechanism.

Minor comments:

• Page 5: "605-residue adaptor Spindly". State that "605-residue" refers to the human protein.
• Page 88: "The region of Spindly downstream of the Spindly box (residues 281-322) is very conserved among Spindly orthologues, but not among other members of the BICD adaptor family (Figure 1 - Supplement 1L)." This is not very obvious from the alignment shown in the figure.
• Page 13: "...(A23V-A24V) mutant, which has been previously shown to inhibit the interaction with the LIC2 in a similar assay (Gama et al., 2017)." The LIC isoform used in the referenced study was LIC1.
• Figure 5J: Information about statistical significance should be added.
• Figure 7B - D: Red on black is not an ideal color choice for these graphs.
• Page 15: When discussing the recently discovered interphase functions of Spindly, also cite Clemente et al. (2018; doi:10.3390/jdb6020009) and Conte et al. (2018; doi:10.1242/bio.033233).
• Page 17: "Evidence supporting this idea is that mutations in the 276-306 region, including the deletion of this entire fragment or the introduction of charge-inverting point mutations at residues 295 and 297 respectively abolish or largely decrease the kinetochore recruitment of Spindly ((Raisch et al., 2021) and this study),...". Sacristan et al. (2018) should also be cited in this context, as this study established the importance of residues 274-287 for Spindly recruitment to kinetochores.
• Page 17: "In vitro, the 276-306 region is also required for the assembly of RZZ-Spindly filaments (this study and (Raisch et al., 2021))." It could also be mentioned here that residues 274-287 of Spindly are necessary for RZZ-Spindly filament formation in cells, as shown by Sacristan et al. (2018).
• Page 17: "Plausibly, the solution to this conundrum will require biochemical reconstitutions addressing the spectrum of interactions that this protein establishes at the kinetochore." Presumably, "this protein" refers to Spindly, but this is not clear since the subject of the preceding sentence is RZZ.

Significance

Cargo transport by cytoskeletal motors must be tightly regulated to establish and maintain intracellular organization and for faithful execution of development, including cell division. Much of this regulation occurs at the motor-cargo interface but remains poorly understood at the molecular level. In recent years it has become clear that adaptor proteins not only provide a physical link between motors and their cargo but also participate in motor activation. Adaptor-coupled activation is particularly important for dynein, because adaptors promote dynein's interaction with its essential co-factor dynactin.

BICD2 (along with other Bicaudal D proteins) is the most intensely studied dynein adaptor and has long been known to be subject to autoinhibition with regard to dynein-dynactin binding, which is relieved by cargo binding to the BICD2 C-terminal region. A important question has been whether the same regulatory logic applies to other dynein adaptors. The study by d'Amico et al. presents the first evidence that conformational inhibition extends to adaptors other than Bicaudal D proteins. The study also reveals that Spindly's autoinhibition mechanism is more complex than that of BICD2. This likely reflects Spindly's dual function in dynein-dynactin recruitment and kinetochore expansion. The results of d'Amico et al. suggest that the Spindly autoinhibition mechanism has evolved to coordinate the two processes, and this idea is further supported by a recent study on the RZZ-Spindly interaction from the same group (Raisch et al. 2021; doi:10.1101/2021.12.03.471119). One of the most important insights from d'Amico et al. is that there must be another binding partner of Spindly at kinetochores besides the RZZ complex that participates in the relief of Spindly autoinhibition. The study has therefore identified an important future research direction. It will be interesting to investigate whether additional adaptors follow the multi-step activation model proposed here for Spindly.

Regarding the technical aspects, the study illustrates that AF2-based structure prediction is a powerful tool for investigating conformational regulation, and it introduces an important innovation: the ability to generate recombinant human dynactin opens the door to the engineering of dynactin mutants, which promises to accelerate mechanistic dissection of this essential dynein co-factor.

In conclusion, the study represents a significant step forward in our understanding of how dynein-cargo interactions are regulated by adaptor proteins and is therefore of general interest for researchers studying the molecular mechanisms of chromosome segregation as well as intracellular transport.

Reviewer expertise keywords: same as the keywords of this 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 #1

Evidence, reproducibility and clarity

In this article, Amico et al. explore how Spindly self-regulates its interaction with Dynein-Dynactin. They propose that Spindly adopts an auto-inhibited, closed conformation that blocks the CC1 box and Spindly motif, preventing its interaction with dynein-dynactin. The authors used a combination of X-ray crystallography, biochemistry, and structure predictions to detail the intramolecular interactions in Spindly that mediate this closed state. They then use analytical SEC to test their proposed auto-inhibition mechanism by monitoring Spindly binding to the pointed end complex. They suggest that auto-inhibited Spindly is unable to bind Dynein-Dynactin regardless of the presence or absence of Spindly's cargo, the RZZ complex. In contrast, by using mutagenesis to prevent this auto-inhibition, the authors show that uninhibited Spindly can interact with members of the Dynein-Dynactin complex. Finally, they use cellular experiments to show that relieving autoinhibition prevents the proper localization of Spindly and Dynein-Dynactin to kinetochores during mitosis, likely due to the formation of ectopic Spindly-Dynein-Dynactin complexes in these cells.

This is an interesting paper that provides important insights into the mechanism of Spindly regulation and its associations with its interacting partners. However, additional work is necessary to support some of their conclusions. In addition, the text is at times quite dense and harder to follow, which prevents their findings as being impactful as could be possible for the bigger picture paradigms of kinetochore function.

Major Points:

1. The crosslinking and mass photometry experiments are done at highly differing concentrations (5 μM vs. 10 nM). The mass photometry should be performed at the same concentration as the crosslinking experiments to determine if Spindly forms a higher order oligomer at the higher concentration. These results will aid in the interpretation of the crosslinking mass spectrometry experiments, as the observed interactions could be intermolecular contacts rather than intramolecular contacts if Spindly is tetrameric at these concentrations, as is suggested in figure 4E for specific Spindly constructs.
2. In figure 2, more conclusive evidence is needed to show that full length Spindly does not form a complex with Dynein-Dynactin. My interpretation of the gels in figure 2D suggests that full length Spindly does form a complex with Dynein-Dynactin, as in the final gel (red outline) it looks as if full length Spindly is indeed peaking with the rest of the Dynein-Dynactin proteins, albeit with excess Spindly eluting later. Figure legends containing protein concentrations used in SEC assays would aid in the interpretation of this data. To conclusively show that full length Spindly doesn't form a complex with Dynein-Dynactin, additional assays will be necessary, such as pull-down assays, or mass photometry.
3. In figure 3C, 3E, and figure 5C, there is a shift in the PE peaks in the presence of Spindly, but it isn't clear why doesn't the complex doesn't elute earlier than Spindly alone. If the complex is dissociating on the column, additional assays are necessary to confirm that these Spindly constructs stably interact with PE. If this shift is also accompanied by a major change in shape, thus allowing Spindly to elute later than it does alone, this needs to be explored or explained further.
4. The authors should provide better a rationale for why the pointed-end complex is used in figure 3 in lieu of the complex used in figure 2.
5. In Figure 5I, WT Spindly also binds to LIC, although less WT Spindly is bound to LIC than Spindly CC2* or Spindly deltaRV. This should be addressed in the text.
6. The authors claim that the mechanism they describe may be a paradigm for dynein activation by other adaptors at various cellular locations, but they aren't able to identify a mechanism for how Spindly converts from its auto-inhibited state to its permissive state. A more thorough examination of this mechanism is necessary to claim that this mechanism could be paradigmatic, or a revision of the text is needed.

Minor Points:

1. The manuscript could benefit from careful review of the text, captions, and figures, as a few minor typos and inconsistencies in the figures and text were present.
2. The list of common structural and functional features of Dynein-Dynactin adaptors could be indicated more clearly.
3. Several times the authors use alpha fold predictions to confirm their data. Although the predictions support several of their conclusions, saying that predictions can confirm the data is an overstatement.
4. Figure 1H would be improved by the addition of the amino acid numbers in the domain diagram.
5. Concentrations used for each protein for the analytical SEC experiments should be listed in the figure or caption.
6. In addition to the caption, it would be helpful to the reader to indicate which experiments use farnesylated Spindly.
7. Error bars are missing from the WT sample in figure 5J. This figure would benefit from statistical analysis.

Significance

This paper builds on recent work from the Mussachio lab and others exploring the nature of the fibrous corona at kinetochores and the molecular basis for dynein recruitment. This paper is focused on the structural nature of the interactions that underlie Spindly recruitment to kinetochores and its interactions with dynein and other factors. Although reductionist in its approach, this paper has the potential to have broad implications for thinking about the control of corona assembly and dynein recruitment with an elegant auto-regulation of Spindly. Researchers interested in cell division, chromosome segregation, kinetochore function, dynein regulation, and the structural basis for core cellular processes should be interested in this paper.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Reviewer #1 (Evidence, reproducibility and clarity (Required)): **Summary:** Techniques to probe the local environment of membrane proteins are sparse, although the influence of lipids on the membrane protein's function are known since many years. Therefore, the paper by Umebayashi et al. is important. The environment-sensitive dye Nile red (NR) coupled to a membrane protein is an appropriate sensor for monitoring the local membrane fluidity. Linking of Nile red to the receptor via a flexible tether was achieved with the acyl carrier protein (ACP)-tag method. Experiments showed that depending on the ACP site a certain linker length is required to have NR inserted in the membrane and thus be an effective sensor for lipid disorder. This technology could be of general usability to study the environment of membrane proteins in the context of their function. As an example, the technique allowed insulin induced membrane disorder in the close insulin receptor vicinity to be observed. Further, results suggested that tyrosine activity is required for this disorder to happen. The experimental results appear to be complete and controls were made.

**Major comments:** 1) Sometimes technical terms are used without explanation: What is the GP value? What is ACP-IR? The spectrum was measured in number of rois? The reader can find those abbreveations out, but it would be nice to have them defined.

We have made a list of abbreviations.

2) Fig. 1d) is confusing. The ACP-IR labelling is evident in 3 panels, but there is no difference in the color (emission spectra of 1992-ACP-IR vs 2031-ACP-IR should be visible??). The DAPI staining is very different. When doing the latter, how difficult is it to get the staining equal?

The differences in spectra cannot be seen because we used pseudo colors for display of the DAPI and CoA-PEG-NR staining. The reviewer’s comments about the unequal DAPI staining is correct. The reason for this is most likely that the cell membrane is unequally permeabilized by PFA treatment. As the point of this figure is just to show that the plasma membrane is labeled, dependent upon the expression of the ACP-tagged insulin receptor, we don’t think that the variable intensities of the DAPI staining is important. DAPI is simply used to indicate the position of the cells.

3) How can one interpret Fig. 4: a) Control goes over 4 frames, at 240" insulin is added, and 10 frames should show a fluctuation difference?

We showed 4 frames after control treatment that showed no significant change was observed by control treatment. We expected that clear changes would be invoked by insulin treatment in GP images, however these changes, while visible in the GP images, are difficult to see for the untrained observer. This is the reason why we used the ZNCC method in the subsequent figures to better visualize the changes.

1. b) A color shift from blue to green is visible after insulin addition. But it is faint - difficult to assess from the pseudo color scheme. What does 1000 pixel top/1000 pixel bottom mean in c). Is it an attempt to better visualize the fluctuation? It is difficult to recognize a difference before and after adding insulin. d) It seems that the kymograph set should show this. What is the color scale? Why is 3 so untypical, i.e., no change? Box 6 is also peculiar: the left side does not show a strong change upon insulin administration, the right side does. Why? We appreciate the helpful comments for improving our manuscript.

As pointed out, the change of GP value is extremely small before and after insulin addition, so it is difficult to fully visualize the change with normal pseudo-color expression. To deal with this, we adopted the following two methods to visualize minute changes.

1) Visualization of local changes of the statistical GP value showed by ZNCC throughout the time-lapse images (Fig. 6 and Fig. S2B).

2) Visualization of the top/bottom 1000 pixels of the sorting ZNCC value in each image (Fig. 7 and Fig. S2C). The top 1000 pixels are the ones that showed the largest changes. The bottom 1000 pixels are the ones that showed the smallest changes.

Owing to these expressions, we found out that the level of the response against the insulin signal was spatially and temporally heterogeneous in the membrane.

As for the color scale, in order to clarify the meaning of the difference of color, we have added the description about the relationship between the color and the ZNCC value in the results section.

4) How is the kymogram calculated? The legend says 'The horizontal dimension represents the averaged ZNCC inside the rectangular area, and the vertical dimension represents time'. The averaged ZNCC is a single value, so it is not clear why the kymogram shows a variation from left to right. May it be the ZNCC was averaged just vertically?

We apologize that we did not provide information regarding making the kymograph.

In the yellow rectangular area (Fig. 6B), the ZNCC values of the pixels with the same x coordinate value were vertically averaged, which were represented as the horizontal direction of the kymograph. That is, one horizontal line of the kymograph holds the spatial distribution of the ZNCC value along the horizontal direction of the membrane, and the vertical direction shows their time changes. To make it easier to understand, we refined the description about the kymograph in the legend of Fig. 6.

5) When calculating cross-correlation values on images, they need to be aligned. What fraction of the total image does the selected 19x19 box represent? As described, I imagine that a rolling CC over 19x19 pixels is calculated over an image from the time lapse series comparing it with the reference Iave(x,y). Compared to the 3x3 median filtered CP image, the ZNCC image should then be much more blurred??

Below we provide more information regarding the calculation of ZNCC.

Each local window for ZNCC calculation is set to a 19x19 pixels centered on every single pixel excluding the edges of an image. The ZNCC value calculated in that window is set to a center pixel of that area. After that, a new window centered on the adjacent pixel is set and calculate the new ZNCC. That is, the calculation window is slid throughout the image. Also, the calculated ZNCC value is not set to all the pixels of the window, but is set to only the center pixel of the window, so there is no blur effect like median filtering.

The figure below shows a schematic view of our ZNCC calculation.

Schematic view of our ZNCC calculation

**Minor comment:** On page 16 supplementary is not spelled properly.

corrected

Reviewer #1 (Significance (Required)):

The key point of this paper is convincing and the new technology appears to have a lot of potential. It can be applied to study membrane protein function in the context of its environment, the lipid bilayer.

Membrane fluidity measurements have been developed (e.g., using fluorescent probes like laurdan). However, the trick to link a probe like nile red by ACP technology to the insulin receptor and to observe its activity is quite new.

A most recent description of such a technology is in TrAC Trends in Analytical Chemistry Volume 133, December 2020, 116092.

This is an interesting review, but not directly impacting on our work.

**Referees cross-commenting**

All comments are constructive and important. The paper is important but needs to be amended as proposed.

Reviewer #2 (Evidence, reproducibility and clarity (Required)): **Summary:** In this manuscript, authors generated an ACP-attached Nile Red probe in order to specifically label Insulin receptor in the membrane. Owing to this specificity, one can measure the lipid membrane properties around a specific protein in the membrane. **Major comments:**

For the conclusions in the manuscript to be convincing, in my opinion, these additional data need to be added. Some of these are new experiments, and some are detailed analysis of existing data. The new experiments are not for new line of investigation, instead it is to confirm their statements and conclusions. The major point is the reliability of spectral shift. In usual environment sensitive probes, it is certain that they are in the membrane whatever is done to the membrane. However, when the probe is attached to a protein, it is not trivial to have the same confidence that the probe is always inside the membrane, and it is in the same plane of the membrane. 1992-ACP-IR is a good example; authors state that it binds to the protein outside the membrane, but when there is cholesterol addition and -maybe more interestingly- cholesterol removal, the dye still reacts and changes its emission (even PreCT changes its emission quite a bit at the 570 nm region). This is a clear indication of a change in localization of the probe upon some changes in the membrane. This implies that observed spectral shifts may not be due to lipid packing differences, but due to localization of the probes. For this reason, it is crucial to know where any environment sensitive probe localize in the membrane with respect to membrane normal, and this knowledge is more important for this probe. Related to this, the spectral difference upon insulin treatment and activation of insulin receptor could be due to changes in probe's localization in the membrane. Especially because authors show in Fig1e, the spectra can change depending on the probe localization. Relatedly, quantum yield of NR should be significantly different when it is inside vs outside membrane. Authors should show QY for 1992-ACP-NR and 2031-ACP-NR with different PEG lengths and upon insulin treatment.

We understand the logic of the request to measure the QY, since the QY of Nile red is much higher in organic solvents than in aqueous solutions, so it might be predicted that the QY of Nile red is higher in a lipid bilayer than when covalently bound to the protein in an aqueous environment. However, this argument depends upon the mechanism for the increase in quantum yield when going from aqueous to a non-polar solution. One possible explanation is based on the intrinsic properties of the dye under the two conditions. The alternative explanation would be that the dye would aggregate (be insoluble) in aqueous solution and therefore either not fluoresce or self-quench. In this case, we believe that the latter is the explanation because we and others have previously shown the turn-on properties of the probe when binding to proteins (SNAP-tag and others). It is not simple to measure QY in the cell under a microscope, but we have done something similar shown in supplementary figure 4. We labeled the three ACP-receptor complexes with PEG11-Nile red and co-stained with antibody to the Insulin Receptor. We then calculated a relative quantum yield. There were very little differences at all between the relative quantum yields, so we conclude that it is not the environment of the probe, which affects the quantum yield under these conditions, but the fact that it is covalently attached to a protein and incapable of forming aggregates. What distinguishes these constructs is the emission spectrum, not the quantum yield. In supplementary Table 2 we also did QY measurements in vitro and we could reproduce the increase of quantum yield by association with liposomes or in organic solvents. We tested whether non-covalent association with a protein would increase the QY by incubation with the lipid binding protein, BSA, in PBS. This was not the case, strongly pointing to the conclusion that it is the covalent association with the protein that increases the QY, not association with a protein. We believe that our demonstration of changes in fluorescent spectra with changes in cholesterol, large changes in fluorescent spectra with linker length for the 1992 construct and voltage sensitivity using patch-clamp prove that the Nile red is reporting on the membrane environment under the conditions we propose.

**Minor comments:** - Fig 1d requires quantification We do not agree on this. This is simply to show that the labeling is dependent upon expression of the relevant ACP-IR constructs. There is no detectable labeling of the control.

• Voltage sensitivity of different PEG length of 2031-ACP probe should be added. We have added this data in figure 2 panel E.

• Fig 3a graph should show all data points, not only bar graphs. Also, the band in 3a for +CoA-PEG-NR is dimmer than other bands, is it specific to this particular gel since quantification does not show any difference?

There is no significant difference- Fig 4d, colour code is needed.

Done

• Fig 5b and Fig3d are basically the same experiments in terms of control measurement, why is the difference in 3b is 0.04 GP unit while it is 0.007 GP unit?

We explain in the MS, but have improved the title of Y-axis in Fig.5 b graph so that the difference in what is plotted is clear. - Why is inhibitor data so noisy? We should discuss.

We don’t know the exact reason why inhibitor data is noisy, but we speculate that the actin cytoskeleton and phosphoinositide-dependent signaling could affect the membrane stability, and the membrane environment would be fluctuated in the presence of latrunculin B or PI3K inhibitor.

Reviewer #2 (Significance (Required)): Overall, this is a very useful approach, and this line of research will yield very useful tools to shed light on how lipids surrounding proteins can change their function. Major advance of the paper is the new chemical biology tool. There is also biological data on how insulin can change the insulin receptor's membrane environment which is contradictory to some old literature claiming that InsR becomes more "rafty" upon insulin treatment (e.g., PMID: 11751579).

If this type of tagging proves robust and reproducible (limitations and concerns listed above and below), it could be used by other researchers to tag their protein of interest and investigate the lipid environment around those proteins.

The downside of this method is that the probe requires ACP tag, a relatively less used tag than others in biology, therefore researchers interested in using this probe should have their proteins with ACP tag. Moreover, the linker length and ACP-tag position are quite crucial parameters (and probably should be optimized for each protein). Longer PEG lengths cannot report on changes efficiently (Fig3b), while shorter lengths are prone to artefacts as they can go out of membrane (Fig1 and Fig2). This might limit its widespread use.

The reason for using the ACP tag is that neither the SNAP tap nor the HALO tag working. The tethered Nile Red preferred to bind to the tqg rather than inserting into the membrane.

**Referees cross-commenting** I agree with all comments and concerns of other reviewers. I see the usability and potential of this new technology along with its limitations as all three reviewers pointed out.

Reviewer #3 (Evidence, reproducibility and clarity (Required)): See below. No concerns on any of these issues.

Reviewer #3 (Significance (Required)): **Critique:** This MS reports a proof-of-principle for using site-directed environmentally sensitive probe technology to assess the local membrane environment of a receptor tyrosine kinase (IR) upon activation. This technology addresses a major gap in our arsenal of tools to study the mechanisms of membrane signaling as the parameters of interest are biophysical parameters rather than purely biochemical ones. How to do this with spatial and temporal resolution is a major challenge. This study builds on previous work by the Riezman group that develops an extrinsic labeling system to tether Nile Red to specific sites on the ectodomain of a signaling receptor and then probe local membrane environments as a function of receptor activity. This is a carefully done study is well-controlled, is clever in design and is well-described. Although the major issues to which such a general technology could contribute involve intracellular (and not extracellular) event, the advances described will be of general interest -- particularly that local membrane order decreases when IR becomes activated. Specific comments for the authors' consideration follow:

**Specific Comments:** (i) As a general comment, the authors are measuring extracellular plasma membrane leaflet properties that may or may not translate to what is happening in the local inner leaflet environment. A general reader may well miss the significance of this. This point needs to be more explicitly emphasized in the Discussion.

This has been discussed in the revised version.

(ii) Why not treat cells with a PLC inhibitor to block PIP2 hydrolysis and ask if that inhibits membrane disorder. It is PIP2 hydrolysis/resynthesis that regulates the actin cytoskeleton at signaling receptors and this seems an attractive candidate for study.

There is a long list of attractive post-signaling events of the insulin receptor and how this works in different cell types that could be tested. We believe that this is beyond the scope of this study and we encourage others to do this.

(iii) The data acquisition time is at least 4 min which is long enough for activated receptors to be recruited to sites of endocytosis. Can the authors exclude the possibility that what they are measuring isn't reflective of such spatial reorganization? Does a clathrin inhibitor block the observed change in local membrane order for activated IR? We determined localization to AP2 adaptor containing clathrin coated pits at the cell surface and showed that during the time-course of the experiment that there is no significant change in co-localization or evidence for endocytosis (new figure 9). Therefore, we decided not to do the clathrin inhibitor blocking experiment because we believe that it could only lead to indirect effects.

(iv) Receptor activation is accompanied by other transitions such as dimerization, etc. Can the authors exclude the possibility that what they are measuring is related to changes in depth of insertion of the NR probe into the plasma membrane outer leaflet that is a consequence of IR conformational transitions associated with activation? This is highly unlikely given the fact that fluidification of the membrane environment is found with all length linkers. Given the intervals in increases in linker length on the 2031 construct, which is the closest to the membrane, it is very difficult to conceive that any of the ones larger than 5 PEGs restrict significantly the membrane insertion of the dye. **Referees cross-commenting**

I think we have a consensus opinion

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

See below. No concerns on any of these issues.

Significance

Critique:

This MS reports a proof-of-principle for using site-directed environmentally sensitive probe technology to assess the local membrane environment of a receptor tyrosine kinase (IR) upon activation. This technology addresses a major gap in our arsenal of tools to study the mechanisms of membrane signaling as the parameters of interest are biophysical parameters rather than purely biochemical ones. How to do this with spatial and temporal resolution is a major challenge. This study builds on previous work by the Riezman group that develops an extrinsic labeling system to tether Nile Red to specific sites on the ectodomain of a signaling receptor and then probe local membrane environments as a function of receptor activity.

This is a carefully done study is well-controlled, is clever in design and is well-described. Although the major issues to which such a general technology could contribute involve intracellular (and not extracellular) event, the advances described will be of general interest -- particularly that local membrane order decreases when IR becomes activated. Specific comments for the authors' consideration follow:

Specific Comments:

(i) As a general comment, the authors are measuring extracellular plasma membrane leaflet properties that may or may not translate to what is happening in the local inner leaflet environment. A general reader may well miss the significance of this. This point needs to be more explicitly emphasized in the Discussion.

(ii) Why not treat cells with a PLC inhibitor to block PIP2 hydrolysis and ask if that inhibits membrane disorder. It is PIP2 hydrolysis/resynthesis that regulates the actin cytoskeleton at signaling receptors and this seems an attractive candidate for study.

(iii) The data acquisition time is at least 4 min which is long enough for activated receptors to be recruited to sites of endocytosis. Can the authors exclude the possibility that what they are measuring isn't reflective of such spatial reorganization? Does a clathrin inhibitor block the observed change in local membrane order for activated IR?

(iv) Receptor activation is accompanied by other transitions such as dimerization, etc. Can the authors exclude the possibility that what they are measuring is related to changes in depth of insertion of the NR probe into the plasma membrane outer leaflet that is a consequence of IR conformational transitions associated with activation?

Referees cross-commenting

I think we have a consensus opinion

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, authors generated an ACP-attached Nile Red probe in order to specifically label Insulin receptor in the membrane. Owing to this specificity, one can measure the lipid membrane properties around a specific protein in the membrane.

Major comments:

For the conclusions in the manuscript to be convincing, in my opinion, these additional data need to be added. Some of these are new experiments, and some are detailed analysis of existing data. The new experiments are not for new line of investigation, instead it is to confirm their statements and conclusions. The major point is the reliability of spectral shift. In usual environment sensitive probes, it is certain that they are in the membrane whatever is done to the membrane. However, when the probe is attached to a protein, it is not trivial to have the same confidence that the probe is always inside the membrane, and it is in the same plane of the membrane. 1992-ACP-IR is a good example; authors state that it binds to the protein outside the membrane, but when there is cholesterol addition and -maybe more interestingly- cholesterol removal, the dye still reacts and changes its emission (even PreCT changes its emission quite a bit at the 570 nm region). This is a clear indication of a change in localization of the probe upon some changes in the membrane. This implies that observed spectral shifts may not be due to lipid packing differences, but due to localization of the probes. For this reason, it is crucial to know where any environment sensitive probe localize in the membrane with respect to membrane normal, and this knowledge is more important for this probe. Related to this, the spectral difference upon insulin treatment and activation of insulin receptor could be due to changes in probe's localization in the membrane. Especially because authors show in Fig1e, the spectra can change depending on the probe localization. Relatedly, quantum yield of NR should be significantly different when it is inside vs outside membrane. Authors should show QY for 1992-ACP-NR and 2031-ACP-NR with different PEG lengths and upon insulin treatment.

Minor comments:

• Fig 1d requires quantification
• Voltage sensitivity of different PEG length of 2031-ACP probe should be added.
• Fig 3a graph should show all data points, not only bar graphs. Also, the band in 3a for +CoA-PEG-NR is dimmer than other bands, is it specific to this particular gel since quantification does not show any difference?
• Fig 4d, colour code is needed.
• Fig 5b and Fig3d are basically the same experiments in terms of control measurement, why is the difference in 3b is 0.04 GP unit while it is 0.007 GP unit?
• Why is inhibitor data so noisy?

Significance

Overall, this is a very useful approach, and this line of research will yield very useful tools to shed light on how lipids surrounding proteins can change their function. Major advance of the paper is the new chemical biology tool. There is also biological data on how insulin can change the insulin receptor's membrane environment which is contradictory to some old literature claiming that InsR becomes more "rafty" upon insulin treatment (e.g., PMID: 11751579).

If this type of tagging proves robust and reproducible (limitations and concerns listed above and below), it could be used by other researchers to tag their protein of interest and investigate the lipid environment around those proteins.

The downside of this method is that the probe requires ACP tag, a relatively less used tag than others in biology, therefore researchers interested in using this probe should have their proteins with ACP tag. Moreover, the linker length and ACP-tag position are quite crucial parameters (and probably should be optimized for each protein). Longer PEG lengths cannot report on changes efficiently (Fig3b), while shorter lengths are prone to artefacts as they can go out of membrane (Fig1 and Fig2). This might limit its widespread use.

Referees cross-commenting

I agree with all comments and concerns of other reviewers. I see the usability and potential of this new technology along with its limitations as all three reviewers pointed out.

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:

Techniques to probe the local environment of membrane proteins are sparse, although the influence of lipids on the membrane protein's function are known since many years. Therefore, the paper by Umebayashi et al. is important. The environment-sensitive dye Nile red (NR) coupled to a membrane protein is an appropriate sensor for monitoring the local membrane fluidity. Linking of Nile red to the receptor via a flexible tether was achieved with the acyl carrier protein (ACP)-tag method. Experiments showed that depending on the ACP site a certain linker length is required to have NR inserted in the membrane and thus be an effective sensor for lipid disorder. This technology could be of general usability to study the environment of membrane proteins in the context of their function. As an example, the technique allowed insulin induced membrane disorder in the close insulin receptor vicinity to be observed. Further, results suggested that tyrosine activity is required for this disorder to happen. The experimental results appear to be complete and controls were made.

Major comments:

1) Sometimes technical terms are used without explanation: What is the GP value? What is ACP-IR? The spectrum was measured in number of rois? The reader can find those abbreveations out, but it would be nice to have them defined.

2) Fig. 1d) is confusing. The ACP-IR labelling is evident in 3 panels, but there is no difference in the color (emission spectra of 1992-ACP-IR vs 2031-ACP-IR should be visible??). The DAPI staining is very different. When doing the latter, how difficult is it to get the staining equal?

3) How can one interpret Fig. 4: a) Control goes over 4 frames, at 240" insulin is added, and 10 frames should show a fluctuation difference? b) A color shift from blue to green is visible after insulin addition. But it is faint - difficult to assess from the pseudo color scheme. What does 1000 pixel top/1000 pixel bottom mean in c). Is it an attempt to better visualize the fluctuation? It is difficult to recognize a difference before and after adding insulin. d) It seems that the kymograph set should show this. What is the color scale? Why is 3 so untypical, i.e., no change? Box 6 is also peculiar: the left side does not show a strong change upon insulin administration, the right side does. Why?

4) How is the kymogram calculated? The legend says 'The horizontal dimension represents the averaged ZNCC inside the rectangular area, and the vertical dimension represents time'. The averaged ZNCC is a single value, so it is not clear why the kymogram shows a variation from left to right. May it be the ZNCC was averaged just vertically?

5) When calculating cross-correlation values on images, they need to be aligned. What fraction of the total image does the selected 19x19 box represent? As described, I imagine that a rolling CC over 19x19 pixels is calculated over an image from the time lapse series comparing it with the reference Iave(x,y). Compared to the 3x3 median filtered CP image, the ZNCC image should then be much more blurred??

Minor comment:

On page 16 supplementary is not spelled properly.

Significance

The key point of this paper is convincing and the new technology appears to have a lot of potential. It can be applied to study membrane protein function in the context of its environment, the lipid bilayer.

Membrane fluidity measurements have been developed (e.g., using fluorescent probes like laurdan). However, the trick to link a probe like nile red by ACP technology to the insulin receptor and to observe its activity is quite new.

A most recent description of such a technology is in TrAC Trends in Analytical Chemistry Volume 133, December 2020, 116092.

Referees cross-commenting

All comments are constructive and important. The paper is important but needs to be amended as proposed.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

First we would like to express our deep gratitude to the reviewers for thoroughly and fairly reviewing our work.

---

Reviewer #1:

Major Concerns

1. A major concern I have is with the use of DAPT to modulate Notch signaling, and investigate the impact on integrins, Yap, cadherins, etc. Gamma-secretase, the target of DAPT, cleaves not only Notch receptors, but also IntegrinB1, Nectins, Cadherins, Ephrins and more. This recent review lists 149 substrates (Guner & Lichtenthaler Seminars in Cell & Developmental Biology 2020). The risk that some of the results reflect DAPT impact on IntegrinB1, Cadherins etc themselves is significant. The authors should validate their findings with more specific modulation of Notch activity, for example with a Notch blocking antibody, with siRNA, or with SAHM1. We agree with the reviewer´s comment and will add additional key experiments using SAHM1 as alternative inhibitor of Notch activity.

Furthermore, EGTA was used to "acutely destabilize VE-Cadherin". But EGTA chelates Calcium, which is essential for Notch structure, and EGTA is thus a well-known activator of Notch signaling (see eg Rand MD et al. (2000) Calcium depletion dissociates and activates heterodimeric notch receptors. Mol Cell Biol). The authors rightfully describe and cite this paper, but the use of EGTA nonetheless confounds interpretation. The authors check for NICD levels (at what timepoint?) but the staining is cytoplasmic (also not labelled in the figure per se, but described in the figure legend? - please label the staining in the panel). And in any case, NICD is very short-lived and nuclear staining cannot be taken as a hallmark of signaling activity. In particular if staining is performed at a time point at which the receptor and NICD may have been exhausted/depleted. The authors should validate these observations/conclusions with the Notch reporter to conclusively demonstrate whether EGTA does not activate Notch in their system.

To test whether transient treatment with EGTA causes Notch activation we will repeat this experiment with Notch reporter activity as readout.

Trans-endocytosis of NECD on different substrates: the authors suggest that trans-endocytosis of NECD by Dll4 increases on softer substrates. But the authors also show that soft substrates lead to spreading out of cells, which could confound interpretation (is overlapping membranes, not internalization). The authors could validate trans-endocytosis by FACS: check if red Dll4+ cells contain more NECD. It is also not clear to me in this experiment whether the authors are looking at green NECD, or Notch1 full length, since they write "overlap of Notch1 and Dll4", which would not reflect trans-endocytosis but interactions at the cell surface for both cells. Please also define "overlay intensity", or explain further.

We will validate the trans-endocytosis by flow cytometry. In addition, we describe the procedure for microscopic analysis more clearly (methods section, p 4; results section, p 17-19)

The authors conclude their introduction with a statement that mechanosensitivity of Notch is linked to endocytosis, but their conclusion from Fig 6C was that Notch stiffness-dependence was independent of endocytosis, using the rhDll4..?

We have now rephrased this sentence.

• *

Minor concerns

1. In the introduction, the authors describe Dll3 as a Notch ligand that activates Notch signaling in trans. To my knowledge, Dll3 has only been described as a cis-inhibitor of Notch signaling. (I think this may have arisen during repeated edits of the manuscript!) This has now been corrected in the current version.

In the introduction, the authors state that Notch1, Dll4 and Jag1 control angiogenesis, but then they only describe what Notch1/Dll4 do in the next few sentences. Perhaps one sentence to describe the role of Jag1 would help avoid the feeling of being "left hanging".

This has now been corrected in the current version.

Data presentation: please show all bar graphs with the individual replicates (dotplots).

We have now changed all bar graphs into scatter plots.

Data analysis/normalization: many graphs represent normalization of values in multiple steps which are not described in the methods/legends/results. For example, Notch reporter gene activity (Fig 1A) is Firefly divided by Renilla, and presumably normalized to the control condition at 1 (or an average of 1 for the three controls?). This is not explained. Also, it is not clear whether the data reported for the Control condition are Huvec on rhDll4 compared (normalized) to Huvec on control substrate (and similar for each other condition). What controls are included in this experiment? Please provide the full data to provide insight into the magnitude of activation by Dll4 itself. Perhaps "Control" is without rhDll4? But the bar underneath A/B implies this rhDll4 was used in all conditions.

We have edited our manuscript accordingly to avoid these ambiguities.

Statistics: data should be presented as means +/- standard deviation, not standard error of the mean (see for example Barde & Barde Perspect Clin Res. 2012): "SEM quantifies uncertainty in estimate of the mean whereas SD indicates dispersion of the data from mean. As readers are generally interested in knowing the variability within sample, descriptive data should be precisely summarized with SD."

We now use SD instead of SEM.

Statistics: In the Methods section, the authors state that one-way ANOVA was followed by Dunnett's multiple comparison test, and two-way ANOVA was followed by Tukey's multiple comparison test. Dunnett is used to compare every mean to a control mean, while Tukey is used to compare every mean with every other mean. Fig 1 describes using Dunnett for Fig 1B, but the end of the legend days Tukey was used. However Fig 1A,C show internal pairwise comparisons to plastic. Please be sure to explain which statistics were used where, and why, and if plastic was set as the comparator, please be explicit about this. Fig 3 uses "Sidak's corrected two-way ANOVA" and "Sidak's multiple comparison test"? I think Sidak is a method to correct alpha or p for multiple comparisons, as stated in the first instance, but it is described why this was used here, and not in other analyses, and whether the authors then applied Tukey's post-hoc test as described in the methods section? Similar comments for Fig 6. It is counter-intuitive that the plastic -1.5kPa PDMS difference with no error-bar overlap in 1A would be 1-star significance, while the plastic-70kPa difference with almost overlapping error bars in 1B would be 4-star significance. Please check/show values. In Fig 1B Figure legend, the authors write "Data is presented in a bar plot and compared with the integrin β____1 intensities without DAPT treatment", but this is not the statistical comparison presented. Fig 3B shows a very minor difference with overlapping error bars as 3-star significance? Is this correct?

We have checked all statistical issues and corrected where necessary. Since the sample size and variance were homogenous in all comparisons we now uniformly use ANOVA and Tukey´s multiple comparison test as post hoc to keep things simple.

How much nuclear NICD (NICD intensity) is there in control conditions? (Control missing from Fig 1B, D).

We will repeat the experiment and compare the NICD levels with those in non-activated cells on plastic.

A DAPI counterstaining for 1B/D right panels would facilitate evaluation of whether NICD nuclear intensity is increased. The same applies for nuclear YAP assessment in Fig 3B. I assume a nuclear counter-stain was done for quantification of nuclear NICD intensity, and nuclear YAP intensity, but this is not described in the Materials and Methods, please add a description of how intensity was quantified, and provide nuclear counterstain images. (Also, what is the unit on the y-axis of "intensity" graphs? Arbitrary units (a.u.)?

The counterstaining method with Hoechst as well as the use of the nuclear staining for quantitative analysis of images are now described in the Methods section and where needed in the figure legends. The y-axis of the intensity graphs now has a dimension (a.u.). We decided against overlay of the nuclear staining with the NICD or YAP images for graphical reasons (visibility of the respective staining).

How much "overall" integrin B1 is there in DAPT-treated conditions in Fig 2C? (related to the concept that DAPT could be cleaving integrin B1, it could be depleted at 24 hours..?)

We will additionally add this experiment and validate the effect of Noch inhibition on the overall intergrin level by the alternative inhibitor SAHM1

More details regarding the analysis procedure need to be added to the Methods Section. Were cells segmented and then mean intensity estimated for the whole cell? Was this done by means of Intensity Ratio Nuclei Cytoplasm Tool plugin for Fiji alone? Were images background corrected, corrected for inhomogeneous illumination, normalized? In the case of Integrin beta 1 active, the expression seems to be patterned, was intensity expressed as mean intensity of every pixel corresponding to cytoplasm? For VE Cadherin staining, how was intensity estimated (only pixels corresponding to membrane were considered or every pixel of the cell)? Many figures are originated from a confocal microscope: were z-stacks acquired and then maximum projections done? Were z-stacks acquired and then fluorescence quantified in 3D images? Was a single plane acquired or analyzed, and if that is the case, how was this plane chosen?

The requested information has now been inserted in the respective results and method sections.

In Fig 4A, how is VE-Cadherin intensity quantified? As an average per field of view? Or per cell? And if per cell, how was each cell delineated? And if not per cell, how were equal cell numbers ensured? In FRAP experiment, how was intensity quantified? Was it per cell, per field of view or per region? Was each bleached region analyzed separately, or each cell? The datapoints should be either added to Figure 4C or as supplementary to assess the fitting. How many bleached regions per cell were done and how many cells were analyzed? In FRAP experiment, was bleaching done with an increased pixel dwell time? Was laser intensity increased? Do you have an estimation of laser power (not percentage) or flux?

These issues are now described in more detail in the respective figure legend.

Figure S2 is not referenced in the manuscript - I think a reference to "Figure S3" in the NECD transendocytosis section (no page numbers or line numbering) should be to Fig S2 instead?

Sorry for this mistake! We corrected this now.

In Figure 5A NICD nuclear intensity normalized somehow (normalization not explained), and stiffness no longer appears to regulate NICD levels as shown in Figure 1B.

We have now described the normalization better in the figure legend. The difference to the results in Fig. 1B is that in Fig. 5A the cells were not activated by Dll4 sender cells or rhDll4 (endogenous Notch activity). This is now stated more clearly.

Fig 6B: From the immuno at right there is a clear stiffness-dependent difference in Transferrin uptake. How were "single cell uptake" and "number of particles" quantified? (How were cell bodies identified?) Uptake could also be verified with FACS.

In this point, we disagree with the reviewer: we really do not see a systematic difference in intensities between the different substrates. The process of image analysis is now better described in the figure legend. The result was so clear that we did not use FACS as complementary approach.

Fig 6C: there appear to be very different numbers of cells in the brightfield image at right. Are the 70, 1.5, and 0.5 kPa Notch reporter activities different from one another or only different from plastic? Might these results reflect cell density/increased Notch signaling due to more cell-cell contacts?

Unfortunately, with decreasing stiffness the PDMS gels become optically more and more cloudy, giving the false impression of a higher cell number. We tried to circumvent this by changing contrast and brightness of the images, but to no satisfying effect. We now mention this issue in the figure legend.

How was the Dll4 coating of the different substrates done?

The coating of the substrates is now described under a specific subheading in the Methods section.

It would be helpful to describe the composition of Collagen G (Collagen I) in the text (it is a risk to expect vendor information to remain available indefinitely).

The role and composition of the Collagen G coatings was included in the text (p 7). Further information on the manufacturer of the product used is included in the methods section.

Please list catalog numbers for all reagents, and dilutions used for antibodies.

We have added this information wherever possible.

Instead of using red and green for images, maybe cyan, yellow and/or magenta could be used to help the reader see what is being shown (especially if the reader might be color blind).

We will of course adhere to the respective policy of the publishing journal, once the manuscript is accepted.

Packages and tools such as Intensity Ratio Nuclei Cytoplasm Tool plugin for FIJI should be referenced.

We have now referenced respective tools.

Reviewer #2:

*Major comments: *

Is there difference on a growth rate of cells on softer vrs stiffer gels that could affect cell morphology/signaling pathways?

This is an important point and we will perform additional respective experiments.

Nuclear localization of NICD and YAP would be good to validate with western blot.

Quantification of Western Blots (especially after nuclear isolation) is – at least in our hands – much less sensitive and reliable then quantitative imaging. We do not think that this experiment would strengthen our study.

In Figure 3 and Figure 5, siRNA experiments would strengthen the data. DAPT is not only an inhibitor of Notch but affects to other proteins as well. This should be stated.

A similar point was raised by Reviewer#1 with the suggestion to use SAHM1 as an alternative to DAPT. As suggested we will add these experiments.

How was the mean VE-cadherin branch length determined? This term often refers to angiogenesis assay/sprout formation and maybe another one should be considered here to describe VE-cadherin junction morphology.

Add to all figure texts how many cells were used for the analyses*. *

The cell number is now added wherever appropriate.

In Fig. 6C the cell morphology of HUVECs look abnormal in comparison to other images and should be re-done.

In contrast to all other experiments the cells where not confluent in this case. The different morphology is a sign of the lack of neighbours, not of some problem with the cells.

Was all the data normally distributed and thus ANOVA was used? Please add more details on the statistics part. Did you remove outliers?

Like also suggested by Reviewer #1 we have added more information on statistics and streamlined this. The data are normally distributed, outliers wer not removed.

MTT assay of DAPT would need to be presented as it can be cytotoxic. Cells are not well visible in Fig 2C with DAPT. DAPI and F-actin staining would help to see the cell morphology.

We will add respective data on cell viability after DAPT (and SAHM1) treatment in a revised version of the manuscript.

Minor comments:

Please clarify how coating with rhDDL4 is done as this was unclear at least for this reviewer.

The coating of the substrates is now described under a specific subheading in the Methods section.

HUVECs are known to be hard to transfect. Please provide data on transfection efficiencies of all transiently transfected cells.

We did not systematically monitor transfection efficiencies in this context, since there was always an internal control (e.g. co-reporter in the reporter gene assay) or the data were obtained on a single cell based quantification. Generally, we yield transfection efficiencies around 30% with HUVECs.

Reviewer #3:

Major comments:

• *

1) The authors use recombinant Dll4 or Dll4-expressing ("sender") cells to activate Notch in co-cultured cells. This is per se fine however, one might over-estimate all other observed downstream effects as endogenous Notch activity is lower. It would be important to see how naïve HUVEC or other primary endothelial cells respond to changes in stiffness. qPCR of Notch target genes such as Hey1, Hey2, Hes5, Dll4 is frequently used as a readout of Notch activity in this context. Also. the Notch transcriptional reporter assay might be a suitable read-out-

In Fig.5A we show data on endogenous Notch activity (- EGTA) on substrates with different stiffness. In this case NICD levels in the nucleus do not differ. It will definitely be interesting to repeat this experiment based on the reporter gene assay.

2) As the authors mention in the Discussion, cell density could be of utmost importance given the fact that Notch signaling usually is assumed as an in trans signaling event between adjacent cell membranes. However, also other signaling modes (in cis, cis inhibition, JAG1 vs DLL4 ratio) might be important. As such, the authors should carefully document an report on cell density in all experiments. Secondly, the authors should use other conditions such as sparse cell density and thirdly the authors should measure transcriptional effects of stiffness on Notch ligand expression.

In all experiments (with the exception of Fig. 6C) we used confluent cells. With the sparse cells (Fig. 6C) we also observe stiffness dependency. Investigating Notch ligand expression is definitely a good idea and will be investigated in the revised manuscript.

3) The authors need to compare stiffness in their model with physiological conditions in developing tissues and ideally also in tumor which often have increased tissue stiffness.

*Good point! We have now integrated such comparisons in the Discussion. *

4) Is Notch activation due to changes in stiffness dependent on the presence of ligands or could it be that (unspecific) binding of Notch receptors to ECM could trigger cleavage just by conformational change?

Since there is no stiffness dependent response on collagen (Fig. 6C, left panel), an effect of unspecific binding is highly unlikely.

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

The authors use different cell culture conditions to alter stiffness (DPMS model) and to measure the effect on Notch signaling and potential upstream and downstream factors. The experiments suggest that softer stiffness leads to higher Notch signaling activity in cultured endothelial cells which had been further stimulated by the Notch ligand DLL4. The data suggest that beta1 integrin activity is promoted by Notch which supports previous findings by others. Also, there is a bidirectional interaction with VE-Cadherin also supporting previous findings. This is a solid study using cultured cells only. The topic is of interest for researches investigating vascular biology, potentially also tumor vascular biology, ECM stiffness and its effect on signaling and Notch signaling per se.

Major comments:

• Are the key conclusions convincing? YES
• Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether? NO
• Would additional experiments be essential to support the claims of the paper? YES, SEE BELOW-
• Are the suggested experiments realistic in terms of time and resources? YES within about a six months' time period.
• Are the data and the methods presented in such a way that they can be reproduced? YES, however, more information is needed about cell density on the plates and the DLL4 expression level on the sender cells.

• Are the experiments adequately replicated and statistical analysis adequate? YES, however showing data points within the bar graphs would improve this study.

• The authors use recombinant Dll4 or Dll4-expressing ("sender") cells to activate Notch in co-cultured cells. This is per se fine however, one might over-estimate all other observed downstream effects as endogenous Notch activity is lower. It would be important to see how naïve HUVEC or other primary endothelial cells respond to changes in stiffness. qPCR of Notch target genes such as Hey1, Hey2, Hes5, Dll4 is frequently used as a readout of Notch activity in this context. Also. the Notch transcriptional reporter assay might be a suitable read-out-

• As the authors mention in the Discussion, cell density could be of utmost importance given the fact that Notch signaling usually is assumed as an in trans signaling event between adjacent cell membranes. However, also other signaling modes (in cis, cis inhibition, JAG1 vs DLL4 ratio) might be important. As such, the authors should carefully document an report on cell density in all experiments. Secondly, the authors should use other conditions such as sparse cell density and thirdly the authors should measure transcriptional effects of stiffness on Notch ligand expression.
• The authors need to compare stiffness in their model with physiological conditions in developing tissues and ideally also in tumor which often have increased tissue stiffness.
• Is Notch activation due to changes in stiffness dependent on the presence of ligands or could it be that (unspecific) binding of Notch receptors to ECM could trigger cleavage just by conformational change?

Significance

It was shown that Notch1 acts as a mechanosensor in endothelial cells. However, it is unclear how blood flow activates Notch1. Also, it is clear that stiffness influences blood vessel formation, which is under genetic control of Notch signaling. The importance of this study is to show that stiffness has a strong effect on Notch1 activation (maybe by increasing pulling force of ligands and subsequent endocytosis).

The major limitations of this study are:

1. work was only performed in cell culture, unclear whether there is any relevance in vivo
2. there is an artificial (over)-activation of endothelial Notch signaling by Dll4 expressing cells. Unclear whether this reflects physiological Notch signaling activity.
3. The mechanism how Notch1 gets activated remained elusive.

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

Kretchmer et al. investigates the role of substrate stiffness on Notch signalling pathway. They show increased Notch activity on softer substrates. Transendocytosis of NECD is suggested to be regulated by the substrate stiffness. They also conclude that the softer the substrate the more integrin beta 1 is activated.

Major comments:

Is there difference on a growth rate of cells on softer vrs stiffer gels that could affect cell morphology/signaling pathways?

Nuclear localization of NICD and YAP would be good to validate with western blot.

In Figure 3 and Figure 5, siRNA experiments would strengthen the data. DAPT is not only an inhibitor of Notch but affects to other proteins as well. This should be stated.

How was the mean VE-cadherin branch length determined? This term often refers to angiogenesis assay/sprout formation and maybe another one should be considered here to describe VE-cadherin junction morphology.

Add to all figure texts how many cells were used for the analyses.

In Fig. 6C the cell morphology of HUVECs look abnormal in comparison to other images and should be re-done.

Was all the data normally distributed and thus ANOVA was used? Please add more details on the statistics part. Did you remove outliers?

MTT assay of DAPT would need to be presented as it can be cytotoxic. Cells are not well visible in Fig 2C with DAPT. DAPI and F-actin staining would help to see the cell morphology.

Minor comments:

Please clarify how coating with rhDDL4 is done as this was unclear at least for this reviewer. HUVECs are known to be hard to transfect. Please provide data on transfection efficiencies of all transiently transfected cells.

Significance

The paper is interesting for the researchers studying angiogenesis and also cancer as the matrix stiffness regulates cancer progression.

My expertise lies in understanding mechanisms of angiogenesis, endothelial cell function and crosstalk with other cell types of the vessel wall. My group also studies Hippo signaling and has vast experience on isolation, culturing and doing experiments on HUVECs and other types of endothelial 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 manuscript, Kretschmer and colleagues investigate the role of matrix stiffness in Notch signaling using a series of gain and loss of function experiments (over-expression and inhibitors). As read-outs they use Notch reporter assays, FRAP, transferrin uptake, and immunofluorescence analyses. The authors conclude that softer substrates potentiate Notch signaling. While the questions are interesting and important, I am concerned with the use of inhibitors with off-target or unintended effects, as listed below. There is also some information missing from Materials and Methods which makes it difficult to assess the methodology and resulting conclusions.

Major Concerns

1. A major concern I have is with the use of DAPT to modulate Notch signaling, and investigate the impact on integrins, Yap, cadherins, etc. Gamma-secretase, the target of DAPT, cleaves not only Notch receptors, but also IntegrinB1, Nectins, Cadherins, Ephrins and more. This recent review lists 149 substrates (Guner & Lichtenthaler Seminars in Cell & Developmental Biology 2020). The risk that some of the results reflect DAPT impact on IntegrinB1, Cadherins etc themselves is significant. The authors should validate their findings with more specific modulation of Notch activity, for example with a Notch blocking antibody, with siRNA, or with SAHM1.
2. Furthermore, EGTA was used to "acutely destabilize VE-Cadherin". But EGTA chelates Calcium, which is essential for Notch structure, and EGTA is thus a well-known activator of Notch signaling (see eg Rand MD et al. (2000) Calcium depletion dissociates and activates heterodimeric notch receptors. Mol Cell Biol). The authors rightfully describe and cite this paper, but the use of EGTA nonetheless confounds interpretation. The authors check for NICD levels (at what timepoint?) but the staining is cytoplasmic (also not labelled in the figure per se, but described in the figure legend? - please label the staining in the panel). And in any case, NICD is very short-lived and nuclear staining cannot be taken as a hallmark of signaling activity. In particular if staining is performed at a time point at which the receptor and NICD may have been exhausted/depleted. The authors should validate these observations/conclusions with the Notch reporter to conclusively demonstrate whether EGTA does not activate Notch in their system.
3. Trans-endocytosis of NECD on different substrates: the authors suggest that trans-endocytosis of NECD by Dll4 increases on softer substrates. But the authors also show that soft substrates lead to spreading out of cells, which could confound interpretation (is overlapping membranes, not internalization). The authors could validate trans-endocytosis by FACS: check if red Dll4+ cells contain more NECD. It is also not clear to me in this experiment whether the authors are looking at green NECD, or Notch1 full length, since they write "overlap of Notch1 and Dll4", which would not reflect trans-endocytosis but interactions at the cell surface for both cells. Please also define "overlay intensity", or explain further.
4. The authors conclude their introduction with a statement that mechanosensitivity of Notch is linked to endocytosis, but their conclusion from Fig 6C was that Notch stiffness-dependence was independent of endocytosis, using the rhDll4..?

Minor concerns

1. In the introduction, the authors describe Dll3 as a Notch ligand that activates Notch signaling in trans. To my knowledge, Dll3 has only been described as a cis-inhibitor of Notch signaling. (I think this may have arisen during repeated edits of the manuscript!)
2. In the introduction, the authors state that Notch1, Dll4 and Jag1 control angiogenesis, but then they only describe what Notch1/Dll4 do in the next few sentences. Perhaps one sentence to describe the role of Jag1 would help avoid the feeling of being "left hanging".
3. Data presentation: please show all bar graphs with the individual replicates (dotplots).
4. Data analysis/normalization: many graphs represent normalization of values in multiple steps which are not described in the methods/legends/results. For example, Notch reporter gene activity (Fig 1A) is Firefly divided by Renilla, and presumably normalized to the control condition at 1 (or an average of 1 for the three controls?). This is not explained. Also, it is not clear whether the data reported for the Control condition are Huvec on rhDll4 compared (normalized) to Huvec on control substrate (and similar for each other condition). What controls are included in this experiment? Please provide the full data to provide insight into the magnitude of activation by Dll4 itself. Perhaps "Control" is without rhDll4? But the bar underneath A/B implies this rhDll4 was used in all conditions.
5. Statistics: data should be presented as means +/- standard deviation, not standard error of the mean (see for example Barde & Barde Perspect Clin Res. 2012): "SEM quantifies uncertainty in estimate of the mean whereas SD indicates dispersion of the data from mean. As readers are generally interested in knowing the variability within sample, descriptive data should be precisely summarized with SD."
6. Statistics:
   • a. In the Methods section, the authors state that one-way ANOVA was followed by Dunnett's multiple comparison test, and two-way ANOVA was followed by Tukey's multiple comparison test. Dunnett is used to compare every mean to a control mean, while Tukey is used to compare every mean with every other mean. Fig 1 describes using Dunnett for Fig 1B, but the end of the legend days Tukey was used. However Fig 1A,C show internal pairwise comparisons to plastic. Please be sure to explain which statistics were used where, and why, and if plastic was set as the comparator, please be explicit about this.
   • b. Fig 3 uses "Sidak's corrected two-way ANOVA" and "Sidak's multiple comparison test"? I think Sidak is a method to correct alpha or p for multiple comparisons, as stated in the first instance, but it is described why this was used here, and not in other analyses, and whether the authors then applied Tukey's post-hoc test as described in the methods section? Similar comments for Fig 6.
   • c. It is counter-intuitive that the plastic -1.5kPa PDMS difference with no error-bar overlap in 1A would be 1-star significance, while the plastic-70kPa difference with almost overlapping error bars in 1B would be 4-star significance. Please check/show values.
   • d. In Fig 1B Figure legend, the authors write "Data is presented in a bar plot and compared with the integrin β1 intensities without DAPT treatment", but this is not the statistical comparison presented.
   • e. Fig 3B shows a very minor difference with overlapping error bars as 3-star significance? Is this correct?
7. How much nuclear NICD (NICD intensity) is there in control conditions? (Control missing from Fig 1B, D).
8. A DAPI counterstaining for 1B/D right panels would facilitate evaluation of whether NICD nuclear intensity is increased. The same applies for nuclear YAP assessment in Fig 3B. I assume a nuclear counter-stain was done for quantification of nuclear NICD intensity, and nuclear YAP intensity, but this is not described in the Materials and Methods, please add a description of how intensity was quantified, and provide nuclear counterstain images. (Also, what is the unit on the y-axis of "intensity" graphs? Arbitrary units (a.u.)?
9. How much "overall" integrin B1 is there in DAPT-treated conditions in Fig 2C? (related to the concept that DAPT could be cleaving integrin B1, it could be depleted at 24 hours..?)
10. More details regarding the analysis procedure need to be added to the Methods Section. Were cells segmented and then mean intensity estimated for the whole cell? Was this done by means of Intensity Ratio Nuclei Cytoplasm Tool plugin for Fiji alone? Were images background corrected, corrected for inhomogeneous illumination, normalized? In the case of Integrin beta 1 active, the expression seems to be patterned, was intensity expressed as mean intensity of every pixel corresponding to cytoplasm? For VE Cadherin staining, how was intensity estimated (only pixels corresponding to membrane were considered or every pixel of the cell)? Many figures are originated from a confocal microscope: were z-stacks acquired and then maximum projections done? Were z-stacks acquired and then fluorescence quantified in 3D images? Was a single plane acquired or analyzed, and if that is the case, how was this plane chosen?
11. In Fig 4A, how is VE-Cadherin intensity quantified? As an average per field of view? Or per cell? And if per cell, how was each cell delineated? And if not per cell, how were equal cell numbers ensured?
12. In FRAP experiment, how was intensity quantified? Was it per cell, per field of view or per region? Was each bleached region analyzed separately, or each cell? The datapoints should be either added to Figure 4C or as supplementary to assess the fitting. How many bleached regions per cell were done and how many cells were analyzed?
13. In FRAP experiment, was bleaching done with an increased pixel dwell time? Was laser intensity increased? Do you have an estimation of laser power (not percentage) or flux?
14. Figure S2 is not referenced in the manuscript - I think a reference to "Figure S3" in the NECD transendocytosis section (no page numbers or line numbering) should be to Fig S2 instead?
15. In Figure 5A NICD nuclear intensity normalized somehow (normalization not explained), and stiffness no longer appears to regulate NICD levels as shown in Figure 1B.
16. Fig 6B: From the immuno at right there is a clear stiffness-dependent difference in Transferrin uptake. How were "single cell uptake" and "number of particles" quantified? (How were cell bodies identified?) Uptake could also be verified with FACS.
17. Fig 6C: there appear to be very different numbers of cells in the brightfield image at right. Are the 70, 1.5, and 0.5 kPa Notch reporter activities different from one another or only different from plastic? Might these results reflect cell density/increased Notch signaling due to more cell-cell contacts?
18. How was the Dll4 coating of the different substrates done?
19. It would be helpful to describe the composition of Collagen G (Collagen I) in the text (it is a risk to expect vendor information to remain available indefinitely).
20. Please list catalog numbers for all reagents, and dilutions used for antibodies.
21. Instead of using red and green for images, maybe cyan, yellow and/or magenta could be used to help the reader see what is being shown (especially if the reader might be color blind).
22. Packages and tools such as Intensity Ratio Nuclei Cytoplasm Tool plugin for FIJI should be referenced. https://github.com/MontpellierRessourcesImagerie/imagej_macros_and_scripts/wiki/Intensity-Ratio-Nuclei-Cytoplasm-Tool#how-to-cite-the-tool

Significance

The concept of how stiffness regulates Notch signaling is of timely interest. While the mechanobiology of Notch has attracted a fair amount of attention (publications), less is known of how stiffness impacts Notch signaling.

The work could be of interest to the Notch field, biomechanics, cell biology/adhesion experts. It could be relevant for designing cellular scaffolds for biological or medical applications.

The expertise of this reviewer is Notch and imaging.