Reviewer #2 (Public review):

Summary

In this extensive comparative study, Moreno-Borrallo and colleagues examine the relationships between plasma glucose levels, albumin glycation levels, diet and life-history traits across birds. Their results confirmed the expected positive relationship between plasma blood glucose level and albumin glycation rate but also provided findings that are somewhat surprising or contrast with findings of some previous studies (positive relationships between blood glucose and lifespan, or absent relationships between blood glucose and clutch mass or diet). This is the first extensive comparative analysis of glycation rates and their relationships to plasma glucose levels and life history traits in birds that is based on data collected in a single study, with blood glucose and glycation measured using unified analytical methods (except for blood glucose data for 13 species collected from a database).

Strengths

This is an emerging topic gaining momentum in evolutionary physiology, which makes this study a timely, novel and important contribution. The study is based on a novel data set collected by the authors from 88 bird species (67 in captivity, 21 in the wild) of 22 orders, except for 13 species, for which data were collected from a database of veterinary and animal care records of zoo animals (ZIMS). This novel data set itself greatly contributes to the pool of available data on avian glycemia, as previous comparative studies either extracted data from various studies or a ZIMS database (therefore potentially containing much more noise due to different methodologies or other unstandardised factors), or only collected data from a single order, namely Passeriformes. The data further represents the first comparative avian data set on albumin glycation obtained using a unified methodology. The authors used LC-MS to determine glycation levels, which does not have problems with specificity and sensitivity that may occur with assays used in previous studies. The data analysis is thorough, and the conclusions are substantiated. Overall, this is an important study representing a substantial contribution to the emerging field evolutionary physiology focused on ecology and evolution of blood/plasma glucose levels and resistance to glycation.

Weaknesses

Unfortunately, the authors did not record handling time (i.e., time elapsed between capture and blood sampling), which may be an important source of noise because handling-stress-induced increase in blood glucose has previously been reported. Moreover, the authors themselves demonstrate that handling stress increases variance in blood glucose levels. Both effects (elevated mean and variance) are evident in Figure ESM1.2. However, this likely makes their significant findings regarding glucose levels and their associations with lifespan or glycation rate more conservative, as highlighted by the authors.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews:

Reviewer #1 (Public review):

The paper explored cross-species variance in albumin glycation and blood glucose levels in the function of various life-history traits. Their results show that

(1) blood glucose levels predict albumin gylcation rates

(2) larger species have lower blood glucose levels

(3) lifespan positively correlates with blood glucose levels and

(4) diet predicts albumin glycation rates.

The data presented is interesting, especially due to the relevance of glycation to the ageing process and the interesting life-history and physiological traits of birds. Most importantly, the results suggest that some mechanisms might exist that limit the level of glycation in species with the highest blood glucose levels.

While the questions raised are interesting and the amount of data the authors collected is impressive, I have some major concerns about this study:

(1) The authors combine many databases and samples of various sources. This is understandable when access to data is limited, but I expected more caution when combining these. E.g. glucose is measured in all samples without any description of how handling stress was controlled for. E.g glucose levels can easily double in a few minutes in birds, potentially introducing variation in the data generated. The authors report no caution of this effect, or any statistical approaches aiming to check whether handling stress had an effect here, either on glucose or on glycation levels.

(2) The database with the predictors is similarly problematic. There is information pulled from captivity and wild (e.g. on lifespan) without any confirmation that the different databases are comparable or not (and here I'm not just referring to the correlation between the databases, but also to a potential systematic bias (e.g. captivate-based sources likely consistently report longer lifespans). This is even more surprising, given that the authors raise the possibility of captivity effects in the discussion, and exploring this question would be extremely easy in their statistical models (a simple covariate in the MCMCglmms).

(3) The authors state that the measurement of one of the primary response variables (glycation) was measured without any replicability test or reference to the replicability of the measurement technique.

(4) The methods and results are very poorly presented. For instance, new model types and variables are popping up throughout the manuscript, already reporting results, before explaining what these are e.g. results are presented on "species average models" and "model with individuals", but it's not described what these are and why we need to see both. Variables, like "centered log body mass", or "mass-adjusted lifespan" are not explained. The results section is extremely long, describing general patterns that have little relevance to the questions raised in the introduction and would be much more efficiently communicated visually or in a table.

Reviewer #2 (Public review):

Summary

In this extensive comparative study, Moreno-Borrallo and colleagues examine the relationships between plasma glucose levels, albumin glycation levels, diet, and lifehistory traits across birds. Their results confirmed the expected positive relationship between plasma blood glucose level and albumin glycation rate but also provided findings that are somewhat surprising or contradicting findings of some previous studies (relationships with lifespan, clutch mass, or diet). This is the first extensive comparative analysis of glycation rates and their relationships to plasma glucose levels and life history traits in birds that are based on data collected in a single study and measured using unified analytical methods.

Strengths

This is an emerging topic gaining momentum in evolutionary physiology, which makes this study a timely, novel, and very important contribution. The study is based on a novel data set collected by the authors from 88 bird species (67 in captivity, 21 in the wild) of 22 orders, which itself greatly contributes to the pool of available data on avian glycemia, as previous comparative studies either extracted data from various studies or a database of veterinary records of zoo animals (therefore potentially containing much more noise due to different methodologies or other unstandardised factors), or only collected data from a single order, namely Passeriformes. The data further represents the first comparative avian data set on albumin glycation obtained using a unified methodology. The authors used LC-MS to determine glycation levels, which does not have problems with specificity and sensitivity that may occur with assays used in previous studies. The data analysis is thorough, and the conclusions are mostly wellsupported (but see my comments below). Overall, this is a very important study representing a substantial contribution to the emerging field of evolutionary physiology focused on the ecology and evolution of blood/plasma glucose levels and resistance to glycation.

Weaknesses

My main concern is about the interpretation of the coefficient of the relationship between glycation rate and plasma glucose, which reads as follows: "Given that plasma glucose is logarithm transformed and the estimated slope of their relationship is lower than one, this implies that birds with higher glucose levels have relatively lower albumin glycation rates for their glucose, fact that we would be referring as higher glycation resistance" (lines 318-321) and "the logarithmic nature of the relationship, suggests that species with higher plasma glucose levels exhibit relatively greater resistance to glycation" (lines 386-388). First, only plasma glucose (predictor) but not glycation level (response) is logarithm transformed, and this semi-logarithmic relationship assumed by the model means that an increase in glycation always slows down when blood glucose goes up, irrespective of the coefficient. The coefficient thus does not carry information that could be interpreted as higher (when <1) or lower (when >1) resistance to glycation (this only can be done in a log-log model, see below) because the semi-log relationship means that glycation increases by a constant amount (expressed by the coefficient of plasma glucose) for every tenfold increase in plasma glucose (for example, with glucose values 10 and 100, the model would predict glycation values 2 and 4 if the coefficient is 2, or 0.5 and 1 if the coefficient is 0.5). Second, the semi-logarithmic relationship could indeed be interpreted such that glycation rates are relatively lower in species with high plasma glucose levels. However, the semi-log relationship is assumed here a priori and forced to the model by log-transforming only glucose level, while not being tested against alternative models, such as: (i) a model with a simple linear relationship (glycation ~ glucose); or (ii) a loglog model (log(glycation) ~ log(glucose)) assuming power function relationship (glycation = a * glucose^b). The latter model would allow for the interpretation of the coefficient (b) as higher (when <1) or lower (when >1) resistance in glycation in species with high glucose levels as suggested by the authors.

Besides, a clear explanation of why glucose is log-transformed when included as a predictor, but not when included as a response variable, is missing.

We apologize for missing an answer to this part before. Indeed, glucose is always log transformed and this is explained in the text.

The models in the study do not control for the sampling time (i.e., time latency between capture and blood sampling), which may be an important source of noise because blood glucose increases because of stress following the capture. Although the authors claim that "this change in glucose levels with stress is mostly driven by an increase in variation instead of an increase in average values" (ESM6, line 46), their analysis of Tomasek et al.'s (2022) data set in ESM1 using Kruskal-Wallis rank sum test shows that, compared to baseline glucose levels, stress-induced glucose levels have higher median values, not only higher variation.

Although the authors calculated the variance inflation factor (VIF) for each model, it is not clear how these were interpreted and considered. In some models, GVIF^(1/(2*Df)) is higher than 1.6, which indicates potentially important collinearity; see for example https://www.bookdown.org/rwnahhas/RMPH/mlr-collinearity.html). This is often the case for body mass or clutch mass (e.g. models of glucose or glycation based on individual measurements).

It seems that the differences between diet groups other than omnivores (the reference category in the models) were not tested and only inferred using the credible intervals from the models. However, these credible intervals relate to the comparison of each group with the reference group (Omnivore) and cannot be used for pairwise comparisons between other groups. Statistics for these contrasts should be provided instead. Based on the plot in Figure 4B, it seems possible that terrestrial carnivores differed in glycation level not only from omnivores but also from herbivores and frugivores/nectarivores.

Given that blood glucose is related to maximum lifespan, it would be interesting to also see the results of the model from Table 2 while excluding blood glucose from the predictors. This would allow for assessing if the maximum lifespan is completely independent of glycation levels. Alternatively, there might be a positive correlation mediated by blood glucose levels (based on its positive correlations with both lifespan and glycation), which would be a very interesting finding suggesting that high glycation levels do not preclude the evolution of long lifespans.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

(1) Line 84: "glycation scavengers" such as polyamines - can you specify what these polyamines do exactly?

A clarification of what we mean with "glycation scavengers" is added.

(2) Line 87-89: specify that the work of Wein et al. and this sentence is about birds.

This is now clarified.

(3) Line 95: "88 species" add "OF BIRDS". Also, I think it would be nice if you specified here that you are relying on primary data.

This is now clarified (line 96).

(4) Line 90-119: I find this paragraph very long and complex, with too many details on the methodology. For instance, I agree with listing your hypothesis, e.g. that with POL, but then what variables you use to measure the pace of life can go in the materials and methods section (so all lines between 112-119).

This is explained here as a previous reviewer considered this presentation was indeed needed in the introduction.

(5) Line 122-124: The first sentence should state that you collected blood samples from various sources, and list some examples: zoos? collaborators? designated wild captures? Stating the sample size before saying what you did to get them is a bit weird. Besides, you skipped a very important detail about how these samples were collected, when, where, and using what protocols. We know very well, that glucose levels can increase quickly with handling stress. Was this considered during the captures? Moreover, you state that you had 484 individuals, but how many samples in total? One per individual or more?

We kindly ask the reviewer to read the multiple supplementary materials provided, in which the questions of source of the samples, potential stress effects and sample sizes for each model are addressed. All individuals contributed with one sample. More details about the general sources employed are given now in lines 125-127.

(6) Line 135-36: numbers below 10 should be spelled out.

Ok. Now that is changed.

(7) Line 136: the first time I saw that you had both wild and captive samples. This should be among the first things to be described in the methods, as mentioned above.

As stated above, details on this are included in the supplementary materials, but further clarifications have now been included in the main text (question 5).

(8) Line 137-138: not clear. So you had 46 samples and 9 species. But what does the 3-3-3 sample mean? or for each species you chose 9 samples (no, cause that would be 81 samples in total)?

This has now been clarified (lines 139-140).

(9) Line 139-141: what methodological constraints? Too high glucose levels? Too little plasma?

There were cases in which the device (glucometer) produced an unspecific error. This did not correspond to too high nor too low glucose levels, as these are differently signalled errors. Neither the manual nor the client service provided useful information to discern the cause. This may perhaps be related to the composition of the plasma of certain species, interfering with the measurement. Some clarifications have been added (lines 143-146).

(10) Line 143: should be ZIMS.

Corrected.

(11) Line 120-148: you generally talk about individuals here, but I feel it would be more precise to use 'samples'.

The use is totally interchangeable, as we never measured more than one sample for a given individual within this study. Besides, in some cases, saying “sample” could result less informative.

(12) Line 150: missing the final number of measurements for glucose and glycation.

Please, read the ESM6 (Table ESM6.1), where this information is given.

(13) Line 154-155: so you took multiple samples from the same individual? It's the first time the text indicates so. Or do you mean technical replicates were not performed on the same samples?

As previously indicated, each individual included only one sample. Replicates were done only for some individuals to validate the technique, as it would be unfeasible to perform replicates of all of them. This part of the text is referring to the fact that not all samples were analysed at the same time, as it takes a considerable amount of time, and the mass spectrometry devices are shared by other teams and project. Clarifications in this sense are now added (lines 160-163).

(14) Line 171-172: "After realizing that diet classifications from AVONET were not always suitable for our purpose" - too informal. Try rephrasing, like "After determining that AVONET diet classifications did not align with our research needs...", but you still need to specify what was wrong with it and what was changed, based on what argument?

The new formulation suggested by the reviewer has now been applied (lines 181-183). The details are given in the ESM6, as indicated in the text. 

(15) Line 174-176: You start a new paragraph, talking about missing values, but you do not specify what variable are you talking about. you talk about calculating means, but the last variable you mentioned was diet, so it's even more strange.

We refer to life history traits. It has now been clarified in the text (line 185).

(16) Line 177: what longevity records? Coming from where? How did you measure longevity? Maximum lifespan ever recorded? 80-90% longevity, life expectancy???

We refer to maximum lifespan, as indicated in the introduction and in every other case throughout the manuscript. Clarifications have now been introduced (188-190).

(17) Line 180-183: using ZIMS can be problematic, especially for maximum longevity. There are often individuals who had a wrong date of birth entered or individuals that were failed to be registered as dead. The extremes in this database are often way off. If you want to combine though, you can check the correlation of lifespans obtained from different sources for the overlapping species. If it's a strong correlation it can be ok, but intuitively this is problematic.

The species for which we used ZIMS were those for which no other databases reported any values. We could try correlations for other species, but this issue is not necessarily restricted to ZIMS, as the primary origin of the data from other databases is often difficultly traceable. Also, ZIMS is potentially more updated that some of the other databases, mainly Amniotes database, from which we rely the most, as it includes the highest number of species in the most easily accessible format.

(18) Line 181-186: in ZIMS you calculate the average of the competing records, otherwise you choose the max. Why use different preferences for the same data?

This constitutes a misunderstanding, for which we include clarifications now (line 196). We were referring here to the fact that for maximum lifespan the maximum is always chosen, while for other variables an average is calculated. 

(19) Line 198: Burn-in and thinning interval is quite low compared to your number of iterations. How were model convergences checked?

Please, check ESM1.

(20) Line 201-203: What's the argument using these priors? Why not use noninformative ones? Do you have some a priori expectations? If so, it should be explained.

Models have now been rerun with no expectations on the variance partitions so the priors are less informative, given the lack of firm expectations, and results are similar. Smaller nu values are also tried.

(21) Line 217: "carried" OUT.

Corrected (now in line 229).

(22) Line 233-234: "species average model" - what is this? it was not described in the methods.

Please, read the ESM6.

(23) Line 232-246: (a) all this would be better described by a table or plot. You can highlight some interesting patterns, but describing it all in the text is not very useful I think, (b) statistically comparing orders represented by a single species is a bit odd.

(a) Figure 1 shows this graphically, but this part was found to be quite short without descriptions by previous reviewers. (b) We recognise this limitation, but this part is not presented as one of the main results of the article, and just constitutes an attempt to illustrate very general patterns, in order to guide future research, as in most groups glycation has never been measured, so this still constitutes the best illustration of such patterns in the literature.

(24) Line 281: the first time I saw "mass-adjusted maximum lifespan" - what is this, and how was it calculated? It should be described in the methods. But in any case, neither ratios, nor residuals should be used, but preferably the two variables should be entered side by side in the model.

Please, see ESM6 for the explanations and justifications for all of this.

(25) Line 281: there was also no mention of quadratic terms so far. How were polynomial effects tested/introduced in the models? Orthogonal polynomials? or x+ x^2?

Please, read ESM6.

(26) Table 1. What is 'Centred Log10Body mass', should be added in the methods.

Please, read ESM6.

(27) Table 1: what's the argument behind separating terrestrial and aquatic carnivores?

This was mostly based on the a priori separation made in AVONET, but it is also used in a similar way by Szarka and Lendvai 2024 (comparative study on glucose in birds), where differences in glucose levels between piscivorous and carnivorous are reported. We had some reasons to think that certain differences in dietary nutrient composition, as discussed later, can make this difference relevant.

(28) Table 1: The variable "Maximum lifespan" is discussed and plotted as 'massadjusted maximum lifespan' and 'residual maximum lifespan'. First, this is confusing, the same name should be used throughout and it should be defined in the methods section. Second, it seems that non-linear effects were tested by using x + x^2. This is problematic statistically, orthogonal polynomials should be used instead (check polyfunction in R). Also, how did you decide to test for non-linear effects in the case of lifespan but not the other continuous predictors? Should be described in the methods again.

Please, read ESM6. Data exploration was performed prior to carry out these models. Orthogonal polynomials were considered to difficult the interpretation of the estimates and therefore the patterns predicted by the models, so raw polynomials were used. Clarifications have now been included in line 297.

(29) Figure 2. From the figure label, now I see that relative lifespan is in fact residual. This is problematic, see Freckleton, R. P. (2009). The seven deadly sins of comparative analysis. Journal of evolutionary biology, 22(7), 1367-1375. Using body mass and lifespan side by side is preferred. This would also avoid forcing more emphasis on body mass over lifespan meaning that you subjectively introduce body mass as a key predictor, but lifespan and body size are highly correlated, so by this, you remove a large portion of variance that might in fact be better explained by lifespan.

Please, read ESM6 for justifications on the use of residuals.

Reviewer #2 (Recommendations for the authors):

(1) If the semi-logarithmic relationship (glycation ~ log10(glucose)) is to be used to support the hypothesis about higher glycation resistance in species with high blood glucose (lines 318-321 and 386-388), it should be tested whether it is significantly better than the model assuming a simple linear relationship (i.e., glycation ~ glucose). Alternatively, if the coefficient is to be used to determine whether glycation rate slows down or accelerates with increasing glucose levels, log-log model (log10(glycation) ~ log10(glucose)) assuming power function relationship (glycation = a * glucose^b) should be used (as is for example in the literature about relationships between metabolic rates and body size). Probably the best approach would be to compare all three models (linear, semi-logarithmic, and log-log) and test if one performs significantly better. If none of them, then the linear model should be selected as the most parsimonious.

Different options (linear, both semi-logarithmic combinations and log-log) have now been tested, with similar results. All of the models confirm the pattern of a significant positive relationship between glucose and glycation. Moreover, when standardizing the variables (both glucose and glycation, either log transformed or not), the estimate of the slope is almost equal for all the models. It is also lower than one, which in the case of both the linear and log-log confirms the stated prediction. The log-log model, showing a much lower DIC than the linear version, is now shown as the final model.

(2) ESM6, line 46: Please note that Kruskal-Wallis rank sum test in ESM1 shows that, compared to baseline glucose levels, stress-induced glucose levels have higher median values (not only higher variation). With this in mind, what is the argument here about increased variation being the main driver of stress-induced change in glucose levels based on? It seems that both the median values and variation differ between baseline and stress-induced levels, and this should be acknowledged here.

As discussed in the public answers, Kruskal Wallis does not allow to determine differences in mean, but just says that the groups are “different” (implicitly, in their ranksums, which does not mean necessarily in mean), while the Levene test performed signals heteroskedasticity. This makes this feature of the data analytically more grounded. Of course, when looking at the data, a higher mean can be perceived, but nothing can be said about its statistical significance. Still, some subtle changes have been introduced in corresponding section of the ESM6.

(3) Have you recorded the sampling times? If yes, why not control them in the models? It is at least highly advisable to include the sampling times in the data (ESM5).

As indicated in ESM6 lines 42-43, we do not have sampling times for most of the individuals (only zebra finches and swifts), so this cannot be accounted for in the models.

(4) If sampling times will remain uncontrolled statistically, I recommend mentioning this fact and its potential consequences (i.e., rather conservative results) in the Methods section of the main text, not only in ESM6.

A brief description of this has now been included in the main text (lines 129-132), referencing the more detailed discussion on the supplementary materials. Some subtle changes have also been included in the “Possible effects of stress” section of the ESM6.

(5) ESM6, lines 52-53: The lower repeatability in Tomasek et al.' study compared to your study is irrelevant to the argument about the conservative nature of your results (the difference in repeatability between both studies is most probably due to the broader taxonomic coverage of the current study). The important result in this context is that repeatability is lower when sampling time is not considered within Tomasek et al's data set (ESM1). Therefore, I suggest rewording "showing a lower species repeatability than that from our data" to "showing lower species repeatability when sampling time is not considered" to avoid confusion. Please also note that you refer here to species repeatability but, in ESM1, you calculate individual repeatability. Nevertheless, both individual and species repeatabilities are lower when not controlling for sampling time because the main driver, in that case, is an increased residual variance.

We recognize the current confusion in the way the explanation is exposed, and have significantly changed the redaction of the section. However, we would like to indicate that ESM1 shows both species and individual repeatability (for Tomasek et al. 2022 data, for ours only species as we do not have repeated individual values). Changes are now made to make it more evident.

(6) I recommend providing brief guidelines for the interpretation of VIFs to the readers, as well as a brief discussion of the obtained values and their potential importance.

Thank you for the recommendation. We included a brief description in lines 230-231. Also in the results section (lines 389-393).

(7) Line: 264: Please note that the variance explained by phylogeny obtained from the models with other (fixed) predictors does not relate to the traits (glucose or glycation) per se but to model residuals.

We appreciate the indication, and this has been rephrased accordingly (lines 280-286).

(8) Change the term "confidence intervals" to "credible intervals" throughout the paper, since confidence interval is a frequentist term and its interpretations are different from Bayesian credible interval.

Thank you for the remark, this has now been changed.

(9) Besides lifespan, have you also considered quadratic terms for body mass? The plot in Figure 2A suggests there might be a non-linear relationship too.

A quadratic component of body mass has not shown any significant effect on glucose in an alternative model. Also, a model with linear instead of log glucose (as performed in other studies) did not perform better by comparing the DICs, despite both showing a significant relationship between glucose and body mass. Therefore, this model remains the best option considered as presented in the manuscript.

(10) ESM6, lines 115-116: It is usually recommended that only factors with at least 6 or 8 levels are included as random effects because a lower number of levels is insufficient for a good estimation of variance.

In a Bayesian approach this does not apply, as random and fixed factors are estimated similarly. 

(11) Typos and other minor issues:

a) Line 66: Delete "related".

b) Figure 2: "B" label is missing in the plot.

c) Reference 9: Delete "Author".

d) References 15 and 83 are duplicated. Keep only ref. 83, which has the correct citation details.

e) ESM6, line 49: Change "GLLM" to "GLMM".

Thank you for indicating this. Now it’s corrected.

eLife Assessment

This important study introduces a fully differentiable variant of the Gillespie algorithm as an approximate stochastic simulation scheme for complex chemical reaction networks, allowing kinetic parameters to be inferred from empirical measurements of network outputs using gradient descent. The concept and algorithm design are convincing and innovative. While the proofs of concept are promising, some questions are left open about implications for more complex systems that cannot be addressed by existing methods. This work has the potential to be of significant interest to a broad audience of quantitative and synthetic biologists.

Reviewer #1 (Public review):

Summary:

This work introduces the differentiable Gillespie algorithm, DGA, which is a differentiable variant of the celebrated (and exact) Gillespie algorithm commonly used to perform stochastic simulations across numerous fields, notably in the life sciences. The proposed DGA approximates the exact Gillespie algorithm using smooth functions yielding a suitable approximate differentiable stochastic system as a proxy for the underlying discrete stochastic system, where DGA stochastic reactions have continuous reaction index and the species abundances. To illustrate their methodology, the authors specifically consider in detail the case of a well-studied two-state promoter gene regulation system that they analyze using a machine learning approach, and by combining simulation data with analytical results. For the two-state promoter gene system, the DGA is benchmarked by accurately reproducing the results of the exact Gillespie algorithm. For this same simple system, the authors also show how the DGA can be used for estimating kinetic parameters of both simulated and real noisy experimental data. This lets them argue convincingly that the DGA can become a powerful computation tool for applications in quantitative and synthetic biology. In order to argue that the DGA can be employed to design circuits with ad-hoc input-output relations, these considerations are then extended to a more complex four-state promoter model of gene regulation. The main strength of the paper is its clarity and its pedagogical presentation of the simulation methods.

Strengths:

The main strength of the paper is its clarity and its pedagogical presentation of the simulation methods.

Weaknesses:

It would have been useful to have a brief discussion, based on a concrete example, of what can be achieved with the DGA and is totally beyond the reach of the Gillespie algorithm and the numerous existing stochastic simulation methods. A more comprehensive and quantitative analysis of the limitations of the DGA, e.g. for rare events, and how it might be used for stochastic spatial systems would have also been helpful. However, this is arguably beyond the scope of this study whose primary goal is to introduce the DGA and demonstrate that it can achieve tasks like parameter estimation and network design.

Comments on revisions:

The authors have made a sound effort to address many of the comments raised in the previous reports. This has helped improve the clarity of the discussion.

Reviewer #2 (Public review):

Summary:

In this work, the authors present a differentiable version of the widely-used Gillespie Algorithm. The Gillespie Algorithm has been used for decades to simulate the behavior of stochastic biochemical reaction networks. But while the Gillespie Algorithm is a powerful tool for the forward simulation of biochemical systems given some set of known reaction parameters, it cannot be used for reverse process, i.e. inferring reaction parameters given a set of measured system characteristics. The Differentiable Gillespie Algorithm ("DGA") overcomes this limitation by approximating two discontinuous steps in the Gillespie Algorithm with continuous functions. This makes it possible to calculate of gradients for each step in the simulation process which, in turn, allows the reaction parameters to be optimized via powerful backpropagation techniques. In addition to describing the theoretical underpinnings of DGA, the authors demonstrate different potential use-cases for the algorithm in the context of simple models of stochastic gene expression.

Overall, the DGA represents an important conceptual step forward for the field and should lay the groundwork for exciting innovations in the analysis and design of stochastic reaction networks. At the same time, significantly more work is needed to establish when the approximations made by DGA are valid and to demonstrate the viability of the algorithm in the context of complicated reaction networks.

Strengths:

This work makes an important conceptual leap by introducing a version of the Gillespie Algorithm that is end-to-end differentiable. This idea alone has the potential to drive a number of exciting innovations in the analysis, inference, and design of biochemical reaction networks. Beyond the theoretical adjustments, the authors also implement their algorithm in a Python-based codebase that combines DGA powerful optimization libraries like PyTorch. This codebase has the potential to be of interest to a wide range of researchers, even if the true scope of the method's applicability remains to be fully determined.

The authors also demonstrate how DGA can be used in practice both to infer reaction parameters from real experimental data (Figure 7) and to design networks with user-specified input-output characteristics (Figure 8). These illustrations should provide a nice roadmap for researchers interested in applying DGA to their own projects/systems.

Finally, although it does not stem directly from DGA, the exploration of pairwise parameter dependencies in different network architectures provides an interesting window into the design constraints (or lack thereof) that shape the architecture of biochemical reaction networks.

Weaknesses:

While it is clear that the DGA represents an important conceptual advancement, the authors do not do enough in the present manuscript to (i) validate the robustness of DGA inference and (ii) demonstrate that DGA inference works in the kinds of complex biochemical networks where it would actually be of legitimate use.

It is to the authors' credit that they are open and explicit about the potential limitations of DGA due to breakdowns in its continuous approximations. However they do not provide the reader with nearly enough empirical (i.e. simulation-based) or theoretical context to assess when, why, and to what extent DGA will fail in different situations. In Figure 2, they compare DGA to GA (i.e. ground-truth) in the context of a simple two state model of a stochastic transcription. Even in this minimal system, we see that DGA deviates notably from ground-truth both in the simulated mRNA distributions (Figure 2A) and in the ON/OFF state occupancy (Figure 2C). This begs the question of how DGA will scale to more complicated systems, or systems with non-steady state dynamics. Will the deviations become more severe? This is important because, in practice, there is really not much need for using DGA with a simple 2 state system-we have analytic solutions for this case. It is the more complex systems where DGA has the potential to move the needle.

A second concern is that the authors' present approach for parameter inference and error calculation does not seem to be reliable. For example, in Figure 5A, they show DGA inference results for the ON rate of a two-state system. We see substantial inference errors in this case, even though the inference problem should be non-degenerate in this case. One reason for this seems to be that the inference algorithm does not reliably find the global minimum of the loss function (Figure 2B). To turn DGA into a viable approach, it is paramount that the authors find some way to improve this behavior, perhaps by using multiple random initializations to better search the loss space.

Finally, the authors do a good job of illustrating how DGA might be used to infer biological parameters (Figure 7) and design reaction networks with desired input-output characteristics (Figure 8). However, analytic solutions exist for both of the systems they select for examples. This means that, in practice, there would be no need for DGA in these contexts, since one could directly optimize, e.g., the expressions for the mean and Fano Factor of the system in Figure 7A. I still believe that it is useful to have these examples, but it seems critical to add a use-case where DGA is the only option.

Comments on revisions:

I am concerned that the results in Figure 8D may not be correct, or that the authors may be mis-interpreting them. From my reading of the paper they cite (Lammers & Flamholz 2023), the equilibrium sharpness limit for the network they consider in Figure 8 should be 0.25. But both solutions shown in Figure 8D fall below this limit, which means that they have sharpness levels that could have been achieved with no energy expenditure. If this is the case, then it would imply that while both systems do dissipate energy, they are not doing so productively; meaning that the same results could be achieved while holding Phi=0.

I acknowledge that this could be due to a difference in how they measure sharpness, but wanted to raise it here in case it is, in fact, a genuine issue with the analysis.

There should be an easy fix for this: just set the sharper "desired response" curve in 8b to be such that it demands non-equilibrium sharpness levels (0.25)

Reviewer #3 (Public review):

Summary:

This manuscript introduces a differentiable variant of the Gillespie algorithm (DGA) that allows gradient calculation using backpropagation. The most significant contribution of this work is the development of the DGA itself, a novel approach to making stochastic simulations differentiable. This is achieved by replacing discontinuous operations in the traditional Gillespie algorithm with smooth, differentiable approximations using sigmoid and Gaussian functions. This conceptual advance opens up new avenues for applying powerful gradient-based optimization techniques, prevalent in machine learning, to studying stochastic biological systems.

The method was tested on a simple two-state promoter model of gene expression. The authors found that the DGA accurately captured the moments of the steady-state distribution and other major qualitative features. However, it was less accurate at capturing information about the distribution's tails, potentially because rare events result from frequent low-probability reaction events where the approximations made by the DGA have a greater impact. The authors also used the DGA to design a four-state promoter model of gene regulation that exhibited a desired input-output relationship. The DGA could learn parameters that produced a sharper response curve, which was achieved by consuming more energy.

The authors conclude that the DGA is a powerful tool for analyzing and designing stochastic systems. The discussion lays several open questions in the field and constructively addresses shortcomings of the proposed method as well as potential ways forward.

Strengths:

The DGA allows gradient-based optimization techniques to estimate parameters and design networks with desired properties.

The DGA efficacy in estimating kinetic parameters from both synthetic and experimental data. This capability highlights the DGA's potential to extract meaningful biophysical parameters from noisy biological data.

The DGA's ability to design a four-state promoter architecture exhibits a desired input-output relationship. This success indicates the potential of the DGA as a valuable tool for synthetic biology, enabling researchers to engineer biological circuits with predefined behaviours.

Weaknesses:

The study primarily focuses on analysing the steady-state properties of stochastic systems.

Comments on revisions:

Thank you for addressing all the points raised. I am looking forward to seeing the next steps in DGAs development and performance!

Author response:

The following is the authors’ response to the current reviews.

Response to Reviewer 2’s comments:

I am concerned that the results in Figure 8D may not be correct, or that the authors may be mis-interpreting them. From my reading of the paper they cite (Lammers & Flamholz 2023), the equilibrium sharpness limit for the network they consider in Figure 8 should be 0.25. But both solutions shown in Figure 8D fall below this limit, which means that they have sharpness levels that could have been achieved with no energy expenditure. If this is the case, then it would imply that while both systems do dissipate energy, they are not doing so productively; meaning that the same results could be achieved while holding Phi=0.

I acknowledge that this could be due to a difference in how they measure sharpness, but wanted to raise it here in case it is, in fact, a genuine issue with the analysis.There should be an easy fix for this: just set the sharper "desired response" curve in 8b to be such that it demands non-equilibrium sharpness levels (0.25<S<0.5).

Thank you for raising this point regarding the interpretation of our results in Figure 8D. We agree that if the equilibrium sharpness limit for this particular network is around 0.25 (as shown by Lammers & Flamholz 2023), then achieving a sharpness below this threshold could, in principle, be accomplished without any energy expenditure. However, in our current design approach, the loss function is solely designed to enforce agreement with a target mean mRNA level at different input concentrations; it does not explicitly constrain energy dissipation, noise, or other metrics. Consequently, the DGA has no built-in incentive to minimize or optimize energy consumption, which means the resulting solutions may dissipate energy without exceeding the equilibrium sharpness limit.

In other words, the same input–output relationship could theoretically be achieved with \Phi =0 if an explicit constraint or regularization term penalizing energy usage had been included. As noted, adding such a term (e.g., penalizing \Phi^2) is conceptually straightforward but falls outside the scope of this study. Our primary goal is to demonstrate the flexibility of the DGA in designing a desired response, rather than to delve into energy–sharpness trade-offs or other biological considerations

While we appreciate the suggestion to set a higher target sharpness that exceeds the equilibrium limit, we believe the current example effectively demonstrates the DGA’s ability to design circuits with desired input-output relationships, which is the primary focus of this study. Researchers interested in optimizing energy efficiency, burst size, burst frequency, noise, response time, mutual information, or other system properties can easily extend our approach by incorporating additional terms into the loss function to target these specific objectives.

We hope this explanation addresses your concern and clarifies that the manuscript provides sufficient context for readers to interpret the results in Figure 8D correctly.

The following is the authors’ response to the original reviews.

Reviewer #1 (Public review):

We thank Reviewer #1 for their thoughtful feedback and appreciation of the manuscript's clarity. Our primary goal is to introduce the DGA  as a foundational tool for integrating stochastic simulations with gradient-based optimization. While we recognize the value of providing detailed comparisons with existing methods and a deeper analysis of the DGA’s limitations (such as rare event handling), these topics are beyond the scope of this initial work. Our focus is on presenting the core concept and demonstrating its potential, leaving more extensive evaluations for future research.

Reviewer #2 (Public review):

We thank Reviewer #2 for their detailed and constructive feedback. We appreciate the recognition of the DGA as a significant conceptual advancement for stochastic biochemical network analysis and design.

Weaknesses:

(1) Validation of DGA robustness in complex systems:

Our primary goal is to introduce the DGA framework and demonstrate its feasibility. While validation on high-dimensional and non-steady-state systems is important, it is beyond the scope of this initial work. Future studies may improve scalability by employing techniques such as dynamically adjusting the smoothness of the DGA's approximations during simulation or using surrogate models that remain differentiable but more accurately capture discrete behaviors in critical regions, thus preserving gradient computation while improving accuracy.

(2) Inference accuracy and optimization:

We acknowledge that the non-convex loss landscape in the DGA can hinder parameter inference and convergence to global minima, as seen in Figure 5A. While techniques like multi-start optimization or second-order methods (e.g., L-BFGS) could improve performance, our focus here is on establishing the DGA framework. We plan to explore better optimization methods in future work to improve the accuracy of parameter inference in complex systems.

(3) Use of simple models for demonstration:

We selected well-understood systems to clearly illustrate the capabilities of the DGA. These examples were intended to demonstrate how the DGA can be applied, rather than to solve problems better addressed by analytical methods. Applying DGA to more complex, analytically intractable systems is an exciting avenue for future work, but introducing the method was our main objective in this study.

Reviewer #3 (Public review):

We thank the reviewer for their detailed and insightful feedback. We appreciate the recognition of the DGA as a significant advancement for enabling gradient-based optimization in stochastic systems.

Weaknesses:

(1) Application beyond steady-state analysis

We acknowledge the limitation of focusing solely on steady-state properties. To extend the DGA for analyzing transient dynamics, time-dependent loss functions can be incorporated to capture system evolution over time. This could involve aligning simulated trajectories with experimental time-series data or using moment-matching across multiple time points. 

(2) Numerical instability in gradient computation

The reviewer correctly highlights that large sharpness parameters (a and b) in the sigmoid and Gaussian approximations can induce numerical instability due to vanishing or exploding gradients. To address this, adaptive tuning of a and b during optimization could balance smoothness and accuracy. Additionally, alternative smoothing functions (e.g., softmax-based reaction selection) and gradient regularization techniques (such as gradient clipping and trust-region methods) could improve stability and convergence.

Reviewer #1 (recommendations):

We thank the reviewer for their thoughtful and constructive feedback on our manuscript. Below, we address each of the comments and suggestions raised.

Main points:

(1) It would have been useful to have a brief discussion, based on a concrete example, of what can be achieved with the DGA and is totally beyond the reach of the Gillespie algorithm and the numerous existing stochastic simulation methods.

Thank you for your comment. We would like to clarify that the primary aim of this work is to introduce the DGA and demonstrate its feasibility for tasks such as parameter estimation and network design. Unlike traditional stochastic simulation methods, the DGA’s differentiable nature enables gradient-based optimization, which is not possible with the classical Gillespie algorithm or its variants.

(2) As often with machine learning techniques, there is a sense of black box, with a lack of mathematical details of the proposed method: as opposite to the exact Gillespie algorithm, whose foundations lie on solid mathematical results (exponentially-distributed waiting times of continuous-time Markov processes), the DGA involves uncontrolled approximations, that are only briefly mentioned in the paper. For instance, it is currently simply noted that "the approximations introduced by the DGA may be pronounced in more complex settings such as the calculation of rare events", without specifying how limiting these errors are. It would be useful to include a clearer and more comprehensive discussion of the limitations of the DGA: When does it work accurately? What are the approximations/errors and can they be controlled? When is it worth paying the price for those approximations/errors, and when is it better to stick to the Gillespie algorithm? Is this notably the case for problems involving rare events? Clearly, these are difficult questions, and the answers are problem specific. However, it would be important to draw the readers' attention on the issues, especially if the DGA is presented as a potentially significant tool in computational and synthetic biology.

We acknowledge the importance of discussing the limitations of the DGA in more detail. While we have noted that the approximations introduced by the DGA may impact its accuracy in certain scenarios, such as rare-event problems, a deeper exploration of these trade-offs is outside the scope of this work. Instead, we provide sufficient context in the manuscript to guide readers on when the DGA is appropriate.

(3) The DGA is here introduced and discussed in the context of non-spatial problems (simple gene regulatory networks). However, numerous problems in the life sciences and computational/synthetic biology, involve stochasticity and spatial degrees of freedom (e.g. for problems involving diffusion, migration, etc). It is notoriously challenging to use the Gillespie algorithm to efficiently simulate stochastic spatial systems, especially in the context of rare events (e.g., extinction or fixation problems). It would be useful to comment on whether, and possibly how, the DGA can be used to efficiently simulate stochastic spatial systems, and if it would be better suited than the Gillespie algorithm for this purpose.

Thank you for pointing this out. Although our current work centers on non-spatial systems, we agree that many biological contexts incorporate both stochasticity and spatial degrees of freedom. Extending the DGA to efficiently simulate such systems would indeed require substantial modifications—for instance, coupling it with reaction-diffusion frameworks or spatial master equations. We believe this is an exciting direction for future research and mention it briefly in the discussion as a potential extension.

Minor suggestions:

(1) After Eq.(10): it would be useful to explain and motivate the choice of the ratio JSD/H.

Done.

(2) On page 6, just below the caption of Fig.4: it would be useful to clarify what is actually meant by "... convergence towards the steady-state distribution of the exact Gillespie simulation, which is obtained at a simulation time of 10^4".

Done.

(3) At the end of Section B on page 7: please clarify what is meant here by "soft directions".

Done.

Reviewer #2 (recommendations):

We thank the reviewer for their thoughtful comments and constructive feedback. Below, we address each of the comments/suggestions.

Main points:

(1) Enumerate the conditions under which DGA assumptions hold (and when they do not). There is currently not enough information for the interested reader to know whether DGA would work for their system of interest. Without this information, it is difficult to assess what the true scope of DGA's impact will be. One simple idea would be to test DGA performance along two axes: (i) increasing number of model states and (ii) presence/absence of non-steady state dynamics. I acknowledge that these are very open-ended directions, but looking at even a single instance of each would greatly strengthen this work. Alternatively, if this is not feasible, then the authors should provide more discussion of the attendant difficulties in the main text.

We agree that a detailed exploration of the conditions under which the DGA assumptions hold would be a valuable addition to the field. However, this paper primarily aims to introduce the DGA methodology and demonstrate its proof-of-concept applications. A comprehensive analysis along axes such as increasing model states or non-steady-state dynamics, while important, would require significant additional simulations and is beyond the scope of this work. In Appendix A, we have discussed the trade-off between accuracy and numerical stability. Additionally, we encourage future users to tune the hyperparameters a and b for their specific systems.

(2) Demonstrate DGA performance in a more complex biochemical system. Clearly the authors were aware that analytic solutions exist for the 2-state system in Figure 7, but it this is actually also the case (I think) for mean mRNA production rate of the non-equilibrium system in Figure 8. To really demonstrate that DGA is practically viable, I encourage the authors to seek out an interesting application that is not analytically tractable.

We appreciate the suggestion to validate DGA on a more complex biochemical system. However, the goal of this study is not to provide an exhaustive demonstration of all possible applications but to introduce the DGA and validate it in systems where ground-truth comparisons are available. While the non-equilibrium system in Figure 8 might be analytically tractable, its complexity already provides a meaningful demonstration of DGA’s ability to optimize parameters and design systems. Extending this work to analytically intractable systems is an exciting direction for future studies, and we hope this paper will inspire others to explore these applications.

(3) Take steps to improve the robustness of parameter optimization and error bar calculations. (3a) When the loss landscape is degenerate, shallow, or otherwise "difficult," a common solution is to perform multiple (e.g. 25-100) inference runs starting from different random positions in parameter space. Doing this, and then taking the parameter set that minimizes the loss should, in theory, lead to a more robust recovery of the optimal parameter set.

(3b) It seems clear that the Hessian approximation is underestimating the true error in your inference results. One alternative is to use a "brute force" approach like bootstrap resampling to get a better estimate for the statistical dispersion in parameter estimates. But I recognize that this is only viable if the inference is relatively fast. Simply recovering the true minimum will, of course, also help.

(3a) We acknowledge the challenge posed by degenerate or shallow loss landscapes during parameter optimization. While performing multiple inference runs from different initializations is a common strategy, this approach is computationally intensive. Instead, we rely on standard optimization techniques (e.g., ADAM) to find a robust local minimum. 

(3b) Thank you for your comment. We agree that Hessian-based error bars can underestimate uncertainty, particularly in degenerate or poorly conditioned loss landscapes. While methods like bootstrap and Monte Carlo can provide more robust estimates, they can be computationally prohibitive for larger-scale simulations. A simpler reason for not using them is the high resource demand from repeated simulations, which quickly becomes infeasible for complex or high-dimensional models. We note these trade-offs between robust estimation and practicality as an important area for further exploration.

Moderate comments:

(1) Figure 7: is it possible to also show the inferred kon values? Specifically, it would be of interest to see how kon varies with repressor concentration.

Thank you for the suggestion. We have updated Figure 7 to include the inferred kon values, showing their variation with the mean mRNA copy number. However, we could not plot them against repressor concentration due to the lack of available data.

(2) Figure 8B & D: the authors claim that the sharper system dissipates more energy, but doesn't 8D show the opposite of this? More importantly, it does not look like either network drives sharpness levels that exceed the upper equilibrium limit cited in [36]. So it is not clear that it is appropriate to look at energy dissipation here. In fact, it is likely that equilibrium networks could produce the curves in 8B, and might be worth checking.

Thank you for pointing this out. We realized that the plotted values in Figure 8D were incorrect, as we had mistakenly plotted noise instead of energy dissipation. The plot has now been corrected. 

(3) Figure 8: I really like this idea of using DGA to "design" networks with desired input-output properties, but I wonder if you could explore more a biologically compelling use-case. Specifically, what about some kind of switch-like logic where, as the activator concentration increases, you have first 0 genes on, then 1 promoter on, then 2 promoters on. This would achieve interesting regulatory logic, and having DGA try to produce step functions would ensure that you force the networks to be maximally sharp (i.e. about double what you're currently achieving).

Thank you for this intriguing suggestion. While the proposed switch-like logic use case is indeed compelling, implementing such a system would require significant work. This goes beyond the scope of the current study, which focuses on demonstrating the feasibility of DGA for network design with simple input-output properties.

Minor comments:

(1) Figure 4B & C: the bar plots do not do a good job conveying the points made by the authors. Consider alternatives, such as scatter plots or box plots that could convey inference uncertainty.

Done.

(2) Figure 4B: consider using a log y-axis.

The y-axis in Figure 4B is already plotted on a log scale.

(3) Figure 4D is mentioned prior to 4C in the text. Consider reordering.

Done. 

(4) Figure 5B: it is difficult to assess from this plot whether or not the landscape is truly "flat," as the authors claim. Flat relative to what? Consider alternative ways to convey your point.

Thank you for highlighting this ambiguity. By describing the loss landscape as “flat,” we intend to convey its relative insensitivity to parameter variations in certain regions, rather than implying a completely level surface. While we believe Figure 5B still provides a useful qualitative depiction of this behavior, we acknowledge that it does not quantitatively establish “flatness.” In future work, we plan to incorporate more rigorous measures—such as gradient magnitudes or Hessian eigenvalues—to more accurately characterize and communicate the geometry of the loss landscape.

Reviewer #3 (recommendations):

We sincerely thank the reviewer for their thoughtful feedback and constructive suggestions, which have helped us improve the clarity and rigor of our manuscript. Below, we address each of the comments.

(1) Precision is lacking in the introduction section. Do the authors mean the Direct SSA, sorted SSA, which is usually faster, and how about rejection sampling methods?

Thank you for pointing this out. We have updated the introduction to explicitly mention the Direct SSA.

(2) When mentioning PyTorch and Jax, would be good to also talk about Julia, as they have fast stochastic simulators.

We have now mentioned Julia alongside PyTorch and Jax.

(3) Mentioned references 22-27. Reference 26 is an odd choice; a better reference is from the same author the Automatic Differentiation of Programs with Discrete Randomness, G Arya, M Schauer, F Schäfer, C Rackauckas, Advances in Neural Information Processing Systems, NeurIPS 2022

We have now cited the suggested reference.

(4) Page 1, Section: 'To circumnavigate these difficulties, the DGA modifies....' Have you thought about how you would deal with the bias that will be introduced by doing this?

Thank you for your insightful comment. We acknowledge the potential for bias due to the differentiable approximations in the DGA; however, our analysis has not revealed any systematic bias compared to the exact Gillespie algorithm. Instead, we observe irregular deviations from the exact results as the smoothness of the approximations increases.

(5) Page 2, first sentence '... traditional Gillespie...' be more precise here - the direct algorithm.

Thank you for your comment. We believe that the context of the paper, particularly the schematic in Figure 1, makes it clear that we are focusing on the Direct SSA. 

(6) Page 2, second paragraph: ' In order to simulate such a system...' This doesn't fit here as this section is about tau-leaping. As this approach approximates discrete operations, it is unclear if it would work for large models, snap-shot data of larger scale and if it would be possible to extend it for time-lapse data

Thank you for your comment. We respectfully disagree that this paragraph is misplaced. The purpose of this paragraph is to explain why the standard Gillespie algorithm does not use fixed time intervals for simulating stochastic processes. By highlighting the inefficiency of discretizing time into small intervals where reactions rarely occur, the paragraph provides necessary context for the Gillespie algorithm’s event-driven approach, which avoids this inefficiency.

Regarding the applicability of the DGA to larger models, snapshot data, or time-lapse data, we acknowledge these are important directions and have noted them as potential extensions in the discussion section.

(7) Page 2 Section B: 'In order to make use of modern deep-learning techniques...' It doesn't appear from the paper that any modern deep learning is used.

Thank you for your comment. Although the DGA does not utilize deep learning architectures such as neural networks, it employs automatic differentiation techniques provided by frameworks like PyTorch and Jax. These tools allow efficient gradient computations, making the DGA compatible with modern optimization workflows.

(8) Page 3, Fig 1(a). S matrix last row, B and C should swap places: B should be 1 and C is -1.

Corrected the typo.

(9) Fig1 needs a more detailed caption.

Expanded the caption slightly for clarity.

(10) Page 3 last paragraph: 'The hyperparameter b...' Consequences of this are relevant, for example can we now go below zero. Also, we lose more efficient algorithms here. It would be good to discuss this in more detail that this is an approx.. algorithm that is good for our case study, but for other to use it more tests are needed.

Thank you for the comment. Appendix A discusses the trade-offs related to a and b, but we agree that more detailed analysis is needed. The hyperparameters are tailored to our case study and must be tuned for specific systems.

(11) Page 4, Section C, first paragraph, 'The goal of making...' This is snapshot data. Would the framework also translate to time-lapse data? Also, it would be better to make it clearer earlier which type of data are the target of this study.

Thank you for your suggestion. While the current study focuses on snapshot data and steady-state properties, we believe the DGA could be extended to handle time-lapse data by incorporating multiple recorded time points into its inference objective. Specifically, one could modify the loss function to penalize discrepancies across observed transitions between these time points, effectively capturing dynamic trajectories. We consider this an exciting area for future development, but it lies beyond our present scope.

(12) Page 4 Section C, sentence '...experimentally measured moments'. Should later be mentioned as error, as moments are imperfect

Thank you for your comment. We agree that experimentally measured moments are inherently noisy and may not perfectly represent the true system. However, within the context of the DGA, these moments serve as target quantities, and the discrepancy between simulated and measured moments is already accounted for in the loss function. 

(13) Page 4 Section C, last sentence '...second-order...such as ADAM'. Another formulation would be better as second order can be confusing, especially in the context of parameter estimation

We have revised the language to avoid confusion regarding “second-order” methods.

(14) Fig 4(a) a density plot would fit better here

Fig. 4(a) has been updated to a scatter density plot as suggested.

(15) Fig 4(c) Would be interesting to see closer analysis of trade of between gradient and accuracy when changing a and b parameters

Thank you for this suggestion. We acknowledge that an in-depth exploration of these trade-offs could provide deeper insights into the method’s performance. However, for now, we believe the current analysis suffices to highlight the utility of the DGA in the contexts examined.

(16) Page 6 Section III, first sentence: This fits more to intro. Further the reference list is severely lacking here, with no comparison to other methods for actually fitting stochastic models.

Thank you for the suggestion. We have added a few references there.

(17) Page 6, Section A, sentence: '....experimental measured mean...' Why is it a good measure here (moment matching is not perfect), also do you have distribution data, would that not be better? How about accounting for measurement error?

Thank you for the comment. While we do not have full distribution data, we acknowledge that incorporating experimental measurement error could enhance the framework. A weighted loss function could model uncertainty explicitly, but this is beyond the scope of the current study. 

(18) Page 7, section B, first paragraph: 'Motivated by this, we defined the...'Why using Fisher-Information when profile-likelihood have proven to be better, especially for systems with few parameters like this.

Thank you for the suggestion. While profile-likelihood is indeed a powerful tool for parameter uncertainty analysis, we chose Fisher Information due to its computational efficiency and compatibility with the differentiable nature of the DGA framework.

(19)  Page 7, section C, sentence '...set kR/off=1..'. In this case, we cannot infer this parameter.

Thank you for the comment. You are correct that setting kR/off = 1 effectively normalizes the rates, making this parameter unidentifiable. In steady-state analyses, not all parameters can be independently inferred because observable quantities depend on relative—rather than absolute—rate values (as evident when setting the time derivative to zero in the master equation). To infer all parameters, one would need additional information, such as time-series data or moments at finite time.

(20)  Page 7 Section 2. Estimating parameters .... Sentence: '....as can be seen, there is very good agreement..' How many times the true value falls within the CI (because corr 0.68 is not great).

Thank you for your comment. While a correlation coefficient of 0.68 indicates moderate agreement, the primary goal was to demonstrate the feasibility of parameter estimation using the DGA rather than achieving perfect accuracy. The coverage of the CI was not explicitly calculated, as the focus was on the overall trends and relative agreement.

(21) Page 7 Section 2. Estimating parameters .... Sentence: 'Fig5(c) shows....' Is this when using exact simulator?

Thank you for your question. Yes, the exact values in x-axis of Fig. 5(c) are obtained using the exact Gillespie simulation.

(22) Page 7 Section 3 Estimating parameters for the... Sentence: 'Fig6(a) shows...' Why Cis are not shown?

Thank you for your comment. CIs are not shown in Fig. 6(a) because this particular case is degenerate, making the calculation and meaningful representation of CIs challenging. 

(23) Page 10, Sentence: 'As can be seen in Fig 7(b)...' Can you show uncertainty in measured value? It would be good to see something of a comparison against an exact method, at least on simulated synthetic data

Thank you for the comment. Fig. 7(a) already includes error bars for the experimental data, which account for measurement uncertainty. However, in Fig. 7(b), we do not include error bars for the experimental values due to limitations in the available data.

(24) Page 12, Section B Loss function '...n=600...' This is on a lower range. Have you tested with n=1000?

Yes, we have tested with n=1000 and observed no significant difference in the results. This indicates that n=600 is sufficient for the purposes of this study. 

(25) Fig 8(c) why there are no CI shown?

Thank you for your comment. CIs were not included in Fig. 8(c) due to degeneracy, which makes meaningful confidence intervals difficult to compute.

(26) Page 12 Conclusion, sentence: '..gradients via backpropagation...' Actually, by making the function continuous, both forward and reverse mode might be used. And in this case, forward-mode would actually be the fastest by quite a margin

Thank you for your insightful comment. You are correct that by making the function continuous, both forward-mode and reverse-mode automatic differentiation can be used. We have now mentioned this point in the discussion.

(27) Overall comment for the Conclusion section: It would be good to discuss how this framework compares to other model-fitting frameworks for models with stochastic dynamics. The authors mention dynamic data and more discussion on this would be very welcomed. Why use ADAM and not something established like BFGS for model fitting? It would be interesting to discuss how this can fit with other SSA algorithms (e.g. in practice sorting SSA is used when models get larger). Also, inference comparison against exact approaches would be very nice. As it is now, the authors truly only check the accuracy of the SSA on 1 model -it would be interesting to see for other models.

Thank you for your detailed comments. While this study focuses on introducing the DGA and demonstrating its feasibility, we agree that comparisons with other model-fitting frameworks, testing on additional models, and integrating with other SSA variants like sorted SSA are important directions for future work. Similarly, extending the DGA to handle transient dynamics and exploring alternatives to ADAM, such as BFGS, are promising areas to investigate further.

Reviewer #1 (Public review):

Summary:

The manuscript puts forward a statistical method to more accurately report the significance of correlations within data. The motivation for this study is two-fold. First, the publication of biological studies demands the report of p-values, and it is widely accepted that p-values below the arbitrary threshold of 0.05 give the authors of such studies justification to draw conclusions about their data. Second, many biological studies are limited by the number of replicate samples that are feasible, with replicates of less than 5 typical. The authors report a statistical tool that uses a permute-match approach to calculate p-values. Notably, the proposed method reduces p-values from around 0.2 to 0.04 as compared to a standard permutation test with a small sample size. The approach is clearly explained, including detailed mathematical explanations and derivations. The advantage of the approach is also demonstrated through analysis of computer-generated synthetic data with specified correlation and analysis of previously published data related to fish schooling. The authors make a clear case that this method is an improvement over the more standard approach currently used, and also demonstrate the impact of this methodology on the ability to obtain p-values that are the standard for biological research. Overall, this paper is very strong. While the subject matter seems somewhat specialized, I would make the case that this will be an important study that has broad general interest to readers. The findings are very general and applicable to many research contexts. Experimentalists also want to report accurate p-values in their work and better understand how these values are calculated. Although I believe the previous statement is true, I am not sure that many research groups doing biological work are reading specialized statistics journals regularly. Therefore a useful and broadly applicable statistical tool is well placed in this journal.<br /> Strengths:

The proposed method is broadly applicable to many realistic datasets in many experimental contexts.

The power of this method was demonstrated with both real experimental data and "synthetic" data. The advantages of the tool are clearly reported. The zebrafish data is a great example dataset.

The method solves a real-life problem that is frequently encountered by many experimental groups in the biological sciences.

The writing of the paper is surprisingly clear, given the technical nature of the subject matter. I would not at all consider myself a statistician or mathematician, but I found the text easy to follow. The authors did an impressive job guiding the reader through material that would often be difficult to grasp. The introduction was also well-written and clearly motivated the goals of the study.

Weaknesses:

A few changes could be made if the manuscript is revised. I would consider all of these points minor, but the paper could be improved if these points were addressed.

(1) The caption of Figure 2 doesn't seem to mention panel D. Figure A-2 also does not mention C in the caption.

(2) Figure 2D is a little hard to follow. First, the definition of "Power" is not clear, and I couldn't find the precise definition in the text. Second, the legend for the different lines in 2D is only given in Figure A-2. Perhaps a portion of the caption for Figure 2 is missing?

(3) The concept of circular variance for the fish data was heard to understand/visualize. The equation on line 326 did not help much. If there is a very simple picture that could be added near line 326 that helps to explain Ct and theta, that could be a big help for some readers who do not work on related systems. The analysis performed is understandable, the reader just has to accept that circular variance captions the degree of alignment of the fish.

(4) For the data discussed in Figure 3, I wasn't 100% sure how the time windows were selected. In the caption, it says "time series to different lengths starting from the first frame". So the 20 s time window was from t=0 to t= 20 s. Would a different result be obtained if a different 20 s window was chosen (from t = 4 min to t = 4 min 20 s just to give a specific example). I suppose by chance one of the time windows would give a p-value less than the target 0.05, that wouldn't be surprising. Maybe a random time window should be selected (although I am not indicating what was reported was incorrect)? A little more discussion on this aspect of the study may be helpful.

Reviewer #2 (Public review):

Summary:

This paper presented a hypothesis testing procedure for the independence of two time-series that was potentially suitable for nonlinear dependence and for small-sample cases. This should bring potential benefits for biology data.

Strengths:

The test offers good flexibility for different kinds of dependence (through adjusting \rho), and seems to have good finite sample performance compared to the literature. The justification regarding the validity of the test procedure is clear.

Weaknesses:

(1) The size of the test is not guaranteed to (asymptotically) equal \alpha, which may damage the power.

(2) The computational time can be an issue for a moderately large sample size when calculating the X / Y-perfect match. It will be beneficial to include discussions on the implementations of the test.

eLife Assessment

This study presents important insights into the regulation of left-right organ formation. By combining genetic perturbation of all three Meteorin genes in zebrafish and timelapse imaging, the authors identify an essential role for this protein family in the establishment of left-right patterning. They provide convincing evidence that Meteorins are required for the morphogenesis of dorsal forerunner cells, the precursors of the left-right organizer (also named Kupffer's vesicle) in zebrafish. In line with this, Meteorins were shown to genetically interact with integrins ItgaV and Itgb1b to regulate dorsal forerunner cell clustering.

Reviewer #1 (Public review):

Summary:

Meteorin proteins were initially described as secreted neurotrophic factors. In this manuscript, Eggeler et al. demonstrate a novel role for Meteorins in establish left-right axis formation in the zebrafish embryo. The authors generated null mutations in each of the three zebrafish meteorin genes - metrn, metrnla, and metrnlab. Triple mutant embryos displayed phenotypes strongly associated with left-right defects such as heart looping and visceral organ placement, and disrupted expression of Nodal-responsive genes, as did single mutants for metrn and metrnla. The authors then go on to demonstrate that these defects in left-right asymmetry are likely to due to defects in Kupffer's Vesicle and the progenitor dorseal forerunner cells including impaired lumen formation and reduced fluid flow, reduced clustering among DFCs, impaired DFC migration, mislocalization of apical proteins ZO-1 and aPKC, and detachment of DFCs from the EVL. Notably, the authors found that expression of marker genes sox32 and sox17 were not affected, suggesting Meteorins are required for DFC/KV morphogenesis but not necessarily fate specification. Finally, the authors show genetic interaction between Meteorins and integrin receptors, which were previously implicated in left-right patterning. In a supplemental figure, the manuscript also presents data showing expression of meteorin genes around the chick Hensen's node, suggesting that the left-right patterning functions may be conserved among vertebrates.

Strengths:

Strengths of this study include the generation of a triple mutant line that targets all known zebrafish meteorin family members. The experiments presented in this study were rigorous, especially with respect to quantification and statistical analysis.

Weaknesses:

Although the authors convincingly demonstrate a role for Meteorins in zebrafish left-right patterning, data supporting a conserved role in other vertebrates is compelling but limited to one supplemental figure.

Reviewer #2 (Public review):

Summary:

In this manuscript the authors describe their study on the role of meteorins in establishing the left-right organizer. The left-right organizer is a transient organ in vertebrate embryos in which rotating cilia cause a fluid flow that breaks the left-right symmetry and coordinates lateralization of internal organs such as gut and heart. In zebrafish, the left-right organizer (also named Kupffer's vesicle) is formed by dorsal forerunner cells, but very little is known about how dorsal forerunner cells coalles and form this ciliated vesicle in the embryo. The authors mutated the three meteorin-coding genes in zebrafish and observed that mutations in each one of these causes laterality defects with the strongest defects observed in the triple mutant. Loss of meteorins affects nodal gene expression, which play essential roles in establishing organ laterality. Meteorins are widely expressed in developing embryos and expression in lateral plate mesoderm and dorsal forerunner cells was observed. The meteorin triple mutant embryos display defects in the migration and clustering of the dorsal forerunner cells impairing kupffer's vesicle formation and cilia rotation. Finally, the authors show that meteorins genetically interact with integrins.

Strengths:

- These authors went through the lengthy process of generating triple mutants affecting all three meteorin genes. This provides robust genetic evidence on the role of meteorins in establishing organ laterality and circumvented that interpretation of the results would be hard due to redundant functions of meteorins.<br /> - The use of life imaging on triple mutants is appreciated<br /> - High-quality imaging of dorsal forerunner to quantify cell migrations and its relation to Kupffer's vesicle formation.

Weaknesses:

- Lack of a model how meteorins regulate dorsal forerunner cell migration.<br /> - Only genetic data to suggest a link between meteorins and integrins<br /> - Besides its role in DFC migration, meteorins may also play a more direct role in regulating Nodal signaling, which is not addressed here.

eLife Assessment

This study maps the genotype-phenotype landscapes of three E. coli transcription factors and the topographical features of these landscapes. It shows that ruggedness and epistasis do not hinder the evolution of strong transcription factor binding sites. These convincing findings contribute valuable insights into fitness landscape theories and highlight the role of chance, contingency, and evolutionary biases in gene regulation. The authors then study the topographical features of these landscapes, especially the number and distribution of local maxima, as well as the statistical properties of evolutionary paths on these landscapes.

Reviewer #1 (Public review):

Summary:

For each of the three key transcription factor (TF) proteins in E. coli, the authors generate a large library of TF binding site (TFBS) sequences on plasmids, such that each TFBS is coupled to the expression of a fluorescence reporter. By sorting the fluorescence of individual cells and sequencing their plasmids to identify each cell's TFBS sequence (sort-seq), they are able to map the landscape of these TFBSs to the gene expression level they regulate. The authors then study the topographical features of these landscapes, especially the number and distribution of local maxima, as well as the statistical properties of evolutionary paths on these landscapes. They find the landscapes to be highly rugged, with about as many local peaks as a random landscape would have, and with those peaks distributed approximately randomly in sequence space. The authors find that there are a number of peaks that produce regulation stronger than that of the wild-type sequence for each TF and that it is not too unlikely to reach one of those "high peaks" from a random starting sequence. Nevertheless, the basins of attractions for different peaks have significant overlap, which means that chance plays a major role in determining which peak a population will evolve to.

Strengths:

(1) The experiments and analysis of this paper are very well-executed and, by and large, very thorough (with an important exception identified below). I appreciated the systematic nature of the project, both the large-scale experiments done on three TFs with replicates and the systematic analysis of the resulting landscapes. This not only makes the paper easy to follow but also inspires confidence in their results since there is so much data and so many different ways of analyzing it. It's a great recipe for other studies of genotype-phenotype landscapes to follow.

(2) Considering how technical the project was, I am really impressed at how easy to read I found the paper, and the authors deserve a lot of credit for making it so. They do a great job of building up the experiments and analyses step-by-step and explaining enough of the basics of the experimental design and the essence of each analysis in the main text without getting too complicated with details that can be left to the Methods or SI. Compared to other big data papers, this one was refreshingly not overwhelming.

Weaknesses:

(1) The main weakness of this paper, in my view, is that it felt disconnected from the larger body of work on fitness and genotype-phenotype landscapes, including previous data on TFBSs in E. coli, genotype-phenotype maps of TFBSs in other systems, protein sequence landscapes (e.g., from mutational scans or combinatorially-complete libraries), and fitness landscapes of genomic mutations (e.g., combinatorially-complete landscapes of antibiotic resistance alleles). I have no doubt the authors are experts in this literature, and they probably cite most of it already given the enormous number of references. But they don't systematically introduce and summarize what was already known from all that work, and how their present study builds on it, in the Abstract and Introduction, which left me wondering for most of the paper why this project was necessary. Eventually, the authors do address most of these points, but not until the end, in the Discussion. Readers who have no familiarity with this literature might read this paper thinking that it's the first paper ever to study topography and evolutionary paths on genotype-phenotype landscapes, which is not true.

There were two points that made this especially confusing for me. First, in order to choose which nucleotides in the binding sites to vary, the authors invoke existing data on the diversity of these sequences (position-weight matrices from RegulonDB). But since those PWMs can imply a genotype-phenotype map themselves, an obvious question I think the authors needed to have answered right away in the Introduction is why it is insufficient for their question. They only make a brief remark much later in the Results that the PWM data is just observed sequence diversity and doesn't directly reflect the regulation strength of every possible TFBS sequence. But that is too subtle in my opinion, and such a critical motivation for their study that it should be a major point in the Introduction.

The second point where the lack of motivation in the Introduction created confusion for me was that they report enormous levels of sign epistasis in their data, to the point where these landscapes look like random uncorrelated landscapes. That was really surprising to me since it contrasts with other empirical landscape data I'm familiar with. It was only in the Discussion that I found some significant explanation of this - namely that this could be a difference between prokaryotic TFBSs, as this paper studies, and the eukaryotic TFBSs that have been the focus of many (almost all?) previous work. If that is in fact the case - that almost all previous studies have focused on eukaryotic TFBSs or other kinds of landscapes, and this is the first to do a systematic test of prokaryotic TFBS, then that should be a clear point made in the Abstract and Introduction. (I find a comparable statement only in the very last paragraph of the Discussion.) If that's the case, then I would also find that point to be a much stronger, more specific conclusion of this paper to emphasize than the more general result of observing epistasis and contingency (as is currently emphasized in the Abstract), which has been discussed in tons of other papers. This raises all sorts of exciting questions for future studies - why do the landscapes of prokaryotic TFBSs differ so dramatically from almost all the other landscapes we've observed in biology? What does that mean for the evolutionary dynamics of these different systems?

(2) I am a bit concerned about the lack of uncertainties incorporated into the results. The authors acknowledge several key limitations of their approach, including the discreteness of the sort-seq bins in determining possible values of regulation strength, the existence of a large number of unsampled sequences in their genotype space, as well as measurement noise in the fluorescence readouts and sequencing. While the authors acknowledge the existence of these factors, I do not see much attempt to actually incorporate the effect of these uncertainties into their conclusions, which I suspect may be important. For example, given the bin size for the fluorescence in sort-seq, how confident are they that every sequence that appears to be a peak is actually a peak? Is it possible that many of the peak sequences have regulation strengths above all their neighbors but within the uncertainty of the fluorescence, making it possible that it's not really a peak? Perhaps such issues would average out and not change the statistical nature of their results, which are not about claiming that specific sequences are peaks, just how many peaks there are. Nevertheless, I think the lack of this robustness analysis makes the results less convincing than they otherwise would be.

Reviewer #2 (Public review):

The authors aim to investigate the ability of evolution to create strong transcription factor binding sites (TFBSs) de novo in E. coli. They focus on three global transcriptional regulators: CRP, Fis, and IHF, using a massively parallel reporter assay to evaluate the regulatory effects of over 30,000 TFBS variants. By analyzing the resulting genotype-phenotype landscapes, they explore the ruggedness, accessibility, and evolutionary dynamics of regulatory landscapes, providing insights into the evolutionary feasibility of strong gene regulation. Their experiments show that de novo adaptive evolution of new gene regulation is feasible. It is also subject to a blend of chance, historical contingency, and evolutionary biases that favor some peaks and evolutionary paths.

(1) Strengths of the methods and results:

The authors successfully employed a well-designed sort-seq assay combined with high-throughput sequencing to map regulatory landscapes. The experimental design ensures reliable measurement of regulation strengths. Their system accounts for gene expression noise and normalizes measurements using appropriate controls.

Comprehensive Landscape Mapping:<br /> The study examines ~30,000 TFBS variants per transcription factor, providing statistically robust and thorough maps of the regulatory landscapes for CRP, Fis, and IHF. The landscapes are rigorously analyzed for ruggedness (e.g., number of peaks) and epistasis, revealing parallels with theoretical uncorrelated random landscapes.

Evolutionary Dynamics Simulations:<br /> Through simulations of adaptive walks under varying population dynamics, the authors demonstrate that high peaks in regulatory landscapes are accessible despite ruggedness. They identify key evolutionary phenomena, such as contingency (multiple paths to peaks) and biases toward specific evolutionary outcomes.

Biological Relevance and Novelty:<br /> The author's work is novel in focusing on global regulators, which differ from previously studied local regulators (e.g., TetR). They provide compelling evidence that rugged landscapes are navigable, facilitating de novo evolution of regulatory interactions. The comparison of landscapes for CRP, Fis, and IHF underscores shared topographical features, suggesting general principles of global transcriptional regulation in bacteria.

(2) Weaknesses of the methods and results:

Undersampling of Genotype Space:<br /> While the quality filtering of the data ensures robustness, ~40% of the TFBS space remains uncharacterized. The authors acknowledge this limitation but could improve the analysis by employing subsampling or predictive modeling.

Simplified Regulatory Architecture:<br /> The study considers a minimal system of a single TFBS upstream of a reporter gene. While this may have been necessary for clarity, this simplification may not reflect the combinatorial complexity of transcriptional regulation in vivo.

Lack of Experimental Validation of Simulations:<br /> The adaptive walks are based on simulated dynamics rather than experimental evolution. Incorporating in vivo experimental evolution studies would strengthen the conclusions. Although this is a large request for the paper, that would not prevent publication.

Impact on the Field:<br /> This study advances our understanding of adaptive landscapes in gene regulation and offers a critical step toward deciphering how global regulators evolve de novo binding sites. The findings provide foundational insights for synthetic biology, evolutionary genetics, and systems biology by highlighting the evolutionary accessibility of strong regulation in bacteria.

Utility of Methods and Data:<br /> The sort-seq approach, combined with landscape analysis, provides a robust framework that can be extended to other transcription factors and systems. If made publicly available, the study's data and code would be valuable for researchers modeling transcriptional regulation or studying evolutionary dynamics.

Additional Context:<br /> The study builds on a growing body of work exploring regulatory evolution. For instance, recent studies on local regulators like TetR and AraC have revealed high ruggedness and epistasis in TFBS landscapes. This study distinguishes itself by focusing on global regulators, which are more biologically complex and influential in bacterial gene networks. The observed evolutionary contingency aligns with findings in other biological systems, such as protein evolution and RNA folding landscapes, underscoring the generality of these evolutionary principles.

Conclusion:<br /> The authors successfully mapped the genotype-phenotype landscapes for three global regulators and simulated evolutionary dynamics to assess the feasibility of strong TFBS evolution. They convincingly demonstrate that ruggedness and epistasis, while prominent, do not preclude the evolution of strong regulation. Their results support the notion that gene regulation evolves through a blend of chance, contingency, and evolutionary biases.

This paper makes a significant contribution to the understanding of regulatory evolution in bacteria. While minor limitations exist, the authors' methods are robust, and their findings are well-supported. The work will likely be of broad interest to researchers in molecular evolution, synthetic biology, and gene regulation.

eLife Assessment

This important study characterizes and validates a new activity marker - fast labelling of engram neurons (FLEN) - which is transiently active and driven by cFos, allowing the monitoring of intrinsic and synaptic properties of engram neurons shortly after the learning experience. The results convincingly demonstrate the utility of this novel viral tool for studying early changes in the properties of engram cells. However, the study would benefit from exploring how accurately FLEN reflects endogenous cFos activity, how this labelling technique compares to previous versions, and from careful consideration of alternative explanations such as changes in release probability.

Reviewer #1 (Public review):

Summary:

The manuscript by Cupollilo et al describes the development, characterization, and application of a novel activity labeling system; fast labelling of engram neurons (FLEN). Several such systems already exist but this study adds additional capability by leveraging an activity marker that is destabilized (and thus temporally active) as well as being driven by the full-length promoter of cFos. The authors demonstrate the activity-dependent induction and time course of expression, first in cultured neurons and then in vivo in hippocampal CA3 neurons after one trial of contextual fear conditioning. In a series of ex vivo experiments, the authors perform patch clamp analysis of labeled neurons to determine if these putative engram neurons differ from non-labelled neurons using both the FLEN system as well as the previously characterized RAM system. Interestingly the early labelled neurons at 3 h post CFC (FLEN+) demonstrated no differences in excitability whereas the RAM-labelled neurons at 24h after CFC had increased excitability. Examination of synaptic properties demonstrated an increase in sEPCS and mEPSC frequencies as well as those for sIPSCs and mIPSCs which was not due to a change in the mossy fiber input to these neurons.

Strengths:

Overall the data is of high quality and the study introduces a new tool while also reassessing some principles of circuit plasticity in the CA3 that have been the focus of prior studies.

Weaknesses:

No major weaknesses were noted.

Reviewer #2 (Public review):

Summary:

Cupollilo et al. investigate the properties of hippocampal CA3 neurons that express the immediate early gene cFos in response to a single foot shock. They compare ex-vivo the electrophysiological properties of these "engram neurons" labeled with two different cFos promoter-driven green markers: Their new tool FLEN labels neurons 2-6 h after activity, while RAM contains additional enhancers and peaks considerably later (>24 h). Since the fraction of labeled CA3 cells is comparable with both constructs, it is assumed (but not tested) that they label the same population of activated neurons at different time points. Both FLEN+ and RAM+ neurons in CA3 receive more synaptic inputs compared to non-expressing control neurons, which could be a causal factor for cFos activation, or a very early consequence thereof. Frequency facilitation and E/I ratio of mossy fiber inputs were also tested, but are not different in both cFos+ groups of neurons. One day after foot shock, RAM+ neurons are more excitable than RAM- neurons, suggesting a slow increase in excitability as a major consequence of cFos activation.

Strengths:

The study is conducted to high standards and contributes significantly to our understanding of memory formation and consolidation in the hippocampus. Modifications of intrinsic neuronal properties seem to be more salient than overall changes in the total number of (excitatory and inhibitory) inputs, although a switch in the source of the synaptic inputs would not have been detected by the methods employed in this study

Weaknesses:

With regard to the new viral tool, a direct comparison between the new tool FLEN and existing cFos reporters is missing.

eLife Assessment

This study provides important evidence that the postmating behavioral switch in male mice is mediated by distinct stages of synaptic plasticity within the medial amygdala-MPOA-BSTrh pathway. The findings are convincing, supported by rigorous behavioral characterization and electrophysiological approaches that disentangle the contributions of mating, cohabitation, and parental experience to neural circuit changes. While some methodological details and statistical reporting require clarification, the study significantly advances our understanding of the neural mechanisms underlying paternal behavior.

Reviewer #1 (Public review):

Summary:

After mating, male mice undergo a behavioral switch from infanticide to parental behavior (postmating switch). The neural mechanisms underlying this switch are still largely unknown. Studies performed in different mouse strains have also resulted in mixed evidence for whether mating (specifically: ejaculation) itself is sufficient for this switch, or whether subsequent cohabitation with the pregnant female, and parental experience with pups is required. Recent work found that while lesions to the central part of the medial preoptic area (cMPOA) promote infanticidal behavior, lesions to the rhomboid nucleus of the bed nucleus of the stria terminalis (BSTrh) inhibit infanticide. The current work convincingly adds to this evidence by showing that mating and cohabitation lead to reduced inhibition from Cart-positive medial amygdala neurons onto cMPOA neurons, and that this synaptic change is in fact critical for the postmating switch. Further, the authors demonstrate that parental experience increases inhibitory synaptic transmission onto BSTrh neurons. The male postmating switch thus appears to rely on two sequential stages of synaptic plasticity.

Strengths:

(1) The behavioral characterization is thorough and the authors nicely manage to disentangle the relative contributions of mating, cohabitation, and parental experience to the postmating switch. Their finding of dissociable plasticity mechanisms underlying mating/cohabitation vs pup experience is intriguing.

(2) Most conclusions are based on complementary evidence from different experimental approaches and are compelling.

Weaknesses:

(1) The authors do not provide an explicit synthesis/model of the circuit-level changes underlying this switch. For instance, how does cMPOA-to-BSTrh connectivity change in fathers, and how does the necessity of the cMPOA for the exposure/sensitisation effect square with the effect being postsynaptic in the BSTrh?

(2) The presentation of the manuscript (clarity of language, grammar, reporting of stats in figures etc.) needs to be improved.

Reviewer #2 (Public review):

Summary:

The present study identifies how mating and pup experience are correlated with differences in inhibitory neurotransmission underlying the promotion of paternal behavior toward pups. The study builds on existing knowledge about the circuit between the medial amygdala, medial preoptic area, and the bed nucleus of stria terminalis to uncover synaptic changes correlated with behavior. The authors find that inhibition from the medial amygdala is decreased in the medial preoptic area and increased in the bed nucleus of stria terminalis to promote paternal behavior in mated males.

Strengths:

The authors use a combination of in vivo activity manipulation and slice electrophysiology to study the role of inhibition in this circuit in dynamic infant-directed behavior induced by mating.

Weaknesses:

(1) Some technical and methodological details are incomplete or missing for interpretation of the significance of the findings. Statistical details are also left out.

(2) The rationale for using Cartpt as a marker is not fully explained. This marker has activity-dependent expression and this possibility is not explored experimentally--for example, could exposure to objects or pups change expression (or the number of cells expressing) cartpt alone?

(3) The cfos experiment is quantified by exposing a male to a pup inside a tea ball. Therefore, it is unclear how the male was classified as infanticidal or parental based on the available criteria provided in the methods section.

(4) There is no information about inclusion/exclusion criteria for chemical and viral experiments. Specifically, there is no information provided about the validation of the lesion experiment--how large were the lesions? Is there concern about leakage of the chemical into the recorded region (MPOA and BNST are adjacent).

(5) The authors do not provide information about how long rAAV is allowed to express before quantifying retrograde transport.

(6) For statistics, the authors do not provide information distinguishing the main effects from multiple comparisons post hoc testing for the ANOVA analyses.

Reviewer #3 (Public review):

Ito et al. investigate the role of synaptic plasticity in the medial preoptic area (MPOA) pathway of male mice and its involvement in transitions from infanticidal aggression to parental behavior. Using optogenetics, whole-cell patch-clamp recordings, and behavioral assays, they demonstrate that inhibitory synaptic transmission from the posterior-dorsal medial amygdala (MePD) to the central MPOA (cMPOA) decreases following mating and cohabitation with pregnant females. This synaptic disinhibition is correlated with a reduction in aggressive behavior toward pups. They further show that paternal experience induces enhanced inhibitory transmission in the rhomboid nucleus of the bed nucleus of the stria terminalis (BSTrh), downstream of the MPOA, through postsynaptic mechanisms. These findings suggest a circuit-based model where social experiences and mating induce synaptic changes in the Me-cMPOA-BSTrh pathway, mediating the transition to parental behavior.

The conclusions of this paper are largely supported by the data, but several methodological and conceptual aspects require clarification or additional experiments.

(1) When evaluating the Me Cartpt-expressing neuron projection to the cMPOA, the authors compared excitatory postsynaptic currents (EPSCs) and inhibitory postsynaptic currents (IPSCs). However, the standard procedure for isolating these currents is to hold the membrane potential at the reversal potential for inhibitory or excitatory currents, respectively. The authors appear not to have followed this procedure, making it unclear how EPSCs and IPSCs were calculated. This requires clarification to ensure the validity of their reported E/I balance changes.

(2) The authors chose to assess parental behavior over four consecutive days. It is unclear why this specific timeframe was selected. A justification for this choice would strengthen the interpretation of the behavioral data.

(3) The experimental design in Figure 5, where the authors lesioned the entire cMPOA to assess its role in BSTrh inhibition, presents several limitations: First, the effects on BSTrh activity could result from indirect circuit alterations rather than direct cMPOA projections. The current lesion approach cannot disentangle these possibilities. Second, the cMPOA is a heterogeneous region containing diverse neuronal subtypes. Full lesions prevent the differentiation of the roles played by distinct populations within this region. Third, lesion specificity is questionable, as some lesions extended beyond the cMPOA boundaries (Figure S5). This overextension complicates the interpretation of the results and requires tighter control.

(4) In Figure 3, the authors show that optogenetic inhibition of Me projections to the cMPOA modifies the frequency of spontaneous inhibitory postsynaptic currents (sIPSCs). However, the proposed mechanism that this modulation reflects inter-neuronal network activity within the cMPOA lacks sufficient experimental validation. Additional experiments assessing circuit-level interactions could substantiate these claims.

(5) While the paper highlights synaptic changes in the cMPOA, it does not establish a direct relationship between these changes and the social experience. How do mating and cohabitation with females impact this pathway and modulate synaptic strength? The discussion could benefit from integrating these factors into their proposed model.

Overall, the paper offers valuable insights into the neural circuitry underlying male parental behavior, particularly the synaptic dynamics of the Me-cMPOA-BSTrh pathway. However, addressing these methodological and conceptual limitations would significantly enhance the clarity and impact of the work.

eLife Assessment

This study provides valuable observations indicating that human pyramidal neurons propagate information as fast as rat pyramidal neurons despite their larger size. Convincing evidence demonstrates that this property is due to several biophysical properties of human neurons. This study will be of interest to neurophysiologists.

Reviewer #1 (Public review):

The propagation of electrical signals within neuronal circuits is tightly regulated by the physical and molecular properties of neurons. Since neurons vary in size across species, the question arises whether propagation speed also varies to compensate for it. The present article compares numerous speed-related properties in human and rat neurons. They found that the larger size of human neurons seems to be compensated by a faster propagation within dendrites but not axons of these neurons. The faster dendritic signal propagation was found to arise from wider dendritic diameters and greater conductance load in human neurons. In addition, the article provides a careful characterization of human dendrites and axons, as the field has only recently begun to characterize post-operative human cells. There are only a few studies reporting dendritic properties and these are not all consistent, hence there is added value of reporting these findings, particularly given that the characterization is condensed in a compartmental model.

Strengths

The study was performed with great care using standard techniques in slice electrophysiology (pharmacological manipulation with somatic patch-clamp) as well as some challenging ones (axonal and dendritic patch-clamp). Modeling was used to parse out the role of different features in regulating dendritic propagation speed. The finding that propagation speed varies across species is novel as previous studies did not find a large change in membrane time constant nor axonal diameters (a significant parameter affecting speed). A number of possible, yet less likely factors were carefully tested (Ih, membrane capacitance). The main features outlined here are well known to regulate speed in neuronal processes. The modeling was also carefully done to verify that the magnitude of the effects is consistent with the difference in biophysical properties. Hence, the findings appear very solid to me.

Weaknesses

The role of diameter in regulating propagation speed is well known in the axon literature.

Comment on the revised version: the authors have now made clearer that the role of diameter was well known in the manuscript.

Reviewer #2 (Public review):

Summary:

In this paper, Oláh and colleagues introduce new research data on the cellular and biophysical elements involved in transmission within the pyramidal circuits of the human neocortex. They gathered a comprehensive set of patch-clamp recordings from human and rat pyramidal neurons to compare how the temporal aspect of neuronal processing is maintained in the larger human neocortex. A range of experimental techniques have been used, including two-photon guided dual whole-cell recordings, electron microscopy, complemented by theoretical and computational methods.

The authors find that synaptically connected pyramidal neurons within the human neocortex have longer intercellular path lengths. They go on to show that the short soma to soma latencies is not due to propagation velocity along the axon but instead reflects a higher propagation speed of synaptic potentials from dendrite to soma. Next, in a series of extensive computational modeling studies focusing on the synaptic potentials, the authors show that the shorter latency may be explained by larger diameters, affecting the cable properties and resulting is relatively faster propagation of EPSPs in the human neuron. The manuscript is well-written, and the physiological experiments and in-depth theoretical steps for the simulations are clear. Whether passive cable properties of the dendrites alone are responsible for higher velocities remains to be further investigated. Based on the present data the contribution of active membrane properties cannot be excluded.

Strengths:

The authors used complex 2P-guided dual whole-cell recordings in human neurons. In combination with detailed reconstructions, these approaches represent the next steps in unravelling the information processing in human circuits.

The computational modelling and cable theory application to the experimentally constrained simulations provides an integrated view of the passive membrane properties of human neurons.

Weaknesses:

Whether the cable properties alone are the main explanation for speeding the electrical signaling in human pyramidal neurons deserves further studies.

Reviewer #3 (Public review):

Summary:

This study indicates that connections across human cortical pyramidal cells have identical latencies despite a larger mean dendritic and axonal length between somas in human cortex. A precise demonstration combining detailed electrophysiology and modeling, indicates that this property is due to faster propagation of signals in proximal human dendrites. This faster propagation is itself due to a slightly thicker dendrite, to a larger capacitive load, and to stronger hyperpolarizing currents. Hence, the biophysical properties of human pyramidal cells are adapted such that they do not compromise information transfer speed.

Strengths:

The manuscript is clear and very detailed. The authors have experimentally verified a large number of aspects that could affect propagation speed and have pinpointed the most important one. This paper provides an excellent comparision of biophysical properties between rat and human pyramidal cells. Thanks to this approach a comprehensive description of the mechanisms underlying the acceleration of propagation in human dendrite is provided.

Weaknesses:

The weaknesses I had identified have been addressed by the authors.

Author response:

The following is the authors’ response to the previous reviews.

We are grateful for the positive evaluation of the work and the critical points raised by the reviewers. We thank all reviewers for their excellent comments. We believe that these revisions have significantly improved the quality of our study.

In response to the 2nd reviewer, we apologise for the missing data, we failed to provide a P-value of the RM ANOVA post-hoc test, we are very grateful that this was brought to our attention. We have revised the RM ANOVA by using the Tukey HSD post-hoc test, which is generally recommended for pairwise comparisons as it is more robust to unequal sample sizes. The controversial statistical analysis of the overall comparison of speed differences was deleted, as were three supplementary figures (Fig. S4, Fig. S9 and S10), which are less informative in support of the manuscript.

eLife Assessment

This study is valuable as it provides information about the genes regulated by sex hormone treatment in song nuclei and other brain regions and suggests candidate genes that might induce sexual dimorphism in the zebra finch brain. The analysis presented is thorough and detailed. Whereas the evidence for gene regulation by hormone treatment is well supported, the evidence for an association of those genes with song learning (as written in the title) is incompletely supported as no manipulation of song learning or song analysis was conducted.

Reviewer #3 (Public review):

Summary:

Davenport et al have investigated how a masculinizing dose of estrogen changes the transcriptomes of several key song nuclei song and adjacent brain areas in juvenile zebra finches of both sexes. Only male zebra finches sing, learn song, and normally have a fully developed song control circuitry, so the study was aimed at further understanding how genetic and hormonal factors contribute to the dimorphism in song behavior and related brain circuitry in this species. Using WGCNA and follow-up correlations to re-analyze published transcriptome datasets, the authors provide evidence that the main variance of several identified gene co-expression modules significantly correlates with one or some of the factors examined, including sex, estrogen treatment, regional neuroanatomy, chromosomal placement, or vocal learning, noting that the latter is largely based on inference due to expression in song control nuclei.

Strengths:

Among the main strengths are the thorough gene co-expression module and correlation analyses, and the inclusion of both song nuclei and adjacent areas, the latter serving as sort of controls for areas that are not dimorphic and likely broadly present in birds in general. In situ hybridization data discussed in a previous publication (Choe et al., Hormones and Behavior, 2021) provides some support for the neuroanatomical specializations of gene expression. It is also significant that the transcriptome re-analysis was performed with an improved genome assembly that also includes the sex chromosomes, thus expanding the Z/W chromosome gene analyses in Friedrich et al, Cell Reports, 2022. The most relevant finding is arguably the identification of some modules where gene expression variation within song nuclei correlates with hormonal effects and/or gene location on sex chromosomes, which are present at different dosages between sexes. Sex differences in gene expression in areas that are not song nuclei may also bring insights into functions other than song behavior or vocal learning. The study also shows how a published RNA-seq dataset can be reanalyzed in novel and informative ways.

Weaknesses:

The validation of the inferred direction of regulation in the identified co-expression modules is limited to the in situ data mentioned above. Further evidence that representative genes in the main modules differ in expression when comparing sexes or E2- vs VEH-treated tissues using independent samples and/or methods would provide further validation and enhance rigor. Most importantly, E2 is known to exert various actions on brain physiology and neuronal function. Because there was no manipulation of candidate genes, nor assessment/manipulation of vocal behavior or vocal learning, an involvement of the identified candidate genes in setting up the sexual dimorphism of the song system or song behavior was not directly tested in this study. For the latter reason, the implication of the Title (..."gene expression associated with vocal learning...") is not well supported. While novel insights were gained into brain expression of Z chromosome genes, it cannot be excluded that the higher male expression of some Z genes may not affect brain cell function and thus may not require active compensation (as discussed for nucleus RA in Friedrich et al, Cell Reports, 2022).

Author response:

The following is the authors’ response to the original reviews.

eLife Assessment

This study is useful as it provides further analysis of previously published data to address which specific genes are part of the masculinizing actions of E2 on female zebra finches, and where these key genes are expressed in the brain. However the data supporting the conclusion of masculinizing the song system are incomplete as the current manuscript is a re-analysis of differential gene expression modulated by E2 treatment between male/female zebra finches without manipulation of gene expression. The conclusions (and title) regarding song learning are also incompletely supported with no gene manipulation or song analysis. Importantly, the use of WGCNA for a question of sex-chromosome expression in species without dosage compensation is considered inadequate. As the experimental design did not include groups to directly test for song learning, and there was also no analysis of song performance, these data were also considered inadequate in that regard.

We are sorry the editor felt the manuscript so incomplete and inadequate. Though the tone of this assessment seems more severe than the below reviewer comments, we are also happy to see that the editor has considered our paper further for a revised publication, based on the reviewer’s comments. We address the editor’s comments as follows:

While we agree that manipulation of some of the genes we discovered, whose expression levels are E2-sensitive in the song system, would take the study further in validating some proposed hypothesis in the discussion of the paper, we don’t think the outcome of gene manipulations would change the major conclusions from the results of the paper. In this study we performed estrogen hormone manipulations, with causal consequences on gene expression in song nuclei and associated song behavior. In a way this is analogous to gene manipulations, but manipulating directly the action of estrogen. The categories of genes impacted, and the differences among the sex chromosomes wouldn’t change.

For the comment on WGCNA being inadequate for addressing questions on sex chromosome expression in species without dosage compensation, we think the evidence in our data does not bear that out. One main result of this paper is the separation of Z chromosome transcripts whose expression is most strongly regulated by chromosomal dosage (WGCNA module E) across regions from those subject to additional sources of regulation in song nuclei (other modules). It seems to us that rather than being confounded by the lack of dosage compensation, WGCNA allowed us to better resolve the effects of dosage on different genes within the sex chromosomes. We have added a new figure more directly examining sex chromosome transcript abundance within different modules. Briefly, we found that module E assigned Z chromosome genes exhibited almost exactly the male-biased expression ratio expected from no dosage compensation while the Z chromosome genes in song nuclei assigned to other modules were expressed below the dosage predicted value, consistent with module E containing those genes whose expression are most strongly regulated by dose across all brain regions sampled.

At its core, WGCNA finds sets of correlated genes. The biological reality of the zebra finch transcriptome is that Z chromosome expression is largely anti-correlated with W chromosome due to dosage. However, this dosage effect is not felt equally by all genes and WGCNA provides an unbiased computational framework which can be used to separate dose from other potential sources of gene regulation. This is why roughly ⅓ of Z chromosome genes are not assigned to module E; for example the growth hormone receptor is assigned to module G based on its correlation with genes upregulated within HVC.

“As the experimental design did not include groups to directly test for song learning, and there was also no analysis of song performance, these data were also considered inadequate in that regard.”

Concerning the comment on no analysis on song performance in the paper, all such analyses were conducted on our previous study on the same animals (Choe et al. 2021, Hormones & Behavior). The birds considered here were sacrificed at PHD30, prior to the onset of learned song behavior. However, females treated with E2 the same at the same time and allowed to mature into adulthood, went onto to develop rudimentary song. Further, induction of rudimentary song learning in females following E2 treatment has been well established since the early ‘80s. We have added the following text toward the end of the intro to make this more clear:

“While the birds for this study were sacrificed prior to the developmental presentation of song behavior, we have previously shown that female finches treated in exactly the say way with E2 go on to produce rudimentary imitative songs as adults (Choe et al 2021), consistent with the known induction of vocal learning in females by E2 (REF).”

Reviewer #1 (Recommendations For The Authors):

Overall, this is a wonderfully designed and executed study that takes full advantage of new resources, such as the most complete zebra finch genome assembly yet, as well as the latest methods. I have very few suggestions as to the improvement of the manuscript. They are as follows:

Results Section:

In the paragraph "Identification of gene expression modules in song nuclei":

"The E2-treated females in this study had similarly sized song system nuclei as males, indicating that E2 treatment prevented atrophy."

Clarify if this comparison is to treated and/or untreated males.

We thank the reviewer for their comment. The relative differences in the song nuclei sizes between the E2-treated females and the other groups is more complex that our original sentence implied. We have revised the main the text as follows

“In our previous study, we found that estradiol treatment in PHD30 females caused HVC to enlarge and Area X to appear when it normally does not develop in females, but both at sizes less than in untreated or treated males.The sizes of PHD30 female LMAN RA were already the sizes as seen in males, as the later has not atrophied yet at this age(25).”

In the paragraph "Sex- and micro-chromosome gene expression across the telencephalon": "These animal and chromosome specific shifts in the transcriptomes could represent the systemic effects of allelic chromosomal structural variation..."

The authors should clarify the meaning of a"llelic chromosomal structural variation" in this context, as it is an unusual phrase. Major chromosomal structural variation seems unlikely to produce these effects. Is it also possible that animal-specific modules with brain-wide higher could also result from laboratory contamination between all samples from one animal? This is not too likely but perhaps should be acknowledged or ruled out.

We have removed the word allelic, which was unnecessary. We can’t envision how laboratory contamination could occur such that all of one animal’s samples would be affected to produce the observed result which is module and chromosome specific. An animal wide effect could emerge during sacrifice, but we can think of no reason that would affect these modules and not others. Rather, the most likely explanation is biological natural difference between animals. We have added this consideration of alternative explanations.

In the section "Candidate gene drivers of HVC specialization in E2-treated females":

When discussing GHR's role in cell growth and proliferation, the authors' argument could be expanded by including the documented role of GH signaling in anti-apoptotic protection of neurons from rounds of neural pruning during development as documented in the chicken, e.g. • Harvey S, Baudet M-L, Sanders EJ. 2009. Growth Hormone-induced Neuroprotection in the Neural Retina during Chick Embryogenesis. Annals of the New York Academy of Sciences, 1163: 414-416. https://doi.org/10.1111/j.1749-6632.2008.03641.x

We thank the reviewer for sharing this publication with us.. We have added the following sentence to our discussion with the above citation. “Further, our results are consistent with growth hormone’s known role in avian anti-apoptotic protection, with elevated signaling associated with the survival of chicken neurons during rounds of pruning in the developing

retina.”

The authors' argument of the relevance of the passerine GH duplication would be strengthened by citing:

• Rasband SA, Bolton PE, Fang Q, Johnson PLF, Braun MJ. 2023. Evolution of the Growth Hormone Gene Duplication in Passerine Birds, Genome Biol Evol, 15(3) https://doi.org/10.1093/gbe/evad033. Greatly expands on the Yuri et al. paper cited by characterizing of the molecular evolution of these genes across hundreds of avian species, supporting positive selection on multiple amino acid sites identified in both ancestral and duplicate (passerine) growth hormone.

• Xie F, London SE, Southey BR et al. 2010. The zebra finch neuropeptidome: prediction, detection and expression. BMC Biol 8, 28. https://doi.org/10.1186/1741-7007-8-28 The authors report significantly different expression of the ancestral GH gene in the adult male zebra finch auditory forebrain after different song exposure experiences.

We have amended the results section sentence and added all suggested citations. The sentence now reads: “The gene which encodes growth hormone receptor’s ligand, growth hormone, is interestingly duplicated and undergoing accelerated evolution in the genomes of songbirds (Rasband et al 2023); the GH ligand has been found to be upregulated in the zebra finch auditory forebrain following the presentation of familiar song (Xie et al 2010).”

Figures:

- Figure 1B. "Duration of sex typing" being a shorter bar compared to the others is not fully explained in the experimental design. Presumably at the end of this time period, the sex is non-invasively, phenotypically evident. I suggest an arrow pointing to the PHD/PHD range when sex is apparent in plumage/anatomy.

- Figure 4. Caption appears to be truncated; "across all... genes"?

Fixed

- Figure 5. For 5E, 5F, 5G, 5H, consider enlarging the plots so overlapping gene symbols are readable. Alternately, smaller numbers or symbols could be used with a key in areas where overlapping symbols are hard to prevent.

We agree that these are not the easiest to read; we originally offset the symbols in R to minimize overlaps, but it can only do so much for the more crammed panels. We have now added a supplemental .xlsx file with the underlying data from each of the 4 tests for readers that want to examine the data in more detail.

Reviewer #2 (Recommendations For The Authors):

Since WGCNA methods will inherently draw together sex-chromosome genes into the same module in systems without dosage compensation, I suggest the authors rerun the WGCNA using only female samples and only male samples. Then identify the composition of modules that differ between E2 and vehicle-treated females and compare these genes to males. Then from male WGCNA identify the composition of modules that differ between E2 and vehicle-treated males and compare to female modules.

We thank the reviewer for their suggestions. However, we believe it is not as strong as the approach we used, which is grouping data from both sexes in the WGCNA analyses in a study that is looking for sex differences. The reviewer's proposed approach amounts to computing modules twice (once per sex), determining song system specialized modules and E2 responsive modules in both settings, then intersecting the two sets to find corresponding modules, all done to prevent the non-dose compensated sex chromosome genes from being drawn into the same module.

While WGCNA does group the majority of sex chromosome genes into module E, it does not categorize them all this way (Fig 3). The module classification instead differentiates those sex chromosome genes whose expression are most explained by chromosome dosage / sex across regions (modE) from those whose expression is controlled by other sources of regulation; for an example of the latter, the growth hormone receptor (GHR) is one of several Z chromosome genes classified into modG as its expression better correlates with the genes specialized to HVC than it does with the majority of dosage-dependent Z chromosome genes found in modE. Further, to remove biological sex as a variable in a WGCNA analysis that is focused on sex differences seems counterintuitive.

Instead, to quantitatively address the reviewer’s concern, we conducted additional analyses, that led to an added new figure, associated text, and tables, that better describes sex/chromosome dosage effects on the abundance (FPKM) and expression ratios of sex chromosome transcripts by module irrespective of brain region (Fig. 5). We find that the Z chromosome genes in modE were expressed at the expected chromosome dosage in the non-vocal surrounding regions (65.06% observed vs 66.6% expected) while in other modules, other Z chromosome genes were expressed at intermediate levels between equal expression and the expected chromosomal dosage. For example, the Z chromosome content of modules D and H exhibited near equal expression between sexes. Within the song system, Z chromosome gene content of modG was highly expressed in males beyond what is expected from chromosome dosage, consistent with modG’s male-specific upregulation in song nuclei relative to surrounds in the absence of E2. These results better demonstrate that in our WGCNA on the combined dataset we are able to separate those Z chromosome genes whose expression is predominantly dosage controlled from those subject to additional regulation such as song system specialization.

Fig. S3 Legend: 'Black arrow' -> 'Red arrow'

Change made.

Fig. S5 - What part of the figure shows the 'human convergent signature'? Also, simply listing the number of genes mapped to a chromosome is misleading to readers unfamiliar with the zebra finch genome, you should either provide the number of genes on each chromosome or present as corrected by that number.

Fig. S5 was the same type of analyses in Fig. 3 but with an older zebra finch genome assembly, where we had not included the panel a for enrichments with genes convergent in expression between songbird song regions and humans speech brain regions. However, we see that Fig. S5 was not adding any new important information to the paper, so we removed it.

For the chromosome analyses in Fig. 3b, we provide both the total raw number of module assigned genes broken down by chromosome (The black bar plots on the right) as well as a statistical fold-enrichment value of modules per chromosome. Given the number of genes per chromosome and genes per module in our data, we computed the fold-enrichment for each intersection (observed intersection size / expected intersection size). To test for the significance of these enrichments, we bootstrapped FDR corrected p values for the enrichment of each chromosome-module pairing by randomizing the mapping of genes to modules to construct a null distribution of fold enrichments for each intersection. Our intent was not to describe the size of the chromosomes themselves, information readily available elsewhere, but to show the disproportionate chromosomal origins of the gene sets considered by this study. Performing this enrichment test using all annotated genes per chromosome would artificially increase enrichment values and make the analysis less conservative by confounding the results with the inherent enrichment for “brain function” in the assigned genes relative to all genes.

At several places you say "we correlated expression of each sex chromosome transcript with sexual dimorphism within each region, such that expressed W genes would be positively correlated and depleted Z chromosome genes would be anticorrelated." What was the sexual dimorphism that was being correlated with? Is this the eigengene?

We thank you for this comment. Our language was less clear than it could be. We tested for correlations of both the eigengene and the individual gene expression profiles with the biological sex of the animals. We have changed the text to:

“To do this, we tested for a correlation between the expression of each sex chromosome transcript to the animals’ sex within each brain region. We found that female-enriched transcripts were positively correlated with sex and male-enriched transcripts were anticorrelated (Fig. 4f,g).”

Fig. 4A: The 'true/false' boxes and animal A-L is confusing and unnecessary. I'd suggest just using M and F (or sex symbols) with a horizontal line below each set of 3 for respective E2 and Veh.

Change made.

Reviewer #3 (Recommendations For The Authors):

General comments:

After the initial characterization of the datasets and module identification, it is quite hard to follow the logic of the data presentation in the various other Results sections or to clearly understand how they relate to the main stated goal to identify factors related to sex differences in vocal learning. The most relevant findings relate to the presumed actions of hormone treatment and sex chromosome gene dosage in song nuclei, whereas analyses of other brain areas, other chromosomes, or speech-related genes serve more as controls and/or appear as distractions from the main theme. A suggestion to increase the clarity of the presentation and potential impact of the study is to change the order of the presentation, focusing first on the specific analyses and comparisons that most directly speak to the main goals of the study, and then secondarily and more briefly presenting the controls or less related comparisons.

The reviewer’s suggestion for the results section organization is exactly what we had tried to do. We opened the first paragraph on identification of modules, then presented the song nuclei specific modules, followed by E2-changes to those modules; and the followed by other specific results for the remainder of the paper, including module enrichments to specific chromosomes. The reviewer mentioned our analyses of “other brain areas” (which we assume to mean the non-vocal surround regions), other chromosomes (which we assume means autosomes) and speech-related genes as controls were a distraction in the paper; but within our analysis, these other brain regions are essential controls needed to assess the song-system specificity of any observed sex differences observed from the very first paragraphs of the results; the autosomes were not controls for sex chromosome results, but primary results in of themselves; the overlap with speech-related genes was also not a control, but a novel discovery. We have revised these points in the paper to make them clearer, and revised some of the section titles and transitions between sections to help increase clarity of the main storyline of the paper.

A related comment is that many of the inferences drawn from the WGCNA analysis were quite complex, thus independent verification of some predictions would be quite valuable. For example, consider the passage: "In non-vocal learning juvenile females, interestingly LMAN was specialized relative to the AN by the same gene modules as in males (B, F, and I) as well as an additional module G (Fig. 2b); RA was specialized by module A as in males, but not module L and by additional modules A and G. In contrast, neither juvenile female HVC nor Area X exhibited significant gene module expression specializations relative to their surrounds." Providing in situ hybridization verification of these regional gene expression predictions with a few representative genes seems quite feasible given the group's expertise and would considerably strengthen confidence in the module-based inferences.

We performed in-situ independent validation of 36 candidate genes in our first study with this dataset (Choe et al 2021). We now mention this validation in the revised paper. The reviewer’s selection of one of our sentences though made us realize that our grammar used to explain the results was not as clear as it needs to be. We thus cleaned up the grammar of our module descriptions so that it should be communicated with less complexity, the main issue noted by the reviewer.

Because this is a re-analysis of a previously published dataset, the authors should more explicitly describe somewhere in the Discussion how the present analysis advances the understanding of sex differences in songbird neuroanatomy and behavior beyond the previous analysis.

We have added an additional sentence into the discussion more clearly separating the results of the current study from our previous work.

Specific comments:

Abstract:

There is evidence (from Frank Johnson's lab) that RA does not completely atrophy in female zebra finches, but is still present with more preserved connectivity than previously thought, possibly related to non-singing function(s). A term like 'marked reduction' of female RA may more accurately reflect the current state of knowledge.

We have changed the text to “partial atrophy”.

The term "driver" is undefined and unclear at this point of the paper; a clear definition for "driver" is also lacking in the Intro.

We now define “driver” or “genetic driver” as understood to mean “a genetic locus whose expression and/or inheritance strongly regulates the trait of interest”.

When citing the literature on studies that identified "specific genes with specialized up- or down-regulated expression in song and speech circuits relative to the surrounding motor control circuits", the authors should also cite studies from other labs (e.g. Li et al., PNAS, 2007; Lovell et al, Plos One 2008; Lovell et al, BMC Genomics 2018; Nevue et al, Sci Rep. 2020), to be accurate and fair.

Citations added

For clarity, the authors should explicitly formulate the hypothesis they are proposing at the end of the Summary.

We thank the reviewer for this comment. We have replaced the final sentence of the summary with: “We present a hypothesis where reduced dosage and expression of these Z chromosome genes changes the developmental trajectory of female HVC, partially preventable by estrogen treatment, contributing to the loss of song learning behavior.”

Introduction:

Vocal learning is arguably the ability to imitate 'vocal' sounds, this could be clarified here.

We have amended the sentence to “Vocal learning is the ability to imitate heard sounds using a vocal organ…”

Given they are currently considered sister taxa, can the author briefly explain what is the basis for assuming that songbirds and parrots independently evolved vocal learning?

Although songbirds and parrots belong to a monophyletic clade, they are not sister taxa. There are two clades separating them that are vocal non-learners. We have cited the reference that demonstrated this (e.g. Jarvis et al 2014 Science).

Why use Taeniopygia castanotis rather than the more broadly used Taeniopygia guttata?

Zebra finches were recently reclassified and T.castanotis is now more accurate. The Indonesian Timor zebra finch retained T.guttata while the Australian finch, used here, was classified as T.castanotis.

The authors state: "...vocal learning is strongly sexually dimorphic in zebra finches and many other vocal learning species" and cite Nottebohm and Arnold, Science, 1978. That landmark paper only shows dimorphism in song nuclei (not learning) in two songbird species. The authors should provide citations for other species and behavior, or modify the statement.

We have added an additional citation (Odom et al.) to this sentence which covers the phylogeny more broadly.

The authors refer to the nucleus RA as being located in the lateral intermediate arcopallium (LAI). Other labs have described this domain as the dorsal part of the intermediate arcopallium, thus AId or AID (Mello et al., JCN, 2019; Yuan and Bottjer, J Neurophys 2019; Yuan and Bottjer, eNeuro, 2020; Nevue et al., BCM Genomics, 2020). The authors should acknowledge this discrepancy in nomenclature so that data and conclusions can be more readily compared across studies.

We thank the reviewer and agree that this is helpful. We have added a note at the first mention of LAI.

The authors state that data from the gynandromorph bird described by Agate et al implicates "sex chromosome gene expression within the song system" as involved in the song system sexual dimorphism. That study, however, only rules out circulating gonadal steroids, and while suggesting a cell-autonomous mechanism like sex chromosome genes, it does not necessarily exclude other brain-autonomous factors like sex differences in local production of sex steroids.

We say that this study “implicated” sex chromosome gene expression, which is accurate per the results and discussion of that study. We are unsure what “brain autonomous factors like sex differences in local production of sex steroids” means?. “Brain autonomous” and “local production” in the brain seem contradictory in this context?

Results:

The authors state that "the E2-treated females in this study had similarly sized song system nuclei as males, indicating that E2 treatment prevented atrophy". Can they clarify whether the VEH-treated females actually had smaller RAs than E2-treated females or VEH-treated males at this age? This is still quite early in development and it is unclear to what extent RA's marked sexual dimorphism in adults or later developmental ages has already taken place in untreated (or VEH-treated) birds. A related comment is that the authors state later on: "We interpret these findings to indicate that: LMAN and RA atrophy later in juvenile female development..." Does this mean these nuclei actually did not show the marked decreases predicted earlier in the text? Clarifying this point would be helpful.

We thank the reviewer for pointing out this discrepancy, which reviewer #1 asked for clarification as well. RA size at this age is similar in males and females. However, HVC and Area X is smaller and absent respectively in females and E2 treatment partially prevents this atrophy. The text now reads:

“In our previous study, we found that estradiol treatment in PHD30 females caused HVC to enlarge and Area X to appear when it normally does not develop in females, but both at sizes less than in untreated or treated males.The sizes of PHD30 female LMAN RA were already the sizes as seen in males, as the later has not atrophied yet at this age(25).”

The authors acknowledge that area X is absent in untreated and VEH-treated females. Could they please clarify how area X and the surrounding stratal tissue that excludes area X were identified for laser capture dissections in juvenile females?

We have added the following statement to the main text portion discussing the dissections.

“In the case of vehicle-treated females which lack Area X, a piece of striatum from the same location of where Area X is found in males was taken. “

Some passages in Results discussing the authors' interpretation of the modules seem quite speculative and possibly belong instead in the Discussion. For example: "... that module A and G genes could be associated with the start of this atrophy; HVC and Area X are likely the first to atrophy or not develop; and lack of any gene module specialization in them at this age could mean that they would be more sensitive to estrogen prevention of vocal learning loss."

As suggested, we have removed this text from the results; these ideas were already presented in the Discussion. We have merged the resulting small paragraph with the preceding paragraph.

The authors state: "To assess the effects of chronic exogenous estrogen on the developing song system, we first performed a control analysis of modules in the E2-treated juvenile males." How can an assessment of estrogen effects be a "control" analysis? Does this refer to a contrast with females? Please clarify the language here.

The reviewer is correct, that E2 treatment in males should not be considered a control experiment. We removed the word “control”.

When discussing the GO-enriched terms for module G, it is unclear how the authors reached the conclusion about "proliferative", as the enriched terms do not refer to processes more directly indicative of proliferation like "cell division" or "cell cycle regulation". Rather, these terms seem more related to differentiation and growth, which do not necessarily imply proliferation. The authors also refer to "HVC proliferation" later on in the Discussion. However, there is conclusive evidence from several labs that proliferative events associated with postnatal neuronal addition and/or replacement in song nuclei occur in the subventricular zone, not in song nuclei like HVC itself, and that the growth of song nuclei largely reflects cell survival, as well as growth in size and complexity under the regulation of sex steroids.

We agree that “proliferative” may have been a poor word choice here. We did not mean to indicate that cell division was occuring in HVC itself. Instead we meant to indicate that HVC is able to accommodate the new born neurons from the SVZ. We have replaced the word “proliferative” throughout. In the instance the reviewer mentions specifically we replaced it with,“...potentially act to integrate and differentiate late born neurons.”

With regard to module E, referring to a telencephalon-wide sexually dimorphic gene expression program seems quite a stretch, given that only a few regions were sampled and compared between sexes. These related statements should be toned down.

We have replaced “telencephalon-wide” with “more distributed across the finch telencephalon” and other similar language in each instance.

The following passage is very speculative and should shortened and/or moved to the Discussion: "Based on the findings in these gene sets, we hypothesize that without excess estrogen in females, HVC expansion is prevented by not specializing the growth and neuronal migration promoting genes in module G to the HVC lineage by late development. This is potentially enacted by depleting necessary gene products from the Z sex chromosome, such as GHR, which are already present in only one copy."

We have deleted this portion of the text, as the idea is already present in the discussion.

Figure 5: To this reviewer, the comparisons of sex differences and of female response to E2 are the most relevant and informative ones, whereas the regional differences between song nuclei and surrounds refer to different cell populations and cell types where other processes may be occurring, independently of what occurs in song nuclei. It thus seems like the intersection analysis in panel 5i may be subtracting out important "core genes" in terms of E2 effects and/or sex differences in the most relevant cell populations, i.e. in this case within song nucleus HVC.

Song learning and the vocal learning brain regions are specialized behaviors and associated nuclei which have a set of hundreds of specialized genes compared to the surrounds. Our previous findings shows that E2 drives the appearance of these specializations in female zebra finches. Thus, we considered this the most interesting question to focus on, which we have further highlighted. Nevertheless, in response to the reviewers suggestion, we have added a .xlsx supplemental file containing the results from each of the individual tests so readers may examine any single comparison, or set of comparisons, in more detail.

Discussion:

It is unclear what the term "critical period" refers to in: "during the critical period of atrophy for the female vocal circuit"; please clarify.

We agree that our language was nebulous. We have replaced it with “as several male song control nuclei begin to expand and female nuclei partially atrophy”

In: "HVC appeared unspecialized at the level of gene module expression in control females", does "unspecialized" refer to a lack of difference in gene expression when compared to surroundings? Please clarify. The same comment applies to other uses of "unspecialized" in this paragraph.

Yes, unspecialized means lack of difference in gene expression in the song nucleus. To clarify this point, we have reworked that and the following sentence as follows:

“HVC appeared unspecialized compared to the surrounding nidopallium at the level of gene module expression in control females, with no significantly differentially expressed MEGs . However, in E2-treated females, HVC exhibited a subset of the observed male HVC gene expression specializations. Similarly, the vehicle-treated female striatum located where Area X would be also lacked any specialized gene module expression, but the E2-treated female Area X exhibited a subset of the male Area X specializations, consistent with the known absence of Area X in vehicle-treated females and presence in E2-treated females.”

The authors state: "...we surprisingly found that the most specialized genes were disproportionately from the Z chromosome", when discussing module G in HVC. Why is this so surprising? In a sense, this could be taken as consistent with the findings of Friedrich et al, 2022, where sex differences in the RA transcriptome were predominantly Z related on 20 dph. Arguably 20 dph is still quite close to 30 dph in the present study, when compared to 50 dph in Friedrich et al, when autosomes predominate.

Our bioRxiv was originally posted in July 2021, prior to the publication of Friedrich et al, 2022; however we had previously added to our discussion that several of our results are consistent with the observations of Friedrich et al..

We have a different interpretation of Z chromosome gene results in Friedrich et al.. While the percentage of specialized genes from the Z chromosome decreased, the absolute number of specialized Z chromosome genes actually increased over this interval. In Fig. 3a from Friedrich et al. it appears that ~28% of Z chromosome genes were sexually dimorphic in their expression in RA at PHD20 but that ~39% of Z chromosome genes were similarly dimorphic at PHD50. We interpret this result as the Z chromosome genes being among the earliest genes differentially expressed between the sexes, not that their differential expression or role ever subsequently decreased. We have reworked this portion of the discussion to make our point more clear:

“This model of sex chromosome influenced song system development is consistent with recent observations comparing male and female zebra finch transcriptomes from RA at young juvenile (PHD20) and young adult (PHD50) ages in un-manipulated birds (Friedrich et al. 2022)57. While that study proposes that the role of the sex chromosome in maintaining transcriptomic sex differences diminishes across development, as the proportion of specialized genes that originate on the sex chromosomes diminishes, this effect was driven by large increases in differentially expressed autosomal genes rather than by any reduction in sex chromosome dimorphism; the percentage of differentially expressed Z chromosome genes increased from PHD20 (28%) to PHD50 (39%) (Friedrich et al). This leads us to conclude that sexually dimorphic Z chromosome expression at juvenile ages precedes the sexually dimorphic expression of the autosomes seen in adults. This is consistent with our hypothesis that sufficient expression of select Z chromosome gene products (GHR, etc..) is necessary for subsequent autosomal song system specializations (modG).”

Further, when we write ”When examining the module G HVC specialization induced by E2-treatment in female HVC, we surprisingly found that the most specialized genes were disproportionately from the Z chromosome” we are referring to the upregulation of module G by E2 in female HVC, not the sex difference described in RA by Friedrich et al. which only utilized un-treated RA samples and thus is more likely related to our observations of module E.

The term "sexual dimorphism" has been more traditionally used for sex differences that are very marked, like features that are highly regressed or absent in one sex, most often in females. Quantitative differences in gene expression, including dosage differences like those related to module E, are more appropriately described as sex differences rather than dimorphisms. That usage would be more consistent with most of the literature, and thus preferable.

We did a google search for common definitions, and found more the opposite. Sexual dimorphism being used more often as differences of degree (with the zebra finch example as one of the top hits), and sex differences being used often as more absolute differences (like presence vs absence of the Y chromosome). Further, as in the reviewer’s first sentence, the definition of sexual dimorphism is a sex difference. That is, the two phrases can be interchangeable. Thus, we prefer to keep sexual dimorphism.

Several references are incomplete or seem truncated, like 9 and 10.

Fixed

Table S2: Please examine and take into account the W gene curation presented in Table S3 of Friedrich et al., 2022.

We have added additional supplementals (supplemetal_w_chrom_express.csv and supplemetal_z_chrom_express.csv) of the data provided in new Fig 5 incorporating the curation information from Table S3 from Friedrich et al.

Data availability:

Genes for all the main modules identified should be presented in a Supplemental Table, or through a link to a stable data repository.

We have added an additional Supplemental Table supplemental_gene_module_assignment.csv with this information.

eLife Assessment

This valuable paper introduces Heron, lightweight scientific software that is designed to streamline the implementation of complex experimental pipelines. The software is tailored for workflows that require coordinating many logical steps across interconnected hardware components with heterogeneous computing environments. The authors convincingly demonstrate Heron's utility and effectiveness in the context of behavioral experiments, addressing a growing need among experimentalists for flexible and scalable solutions that accommodate diverse and evolving hardware requirements.

Reviewer #2 (Public review):

Summary:

The authors provide an open-source graphic user interface (GUI) called Heron, implemented in Python, that is designed to help experimentalists to:

(1) Design experimental pipelines and implement them in a way that is closely aligned with their mental schemata of the experiments<br /> (2) Execute and control the experimental pipelines with numerous interconnected hardware and software on a network.

The former is achieved by representing an experimental pipeline using a Knowledge Graph and visually representing this graph in the GUI. The latter is accomplished by using an actor model to govern the interaction among interconnected nodes through messaging, implemented using ZeroMQ. The nodes themselves execute user-supplied code in, but not limited to, Python.

Using three showcases of behavioral experiments on rats, the authors highlighted four benefits of their software design:

(1) The knowledge graph serves as a self-documentation of the logic of the experiment, enhancing the readability and reproducibility of the experiment,<br /> (2) The experiment can be executed in a distributed fashion across multiple machines that each has different operating system or computing environment, such that the experiment can take advantage of hardware that sometimes can only work on a specific computer/OS, a commonly seen issue nowadays,<br /> (3) The users supply their own Python code for node execution that is supposed to be more friendly to those who do not have a strong programming background,<br /> (4) The GUI can also be used as an experiment control panel for users to control/update parameters on the fly.

Strengths:

(1) The software is light-weight and open-source, provides a clean and easy-to-use GUI,<br /> (2) The software answers the need of experimentalists, particularly in the field of behavioral science, to deal with the diversity of hardware that becomes restricted to run on dedicated systems. It can also be widely adopted in many other experimental settings.<br /> (3) The software has a solid design that seems to be functionally reliable and useful under many conditions, demonstrated by a number of sophisticated experimental setups.<br /> (4) The software is well documented. The authors pay special attention to documenting the usage of the software and setting up experiments using this software.

Comments on revisions: The authors have addressed my concerns from the initial review.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews

Reviewer #1 (Public Review):

Summary:

The authors have created a system for designing and running experimental pipelines to control and coordinate different programs and devices during an experiment, called Heron. Heron is based around a graphical tool for creating a Knowledge Graph made up of nodes connected by edges, with each node representing a separate Python script, and each edge being a communication pathway connecting a specific output from one node to an iput on another. Each node also has parameters that can be set by the user during setup and runtime, and all of this behavior is concisely specified in the code that defines each node. This tool tries to marry the ease of use, clarity, and selfdocumentation of a purely graphical system like Bonsai with the flexibility and power of a purely code-based system like Robot Operating System (ROS).

Strengths:

The underlying idea behind Heron, of combining a graphical design and execution tool with nodes that are made as straightforward Python scripts seems like a great way to get the relative strengths of each approach. The graphical design side is clear, selfexplanatory, and self-documenting, as described in the paper. The underlying code for each node tends to also be relatively simple and straightforward, with a lot of the complex communication architecture successfully abstracted away from the user. This makes it easy to develop new nodes, without needing to understand the underlying communications between them. The authors also provide useful and well-documented templates for each type of node to further facilitate this process. Overall this seems like it could be a great tool for designing and running a wide variety of experiments, without requiring too much advanced technical knowledge from the users.

The system was relatively easy to download and get running, following the directions and already has a significant amount of documentation available to explain how to use it and expand its capabilities. Heron has also been built from the ground up to easily incorporate nodes stored in separate Git repositories and to thus become a large community-driven platform, with different nodes written and shared by different groups. This gives Heron a wide scope for future utility and usefulness, as more groups use it, write new nodes, and share them with the community. With any system of this sort, the overall strength of the system is thus somewhat dependent on how widely it is used and contributed to, but the authors did a good job of making this easy and accessible for people who are interested. I could certainly see Heron growing into a versatile and popular system for designing and running many types of experiments.

Weaknesses:

(1) The number one thing that was missing from the paper was any kind of quantification of the performance of Heron in different circumstances. Several useful and illustrative examples were discussed in depth to show the strengths and flexibility of Heron, but there was no discussion or quantification of performance, timing, or latency for any of these examples. These seem like very important metrics to measure and discuss when creating a new experimental system.

Heron is practically a thin layer of obfuscation of signal passing across processes. Given its design approach it is up to the code of each Node to deal with issues of timing, synching and latency and thus up to each user to make sure the Nodes they author fulfil their experimental requirements. Having said that, Heron provides a large number of tools to allow users to optimise the generated Knowledge Graphs for their use cases. To showcase these tools, we have expanded on the third experimental example in the paper with three extra sections, two of which relate to Heron’s performance and synching capabilities. One is focusing on Heron’s CPU load requirements (and existing Heron tools to keep those at acceptable limits) and another focusing on post experiment synchronisation of all the different data sets a multi Node experiment generates.   

(2) After downloading and running Heron with some basic test Nodes, I noticed that many of the nodes were each using a full CPU core on their own. Given that this basic test experiment was just waiting for a keypress, triggering a random number generator, and displaying the result, I was quite surprised to see over 50% of my 8-core CPU fully utilized. I don’t think that Heron needs to be perfectly efficient to accomplish its intended purpose, but I do think that some level of efficiency is required. Some optimization of the codebase should be done so that basic tests like this can run with minimal CPU utilization. This would then inspire confidence that Heron could deal with a real experiment that was significantly more complex without running out of CPU power and thus slowing down.

The original Heron allowed the OS to choose how to manage resources over the required process. We were aware that this could lead to significant use of CPU time, as well as occasionally significant drop of packets (which was dependent on the OS and its configuration). This drop happened mainly when the Node was running a secondary process (like in the Unity game process in the 3rd example). To mitigate these problems, we have now implemented a feature allowing the user to choose the CPU that each Node’s worker function runs on as well as any extra processes the worker process initialises. This is accessible from the Saving secondary window of the node. This stops the OS from swapping processes between CPUs and eliminates the dropping of packages due to the OS behaviour. It also significantly reduces the utilised CPU time. To showcase this, we initially run the simple example mentioned by the reviewer. The computer running only background services was using 8% of CPU (8 cores). With Heron GUI running but with no active Graph, the CPU usage went to 15%. With the Graph running and Heron’s processes running on OS attributed CPU cores, the total CPU was at 65% (so very close to the reviewer’s 50%). By choosing a different CPU core for each of the three worker processes the CPU went down to 47% and finally when all processes were forced to run on the same CPU core the CPU load dropped to 30%.  So, Heron in its current implementation running its GUI and 3 Nodes takes 22% of CPU load. This is still not ideal but is a consequence of the overhead of running multiple processes vs multiple threads. We believe that, given Heron’s latest optimisation, offering more control of system management to the user, the benefits of multi process applications outweigh this hit in system resources. 

We have also increased the scope of the third example we provide in the paper and there we describe in detail how a full-scale experiment with 15 Nodes (which is the upper limit of number of Nodes usually required in most experiments) impacts CPU load. 

Finally, we have added on Heron’s roadmap projects extra tasks focusing only on optimisation (profiling and using Numba for the time critical parts of the Heron code).

(3) I was also surprised to see that, despite being meant specifically to run on and connect diverse types of computer operating systems and being written purely in Python, the Heron Editor and GUI must be run on Windows. This seems like an unfortunate and unnecessary restriction, and it would be great to see the codebase adjusted to make it fully crossplatform-compatible.

This point was also mentioned by reviewer 2. This was a mistake on our part and has now been corrected in the paper. Heron (GUI and underlying communication functionality) can run on any machine that the underlying python libraries run, which is Windows, Linux (both for x86 and Arm architectures) and MacOS. We have tested it on Windows (10 and 11, both x64), Linux PC (Ubuntu 20.04.6, x64) and Raspberry Pi 4 (Debian GNU/Linux 12 (bookworm), aarch64). The Windows and Linux versions of Heron have undergone extensive debugging and all of the available Nodes (that are not OS specific) run on those two systems. We are in the process of debugging the Nodes’ functionality for RasPi. The MacOS version, although functional requires further work to make sure all of the basic Nodes are functional (which is not the case at the moment). We have also updated our manuscript (Multiple machines, operating systems and environments) to include the above information. 

(4) Lastly, when I was running test experiments, sometimes one of the nodes, or part of the Heron editor itself would throw an exception or otherwise crash. Sometimes this left the Heron editor in a zombie state where some aspects of the GUI were responsive and others were not. It would be good to see a more graceful full shutdown of the program when part of it crashes or throws an exception, especially as this is likely to be common as people learn to use it. More problematically, in some of these cases, after closing or force quitting Heron, the TCP ports were not properly relinquished, and thus restarting Heron would run into an "address in use" error. Finding and killing the processes that were still using the ports is not something that is obvious, especially to a beginner, and it would be great to see Heron deal with this better. Ideally, code would be introduced to carefully avoid leaving ports occupied during a hard shutdown, and furthermore, when the address in use error comes up, it would be great to give the user some idea of what to do about it.

A lot of effort has been put into Heron to achieve graceful shut down of processes, especially when these run on different machines that do not know when the GUI process has closed. The code that is being suggested to avoid leaving ports open has been implemented and this works properly when processes do not crash (Heron is terminated by the user) and almost always when there is a bug in a process that forces it to crash. In the version of Heron available during the reviewing process there were bugs that caused the above behaviour (Node code hanging and leaving zombie processes) on MacOS systems. These have now been fixed. There are very seldom instances though, especially during Node development, that crashing processes will hang and need to be terminated manually. We have taken on board the reviewer’s comments that users should be made more aware of these issues and have also described this situation in the Debugging part of Heron’s documentation. There we explain the logging and other tools Heron provides to help users debug their own Nodes and how to deal with hanging processes.

Heron is still in alpha (usable but with bugs) and the best way to debug it and iron out all the bugs in all use cases is through usage from multiple users and error reporting (we would be grateful if the errors the reviewer mentions could be reported in Heron’s github Issues page). We are always addressing and closing any reported errors, since this is the only way for Heron to transition from alpha to beta and eventually to production code quality.

Overall I think that, with these improvements, this could be the beginning of a powerful and versatile new system that would enable flexible experiment design with a relatively low technical barrier to entry. I could see this system being useful to many different labs and fields. 

We thank the reviewer for positive and supportive words and for the constructive feedbacks. We believe we have now addressed all the raised concerns.  

Reviewer #2 (Public Review):

Summary:

The authors provide an open-source graphic user interface (GUI) called Heron, implemented in Python, that is designed to help experimentalists to

(1) design experimental pipelines and implement them in a way that is closely aligned with their mental schemata of the experiments,

(2) execute and control the experimental pipelines with numerous interconnected hardware and software on a network.

The former is achieved by representing an experimental pipeline using a Knowledge Graph and visually representing this graph in the GUI. The latter is accomplished by using an actor model to govern the interaction among interconnected nodes through messaging, implemented using ZeroMQ. The nodes themselves execute user-supplied code in, but not limited to, Python.

Using three showcases of behavioral experiments on rats, the authors highlighted three benefits of their software design:

(1) the knowledge graph serves as a self-documentation of the logic of the experiment, enhancing the readability and reproducibility of the experiment,

(2) the experiment can be executed in a distributed fashion across multiple machines that each has a different operating system or computing environment, such that the experiment can take advantage of hardware that sometimes can only work on a specific computer/OS, a commonly seen issue nowadays,

(3) he users supply their own Python code for node execution that is supposed to be more friendly to those who do not have a strong programming background.

Strengths:

(1) The software is light-weight and open-source, provides a clean and easy-to-use GUI,

(2) The software answers the need of experimentalists, particularly in the field of behavioral science, to deal with the diversity of hardware that becomes restricted to run on dedicated systems.

(3) The software has a solid design that seems to be functionally reliable and useful under many conditions, demonstrated by a number of sophisticated experimental setups.

(4) The software is well documented. The authors pay special attention to documenting the usage of the software and setting up experiments using this software.

Weaknesses:

(1) While the software implementation is solid and has proven effective in designing the experiment showcased in the paper, the novelty of the design is not made clear in the manuscript. Conceptually, both the use of graphs and visual experimental flow design have been key features in many widely used softwares as suggested in the background section of the manuscript. In particular, contrary to the authors’ claim that only pre-defined elements can be used in Simulink or LabView, Simulink introduced MATLAB Function Block back in 2011, and Python code can be used in LabView since 2018. Such customization of nodes is akin to what the authors presented.

In the Heron manuscript we have provided an extensive literature review of existing systems from which Heron has borrowed ideas. We never wished to say that graphs and visual code is what sets Heron apart since these are technologies predating Heron by many years and implemented by a large number of software. We do not believe also that we have mentioned that LabView or Simulink can utilise only predefined nodes. What we have said is that in such systems (like LabView, Simulink and Bonsai) the focus of the architecture is on prespecified low level elements while the ability for users to author their own is there but only as an afterthought. The difference with Heron is that in the latter the focus is on the users developing their own elements. One could think of LabView style software as node-based languages (with low level visual elements like loops and variables) that also allow extra scripting while Heron is a graphical wrapper around python where nodes are graphical representations of whole processes. To our knowledge there is no other software that allows the very fast generation of graphical elements representing whole processes whose communication can also be defined graphically. Apart from this distinction, Heron also allows a graphical approach to writing code for processes that span different machines which again to our knowledge is a novelty of our approach and one of its strongest points towards ease of experimental pipeline creation (without sacrificing expressivity). 

(2) The authors claim that the knowledge graph can be considered as a self-documentation of an experiment. I found it to be true to some extent. Conceptually it’s a welcoming feature and the fact that the same visualization of the knowledge graph can be used to run and control experiments is highly desirable (but see point 1 about novelty). However, I found it largely inadequate for a person to understand an experiment from the knowledge graph as visualized in the GUI alone. While the information flow is clear, and it seems easier to navigate a codebase for an experiment using this method, the design of the GUI does not make it a one-stop place to understand the experiment. Take the Knowledge Graph in Supplementary Figure 2B as an example, it is associated with the first showcase in the result section highlighting this self-documentation capability. I can see what the basic flow is through the disjoint graph where 1) one needs to press a key to start a trial, and 2) camera frames are saved into an avi file presumably using FFMPEG. Unfortunately, it is not clear what the parameters are and what each block is trying to accomplish without the explanation from the authors in the main text. Neither is it clear about what the experiment protocol is without the help of Supplementary Figure 2A.

In my opinion, text/figures are still key to documenting an experiment, including its goals and protocols, but the authors could take advantage of the fact that they are designing a GUI where this information, with properly designed API, could be easily displayed, perhaps through user interaction. For example, in Local Network -> Edit IPs/ports in the GUI configuration, there is a good tooltip displaying additional information for the "password" entry. The GUI for the knowledge graph nodes can very well utilize these tooltips to show additional information about the meaning of the parameters, what a node does, etc, if the API also enforces users to provide this information in the form of, e.g., Python docstrings in their node template. Similarly, this can be applied to edges to make it clear what messages/data are communicated between the nodes. This could greatly enhance the representation of the experiment from the Knowledge graph.

In the first showcase example in the paper “Probabilistic reversal learning.

Implementation as self-documentation” we go through the steps that one would follow in order to understand the functionality of an experiment through Heron’s Knowledge Graph. The Graph is not just the visual representation of the Nodes in the GUI but also their corresponding code bases. We mention that the way Heron’s API limits the way a Node’s code is constructed (through an Actor based paradigm) allows for experimenters to easily go to the code base of a specific Node and understand its 2 functions (initialisation and worker) without getting bogged down in the code base of the whole Graph (since these two functions never call code from any other Nodes). Newer versions of Heron facilitate this easy access to the appropriate code by also allowing users to attach to Heron their favourite IDE and open in it any Node’s two scripts (worker and com) when they double click on the Node in Heron’s GUI. On top of this, Heron now (in the versions developed as answers to the reviewers’ comments) allows Node creators to add extensive comments on a Node but also separate comments on the Node’s parameters and input and output ports. Those can be seen as tooltips when one hovers over the Node (a feature that can be turned off or on by the Info button on every Node).  

As Heron stands at the moment we have not made the claim that the Heron GUI is the full picture in the self-documentation of a Graph. We take note though the reviewer’s desire to have the GUI be the only tool a user would need to use to understand an experimental implementation. The solution to this is the same as the one described by the reviewer of using the GUI to show the user the parts of the code relevant to a specific Node without the user having to go to a separate IDE or code editor. The reason this has not been implemented yet is the lack of a text editor widget in the underlying gui library (DearPyGUI). This is in their roadmap for their next large release and when this exists we will use it to implement exactly the idea the reviewer is suggesting, but also with the capability to not only read comments and code but also directly edit a Node’s code (see Heron’s roadmap). Heron’s API at the moment is ideal for providing such a text editor straight from the GUI.

(3) The design of Heron was primarily with behavioral experiments in mind, in which highly accurate timing is not a strong requirement. Experiments in some other areas that this software is also hoping to expand to, for example, electrophysiology, may need very strong synchronization between apparatus, for example, the record timing and stimulus delivery should be synced. The communication mechanism implemented in Heron is asynchronous, as I understand it, and the code for each node is executed once upon receiving an event at one or more of its inputs. The paper, however, does not include a discussion, or example, about how Heron could be used to address issues that could arise in this type of communication. There is also a lack of information about, for example, how nodes handle inputs when their ability to execute their work function cannot keep up with the frequency of input events. Does the publication/subscription handle the queue intrinsically? Will it create problems in real-time experiments that make multiple nodes run out of sync? The reader could benefit from a discussion about this if they already exist, and if not, the software could benefit from implementing additional mechanisms such that it can meet the requirements from more types of experiments.

In order to address the above lack of explanation (that also the first reviewer pointed out) we expanded the third experimental example in the paper with three more sections. One focuses solely on explaining how in this example (which acquires and saves large amounts of data from separate Nodes running on different machines) one would be able to time align the different data packets generated in different Nodes to each other. The techniques described there are directly implementable on experiments where the requirements of synching are more stringent than the behavioural experiment we showcase (like in ephys experiments). 

Regarding what happens to packages when the worker function of a Node is too slow to handle its traffic, this is mentioned in the paper (Code architecture paragraph): “Heron is designed to have no message buffering, thus automatically dropping any messages that come into a Node’s inputs while the Node’s worker function is still running.” This is also explained in more detail in Heron’s documentation. The reasoning for a no buffer system (as described in the documentation) is that for the use cases Heron is designed to handle we believe there is no situation where a Node would receive large amounts of data in bursts while very little data during the rest of the time (in which case a buffer would make sense). Nodes in most experiments will either be data intensive but with a constant or near constant data receiving speed (e.g. input from a camera or ephys system) or will have variable data load reception but always with small data loads (e.g. buttons). The second case is not an issue and the first case cannot be dealt with a buffer but with the appropriate code design, since buffering data coming in a Node too slow for its input will just postpone the inevitable crash. Heron’s architecture principle in this case is to allow these ‘mistakes’ (i.e. package dropping) to happen so that the pipeline continues to run and transfer the responsibility of making Nodes fast enough to the author of each Node. At the same time Heron provides tools (see the Debugging section of the documentation and the time alignment paragraph of the “Rats playing computer games”  example in the manuscript) that make it easy to detect package drops and either correct them or allow them but also allow time alignment between incoming and outgoing packets. In the very rare case where a buffer is required Heron’s do-it-yourself logic makes it easy for a Node developer to implement their own Node specific buffer.

(4) The authors mentioned in "Heron GUI’s multiple uses" that the GUI can be used as an experimental control panel where the user can update the parameters of the different Nodes on the fly. This is a very useful feature, but it was not demonstrated in the three showcases. A demonstration could greatly help to support this claim.

As the reviewer mentions, we have found Heron’s GUI double role also as an experimental on-line controller a very useful capability during our experiments. We have expanded the last experimental example to also showcase this by showing how on the “Rats playing computer games” experiment we used the parameters of two Nodes to change the arena’s behaviour while the experiment was running, depending on how the subject was behaving at the time (thus exploring a much larger set of parameter combinations, faster during exploratory periods of our shaping protocols construction). 

(5) The API for node scripts can benefit from having a better structure as well as having additional utilities to help users navigate the requirements, and provide more guidance to users in creating new nodes. A more standard practice in the field is to create three abstract Python classes, Source, Sink, and Transform that dictate the requirements for initialisation, work_function, and on_end_of_life, and provide additional utility methods to help users connect between their code and the communication mechanism. They can be properly docstringed, along with templates. In this way, the com and worker scripts can be merged into a single unified API. A simple example that can cause confusion in the worker script is the "worker_object", which is passed into the initialise function. It is unclear what this object this variable should be, and what attributes are available without looking into the source code. As the software is also targeting those who are less experienced in programming, setting up more guidance in the API can be really helpful. In addition, the self-documentation aspect of the GUI can also benefit from a better structured API as discussed in point 2 above.

The reviewer is right that using abstract classes to expose to users the required API would be a more standard practice. The reason we did not choose to do this was to keep Heron easily accessible to entry level Python programmers who do not have familiarity yet with object oriented programming ideas. So instead of providing abstract classes we expose only the implementation of three functions which are part of the worker classes but the classes themselves are not seen by the users of the API. The point about the users’ accessibility to more information regarding a few objects used in the API (the worker object for example) has been taken on board and we have now addressed this by type hinting all these objects both in the templates and more importantly in the automatically generated code that Heron now creates when a user chooses to create a Node graphically (a feature of Heron not present in the version available in the initial submission of this manuscript).  

(6) The authors should provide more pre-defined elements. Even though the ability for users to run arbitrary code is the main feature, the initial adoption of a codebase by a community, in which many members are not so experienced with programming, is the ability for them to use off-the-shelf components as much as possible. I believe the software could benefit from a suite of commonly used Nodes.

There are currently 12 Node repositories in the Heron-repositories project on Github with more than 30 Nodes, 20 of which are general use (not implementing a specific experiment’ logic). This list will continue to grow but we fully appreciate the truth of the reviewer’s comment that adoption will depend on the existence of a large number of commonly used Nodes (for example Numpy, and OpenCV Nodes) and are working towards this goal.

(7) It is not clear to me if there is any capability or utilities for testing individual nodes without invoking a full system execution. This would be critical when designing new experiments and testing out each component.

There is no capability to run the code of an individual Node outside Heron’s GUI. A user could potentially design and test parts of the Node before they get added into a Node but we have found this to be a highly inefficient way of developing new Nodes. In our hands the best approach for Node development was to quickly generate test inputs and/or outputs using the “User Defined Function 1I 1O” Node where one can quickly write a function and make it accessible from a Node. Those test outputs can then be pushed in the Node under development or its outputs can be pushed in the test function, to allow for incremental development without having to connect it to the Nodes it would be connected in an actual pipeline. For example, one can easily create a small function that if a user presses a key will generate the same output (if run from a “User Defined Function 1I 1O” Node) as an Arduino Node reading some buttons. This output can then be passed into an experiment logic Node under development that needs to do something with this input. In this way during a Node development Heron allows the generation of simulated hardware inputs and outputs without actually running the actual hardware. We have added this way of developing Nodes also in our manuscript (Creating a new Node).

Reviewer #3 (Public Review):

Summary:

The authors present a Python tool, Heron, that provides a framework for defining and running experiments in a lab setting (e.g. in behavioural neuroscience). It consists of a graphical editor for defining the pipeline (interconnected nodes with parameters that can pass data between them), an API for defining the nodes of these pipelines, and a framework based on ZeroMQ, responsible for the overall control and data exchange between nodes. Since nodes run independently and only communicate via network messages, an experiment can make use of nodes running on several machines and in separate environments, including on different operating systems.

Strengths:

As the authors correctly identify, lab experiments often require a hodgepodge of separate hardware and software tools working together. A single, unified interface for defining these connections and running/supervising the experiment, together with flexibility in defining the individual subtasks (nodes) is therefore a very welcome approach. The GUI editor seems fairly intuitive, and Python as an accessible programming environment is a very sensible choice. By basing the communication on the widely used ZeroMQ framework, they have a solid base for the required non-trivial coordination and communication. Potential users reading the paper will have a good idea of how to use the software and whether it would be helpful for their own work. The presented experiments convincingly demonstrate the usefulness of the tool for realistic scientific applications.

Weaknesses:

(1) In my opinion, the authors somewhat oversell the reproducibility and "selfdocumentation" aspect of their solution. While it is certainly true that the graph representation gives a useful high-level overview of an experiment, it can also suffer from the same shortcomings as a "pure code" description of a model - if a user gives their nodes and parameters generic/unhelpful names, reading the graph will not help much. 

This is a problem that to our understanding no software solution can possibly address. Yet having a visual representation of how different inputs and outputs connect to each other we argue would be a substantial benefit in contrast to the case of “pure code” especially when the developer of the experiment has used badly formatted variable names.

(2) Making the link between the nodes and the actual code is also not straightforward, since the code for the nodes is spread out over several directories (or potentially even machines), and not directly accessible from within the GUI. 

This is not accurate. The obligatory code of a Node always exists within a single folder and Heron’s API makes it rather cumbersome to spread scripts relating to a Node across separate folders. The Node folder structure can potentially be copied over different machines but this is why Heron is tightly integrated with git practices (and even politely asks the user with popup windows to create git repositories of any Nodes they create whilst using Heron’s automatic Node generator system). Heron’s documentation is also very clear on the folder structure of a Node which keeps the required code always in the same place across machines and more importantly across experiments and labs. Regarding the direct accessibility of the code from the GUI, we took on board the reviewers’ comments and have taken the first step towards correcting this. Now one can attach to Heron their favourite IDE and then they can double click on any Node to open its two main scripts (com and worker) in that IDE embedded in whatever code project they choose (also set in Heron’s settings windows). On top of this, Heron now allows the addition of notes both for a Node and for all its parameters, inputs and outputs which can be viewed by hovering the mouse over them on the Nodes’ GUIs. The final step towards GUI-code integration will be to have a Heron GUI code editor but this is something that has to wait for further development from Heron’s underlying GUI library DearPyGUI.

(3) The authors state that "[Heron’s approach] confers obvious benefits to the exchange and reproducibility of experiments", but the paper does not discuss how one would actually exchange an experiment and its parameters, given that the graph (and its json representation) contains user-specific absolute filenames, machine IP addresses, etc, and the parameter values that were used are stored in general data frames, potentially separate from the results. Neither does it address how a user could keep track of which versions of files were used (including Heron itself).

Heron’s Graphs, like any experimental implementation, must contain machine specific strings. These are accessible either from Heron’s GUI when a Graph json file is opened or from the json file itself. Heron in this regard does not do anything different to any other software, other than saving the graphs into human readable json files that users can easily manipulate directly.

Heron provides a method for users to save every change of the Node parameters that might happen during an experiment so that it can be fully reproduced. The dataframes generated are done so in the folders specified by the user in each of the Nodes (and all those paths are saved in the json file of the Graph). We understand that Heron offers a certain degree of freedom to the user (Heron’s main reason to exist is exactly this versatility) to generate data files wherever they want but makes sure every file path gets recorded for subsequent reproduction. So, Heron behaves pretty much exactly like any other open source software. What we wanted to focus on as the benefits of Heron on exchange and reproducibility was the ability of experimenters to take a Graph from another lab (with its machine specific file paths and IP addresses) and by examining the graphical interface of it to be able to quickly tweak it to make it run on their own systems. That is achievable through the fact that a Heron experiment will be constructed by a small amount of Nodes (5 to 15 usually) whose file paths can be trivially changed in the GUI or directly in the json file while the LAN setup of the machines used can be easily reconstructed from the information saved in the secondary GUIs.

Where Heron needs to improve (and this is a major point in Heron’s roadmap) is the need to better integrate the different saved experiments with the git versions of Heron and the Nodes that were used for that specific save. This, we appreciate is very important for full reproducibility of the experiment and it is a feature we will soon implement. More specifically users will save together with a graph the versions of all the used repositories and during load the code base utilised will come from the recorded versions and not from the current head of the different repositories. This is a feature that we are currently working on now and as our roadmap suggests will be implemented by the release of Heron 1.0. 

(4) Another limitation that in my opinion is not sufficiently addressed is the communication between the nodes, and the effect of passing all communications via the host machine and SSH. What does this mean for the resulting throughput and latency - in particular in comparison to software such as Bonsai or Autopilot? The paper also states that "Heron is designed to have no message buffering, thus automatically dropping any messages that come into a Node’s inputs while the Node’s worker function is still running."- it seems to be up to the user to debug and handle this manually?

There are a few points raised here that require addressing. The first is Heron’s requirement to pass all communication through the main (GUI) machine. We understand (and also state in the manuscript) that this is a limitation that needs to be addressed. We plan to do this is by adding to Heron the feature of running headless (see our roadmap). This will allow us to run whole Heron pipelines in a second machine which will communicate with the main pipeline (run on the GUI machine) with special Nodes. That will allow experimenters to define whole pipelines on secondary machines where the data between their Nodes stay on the machine running the pipeline. This is an important feature for Heron and it will be one of the first features to be implemented next (after the integration of the saving system with git). 

The second point is regarding Heron’s throughput latency. In our original manuscript we did not have any description of Heron’s capabilities in this respect and both other reviewers mentioned this as a limitation. As mentioned above, we have now addressed this by adding a section to our third experimental example that fully describes how much CPU is required to run a full experimental pipeline running on two machines and utilising also non python code executables (a Unity game). This gives an overview of how heavy pipelines can run on normal computers given adequate optimisation and utilising Heron’s feature of forcing some Nodes to run their Worker processes on a specific core. At the same time, Heron’s use of 0MQ protocol makes sure there are no other delays or speed limitations to message passing. So, message passing within the same machine is just an exchange of memory pointers while messages passing between different machines face the standard speed limitations of the Local Access Network’s ethernet card speeds. 

Finally, regarding the message dropping feature of Heron, as mentioned above this is an architectural decision given the use cases of message passing we expect Heron to come in contact with. For a full explanation of the logic here please see our answer to the 3rd comment by Reviewer 2.

(5) As a final comment, I have to admit that I was a bit confused by the use of the term "Knowledge Graph" in the title and elsewhere. In my opinion, the Heron software describes "pipelines" or "data workflows", not knowledge graphs - I’d understand a knowledge graph to be about entities and their relationships. As the authors state, it is usually meant to make it possible to "test propositions against the knowledge and also create novel propositions" - how would this apply here?

We have described Heron as a Knowledge Graph instead of a pipeline, data workflow or computation graph in order to emphasise Heron’s distinct operation in contrast to what one would consider a standard pipeline and data workflow generated by other visual based software (like LabView and Bonsai). This difference exists on what a user should think of as the base element of a graph, i.e. the Node. In all other visual programming paradigms, the Node is defined as a low-level computation, usually a language keyword, language flow control or some simple function. The logic in this case is generated by composing together the visual elements (Nodes). In Heron the Node is to be thought of as a process which can be of arbitrary complexity and the logic of the graph is composed by the user both within each Node and by the way the Nodes are combined together. This is an important distinction in Heron’s basic operation logic and it is we argue the main way Heron allows flexibility in what can be achieved while retaining ease of graph composition (by users defining their own level of complexity and functionality encompassed within each Node). We have found that calling this approach a computation graph (which it is) or a pipeline or data workflow would not accentuate this difference. The term Knowledge Graph was the most appropriate as it captures the essence of variable information complexity (even in terms of length of shortest string required) defined by a Node.

Recommendations for the authors:  

Reviewer #1 (Recommendations For The Authors):

-  No buffering implies dropped messages when a node is busy. It seems like this could be very problematic for some use cases... 

This is a design principle of Heron. We have now provided a detailed explanation of the reasoning behind it in our answer to Reviewer 2 (Paragraph 3) as well as in the manuscript. 

-  How are ssh passwords stored, and is it secure in some way or just in plain text?  

For now they are plain text in an unencrypted file that is not part of the repo (if one gets Heron from the repo). Eventually, we would like to go to private/public key pairs but this is not a priority due to the local nature of Heron’s use cases (all machines in an experiment are expected to connect in a LAN).  

Minor notes / copyedits:

-  Figure 2A: right and left seem to be reversed in the caption. 

They were. This is now fixed. 

-  Figure 2B: the text says that proof of life messages are sent to each worker process but in the figure, it looks like they are published by the workers? Also true in the online documentation.  

The Figure caption was wrong. This is now fixed.

-  psutil package is not included in the requirements for GitHub

We have now included psutil in the requirements.

-  GitHub readme says Python >=3.7 but Heron will not run as written without python >= 3.9 (which is alluded to in the paper)

The new Heron updates require Python 3.11. We have now updated GitHub and the documentation to reflect this.

-  The paper mentions that the Heron editor must be run on Windows, but this is not mentioned in the Github readme.  

This was an error in the manuscript that we have now corrected.

-  It’s unclear from the readme/manual how to remove a node from the editor once it’s been added.  

We have now added an X button on each Node to complement the Del button on the keyboard (for MacOS users that do not have this button most of the times).

-  The first example experiment is called the Probabilistic Reversal Learning experiment in text, but the uncertainty experiment in the supplemental and on GitHub.  

We have now used the correct name (Probabilistic Reversal Learning) in both the supplemental material and on GitHub

-  Since Python >=3.9 is required, consider using fstrings instead of str.format for clarity in the codebase  

Thank you for the suggestion. Latest Heron development has been using f strings and we will do a refactoring in the near future.

-  Grasshopper cameras can run on linux as well through the spinnaker SDK, not just Windows.  

Fixed in the manuscript. 

-  Figure 4: Square and star indicators are unclear.

Increased the size of the indicators to make them clear.

-  End of page 9: "an of the self" presumably a typo for "off the shelf"?  

Corrected.

-  Page 10 first paragraph. "second root" should be "second route"

Corrected.

-  When running Heron, the terminal constantly spams Blowfish encryption deprecation warnings, making it difficult to see the useful messages.  

The solution to this problem is to either update paramiko or install Heron through pip. This possible issue is mentioned in the documentation.

-  Node input /output hitboxes in the GUI are pretty small. If they could be bigger it would make it easier to connect nodes reliably without mis-clicks.

We have redone the Node GUI, also increasing the size of the In/Out points.

Reviewer #2 (Recommendations For The Authors):

(1) There are quite a few typos in the manuscript, for example: "one can accessess the code", "an of the self", etc.  

Thanks for the comment. We have now screened the manuscript for possible typos.

(2) Heron’s GUI can only run on Windows! This seems to be the opposite of the key argument about the portability of the experimental setup.  

As explained in the answers to Reviewer 1, Heron can run on most machines that the underlying python libraries run, i.e. Windows and Linux (both for x86 and Arm architectures). We have tested it on Windows (10 and 11, both x64), Linux PC (Ubuntu 20.04.6, x64) and Raspberry Pi 4 (Debian GNU/Linux 12 (bookworm), aarch64). We have now revised the manuscript and the GitHub repo to reflect this.

(3) Currently, the output is displayed along the left edge of the node, but the yellow dot connector is on the right. It would make more sense to have the text displayed next to the connectors.  

We have redesigned the Node GUI and have now placed the Out connectors on the right side of the Node.

(4) The edges are often occluded by the nodes in the GUI. Sometimes it leads to some confusion, particularly when the number of nodes is large, e.g., Fig 4.

This is something that is dependent on the capabilities of the DearPyGUI module. At the moment there is no way to control the way the edges are drawn.

Reviewer #3 (Recommendations For The Authors):

A few comments on the software and the documentation itself:

- From a software engineering point of view, the implementation seems to be rather immature. While I get the general appeal of "no installation necessary", I do not think that installing dependencies by hand and cloning a GitHub repository is easier than installing a standard package.

We have now added a pip install capability which also creates a Heron command line command to start Heron with. 

-The generous use of global variables to store state (minor point, given that all nodes run in different processes), boilerplate code that each node needs to repeat, and the absence of any kind of automatic testing do not give the impression of a very mature software (case in point: I had to delete a line from editor.py to be able to start it on a non-Windows system).  

As mentioned, the use of global variables in the worker scripts is fine partly due to the multi process nature of the development and we have found it is a friendly approach to Matlab users who are just starting with Python (a serious consideration for Heron). Also, the parts of the code that would require a singleton (the Editor for example) are treated as scripts with global variables while the parts that require the construction of objects are fully embedded in classes (the Node for example). A future refactoring might make also all the parts of the code not seen by the user fully object oriented but this is a decision with pros and cons needing to be weighted first. 

Absence of testing is an important issue we recognise but Heron is a GUI app and nontrivial unit tests would require some keystroke/mouse movement emulator (like QTest of pytest-qt for QT based GUIs). This will be dealt with in the near future (using more general solutions like PyAutoGUI) but it is something that needs a serious amount of effort (quite a bit more that writing unit tests for non GUI based software) and more importantly it is nowhere as robust as standard unit tests (due to the variable nature of the GUI through development) making automatic test authoring an almost as laborious a process as the one it is supposed to automate.

-  From looking at the examples, I did not quite see why it is necessary to write the ..._com.py scripts as Python files, since they only seem to consist of boilerplate code and variable definitions. Wouldn’t it be more convenient to represent this information in configuration files (e.g. yaml or toml)?  

The com is not a configuration file, it is a script that launches the communication process of the Node. We could remove the variable definitions to a separate toml file (which then the com script would have to read). The pros and cons of such a set up should be considered in a future refactoring.

Minor comments for the paper:

-  p.7 (top left): "through its return statement" - the worker loop is an infinite loop that forwards data with a return statement?  

This is now corrected. The worker loop is an infinite loop and does not return anything but at each iteration pushes data to the Nodes output.

-  p.9 (bottom right): "of the self" → "off-the-shelf"  

Corrected.

-  p.10 (bottom left): "second root" → "second route"  

Corrected.

-  Supplementary Figure 3: Green start and square seem to be swapped (the green star on top is a camera image and the green star on the bottom is value visualization - inversely for the green square).  

The star and square have been swapped around.

-  Caption Supplementary Figure 4 (end): "rashes to receive" → "rushes to receive"  

Corrected.

eLife Assessment

This important study advances our understanding of the role of dopamine in modulating pair bonding in mandarin voles by examining dopamine signaling within the nucleus accumbens across various social stimuli using state-of-the-art causal perturbations. The evidence supporting the findings is compelling, particularly cutting-edge approaches for measuring dopamine release as well as the activity of dopamine receptor populations during social bonding. Some concerns remain about the statistical analyses.

Reviewer #2 (Public review):

Summary:

Using in vivo fiber-photometry the authors first establish that DA release when contacting their partner mouse increases with days of cohabitation while this increase is not observed when contacting a stranger mouse. Similar effects are found in D1-MSNs and D2-MSNs with the D1-MSN responses increasing and D2-MSN responses decreasing with days of cohabitation. They then use slice physiology to identify underlying plasticity/adaptation mechanisms that could contribute to the changes in D1/D2-MSN responses. Last, to address causality the authors use chemogenetic tools to selectively inhibit or activate NAc shell D1 or D2 neurons that project to the ventral pallidum. They found that D2 inhibition facilitates bond formation while D2 excitation inhibits bond formation. In contrast, both D1-MSN activation and inhibition inhibits bond formation.

Strengths:

The strength of the manuscript lies in combining in vivo physiology to demonstrate circuit engagement and chemogenetic manipulation studies to address circuit involvement in pair bond formation in a monogamous vole.

Weaknesses:

Weaknesses include that a large set of experiments within the manuscript are dependent on using short promoters for D1 and D2 receptors in viral vectors. As the authors acknowledge this approach can lead to ectopic expression and the presented immunohistochemistry supports this notion. It seems to me that the presented quantification underestimates the degree of ectopic expression that is observed by eye when looking at the presented immunohistochemistry. However, given that Cre transgenic animals are not available for Microtus mandarinus and given the distinct physiological and behavioral outcomes when imaging and manipulating both viral-targeted populations this concern is minor.

The slice physiology experiments provide some interesting outcomes but it is unclear how they can be linked to the in vivo physiological outcomes and some of the outcomes don't match intuitively (e.g. cohabitation enhances excitatory/inhibitory balance in D2-MSNs but the degree of contact-induced inhibition is enhanced in D2-MSN).

One interesting finding is that the relationship between D2-MSN and pair bond formation is quite clear (inhibition facilitates while excitation inhibits pair bond formation). In contrast, the role of D1-MSNs is more complicated since both excitation and inhibition disrupts pair bond formation. This is not convincingly discussed.

It seemed a missed opportunity that physiological read out is limited to males. I understand though that adding females may be beyond the scope of this manuscript.

Comments on revised version:

The authors addressed most of my comments, some would still need to be addressed.

(1) Previous comment: "The authors do not use an isosbestic control wavelength in photometry experiments, although they do use EGFP control mice which show no effects of these interventions, a within-subject control such as an isosbestic excitation wavelength could give more confidence in these data and rule out motion artefacts within subjects."

The authors should include a paragraph in the discussion addressing the limitations of not using an internal control for the fiberphotometric measurements.

(2) Previous Comment: The slice physiology experiments provide some interesting outcomes but it is unclear how they can be linked to the in vivo physiological outcomes and some of the outcomes don't match intuitively (e.g. cohabitation enhances excitatory/inhibitory balance in D2-MSNs but the degree of contact-induced inhibition is enhanced in D2-MSN).

My comment may not have been clear and the response didn't address my comment. What is missing in the discussion is an explanation of why a relative increase in excitation of D2-MSNs in the slice (Fig. 4J) is associated with an increased inhibition in vivo (Fig. 2H)?

(3) Previous Comment: One interesting finding is that the relationship between D2-MSN and pair bond formation is quite clear (inhibition facilitates while excitation inhibits pair bond formation). In contrast, the role of D1-MSNs is more complicated since both excitation and inhibition disrupt pair bond formation. This is not convincingly discussed.

Similarly, here the response provided does not address my question. Please focus on discussing why both excitation and inhibition of D1-MSNs can disrupt pair bond formation (Figure 7).

Reviewer #3 (Public review):

Summary:

The manuscript is evaluating changes in dopamine signaling in the nucleus accumbens following pair bonding and exposure to various stimuli in mandarin voles. In addition, the authors present chemogenetic data which demonstrates excitation and inhibition of D1 and D2 MSN affect pair bond formation.

Strengths:

The experimental designs are strong. The approaches are innovative and use cutting-edge methods. The manuscript is well written.

Comments on revised version:

I appreciate the efforts by the authors to address many of my previous comments. The issues that remain are those associated with the statistics. It seems that not all statistical analyses were performed with the correct test. For example, the photometry data comparing emissions during partner vs stranger investigation over time would be best performed as a two-way ANOVA with odor type and time being separate variables. Also, there are paired t-tests being performed by calculating an average deltaF/F during the 4 second window following the being of a behavioral event. I think an area-under-the-curve calculation of these events would better capture the fluorescent emissions of these events as an index. Details in the Result describing the data being analyzed via ANOVA vs t-tests when reporting the results would be useful for the reviewer to understand each analysis.

eLife Assessment

The authors aim to elucidate the mechanism by which pyroptosis (through the formation of Gasdermin D (GSDMD) pores in the plasma membrane) contributes to increased release of procoagulant Tissue Factor-containing microvesicles. The data offers solid mechanistic insights as to the interplay between pyroptosis and microvesicle release with NINJ1. The study provides useful insights into the potential of targeting Ninj1 as a therapeutic strategy.

Reviewer #1 (Public review):

The authors demonstrated that NINJ1 promotes TF-positive MV release during pyroptosis and thereby triggers coagulation. Coagulation is one of the risk factors that can cause secondary complications in various inflammatory diseases, making it a highly important therapeutic target in clinical treatment. This paper effectively explains the connection between pyroptosis and MV release with Ninj1, which is a significant strength. It provides valuable insight into the potential of targeting Ninj1 as a therapeutic strategy.

Although the advances in this paper are valuable, several aspects need to be clarified. Some comments are discussed below.

(1) Since it is not Ninj1 directly regulating coagulation but rather the MV released by Ninj1 playing a role, the title should include that. The current title makes it seem like Ninj1 directly regulates inflammation and coagulation. It would be better to revise the title.

(2) Ninj1 is known to be an induced protein that is barely expressed in normal conditions. As you showed in "Fig1G" data, control samples showed no detection of Ninj1. However, in "Figure S1", all tissues (liver, lung, kidney and spleen) expressed Ninj1 protein. If the authors stimulated the mice with fla injection, it should be mentioned in the figure legend.

(3) In "Fig3A", the Ninj1 protein expression was increased in the control of BMDM +/- cell lysate rather than fla stimulation. However, in MV, Ninj1 was not detected at all in +/- control but was only observed with Fla injection. The authors need to provide an explanation for this observation. Additionally, looking at the MV β-actin lane, the band thicknesses appear to be very different between groups. It seems necessary to equalize the protein amounts. If that is difficult, at least between the +/+ and +/- controls.

(4) Since the authors focused Ninj1-dependent microvesicle (MV) release, they need to show MV characterizations (EM, NTA, Western for MV markers, etc...).

(5) To clarify whether Ninj1-dependent MV induces coagulation, the authors need to determine whether platelet aggregation is reduced with isolated +/- MVs compared to +/+ MVs.

(6) Even with the authors well established experiments with haploid mice, it is a critical limitation of this paper. To improve the quality of this paper, the authors should consider confirming the findings using mouse macrophage cell lines, such as generating Ninj1-/- Raw264.7 cell lines, to examine the homozygous effect.

(7) There was a paper reported in 2023 (Zhou, X. et al., NINJ1 Regulates Platelet Activation and PANoptosis in Septic Disseminated Intravascular Coagulation. Int. J. Mol. Sci. 2023) that revealed the relationship between Ninj1 and coagulation. According to this paper, inhibition of Ninj1 in platelets prevents pyroptosis, leading to reduced platelet activation and, consequently, the suppression of thrombosis. How about the activation of platelets in Ninj1 +/- mice? The author should add this paper in the reference section and discuss the platelet functions in their mice.

Reviewer #2 (Public review):

Summary:

The authors main goal is to understand the mechanism by which pyroptosis (through the formation of Gasdermin D (GSDMD) pores in the plasma membrane) contributes to increased release of procoagulant Tissue Factor-containing microvesicles (MV). Their previous data demonstrate that GSDMD is critical for the release of MV that contains Tissue Factor (TF), thus making a link between pyroptosis and hypercoagulation. Given the recent identification of NINJ1 being responsible for plasma membrane rupture (Kayagaki et al. Nature 2011), the authors wanted to determine if NINJ1 is responsible for TF-containing MV release. Given the constitutive ninj1 KO mouse leads to partial embryonic lethality, the authors decide to use a heterozygous ninj1 KO mouse (ninj1+/-), and demonstrate that Ninj1 plays a role in release of TF-containing MV.

eLife Assessment

In this valuable study, the authors integrate several datasets to describe how the genome interacts with nuclear bodies across distinct cell types and in Lamin A and LBR knockout cells. They provide convincing evidence to support their claims and particularly find that specific genomic regions segregate relative to the equatorial plane of the cell when considering their interaction with various nuclear bodies. The authors are encouraged to consider citing the relevant work of other labs who have shown the presence of different types of Lamin Associated Domains (LADs).

Reviewer #2 (Public review):

Summary:

Golamalamdari, van Schaik, Wang, Kumar Zhang, Zhang and colleagues study interactions between the speckle, nucleolus and lamina in multiple cell types (K562, H1, HCT116 and HFF). Their datasets define how interactions between the genome and the different nuclear landmarks relate to each other and change across cell types. They also identify how these relationships change in K562 cells in which LBR and LMNA are knocked out.

Strengths:

Overall, there are a number of datasets that are provided, and several "integrative" analyses performed. This is a major strength of the paper, and I imagine the datasets will be of use to the community to further probed and the relationships elucidated here further studied. An especially interesting result was that specific genomic regions (relative to their association with the speckle, lamina, and other molecular characteristics) segregate relative to the equatorial plane of the cell.

Weaknesses:

The experiments are primarily descriptive, and the cause-and-effect relationships are limited (though the authors do study the role of LMNA/LBR knockdown with their technologies).

Author response:

The following is the authors’ response to the original reviews.

eLife Assessment

(1) This is a valuable manuscript that successfully integrates several data sets to determine genomic interactions with nuclear bodies.

In this paper we both challenge and/or revise multiple long-standing “textbook” models of nuclear genome organization while also revealing new features of nuclear genome organization. Therefore, we argue that the contributions of this paper extend well beyond “valuable”. Specifically, these contributions include:

a. We challenge a several decades focus on the correlation of gene positioning relative to the nuclear lamina. Instead, through comparison of cell lines, we show a strong correlation of di4erences in gene activity with di4erences in relative distance to nuclear speckles in contrast to a very weak correlation with di4erences in relative distance to the nuclear lamina. This inference of little correlation of gene expression with nuclear lamina association was supported by direct experimental manipulation of genome positioning relative to the nuclear lamina. Despite pronounced changes in relative distances to the nuclear lamina there was little change relative to nuclear speckles and little change in gene expression.

b. We similarly challenge the long-standing proposed functional correlation between the radial positioning of genes and gene expression. Here, and in a now published companion paper (doi.org/10.1038/s42003-024-06838-7), we demonstrate how nuclear speckle positioning relative to nucleoli and the nuclear lamina varies among cell types, as does the inverse relationship between genome positioning relative to nuclear speckles and the nuclear lamina. Again, this is consistent with the primary correlation of gene activity being the positioning of genes relative to nuclear speckles and also explains previous observations showing a strong relationship between radial position and gene expression only in some cell types.

c. We identified a new partially repressed, middle to late DNA replicating type of chromosome domain- “p-w-v fILADs”- by their weak interaction with the nuclear lamina, which, based on our LMNA/LBR KO experimental results, compete with LADs for nuclear lamina association. Moreover, we show that when fLADs convert to iLADs, most conversions are to this p-w-v fiLAD state, although ~ one third are to a normal, active, early replicating iLAD state. Thus, fLADs can convert between repressed, partially repressed, and active states, challenging the prevailing assumption of the division of the genome into two states – active, early replicating A compartment/iLAD regions versus inactive, late replicating, B compartment/LAD regions.

d. We identified nuclear speckle associated domains as DNA replication initiation zones, with the domains showing strongest nuclear speckle attachment initiating DNA replication earliest in S-phase.

e. We describe for the first time an overall polarization of nuclear genome organization in adherent cells with the most active, earliest replicating genomic regions located towards the equatorial plane and less expressed genomic regions towards the nuclear top or bottom surfaces. This includes polarization of some LAD regions to the nuclear lamina at the equatorial plane and other LAD regions to the top or bottom nuclear surfaces.

We have now rewritten the text to make the significance of these new findings clearer.

(2) Strength of evidence: The evidence supporting the central claims is varied in its strength ranging from solid to incomplete. Orthogonal evidence validating the novel methodologies with alternative approaches would better support the central claims.

We argue that our work exploited methods, data, and analyses equal to or more rigorous than the current state-of-the-art. This indeed includes orthogonal evidence using alternative methods which both supported our novel methodologies as well as demonstrating their robustness relative to more conventional approaches. This explains how we were able to challenge/revise long-standing models and discover new features of nuclear genome organization. More specifically:

a. Unlike most previous analyses, we have integrated both genomic and imaging approaches to examine the nuclear genome organization relative to not one, but several di4erent nuclear locales and we have done this across several cell types. To our knowledge, this is the first such integrated approach and has been key to our success in appreciating new features of nuclear genome organization.

b. The 16-fraction DNA replication Repli-seq data we developed and applied to this project represents the highest temporal mapping of DNA replication timing to date.

c. The TSA-seq approach that we used remains the most accurate sequence-based method for estimating microscopic distance of chromosome regions to di4erent nuclear locales. As implemented, this method is unusually robust and direct as it exploits the exponential micron-scale gradient established by the di4usion of the free-radicals generated by peroxidase labeling to measure relative distances of chromosome regions to labeled nuclear locales. We had previously demonstrated that TSA-seq was able to estimate the average distances of genomic regions to nuclear speckles with an accuracy of ~50 nm, as validated by light microscopy. The TSA-seq 2.0 protocol we developed and applied to this project maintained the original resolution of TSA-seq to estimate to an accuracy of ~50 nm the average distances of genomic regions from nuclear speckles, as validated by light microscopy, while achieving more than a 10-fold reduction in the required number of cells.

We have rewritten the text to address the reviewer concerns that led them to their initial characterization of the TSA-seq as novel and not yet validated.

First, we have added a discussion of how the use of nuclear speckle TSA-seq as a “cytological ruler” was based on an extensive initial characterization of TSA-seq as described in previous published literature. In that previous literature we showed how the conventional molecular proximity method, ChIP-seq, instead showed local accumulation of the same marker proteins over short DNA regions unrelated to speckle distances. Second, we reference our companion paper, now published, and describe how the extension of TSA-seq to measure relative distances to nucleoli was further validated and shown to be robust by comparison to NAD-seq and extensive multiplexed immuno-FISH data. We further discuss how in the same companion paper we show how nucleolar DamID instead was inconsistent with both the NAD-seq and multiplexed immuno-FISH data as well as the nucleolar TSA-seq.

Third, we have added scatterplots showing exactly how highly the estimated microscopic distances to all three nuclear locales, measured in IMR90 fibroblasts, correlate with the TSA-seq measurements in HFF fibroblasts. This addresses the concern that we were not using the exact same fibroblast cell line for the TSA-seq versus microscopic measurements. The strong correlation already observed would only be expected to become even stronger with use of the exact same fibroblast cell lines for both measurements.

Fourth, we have addressed the reviewer concern that the nuclear lamin TSA-seq was not properly validated because it did not match nuclear lamin Dam-ID. We have now added to the text a more complete explanation of how microscopic proximity assays such as TSA-seq measure something di4erent from molecular proximity assays such as DamID or NAD-seq. We have added further explanation of how TSA-seq complements molecular proximity assays such as DamID and NAD-seq, allowing us to extract further information than either measurement alone. We also briefly discuss why TSA-seq succeeds for certain nuclear locales using multiple independent markers whereas molecular proximity assays may fail against the same nuclear locales using the same markers. This includes brief discussion from our own experience attempting unsuccessfully to use DamID against nucleoli and nuclear speckles.

Reviewer #1 (Public Review):

(1) The weakness of this study lies in the fact that many of the genomic datasets originated from novel methods that were not validated with orthogonal approaches, such as DNAFISH. Therefore, the detailed correlations described in this work are based on methodologies whose efficacy is not clearly established. Specifically, the authors utilized two modified protocols of TSA-seq for the detection of NADs (MKI67IP TSA-seq) and LADs (LMNB1-TSA-seq).

We disagree with the statement that the TSA-seq approach and data has not been validated by orthogonal approaches. We have now addressed this point in the revised manuscript text:

a) We added text to describe how previously FISH was used to validate speckle TSA-seq by demonstrating a residual of ~50 nm between the TSA-seq predicted distance to speckles and the distance measured by light microscopy using FISH:

"In contrast, TSA-seq measures relative distances to targets on a microscopic scale corresponding to 100s of nm to ~ 1 micron based on the measured diffusion radius of tyramide-biotin free-radicals (Chen et al., 2018). Exploiting the measured exponential decay of the tyramide-biotin free-radical concentration, we showed how the mean distance of chromosomes to nuclear speckles could be estimated from the TSA-seq data to an accuracy of ~50 nm, as validated by FISH (Chen et al., 2018)."

b) We note that we also previously have validated lamina (Chen et al, JCB 2018) and nucleolar (Kumar et al, 2024) TSA-seq and further validated speckle TSA-seq (Zhang et al, Genome Research 2021) by traditional immuno-FISH and/or immunostaining. The overall high correlation between lamina TSA-seq and the orthogonal lamina DamID method was also extensively discussed in the first TSA-seq paper (Chen et al, JCB 2018). Included in this discussion was description of how the di4erences between lamina TSA-seq and DamID were expected, given that DamID produces a signal more proportional to contact frequency, and independent of distance from the nuclear lamina, whereas TSA-seq produces a signal that is a function of microscopic distance from the lamina, as validated by traditional FISH.

c) We added text to describe how the nucleolar TSA-seq previously was validated by two orthogonal methods- NAD-seq and multiplexed DNA immuno-FISH:

"We successfully developed nucleolar TSA-seq, which we extensively validated using comparisons with two different orthogonal genome-wide approaches (Kumar et al., 2024)- NAD-seq, based on the biochemical isolation of nucleoli, and previously published direct microscopic measurements using highly multiplexed immuno-FISH (Su et al., 2020)."

d) We have now added panels A&B to Fig. 7 and a new Supplementary Fig. 7 demonstrating further validation of TSA-seq based on showing the high correlation between the microscopically measured distances of many hundreds of genomic sites across the genome from di4erent nuclear locales and TSA-seq scores. As discussed in response #2 below, we have used comparison of distances measured in IMR90 fibroblasts with TSA-seq scores measured in HFF fibroblasts. We would argue therefore that these correlations are a lower estimate and therefore the correlation between microscopic distances and TSAseq scores would likely have been still higher if we had performed both assays in the exact same cell line.

(2) Although these methods have been described in a bioRxiv manuscript by Kumar et al., they have not yet been published. Moreover, and surprisingly, Kumar et al., work is not cited in the current manuscript, despite its use of all TSA-seq data for NADs and LADs across the four cell lines.

The Kumar et al, Communications Biology, 2024 paper is now published and is cited properly in our revision. We apologize for this oversight and confusion our initial omission of this citation may have created. We had been writing this manuscript and the Kumar et al manuscript in parallel and had intended to co-submit. We planned to cross-reference the two at the time we co-submitted, adding the Kumar et al reference to the first version of this manuscript once we obtained a doi from bioRxiv. But we then submitted the Kumar et al manuscript several months earlier, but meanwhile forgot that we had not added the reference to our first manuscript version.

(3) Moreover, Kumar et al. did not provide any DNA-FISH validation for their methods.

As we described in our response to Reviewer 1's comment #1, we had previously provided traditional FISH validation of lamina TSA-seq in our first TSA- seq paper as well as validation by comparison with lamina DamID (Chen et al, 2018).

We also described how the nucleolar TSA-seq was extensively cross-validated in the Kumar et al, 2024 paper by both NAD-seq and the highly multiplexed immuno-FISH data from Su et al, 2020).

We note additionally that in the Kumar et al, 2024 paper the nucleolar TSA-seq was additionally validated by correlating the predicted variations in centromeric association with nucleoli across the four cell lines predicted by nucleolar TSA-seq with the variations observed by traditional immunofluorescence microscopy.

(4) Therefore, the interesting correlations described in this work are not based on robust technologies.

This comment was made in reference to the Kumar et al paper not having been published, and, as noted in responses to points #2 and #3, the paper is now published.

But we wanted to specifically note, however, that our experience is that TSA-seq has proven remarkably robust in comparison to molecular proximity assays. We've described in our responses to the previous points how TSA-seq has been cross-validated by both microscopy and by comparison with lamina DamID and nucleolar NAD-seq. We note also that in every application of TSA-seq to date, all antibodies that produced good immunostaining showed good TSA-seq results. Moreover, we obtained nearly identical results in every case in which we performed TSA-seq with different antibodies against the same target. Thus anti-SON and antiSC35 staining produced very similar speckle TSA-seq data (Chen et al, 2018), anti-lamin A and anti-lamin B staining produced very similar lamina TSA-seq data (Chen et al, 2018), antinucleolin and anti-POL1RE staining produced very similar DFC/FC nucleolar TSA-seq data (Kumar et al, 2024), and anti-MKI67IP and anti-DDX18 staining produced very similar GC nucleolar TSA-seq data (Kumar et al, 2024).

This independence of results with TSA-seq to the particular antibody chosen to label a target differs from experience with methods such as ChIP, DamID, and Cut and Run/Tag in which results can differ or be skewed based on variable distance and therefore reactivity of target proteins from the DNA or due to other factors such as non-specific binding during pulldown (ChIP) or differential extraction by salt washes (Cut and Tag).

Our experience in every case to date is that antibodies that produce similar immunofluorescence staining produce similar TSA-seq results. We attribute this robustness to the fact that TSA-seq is based only on the original immunostaining specificity provided by the primary and secondary antibodies plus the diffusion properties of the tyramide-free radical.

We've now added the following text to our revised manuscript:

"As previously demonstrated for both SON and lamin TSA-seq (Chen et al., 2018), nucleolar TSA-seq was also robust in the sense that multiple target proteins showing similar nucleolar staining showed similar TSA-seq results (Kumar et al., 2024); this robustness is intrinsic to TSA-seq being a microscopic rather than molecular proximity assay, and therefore not sensitive to the exact molecular binding partners and molecular distance of the target proteins to the DNA."

(5) An attempt to validate the data was made for SON-TSA-seq of human foreskin fibroblasts (HFF) using multiplexed FISH data from IMR90 fibroblasts (from the lung) by the Zhuang lab (Su et al., 2020). However, the comparability of these datasets is questionable. It might have been more reasonable for the authors to conduct their analyses in IMR90 cells, thereby allowing them to utilize MERFISH data for validating the TSA-seq method and also for mapping NADs and LADs.

We disagree with the reviewer's overall assessment that that the use of the IMR90 data to further validate the TSA-seq is questionable because the TSA-seq data from HFF fibroblasts is not necessarily comparable with multiplexed immuno-FISH microscopic distances measured in IMR90 fibroblasts.

In response we have now added panels to Fig. 7 and Supplementary Fig. 7, showing:

a) There is very little di4erence in correlation between speckle TSA-seq and measured distances from speckles in IMR90 cells whether we use IMR90 or HFF cells SON TSA-seq data (R² = 0.81 versus 0.76) (new Fig. 7A).

b) There is also a high correlation between lamina (R² = 0.62) and nucleolar (R² = 0.73) HFF TSA-seq and measured distances in IMR90 cells. Thus, we conclude that this high correlation shows that the multiplexed data from ~1000 genomic locations does validate the TSA-seq. These correlations should be considered lower bounds on what we would have measured using IMR90 TSA-seq data. Thus, the true correlation between distances of loci from nuclear locales and TSA-seq would be expected to be either comparable or even stronger than what we are seeing with the IMR90 versus HFF fibroblast comparisons.

c) This correlation is cell-type specific (Fig. 7B, new SFig. 7). Thus, even for speckle TSAseq, highly conserved between cell types, the highest correlation of IMR90 distances with speckle TSA-seq is with IMR90 and HFF fibroblast data. For lamina and nucleolar TSA-seq, which show much lower conservation between cell types, the correlation of IMR90 distances is high for HFF data but much lower for data from the other cell types. This further justifies the use of IMR90 fibroblast distance measurements as a proxy for HFF fibroblast measurements.

Thus, we have added the following text to the revised manuscript:

"We reasoned that the nuclear genome organization in the two human fibroblast cell lines would be sufficiently similar to justify using IMR90 multiplexed FISH data [43] as a proxy for our analysis of HFF TSA-seq data. Indeed, the high inverse correlation (R= -0.86) of distances to speckles measured by MERFISH in IMR90 cells with HFF SON TSA-seq scores is nearly identical to the inverse correlation (R= -0.89) measured instead using IMR90 SON TSA-seq scores (Fig. 7A). Similarly, distances to the nuclear lamina and nucleoli show high inverse correlations with lamina and nucleolar TSA-seq, respectively (Fig. 7A). These correlations were cell type specific, particularly for the lamina and nucleolar distance correlations, as these correlations were reduced if we used TSA-seq data from other cell types (SFig. 7A). Therefore, the high correlation between IMR90 microscopic distances and HFF TSA-seq scores can be considered a lower bound on the likely true correlation, justifying the use of IMR90 as a proxy for HFF for testing our predictions."

Reviewer #2 (Public Review):

Weaknesses:

(1) The experiments are largely descriptive, and it is difficult to draw many cause-andeffect relationships...The study would benefit from a clear and specific hypothesis.

This study was hypothesis-generating rather than hypothesis-testing in its goal. Our research was funded through the NIH 4D-Nucleome Consortium, which had as its initial goal the development, benchmarking, and validation of new genomic technologies. Our Center focused on the mapping of the genome relative to different nuclear locales and the correlation of this intranuclear positioning of the genome with functions- specifically gene expression and DNA replication timing. By its very nature, this project took a discovery-driven versus hypothesis-driven scientific approach. Our question fundamentally was whether we could gain new insights into nuclear genome organization through the integration of genomic and microscopic measurements of chromosome positioning relative to multiple different nuclear compartments/bodies and their correlation with functional assays such as RNA-seq and Repliseq.

Indeed, this study resulted in multiple new insights into nuclear genome organization as summarized in our last main figure. We believe our work and conclusions will be of general interest to scientists working in the fields of 3D genome organization and nuclear cell biology. We anticipate that each of these new insights will prompt future hypothesis-driven science focused on specific questions and the testing of cause-and-effect relationships.

However, we do want to point out that our comparison of wild-type K562 cells with the LMNA/LBR double knockout was designed to test the long-standing model that nuclear lamina association of genomic loci contributes to gene silencing. This experiment was motivated by our surprising result that gene expression differences between cell lines correlated strongly with differences in positioning relative to nuclear speckles rather than the nuclear lamina. Despite documenting in these double knockout cells a decreased nuclear lamina association of most LADs, and an increased nuclear lamina association of the “p-w-v” fiLADs identified in this manuscript, we saw no significant change in gene expression in any of these regions as compared to wild-type K562 cells. Meanwhile, distances to nuclear speckles as measured by TSA-seq remained nearly constant.

We would argue that this represents a specific example in which new insights generated by our genomics comparison of cell lines led to a clear and specific hypothesis and the experimental testing of this hypothesis.

(2) Similarly, the paper would be very much strengthened if the authors provided additional summary statements and interpretation of their results (especially for those not as familiar with 3D genome organization).

We appreciate this feedback and agree with the reviewer that this would be useful, especially for those not familiar with previous work in the field of 3D genome organization. In an earlier draft, we had included additional summary and interpretation statements in both the Introduction and Results sections. At the start of each Results section, we had also previously included brief discussion of what was known before and the context for the subsequent analysis contained in that section. However, we had thought we might be submitting to a journal with specific word limits and had significantly cut out that text.

We have now restored this text and, in certain cases, added additional explanations and context.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

Figures 1C and D. Please add the units at the values of each y-axis.

We have done that.

The representation of Figure 2C lacks clarity and is diJicult to understand. The x-axis labeling regarding the gene fraction number needs clarification.

We've modified the text to the Fig. 2C legend: "Fraction of genes showing significant di=erence in relative positioning to nuclear speckles (gene fraction, x-axis) versus log2 (HFF FKPM / H1 FKPM) (y-axis);"

"We next used live-cell imaging to corroborate that chromosome regions close to nuclear speckles, primarily Type I peaks, would show the earliest DNA replication timing." This sentence requires modification as Supplementary Figure 3F does not demonstrate that Type I peaks exhibit the earliest DNA replication timing; it only indicates that the first PCNA foci in S-phase are in proximity to nuclear speckles.

We've modified the text to: "We next used live-cell imaging to show that chromosome regions close to nuclear speckles show the earliest DNA replication timing; this is consistent with the earliest firing DNA replication IZs, as determined by Repli-seq, aligning with Type 1 peaks that are closely associated with nuclear speckles."

In Figure 5, the authors employed LaminB1-DamID to quantify LADs in LBR-KO and LMNA/LBR-DKO K562 cells. These are interesting results. However, for these experiments, it is crucial to assess LMNB1 signal at the nuclear periphery via immunofluorescence (IF) to confirm the absence of changes, ensuring that the DamID signal solely reflects contacts with the nuclear lamina. Furthermore, in this instance, their findings should be validated through DNA-FISH.

Immunostaining of LMNB1 was performed and showed a normal staining pattern as a ring adjacent to the nuclear periphery. Images of this staining were included in the metadata tied to the sequencing data sets deposited on the 4D Nucleome Data portal. We thank the reviewer for bringing up this point, and have added a sentence mentioning this result in the Results Section:

"Immunostaining against LMNB1 revealed the normal ring of staining around the nuclear periphery seen in wt cells (images deposited as metadata in the deposited sequencing data sets)."

Because both TSA-seq and DamID have been extensively validated by FISH, as detailed in our previous responses to the public reviewer comments, we feel it is unnecessary to validate these findings by FISH.

p-w-v-fiLADs should be labelled in Figure 5B.

We've added labeling as suggested.

"The consistent trend of slightly later DNA replication timing for regions (primarily p-w-v fiLADs) moving closer to the lamina" is not visible in the representation of the data of Figure 5G.

We did not make a change as we believed this trend was apparent in the Figure.

To reduce the descriptive nature of the data, it would be pertinent to conduct H3K9me3 and H3K27me3 ChIP-seq analyses in both the parental and DKO mutant cells. This would elucidate whether p-w-v-fiLADs and NADs anchoring to the nuclear lamina undergo changes in their histone modification profile.

We believe further analysis of the reasons underlying these shifts in positioning, including such ChIP-seq or equivalent analysis, is of interest but beyond the scope of this publication. We see such measurements as the beginning of a new story but insuJicient alone to determine mechanism. Therefore we believe such experiments should be part of that future study.

The description of Figure 7 lacks clarity. Additionally, it appears that TSA-seq for NADs and LADs may not be universally applicable across all cell types, particularly in flat cells, whereas DamID scores demonstrate less variation across cell lines, as also stated by the authors.

TSA-seq is a complement to rather than a replacement for either DamID or NAD-seq. TSAseq reports on microscopic distances whereas both DamID and NAD-seq instead are more proportional to contact frequency with the nuclear lamina or nucleoli, respectively, and insensitive to distances of loci away from the lamina or nucleoli. Thus, TSA-seq provides additional information based on the intrinsic diJerences in what TSA-seq measures relative to molecular proximity methods such as DamID or NAD-seq. The entire point is that the convolution of the exponential point-spread-function of the TSA-seq with the shape of the nuclear periphery allows us to distinguish genomic regions in the equatorial plane versus the top and bottom of the nuclei. The TSA-seq is therefore highly "applicable" when properly interpreted in discerning new features of genome organization. As we stated in the revised manuscript, the lamina DamID and TSA-seq are complementary and provide more information together then either method along. The same is true for the NAD-seq and nucleolar TSA-seq comparison, as described in more detail in the Kumar, et al, 2024 paper.

Introduction:

The list of methodologies for mapping genomic contacts with nucleoli (NADs) should also include recent technologies, such as Nucleolar-DamID (Bersaglieri et al., PMID: 35304483), which has been validated through DNA-FISH.

We did not include nucleolar DamID in the mention in the Introduction of methods for identifying diJerential lamina versus nucleolar interactions of heterochromatin- either from our own collaborative group or from the cited reference- because we did not have confidence in the accuracy of this method in identifying NADs. In the case of the published nucleolar DamID from our collaborative group, published in Wang et al, 2021, we later discovered that despite apparent agreement of the nucleolar DamID with a small number of published FISH localization the overall correlation of the nucleolar DamID with nucleolar localization was poor. As described in detail in the Kumar et al, 2024 publication, this poor correlation of the nucleolar DamID was established using three orthogonal methods- nucleolar TSA-seq, NAD-seq, and multiplexed immuno-FISH measurements from ~1000 genomic locations. Instead, we found that this nucleolar DamID showed high correlation with lamina DamID. We note that many strong NADs are also LADs, which we think is why validation with only several FISH probes is inadequate to demonstrate overall validation of the approach.

We could not compare our nucleolar-DamID data in human cells with the alternative nucleolar-DamID results cited by the reviewer which were performed in mouse cells. We note that in this paper the nucleolar DamID FISH validation only included several putative NAD chromosome regions and, I believe, one LAD region. However, our initial comparison of the nucleolar DamID cited by the reviewer with unpublished TSA-seq data from mouse ESCs produced by the Belmont laboratory and with NAD-seq data from the Kaufman laboratory shows a similar lack of correlation of the nucleolar DamID signal with nucleolar TSA-seq and NAD-seq, as well as multiplexed immuno-FISH data from the Long Cai laboratory, as we saw in our analysis of own nucleolar DamID data in human cells.

We have added explanation concerning the lack of correlation of our nucleolar DamID with orthogonal measurements of nucleolar proximity in the added text (below) to our revised manuscript:

"Nucleolar DamID instead showed broad positive peaks over large chromatin domains, largely overlapping with LADs mapped by LMNB1 DamID (Wang et al., 2021). However, this nucleolar DamID signal, while strongly correlated with lamin DamID, showed poor correlation with either NAD-seq or nucleolar distances mapped by multiplexed immunoFISH (Kumar et al., 2024). We suspect the problem is that with molecular proximity assays the output signals are disproportionally dominated by the small fraction of target proteins juxtaposed in su=icient proximity to the DNA to produce a signal rather than the amount of protein concentrated in the target nuclear body. "

Our mention of nucleolar TSA-seq was in the context of why we focused on nucleolar TSAseq and excluded our own nucleolar DamID. We chose not to discuss the second nucleolar DamID method cited above 1) because it was not appropriate to our discussion of our own experimental approach and 2) also because we cannot yet make a definitive statement of its accuracy for nucleolar mapping.

Reviewer #2 (Recommendations For The Authors):

(1) The authors start the manuscript by describing the 'radial genome organization' model and contrast it with the 'binary model' of genome organization. It would be helpful for the authors to contextualize their results a bit more with regard to these two diJerent models in the discussion.

We have added several sentences in the first paragraph of the Discussion to accomplish this contextualization. The new paragraph reads:

"Here we integrated imaging with both spatial (DamID, TSA-seq) and functional (Repli-seq, RNA-seq) genomic readouts across four human cell lines. Overall, our results significantly extend previous nuclear genome organization models, while also demonstrating a cell-type dependent complexity of nuclear genome organization. Briefly, in contrast to the previous radial model of genome organization, we reveal a primary correlation of gene expression with relative distances to nuclear speckles rather than radial position. Additionally, beyond a correlation of nuclear genome organization with radial position, in cells with flat nuclei we show a pronounced correlation of nuclear genome organization with distance from the equatorial plane. In contrast to previous binary models of genome organization, we describe how both iLAD / A compartment and LAD / B compartment contain within them smaller chromosome regions with distinct biochemical and/or functional properties that segregate di=erentially with respect to relative distances to nuclear locales and geometry."

(2) Data should be provided demonstrating KO of LBR and LMNA - immunoblotting for both proteins would be one approach. In addition, it would be helpful to provide additional nuclear morphology measurements of the DKO cells (volume, surface area, volume of speckles/nucleoli, number of speckles/nucleoli).

We've added additional description describing the generation and validation of the KO lines:

"To create LMNA and LBR knockout (KO) lines and the LMNA/LBR double knockout (DKO) line, we started with a parental "wt" K562 cell line, clone #17, expressing an inducible form of Cas9 (Brinkman et al., 2018). The single KO and DKO were generated by CRISPR-mediated frameshift mutation according to the procedure described previously (Schep et al., 2021). The "wt" K562 clone #17 was used for comparison with the KO clones.

The LBR KO clone, K562 LBR-KO #19, was generated, using the LBR2 oligonucleotide GCCGATGGTGAAGTGGTAAG to produce the gRNA, and validated previously, using TIDE (Brinkman et al., 2014) to check for frameshifts in all alleles as described elsewhere (Schep et al., 2021). The LMNA/LBR DKO, K562 LBR-LMNA DKO #14, was made similarly, starting with the LBR KO line and using the combination of two oligonucleotides to produce gRNAs:

LMNA-KO1: ACTGAGAGCAGTGCTCAGTG, LMNA-KO2: TCTCAGTGAGAAGCGCACGC.

Additionally, the LMNA KO line, K562 LMNA-KO #14, was made the same way but starting with the "wt" K562 cell line. Validation was as described above; additionally, for the new LMNA KO and LMNA/LBR DKO lines, immunostaining showed the absence of anti-LMNA antibody signal under confocal imaging conditions used to visualize the wt LMNA staining while the RNA-seq from these clones revealed an ~20-fold reduction in LMNA RNA reads relative to the wt K562 clone."

As suggested, we also added morphological data for the DKO line in a modified SFig.5.

(3) The rationale for using LMNB1 TSA-seq and LMNB1 DAMID is not immediately clear. The LMNB1 TSA-seq is more variable across cell types and replicates than the DAMID. Could the authors please compare the datasets a bit more to understand the diJerences? For example, the authors demonstrate that "40-70% of the genome shows statistically significant diJerences in Lamina TSA-seq over regions 100 kb or larger, with most of these regions showing little or no diJerences in speckle TSA-seq scores." If the LMNB1 DAMID data is used for this analysis or Figure 2D, is the same conclusion reached? Also, in Figure 6, the authors conclude that C1 and C3 LAD regions are enriched for constitutive LADs, while C2 and C4 LAD regions are fLADs. This is a bit surprising because the authors and others have previously shown that constitutive LADs have higher LMNB1 contact frequency than facultative LADs (Kind, et al Cell 2015, Figure 3C).

Indeed, in the first TSA-seq paper (Chen et al, 2018) we did observe that cLADs had the highest LMNB TSA-seq scores; this was for K562 cells with round nuclei in which there is therefore no diJerence in lamina TSA-seq scores produced by nuclear shape over the entire nucleus.

However, there are diJerences between TSA-seq and DamID in terms of what they measure and we refer the reviewer to the first TSA-seq paper (Chen et al, 2018) that explains in greater depth these diJerences. This first paper explains how DamID is indeed related to contact frequency but how the TSA-seq instead estimates mean distances from the target, in this case the nuclear lamina. This is because the diJusion of tyramide free radicals from the site of their constant HRP production produces an exponential decay gradient of tyramide free radical concentration at steady state.

We have summarized these diJerences in in text we have added to introduce both DamID and TSA-seq in the second Results section:

"DamID is a well-established molecular proximity assay; DamID applied to the nuclear lamina divides the genome into lamina-associated domains (LADs) versus nonassociated “inter-LADs” or “iLADs” (Guelen et al., 2008; van Steensel and Belmont, 2017). In contrast, TSA-seq measures relative distances to targets on a microscopic scale corresponding to 100s of nm to ~ 1 micron based on the measured diJusion radius of tyramide-biotin free-radicals (Chen et al., 2018)... While LMNB1 DamID segments LADs most accurately, lamin TSA-seq provides distance information not provided by DamID- for example, variations in relative distances to the nuclear lamina of diJerent iLADs and iLAD regions. These diJerences between the LMNB1 DamID and LMNB TSA-seq signals are also crucial to a computational approach, SPIN, that segments the genome into multiple states based on their varying nuclear localization, including biochemically and functionally distinct lamina-associated versus near-lamina states (Consortium et al., 2024; Wang et al., 2021).

Thus, lamin DamID and TSA-seq complement each other, providing more information together than either one separately."

We note that these diJerences in lamina DamID and TSA-seq are crucial to being able to gain additional information by comparing variations in the lamina TSA-seq for LADs in Figs. 6&7. See our response to point (4) below, for further explanation.

(4) In 7B/C, the authors show that the highest LMNB1 regions in HFF are equator of IMR90s. However, in Figure 7G, their cLAD score indicates that constitutive LADs are not at the equator. This is a bit surprising given the point above and raises the possibility that SON signals (as opposed to LMNB1 signals) might be more responsible for correlation to localization relative to the equator. Hence, it might be helpful if the authors repeat the analyses in Figures 7B/C in regions with diJering LMNB1 signals but similar SON signals (and vice versa).

Again, this is based on the apparent assumption by the reviewer that DamID and TSA-seq work the same way and measure the same thing. But as explained above in the previous point, this is not true.

In our first TSA-seq paper (Chen et al, 2018) we showed how we could use the exponential decay point-spread-function produced by TSA, measured directly by light microscopy, to convert sequencing reads from the TSA-seq into a predicted mean distance from nuclear speckles, approximated as point sources. These mean distances predicted from the SON TSA-seq data agreed with measured FISH distances to nuclear speckles to within ~50 nm for a set of DNA probes from diJerent chromosome regions. Moreover, varying TSA staining conditions changed the decay constants of this exponential decay, thus producing diJerences in the SON TSA-seq signals. By using these diJerent exponential decay functions to convert the TSA-seq scores from these independent data sets to estimated distances from nuclear speckles, we again observed a distance residual of ~50 nm; in this case though this distance residual of ~50 nm represented the mean residual observed genome-wide. This gives us great confidence that the TSA-seq is working as we have modeled it.

As we mentioned in our response to point 3 above, we did see the highest LMNB TSA-seq signal for cLADs in K562 cells with round nuclei (Chen et al, 2018).

But as we now show in our simulation performed in this paper for Fig. 7, the observed tyramide free radical exponential decay gradient convolved with the flat nuclear lamina shape produces a higher equatorial LMNB1 TSA-seq signal for LADs at the equatorial plane. We confirmed that LADs with this higher TSA-seq signal were enriched at the equatorial plane by mining the multiplexed IMR90 imaging data. Similar mining of the multiplexed FISH IMR90 data showed localization of cLADs away from the equatorial plane.

We are not clear about the rationale for what the reviewer is suggesting about SON signals "being more responsible for correlation to localization to the equator". We have provided an explanation for the higher lamina TSA-seq scores for LADs near the equator based on the measured spreading of the tyramide free radicals convolved with the eJect of the nuclear shape. This makes a prediction that the observed variation in lamina TSA-seq scores for LADs with similar DamID scores is related to their positioning relative to the equatorial plane as we then validated through our mining of the IMR90 multiplexed FISH data.

(5) FISH of individual LADs, v-fiLADs, and p-w-v-fiLADs relative to the lamina and speckle would be helpful to understand their relative positioning in control and LBR/LMNA double KO cells. This would significantly bolster the claim that "histone mark enrichments..more precisely revealed the diJerential spatial distribution of LAD regions...".

Adequately testing these predictions made from the lamina/SON TSA-seq scatterplots by direct FISH measurements would require measurements from large numbers of diJerent chromosome regions through a highly multiplexed immuno-FISH approach. We are not equipped currently in any of our laboratories to do such measurements and we leave this therefore for future studies.

Rather our statement is based on our use of TSA-seq analyzed through these 2D scatterplots and should be valid to the degree that our TSA-seq measurements do indeed correlate with microscopy derived distances.

However, we do now include demonstration of a high correlation of speckle, lamina, and nucleolar TSA-seq with highly multiplexed immuno-FISH measurement of distances to these locales in a revised Fig. 7. The high correlation shown between the TSA-seq scores and measured distances does therefore add additional support to our claim that the reviewer is discussing, even without our own multiplexed FISH validation.

(6) "In contrast, genes within genomic regions which in pair-wise comparisons of cell lines show a statistically significant diJerence in lamina TSA-seq show no obvious trend in their expression diJerences (Figure 2C).". This appears to be an overstatement based on the left panel of 2D.

We do not follow the reviewer's point. In Fig. 2C we show little bias in the diJerences in gene expression between the two cell types for regions that showed diJerences in lamina TSA-seq. The reviewer is suggesting something otherwise based on their impression, not explicitly stated, of the left panel of Fig. 2D. But we see similar shades of blue extending vertically at low SON values and similar shades of red extending vertically at high SON values, suggesting a correlation of gene expression only with the SON TSA-seq score but not with the LMNB1 TSA-seq score displayed on the y-axis. This is also consistent with the very small and/or insignificant correlation coeJicients measured in our linear model relating diJerences in LMNB1 TSA-seq to diJerences in expression but the large correlation coeJicient observed for SON TSA-seq (Fig. 2E). Thus, we see Fig. 2C-E as self-consistent.

(7) In the section on "Polarity of Nuclear Genome Organization" - "....Using the IMR90 multiplexed FISH data set [43]...." - The references are not numbered.

We thank the reviewer for this correction.

(8) I believe there is an error in the Figure 7B legend. The descriptions of Cluster 1 and 2 do not match those indicated in the figure.

We again thank the reviewer for this correction.

eLife Assessment

This important study allows for a better understanding of anthelmintic drug resistance in nematodes. The authors provide a detailed analysis of the role of UBR-1 and its underlying mechanism in ivermectin resistance using convincing behavioural and genetic experiments with C. elegans. Although the authors have addressed the concerns of the reviewers, it would be prudent for the authors to disclose the Dyf phenotype in ubr-1 mutants. The authors should at the very least report the Dyf phenotype and the experiment on which they base the argument that the Dyf phenotype does not affect their conclusions.

Reviewer #1 (Public review):

Summary:

The drug Ivermectin is used to effectively treat a variety of worm parasites in the world, however resistance to Ivermectin poses a rising challenge for this treatment strategy. In this study, the authors found that loss of the E3 ubiquitin ligase UBR-1 in the worm C. elegans results in resistance to Ivermectin. In particular, the authors found that ubr-1 mutants are resistant to the effects of Ivermectin on worm viability, body size, pharyngeal pumping and locomotion. The authors previously showed that loss of UBR-1 disrupts homeostasis of the amino acid and neurotransmitter glutamate resulting in increased levels of glutamate in C. elegans. Here, the authors found that the sensitivity of ubr-1 mutants to Ivermectin can be restored if glutamate levels are reduced using a variety of different methods. Conversely, treating worms with exogenous glutamate to increase glutamate levels also results in resistance to Ivermectin supporting the idea that increased glutamate promotes resistance to Ivermectin. The authors found that the primary known targets of Ivermectin, glutamate-gated chloride channels (GluCls), are downregulated in ubr-1 mutants providing a plausible mechanism for why ubr-1 mutants are resistant to Ivermectin. Although it is clear that loss of GluCls can lead to resistance to Ivermectin, this study suggests that one potential mechanism to decrease GluCl expression is via disruption of glutamate homeostasis that leads to increased glutamate. This study suggests that if parasitic worms become resistant to Ivermectin due to increased glutamate, their sensitivity to Ivermectin could be restored by reducing glutamate levels using drugs such as Ceftriaxone in a combination drug treatment strategy.

Strengths:

- The use of multiple independent assays (i.e., viability, body size, pharyngeal pumping, locomotion and serotonin-stimulated pharyngeal muscle activity) to monitor the effects of Ivermectin<br /> - The use of multiple independent approaches (got-1, eat-4, ceftriaxone drug, exogenous glutamate treatment) to alter glutamate levels to support the conclusion that increased glutamate in ubr-1 mutants contributes to Ivermectin resistance

Weaknesses:

- The primary target of Ivermectin is GluCls so it is not surprising that alteration of GluCl expression or function would lead to Ivermectin resistance<br /> - It remains to be seen what percent of Ivermectin resistant parasites in the wild have disrupted glutamate homeostasis as opposed to mutations that more directly decrease GluCl expression or function.

Comments on revisions: All my concerns have been addressed by the authors.

Reviewer #2 (Public review):

Summary:

The authors provide a very thorough investigation on the role of UBR-1 in anthelmintic resistance using the non-parasitic nematode, C. elegans. Anthelmintic resistance to macrocyclic lactones is a major problem in veterinary medicine and likely just a matter of time until resistance emerges in human parasites too. Therefore, this study providing novel insight into the mechanisms of ivermectin resistance is particularly important and significant.

Strengths:

The authors use very diverse technologies (behavior, genetics, pharmacology, genetically encoded reporters) to dissect the role of UBR-1 in ivermectin resistance. Deploying such a comprehensive suite of tools and approaches provides exceptional insight into the mechanism of how UBR-1 functions in terms of ivermectin resistance.

Weaknesses:

I do not see any major weaknesses in this study. My only concern is whether the observations made by the authors would translate to any of the important parasitic helminths in which resistance has naturally emerged in the field. This is always a concern when leveraging a non-parasitic nematode to shed light on a potential mechanism of resistance of parasitic nematodes, and I understand that it is likely beyond the scope of this paper to test some of their results in parasitic nematodes.

Comments on revisions: The authors have now addressed all my concerns.

Reviewer #3 (Public review):

Summary:

Li et al propose to better understand the mechanisms of drug resistance in nematode parasites by studying mutants of the model roundworm C. elegans that are resistant to the deworming drug ivermectin. They provide compelling evidence that loss-of-function mutations in the E3 ubiquitin ligase encoded by the UBR-1 gene make worms resistant to the effects of ivermectin (and related compounds) on viability, body size, pharyngeal pumping rate, and locomotion and that these mutant phenotypes are rescued by a UBR-1 transgene. They propose that the mechanism is resistance is indirect, via the effects of UBR-1 on glutamate production. They show mutations (vesicular glutamate transporter eat-4, glutamate synthase got-1) and drugs (glutamate, glutamate uptake enhancer ceftriaxone) affecting glutamate metabolism/transport modulate sensitivity to ivermectin in wild type and ubr-1 mutants. The data are generally consistent with greater glutamate tone equating to ivermectin resistance. Finally, they show that manipulations that are expected to increase glutamate tone appear to reduce expression of the targets of ivermectin, the glutamate-gated chloride channels, which is known to increase resistance.

There is a need for genetic markers of ivermectin resistance in livestock parasites that can be used to better track resistance and to tailor drug treatment. The discovery of UBR-1 as a resistance gene in C. elegans will provide a candidate marker that can be followed up in parasites. The data suggest Ceftriaxone would be a candidate compound to reverse resistance.

Strengths:

The strength of the study is the thoroughness of the analysis and the quality of the data. There can be little doubt that ubr-1 mutations do indeed confer ivermectin resistance. The use of both rescue constructs and RNAi to validate mutant phenotypes is notable. Further, the variety of manipulations they use to affect glutamate metabolism/transport makes a compelling argument for some kind of role for glutamate in resistance.

Weaknesses:

The use of single ivermectin dose assays can be misleading. A response change at a single dose shows that the dose-response curve has shifted, but the response is not linear with dose, so the degree of that shift may be difficult to discern and may result from a change in slope but not EC50.

eLife Assessment

This manuscript describes a resource detailing the econstitution of Holothuria glaberrima gut following self-evisceration in response to a potassium chloride injection, using scRNAseq and fluorescent RNA localization in situ. It provides some new findings about organ regeneration, as well as the origins of pluripotent cells, and places these findings in the context of regeneration across species. The paper's schematic model and HCR images are a valuable foundation for future work. The authors provide convincing RNA localization images to validate their data and to provide spatial context. These validation experiments are of good quality but remain challenging to connect to the complex spatial organization of complex tissues. This resource will be of interest to the field of regeneration, particularly in invertebrates, but also in comparative studies in other species, including evolutionary studies.

Reviewer #1 (Public review):

Summary:

Joshua G. Medina-Feliciano et al. investigated the single-cell transcriptomic profile of holoturian regenerating intestine following evisceration, a process used to expel their viscera in response to predation. Using single-cell RNA-Sequencing and standard analysis such as "Find cluster markers", "Enrichment analysis of Gene Ontology" and "RNA velocity", they identify 13 cell clusters and their potential cell identity. Based on bioinformatic analysis they identified potentially proliferating clusters and potential trajectories of cell differentiation. This manuscript represents a useful dataset that can provide candidate cell types and cell markers for more in-depth functional analysis of the holoturian intestine regeneration.

The conclusions of this paper are supported only by bioinformatic analyses since the in vivo validation through HCR is not sufficient to support them.

Strengths:

- The Authors are providing a single-cell dataset obtained from sea cucumbers regenerating their intestines. This represents the first fundamental step to an unbiased approach to better understand this regeneration process and the cellular dynamics taking part in it.

- The Authors run all the standard analyses providing the reader with a well digested set of information about cell clusters, potential cell types, potential functions and potential cell differentiation trajectories.

Weaknesses:

- The Authors frequently report the percentage of cells with a specific feature (either labelled or expressing a certain gene or belonging to a certain cluster). This number can be misleading since that is calculated after cell dissociation and additional procedures (such as staining or sequencing and dataset cleanup) that can heavily bias the ratio between cell types. Similarly, the Authors cannot compare cell percentage between anlage and mesentery samples since that can be affected by technical aspects related to cell dissociation, tissue composition and sequencing depth.

- The Authors did not validate all the clusters.

- There is no validation of the trajectory analysis and there is no validation of the proliferating cluster with H3P or EdU co-labeling.

Reviewer #2 (Public review):

Summary:

This research offers a comprehensive analysis of the regenerative process in sea cucumbers and builds upon decades of previous research. The approach involves a detailed examination using single-cell sequencing, making it a crucial reference paper while shedding new light on regeneration in this organism.

Strengths:

Detailed analysis of single-cell sequencing data and high-quality RNA localization images provide significant new insights into regeneration in sea cucumbers and, more broadly, in animals. Identifying a proliferating cluster of cells is very interesting and may open avenues to identify the cell lineage history and deeper molecular properties of the cells that regenerate the intestine.

Weaknesses:

The spatial context of the RNA localization images is challenging to interpret in this spatially complex tissue organization. Although the authors have taken care to perform RNA localization staining, it is still challenging to relate these data to their schematic model. This is only a minor weakness that will almost certainly be clarified by future work from the authors as they follow up on findings.

Reviewer #3 (Public review):

Summary:

The authors have done a good job at creating a "resource" paper for the study of gut regeneration in sea cucumbers. They present a single-cell RNAseq atlas for the reconstitution of Holothuria glaberrima gut following self-evisceration in response to a potassium chloride injection. The authors provide data characterizing cellular populations and precursors of the regenerating anlage at 9 days post evisceration. As a "Tools and Resources" contribution to eLife, this work, with some revisions, could be appropriate. It will be impactful in the fields of regeneration, particularly in invertebrates, but also in comparative studies in other species, including evolutionary studies. Some of these comparative studies could extend to vertebrates and could therefore impact regenerative medicine in the future.

Strengths:

• Novel and useful information for a model organism and question for which this type of data has not yet been reported<br /> • Single-cell gene expression data will be valuable for developing testable hypotheses in the future<br /> • Marker genes for cell types provided to the field<br /> • Interesting predictions about possible lineage relationships between cells during sea cucumber gut regeneration<br /> • Authors have done a good job in the revision of making sure not to overstate the lineage claims in absence of definitive lineage-tracing experiments<br /> • Authors have improved the figures and the overall readability of the figures and text

Specific questions:

- Is there any way to systematically compare these cells to evolutionarily-diverged cells in distant relatives to sea cucumbers? Or even on a case-by-case basis? For example, is there evidence for any of these transitory cell types to have correlate(s) in vertebrate gut regeneration?

• Authors acknowledged this would be interesting and important, but they say in the response document this is outside the scope of the current manuscript and more data would be needed to do this well.

- Line 808: The authors may make a more accurate conclusion by saying that the characteristics are similar to blastemas or behaves like a blastema rather than it is blastema. There is ambiguity about the meaning of this term in the field, but most researchers seem to currently have in mind that the "blastema" definitions includes a discrete spatial organization of cells, and here these cells are much more spread out. This could be a good opportunity for the authors to engage in this dialogue, perhaps parsing out the nuances of what a "blastema" is, what the term has traditionally referred to, and how we might consider updating this term or at least re-framing the terminology to be inclusive of functions that "blastemas" have traditionally had in the literature and how they may be dispersed over geographical space in an organism more so than the more rigid, geographically-restricted definition many researchers have in mind. However, if the authors choose to elaborate on these issues, those elaborations do belong in the discussion, and the more provisional terminology we mention here could be used throughout the paper until that element of the revised discussion is presented. We would welcome the authors to do this as a way to point the field in this direction as this is also how we view the matter. For example, some of the genes whose expression has been observed to be enriched following removal of brain tissue in axolotls (such as kazald2, Lust et al.), are also upregulated in traditional blastemas, for instance, in the limb, but we appreciate that the expression domain may not be as localized as in a limb blastema. Additionally, since there is now evidence that some aspects of progenitor cell activation even in limb regeneration extend far beyond the local site of amputation injury (Johnson et al., Payzin-Dogru et al.), there is an opportunity to connect the dots and make the claim that there could be more dispersion of "blastema function" than previously appreciated in the field. Diving a bit more into these nuances may also enable a better conceptual framework of how blastema function may evolve across vast evolutionary time and between different injury contexts in super-regenerative organisms.

• Authors addressed this comment and agree it is interesting, but given how much territory they had to cover and space limitations, they will save this type of discussion and comparative theoretical work for the future.

Overall, the manuscript is much improved.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews:

Reviewer #1:

The entire study is based on only 2 adult animals, that were used for both the single cell dataset and the HCR. Additionally, the animals were caught from the ocean preventing information about their age or their life history. This makes the n extremely small and reduces the confidence of the conclusions. 

This statement is incorrect.  While the scRNAseq was indeed performed in two animals (n=2), the HCR-FISH was performed in 3-5 animals (depending on the probe used).  These were different animals from those used for the scRNAseq.  The number of animals used has now been included in the manuscript.

All the fluorescent pictures present in this manuscript present red nuclei and green signals being not color-blind friendly. Additionally, many of the images lack sufficient quality to determine if the signal is real. Additional images of a control animal (not eviscerated) and of a negative control would help data interpretation. Finally, in many occasions a zoomed out image would help the reader to provide context and have a better understanding of where the signal is localized. 

Fluorescent photos have been changed to color-blind friendly colors.  Diagrams, arrows and new photos have been included as to guide readers to the signal or labeling in cells. Controls for HCR-FISH and labeling in normal intestines have been included.  

Reviewer #2:

The spatial context of the RNA localization images is not well represented, making it difficult to understand how the schematic model was generated from the data. In addition, multiple strong statements in the conclusion should be better justified and connected to the data provided.

As explained above we have made an effort to provide a better understanding of the cellular/tissue localization of the labeled cells. Similarly, we have revised the conclusions so that the statements made are well justified.

Reviewer #3:

Possible theoretical advances regarding lineage trajectories of cells during sea cucumber gut regeneration, but the claims that can be made with this data alone are still predictive.

We are conscious that the results from these lineage trajectories are still predictive and have emphasized this in the text. Nonetheless, they are important part of our analyses that provide the theoretical basis for future experiments.

Better microscopy is needed for many figures to be convincing. Some minor additions to the figures will help readers understand the data more clearly.

As explained above we have made an effort to provide a better understanding of the cellular/tissue localization of the labeled cells.  

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

-  Page 4, line 70-81: if the reader is not familiar with holothurian anatomy and regeneration process, this section can be complicated to fully understand. An illustration, together with clear definitions of mesothelium, coelomic epithelium, celothelium and luminal cells would help the reader. 

A figure (now Figure 1) detailing the holothurian anatomy of normal and regenerating animals has been added. A figure detailing the intestinal regeneration process has also been included (S1).

-  Page 5 line 92-104: this paragraph could be shortened. It would be more important to explain what the main question is the Authors would like to answer and why single cell would be the best technique to answer it, than listing previous studies that used scRNA-Seq. 

The paragraph has been shortened and the focus has been shifted to the question of cellular components of regenerative tissues in holothurians.

-  Page 6, line 125-127 and line 129-132: this belongs to the method section. 

This information is now provided in the Materials and Methods section.

-  Page 11, line 210-217: this belongs to the discussion. 

This section has now been included in the Discussion.

-  How many mesenteries are present in one animal? 

This has now been included as part of Figure S1.

-  In the methods there are no information about the quality of the dataset and the sequencing and the difference between the 2 samples coming from the 2 animals. How many cells from each sample and which is the coverage? The Authors provided this info only between mesentery and anlage but not between animals. 

We have added additional information about the sequencing statistics in S4 Fig and S15 Table. Description has also been added in the methods in lines 922-926 under Single Cell RNA Sequencing and Data Analysis section.

-  The result section "An in-depth analysis of the various cluster..." is particularly long and very repetitive. I would encourage to Authors to remove a lot of the details (list of genes and GO terms) that can be found in the figures and stressed only the most important elements that they will need to support their conclusions. Having full and abbreviated gene names and the long list of references makes the text difficult to read and it is challenging to identify the main point that the Authors are trying to highlight. 

This section has been abbreviated.

-  Figure 1: I would suggest adding a graph of holothurian anatomy before and after the evisceration to provide more context of the process we are looking at and remove 1C. 

Information on the holothurian anatomy has been included in a new Fig 1 and in supplementary figure S1

-  Figure 2: I would suggest removing this figure that is redundant with Figure 3 and several genes are not cluster specific. Figure 3 is doing a better job in showing similar concepts. 

Figure 2 was removed and placed in the Supplement section. 

- In figure 3 how were the 3 cell types defined? Was this done manually or through a bioinformatic analysis? 

The cell definition was done following the analysis of the highly expressed transcripts and comparisons to what has been shown in the scientific literature.

-  Figure 2O shows that one of the supra-cluster is made of C2, C7, C6 and C10. This contradicts the text page 9, line 195. 

The transcript chosen for this figure gives the wrong idea that these 4 clusters are similar. We have now addressed this in the manuscript.

-  Figure 4A and 4C: if these are representing a subset of Figure 3, they should be removed in one or the other. The same comment is valid also for Figures 5, 6 and 7. In general the manuscript is very redundant both in terms of Figures and text. 

These are indeed subsets of Fig 3 that were added with the purpose of clarifying the findings, however, in view of the reviewer’s comment we have deleted the redundant information from all figures.

-  Figure 9: since the panels are not in order, it is difficult to follow the flow of the figure.  - All UMAP should have the number of the cluster on the UMAP itself instead of counting only on the color code in order to be color-blind friendly. 

The figure has been modified and clusters are now identified in the UMAP by their number.

-  Figure S1F seems acquired in very different conditions compared to the other images in the same figure. 

Fig S1F (now S2 Fig) is an overlay of fluorescent immune-histochemistry (UV light detected) with “classical” toluidine blue labeling (visible light detected).  This has now been explained in the figure legend.

-  Table S7 is lacking some product numbers. 

The toluidine blue product number has now been added to the table.  The antibodies that lack product number correspond to antibodies generated in our lab  and described in the references provided.

-  The discussion is pretty long and partially redundant with the result section. I would encourage the Authors to shorten the text and shorten paragraphs that have repeating information.  - It might be out of the scope of the Authors but the readers would benefit from having a manuscript that focuses more on the novel aspects discovered with the single-cell RNA-Seq and then have a review that will bring together all the literature published on this topic and integrating the single-cell data with everything that is known so far. 

We have tried to shorten the discussion by eliminating redundant text.

Reviewer #2 (Recommendations For The Authors): 

-  An intriguing finding is the lack of significant difference in the cell clusters between the anlage and mesentery during regeneration. This discovery raises important questions about the regenerative process. The authors should provide a more detailed explanation of the implications of this finding. For example, does it suggest that both organs contribute equally to the regenerated tissues? 

The lack of significant differences in the cell clusters between the anlage and the mesentery is somewhat surprising but can be explained by two different facts. First, we have previously shown that many of the cellular processes that take place in the anlage, including cell proliferation, apoptosis, dedifferentiation and ECM remodeling occur in a gradient that begins at the tip of the mesentery where the anlage forms and extends to various degrees into the mesentery.  Similarly, migrating cells move along the connective tissue of the mesentery to the anlage.  Thus, there is no clear partition of the two regions that would account for distinct cell populations associated with the regenerative stage.  Second, the two cell populations that would have been found in the mesentery but not in the regenerating anlage, mature muscle and neurons, were not dissociated by our experimental protocol as to allow for their sequencing.  Our current experiments are being done using single nuclei RNA sequencing to overcome this hurdle. This has now been included in the discussion.

-  Proliferating cells are obviously important to the study of regeneration as it is assumed these form the regenerating tissue. The authors describe cluster 8 as the proliferative cells. Is there evidence of proliferation in other cell types or are these truly the only dividing cells? Is c8 of multiple cell types but the clustering algorithm picks up on the markers of cell division i.e. what happens if you mask cell division markers - does this cluster collapse into other cluster types? This is important as if there is only one truly proliferating cell type then this may be the origin of the regenerative tissues and is important for this study to know this. 

As the reviewer highlights, we also believe this to be an important aspect to discuss. We have addressed this in the manuscript discussion with the following: “Our data suggest that there appears to be a specific population of only proliferative cells (C8) characterized by a large number of cell proliferation genes, which can be visualized by the top genes shown in Fig 3. These cell proliferation genes are specific to C8, with minimum representation in other populations. Interestingly, as mentioned before C8 expresses at lower levels many of the genes of other coelomic epithelium populations. Nevertheless, even if we mask the top 38 proliferation genes (not shown), this cluster is maintained as an independent cluster, suggesting that its identity is conferred by a complex transcriptomic profile rather than only a few proliferation-related genes. Therefore, the identity and potential role of C8 could be further described by two distinct alternatives: (1) cells of C8 could be an intermediate state between the anlage precursor cells (discussed below) and the specialized cell populations or (2) cells of C8 are the source of the anlage precursor populations from which all other populations arise. The pseudotime data is certainly complex and challenging to interpret with our current dataset, yet the RNA velocity analysis showed in Fig 11B would suggests that cells of C8 transition into the anlage precursor populations, rather than being an intermediate state. This is also supported by the Slingshot pseudotime analysis that incorporates C8 (S13 Fig).

Nevertheless, additional experiments are needed to confirm this hypothesis.”

-  The schematic model presented in Fig 10 is essential for clarifying the paper's findings and will provide a crucial baseline model for future research. However, the comparison of the data shown in the HCR figures with the schematic is challenging due to the lack of spatial context in the HCR figures. The authors should find a way to provide better context in the figures, such as providing two-color in situ images to compare spatial relationships of cell types and/or including lower resolution and side-by-side fluorescent and bright field images if possible. 

The figure has been modified to explain the spatial arrangement of the tissues.

The authors make several strong statements in the discussion that weren't well connected to the findings in the data. Specifically: 

“Regardless of which cell population is responsible for giving rise to the cells of the regenerating intestine, our study reveals that the coelomic epithelium, as a tissue layer, is pluripotent.” 

This has now been expanded to better explain the statement.

738 “…we postulate that cells from C1 stand as the precursor cell population from which the rest of the cells in the coelomic epithelium arise”. 

This has now been expanded to better explain the statement.

748 “differentiation: muscle, neuroepithelium, and coelomic epithelium cells. We also propose the presence of undifferentiated and proliferating cell populations in the coelomic epithelia, which give rise to the cells in this layer…”

This has now been expanded to better explain the statement.

777 “amphibians, the cells of the holothurian anlage coelomic epithelium are proliferative undifferentiated cells and originated via a dedifferentiation process…”

This has now been expanded to better explain the statement.

Reviewer #3 (Recommendations For The Authors): 

Specific questions: 

- Is there any way to systematically compare these cells to evolutionarily-diverged cells in distant relatives to sea cucumbers? Or even on a case-by-case basis? For example, is there evidence for any of these transitory cell types to have correlate(s) in vertebrate gut regeneration? 

This is a most interesting question but one that is perhaps a bit premature to answer due to multiple reasons.  First, most of the studies in vertebrates focus on the regeneration of the luminal epithelium, a layer that we are not studying in our system since it appears later in the regeneration process.  Second, there is still too little data from adult echinoderms to fully comprehend which cells are cell orthologues to vertebrates. Third, we are only analyzing one regenerative stage.  It is our hope that this is just the start of a full description of what cell types/stages are found and how they function in regeneration and that this will lead us to identify the cellular orthologues among animal species.

Major revisions: 

- If lineage tracing is within the scope of this paper, it would provide more definitive evidence to the conclusions made about the precursor populations of the regenerating anlage. 

Response:  This is certainly one of the next steps, however at present, it is not possible due to technical limitations.

Minor revisions: 

- Line 47: "for decades" even longer! Could the authors also cite some other amphibians, such as other salamanders (newts) and larval frogs? 

References have been added.

- Line 85: "specially"-could authors potentially change to "specifically" 

Corrected

- Line 122: Authors should add the full words of what these abbreviations stand for in the caption for Figure 1 or in Figure 1A itself. 

Corrected

- Lines 153: What conclusions are the authors trying to make from one type of tubulin presence compared to the others? It's unclear from the text. 

The authors are not trying to reach any particular conclusion.  They are just stating what was found using several markers, and the possibility that what might be viewed first hand as a single cell population might be more heterogenous.  Although the tubulin-type information might not be relevant for the conclusions in the present manuscript, it might be important for future work on the cell types involved in the regeneration process.

- Line 226: Could the authors clarify if "WNT9" is "WNT9a". Figure 3 lists WNT9a but authors refer to WNT9 in the text. 

The gene names in Fig 3 are based on the human identifiers. H. glaberrima only has one sequence of Wnt9 (Auger et al. 2023) and this sequence shares the highest similarity to human Wnt9a, thus the name in the list. We have now identified the gene as Wnt9 to avoid confusion.

- Lines 236-237: Can authors rule out that some immune cells might infiltrate the mesenchymal population? 

No, this cannot be ruled out.  In fact, we believe that most of the immune cells found in our scRNA-seq are indeed cells that have infiltrated the anlage and are part of the mesenchyma.  This has been reported by us previously (see Garcia-Arraras et al. 2006). We have now included this in the text.

- Line 452-453: The over-representation of ribosomal genes not shown. Would it be possible to show this information in the supplementary figures? 

The sentence has been modified, the data is being prepared as part of a separate publication that focuses on the ribosomal genes.

- Line 480: Could authors clarify if it's WNT9a or just WNT9?

It is indeed Wnt9. See previous response above.

- Line 500: In future experiments, it would be interesting to compare to populations at different timepoints in order see how the populations are changing or if certain precursors are activated at different times. 

We fully agree with the reviewer. These are ongoing experiments or are part of new grant proposals.

- Line 567-568: Choosing 9-dpe allowed for 13 clusters, but do authors expect a different number of clusters at different timepoints as things become more terminally differentiated? 

Definitely, we believe that clusters related to the different regenerative stages of cells can be found by looking at earlier or later regeneration stages of the organ.  A clear example is that if the experiment is done at 14-dpe, when the lumen is forming, cells related to luminal epithelium populations will appear. It is also possible that different immune cells will be associated with the different regeneration stages.

- Line 653: References Figure 10D (not in this manuscript). Are authors referring to only 1D or 9D or an old draft figure number? 

As the reviewer correctly points out, this was a mistake where the reference is to a previous draft. It has now been corrected.

- Line 701: "our study reveals that the coelomic epithelium, as a tissue layer, is pluripotent." Phrasing may be better as referring to the cell population making up the tissue layer as pluripotent/multipotent or that the cells it contains would likely be pluripotent or multipotent. Additionally, lineage tracing may be needed to definitively demonstrate this. 

This has been modified.

- Line 808: The authors may make a more accurate conclusion by saying that the characteristics are similar to blastemas or behave like a blastema rather than it is blastema. There is ambiguity about the meaning of this term in the field, but most researchers seem to currently have in mind that the "blastema" definition includes a discrete spatial organization of cells, and here these cells are much more spread out. This could be a good opportunity for the authors to engage in this dialogue, perhaps parsing out the nuances of what a "blastema" is, what the term has traditionally referred to, and how we might consider updating this term or at least re-framing the terminology to be inclusive of functions that "blastemas" have traditionally had in the literature and how they may be dispersed over geographical space in an organism more so than the more rigid, geographically-restricted definition many researchers have in mind. However, if the authors choose to elaborate on these issues, those elaborations do belong in the discussion, and the more provisional terminology we mention here could be used throughout the paper until that element of the revised discussion is presented. We would welcome the authors to do this as a way to point the field in this direction as this is also how we view the matter. For example, some of the genes whose expression has been observed to be enriched following removal of brain tissue in axolotls (such as kazald2, Lust et al.), are also upregulated in traditional blastemas, for instance, in the limb, but we appreciate that the expression domain may not be as localized as in a limb blastema. Additionally, since there is now evidence that some aspects of progenitor cell activation even in limb regeneration extend far beyond the local site of amputation injury (Johnson et al., Payzin-Dogru et al.), there is an opportunity to connect the dots and make the claim that there could be more dispersion of "blastema function" than previously appreciated in the field. Diving a bit more into these nuances may also enable better conceptual framework of how blastema function may evolve across vast evolutionary time and between different injury contexts in super-regenerative organisms. 

We have followed the reviewer’s suggestion and stated that the holothurian anlage behaves as a blastema. Though we would love to elaborate on the blastema topic, as suggested by the reviewer, we believe that it would extend the discussion too much and that the topic might be better served in a different publication.

- In the discussion, it would be important not to leave the reader with the impression that all amphibian blastema cells originate via dedifferentiation. This is not the case. For example, in axolotls (Sandoval-Guzman et al.) and in larval/juvenile newts, muscle progenitors within the blastema structure have been shown to originate from muscle satellite cells, a kind of stem cell, in stump tissues (while adult newts use dedifferentiation of myofibers to generate muscle progenitors in the blastema). Most cell lineages simply have not been evaluated in the level of detail that would be required to definitively conclude one way or the other, and the door is open for a more substantial contribution from stem cell populations than previously appreciated especially because new tools exist to detect and study them. Providing the reader with a more nuanced view of this situation will not negatively impact the findings in this paper, but it will show that there is biological complexity still waiting to be discovered and that we don't have all the answers at this point. 

This has now been corrected. 

Figures: Overall, the figures need minor work. 

- Figure 1A: Can the authors draw a smaller, full-body cartoon and feature the current high-mag cartoon as an inset to that? Can they label the axes and make it clear how the geometry works here?

Fig 1 has been re-done and now is split into Fig 1 and Fig 2.

- Figure 1B: Can the authors label the UMAP with cluster identities on the map itself? This will make it easier to identify each cluster (especially to make sure cluster 11 is easier to find). 

This has been corrected.

- Figure 2: Could the authors put boxes/clearly distinguish panel labels around each cluster (AO), so that there are clear boundaries? 

Fig 2 has been moved to Supplement, following another reviewer recommendation.

- "Gene identifiers starting with "g" correspond to uncharacterized gene models of H. glaberrima." - The sentence is from another figure caption but this figure would benefit from having this sentence in the figure caption as well. 

This has been added to other figures as suggested.

- Figure 3A: Can the authors potentially bold, highlight, or underline genes you discuss in text, so it's easier for the reader to reference? 

This has been added as suggested.

- Figure 3C: Can the authors please label the cell types directly on the UMAP here as well? 

The changes were made following the reviewer’s recommendation.

- Figure 4D-E: There's not much context here to determine if this HCR-FISH validation can tell us anything about these cells besides some of them appear to be there. Do authors expect the coelomocyte morphology to look different in regenerating/injured tissue versus normal animals? Can the authors provide some double in situs, as well as some lower-magnification views showing where the higher-magnification insets are located? Is there any spatial pattern to where these cells are found? Counter stains would be helpful. 

- Figure 6C: If clusters C5, C8, C9 are part of the coelomic epithelium, then authors could show a smaller diagram above with blue and grey to show types and then show clusters separately to help get their point across better. 

- Figure 6G: This image appears to have high background- would it be possible for authors to repeat phalloidin stain or reimage with a lower exposure/gain. Additionally, imaging with Zstacks would help to obtain maximum intensity projections. It would greatly aid the reader if each image was labeled with HCR probes/antibodies that have been applied to the sample. 

- Figure 7E: The cells appear to be out of focus and have high background. Additionally, they are lacking the speckled appearance expected to be seen with HCR-FISH. Would it be possible for authors to collect another image utilizing z-stacks? 

HCR-FISH figures identifying the gene expression characteristic of cell clusters have been modified following the reviewer’s concerns.  The changes include:

(1) Additional clusters have been verified with probes to gene identifiers. These include clusters 8, 9 and 12.

(2) Redundant information has been removed.

(3) Colors have been changed to make figures friendlier to color-impaired readers.

(4) Spatial context has been added or identified.

(5) In some cases, improved photos have been added

(6) Better labels have been included

(7) When necessary individual photos used for the overlay have been included.

- Figure 9A: Could authors add cluster labels onto UMAP directly? 

This change was made to Fig 2A. UMAP in Fig 9A is the same and used just as reference of the subset.

- Figure 10: It could be useful if authors put a small map of the sea cucumber like in other images so that readers know where in the anlage this zoomed in model represents. 

Added as suggested by the reviewer.

- Supplementary figure 1F: Could authors add an arrow to the dark cell that's being pointed out? 

Changed made as suggested by the reviewer.

- Supplementary figure 1: Could authors label clearly what color is labeled with what marker? 

Changed made as suggested by the reviewer.

eLife Assessment

The authors present convincing findings on trends in hind limb morphology through the evolution of titanosaurian sauropod dinosaurs, the land animals that reached the most remarkable gigantic sizes. The important results include the use of 3D geometric morphometrics to examine the femur, tibia, and fibula to provide new information on the evolution of this clade and on evolutionary trends between morphology and allometry.

Joint Public Review:

Páramo et al. used 3D geometric morphometric analyses of the articulated femur, tibia, and fibula of 17 macronarian taxa (known to preserve these three skeletal elements) to investigate morphological changes that occurred in the hind limb through the evolutionary history of this sauropod clade. A principal components analysis was completed to understand the distribution of the morphological variation. A supertree was constructed to place evolutionary trends in morphological variation into phylogenetic context, and hind limb centroid size was used to investigate potential relationships between skeletal anatomy and gigantism. The majority of the results did not yield statistically significant differences, but they did identify interesting shape-change trends, especially within subclades of Titanosauria. Many previous studies have attempted to elucidate a link between wide-gauge posture and gigantism, which in this study Páramo et al. investigate among several titanosaurian subclades. They propose that morphologies associated with wide-gauge posture arose in parallel with increasing body size among basal members of Macronaria and that this connection became less significant once wide-gauge posture was acquired within Titanosauria. The authors also suggest that other biomechanical factors influenced the independent evolution of subclades within Titanosauria and that these influences resulted in instances of convergent evolution. Therefore, they infer that, overall, wide-gauge posture was not significantly correlated with gigantism, though some morphological aspects of hind limb skeletal anatomy appear to have been associated with gigantism. Their work also supports previous findings of a decrease in body size within Titanosauriformes (which they found to be not significant with shape variables but significant with Pagel's lambda). Collectively, their results support and build on previous work to elucidate more specifics on the evolution of this enigmatic clade. Further study will show if their hypotheses stand or if the inclusion of additional specimens and taxa yields alternative results.

[Editors' note: One of the original reviewers, Reviewer 2, reviewed this revised version of the manuscript; they reported satisfaction with the changes made by the authors in response to the original reviewer comments.]

Author response:

The following is the authors’ response to the original reviews.

eLife Assessment

The authors present valuable findings on trends in hind limb morphology throughout the evolution of titanosaurian sauropod dinosaurs, the land animals that reached the most remarkable gigantic sizes. The solid results include the use of 3D geometric morphometrics to examine the femur, tibia, and fibula to provide new information on the evolution of this clade and understand the evolutionary trends between morphology and allometry. Further justification of the ontogenetic stages of the sampled individuals would help strengthen the manuscript's conclusions, and the inclusion of additional large-body mass taxa could provide expanded insights into the proposed trends.

Most of the analyzed specimens, especially from the smaller taxa, come from adult or subadult specimens. None exhibit features that may indicate juvenile status. However, we lack information of the paleohistology that may be a stronger indicator on the ontogenetic status of the individual, and some of operative taxonomic units used in the study come from mean shape of all the sampled specimens.

Current information on morphological differences between adult and subadult or juvenile specimens indicates that even early juvenile specimens may share same morphological features and overall morphology as the adult (e.g., see Curry-Rogers et al., 2016; Appendix S3). We included a comprehensive analysis of the impact of juvenile specimens as one of the aspects of the intraspecific variability that may alter our results in Appendix S3.

Public Reviews:

Reviewer #1:

Weaknesses:

Several sentences throughout the manuscript could benefit from citations. For example, the discussion of using hind limb centroid size as a proxy for body mass has no citations attributed. This should be cited or described as a new method for estimating body mass with data from extant taxa presented in support of this relationship. This particular instance is a very important point to include supporting documentation because the authors' conclusions about evolutionary trends in body size are predicated on this relationship.

We address this issue in the text (Line 32 & 64). Centroid size seems a good indication as it’s the overall size of the entire hind limb, and the length of the femur and tibia is well correlated independently with the body size/mass. Also, as we use few landmarks and only those that are purely type I or II landmarks, with curves of semilandmarks bounded or limited by them, centroid size is not sensible to landmark number differences across the sample in our study (as the centroid size is dependent of the number of landmarks of the current study as well as the physical dimensions of the specimens).

We have sampled and repeated all the analyses using other proxies like the femoral length and the body mass estimated from the Campione & Evans (2020) and Mazzeta et al. (2004) methods. The comprehensive description of the method is in Appendix S2, the alternative analyses can be accessed in the Appendix S3 and S4; and the code for the alternative analyses can be accessed in the modified Appendix S5. All offer similar results than the ones obtained in our analyses with the body size proxied with the hind limb landmark configuration centroid size.

An additional area of concern is the lack of any discussion of taphonomic deformation in Section 3.3 Caveats of This Study, the results, or the methods. The authors provide a long and detailed discussion of taphonomic loss and how this study does a good job of addressing it; however, taphonomic deformation to specimens and its potential effects on the ensuing results were not addressed at all. Hedrick and Dodson (2013) highlight that, with fossils, a PCA typically includes the effects of taphonomic deformation in addition to differences in morphology, which results in morphometric graphs representing taphomorphospaces. For example, in this study, the extreme negative positioning of Dreadnoughtus on PC 2 (which the authors highlight as "remarkable") is almost certainly the result of taphonomic deformation to the distal end of the holotype femur, as noted by Ullmann and Lacovara (2016).

We included a brief commentary in the Caveats of This Study (Line 467) and greatly expanded this issue in the Appendix S3. We followed the methodology proposed by Lefebvre et al. (2020) to discuss the effects of taphonomic deformation in the shape analyses.

Our shape variables (PCs obtained from the shape PCA) should be viewed as taphomorphospaces as Hedrick and Dodson, as well as the reviewer, points in such cases.

The analysis of the effects of taphonomy or errors induced by the landmark estimation method indicate that Dreadnoughtus schrani is one of the few sampled taxa that may have a noticeable impact on our analyses due lithostatic deformation. Other taxa like Mendozasaurus neguyelap or Ampelosaurus atacis may also induce some alterations to the PCs. In general, the trends of those PCs slightly altered by taphonomy, where D. scharni is the only sauropod that may alter an entire PC like PC2, did not exhibit phylogenetic signal and are a small proportion of the sample variance.

The authors investigated 17 taxa and divided them into 9 clades, with only Titanosauria and Lithostrotia including more than two taxa (and four clades are only represented by one taxon). While some of these clades represent the average of multiple individuals, the small number of plotted taxa can only weakly support trends within Titanosauria. If similar general trends could be found when the taxa are parsed into fewer, more inclusive clades, it would support and strengthen their claims. Of course, the authors can only study what is preserved in the fossil record, and titanosaurian remains are often highly fragmentary; these deficiencies should therefore not be held against the authors. They clearly put effort and thought into their choices of taxa to include in this study, but there are limitations arising from this low sample size that inherently limit the confidence that can be placed on their conclusions, and this caveat should be more clearly discussed. Specifically, the authors note that their dataset contains many lithostrotians, but they do not discuss unevenness in body size sampling. As neither their size-category boundaries nor the taxa which fall into each of them are clearly stated, the reader must parse the discussion to glean which taxa are in each size category. It should be noted that the authors include both Jainosaurus and Dreadnoughtus as 'large' taxa even though the latter is estimated to have been roughly five times the body mass of the former, making Dreadnoughtus the only taxon included in this extreme size category. The effects that this may have on body size trends are not discussed. Additionally, few taxa between the body masses of Jainosaurus and Dreadnoughtus have been included even though the hind limbs of several such macronarians have been digitized in prior studies (such as Diamantinasaurus and Giraffititan; Klinkhamer et al. 2018). Also, several members of Colossosauria are more similar in general body size to Dreadnoughtus than Jainosaurus, but unfortunately, they do not preserve a known femur, tibia, and fibula, so the authors could not include them in this study. Exclusion of these taxa may bias inferences about body size evolution, and this is a sampling caveat that could have been discussed more clearly. Future studies including these and other taxa will be important for further evaluating the hypotheses about macronarian evolution advanced by Páramo et al. in this study.

Sadly, we could not include some larger sized titanosaurians sauropods. As the reviewers points out, the lack of larger sauropods among the sampled taxa may hinder our results, as the “large-bodied” category is filled with some mid-sized taxa and the former Dreadnoughtus schrani which is five times larger than some of them. We tried to include Elaltitan lilloi, digitized for this study and included in preliminary analyses, but the fragmentary status increased greatly the error by the estimation method as there is only a proximal third or mid femur preserved from this taxon. Therefore we opted to exclude it from our database.

Other taxa considered, as the reviewer suggest, was not readily available for the authors as the time of this study was conducted and including now may have increased the possible bias of our study. Giraffatitan brancai is an Late Jurassic brachiosaurid, which may again increase the number of early-branching titanosauriforms with large body masses while most of the smaller taxa sampled are recovered in deeply-branching macronarians (including Diamantinasaurus matildae if we would have also included it). Future analyses may include a wider sample of the mid to large-bodied titanosaurians, especially lithostrotians, as well as some colossosaurs like Patagotitan mayorum.

Reviewer #1 (Recommendations For The Authors):

These are all minor comments that would improve the manuscript.

- There are a few typos throughout the manuscript such as: line 70 should be 2016 and line 242 should be forelimb.

Corrected.

- To me, the most interesting aspect of your study is the diversity and trends recovered in titanosaurian subclades and I would highlight this, not gigantism, in the title if you choose to revise the title.

It has been addressed. The specificality of some of the tests and the implication to the acquisition of the spread limb posture and gigantism in early-branching taxa is important nonetheless, so we think that it may remain in the title.

- The abstract should provide more details on the results such as none of the listed trends were statistically significant.

Many of the trends exhibit phylogenetic signal, but not the allometric components. We have briefly addressed them.

- Several sentences in the manuscript need citations such as: line 48 the reference to other megaherbivores, line 66 the discussion of poor understanding of the relationship of wide gauge posture and gigantism, and the use of centroid size as an estimate of body mass (see Public Review).

We changed the line 66 to improve the focus on the current state of the art in the hypothesis of a relationship between arched limbs and in the increase of body size. We included a section relating centroid size as a proxy (due the good correlation between the femur and tibia length and the body mass) and the caveats of using it. We also expanded in the Appendix S2 the use of centroid size and the alternative models.

- With titanosaur evolution, you mention that they are adapting to new niches and topography (line 64). What support is there for this versus they are adapting to be more successful in their current environment?

Noted, we have changed the phrase to improved efficiency exploiting of inland environments, as thy can be either opening new inland niches or adapting better to current inland niches that were already exploited for less deeply branching sauropods. However, its testing is beyond the scope of the current work.

- Line 384-385: the discussion of Rapetosaurus should mention that it is a juvenile and some studies have suggested that titanosaur limbs grow allometrically.

We have included a small line. Whether Rapetosaurus krausei exhibit allometric growth or not may not change greatly the discussion, maybe only excluding it as morphologically convergent to Lirainosaurus and Muyelensaurus. But if that so, it will be further proof that small-sized titanosaurs exhibit the robust skeleton expected in the giant titanosaurs.

- I would consider addressing the question of if we are certain enough in our understanding of titanosaurian phylogeny to rule out homology, especially when you discuss the uncertainty of the placement of specific taxa. Also, Diamantinasaurus is not the only titanosaur that has been proposed as a member of both basal and more derived subclades (e.g., Dreadnoughtus).

We tried to assume a more conservative approach. We could not fully rule out that some of the features observed in the sampled deeply branching lithostrotians, especially saltasauroids, cannot be present in the entire somphospondylan lineage. However, none of the less deeply-branching or early-branching titanosaurs exhibit this kind of morphology. Recent studies propose the possibility that entire groups, included in this study like the Colossosauria, change its position in the phylogeny. However, despite the debated phylogenetic position of Diamantinasaurus or Dreadnoughtus, or even the inclusion of Colossosauria within the saltasauroids and the inclusion of the Ibero-Armorican lithostrotians as putative saltasaurids (Mocho et al. 2024). However, even considering these changes we did not notice any relevant differences in our conclusions about hind limb arched morphology nor about size. Distal hind limb overall robustness should indeed be addressed in the light of shifts in phylogenetic position and include some interesting sauropods like Diamantinasaurus or expand the large-sized Colossosauria or early-branching somphospondyls as it may have profound implications on the morphofunctional adaptations to specific feeding niches, e.g., see current hypotheses about rearing as mentioned in Bates et al. (2016), Ullmann et al. (2017) or Vidal et al. (2020). We had not enough information to conclude the presence of any plesiomorphic condition or analogous feature with our current sample and the debated titanosaurian phylogeny.

- I understand this is not standard in the field, but your study provides the opportunity to conduct sensitivity testing of the effects of cartilage thickness and user articulation of the bones on PCA results. This would be an inciteful addition to the field of GMM.

We are currently developing such a comprehensive analysis and several other implications on our past results. However, we feel that it is beyond the scope of the current study. We appreciate the suggestion nonetheless, as it would be a sensitivity test of the impact of several of our assumptions in the final results that is often not considered.

- In Figure 1, if all the limbs were arranged the same way it would be easier to interpret. Consider flipping panels B and D to match A and C.

Accepted.

- In Figures 2-4, the views in C should be labeled in the figure or caption. Oceanotitan is also in the PCA plot but not included in the figure caption. Also, consider changing the names to represent the paraphyletic groupings you are using instead of formal clade names. For example, change 'Titanosauria' to 'Basal Titanosaurs' to reflect that it is not including all titanosaurs in the sample.

Changes accepted for the shape PCA results. The informal (i.e., paraphyletic) terms such as “Basal Titanosaurs” were only used in the shape analyses as in the RMA, the Titanosauria (and other more inclusive groups) were used as natural groups. Each partial RMA model is based on a sample of all the taxa that are included within that particular clade (e.g., Titanosauria includes both Dreadnoughtus and Saltasaurus; Lithostrotia excludes the former).

- I am concerned that centroid size does not scale evenly across the wide-ranging body mass of titanosaurs. I do not know if this affects your size trends or their significance, but as I mentioned above Dreadnoughtus is much bigger than most of the taxa included and that isn't as drastically apparent in centroid size (in Figure 5) as it is when taxa are plotted by body mass.

Main problematic with centroid size of the hind limb is the shift in the body plan of deeply-branching titanosaurs as the Center of Masses is displaced toward the anterior portion of the body and it has been proposed due a large development of the forelimb region (e.g., Bates et al. 2016). However, it would only increase the effects of the phyletic body size reduction, as smaller taxa tend to have a 1:1 fore limb and hind limb ratio, e.g., from our past analyses as in Páramo et al. (2019), and the sacrum is not as beveled as in earlier somphospondyls, e.g., Vidal et al. (2020). The role of the low-browsing feeding habits of deeply-branching lithostrotians shall be explored elsewhere, as it may be the main driving force of this effect. Our point is, the proxy used may have some slight offset due some high-browsing giant early-branching titanosaurs which has a greater cranial region development which increase its body size and mass beyond our bare-minimum estimation based on the hind limb region. But, overall, this offset is assumed to be low. We repeated the analyses with the femoral length as proxy of body size and a mass estimation, including the quadratic equation based on both humeral and femoral lengths, and the results remain similar. Another problem that arises with the use of centroid size is the way it shall be calculated, but as we used an even number of landmarks and curve semilandmarks, and all of them bounded to anatomical features, it remains equal at least for our sample (but cannot be extrapolated to other geometric morphometric studies that do not use the same configurations)

We appreciate the reviewer concerns nonetheless, as it was on of our own when designing this study, and we in the future will try to expand the analyses, or advise anyone expanding on this study, using total body size/volume estimations following Bates et al. (2016). Which also includes test of the effects of the different whole-body estimation models.

Cites:

Bates KT, Mannion PD, Falkingham PL, Brusatte SL, Hutchinson JR, Otero A, Sellers WI, Sullivan C, Stevens KA, Allen V. 2016. Temporal and phylogenetic evolution of the sauropod dinosaur body plan. Royal Society Open Science 3:150636. doi:10.1098/rsos.150636

Mocho P, Escaso F, Marcos-Fernández F, Páramo A, Sanz JL, Vidal D, Ortega F. 2024. A Spanish saltasauroid titanosaur reveals Europe as a melting pot of endemic and immigrant sauropods in the Late Cretaceous. Commun Biol 7:1016. doi:10.1038/s42003-024-06653-0

Páramo A, Ortega F, Sanz JL. 2019. A Niche Partitioning Scenario for the Titanosaurs of Lo Hueco (Upper Cretaceous, Spain). International Congress of Vertebrate Morphology (ICVM) - Abstract Volume, Journal of Morphology. Prague. p. S197.

Ullmann PV, Bonnan MF, Lacovara KJ. 2017. Characterizing the Evolution of Wide-Gauge Features in Stylopodial Limb Elements of Titanosauriform Sauropods via Geometric Morphometrics. The Anatomical Record 300:1618–1635. doi:10.1002/ar.23607

Vidal D, Mocho P, Aberasturi A, Sanz JL, Ortega F. 2020. High browsing skeletal adaptations in Spinophorosaurus reveal an evolutionary innovation in sauropod dinosaurs. Sci Rep 10:6638. doi:10.1038/s41598-020-63439-0

Reviewer #2:

The authors report a quantitative comparative study regarding hind limb evolution among titanosaurs. I find the conclusions and findings of the manuscript interesting and relevant. The strength of the paper would be increased if the authors were to improve their reporting of taxon sampling and their discussion of age estimation and the potential implications that uncertainty in these estimates would have for their conclusions regarding gigantism (vs. ontogenetic patterns).

Considering the observations made by reviewer #1, we included a data about the impact of ontogenetic patterns and other intraspecific variability in the Appendix S3. We considered to increase the sample but it has not been possible at the time of this study was carried out.

Reviewer #2 (Recommendations For The Authors):

I have a few concerns/requests for the authors, that I hope can be easily resolved.

Comments:

- What drove taxon sampling?

Random sampling of somphospondylan sauropods focused on the Lithostrotia clade for the thesis project of one of the authors, APB. Logistics were also one of the bias on our sample, and based on the available titanosaurian material we left out several macronarians that has been already sampled but would further induce a early-branching large sauropod, deeply-branching small sauropod that may alter our results.

- Which phylogenies were used to create the supertree applied to the analyses? What references were used to time-calibrate the tips and deeper nodes? I couldn't find any reference to this. Additionally, more information regarding the R packages and analytical pipeline would be appreciated: e.g. were measurements used in the analyses log-transformed?

A comprehensive description of the methodology is provided in Appendix S2.

- Age estimate: can the author confirm the skeletal maturity of the sampled individuals? If this is not the case, how can the author be sure that the patterns towards gigantism are not reflecting different ontogenetic stages? I believe this should be part of both methods and discussion.

As commented before, we excluded small, probable juvenile specimens from our sample. We have no paleohistological sample backing the claims of the ontogenetic status of some of the specimens that were included or excluded were calculating the mean shape for the operative taxonomic units. However, we followed a criteria to identify the relative ontogenetic status and it has been included in Appendix S3.

- The authors used the centroid size for regressions in Figure 6. Although I believe that this is a good variable, would the author be willing to use body mass and log-transformed femur length in addition to what was done? These would be very useful considering that these variables are (relatively) independent from shape/morphology.

Accepted, we tested our hypotheses with three alternative models based on femoral length, combined femoral and humeral lengths for body mass estimations. Methodology can be found in Appendix S2, results on Appendix S4, code for the alternative methods in Appendix S5.

- Data access: will stl. Files of the limb elements be shared and freely available? In this case, where the files will be deposited?

At the time of the current study, some of the sampled specimens cannot be available (material under study) but the mean shapes can be generated after the landmarks and semilandmark curves and the “atlas” mesh.

- Additionally, outstanding references regarding limb evolution, GMM, role of ontogeny, and evolution of columnar gait are missing. The authors should reinforce the literature review with the following (alphabetical order):

Bonnan, M. F. (2003). The evolution of manus shape in sauropod dinosaurs: implications for functional morphology, forelimb orientation, and phylogeny. Journal of Vertebrate Paleontology, 23(3), 595-613.

Botha, J., Choiniere, J. N., & Benson, R. B. (2022). Rapid growth preceded gigantism in sauropodomorph evolution. Current Biology, 32(20), 4501-4507.

Curry Rogers, K., Whitney, M., D'Emic, M., & Bagley, B. (2016). Precocity in a tiny titanosaur from the Cretaceous of Madagascar. Science, 352(6284), 450-453.

Day, J. J., Upchurch, P., Norman, D. B., Gale, A. S., & Powell, H. P. (2002). Sauropod trackways, evolution, and behavior. Science, 296(5573), 1659-1659.

Fabbri, M., Navalón, G., Benson, R. B., Pol, D., O'Connor, J., Bhullar, B. A. S., ... & Ibrahim, N. (2022). Subaqueous foraging among carnivorous dinosaurs. Nature, 603(7903), 852-857.

Fabbri, M., Navalón, G., Mongiardino Koch, N., Hanson, M., Petermann, H., & Bhullar, B. A. (2021). A shift in ontogenetic timing produced the unique sauropod skull. Evolution, 75(4), 819-831.

González Riga, B. J., Lamanna, M. C., Ortiz David, L. D., Calvo, J. O., & Coria, J. P. (2016). A gigantic new dinosaur from Argentina and the evolution of the sauropod hind foot. Scientific Reports, 6(1), 19165.

Lefebvre, R., Allain, R., & Houssaye, A. (2023). What's inside a sauropod limb? First three‐dimensional investigation of the limb long bone microanatomy of a sauropod dinosaur, Nigersaurus taqueti (Neosauropoda, Rebbachisauridae), and implications for the weight‐bearing function. Palaeontology, 66(4), e12670.

McPhee, B. W., Benson, R. B., Botha-Brink, J., Bordy, E. M., & Choiniere, J. N. (2018). A giant dinosaur from the earliest Jurassic of South Africa and the transition to quadrupedality in early sauropodomorphs. Current Biology, 28(19), 3143-3151.

Martin Sander, P., Mateus, O., Laven, T., & Knötschke, N. (2006). Bone histology indicates insular dwarfism in a new Late Jurassic sauropod dinosaur. Nature, 441(7094), 739-741.

Remes, K. (2008). Evolution of the pectoral girdle and forelimb in Sauropodomorpha (Dinosauria, Saurischia): osteology, myology and function (Doctoral dissertation, München, Univ., Diss., 2008).

Sander, P. M., & Clauss, M. (2008). Sauropod gigantism. Science, 322(5899), 200-201.

Yates, A. M., & Kitching, J. W. (2003). The earliest known sauropod dinosaur and the first steps towards sauropod locomotion. Proceedings of the Royal Society of London. Series B: Biological Sciences, 270(1525), 1753-1758.

We appreciate this suggestion and we already used some of the articles in our study but the selection of cites were based also in the available manuscript space enforced by the edition guidelines. We would have like to include several of these works but we had opted to include some of the works that summarize some of them, whereas excluding others.

eLife Assessment

This is a valuable study that tests the functional role of food-washing behavior in removing tooth-damaging sand and grit in long-tailed macaques and whether dominance rank predicts level of investment in the behavior. The evidence that food-washing is deliberate is compelling, but the evidence for variable and adaptive investment depending on rank, including the fitness-relevance and ultimate evolutionary implications of the findings, is incomplete given limitations of the experimental design. Overall, the paper should be of interest to researchers interested in foraging behavior, cognition, and primate evolution.

Reviewer #1 (Public review):

In this paper, the authors had 2 aims:

(1) Measure macaques' aversion to sand and see if its' removal is intentional, as it likely in an unpleasurable sensation that causes tooth damage.

(2) Show that or see if monkeys engage in suboptimal behavior by cleaning foods beyond the point of diminishing returns, and see if this was related to individual traits such as sex and rank, and behavioral technique.

They attempted to achieve these aims through a combination of geochemical analysis of sand, field experiments, and comparing predictions to an analytical model.

The authors' conclusions were that they verified a long-standing assumption that monkeys have an aversion to sand as it contains many potentially damaging fine grained silicates, and that removing it via brushing or washing is intentional.

They also concluded that monkeys will clean food for longer than is necessary, i.e. beyond the point of diminishing returns, and that this is rank-dependent.

High and low-ranking monkeys tended not to wash their food, but instead over-brushed it, potentially to minimize handling time and maximize caloric intake, despite the long-term cumulative costs of sand.

This was interpreted through the *disposable soma hypothesis*, where dominants maximize immediate needs to maintain rank and increase reproductive success at the potential expense of long-term health and survival.

# Strengths

The field experiment seemed well designed, and their quantification of the physical and mineral properties of quartz particles (relative to human detection thresholds) seemed good relative to their feret diameter and particle circularity (to a reviewer that is not an expert in sand). The *Rank Determination* and *Measuring Sand* sections were clear.

In achieving Aim 1, the authors validated a commonly interpreted, but unmeasured function, of macaque and primate behavior-- a key study/finding in primate food processing and cultural transmission research.

I commend their approach in trying to develop a quantitative model to generate predictions to compare to empirical data for their second aim.<br /> This is something others should strive for.

I really appreciated the historical context of this paper in the introduction and found it very enjoyable and easy to read.

I do think that interpreting these results in the context of the *disposable soma hypothesis* and the potential implications in the *paleolithic matters* section about interpreting dental wear in the fossil record are worthwhile.

# Weaknesses

Several of my concerns in an earlier review were addressed in revision, which I appreciate. One thing I think could strengthen this paper is a clearer link to social foraging theory to explore heterogeneity in handling times (as the currency they are trying to maximize).

I am satisfied with the improvements in statistics and that I can access the code and data.

I am still struck that there was an analysis of only trials where <3 individuals are present. If rank was important, I would imagine that behavior might be different in social contexts when theft, scrounging, policing, aggression, or other distractions might occur-- where rank would have effects on foraging behavior. Maybe lower rankers prioritize rapid food intake then. If rank should be related to investment in this behavior, we might expect this to be magnified (or different) in social contexts where it would affect foraging. It might just be that the data was too hard to score or process in those settings, or the analysis was limited. Additionally, I think that more robust metrics of rank from more densely sampled focal follow data would be a better measure, but I acknowledge the limitations in getting the ideal . Since rank is central to the interpretation of these results, I think that reduced social contexts in which rank was analyzed and the robustness of the data from which rank was calculated and analyzed are the main weaknesses of the evidence presented in this paper.

While some of the boxes about raccoons and Concorde Fallacy were interesting, they did feel like a bit of a distraction from the main message in the paper.

Reviewer #3 (Public review):

This revised paper provides evidence that food washing and brushing in wild long-tailed macaques are deliberate behaviors to remove sand that can damage tooth enamel. The demonstration of the immediate functional importance of these behaviors is nicely done, and there is some interesting initial evidence that macaques differ systematically in their investment in food cleaning based on dominance rank.

The authors interpret this evidence as support for "disposable soma" effects: that reduced time and effort invested food washing in high-ranking individuals is attributable to prioritizing reproductive effort. Given that the analysis is on a single group with no longitudinal data, there are no fitness measures or fitness proxies, the energetic constraints faced by this population are not clear, and both sexes are combined into a single dominance hierarchy (trade-offs between different forms of investment are typically thought to differ between sexes), this conclusion is premature, although an interesting foundation for future studies.

More generally, the results directly supported by the data collection and analysis (grit on Koshima likely damages macaque teeth; processing food helps mitigate the damage; there is some interesting interindividual variation in food processing time, and that time is not always in line with what appears to be optimal) tend to be combined with interpretation that is much more speculative (e.g., the effect sizes observed are consequential for fitness; high-ranking animals are making choices that optimize their long-term fitness at the expense of their soma). This is in part a stylistic choice but can have the effect of drawing attention away from the stronger empirical findings and/or be misleading. Similarly, although I appreciate that the authors were trying to interpret and respond to previous feedback from reviewers, I found the addition of the box text on the raccoon nomenclature and on irrational behavior and the Concorde effect distracting (more intro-textbook style than journal article style).

Author response:

The following is the authors’ response to the original reviews.

We thank the reviewers for their constructive criticism. It is rare and gratifying to receive such thoughtful feedback, and the result is a much stronger paper. We made significant changes to our statistical analyses and figures to better differentiate the effects of sex and dominance rank on food-cleaning behaviors. These revisions uphold our original conclusion––that rank-related variation overwhelms any sex difference in cleaning behavior. We hope that these edits, together with the rest of our responses, provide a convincing demonstration of the tradeoffs of eliminating quartz from food surfaces.

Reviewer #1 (Public Review):

Summary

We have no objections to Reviewer 1’s summary of our manuscript.

Strengths

Reviewer 1 is extremely gracious, and we are grateful for the kind words.

Weaknesses

Reviewer 1 identified several weaknesses, enumerating three types: (1) statistics, (2) insufficient links to foraging theory, and (3) interpretation and validity of the model. The present response is organized around these same categories.

(1) Statistics

We put all of our data and code into the Zenodo repository prior to submission. This content should have been accessible to Reviewer 1 from the outset. But in any event, we are very sorry for the mixup. To ensure access to our data and code during the present stage of review, we included the URL in the main mainscript and here: https://doi.org/10.5281/zenodo.14002737

(a) AIC and outcome distributions

Reviewer 1 criticized our use of AIC for determining model selection. We agree and this aspect of our manuscript is now removed. In lieu of AIC, we produced two data sets consisting of whole number counts (seconds) with means <5. The data were right-skewed due to high concentrations of biologically-meaningful zeros (i.e., bouts of food handling without any cleaning effort). Following the recommendations of Bolker et al. (2008) and others (Brooks et al. 2017, 2019), we chose an outcome distribution (zero-inflated Poisson, see response below) that best matched this data distribution. In addition, we evaluated the post-hoc performance of each of our models using the standardized residual diagnostic tools for hierarchical regression models available in the DHARMa package (Hartig, 2022). To further evaluate our choice of outcome distribution, we generated QQ-plots and residual vs. predicted plots for each model and included them in our revision as Figures S3-S5.

(b) zeros

Reviewer 1 expressed concern over our treatment of biologically-meaningful zeros, and recommended use of a zero-inflated GLMM with either a Poisson or negative binomial outcome distribution. We agree that such models are best for our two data sets. Accordingly, we fit a series of zero-inflated generalized linear mixed models (ZIGLMM) using the glmmTMB package in R, each with a logit-link function, a single zero-inflation parameter applying to all observations, and a Poisson error distribution. For the food-brushing model, we fit a zero-inflated Poisson (ZIP), which produced favorable standardized residual diagnostic plots with no major patterns of deviation (Figure S3) and minor, but non-significant underdispersion (DHARMa dispersion statistic = 0.99, p = 0.80). For our two food-washing models, we used zero-inflated models with Conway-Maxwell Poisson (ZICMP) distributions, an error distribution chosen for its ability to handle data that are more underdispersed (DHARMa dispersion statistic = 8.2E-09, p = 0.74) than the standard zero-inflated Poisson (Brooks et al. 2019). Using this error distribution improved residual diagnostic plots over a standard ZIP model and we view any deviations in the standardized residuals as minor and attributable to the smaller sample size of our food-washing data set (see Figures S4 and S5) (Hartig, 2022). We reported the summarized fixed effects tests for each GLMM in Tables S1-S3 as Analysis of Deviance Tables (Type II Wald chi square tests, one-sided) along with 𝜒2 values, degrees of freedom, and p-values (one-sided tests). Full model summaries with standard errors and confidence intervals are also included in Tables S4-S6. For all statistical analyses, we set 𝛼 = 0.05.

(2) Absence of Links to Foraging Theory

This critique has three components. The first revisits the absence of code for the optimal cleaning time model. This omission was an unfortunate error at the moment of submission, but our code is available now as a Mathematica notebook in Zenodo (https://doi.org/10.5281/zenodo.14002737). The second pivots around our scholarship, admonishing us for failing to acknowledge the marginal value theorem of Charnov (1976). It is a fair point and we have corrected the oversight with a citation to this classic paper. The third criticism is also rooted in scholarship, with Reviewer 1 asking for greater connection to the existing literature on optimal foraging theory, a point echoed in the summary assessment of the editors at eLife. This comment and the weight given to it by eLife’s editors put us in a difficult spot, as our paper is focused on the optimization of delayed gratification, not food acquisition per se. So, we are in the awkward position of gently resisting this recommendation while simultaneously agreeing with Reviewer 1 that we need to better situate our findings in the landscape of existing literature. To thread this needle, we produced Box 2 with a photograph and 410 words. This display box puts our findings into direct conversation with recent research focused on the sunk cost fallacy.

(3) Interpretation and validity of model relative to data

This critique is focused on the simulated brushing and washing results reported in Figure S1, along with its captioning, which was inadequate. We edited the caption to identify the author (JER) who simulated the brushing and washing behaviors of the monkeys. In addition, we clarified the number of brushing replicates (3) and washing replicates (3) for each of three treatments, for a total of 18 simulations.

We followed Reviewer 1’s suggestion, incorporating the experimental uncertainty of grit removal into our optimal cleaning time model. We drew % grit removed values the % grit removed is used to estimate the cleaning inefficiency≥ 100%parameter 𝑐 for from a distribution, discounting the rare event when values were drawn. As brushing and washing, the included uncertainty now allows us to evaluate these parameters as distributions; and, in turn, obtain a distribution for our predicted brushing and washing optimal cleaning times. As we now describe in the main text, the optimal cleaning time for brushing and washing are 𝑡* \= 0. 98 ± 0. 19 s and * = 2. 40 ± 0. 74 s, respectively. We are grateful for Reviewer 1’s suggestion, for it added𝑡 valuable context to our model predictions. Notably, the inclusion of experimental uncertainty did not change the qualitative nature of our results, or the interpretations of our model predictions compared to observed cleaning behaviors.

We choose to exclude variability in handling time h to generate predicted cleaning time optima, at least in the main text. Our reasoning stems from the observation that handling time variability is long-tailed, with the longer handling times associated with behaviors that we do not account for in our analysis. For example, individuals carrying multiple cucumber slices to the ocean were apt to drop them, struggling at times to re-grasp so many at once. Such moments increased handling times substantially. Still, we acted on Reviewer 1’s suggestion, accounting for the tandem effects of handling time variability and uncertainty in % grit removed (see Figure S6). Drawing handling time estimates from a log-normal distribution fitted to the handling time data, we found that these dual sources of uncertainty did not qualitatively change our results. They added further uncertainty to the predicted washing time, but the mean remains roughly equivalent. (We note that brushing is assumed to have a constant handling time––composed of only assessment time and no travel––such that the results for brushing do not change.) Both analyses are included in the Mathematica notebook at (https://doi.org/10.5281/zenodo.14002737).

Reviewer #2 (Public Review):

Summary

We have no objections to Reviewer 2’s summary of our manuscript.

Strengths

Reviewer 2 is extremely gracious, and we are grateful for the kind words.

Weaknesses

Reviewer 2 noted that our manuscript failed to provide “sufficient background on [our study] population of animals and their prior demonstrations of food-cleaning behavior or other object-handling behaviors (e.g., stone handling).” To address this comment, we edited the introduction (lines 56-58) to alert readers to the onset of regular food-cleaning behaviors sometime after December 26, 2004. In addition, we edited our methods text (lines 155-160) to highlight the onset and limited scope of prior research with this study population:

“The animals are well habituated to human observers due to regular tourism and sustained study since 2013 (Tan et al., 2018). Most of this research has revolved around stone tool-mediated foraging on mollusks, the only activity known to elicit stone handling (Malaivijitnond et al., 2007; Gumert and Malaivijitnond, 2012, 2013; Tan et al., 2015), although infants and juveniles will sometimes use stones during object play (Tan, 2017). There has been no prior examination of food-cleaning behaviors.”

Reviewer #3 (Public Review):

Reviewer 3 identified three weaknesses, which we address in three paragraphs.

Reviewer 3 questioned our methods for determining rank-dependent differences in cleaning behavior, arguing that our conclusions were unsupported. It is a fair point, and it compelled us to combine males and females into a single standardized ordinal rank of 24 individuals. This unified ranking is now reflected in the x-axes of Figure 2 and Figure S2. Plotting the data this way––see Figure S2––underscores Reviewer 3’s concern that sex and dominance rank are confounding variables. To address this problem, our GLMM included rank and sex as predictor variables, which controls for the effect of sex when assessing the relationship between rank and cleaning time across the three treatments. Reported in Tables S1-S3, these findings show that the effect of sex on either brushing or washing time was not significant. This result bolsters our original contention that rank-related variation in cleaning time overwhelms any sex differences.

Relatedly, Reviewer 3 questioned our conclusions on the effects of rank because our study was focused on a single social group. In other words, it is plausible that our results were heavily influenced by the idiosyncrasies of select individuals, not dominance rank per se. It is a fair point, and it compelled us to include individual ID as a random effect in each of our GLMMs. Including individual ID as a random intercept allowed us to control for inter-individual variation in cleaning duration while assessing the effects of rank. An analysis based on additional social groups or longitudinal data are certainly desirable, but also well beyond the scope of a Short Report for eLife.

Finally, Reviewer 3 objected to fragments of sentences in our abstract, introduction, and discussion, combining them into a criticism of claims that we did not and do not make. It probably wasn’t intentional, but it puts us in the awkward position of deconstructing a strawman:

● Review 3 begins, “there is no evidence presented on the actual fitness-related costs of tooth wear or the benefits of slightly faster food consumption”. This statement is true while insinuating that collecting such evidence was our intent. To be clear, our experiment was never designed to measure tooth wear or reproductive fitness, nor do we make any claims of having done so.

● Reviewer 3 adds, “Support for these arguments is provided based on other papers, some of which come from highly resource-limited populations (and different species). But this is a population that is supplemented by tourists with melons, cucumbers, and pineapples!” We were puzzled over these sentences. The first fails to mention that the citations exist in our discussion. Citing relevant work in a discussion is a basic convention of scientific writing. But it seems the underlying intent of these words is to denigrate the value of our study population because two dozen tourists visit Koram Island once a day. Exclamations to the contrary, the amount of tourist-provisioned food in the diet of any one monkey is negligible.

● Last, Reviewer 3 commented on matters of style, objecting to “overly strong claims.” We puzzled over this criticism because the claims in question are broader points of introduction or discussion, not results. The root problem appears to be the final sentence of our abstract:

“Dominant monkeys abstained from washing, balancing the long-term benefits of mitigating tooth wear against immediate energetic requirements, an essential predictor of reproductive fitness.”

This sentence has three clauses. The first is a statement of results, whereas the second and third are meant to mirror our discussion on the importance of our findings. We combined the concepts into a single concluding sentence for the sake of concision, but we can appreciate how a reader could feel deceived, expecting to see data on tooth wear and fitness. So, our impression is that we are dealing with a simple misunderstanding of our own making, and that this single sentence explains Reviewer 3’s criticism and tone––it cast a long shadow over the substance of our paper. To resolve this problem, we edited the sentence:

“Dominant monkeys abstained from washing, a choice consistent with the impulses of dominant monkeys elsewhere: to prioritize rapid food intake and greater reproductive fitness over the long-term benefits of prolonging tooth function.”

eLife Assessment

This important study characterizes the molecular signatures and function of a type of enteric neuron (IPAN) in the mouse colon, identifying molecular markers (Cdh6 and Cdh8) for these cells. A battery of compelling and comprehensive experimental findings suggests data from other species are likely translatable to mice, bridging the abundant literature from humans and other mammals into this experimentally tractable animal model. This work will be of interest to scientists studying the motor control of the colon and more generally the enteric neuromuscular system.

Reviewer #1 (Public review):

Summary:

In their manuscript, Gomez-Frittelli and colleagues characterize the expression of cadherin6 (and -8) in colonic IPANs of mice. Moreover, they found that these cdh6-expressing IPANs are capable of initiating colonic motor complexes in the distal colon, but not proximal and midcolon. They support their claim by morphological, electrophysiological and optogenetic, and pharmacological experiments.

Strengths:

The work is very impressive and involves several genetic models and state-of-the-art physiological setups including respective controls. It is a very well-written manuscript that truly contributes to our understanding of GI-motility and its anatomical and physiological basis. The authors were able to convincingly answer their research questions with a wide range of methods without overselling their results.

Weaknesses:

The authors put quite some emphasis on stating that cdh6 is a synaptic protein (in the title and throughout the text), which interacts in a homophilic fashion. They deduct that cdh6 might be involved in IPAN-IPAN synapses (line 247ff.). However, Cdh6 does not only interact in synapses and is expressed by non-neuronal cells as well (see e.g., expression in the proximal tubuli of the kidney). Moreover, cdh6 does not only build homodimers, but also heterodimers with Chd9 as well as Cdh7, -10, and -14 (see e.g., Shimoyama et al. 2000, DOI: 10.1042/0264-6021:3490159). It would therefore be interesting to assess the expression pattern of cdh6-proteins using immunostainings in combination with synaptic markers to substantiate the authors' claim or at least add the possibility of cell-cell-interactions other than synapses to the discussion. Additionally, an immunostaining of cdh6 would confirm if the expression of tdTomato in smooth muscle cells of the cdh6-creERT model is valid or a leaky expression (false positive).

Comments on revisions:

The authors have updated their manuscript and have provided insights and discussions to my remarks.

Reviewer #2 (Public review):

Summary:

Intrinsic primary afferent neurons are an interesting population of enteric neurons that transduce stimuli from the mucosa, initiate reflexive neurocircuitry involved in motor and secretory functions, and modulate gut immune responses. The morphology, neurochemical coding, and electrophysiological properties of these cells have been relatively well described in a long literature dating back to the late 1800's but questions remain regarding their roles in enteric neurocircuitry, potential subsets with unique functions, and contributions to disease. Here, the authors provide RNAscope, immunolabeling, electrophysiological, and organ function data characterizing IPANs in mice and suggest that Cdh6 is an additional marker of these cells.

Strengths:

This paper would likely be of interest to the enteric neuroscience community and increases information regarding the properties of IPANs in mice. These data are useful and suggest that prior data from studies of IPANs in other species are likely translatable to mice.

Weaknesses:

Major weaknesses:<br /> (1) The novelty of this study is relatively limited. The main point of novelty suggests an additional marker of IPANs (Cdh6) that would add to the known list of markers for these cells. How useful this would be is unclear. Other main findings basically confirm that IPANs in mice display the same classical characteristics that have been known for many years from studies in guinea pigs, rats, mice and humans.

(2) Critical controls are needed to support the optogenetic experiments. Control experiments are needed to show that ChR2 expression 1) does not change the baseline properties of the neurons, 2) that stimulation with the chosen intensity of light elicits physiologically relevant responses in those neurons, and 3) that stimulation via ChR2 elicits comparable responses in IPANs in the different gut regions focused on here. These essential controls remain absent in the study and limit confidence in the data derived from this model.

(3) The motor effects observed in optogenetic experiments are difficult to understand in the absence of good controls for optogenetic control of the proposed neuron population (discussed above). It remains unclear how stimulating IPANs in the distal colon would generate retrograde CMCs while stimulating IPANs in the proximal colon did nothing. Key controls confirming that the optogentic stimulus was adequate, specific, and relevant are needed. In addition, better characterization of the Cdh6+ population of cells in both regions would be needed to understand the mechanisms underlying these effects.

(4) From the data shown, it is clear that expression driven by the Cdh6CreERT2 driver is not confined to IPANs. There is obviously expression of GFP and ChR2 in smooth muscle cells. This is a major limitation for the physiological experiments that attempt to use this model to specifically stimulate IPANs and assess changes in gut motor function. Better characterization of this model is needed and control experiments are necessary to assess whether functional ChR2 is expressed in cells beyond the proposed subtype of enteric IPANs.

(5) Some of the main conclusions of this study are overstated and claims of priority are made that are not true. For example, the authors state on lines 27-28 of the abstract that their findings provide the "first demonstration of selective activation of a single neurochemical and functional class of enteric neurons". This is certainly not true since Gould et al (AJP-GIL 2019) expressed ChR2 in nitrergic enteric neurons and showed that activating those cells disrupted CMC activity. In fact, prior work by the authors themselves (Hibberd et al Gastro 2018) showed that activating calretinin neurons with ChR2 evoked motor responses. Work by other groups has used chemogenetics and optogenetics to show effects of activating multiple other classes of neurons in the gut.

(6) The electrophysiological characterization of mouse IPANs is useful but is limited to a small subset of Cdh6+ neurons in the distal colon myenteric plexus. Therefore, it remains unclear how well the properties reported here might reflect those of other Cdh6+ IPANs in the same or different regions. Similarly, blocking IH with ZD7288 affects all IPANs and does not add specific information regarding the role of the proposed Cdh6+ subtype.

(7) The submucosal plexus (SMP) also contains enteric IPANs and these were not included in the analysis of Cdh6 expression. Whether or not the proposed IPAN marker Cdh6 would be useful for identifying or targeting those cells remains unclear.

[Editor's note: The Reviewing Editor considers that further controls requested from the reviewers have largely been provided already in prior publications by other groups, as they concern specifically tools published years ago but in a different tissue context. Hence the methodology used to deliver the results reported here fall within the standard practices in the field. The comprehensive, multi-technique approach to the results is compelling in and of itself, and ought to suffice, rendering this work reproducible and therefore a basis for further research.]

Author response:

The following is the authors’ response to the original reviews.

Reviewer #1 (Public review):

Summary:

In their manuscript, Gomez-Frittelli and colleagues characterize the expression of cadherin6 (and -8) in colonic IPANs of mice. Moreover, they found that these cdh6-expressing IPANs are capable of initiating colonic motor complexes in the distal colon, but not proximal and midcolon. They support their claim by morphological, electrophysiological, optogenetic, and pharmacological experiments.

Strengths:

The work is very impressive and involves several genetic models and state-of-the-art physiological setups including respective controls. It is a very well-written manuscript that truly contributes to our understanding of GI-motility and its anatomical and physiological basis. The authors were able to convincingly answer their research questions with a wide range of methods without overselling their results.

We greatly appreciate the reviewer’s time, careful reading and support of our study.

Weaknesses:

The authors put quite some emphasis on stating that cdh6 is a synaptic protein (in the title and throughout the text), which interacts in a homophilic fashion. They deduct that cdh6 might be involved in IPAN-IPAN synapses (line 247ff.). However, Cdh6 does not only interact in synapses and is expressed by non-neuronal cells as well (see e.g., expression in the proximal tubuli of the kidney). Moreover, cdh6 does not only build homodimers, but also heterodimers with Chd9 as well as Cdh7, -10, and -14 (see e.g., Shimoyama et al. 2000, DOI: 10.1042/02646021:3490159). It would therefore be interesting to assess the expression pattern of cdh6proteins using immunostainings in combination with synaptic markers to substantiate the authors' claim or at least add the possibility of cell-cell-interactions other than synapses to the discussion. Additionally, an immunostaining of cdh6 would confirm if the expression of tdTomato in smooth muscle cells of the cdh6-creERT model is valid or a leaky expression (false positive).

We agree with the reviewer that Cdh6 could be mediating some other cell-cell interaction besides synapses between IPANs, and we noted it in the discussion. Cdh6 primarily forms homodimers but, as the reviewer points out, has been known to also form heterodimers with some other cadherins. We performed RNAscope in the colonic myenteric plexus with Cdh7 and found no expression (data not shown). Cdh10 is suggested to have very low expression (Drokhlyansky et al., 2020), possibly in putative secretomotor vasodilator neurons, and Cdh14 has not been assayed in any RNAseq screens. We attempted to visualize Cdh6 protein via antibody staining (Duan et al., 2018) but our efforts did not result in sufficient signal or resolution to identify synapses in the ENS, which remain broadly challenging to assay. Similarly, immunostaining with Cdh6 antibody was unable to confirm Cdh6 protein in tdT-expressing muscle cells, or by RNAscope. We have addressed these caveats in the discussion section.

(1) E. Drokhlyansky, C. S. Smillie, N. V. Wittenberghe, M. Ericsson, G. K. Griffin, G. Eraslan, D. Dionne, M. S. Cuoco, M. N. Goder-Reiser, T. Sharova, O. Kuksenko, A. J. Aguirre, G. M. Boland, D. Graham, O. Rozenblatt-Rosen, R. J. Xavier, A. Regev, The Human and Mouse Enteric Nervous System at Single-Cell Resolution. Cell 182, 1606-1622.e23 (2020).

(2) X. Duan, A. Krishnaswamy, M. A. Laboulaye, J. Liu, Y.-R. Peng, M. Yamagata, K. Toma, J. R. Sanes, Cadherin Combinations Recruit Dendrites of Distinct Retinal Neurons to a Shared Interneuronal Scaffold. Neuron 99, 1145-1154.e6 (2018).

Reviewer #2 (Public review):

Summary:

Intrinsic primary afferent neurons are an interesting population of enteric neurons that transduce stimuli from the mucosa, initiate reflexive neurocircuitry involved in motor and secretory functions, and modulate gut immune responses. The morphology, neurochemical coding, and electrophysiological properties of these cells have been relatively well described in a long literature dating back to the late 1800's but questions remain regarding their roles in enteric neurocircuitry, potential subsets with unique functions, and contributions to disease. Here, the authors provide RNAscope, immunolabeling, electrophysiological, and organ function data characterizing IPANs in mice and suggest that Cdh6 is an additional marker of these cells.

Strengths:

This paper would likely be of interest to a focused enteric neuroscience audience and increase information regarding the properties of IPANs in mice. These data are useful and suggest that prior data from studies of IPANs in other species are likely translatable to mice.

We appreciate the reviewer’s support of our study and insightful critiques for its improvement.

Weaknesses:

The advance presented here beyond what is already known is minimal. Some of the core conclusions are overstated and there are multiple other major issues that limit enthusiasm. Key control experiments are lacking and data do not specifically address the properties of the proposed Cdh6+ population.

Major weaknesses:

(1) The novelty of this study is relatively low. The main point of novelty suggests an additional marker of IPANs (Cdh6) that would add to the known list of markers for these cells. How useful this would be is unclear. Other main findings basically confirm that IPANs in mice display the same classical characteristics that have been known for many years from studies in guinea pigs, rats, mice and humans.

We appreciate the already existing markers for IPANs in the ENS and the existing literature characterizing these neurons. The primary intent of this study was to use these well-established characteristics of IPANs in both mice and other species to characterize Cdh6-expressing neurons in the mouse myenteric plexus and confirm their classification as IPANs.

(2) Some of the main conclusions of this study are overstated and claims of priority are made that are not true. For example, the authors state in lines 27-28 of the abstract that their findings provide the "first demonstration of selective activation of a single neurochemical and functional class of enteric neurons". This is certainly not true since Gould et al (AJP-GIL 2019) expressed ChR2 in nitrergic enteric neurons and showed that activating those cells disrupted CMC activity. In fact, prior work by the authors themselves (Hibberd et al., Gastro 2018) showed that activating calretinin neurons with ChR2 evoked motor responses. Work by other groups has used chemogenetics and optogenetics to show the effects of activating multiple other classes of neurons in the gut.

We thank the reviewer for bringing up this important point and apologize if our wording was not clear. Whilst single neurochemical classes of enteric neurons have been manipulated to alter gut functions, all such instances to date do not represent manipulation of a single functional class of enteric neurons. In the given examples, multiple functional classes are activated utilizing the same neurotransmitter, as NOS and calretinin are each expressed to varying degrees across putative motor neurons, interneurons and IPANs. In contrast, Chd6 is restricted to IPANs and therefore this study is the first optogenetic investigation of enteric neurons from a single putative functional class. Our abstract and discussion emphasizes this point and differentiates this study from those previous.

(3) Critical controls are needed to support the optogenetic experiments. Control experiments are needed to show that ChR2 expression a) does not change the baseline properties of the neurons, b) that stimulation with the chosen intensity of light elicits physiologically relevant responses in those neurons, and c) that stimulation via ChR2 elicits comparable responses in IPANs in the different gut regions focused on here.

We completely agree controls are essential. However, our paper is not the first to express ChR2 in enteric neurons. Authors of our paper have shown in Hibberd et al. 2018 that expression of ChR2 in a heterogeneous population of myenteric neurons did not change network properties of the myenteric plexus. This was demonstrated in the lack of change in control CMC characteristics in mice expressing ChR2 under basal conditions (without blue light exposure). Regarding question (b), that it should be shown that stimulation with the chosen intensity of light elicits physiologically relevant responses in those neurons. We show the restricted expression of ChR2 in IPANs and that motor responses (to blue light) are blocked by selective nerve conduction blockade.

Regarding question (c), that our study should demonstrate that stimulation via ChR2 elicits comparable responses in IPANs in the different gut regions. We would not expect each region of the gut to behave comparably. This is because the different gut regions (i.e. proximal, mid, distal) are very different anatomically, as is anatomy of the myenteric plexus and myenteric ganglia between each region, including the density of IPANs within each ganglia, in addition to the presence of different patterns of electrical and mechanical activity [Spencer et al., 2020]. Hence, it is difficult to expect that between regions stimulation of ChR2 should induce similar physiological responses. The motor output we record in our study (CMCs) is a unified motor program that involves the temporal coordination of hundreds of thousands of enteric neurons and a complex neural circuit that we have previously characterized [Spencer et al., 2018]. But, never has any study until now been able to selectively stimulate a single functional class of enteric neurons (with light) to avoid indiscriminate activation of other classes of neurons.

(1) T. J. Hibberd, J. Feng, J. Luo, P. Yang, V. K. Samineni, R. W. Gereau, N. Kelley, H. Hu, N. J. Spencer, Optogenetic Induction of Colonic Motility in Mice. Gastroenterology 155, 514-528.e6 (2018).

(2) N. J. Spencer, L. Travis, L. Wiklendt, T. J. Hibberd, M. Costa, P. Dinning, H. Hu, Diversity of neurogenic smooth muscle electrical rhythmicity in mouse proximal colon. American Journal of Physiology-Gastrointestinal and Liver Physiology 318, G244–G253 (2020).

(3) N. J. Spencer, T. J. Hibberd, L. Travis, L. Wiklendt, M. Costa, H. Hu, S. J. Brookes, D. A. Wattchow, P. G. Dinning, D. J. Keating, J. Sorensen, Identification of a Rhythmic Firing Pattern in the Enteric Nervous System That Generates Rhythmic Electrical Activity in Smooth Muscle. The Journal of Neuroscience 38, 5507–5522 (2018).

(4) The electrophysiological characterization of mouse IPANs is useful but this is a basic characterization of any IPAN and really says nothing specifically about Cdh6+ neurons. The electrophysiological characterization was also only done in a small fraction of colonic IPANs, and it is not clear if these represent cell properties in the distal colon or proximal colon, and whether these properties might be extrapolated to IPANs in the different regions. Similarly, blocking IH with ZD7288 affects all IPANs and does not add specific information regarding the role of the proposed Cdh6+ subtype.

Our electrophysiological characterization was guided to be within a subset of Cdh6+ neurons by Hb9:GFP expression. As in the prior comment (1) above, we used these experiments to confirm classification of Cdh6+ (Hb9:GFP+) neurons in the distal colon as IPANs. We have clarified in the results and methods that these experiments were performed in the distal colon and agree that we cannot extrapolate that these properties are also representative of IPANs in the proximal colon. We apologize that this was confusing. Finally, we agree with the reviewer that ZD7288 affects all IPANs in the ENS and have clarified this in the text.

(5) Why SMP IPANs were not included in the analysis of Cdh6 expression is a little puzzling. IPANs are present in the SMP of the small intestine and colon, and it would be useful to know if this proposed marker is also present in these cells.

We agree with the reviewer. In addition to characterizing Cdh6 in the myenteric plexus, it would be interesting to query if sensory neurons located within the SMP also express Cdh6. Our preliminary data (n=2) show ~6-12% tdT/Hu neurons in Cdh6-tdT ileum and colon (data not shown). We have added a sentence to the discussion.

(6) The emphasis on IH being a rhythmicity indicator seems a bit premature. There is no evidence to suggest that IH and IT are rhythm-generating currents in the ENS.

Regarding the statement there is no evidence to suggest that IH and IT are rhythm-generating currents in the ENS. We agree with the reviewer that evidence of rhythm generation by IH and IT in the ENS has not been explicitly confirmed. We are confident the reviewer agrees that an absence of evidence is not evidence of absence, although the presence of IH has been well described in enteric neurons. We have modified the text in the results to indicate more clearly that IH and IT are known to participate in rhythm generation in thalamocortical circuits, though their roles in the ENS remain unknown. Our discussion of the potential role of IH or IT in rhythm generation or oscillatory firing of the ENS is constrained to speculation in the discussion section of the text.

(7) As the authors point out in the introduction and discuss later on, Type II Cadherins such as Cdh6 bind homophillically to the same cadherin at both pre- and post-synapse. The apparent enrichment of Cdh6 in IPANs would suggest extensive expression in synaptic terminals that would also suggest extensive IPAN-IPAN connections unless other subtypes of neurons express this protein. Such synaptic connections are not typical of IPANs and raise the question of whether or not IPANs actually express the functional protein and if so, what might be its role. Not having this information limits the usefulness of this as a proposed marker.

We agree with the reviewer that the proposed IPAN-IPAN connection is novel although it has been proposed before (Kunze et al., 1993). As detailed in our response to Reviewer #1, we attempted to confirm Cdh6 protein expression, but were unsuccessful, due to insufficient signal and resolution. We therefore discuss potential IPAN interconnectivity in the discussion, in the context of contrasting literature.

(1) W. A. A. Kunze, J. B. Furness, J. C. Bornstein, Simultaneous intracellular recordings from enteric neurons reveal that myenteric ah neurons transmit via slow excitatory postsynaptic potentials. Neuroscience 55, 685–694 (1993).

(8) Experiments shown in Figures 6J and K use a tethered pellet to drive motor responses. By definition, these are not CMCs as stated by the authors.

The reviewer makes a valid criticism as to the terminology, since tethered pellet experiments do not record propagation. We believe the periodic bouts of propulsive force on the pellet is triggered by the same activity underlying the CMC. In our experience, these activities have similar periodicity, force and identical pharmacological properties. Consistent with this, we also tested full colons (n = 2) set up for typical CMC recordings by multiple force transducers, finding that CMCs were abolished by ZD7288, similar to fixed pellet recordings (data not shown).

(9) The data from the optogenetic experiments are difficult to understand. How would stimulating IPANs in the distal colon generate retrograde CMCs and stimulating IPANs in the proximal colon do nothing? Additional characterization of the Cdh6+ population of cells is needed to understand the mechanisms underlying these effects.

We agree that the different optogenetic responses in the proximal and distal colon are challenging to interpret, but perhaps not surprising in the wider context. It is not only possible that the different optogenetic responses in this study reflect regional differences in the Chd6+ neuronal populations, but also differences in neural circuits within these gut regions. A study some time ago by the authors showed that electrical stimulation of the proximal mouse colon was unable to evoke a retrograde (aborally) propagating CMC (Spencer, Bywater, 2002), but stimulation of the distal colon was readily able to. We concluded that at the oral lesion site there is a preferential bias of descending inhibitory nerve projections, since the ascending excitatory pathways have been cut off. In contrast, stimulation of the distal colon was readily able to activate an ascending excitatory neural pathway, and hence induce the complex CMC circuits required to generate an orally propagating CMC. Indeed, other recent studies have added to a growing body of evidence for significant differences in the behaviors and neural circuits of the two regions (Li et al., 2019, Costa et al., 2021a, Costa et al., 2021b, Nestor-Kalinoski et al., 2022). We have expanded this discussion.

(1) N. J. Spencer, R. A. Bywater, Enteric nerve stimulation evokes a premature colonic migrating motor complex in mouse. Neurogastroenterology & Motility 14, 657–665 (2002).

(2) Li Z, Hao MM, Van den Haute C, Baekelandt V, Boesmans W, Vanden Berghe P, Regional complexity in enteric neuron wiring reflects diversity of motility patterns in the mouse large intestine. Elife 8:e42914 (2019).

(3) Costa M, Keightley LJ, Hibberd TJ, Wiklendt L, Dinning PG, Brookes SJ, Spencer NJ, Motor patterns in the proximal and distal mouse colon which underlie formation and propulsion of feces. Neurogastroenterology & Motility e14098 (2021a).

(4) Costa M, Keightley LJ, Hibberd TJ, Wiklendt L, Smolilo DJ, Dinning PG, Brookes SJ, Spencer NJ, Characterization of alternating neurogenic motor patterns in mouse colon. Neurogastroenterology & Motility 33:e14047 (2021b).

(5) Nestor-Kalinoski A, Smith-Edwards KM, Meerschaert K, Margiotta JF, Rajwa B, Davis BM, Howard MJ, Unique Neural Circuit Connectivity of Mouse Proximal, Middle, and Distal Colon Defines Regional Colonic Motor Patterns. Cellular and Molecular Gastroenterology and Hepatology 13:309-337.e303 (2022).

Recommendations for the Authors:

Reviewer #1 (Recommendations for the authors):

As mentioned above, immunolocalization of cdh6 would be helpful to substantiate the claims regarding IPAN-IPAN synapses.

As mentioned in our response to both reviewers’ public reviews, we attempted to visualize Cdh6 protein via antibody staining (Duan et al., 2018), but our efforts did not result in sufficient signal or resolution to identify Cdh6+ synapses.

(1) X. Duan, A. Krishnaswamy, M. A. Laboulaye, J. Liu, Y.-R. Peng, M. Yamagata, K. Toma, J. R. Sanes, Cadherin Combinations Recruit Dendrites of Distinct Retinal Neurons to a Shared Interneuronal Scaffold. Neuron 99, 1145-1154.e6 (2018).

Reviewer #2 (Recommendations for the authors):

(1) The authors repeatedly refer to IPANs as "sensory" neurons (e.g. in title, abstract, and introduction) but there is some debate regarding whether these cells are truly "sensory" because the information they convey never reaches sensory perception. This is why they have classically been referred to as intrinsic primary afferent (IPAN) neurons. It would be more appropriate to stick with this terminology unless the authors have compelling data showing that information detected by IPANs reaches the sensory cortex.

We thank the reviewer for their comment, but respectfully disagree. The term “sensory neuron” is well established in the ENS. The first definitive proof that “sensory neurons” exist in the ENS was published in Kunze et al., 1995. We note that this paper did not use the word “IPAN” but used the term “sensory neuron”. Furthermore, mechanosensory neurons were published in Spencer and Smith (2004).

Regarding the reviewer’s comment that the authors would need compelling data showing that information detected by IPANs reaches the sensory cortex before the term “sensory neuron” should be valid, it is important to note that many sensory neurons do not provide direct information to the cortex.

(1) W. A. A. Kunze, J. C. Bornstein, J. B. Furness, Identification of sensory nerve cells in a peripheral organ (the intestine) of a mammal. Neuroscience 66, 1–4 (1995).

(2) N. J. Spencer, T. K. Smith, Mechanosensory S-neurons rather than AH-neurons appear to generate a rhythmic motor pattern in guinea-pig distal colon. The Journal of Physiology 558, 577–596 (2004).

(2) Important information regarding the gut region shown and other details are absent from many figure legends.

We apologize for this omission. We have updated the figure legends to include information on gut regions.

eLife Assessment

This valuable study reports on the critical role of ANKRD5 (ANKEF1) in sperm motility and male fertility. However, the supporting data remain incomplete. This work will be of interest to biomedical researchers working in sperm biology and andrologists.

Reviewer #1 (Public review):

Summary:

Asthenospermia, characterized by reduced sperm motility, is one of the major causes of male infertility. The "9 + 2" arranged MTs and over 200 associated proteins constitute the axoneme, the molecular machine for flagellar and ciliary motility. Understanding the physiological functions of axonemal proteins, particularly their links to male infertility, could help uncover the genetic causes of asthenospermia and improve its clinical diagnosis and management. In this study, the authors generated Ankrd5 null mice and found that ANKRD5-/- males exhibited reduced sperm motility and infertility. Using FLAG-tagged ANKRD5 mice, mass spectrometry, and immunoprecipitation (IP) analyses, they confirmed that ANKRD5 is localized within the N-DRC, a critical protein complex for normal flagellar motility. However, transmission electron microscopy (TEM) and cryo-electron tomography (cryo-ET) of sperm from Ankrd5 null mice did not reveal any structural abnormalities.

Strengths:

The phenotypes observed in ANKRD5-/- mice, including reduced sperm motility and male infertility, are conversing. The authors demonstrated that ANKRD5 is an N-DRC protein that interacts with TCTE1 and DRC4. Most of the experiments are thoughtfully designed and well executed.

Weaknesses:

The cryo-FIB and cryo-ET analyses require further investigation, as detailed below. The molecular mechanism by which the loss of ANKRD5 affects sperm flagellar motility remains unclear. The current conclusion that Ankrd5 knockout reduces axoneme stability is not well-supported. Specifically, are other axonemal proteins diminished in Ankrd5 knockout sperm? Conducting immunofluorescence analyses and revisiting the quantitative proteomics data may help address these questions.

Reviewer #2 (Public review):

Summary:

The manuscript investigates the role of ANKRD5 (ANKEF1) as a component of the N-DRC complex in sperm motility and male fertility. Using Ankrd5 knockout mice, the study demonstrates that ANKRD5 is essential for sperm motility and identifies its interaction with N-DRC components through IP-mass spectrometry and cryo-ET. The results provide insights into ANKRD5's function, highlighting its potential involvement in axoneme stability and sperm energy metabolism.

Strengths:

The authors employ a wide range of techniques, including gene knockout models, proteomics, cryo-ET, and immunoprecipitation, to explore ANKRD5's role in sperm biology.

Weaknesses:

(1) Limited Citations in Introduction: Key references on the role of N-DRC components (e.g., DRC1, DRC2, DRC3, DRC5) in male infertility are missing, which weakens the contextual background.

(2) Lack of Functional Insights: While interacting proteins outside the N-DRC complex were identified, their potential roles and interactions with ANKRD5 are not adequately explored or discussed.

(3) Mitochondrial Function Uncertainty: Immunofluorescence suggests possible mitochondrial localization for ANKRD5, but experiments on its role in energy metabolism (e.g., ATP production, ROS) are insufficient, especially given the observed sperm motility defects.

(4) Glycolysis Pathway Impact: Proteomic analysis indicates glycolysis pathway disruptions in Ankrd5-deficient sperm, but the link between these changes and impaired motility is not well explained.

(5) Cryo-ET Data Limitations: The structural analysis of the DMT lacks clarity on how ANKRD5 influences N-DRC or RS3. The low quality of RS3 data hinders the interpretation of ANKRD5's impact on axoneme structure.

(6) Discussion of Findings: The manuscript could benefit from a deeper discussion on the broader implications of ANKRD5's interactions and its role in sperm energy metabolism and motility mechanisms.

Author response:

Thank you for the constructive feedback from the reviewers. We are grateful for their insights and are committed to addressing the key concerns raised in the public reviews through the following revisions:

(1) Validating Axoneme Stability Claims

We have procured new antibodies for DRC11, as well as marker proteins for ODA, IDA, and RS. We will conduct quantitative immunofluorescence staining to validate our claims regarding axoneme stability.

(2) Investigating ANKRD5 Expression in Other Ciliated Cells

We plan to examine the expression of ANKRD5 in mouse respiratory cilia to determine whether it is also expressed in these cells.

(3) Supplementing Key Citations for N-DRC Components

We will add references to published studies on N-DRC components (e.g., DRC1, DRC2, DRC3, DRC5) associated with male infertility in the Introduction to strengthen the background context.

(4) Further Analysis and Validation of ANKRD5 Interactome

We will conduct additional analyses and validation of the interactome of ANKRD5 detected by LC-MS.

(5) Elucidating the Function of ANKRD5 in Mitochondria

We will further investigate the role of ANKRD5 in mitochondrial function.

(6) Investigating Mitochondrial Function and Energy Metabolism

We will further explore the role of ANKRD5 in mitochondrial function and energy metabolism.

(7) Improving Cryo-ET Data Quality and Interpretation

We will attempt to further improve the quality of the STA results and try to calculate the DMT structure with a period of 96 nm. We will also use the WT density map with the same period to generate a difference map.

(8) Expanding Discussion and Correcting Terminology

The Discussion section will be revised to elaborate on the implications of ANKRD5 for male contraceptive research, particularly in targeting sperm motility. We will also correct terminology inaccuracies (e.g., changing "9+2 microtubule doublet" to "9+2 structure") and address formatting issues (e.g., capitalizing "Control").

Response to Reviewer #2 Comment 4:

We appreciate the reviewer's careful consideration of our proteomic data. However, our Gene Set Enrichment Analysis (GSEA) of glycolysis/gluconeogenesis pathways showed no significant enrichment (p-value=0.089, NES=0.708; Fig.6D), which does not meet the statistical thresholds for biological significance (|NES|>1, pvalue<0.05). This observation is further corroborated by our direct ATP measurements showing no difference between genotypes (Fig.6E). We agree that further studies on metabolic regulation could be valuable, but current evidence does not support glycolysis disruption as a primary mechanism for the motility defects observed in Ankrd5-null sperm. This misinterpretation likely arose from the reviewer's overinterpretation of non-significant proteomic trends. We request that this specific claim be excluded from the assessment to avoid misleading readers.

We will provide a comprehensive point-by-point response, along with detailed experimental data and revised figures, in the resubmitted manuscript. Thank you once again for the opportunity to address the reviewers' concerns. We are confident that these revisions will strengthen our manuscript and contribute to the scientific community.

eLife Assessment

This study demonstrates the critical role of Afadin on the generation and maintenance of complex cellular layers in the mouse retina. The data are solid, which provides important insights into how cell-adhesion molecules contribute to retinal organization. However, further investigations are needed to clarify the mechanisms underlying the cellular disorganization phenotype in the retina and axonal projection to the brain.

Reviewer #1 (Public review):

Summary:

In this study, the authors examined the role of Afadin, a key adaptor protein associated with cell-adhesion molecules, in retinal development. Using a conditional knockout mouse line (Six3-Cre; AfadinF/F), the authors successfully characterized a disorganized pattern of various neuron types in the mutant retinae. Despite these altered distributions, the retinal neurons maintained normal cell numbers and seemingly preserved some synaptic connections. Notably, tracing results indicated mistargeting of retinal ganglion cell (RGC) axon projections to the superior colliculus, and electroretinography (ERG) analyses suggested deficits in visual functions.

Strengths:

This compelling study provides solid evidence addressing the important question of how cell-adhesion molecules influence neuronal development. Compared to previous research conducted in other parts of the central nervous system (CNS), the clearly defined lamination of cell types in the retina serves as a unique model for studying the aberrant neuronal localizations caused by Afadin knockout. The data suggest that cell-cell interactions are critical for retinal cellular organization and proper axon pathfinding, while aspects of cell fate determination and synaptogenesis remain less understood. This work has broad implications not only for retinal studies but also for developmental biology and regenerative medicine.

Weaknesses:

While the phenotypes observed in the Afadin knockout (cKO) mice are intriguing, I would expect to see evidence confirming that Afadin is indeed knocked out in the retina through immunostaining. Specifically, is Afadin knocked out only in certain retinal regions and not others, as suggested by Figures 4A-B? Are Afadin levels different among distinct neuron types, which could mean that its knockout may have a more pronounced impact on certain cell types, such as rods compared to others?

The authors suggest that synapses may form between canonical synaptic partners, based on the proximity of their processes (Figure 2). However, more solid evidence is needed to verify these synapses through the use of synaptic marker staining or transsynaptic labeling before drawing further conclusions.

Although the Afadin cKO mice displayed dramatic phenotypes, additional experiments are necessary to clarify the details of this process. By manipulating Afadin levels in specific cell types or at different developmental time points, we could gain a better understanding of how Afadin regulates accurate retinal lamination and axonal projection.

Reviewer #2 (Public review):

Summary:

This study by Lum and colleagues reports on the role of Afadin, a cytosolic adapter protein that organizes multiple cell adhesion molecule families, in the generation and maintenance of complex cellular layers in the mouse retina. They used a conditional deletion approach, removing Afadin in retinal progenitors, and allowing them to analyze broad effects on retinal neuron development.

The study presents high-quality and extensive characterization of the cellular phenotypes, supporting the main conclusions of the paper. They show that Afadin loss results in significant disorganization of the retinal cellular layers and the neuropil, producing rosettes and displacement of cells away from their resident layers. The major classes of neurons in the inner retina are affected, and some neurons are, remarkably, displaced to the other side of the inner plexiform layer. Nevertheless, they mostly target their synaptic partners, including the RGCs to distant retinorecipient targets in the brain. The main conclusions are as follows. Afadin is necessary for establishing and maintaining the retinal architecture. It is not necessary for the generation of the correct numbers/densities of retinal neuron subtypes. Moreover, Afadin loss preserves associations between known synaptic partners and preserves axonal targeting to retinorecipient layers. The consequences on photoreceptor viability and visual processing are also interesting, underscoring the essential function for maintaining retinal structure and function. Overall the main conclusions describing the consequences are supported by the results.

Strengths:

The study provides new knowledge on the requirement of Afadin in retinal development. The introduction and discussion effectively set up the rationale for this work, and place it in the context of previous studies of Afadin in other regions of the CNS.

The study presents high-quality and extensive characterizations of the cellular phenotypes resulting from Afadin loss. By analyzing various aspects of retinal organization - from cellular densities to axon targeting to brain - the study narrows down the role of a structure for promoting the establishment of the layers, or maintenance. The data are straightforward and convincing, and the interpretations are bounded by the data shown (though minor weakness re. survival). Another important finding is that the targeting of retinal neuron processes to synaptic partners, including retinorecipient targets in the brain, are intact.

The study is important as it establishes a focused requirement for Afadin to set up and preserve the overall cellular organizations within the retinal tissue. The demonstration that Afadin is needed for photoreceptor viability and overall visual function enhances impact by establishing its functional importance.

The manuscript is well well-written and presented. The images are attractive and compelling, and the figures are well organized.

Weaknesses:

(1) Expanding on the developmental mechanism is beyond the scope of the study, and would not add to the main conclusions. However, the manuscript would be improved by providing more clarity on the developmental emergence of the defects. The study left me questioning whether the rosettes and cell displacements occur during earlier stages of retina development, or are progressive. For instance, do the RGCs migrate and establish within the GCL correctly at first, and then are displaced with the progressive disorganization? Or are they disorganized and delaminate en route? Images of RGC staining at P0, or earlier during their migration, would be informative. Data in Figure 1 is limited to DAPI staining at P7. Figure 4 shows an image of rod photoreceptors at P7, with their displacement in the GCL layer (and not contained within a rosette). Are the progenitors mislocalized due to delamination?

A few additional thoughts on how these defects compare to other mutants with rosettes might give us more context for understanding the results.

(2) The manuscript reports that the densities of major inner retinal classes are unaffected. There are a few details missing for this point. How were the cell densities quantified (in terms of ROI size), and normalized? This information is lacking in the methods. There is a striking thickening of the GCL in the DAPI-labeled images shown in Figure 1. What are these cells?

Author response:

Public Reviews:

Reviewer #1 (Public review):

Summary:

In this study, the authors examined the role of Afadin, a key adaptor protein associated with cell-adhesion molecules, in retinal development. Using a conditional knockout mouse line (Six3-Cre; AfadinF/F), the authors successfully characterized a disorganized pattern of various neuron types in the mutant retinae. Despite these altered distributions, the retinal neurons maintained normal cell numbers and seemingly preserved some synaptic connections. Notably, tracing results indicated mistargeting of retinal ganglion cell (RGC) axon projections to the superior colliculus, and electroretinography (ERG) analyses suggested deficits in visual functions.

Thank you for the summary and highlights of our study. We appreciate the input from Reviewer 1 and the Editor on this study, with focus on laminar choices, synaptic choices and axonal projections.

Strengths:

This compelling study provides solid evidence addressing the important question of how cell-adhesion molecules influence neuronal development. Compared to previous research conducted in other parts of the central nervous system (CNS), the clearly defined lamination of cell types in the retina serves as a unique model for studying the aberrant neuronal localizations caused by Afadin knockout. The data suggest that cell-cell interactions are critical for retinal cellular organization and proper axon pathfinding, while aspects of cell fate determination and synaptogenesis remain less understood. This work has broad implications not only for retinal studies but also for developmental biology and regenerative medicine.

Weaknesses:

While the phenotypes observed in the Afadin knockout (cKO) mice are intriguing, I would expect to see evidence confirming that Afadin is indeed knocked out in the retina through immunostaining. Specifically, is Afadin knocked out only in certain retinal regions and not others, as suggested by Figures 4A-B? Are Afadin levels different among distinct neuron types, which could mean that its knockout may have a more pronounced impact on certain cell types, such as rods compared to others?

The authors suggest that synapses may form between canonical synaptic partners, based on the proximity of their processes (Figure 2). However, more solid evidence is needed to verify these synapses through the use of synaptic marker staining or transsynaptic labeling before drawing further conclusions.

Although the Afadin cKO mice displayed dramatic phenotypes, additional experiments are necessary to clarify the details of this process. By manipulating Afadin levels in specific cell types or at different developmental time points, we could gain a better understanding of how Afadin regulates accurate retinal lamination and axonal projection.

Regarding the antibody confirming the Knockout, we tested the commercially available antibody from Sigma but weren’t able to confirm its specificity. There was a homemade antibody from another Japan-based laboratory, but it was not available to share at the moment when the study was conducted. Nonetheless, the original allele was derived for hippocampal and cortical studies by Louis Reichardt’s Lab (UCSF), with verified efficacies of the KO allele.

Regarding phenotypical penetrance, this may likely come from the mosaicism of the clone and the symmetric cell division, leading to a rosette-like structure. At this moment, we reason that Afadin KO does NOT lead to direct neuronal loss, and the selective rod loss may derive from other issues, but we lack direct evidence to validate this point.

In regards to the specific neuronal types and synaptic pairs, we acknowledge the limitations of the current Figure 2 in linking the mutant phenotypes to circuit changes. However, the current genetic reagents (Six3Cre) are not compatible with neuron-type specific labeling of synaptic labeling – i.e., cell type-specific Cre and additional Cre-dependent AAV tools might be desired. To do so, we will need to initiate cell-type-specific breeding of transgenic markers such as Hb9GFP for ooDSGCs, or Chat-Cre, VGlut3-Cre for starburst amacrine cells, vG3 amacrine cells, followed by retinal physiology. These experiments take multi-allelic genetic crosses for a very low breeding yield (1/16 or 1/32 Mendelian ratio). These extensive genetic tests are beyond the scope of the current manuscript.

Reviewer #2 (Public review):

Summary:

This study by Lum and colleagues reports on the role of Afadin, a cytosolic adapter protein that organizes multiple cell adhesion molecule families, in the generation and maintenance of complex cellular layers in the mouse retina. They used a conditional deletion approach, removing Afadin in retinal progenitors, and allowing them to analyze broad effects on retinal neuron development.

The study presents high-quality and extensive characterization of the cellular phenotypes, supporting the main conclusions of the paper. They show that Afadin loss results in significant disorganization of the retinal cellular layers and the neuropil, producing rosettes and displacement of cells away from their resident layers. The major classes of neurons in the inner retina are affected, and some neurons are, remarkably, displaced to the other side of the inner plexiform layer. Nevertheless, they mostly target their synaptic partners, including the RGCs to distant retinorecipient targets in the brain. The main conclusions are as follows. Afadin is necessary for establishing and maintaining the retinal architecture. It is not necessary for the generation of the correct numbers/densities of retinal neuron subtypes. Moreover, Afadin loss preserves associations between known synaptic partners and preserves axonal targeting to retinorecipient layers. The consequences on photoreceptor viability and visual processing are also interesting, underscoring the essential function for maintaining retinal structure and function. Overall, the main conclusions describing the consequences are supported by the results.

Strengths:

The study provides new knowledge on the requirement of Afadin in retinal development. The introduction and discussion effectively set up the rationale for this work, and place it in the context of previous studies of Afadin in other regions of the CNS.

The study presents high-quality and extensive characterizations of the cellular phenotypes resulting from Afadin loss. By analyzing various aspects of retinal organization - from cellular densities to axon targeting to brain - the study narrows down the role of a structure for promoting the establishment of the layers, or maintenance. The data are straightforward and convincing, and the interpretations are bounded by the data shown (though minor weakness re. survival). Another important finding is that the targeting of retinal neuron processes to synaptic partners, including retinorecipient targets in the brain, are intact.

The study is important as it establishes a focused requirement for Afadin to set up and preserve the overall cellular organizations within the retinal tissue. The demonstration that Afadin is needed for photoreceptor viability and overall visual function enhances impact by establishing its functional importance.

The manuscript is well well-written and presented. The images are attractive and compelling, and the figures are well organized.

Thank you for your high praise on the logic, data presentation, and significance of the current manuscript. We appreciate your comments on the novelty and impact of our study using retinal circuits as a model.

Weaknesses:

(1) Expanding on the developmental mechanism is beyond the scope of the study, and would not add to the main conclusions. However, the manuscript would be improved by providing more clarity on the developmental emergence of the defects. The study left me questioning whether the rosettes and cell displacements occur during earlier stages of retina development, or are progressive. For instance, do the RGCs migrate and establish within the GCL correctly at first, and then are displaced with the progressive disorganization? Or are they disorganized and delaminate en route? Images of RGC staining at P0, or earlier during their migration, would be informative. Data in Figure 1 is limited to DAPI staining at P7. Figure 4 shows an image of rod photoreceptors at P7, with their displacement in the GCL layer (and not contained within a rosette). Are the progenitors mislocalized due to delamination?  A few additional thoughts on how these defects compare to other mutants with rosettes might give us more context for understanding the results.

We chose P7 as our focus due to the lamination in controls. In the revised manuscript, we plan to include earlier time points, as suggested by the reviewer. The data in Figure 1 at P7 utilizes well-established cell type markers (RBPMS, Chx10, Ap2α) and is not limited only to DAPI. Additionally, we will revise the discussion section and place our mutant analyses in the context of other mutants with rosettes (beta-catenin, etc.) in the retina. Finally, we will address the comment on progenitor lamination by exploring earlier developmental time points.

(2) The manuscript reports that the densities of major inner retinal classes are unaffected. There are a few details missing for this point. How were the cell densities quantified (in terms of ROI size), and normalized? This information is lacking in the methods. There is a striking thickening of the GCL in the DAPI-labeled images shown in Figure 1. What are these cells?

We will revise the manuscript, particularly the methods section, to address these comments. Additionally, we will tackle ROI units and normalization. The cells in the thickened GCL were identified as displaced amacrine cells and bipolar cells.

eLife Assessment

Centromeres are specific sites on chromosomes that are essential for mitosis and genome fidelity. This valuable work extends previous studies to convincingly show that the centromere-histone core contributes to force transduction through the kinetochore. The centromere mainly strengthens one of the two paths of force transduction, influenced by the centromeric DNA sequence, the mechanism for which remains to be determined. This work will be of interest to those studying cell division and chromosome segregation.

Reviewer #1 (Public review):

Summary:

The authors address the role of the centromere histone core in force transduction by the kinetochore.

Strengths:

They use a hybrid DNA sequence that combines CDEII and CDEIII as well as Widom 601 so they can make stable histones for biophysical studies (provided by the Widom sequence) and maintain features of the centromere (CDE II and III).

Weaknesses:

The main results are shown in one figure (Figure 2). Indeed the Centromere core of Widom and CDE II and III contribute to strengthening the binding force for the OA-beads. The data are very nicely done and convincingly demonstrate the point. The weakness is that this is the entire paper. It is certainly of interest to investigators in kinetochore biology, but beyond that, the impact is fairly limited in scope.

Reviewer #2 (Public review):

Summary:

This paper provides a valuable addendum to the findings described in Hamilton et al. 2020 (https://doi.org/ 10.7554/eLife.56582). In the earlier paper, the authors reconstituted the budding yeast centromeric nucleosome together with parts of the budding yeast kinetochore and tested which elements are required and sufficient for force transmission from microtubules to the nucleosome. Although budding yeast centromeres are defined by specific DNA sequences, this earlier paper did not use centromeric DNA but instead the generic Widom 601 DNA. The reason is that it has so far been impossible to stably reconstitute a budding yeast centromeric nucleosome using centromeric DNA.

In this new study, the authors now report that they were able to replace part of the Widom 601 DNA with centromeric DNA from chromosome 3. This makes the assay more closely resemble the in vivo situation. Interestingly, the presence of the centromeric DNA fragment makes one type of minimal kinetochore assembly, but not the other, withstand stronger forces.

Which kinetochore assembly turned out to be affected was somewhat unexpected, and can currently not be reconciled with structural knowledge of the budding yeast centromere/kinetochore. This highlights that, despite recent advances (e.g. Guan et al., 2021; Dendooven et al., 2023), aspects of budding yeast kinetochore architecture and function remain to be understood and that it will be important to dissect the contributions of the centromeric DNA sequence.

Given the unexpected result, the study would become yet more informative if the authors were able to pinpoint which interactions contribute to the enhanced force resistance in the presence of centromeric DNA.

Strength:

The paper demonstrates that centromeric DNA can increase the attachment strength between budding yeast microtubules and centromeric nucleosomes.

Weakness:

How centromeric DNA exerts this effect remains unclear.

Author response:

Reviewer #1:

Summary:

The authors address the role of the centromere histone core in force transduction by the kinetochore.

Strengths:

They use a hybrid DNA sequence that combines CDEII and CDEIII as well as Widom 601 so they can make stable histones for biophysical studies (provided by the Widom sequence) and maintain features of the centromere (CDE II and III).

Weaknesses:

The main results are shown in one figure (Figure 2). Indeed the Centromere core of Widom and CDE II and III contribute to strengthening the binding force for the OA-beads. The data are very nicely done and convincingly demonstrate the point. The weakness is that this is the entire paper. It is certainly of interest to investigators in kinetochore biology, but beyond that, the impact is fairly limited in scope.

This reviewer might have missed that this is a Research Advance, not an article.  Research Advances are limited in scope by definition and provide a new development that builds on research reported in a prior paper.  They can be of any length.  Our Research Advance builds on our prior work, Hamilton et al., 2020 and provides the new result that native centromere sequences strengthen the attachment of the kinetochore to the nucleosome.

Reviewer #2:

Summary:

This paper provides a valuable addendum to the findings described in Hamilton et al. 2020 (https://doi.org/ 10.7554/eLife.56582). In the earlier paper, the authors reconstituted the budding yeast centromeric nucleosome together with parts of the budding yeast kinetochore and tested which elements are required and sufficient for force transmission from microtubules to the nucleosome. Although budding yeast centromeres are defined by specific DNA sequences, this earlier paper did not use centromeric DNA but instead the generic Widom 601 DNA. The reason is that it has so far been impossible to stably reconstitute a budding yeast centromeric nucleosome using centromeric DNA.

In this new study, the authors now report that they were able to replace part of the Widom 601 DNA with centromeric DNA from chromosome 3. This makes the assay more closely resemble the in vivo situation. Interestingly, the presence of the centromeric DNA fragment makes one type of minimal kinetochore assembly, but not the other, withstand stronger forces.

We thank the reviewer for their careful and positive assessment of our work.

Which kinetochore assembly turned out to be affected was somewhat unexpected, and can currently not be reconciled with structural knowledge of the budding yeast centromere/kinetochore. This highlights that, despite recent advances (e.g. Guan et al., 2021; Dendooven et al., 2023), aspects of budding yeast kinetochore architecture and function remain to be understood and that it will be important to dissect the contributions of the centromeric DNA sequence.

We couldn’t agree more.

Given the unexpected result, the study would become yet more informative if the authors were able to pinpoint which interactions contribute to the enhanced force resistance in the presence of centromeric DNA.

Strength:

The paper demonstrates that centromeric DNA can increase the attachment strength between budding yeast microtubules and centromeric nucleosomes.

Weakness:

How centromeric DNA exerts this effect remains unclear.

eLife Assessment

In this work, the authors use a Drosophila melanogaster adult ventral nerve cord injury model extending and confirming previous observations. This important study reveals key aspects of adult neural plasticity. Taking advantage of several genetic reporter and fate tracing tools, the authors provide solid evidence for different forms of glial plasticity, that are increased upon injury. The significance of the generated cell types under homeostatic conditions and in response to injury remains to be further explored and open up new avenues of research.

Reviewer #2 (Public review):

Summary:

Casas-Tinto et al., provide new insight into glial plasticity using a crush injury paradigm in the ventral nerve cord (VNC) of adult Drosophila. The authors find that both astrocyte-like glia (ALG) and ensheating glia (EG) divide under homeostatic conditions in the adult VNC and identify ALG as the glial population that specifically ramps up proliferation in response to injury, whereas the number of EGs decreases following the insult. Using lineage-tracing tools, the authors interestingly observe interconversion of glial subtypes, especially of EGs into ALGs, which occurs independent of injury and is dependent on the availability of the transcription factor Prospero in EGs, adding to the plasticity observed in the system. Finally, when tracing the progeny of glia, Casas-Tinto and colleagues detect cells of neuronal identity and provide evidence that such glia-derived neurogenesis is favored following ventral nerve cord injury, which puts forward a remarkable way in which glia can respond to neuronal damage.

Strengths:

This study highlights a new facet of adult nervous system plasticity at the level of the ventral nerve cord, supporting the view that proliferative capacity is maintained in the mature CNS and stimulated upon injury.

The injury paradigm is well chosen, as the organization of the neuromeres allows specific targeting of one segment, compared to the remaining intact and with the potential to later link observed plasticity to behavior such as locomotion.

Numerous experiments have been carried out in 7-day old flies, showing that the observed plasticity is not due to residual developmental remodeling or a still immature VNC.

Different techniques are used to observe proliferation in the VNC.

By elegantly combining different methods, the authors show glial divisions including with mitotic-dependent tracing and find that the number of generated glia is refined by apoptosis later on.

The work identifies prospero in glia as important coordinator of glial cell fate, from development to the adult context, which draws further attention to the upstream regulatory mechanisms.

Weaknesses:

The authors do not discuss their results on gliogenesis or neurogenesis in the adult VNC to previous findings made in the context of the injured adult brain.

The authors speculate about the role of glial inter-conversion for tissue homeostasis or regeneration, but no supportive evidence is cited or provided. Further experiments will be required to test the function of the described glial plasticity.

Elav+ cells originating from glia do not express markers for mature neurons at the analysed time-point. If they will eventually differentiate<br /> or what type of structure is formed by them will have to be followed up in future studies.

Context/Discussion

Highlighting some differences in the reactiveness of glia in the VNC compared to the brain could reveal important differences in repair strategies in different areas of the CNS.

Reviewer #3 (Public review):

In this manuscript, Casas-Tintó et al. explore the role of glial cell in the response to a neurodegenerative injury in the adult brain. They used Drosophila melanogaster as a model organism, and found that glial cells are able to generate new neurons through the mechanism of transdifferentiation in response to injury. This paper provides a new mechanism in regeneration, and gives an understanding to the role of glial cells in the process.

The authors have now addressed all my concerns.

Author response:

The following is the authors’ response to the previous reviews.

eLife Assessment

In this work, the authors use a Drosophila adult ventral nerve cord injury model extending and confirming previous observations; this important study reveals key aspects of adult neural plasticity. Taking advantage of several genetic reporter and fate tracing tools, the authors provide solid evidence for different forms of glial plasticity, that are increased upon injury. The data on detected plasticity under physiologic conditions and especially the extent of cell divisions and cell fate changes upon injury would benefit from validation by additional markers. The experimental part would improve if strengthened and accompanied by a more comprehensive integration of results regarding glial reactivity in the adult CNS.

Thank you very much for your thoughtful comments and constructive feedback regarding our manuscript. We appreciate all the positive remarks on the significance of our findings on neural plasticity in this Drosophila adult ventral nerve cord injury model.

In response to your suggestion, we fully agree that the continuation of this project should address in detail cell fate changes with additional markers if available, or an “omic” approach such as scRNAseq. Unfortunately, these further experiments are beyond the scope of this paper to describe the in vivo phenomena of cell reprogramming, and the cellular events that take glial cells to convert into neurons or neuronal precursors.

Additionally, we agree that the experimental part can be further improved by providing a more comprehensive integration of our results with current knowledge on glial reactivity in the adult CNS. We will revise the manuscript accordingly to include a deeper discussion of the broader implications of our findings and their alignment with existing literature.

Thank you again for your valuable input, which will undoubtedly enhance the quality of our work. We look forward to submitting the revised manuscript for your consideration.

Public Reviews:

Reviewer #1 (Public review):

Summary:

Casas-Tinto et al. present convincing data that injury of the adult Drosophila CNS triggers transdifferentiation of glial cell and even the generation of neurons from glial cells. This observation opens up the possibility to get an handle on the molecular basis of neuronal and glial generation in the vertebrate CNS after traumatic injury caused by Stroke or Crush injury. The authors use an array of sophisticated tools to follow the development of glial cells at the injury site in very young and mature adults. The results in mature adults reveal a remarkable plasticity in the fly CNS and dispels the notion that repair after injury may be only possible in nerve cords which are still developing. The observation of so called VC cells which do not express the glial marker repo could point to the generation of neurons by former glial cells.

Conclusion:

The authors present an interesting story which is technically sound and could form the basis for an in depth analysis of the molecular mechanism driving repair after brain injury in Drosophila and vertebrates.

Strengths:

The evidence for transdifferentiation of glial cells is convincing. In addition, the injury to the adult CNS shows an inherent plasticity of the mature ventral nerve cord which is unexpected.

Weaknesses:

Traumatic brain injury in Drosophila has been previously reported to trigger mitosis of glial cells and generation of neural stem cells in the larval CNS and the adult brain hemispheres. Therefore this report adds to but does not significantly change our current understanding. The origin and identity of VC cells is still unclear. The authors show that VC cells are not GABA- or glutamergic. Yet, there are many other neurotransmitter or neuropetides. It would have been nice to see a staining with another general neuronal marker such as anti-Syt1 to confirm the neuronal identity of Syt1.

We thank the reviewer for the constructive comments and positive feedback. We concur that previous studies have demonstrated glial cell proliferation in response to CNS injury. In contrast, our study focuses on glial transdifferentiation that emerges as a novel phenomenon, particularly in response to injury. We found that neuropile glia lose their glial identity and express the pan-neuronal marker Elav. To investigate the identity of these newly observed elav-positive cells, we employed anti-ChAT, antiGABA and anti-GluRIIA antibodies to determine the functional identity of these cells, besides we stained them with other neuronal markers such Enabled, Gigas or Dac (not shown); however, our attempts yielded limited success. To address this, we have now included a discussion section exploring the potential identity of these cells, considering the possibility that they may represent immature neurons.

Reviewer #2 (Public review):

Summary:

Casas-Tinto et al., provide new insight into glial plasticity using a crush injury paradigm in the ventral nerve cord (VNC) of adult Drosophila. The authors find that both astrocyte-like glia (ALG) and ensheating glia (EG) divide under homeostatic conditions in the adult VNC and identify ALG as the glial population that specifically ramps up proliferation in response to injury, whereas the number of EGs decreases following the insult. Using lineage-tracing tools, the authors interestingly observe interconversion of glial subtypes, especially of EGs into ALGs, which occurs independent of injury and is dependent on the availability of the transcription factor Prospero in EGs, adding to the plasticity observed in the system. Finally, when tracing the progeny of glia, Casas-Tinto and colleagues detect cells of neuronal identity and provide evidence that such gliaderived neurogenesis is specifically favoured following ventral nerve cord injury, which puts forward a remarkable way in which glia can respond to neuronal damage.

Strengths:

This study highlights a new facet of adult nervous system plasticity at the level of the ventral nerve cord, supporting the view that proliferative capacity is maintained in the mature CNS and stimulated upon injury.

The injury paradigm is well chosen, as the organization of the neuromeres allows specific targeting of one segment, compared to the remaining intact and with the potential to later link observed plasticity to behaviour such as locomotion.

Numerous experiments have been carried out in 7-day old flies, showing that the observed plasticity is not due to residual developmental remodelling or a still immature VNC.

By elegantly combining different methods, the authors show glial divisions including with mitotic-dependent tracing and find that the number of generated glia is refined by apoptosis later on.

The work identifies prospero in glia as an important coordinator of glial cell fate, from development to the adult context, which draws further attention to the upstream regulatory mechanisms.

We would like to thank the reviewer for his/her comments and the positive analysis of this work.

Weaknesses:

The authors observe consistent inter-conversion of EG to ALG glial subtypes that is further stimulated upon injury. The authors conclude that these findings have important consequences for CNS regeneration and potentially for memory and learning. However, it remains somewhat unclear how glial transformation could contribute to regeneration and functional recovery.

This is an ongoing question in the laboratory and in the field. We know that glial cells contribute to the regenerative program in the nervous system, and molecular signalling in glial cells is determinant for the functional recovery (Losada-Perez et al 2021). Therefore, we include this concept in the discussion as the evidence indicates that glial cells participate in these programs. However, further investigation is required to clarify and determine the mechanisms underlying this glial contribution. To determine if glial to neuron transformation contributes to functional recovery, we would need to compare the recovery of animals with new VC to animals without VC, however, the  molecular mechanism that produces this change of identity is still unknown, and therefore we are not able to generate injured flies with no new VC

The signal of the Fucci cell cycle reporter seems more complex to interpret based on the panels provided compared to the other methods employed by the authors to assess cell divisions.

We agree that Fly Fucci is a genetic reporter that might be more complex to interpret than EdU staining or other markers. However, glial cells proliferation is a milestone of this manuscript, and we used different available tools to confirm our results. We have revised this specific section to ensure that the text is clear and straightforward.

Elav+ cells originating from glia do not express markers for mature neurons at the analysed time-point. If they will eventually differentiate or what type of structure is formed by them will have to be followed up in future studies.

We fully agree with the reviewer, and we will analyze later days to study neuronal fate and contribution to VNC function.

Context/Discussion

There is some lack of connecting or later comparing the observed forms of glial plasticity in the VNC with respect to plasticity described in the fly brain.

Highlighting some differences in the reactiveness of glia in the VNC compared to the brain could point to relevant differences in repair capacity in different areas of the CNS.

Based on the assays employed, the study points to a significant amount of glial "identity" changes or interconversions under homeostatic conditions. The potential significance of this rather unexpected "baseline" plasticity in adult tissues is not explicitly pointed out and could improve the understanding of the findings.

Some speculations if "interconversion" of glia is driven by the needs in the tissue could enrich the discussion.

We would like to thank the reviewer for these suggestions. We have changed the discussion to introduce these concepts.

Reviewer #3 (Public review):

In this manuscript, Casas-Tintó et al. explore the role of glial cell in the response to a neurodegenerative injury in the adult brain. They used Drosophila melanogaster as a

model organism, and found that glial cells are able to generate new neurons through the mechanism of transdifferentiation in response to injury. This paper provides a new mechanism in regeneration, and gives an understanding to the role of glial cells in the process.

Comments on revisions:

In the previous version of the manuscript, I had suggested several recommendations for the authors. Unfortunately, none of these were addressed in the author's revision.

We are sorry for this error. We apologize but we never received these comments. We have now found them, and we have incorporated these comments in the new version of the manuscript.

(1) Have you tried screening for other markers for the EdU+ Repo+ Pros- cells?

We have identified these cells as glial cells (Repo +), and not astrocyte-like glia (pros-). But we have not further characterized  the identity of these cells. Our aim was to identify these proliferating glial cells as NPG (Neuropile glia), which are Astrocyte-Like Glia (ALG), as previous works suggest in larvae (Kato et al., 2020; Losada-Perez et al., 2016), or Ensheathing Glia (EG). To discard the ALG identity, we used prospero as the best marker. The results indicate that there are ALG among the proliferating population, but in addition, we also found pros- glial cells that were EdU positive. These cells are located in the interface between cortex and neuropile, where the neuropile glia position is described. The anti-pros staining indicated they were no ALG which suggest that they are EG.

There is no specific nuclear marker for EG cells, therefore we used FLY_FUCCI under the control of a EG specific promoter (R56F03-Gal4) to determine if the other dividing cells were EG. These results indicate that EG glia divide although their proliferation does not increase upon injury.

The R56F03 Gal4 construct is described as ensheathing glia specific by previous publications, including:

(1) Kremer M. C., Jung C., Batelli S., Rubin G. M. and Gaul U. (2017). The glia of the adult Drosophila nervous system. Glia 65, 606-638. 10.1002/glia.23115

(2) Qingzhong Ren, Takeshi Awasaki, Yu-Chun Wang, Yu-Fen Huang, Tzumin Lee. Lineage-guided Notch-dependent gliogenesis by Drosophila multi-potent progenitors. Development. 2018 Jun 11;145(11):dev160127. doi: 10.1242/dev.160127   

To summarize, our results suggest that part of these proliferating glial cells are ALG and EG. Our results can not discard that a residual part of these proliferating cells are not AG nor EG.

(2) You mentioned that ALG are heterogenous in size and shape, does that mean that you may have different subpopulations of ALG? Would that also mean that only a portion of them responds to injury?

Yes, as in Astrocytes in vertebrates this population is highly heterogeneous. Currently there are no molecular tools to specifically identify these subpopulations and characterize their distinct roles. However, emerging research suggests that differences in size, shape, and potentially molecular markers could correlate with functional diversity. This implies that certain subpopulations of ALG may be more specialized or primed to respond to injury, while others may play roles in homeostasis or other processes. Understanding this heterogeneity will require advanced techniques such as single-cell RNA sequencing, spatial transcriptomics, or live imaging to unravel how these subpopulations contribute to injury responses and overall tissue dynamics.

(3) You mentioned that NP-like cells have similar nuclear shape and size to ALG and EG, while Ventral cortex cells have larger nuclei. Can you please show a quantification of the NP-like cells and Ventral cortex cells size, and show a direct comparison with ALG and EG cells to support those claims (images, quantification and analysis)?

We added a new supplementary figure with a graph showing nuclei size differences between VC and NP-like cells, and a diagram showing VC cell localization. Images in figure 2A-A’ and 2B-B’ show both types of cells with the same scale, additionally, NPG cells are shown in red (current expression of the specific Gal4 line). A direct comparison between EG and NP-like glia can be observed in Figure 3 as well.

Besides of size and localization, we conclude  that VC and N-like cells present different molecular markers as VC are elav-positive and reponegative whereas NP-like cells are repo-positive elav-negative

(4) In Figure 2B, the repo expression is not very clear. I suggest using a different example to support the claim that NP cells are Repo+.

We have changed the color of anti-elav staining to facilitate visualisation

(5) Again, in Figure 2C, you need quantification and analysis to support the claim that you used nuclear shape and size to identify VC vs. NP like cells.

Quantification in point 3, criteria in Figure S1

(6) What is the identity of the newly formed neurons? Other than Elav, have you tried using other markers of neurons that are typically found in this area?

This question is of great interest and relevance. We have done great efforts to solve this open question and so far, our data suggest that these neurons might be in an immature state. In this last version of the manuscript, we included the results (Figure S1) with several different markers. 

The molecular identity of these cell populations, glia and neurons, is currently under investigation.

Minor comments:

(1) In the abstract, EG and ALG abbreviations are not introduced properly.

Thank you very much for noticing this missing information, we have now included it in the abstract.

(2) Please include a representation of the NPG somata location in Figure 1A.

We have included this information in the figure

(3) A schematic showing the differences between ALG and EG cells would be helpful as well.

We have included in the introduction references and reviews where other authors describe in detail the differences.

(4) In Figure 1 E, G, H- please indicated the genotype of the fly used in the panel as well as the cell type studied.

The complete genotype is included in the corresponding figure legend. We have added a simplified genotype in the figure for clarity.

(5) Please show the genotype used for images in Figure 2: ALG or EG specific drivers.

This information is included in the corresponding figure legend. We believe that it is better to keep the figure clean so we decided to keep the complete genotype, which is considerably long, only in the figure legend.

eLife Assessment

This study presents valuable findings by using Fmr1 knockout mice as a model to investigate the role of Fmr1 in sleep regulation. These mice exhibited clear evidence of sleep and circadian disturbances, including abnormal retinal innervation of the SCN, which may provide a potential mechanistic explanation for the observed behavioral deficits. Interestingly, the results suggest that a scheduled feeding approach could improve sleep and circadian rhythms while enhancing social interactions and reducing repetitive behaviors in a mouse model of Fragile X syndrome. The topic is both intriguing and highly significant; however, while the evidence supporting the authors' claims is solid, several issues hinder the manuscript's clarity and impact.

Reviewer #1 (Public review):

Summary:

The authors investigated sleep and circadian rhythm disturbances in Fmr1 KO mice. Initially, they monitored daily home cage behaviors to assess sleep and circadian disruptions. Next, they examined the adaptability of circadian rhythms in response to photic suppression and skeleton photic periods. To explore the underlying mechanisms, they traced retino-suprachiasmatic connectivity. The authors further analyzed the social behaviors of Fmr1 KO mice and tested whether a scheduled feeding strategy could mitigate sleep, circadian, and social behavior deficits. Finally, they demonstrated that scheduled feeding corrected cytokine levels in the plasma of mutant mice.

Strengths:

(1) The manuscript addresses an important topic-investigating sleep deficits in an FXS mouse model and proposing a potential therapeutic strategy.

(2) The study includes a comprehensive experimental design with multiple methodologies, which adds depth to the investigation.

Weaknesses:

(1) The first serious issue in the manuscript is the lack of a clear description of how they performed the experiments and the missing definitions of various parameters in the results. Given that monitoring and analyzing sleep behaviors are the key experiments of this manuscript, I use the "Immobility-Based Sleep Behavior" section of Methods as an example to elaborate:

Incomplete or Incorrect Description of Tracking Threshold:<br /> o The phrase "tracked the (40 sec or greater as previously described" is incomplete and does not clarify what is being tracked. This appears to be an error in writing or editing.<br /> Unclear Relationship Between Threshold and EEG Validation:<br /> o The threshold "40 sec or greater" is mentioned without context or explanation of what it represents (e.g., sleep bout duration, inactivity, or another parameter). The reference to Fisher et al. (2016) and "99% correlation with EEG-defined sleep" seems misaligned with the paragraph's content.

Confusing Definition of Sleep Bout:<br /> o The definition of a sleep bout is unclear. Sleep bouts should logically be based on periods of inactivity, not activity. The sentence suggesting sleep is measured by "activity staying above the threshold" is confusing. The phrase "3 counts of sleep per minute for longer than one minute" requires clarification.

Unclear Data Selection for Analysis:<br /> o The phrase "2 days with the best recording quality" is vague and does not specify how "best" was determined or why only two days out of five were analyzed.

Awkward Grammar and Structure:<br /> o Phrases like "Acquiring data were exported in 1-min bins" are grammatically awkward. "Acquiring" should be "Acquired." Some sentences are overly long and lack clarity, making the text harder to follow.<br /> In addition to this section, the authors should review all paragraphs in the Methods section to improve readability.

(2) Although the manuscript has a relatively long Methods section, some essential information is missing. For instance, the definition of sleep bout, as described above, is unclear. Additional missing information includes:

Figure 2: "Rhythmic strength (%)" and "Cycle-to-cycle variability (min)."<br /> Figure 3: "Activity suppression."<br /> Figure 4: "Rhythmic power (V%)" (is this different from rhythmic strength (%)?) and "Subjective day activity (%)."<br /> Figure 5: Clear labeling of the SCN's anatomical features and an explanation for quantifying only the ventral part instead of the entire SCN. Alternatively, the authors should consider quantifying the whole SCN.<br /> Figure 6: Inconsistencies in terms like "Sleep frag. (bout #)" and "Sleep bouts (#)." Consistent terminology throughout the manuscript is essential.

(3) Figure 1A shows higher mouse activity during ZT13-16. It is unclear why the authors scheduled feeding during ZT15-21, as this seems to disturb the rhythm. Consistent with this, the body weights of WT and Fmr1 KO mice decreased after scheduled feeding. The authors should explain the rationale for this design clearly.

(4) The interpretation of social behavior results in Figure 6 is questionable. The authors claim that Fmr1 KO mice cannot remember the first stranger in a three-chamber test, writing, "The reduced time in exploring and staying in the novel-mouse chamber suggested that the Fmr1 KO mutants were not able to distinguish the second novel mouse from the first now-familiar mouse." However, an alternative explanation is that Fmr1 KO mice do remember the first stranger but prefer to interact with it due to autistic-like tendencies. Data in Table 5 show that Fmr1 KO mice spent more time interacting with the first stranger in the 3-chamber social recognition test, which support this possibility. Similarly, in the five-trial social test, Fmr1 KO mice's preference for familiar mice might explain the reduced interaction with the second stranger.

In Figure 6C (five-trial social test results), only the fifth trial results are shown. Data for trials 1-4 should be provided and compared with the fifth trial. The behavioral features of mice in the 5-trial test can then be shown completely. In addition, the total interaction times for trials 1-4 (154 {plus minus} 15.3 for WT and 150 {plus minus} 20.9 for Fmr1 KO) suggest normal sociability in Fmr1 KO mice (it is different from the results of 3-chamber). Thus, individual data for trials 1-4 are required to draw reliable conclusions.

In Table 6 and Figure 6G-6J, the authors claim that "Sleep duration (Figures 6G, H) and fragmentation (Figures 6I, J) exhibited a moderate-strong correlation with both social recognition and grooming." However, Figure 6I shows a p-value of 0.077, which is not significant. Moreover, Table 6 shows no significant correlation between SNPI of the three-chamber social test and any sleep parameters. These data do not support the authors' conclusions.

(5) Figure 7 demonstrates the effect of scheduled feeding on circadian activity and sleep behaviors, representing another critical set of results in the manuscript. Notably, the WT+ALF and Fmr1 KO+ALF groups in Figure 7 underwent the same handling as the WT and Fmr1 KO groups in Figures 1 and 2, as no special treatments were applied to these mice. However, the daily patterns observed in Figures 7A, 7B, 7F, and 7G differ substantially from those shown in Figures 2B and 1A, respectively. Additionally, it is unclear why the WT+ALF and Fmr1 KO+ALF groups did not exhibit differences in Figures 7I and 7J, especially considering that Fmr1 KO mice displayed more sleep bouts but shorter bout lengths in Figures 1C and 1D.

Furthermore, it is not specified whether the results in Figure 7 were collected after two weeks of scheduled feeding (for how many days?) or if they represent the average data from the two-week treatment period.

The rationale behind analyzing "ZT 0-3 activity" in Figure 7D instead of the parameters shown in Figures 2C and 2D is also unclear.

In Figure 7F, some data points appear to be incorrectly plotted. For instance, the dark blue circle at ZT13 connects to the light blue circle at ZT14 and the dark blue circle at ZT17. This is inconsistent, as the dark blue circle at ZT13 should link to the dark blue circle at ZT14. Similarly, it is perplexing that the dark blue circle at ZT16 connects to both the light blue and dark blue circles at ZT17. Such errors undermine confidence in the data. The authors need to provide a clear explanation of how these data were processed.

Lastly, in the Figure 7 legend, Table 6 is cited; however, this appears to be incorrect. It seems the authors intended to refer to Table 7.

(6) Similar to the issue in Figure 7F, the data for day 12 in Supplemental Figure 2 includes two yellow triangles but lacks a green triangle. It is unclear how the authors constructed this chart, and clarification is needed.

(7) In Figure 8, a 5-trial test was used to assess the effect of scheduled feeding on social behaviors. It is essential to present the results for all trials (1 to 4). Additionally, it is unclear whether the results for familial mice in Figure 8A correspond to trials 1, 2, 3, or 4.<br /> The legend for Figure 8 also appears to be incorrect: "The left panels show the time spent in social interactions when the second novel stranger mouse was introduced to the testing mouse in the 5-trial social interaction test. The significant differences were analyzed by two-way ANOVA followed by Holm-Sidak's multiple comparisons test with feeding treatment and genotype as factors." This description does not align with the content of the left panels. Moreover, two-way ANOVA is not the appropriate statistical analysis for Figure 8A. The authors need to provide accurate details about the analysis and revise the figure legend accordingly.

(8) The circadian activity and sleep behaviors of Fmr1 KO mice have been reported previously, with some findings consistent with the current manuscript, while others contradict it. Although the authors acknowledge this discrepancy, it seems insufficiently thorough to simply state that the reasons for the conflicts are unknown. Did the studies use the same equipment for behavior recording? Were the same parameters used to define locomotor activity and sleep behaviors? The authors are encouraged to investigate these details further, as doing so may uncover something interesting or significant.

(9) Some subtitles in the Results section and the figure legends do not align well with the presented data. For example, in the section titled "Reduced rhythmic strength and nocturnality in the Fmr1 KOs," it is unclear how the authors justify the claim of altered nocturnality in Fmr1 KO mice. How do the authors define changes in nocturnality? Additionally, the tense used in the subtitles and figure legends is incorrect. The authors are encouraged to carefully review all subtitles and figure legends to correct these errors and enhance readability.

Reviewer #2 (Public review):

Summary:

In the present study, the authors, using a mouse model of Fragile X syndrome, explore the very interesting hypothesis that restricting food access over a daily schedule will improve sleep patterns and, subsequently, behavioral capacities. By restricting food access from 12h to 6h over the nocturnal period (active period for mice), they show, in these KO mice, an improvement of the sleep pattern accompanied by reduced systemic levels of inflammatory markers and improved behavior. Using a classical mouse model of neurodevelopmental disorder (NDD), these data suggest that eating patterns might improve sleep quality, reduce inflammation and improve cognitive/behavioral capacities in children with NDD.

Strengths:

Overall, the paper is very well-written and easy to follow. The rationale of the study is generally well-introduced. The data are globally sound. The provided data support the interpretation overall.

Weaknesses:

(1) The introduction part is quite long in the Abstract, leaving limited space for the data provided by the present study.

(2) A couple of points are not totally clear for a non-expert reader:<br /> - The Fmr1/Fxr2 double KO mice are not well described.<br /> - What is the rationale for performing both LD and DD measures?

(3) The data on cytokines and chemokines are interesting. However, the rationale for the selection of these molecules is not given. In addition, these measures have been performed in the systemic blood. Measures in the brain could be very informative.

(4) An important question is the potential impact of fasting vs the impact of the food availability restriction. Indeed fasting has several effects on brain functioning including cognitive functions.

(5) How do the authors envision the potential translation of the present study to human patients? How to translate the 12 to 6 hours of food access in mice to children with Fragile X syndrome?

eLife Assessment

This study presents an important discovery regarding the diversity and evolution of gall-forming microbial effectors. Supported by convincing computational structural predictions and analyses, the research provides insights into the unique mechanisms by which gall-forming microbes exert their pathogenicity in plants. This study also offers guidance that is of value for future studies on pathogen effector function and co-evolution with host plants.

Reviewer #1 (Public review):

Summary:

This manuscript presents a comprehensive structure-guided secretome analysis of gall-forming microbes, providing valuable insights into effector diversity and evolution. The authors have employed AlphaFold2 to predict the 3D structures of the secretome from selected pathogens and conducted a thorough comparative analysis to elucidate commonalities and unique features of effectors among these phytopathogens.

Strengths:

The discovery of conserved motifs such as 'CCG' and 'RAYH' and their central role in maintaining the overall fold is an insightful finding. Additionally, the discovery of a nucleoside hydrolase-like fold conserved among various gall-forming microbes is interesting.

Weaknesses:

Important conclusions are not verified by experiments.

Reviewer #2 (Public review):

Summary:

Soham Mukhopadhyay et al. investigated the protein folding of the secretome from gall-forming microbes using the AI-based structure modeling tool AlphaFold2. Their study analyzed six gall-forming species, including two Plasmodiophorid species and four others spanning different kingdoms, along with one non-gall-forming Plasmodiophorid species, Polymyxa betae. The authors found no effector fold specifically conserved among gall-forming pathogens, leading to the conclusion that their virulence strategies are likely achieved through diverse mechanisms. However, they identified an expansion of the Ankyrin repeat family in two gall-forming Plasmodiophorid species, with a less pronounced presence in the non-gall-forming Polymyxa betae. Additionally, the study revealed that known effectors such as CCG and AvrSen1 belong to sequence-unrelated but structurally similar (SUSS) effector clusters.

Strengths:

(1) The bioinformatics analyses presented in this study are robust, and the AlphaFold2-derived resources deposited in Zenodo provide valuable resources for researchers studying plant-microbe interactions. The manuscript is also logically organized and easy to follow.

(2) The inclusion of the non-gall-forming Polymyxa betae strengthens the conclusion that no effector fold is specifically conserved in gall-forming pathogens and highlights the specific expansion of the Ankyrin repeat family in gall-forming Plasmodiophorids.

(3) Figure 4a and 4b effectively illustrate the SUSS effector clusters, providing a clear visual representation of this finding.

(4) Figure 1 is a well-designed, comprehensive summary of the number and functional annotations of putative secretomes in gall-forming pathogens. Notably, it reveals that more than half of the analyzed effectors lack known protein domains in some pathogens, yet some were annotated based on their predicted structures, despite the absence of domain annotations.

Weaknesses:

(1) The effector families discussed in this paper remain hypothetical in terms of their functional roles, which is understandable given the challenges of demonstrating their functions experimentally. However, this highlights the need for experimental validation as a next step.

(2) Some analyses, such as those in Figure 4e, emphasize motifs derived from sequence alignments of SUSS effector clusters. Since these effectors are sequence-unrelated, sequence alignments might be unreliable. It would be more rigorous to perform structure-based alignments in addition to sequence-based ones for motif confirmation. For instance, methods described in Figure 3E of de Guillen et al. (2015, https://doi.org/10.1371/journal.ppat.1005228) or tools like Foldseek (https://search.foldseek.com/foldmason) could be useful for aligning structures of multiple sequences.

(3) When presenting AlphaFold-generated structures, it is essential to include confidence scores such as pLDDT and PAE. For example, in Figure 1D of Derbyshire and Raffaele (2023, https://doi.org/10.1038/s41467-023-40949-9), the structural representations were colored red due to their high pLDDT scores, emphasizing their reliability.

Author response:

We appreciate the constructive feedback provided by the reviewers and the editorial board. We are delighted by the positive reception of our work and the thoughtful insights shared.

Regarding the validation of our predicted interactions, we are currently conducting yeast two-hybrid (Y2H) assays using a commercially available Arabidopsis thaliana cDNA library to screen for interacting partners of the ANK putative effector PBTT_00818 from Plasmodiophora brassicae. Following this initial screening, we will validate positive interactions through targeted 1-to-1 Y2H assays. In particular, we aim to confirm the AlphaFold Multimer-predicted interaction between PBTT_00818 and MPK3, a key immunity-related kinase in Arabidopsis. 

We are grateful for the reviewers’ thoughtful suggestions regarding clustering visualization, sequence vs. structure-based motif alignments, and structural confidence assessments. We will carefully incorporate these improvements in our planned revisions.

Once again, we thank the editors and reviewers for their rigorous and constructive assessment. We look forward to implementing these refinements and submitting an updated version that further enhances the impact of our study.

eLife Assessment

This important study reports a detailed computational analysis of the CFTR ion channel's permeation mechanism, advancing our understanding of its structure-function relationship. The conclusions are based on extensive molecular dynamics simulations and thorough analysis, but the use of an approximate chloride ion model, known to underestimate key ion-protein interactions, leaves them incomplete without experimental or alternative computational validation. The work will be of interest to biophysicists working on CFTR and cystic fibrosis.

Reviewer #1 (Public review):

Summary:

The goal of this study was to overcome the apparent difficulty in constructing structural models of the open state of the CFTR chloride channel. While several CFTR structural models at near-atomic resolution have been published under a variety of conditions, none of them have demonstrated a pore open across the full dimension of the plasma membrane. Instead, these have routinely been referred to as "near-open" models. In the present study, the authors extended their findings from a prior paper from their group that investigated a series of brief MD simulations, a small number of which exhibited permeation events where chloride ions permeated the pore. This study included massively repeated simulations initiated from these aforementioned Cl permeable conformations. Extensive analysis of the data identified a novel penta-helical structure that comprises the channel pore. This comprehensive study attempted to explain several features of conducting CFTR channels, including single-channel conductance, selectivity, and the mechanisms linking the ATP-induced dimerization of the cytosolic nucleotide-binding domains (NBDs) to the opening of the channel pore (a.k.a., "pore-gating".

Strengths:

The major strength of this study is its comprehensive nature. The approaches applied are cutting-edge and beyond, and are used to explain many different aspects of channel function in CFTR. The strength of evidence is very strong. The paper is extremely well-written, and the arguments are well-supported.

Weaknesses:

The major weakness is that none of the novel conclusions (i.e., those arising solely from this study and not previously published (have been supported by experimental confirmation. That is typical of computational studies such as this.

Reviewer #2 (Public review):

Although recent cryo-EM structures of the CFTR ion channel were reported in a putative open state (ATP-bound, NBD-dimerized), it remains unclear whether these structures explain the conductive properties of the open channel observed in functional experiments. To investigate this, the authors conducted extensive molecular dynamics simulations at different voltages. The simulations are started from snapshots of their prior work, based on the experimental putative open state and including conditions with high negative voltage. Their analysis reveals that the cryo-EM structure represents a near-open metastable state, with most trajectories transitioning to either more closed or more open conformations, leading to the identification of a potential new open state. Permeation rate analysis shows that, unlike the other states, the proposed open state exhibits functional conductive properties of the open channel, although a strong inward rectification, inconsistent with experimental data, is also noted. Further structural analysis and simulations of ATP-unbound closed states offer additional mechanistic insights.

Overall, this work tackles key questions about CFTR: What is the true open conductive state? Does the ATP-bound cryo-EM structure reflect an actual open state? What is the ion permeation mechanism, and what structural changes occur during the closed-to-open transition? Which residues are critical, particularly those linked to diseases like CF? The study, based on a comprehensive set of all-atom molecular dynamics simulations, including a range of physiologically relevant voltages, provides important insights in this regard. It identifies key structural states, permeation pathways, critical residues, and conductance properties that can be directly compared to functional data. Notably, the analysis identifies a new open state of the channel, which, systematic analysis convincingly demonstrates is a conductive conformation of the channel, in line with experimental data at negative voltages. The authors carefully address some of the limitations of their results, exploring and discussing discrepancies with functional experiments, such as inward rectification. The work is also very well written, with a clear and logical presentation of key findings.

The main weakness of this study is that the simulation data rely on the conventional CHARMM36 force field for Cl− ions, which has been shown to significantly underestimate the interaction between Cl− and proteins (J. Chem. Theory Comput. 2021, 17, 6240-6261). For example, the conventional CHARMM36 force field destabilizes the Cl-binding site in CLC-ec1. The latter ion unbinds irreversibly during microseconds-long simulations which is at odds with the experimental binding affinity.

This imbalance in Cl−/protein/water interactions could significantly impact the CFTR simulations, potentially altering state populations and Cl− permeability. Notably, recent work by Levring and Chen (Proc Natl Acad Sci U S A. 2024) identifies a likely Cl− binding site in the bottleneck region of the channel, which contradicts the simulation results showing low occupancy Cl− ions in this region (Fig. 1B and Fig. 6A). This discrepancy may be due to the underestimation of Cl−/protein interactions. Indeed, Orabi et al. have proposed corrections that specifically tune these interactions, including those with aromatic residues, in line with the binding site geometry suggested by Levring and Chen. This imbalance in interactions may also lead to an underestimation of the conductance in the experimental near-open state.<br /> Balanced Cl−/protein interactions could also influence voltage/current relationships, potentially affecting the degree of inward rectification. For example, higher Cl− occupancy in the bottleneck region may stabilize the down state of R334, along with other measured interactions, thereby increasing conductance as the authors have shown.

The experimental evidence reported and discussed by the authors in support of the proposed open state is largely qualitative. For instance, in Figure 4 Supplement 2 there is a significant overlap in the distances and SASA distributions of open and near-open states for the reported residues (are those residues water accessible in the simulations?).

Given the known limitations of the standard CHARMM36 Cl− force field and in the absence of robust experimental validation of the proposed open state, I recommend validating at least part of the results using an independent set of simulations (not started from the previous ones) with an updated Cl− force field. It would be especially important to reassess whether the experimental near-open state is truly metastable and less probable than the new open state, and confirm that the near-open state exhibits negligible conductance.

A minor point worth discussing is whether the observed inward rectification may be influenced by hysteresis or incomplete equilibration, as many simulations were started from prior trajectories at large negative voltages and may not have fully relaxed. For instance, is not uncommon that small structural changes in backbone and sidechains occur in several microseconds (Shaw et al., Science, 2010). That said, discrepancies in current-voltage relationships are not unexpected due to challenges in simulation sampling and force field accuracy (J Gen Physiol 2013 May;141(5):619-32) as the authors stated.

Another minor point to address is the preparation of the simulation setup for the ATP-free structure of the protein. It would be helpful to specify whether any particular controls or steps were taken, given that the structure is based on a relatively low resolution (3.87 Å) model.

Reviewer #3 (Public review):

Background:

Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) is a chloride channel whose dysfunction underlies cystic fibrosis, a life-limiting condition caused by thick, sticky mucus buildup in the lungs and other organs. Despite multiple high-resolution structures of CFTR, these snapshots have all captured the channel in a non-conducting or "closed" conformation - even when the protein was prepared under conditions that should favor channel opening. This discrepancy has posed a key challenge: how can a channel be experimentally observed as closed while physiological tests demonstrate it conducts chloride ions?

Key Findings:

(1) Stable Open Conformation

Through repeated molecular dynamics (MD) simulations of human CFTR in lipid bilayers, researchers observed a reproducible, stable open state. Unlike previous transient openings seen in single-run or short simulations, this conformation remains consistently permeable over extended timescales.

(2) Penta-Helical Arrangement

The authors highlight a "penta-helical" pore-lining arrangement in which five transmembrane helices symmetrically organize to create a clear ion-conduction pathway. This novel configuration resolves the previously puzzling hydrophobic bottleneck found in cryo-EM structures.

(3) Conductance Close to Experimental Values

By analyzing chloride ion flow under near-physiological voltages, they calculate a channel conductance aligning well with electrophysiological measurements. This alignment provides strong support that the observed structure is functionally relevant.

(4) Roles of Key Residues

Several positively charged (cationic) residues in the pore appear crucial for guiding and stabilizing chloride ions. Simultaneously, small kinks in certain helices may act as structural "hinges," allowing or blocking chloride passage.

How to Interpret These Results:

(1) Bridging a Major Gap: The study tackles the mismatch between static "closed" CFTR structures and their known open-channel function. Successfully capturing a stable open state in MD simulations is a significant step toward reconciling what cryo-EM data shows versus what physiological experiments have long told us.

(2) Strength in Multiple Replicas: Running many simulation repeats (rather than relying on a single trajectory) lends credibility. Only if a phenomenon is reproducible across multiple runs can it be considered robust.

(3) Consistency with Mutational Data: Observing that known functional hotspots (e.g., specific charged residues) play a key role in the new pore model further validates these findings.

Important Caveats and Limitations:

(1) Simulation Timescales vs. Biology<br /> Even extended MD (on the microsecond scale) is still much faster, simpler, and more controlled than real cellular processes.

(2) Physiological existence of the penta-helical pore<br /> Although the simulations and results are highly compelling, several factors leave open the possibility of a physiological open conformation differing from the observed penta-helical pore. These factors include ATP hydrolysis, interactions with physiological binding partners, the native membrane environment, and regions not modeled in the CFTR structures, such as the R domain. Most importantly, the transmembrane voltage is very high (500mV).

Bottom Line:

This work delivers a long-awaited, near-physiological view of CFTR's open conformation. It provides a foundational structure against which future experimental and computational studies can be compared. By demonstrating reliable chloride conduction and matching established biophysical data, these simulations bring us closer to understanding - and potentially targeting - CFTR's gating mechanism in health and disease. Readers should applaud the breakthroughs while recognizing that further exploration (including more complex in vitro and in vivo experiments) will still be necessary to capture the full dynamism of CFTR in the living cell environment.

Reviewer #4 (Public review):

Summary:

The structural mechanism of anion permeation through the open CFTR pore has remained unresolved and is subject to ongoing debate. That is because even in CFTR structures obtained under conditions that normally maximally activate the channel (phosphorylation + ATP + non-hydrolytic mutations + potentiator drugs) a bottleneck region in the pore, too narrow to allow passage of hydrated chloride ions, is observed.

The present study uses molecular dynamics (MD) simulations initiated from such "quasi-open" states to address local conformational dynamics of the pore. The authors conclude that the quasi-open structure stably relaxes to a fully open conformation on the sub-microsecond time scale. They provide a detailed analysis of this fully open structure and of the mechanism of chloride permeation. They conclude that two major exit pathways (a central and a peripheral) exist for chloride ions, and that the ions remain near-fully hydrated throughout the pore: chloride-protein interactions displace only 1-2 waters from the first solvation shell. Furthermore, the simulations provide some hints for conformational changes involved in gating.

Strengths:

The findings are interpreted in the context of the large body of published functional studies on CFTR permeation properties, and caveats are adequately discussed.

Weaknesses:

The conclusions on gating would benefit from further discussions. In particular, a fair comparison of the timescale at which channel gating happens, and that of the MD simulations would strengthen the manuscript.

eLife Assessment

Rennert et al. developed a valuable thermodynamic framework to study the force response of branched actin networks from the crucial and unexplored perspective of energetic cost. They used the fact that the entropy production rate must be positive to derive inequalities that set limits on the maximum force produced by branched actin networks, and speculate that the dissipative cost beyond that required to move the load may be necessary to maintain an adaptive steady state. This work is highly innovative, but remains incomplete until the hypotheses of the model are better justified and the conclusions about the dissipative cost of the system are better established.

Reviewer #1 (Public review):

Summary:

This paper investigated the dynamic self-assembly of branched actin networks and the relation between the nonequilibrium features of the dynamics with the thermodynamic cost. The authors constructed a chain model to describe the self-assembly process of a branched actin network, including events like nucleation, polymerization, and capping. The forward and backward transition rates associated with the events allowed them to investigate the entropy production rate of the dynamics. They then used the fact that the entropy production rate has to be greater than zero to derive inequalities that set bounds for the maximum force produced by the branched actin network. The idea is similar to estimating the polymerization force of actin filament via the equation F_{max} = dG/delta, which sets a bound on the maximum force by the thermodynamic potential dG which is the chemical energy associated with ATP hydrolysis and delta is the length increment upon monomer insertion. Furthermore, they speculated the dissipative cost beyond what is necessary to move the load may be necessary to maintain an adaptive steady state.

Strengths:

The authors developed a simple model that is capable of qualitatively reproducing some mechanical phenomena for a branched actin network. The model has captured the essential dynamic elements in the branched actin network and built connections between the maximum load and the adaptation behavior with the energetic cost. It is an interesting study that provides a new perspective to look at the mechanical response of the branched actin network.

Weaknesses:

The text needs to be improved, particularly in the model introduction part. It is unclear to me what happens to the state when the reverse reaction in Figure 2 occurs.

Furthermore, what the authors have done is similar to estimate the polymerization force of actin filaments but in a more complicated scenario. Their conclusion that "dissipative cost in the system beyond what is necessary to move the load may be necessary to maintain an adaptive steady state" is skeptical. The branched actin network is a nonequilibrium system driven by active processes like ATP hydrolysis that converts chemical energy into mechanical work. There has to be a gap between the actual E-C_f curve and that when dissipation rate dot{S} = 0. If the authors want to make the claim, they have to decompose the dissipation into different parts and show that a particular part is associated with adaption. Otherwise, the conclusion about the gap is baseless.

Reviewer #2 (Public review):

Summary:

Rennert et al. developed a thermodynamic framework for the assembly of branched networks to calculate the entropy dissipation associated with this process. They base their model on the simplest possible experimental system consisting of four proteins: actin, Arp2/3, capping protein, and NPF. They decompose the network assembly into a linear model where the order of events (polymerization, capping, and nucleation) is recorded sequentially. Polymerization and capping are sensitive to load and affected by Brownian ratchet effects, while nucleation is not. This simplified model provides an analytical solution that describes the load sensitivity of actin networks and agrees well with experimental data for a given set of transition rates.

Strengths:

(1) These thermodynamic approaches are original and fundamental to our understanding of these non-equilibrium systems.

(2) The fact that the model fits experimental data is encouraging.

Weaknesses:

(1) The possibility of describing branched actin assembly as a Markov process is not well justified.

(2) The choice of parameters controlling the system is open to question. Some parameters are probably completely negligible, while other ignored effects are potentially significant.

(3) The main conclusion of the manuscript, linked to the existence of a dissipation gap, is quite expected. The manuscript would have been more valuable if the authors had been able to decompose dissipation into different components in order to prove that a particular fraction is associated with adaptation.

eLife Assessment

This study presents a potentially fundamental analysis of the original color of a fossil feather from the crest of a 125-million-year-old enantiornithine bird, using sophisticated 3D microscopic and numerical methods to conclude that the feather was iridescent and brightly colored, possibly indicating that this was a male bird that used its crest in sexual displays. At present, the strength of evidence supporting the authors' conclusions is considered incomplete based on methodological incompleteness and questions about taphonomy.

Reviewer #1 (Public review):

Summary:

Li et al describe a novel form of melanosome based iridescence in the crest of an Early Cretaceous enantiornithine avialan bird from the Jehol Group.

Strengths:

Novel set of methods applied to the study of fossil melanosomes.

Weaknesses:

(1) Firstly, several studies have argued that these structures are in fact not a crest, but rather the result of compression. Otherwise, it would seem that a large number of Jehol birds have crests that extend not only along the head but the neck and hindlimb. It is more parsimonious to interpret this as compression as has been demonstrated using actuopaleontology (Foth 2011).<br /> (2) The primitive morphology of the feather with their long and possibly not interlocking barbs also questions the ability of such feathers to be erected without geologic compression.<br /> (3) The feather is not in situ and therefore there is no way to demonstrate unequivocally that it is indeed from the head (it could just as easily be a neck feather)<br /> (4) Melanosome density may be taphonomic; in fact, in an important paper that is notably not cited here (Pan et al. 2019) the authors note dense melanosome packing and attribute it to taphonomy. This paper describes densely packed (taphonomic) melanosomes in non-avian avialans, specifically stating, "Notably, we propose that the very dense arrangement of melanosomes in the fossil feathers (Fig. 2 B, C, and G-I, yellow arrows) does not reflect in-life distribution, but is, rather, a taphonomic response to postmortem or postburial compression" and if this paper was taken into account it seems the conclusions would have to change drastically. If in this case the density is not taphonomic, this needs to be justified explicitly (although clearly these Jehol and Yanliao fossils are heavily compressed).<br /> (5) Color in modern birds is affected by the outer keratin cortex thickness which is not preserved but the authors note the barbs are much thicker (10um) than extant birds; this surely would have affected color so how can the authors be sure about the color in this feather?<br /> (6) Authors describe very strange shapes that are not present in extant birds: "...different from all other known feather melanosomes from both extant and extinct taxa in having some extra hooks and an oblique ellipse shape in cross and longitudinal sections of individual melanosome" but again, how can it be determined that this is not the result of taphonomic distortion?<br /> (7) The authors describe the melanosomes as hexagonally packed but this does not appear to be in fact the case, rather appearing quasi-periodic at best, or random. If the authors could provide some figures to justify this hexagonal interpretation?<br /> (8) One way to address these concerns would be to sample some additional fossil feathers to see if this is unique or rather due to taphonomy<br /> (9) On a side, why are the feet absent in the CT scan image?

Reviewer #2 (Public review):

Summary:

The authors reconstructed the three-dimensional organization of melanosomes in fossilized feathers belonging to a spectacular specimen of a stem avialan from China. The authors then proceed to infer the original coloration and related ecological implications.

Strengths:

I believe the study is well executed and well explained. The methods are appropriate to support the main conclusions. I particularly appreciate how the authors went beyond the simple morphological inference and interrogated the structural implications of melanosome organization in three dimensions. I also appreciate how the authors were upfront with the reliability of their methods, results, and limitations of their study. I believe this will be a landmark study for the inference of coloration in extinct species and how to interrogate its significance in the future.

Weaknesses:

I have a few minor comments.<br /> Introduction: I would suggest the authors move the paragraph on coloration in modern birds (lines 75-97) before line 64, as this is part of the reasoning behind the study. I believe this change would improve the flow of the introduction for the general reader.<br /> Melanosome organization: I was surprised to find little information in the main text regarding this topic. As this is one of the major findings of the study, I would suggest the authors include more information regarding the general geometry/morphology of the single melanosomes and their arrangement in three dimensions.<br /> Keratin: the authors use such a term pretty often in the text, but how is this inference justified in the fossil? Can the authors extend on this? Previous studies suggested the presence of degradation products deriving from keratin, rather than immaculated keratin per se.<br /> Ontogenetic assessment: the authors infer a sub-adult stage for the specimen, but no evidence or discussion is reported in the SI. Can the authors describe and discuss their interpretations?<br /> CT scan data: these data should be made freely available upon publication of the study.

Reviewer #3 (Public review):

Summary:

The paper presents an in-depth analysis of the original colour of a fossil feather from the crest of a 125-million-year-old enantiornithine bird. From its shape and location, it would be predicted that such a feather might well have shown some striking colour and pattern. The authors apply sophisticated microscopic and numerical methods to determine that the feather was iridescent and brightly coloured and possibly indicates this was a male bird that used its crest in sexual displays.

Strengths:

The 3D micro-thin-sectioning techniques and the numerical analyses of light transmission are novel and state-of-the-art. The example chosen is a good one, as a crest feather is likely to have carried complex and vivid colours as a warning or for use in sexual display. The authors correctly warn that without such 3D study feather colours might be given simply as black from regular 2D analysis, and the alignment evidence for iridescence could be missed.

Weaknesses: Trivial.

eLife Assessment

This fundamental manuscript comprehensively examines the roles of nine structural proteins in herpes simplex virus 1 (HSV-1) assembly and nuclear egress. By integrating cryo-light microscopy and soft X-ray tomography, the study presents an innovative approach to investigating viral assembly within cells. The research is thoroughly executed, yielding compelling data that explain previously unknown functions of these structural proteins. This work is of broad interest to virologists, cellular biologists, and structural biologists, offering a robust, contextually rich methodology for studying large protein complex assembly within the cellular environment, serving as an excellent starting point for high-resolution techniques.

Reviewer #1 (Public review):

Summary:

Nahas et al. investigated the roles of herpes simplex virus 1 (HSV-1) structural proteins using correlative cryo-light microscopy and soft X-ray tomography. The authors generated nine viral variants with deletions or mutations in genes encoding structural proteins. They employed a chemical fixation-free approach to study native-like events during viral assembly, enabling observation of a wider field of view compared to cryo-ET. The study effectively combined virology, cell biology, and structural biology to investigate the roles of viral proteins in virus assembly and budding.

Strengths:

(1) The study presented a novel approach to studying viral assembly in cellulo.

(2) The authors generated nine mutant viruses to investigate the roles of essential proteins in nuclear egress and cytoplasmic envelopment.

(3) The use of correlative imaging with cryoSIM and cryoSXT allowed for the study of viral assembly in a near-native state and in 3D.

(4) The study identified the roles of VP16, pUL16, pUL21, pUL34, and pUS3 in nuclear egress.

(5) The authors demonstrated that deletion of VP16, pUL11, gE, pUL51, or gK inhibits cytoplasmic envelopment.

(6) The manuscript is well-written, clearly describing findings, methods, and experimental design.

(7) The figures and data presentation are of good quality.

(8) The study effectively correlated light microscopy and X-ray tomography to follow virus assembly, providing a valuable approach for studying other viruses and cellular events.

(9) The research is a valuable starting point for investigating viral assembly using more sophisticated methods like cryo-ET with FIB-milling.

(10) The study proposes a detailed assembly mechanism and tracks the contributions of studied proteins to the assembly process.

(11) The study includes all necessary controls and tests for the influence of fluorescent proteins.

Weaknesses:

Overall, the manuscript does not have any major weaknesses, just a few minor comments:

(1) The gel quality in Figure 1 is inconsistent for different samples, with some bands not well resolved (e.g., for pUL11, GAPDH, or pUL20).

(2) The manuscript would benefit from a summary figure or table to concisely present the findings for each protein. It is a large body of manuscript, and a summary figure showing the discovered function would be great.

(3) Figure 2 lacks clarity on the type of error bars used (range, standard error, or standard deviation). It says, however, range, and just checking if this is what the authors meant.

(4) The manuscript could be improved by including details on how the plasma membrane boundary was estimated from the saturated gM-mCherry signal. An additional supplementary figure with the data showing the saturation used for the boundary definition would be helpful.

(5) Additional information or supplementary figures on the mask used to filter the YFP signal for Figure 4 would be helpful.

(6) The figure legends could include information about which samples are used for comparison for significance calculations. As the color of the brackets is different from the compared values (dUL34), it would be great to have this information in the figure legend.

(7) In Figure 5B, the association between YFP and mCherry signals is difficult to assess due to the abundance of mCherry signal; single-channel and combined images might improve visualization.

(8) In Figure 6D, staining for tubulin could help identify the cytoskeleton structures involved in the observed virus arrays.

(9) It is unclear in Figure 6D if the microtubule-associated capsids are with the gM envelope or not, as the signal from mCherry is quite weak. It could be made clearer with the split signals to assess the presence of both viral components.

(10) The representation of voxel intensity in Figure 8 is somewhat confusing. Reversion of the voxel intensity representation to align brighter values with higher absorption, which would simplify interpretation.

(11) The visualization in panel I of Figure 8 might benefit from a more divergent colormap to better show the variation in X-ray absorbance.

(12) Figure 9 would be enhanced by images showing the different virus sizes measured for the comparative study, which would help assess the size differences between different assembly stages.

Overall, this is an excellent manuscript and an enjoyable read. It would be interesting to see this approach applied to the study of other viruses, providing valuable insights before progressing to high-resolution methods.

Reviewer #2 (Public review):

Summary:

For centuries, humans have been developing methods to see ever smaller objects, such as cells and their contents. This has included studies of viruses and their interactions with host cells during processes extending from virion structure to the complex interactions between viruses and their host cells: virion entry, virus replication and virion assembly, and release of newly constructed virions. Recent developments have enabled simultaneous application of fluorescence-based detection and intracellular localization of molecules of interest in the context of sub-micron resolution imaging of cellular structures by electron microscopy.

The submission by Nahas et al., extends the state-of-the-art for visualization of important aspects of herpesvirus (HSV-1 in this instance) virion morphogenesis, a complex process that involves virus genome replication, and capsid assembly and filling in the nucleus, transport of the nascent nucleocapsid and some associated tegument proteins through the inner and outer nuclear membranes to the cytoplasm, orderly association of several thousand mostly viral proteins with the capsid to form the virion's tegument, envelopment of the tegumented capsid at a virus-tweaked secretory vesicle or at the plasma membrane, and release of mature virions at the plasma membrane.

In this groundbreaking study, cells infected with HSV-1 mutants that express fluorescently tagged versions of capsid (eYFP-VP26) and tegument (gM-mCherry) proteins were visualized with 3D correlative structured illumination microscopy and X-ray tomography. The maturation and egress pathways thus illuminated were studied further in infections with fluorescently tagged viruses lacking one of nine viral proteins.

Strengths:

This outstanding paper meets the journal's definitions of Landmark, Fundamental, Important, Valuable, and Useful. The work is also Exceptional, Compelling, Convincing, and Solid. The work is a tour de force of classical and state-of-the-art molecular and cellular virology. Beautiful images accompanied by appropriate statistical analyses and excellent figures. The numerous complex issues addressed are explained in a clear and coordinated manner; the sum of what was learned is greater than the sum of the parts. Impacts go well beyond cytomegalovirus and the rest of the herpesviruses, to other viruses and cell biology in general.

Weaknesses:

I have a few suggestions for minor adjustments in the text.

Reviewer #3 (Public review):

Summary:

Kamal L. Nahas et al. demonstrated that pUL16, pUL21, pUL34, VP16, and pUS3 are involved in the egress of the capsids from the nucleous, since mutant viruses ΔpUL16, ΔpUL21, ΔUL34, ΔVP16, and ΔUS3 HSV-1 show nuclear egress attenuation determined by measuring the nuclear:cytoplasmic ratio of the capsids, the dfParental, or the mutants. Then, they showed that gM-mCherry+ endomembrane association and capsid clustering were different in pUL11, pUL51, gE, gK, and VP16 mutants. Furthermore, the 3D view of cytoplasmic budding events suggests an envelopment mechanism where capsid budding into spherical/ellipsoidal vesicles drives the envelopment.

Strengths:

The authors employed both structured illumination microscopy and cellular ultrastructure analysis to examine the same infected cells, using cryo-soft-X-ray tomography to capture images. This combination, set here for the first time, enabled the authors to obtain holistic data regarding a biological process, as a viral assembly. Using this approach, the researchers studied various stages of HSV-1 assembly. For this, they constructed a dual-fluorescently labelled recombinant virus, consisting of eYFP-tagged capsids and mCherry-tagged envelopes, allowing for the independent identification of both unenveloped and enveloped particles. They then constructed nine mutants, each targeting a single viral protein known to be involved in nuclear egress and envelopment in the cytoplasm, using this dual-fluorescent as the parental one. The experimental setting, both the microscopic and the virological, is robust and well-controlled. The manuscript is well-written, and the data generated is robust and consistent with previous observations made in the field.

Weaknesses:

It would be helpful to find out what role the targeted proteins play in nuclear egress or envelopment acquisition in a different orthoherpesvirus, like HSV-2. This would confirm the suitability of the technical approach set and would also act as a way to validate their mechanism at least in one additional herpesvirus beyond HSV-1. So, using the current manuscript as a starting point and for future studies, it would be advisable to focus on the protein functions of other viruses and compare them.

eLife Assessment

This study provides important insights into the regulation of type-I interferon signaling and anti-tumor immunity, demonstrating that ORMDL3 promotes RIG-I degradation to suppress immune responses. The evidence is convincing, with well-executed mechanistic experiments and in vivo validation in syngeneic tumor models. These findings have significant implications for cancer immunotherapy, highlighting ORMDL3 as a potential therapeutic target.

Reviewer #2 (Public review):

Summary:

The authors identified ORMDL3 as a negative regulator of the RLR pathway and anti-tumor immunity. Mechanistically, ORMDL3 interacts with MAVS and further promotes RIG-I for proteasome degradation. In addition, the deubiquitinating enzyme USP10 stabilizes RIG-I and ORMDL3 disturbs this process. Moreover, in subcutaneous syngeneic tumor models in C57BL/6 mice, they showed that inhibition of ORMDL3 enhances anti-tumor efficacy by augmenting the proportion of cytotoxic CD8-positive T cells and IFN production in the tumor microenvironment (TME).

Strengths:

The paper has a clearly arranged structure and the English is easy to understand. It is well written. The results clearly support the conclusion.

Comments on revisions:

All questions have been answered.

eLife Assessment

This study addresses an important and longstanding question regarding the molecular mechanism of protein misfolding in Ig light chain (LC) amyloidosis (AL), a life-threatening condition. By combining advanced techniques, including small-angle X-ray scattering, molecular dynamics simulations, and hydrogen-deuterium exchange mass spectrometry, the authors provide convincing evidence that the "H state" distinguishes amyloidogenic from non-amyloidogenic LCs. These findings not only offer novel insights into LC structural dynamics but also hold promise for guiding therapeutic strategies in amyloidosis and will be of particular interest to structural biologists, biophysicists, and many others working on amyloid diseases.

Reviewer #1 (Public review):

The study investigates light chains (LCs) using three distinct approaches, with a focus on identifying a conformational fingerprint to differentiate amyloidogenic light chains from multiple myeloma light chains. The study's major contribution is the identification of a low-populated "H state," which the authors propose as a unique marker for AL-LCs. While this finding is promising, the review highlights several strengths and weaknesses. Strengths include the valuable contribution of identifying the H state and the use of multiple approaches, which provide a comprehensive understanding of LC structural dynamics. Weaknesses include a lack of physical insights explaining the changes.

Reviewer #2 (Public review):

Summary:

This well-written manuscript addresses an important but recalcitrant problem - molecular mechanism of protein misfolding in Ig light chain (LC) amyloidosis (AL), a major life-threatening form of systemic human amyloidosis. The authors use expertly recorded and analyzed small-angle X-ray scattering (SAXS) data as a restraint for molecular dynamics simulations (called M&M). Six patient-based LC proteins are explored, including four AL and two non-AL. The authors report a partially populated "H-state" determined computationally, wherein the two domains in an LC molecule acquire a straight rather than bent conformation, with an extended interdomain linker; this H-state distinguishes AL from non-AL LCs. H-D exchange mass spectrometry is used to support this conclusion. This is a novel and interesting finding with potentially important translational implications.

Strengths:

Expertly recorded and analyzed SAXS data combined with clever M&M simulations lead to a novel and interesting conclusion, which is supported by limited H-D exchange data.<br /> Stabilization of the CL-CL interface is a good idea that may help protect a subset of AL LCs from misfolding in amyloid.

Computational M&M evidence is convincing and is supported by SAXS data, which are used as restraints for simulations. Although Kratky plots reported in the main MS Fig. 1 show significant differences between the data and the structural model for only one AL protein, AL-55, H-state is also inferred for other AL proteins.

Apparent limitations:

HDX MS results show that residues 35-50 from VL-VL and VL-CL dimerization interface are less protected in AL vs. non-AL proteins, which is consistent with the H-state. However, the small number of proteins yielding useful HDX data (three AL and one non-AL) suggests that this conclusion should be treated with caution. It is unclear whether the conformational heterogeneity depicted in M&M simulations is consistent with HDX results, and whether prior HDX studies of AL and MM LCs are consistent with the conclusions that a particular domain-domain interface is weakened in AL vs. non-AL LCs. The butterfly plots in Fig. 5 could benefit from the X-axis labeling with the peptide fragments.

Reviewer #3 (Public review):

Summary:

This study identifies confirmational fingerprints of amylodogenic light chains, that set them apart from the non-amylodogenic ones.

Strengths:

The research employs a comprehensive combination of structural and dynamic analysis techniques, providing evidence that conformational dynamics at VL-CL interface and structural expansion are distinguished features of amylodogenic LCs.

Weaknesses:

The sample size is limited, which may affect the generalizability of the findings. Additionally, the study could benefit from deeper analysis of specific mutations driving this unique conformation to further strengthen therapeutic relevance.

Furthermore. p-value (statistical significance) of Rg difference should be computer. Finally, significance of mutations (SHM?) at the interface, such as A40G should be compared with previous observations. (Garofalo et al., 2021)

Author response:

The following is the authors’ response to the original reviews.

eLife Assessment

This important study identifies the "H-state" as a potential conformational marker distinguishing amyloidogenic from non-amyloidogenic light chains, addressing a critical problem in protein misfolding and amyloidosis. By combining advanced techniques such as small-angle X-ray scattering, molecular dynamics simulations, and H-D exchange mass spectrometry, the authors provide convincing evidence for their novel findings. However, incomplete experimental descriptions, limitations in SAXS data interpretation, and the way HDX MS data is presented aHect the strength and generalizability of the conclusions. Strengthening these aspects would enhance the impact of this work for researchers in amyloidosis and protein misfolding.

We thank eLife editors and reviewers for their constructive feedback. The manuscript has been improved to provide a more complete description of the experiments and to strengthen the interpretation and presentation of all data. Updated Figures (Figure 2 and Figure 5) and a new Table (Table 2) in the main text provide a more complete and clearer comparison of the SAXS data with MD simulations as well as a clearer representation of the HDX MS data. Additional figures have been added in SI. The text has been extended accordingly and complete materials and methods are now included in the main text. Abstract, introduction and discussion have been revised to improve the overall readability of the manuscript.

Public Reviews:

Reviewer #1 (Public review):

The study investigates light chains (LCs) using three distinct approaches, with a focus on identifying a conformational fingerprint to diHerentiate amyloidogenic light chains from multiple myeloma light chains. The study's major contribution is identifying a low-populated "H state," which the authors propose as a unique marker for AL-LCs. While this finding is promising, the review highlights several strengths and weaknesses. Strengths include the valuable contribution of identifying the H state and using multiple approaches, which provide a comprehensive understanding of LC structural dynamics. However, the study suHers from weaknesses, particularly in interpreting SAXS data, lack of clarity in presentation, and methodological inconsistencies. Critical concerns include high error margins between SAXS profiles and MD fits, unclear validation of oligomeric species in SAXS measurements, and insuHicient quantitative cross-validation between experimental (HDX) and computational data (MD). This reviewer calls for major revisions including clearer definitions, improved methodology, and additional validation, to strengthen the conclusions.

We thank the reviewer for the supportive comments, in the revised version of the manuscript we have focused on improving the clarity and completeness of our work. We are sorry for example to not have made previously clear enough that the comparison of SAXS with MD simulation was not that shown in the main text in Figure 1 and Table 1 (this is the comparison with single structures) but that reported in the SI (previously Figure S1 and Table S2, showing very good fits). These data have been moved in the main text in the reworked Figure 2 and new Table 2.  We have also improved the presentation of the HDX MS data in Figure 5 and in the text adding also additional analysis in SI. Materials and methods are now completely moved in the main text. We generally revised the manuscript for clarity.

Reviewer #2 (Public review):

Summary:

This well-written manuscript addresses an important but recalcitrant problem - the molecular mechanism of protein misfolding in Ig light chain (LC) amyloidosis (AL), a major life-threatening form of systemic human amyloidosis. The authors use expertly recorded and analyzed smallangle X-ray scattering (SAXS) data as a restraint for molecular dynamics simulations (called M&M) and to explore six patient-based LC proteins. The authors report that a highly populated "H-state" determined computationally, wherein the two domains in an LC molecule acquire a straight rather than bent conformation, is what distinguishes AL from non-AL LCs. They then use H-D exchange mass spectrometry to verify this conclusion. If confirmed, this is a novel and interesting finding with potentially important translational implications.

We thank the reviewer for the supportive comments.

Strengths:

Expertly recorded and analyzed SAXS data combined with clever M&M simulations lead to a novel and interesting conclusion. Regardless of whether or not the CL-CL domain interface is destabilized in AL LCs explored in this (Figure 6) and other studies, stabilization of this interface is an excellent idea that may help protect at least a subset of AL LCs from misfolding in amyloid. This idea increases the potential impact of this interesting study.

We thank the reviewer for the supportive comments.

Weaknesses:

The HDX analysis could be strengthened.

We have extended the analysis and improved the presentation of the HDX data. Figure 5 has been reworked, text has been improved accordingly and additional analysis have been reported in SI.

Reviewer #3 (Public review):

Summary:

This study identifies conformational fingerprints of amyloidogenic light chains, that set them apart from the non-amyloidogenic ones.

We thank the reviewer for the supportive comments.

Strengths:

The research employs a comprehensive combination of structural and dynamic analysis techniques, providing evidence that conformational dynamics at the VL-CL interface and structural expansion are distinguished features of amyloidogenic LCs.

We thank the reviewer for the supportive comments.

Weaknesses:

The sample size is limited, which may aHect the generalizability of the findings. Additionally, the study could benefit from deeper analysis of specific mutations driving this unique conformation to further strengthen therapeutic relevance.

We agree, we tried to maximise the size of the sample and this was the best we could do. With respect to the analysis of the mutations, while we tried to discuss some of them also in view of previous works, because our set covers multiple germlines instead than focusing on a single one, this limit our ability to discuss single point mutations systematically, at the same time the discussion of single points mutations has been the focus of many recent works, while our approach provide a diNerent point of view.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

This study provides an investigation of light chains (LCs) using three distinct approaches, focusing primarily on identifying a conformational fingerprint to distinguish amyloidogenic light chains (AL-LCs) from multiple myeloma light chains (MM-LCs). The authors propose that the presence of a low-populated "H state," characterized by an extended quaternary structure and a perturbed CL-CL interface, is unique to AL-LCs. This finding is validated through hydrogendeuterium exchange mass spectrometry (HDX-MS). The study makes a valuable contribution to understanding the structural dynamics of light chains, particularly with the identification of the H state in AL-LCs. However, significant concerns regarding the interpretation of the SAXS data, clarity in presentation, and methodological rigor must be addressed. I recommend major revisions and resubmission of the work.

Major concerns:

(1) A critical concern is how the authors ensure that the SAXS profiles represent only dimeric species, given the high propensity of LCs to aggregate. If higher-order aggregates or monomers were present, this would significantly impact the SAXS data and SAXS-MD integration. Some measurements are bulk SAXS, while others are SEC-SAXS, making the study questionable. The authors need to clarify how only dimeric species were measured for the SEC-SAXS analysis, and all assessments of the dimeric state should be shown in the SI. Additionally, complementary techniques such as DLS or SEC-MALS should be used to verify the oligomeric state of the samples. Without this validation, the SAXS profiles may not be reliable.

We added SEC-MALS and SEC-SAXS data in the SI (Figures S20 and S21) as well the SAXS curves shown in log-log plot (Figure S1) that display a flat trend at low q that exclude aggregation. SAXS is very sensitive to oligomers and aggregates and our data do not indicate the presence of those species. When we had indication of possible aggregation in the sample we used SEC-SAXS.

(2) A major problem with the paper is that the claim of the "H state," which is the novelty of the study and serves as a marker of aggregation, is derived from samples where the error between the SAXS profiles and MD fits is extremely high. This casts doubt on whether the structure is indeed resolved by MD. The main conclusion of the paper is derived from weak consistency between experiment and simulation. In AL55, the error between experiment and simulation is greater than 5; for H7, it is higher than 2.8. The residuals show significant error at mid-q values, suggesting that long-range distance correlations (20-10 Å, CL, VL positioning) are not consistent between simulation and experiment. Furthermore, the FES plots of two independent replicas show deviation in the existence of the H state. One shows a minimum in that region, while the other does not. So, how robust is this conclusion? What is the chi-squared value if each replica is used independently? A separate experimental cross-validation is necessary to claim the existence of the H state.

We apologise for the misunderstanding underlying this reviewer comment. The poor agreement mentioned is not between the SAXS and MD simulations, but with the individual structures, and this disagreement led us to perform MD simulations that are in much better agreement with the data (previously Fig. S1 and Table S2). To avoid this misunderstanding, which would indeed weaken our work, we have now moved both the figure and the table in the main text to the updated Figure 2 and the new Table 2.

Regarding the robustness of the sampling, we believe that Table 3 (previously Table 2) clearly shows the statistical convergence of the data, diNerences in the presentation of the free energy are purely interpolation issues. The chi-squares of each replicate are reported in Table 2 (previously Table S2).

(3) There is insuHicient discussion about SAXS computations from MD trajectories. The accuracy of these calculations is crucial to deriving the existing conclusions, and the study's reliance on the PLUMED plugin, which is known to give inaccurate results for SAXS computations, raises concerns. How the solvent is treated in the SAXS computations needs to be explained. Alternative methods like WAXSiS or Crysol should be explored to check whether the SAXS profiles derived from the MD trajectory are consistent across other SAXS computation methods for the major conformers of the proteins.

We have now clarified that while the SAXS calculation to perform Metainference MD were done using PLUMED (that to our knowledge is as accurate as crysol) SAXS curves used for analysis were calculated using crysol.

(4) The HDX and MD results do not seem to correlate well, and there is a disconnect between Figure 2 (SAXS profiles) and Figure 5 (HDX structural interpretation). The authors should quantitatively assess residue-level dynamics by comparing HDX signals with MD-derived HDX signals for each protein. This would provide a cross-validation between the experimental and computational data.

In our opinion our SAXS, MD and HDX MS data provide a consistent picture. Our HDX-MS do not provide per residue data, making a quantitative comparison out of scope. RMSF data do not necessarily need to correlate with the deuterium uptake.

(5) MD simulations are only used to refine the structure of AlphaFold predictions, but the trajectories could help explain why these structures diHer, what stabilizes the dimer, or what leads to the conformational transition of the H state. A lack of analysis regarding the physical mechanism behind these structural changes is a weakness of the study. The authors should dedicate more eHort to analyzing their data and provide physical insights into why these changes are observed.

Our aim was to identify a property that could discriminate between AL and MM LCs. We used MD simulations, not to refine structures, but to explore the conformational dynamics of LCs (starting from either X-ray structures, homology or AlphaFold models), because SAXS data suggested that conformational dynamics could discriminate between AL- and MM-LCs. Simulations allowed us to propose a hypothesis, which we tested by HDX MS. While more insight is always welcome, we believe that we have achieved our goal for now. In the discussion, we present additional analysis of the simulations to connect with previous literature, we agree that more analysis can be done, and also for this reason, all our data are publicly available.

Minor concerns

(6) The abstract leans heavily on describing the problem and methods but lacks a clear presentation of key results. Providing a concise summary of the main findings (e.g., the identification of the H state) would better balance the abstract.

We agree with the reviewer and we rewrote the abstract.

(7) In the abstract, the term "experimental structure" is used ambiguously. Since SAXS also provides an experimental structure, it is unclear what the authors are referring to. This should be clarified.

We agree with the reviewer and we rewrote the abstract.

(8) Abbreviations such as VL (variable domain) and CL (constant domain) are not defined, making it harder for readers unfamiliar with the field to follow. Abbreviations should be defined when first mentioned.

We agree with the reviewer and we rewrote the abstract.

(9) The introduction provides a good general context but fails to explicitly define the knowledge gap. Specifically, the structural and dynamic determinants of LC amyloidogenicity are not well established, and this study could be framed as addressing that gap.

We thank the reviewer and we agree this could be better framed, we improved the introduction accordingly.

(10) The introduction does not present the novel discovery of the H state early enough. The unique contribution of identifying this state as a marker for AL-LCs should be mentioned upfront to guide the reader through the significance of the study.

We thank the reviewer and we have now made more explicit what we found.

(11) The therapeutic implications of this research should be highlighted more clearly in the discussion. Examples of how these findings could be utilized in drug design or therapeutic approaches would enhance the study's impact.

We thank the reviewer, but while we think that the H-state could be targeted for drug design, since we do not have data yet we do not want to stress this point more than what we are already doing.

(12) There is an overwhelming use of abbreviations such as H3, H7, H18, M7, and M10 without proper introduction. This makes it diHicult for readers to follow the results, and the average reader may become lost in the details. An introductory figure summarizing the sequences under study, along with a schematic of the dimeric structure defining VL and CL domains, would significantly aid comprehension.

We agree and we tried to better introduce the systems and simplify the language without adding a figure that we think would be redundant.

(13) In Figure 1, add labels to each SAXS curve to indicate which protein they correspond to. Also, what does online SEC-SAXS mean?

Done

(14) The caption of Figure 3 is unclear, particularly with abbreviations like Lb, Ls, G, and H, which are not mentioned in the captions. The authors should define these terms for clarity.

Done

(15) The study claims that the dominant structure of the dimer changes between diHerent LCs. However, Figure 5 shows identical structures for all proteins, raising questions about the consistency between the SAXS and HDX data. This inconsistency is a general problem between the MD and HDX sections, where cross-communication and comparisons are not properly addressed.

We do not claim that the dominant structure of the dimer changes between diNerent LCs, this would also be in contradiction with current literature. We claim a diNerence in a low-populated state. From this point of view using always the same structure is consistent and should simplify the representation of the results. We agree that the manuscript may be not always easy to follow and we thank the reviewer in helping us improving it.

(16) The authors show I(q) vs q and residuals for each protein. The Kratky plots are not suHicient to compare the SAXS computations with the measured profile.

Showing Kratky and residuals is a standard and complementary way to present and compare SAXS data to structures. Chi-square values are also reported. Log-log plots have been added to SI in response to previous comments.

(17) The authors need to explain how they estimate the Rg values (from simulation or SAXS profiles). If they are using simulations, they should compute the Rg values from the simulations for comparison.

Rg values reported in Table 1 are derived from SAXS. Rg from simulations have been added in Table 2.

(18) The evolution of the sampling is unclear. The authors need to show the initial starting conformation in each case and the most likely conformation after M&M in the SI, to demonstrate that their approach indeed caused changes in the initial predictions.

Our approach is not structure refinement and as such the proposed analysis would be misleading. Metainference is meant to generate a statistical ensemble representing the equilibrium conformations that as whole reproduce the data. DiNerences (or not) between initial and selected configurations will not be particularly informative in this context.

(19) The authors should also provide a running average of chi-squared values over time to demonstrate that the conformational ensemble converged toward the SAXS profile.

Our simulations are not driven to improve the agreement with SAXS over time, this is not structure refinement. Metainference is meant to generate a statistical ensemble representing the equilibrium conformations that as whole reproduce the data. The suggested analysis would be a misinterpretation of our simulations. The comparison with SAXS is provided in Figure 2 and Table 2 as mentioned above.

(20) The aggregate simulation time of 120 microseconds is misleading, as each replica was only run for 2-3 microseconds. This should be clarified.

The number reported in the text is accurate and represent the aggregated sampling. The number of replicas for each metainference simulation and their length is reported in Table 2 now moved for clarity from the SI to main text.

(21) It is not clear how the replicas were weighted to compute the SAXS profiles and FES. There are two independent runs in each case, and each run has about 30 replicas. How these replicas are weighted needs to be discussed in the SI.

Done

(22) The methods section is unevenly distributed, with detailed explanations of LC production and purification, while other key methodologies like SAXS+MD integration and HDX are not even mentioned in the main text (they are in the Supporting Information). The authors should provide a brief overview of all methodologies in the main text or move everything to the SI for consistency.

We agree with the reviewer, all methods are now in main text. 

Reviewer #2 (Recommendations for the authors):

(1) Computational M&M evidence is strong (Figure 3) and is supported by SAXS (used as restraints). However, Kratky plots reported in the main MS Figure 1 show significant diHerences between the data and the structural model only for one protein, AL-55. It is hard for the general reader to see how these SAXS data support a clear diHerence between AL and non-AL proteins. If possible, please strengthen the evidence; if not, soften the conclusions.

We thank the reviewer for the comments. The chi-square (Table 1) and the residuals (Figure 1) are a strong indication of the diNerence. To strengthen the evidence, following also the comment from reviewer 3 we calculated the p-value (<10^-5) on the significance of the radius of gyration to discriminate AL and MM LCs. We agree that SAXS alone was not enough and this is indeed what prompted us to perform MD simulations.

(2) HDX MS results are cursory and not very convincing as presented. The butterfly plots in Figure 5 are too small to read and are unlabeled so it is unclear which protein is which.  

Figure 5 has been reworked for readability. More data have been added in SI. 

(3) What labeling time was selected to construct these plots and why?

The deuterium uptakes at 30 min HDX time showed the most pronounced diNerences between diNerent proteins, which were chosen to illustrate the key structural features in the main figure panel (Figure 5).

How diHerent are the results at other labeling times? Showing uptake curves (with errors) for more than just two peptides in the supplement Figure S12 might be helpful. 

We found a continuous increase in deuterium uptake as we increased the exchange time from 0.5 to 240 min, which reached saturation at 120 min. Therefore, the exchange follows the same pattern at all time points. Butterfly plots at diNerent HDX times of 0.5 to 240 min are shown in gradient of light blue to dark blue which clearly shows the pattern of deuterium uptake at increasing incubation times (Figure 5). The HDX uptake kinetics of selected peptides with corresponding error bars are shown in Figure S12.

How redundant are the data, i.e. how good is the peptide coverage/resolution in key regions at the domain-domain interface that the authors deem important? Mapping the maximal deuterium uptake on the structures in Figure 5 is not very helpful. Perhaps mapping the whole range of uptake using a gradient color scheme would be more informative.

Overall coverage and redundancy for all four proteins are> 90% and > 4.0, respectively, with an average error margin in fractional uptake among all peptides is 0.04-0.05 Da, which suggests that our data is reliable (Table S3). We modified the main panel figures showing the gradient of deuterium uptake in blue-white-red for 0 to 30% of deuterium uptake on the chain A of the dimeric LCs.

(3) Is the conformational heterogeneity depicted in M&M simulations consistent with HDX results? The authors may want to address this by looking at the EX1/EX2 exchange kinetics for AL vs. non-AL proteins. Do AL proteins show more EX1?

No, we don’t see any EX1 exchange kinetics in our analysis. This is compatible with the prediction of the H-state that is a native like state and not an unfolded/partially folded state. 

(4) Perhaps the main conclusion could be softened given the small number of proteins (six), esp. since only four (3 AL and 1 non-AL) could be explored by HDX. Are other HDX MS data of AL LCs from the same Lambda6 family (e.g. PMID: 34678302) consistent with the conclusions that a particular domain-domain interface is weakened in AL vs. non-AL LCs?

We thank the reviewer for this suggestions. A diNerence in HDX MS data is indeed visible between AL and MM proteins for peptide 33-47 in the suggested paper (Figures 4, S5 and S8). The diNerence is reduced by the mutation identified in the paper as driving the aggregation in that specific case. We now mention this in the discussion.

(5) Please clarify if the H* state is the same for a covalent vs. non-covalent LC dimer.

We do not know because our data are only for covalent dimers. But, interestingly, the state is very similar to what was observed for a model kappa light-chain in Weber, et al., we have better highlighted this point in the discussion.

(6) Please try and better explain why a smaller distance between CL domains in H7 protein and a larger distance in other AL proteins both promote protein misfolding.

We do not have elements to discuss this point in more detail.

(7) Please comment on the Kratky plots data vs. model agreement (see comments above).

Done.

(8) Please find a better way to display, describe, and interpret the HD exchange MS data.

We have generated new main text (new Figure 5) and SI figures that we think allow the reader to better appreciated our observations. Corresponding results sections have been also improved.

Minor points:

(9) Is the population of the H-state with perturbed CL-CL domain interface, which was obtained in M&M simulations, suHicient to be observable by HDX MS?

While populations alone are not enough to determine what is observable by HDX MS, a 10% population correspond roughly to 6 kJ/mol of ΔG and is compatible with EX2 kinetics. Previous works suggested that HDX-MS data should be sensitive to subpopulations of the order of 10%, (https://doi.org/10.1016/j.bpj.2020.02.005, https://doi.org/10.1021/jacs.2c06148)

(10) Typically, an excited intermediate in protein unfolding is a monomer, while here it is an LC dimer. Is this unusual?

This is a good point, we think that intermediates have mostly been studied on monomeric proteins because these are more commonly used as model systems, but we do not feel like discussing this point.

(11) Low deuterium uptake is consistent with a rigid structure but may also reflect buried structure and/or structure that moves on a time scale greater than the labeling time.

We agree.

Reviewer #3 (Recommendations for the authors):

(1) The p-value (statistical significance) of Rg diHerence should be computed.

We thank the reviewer for the suggestion, we calculated the p-value that resulted quite significant.

(2) The significance of mutations (SHM?) at the interface, such as A40G should be compared with previous observations. (Garrofalo et al., 2021).

We thank the reviewer for the suggestion, a sentence has been added in the discussion.

eLife Assessment

The authors present three transgenic models carrying three representative exon deletions of the dystrophin gene. The findings presented are valuable to the field of muscle diseases, particularly muscular dystrophies. The evidence provided in the manuscript is convincing, with rigorous biochemical assays and state-of-the-art microscopy methods.

Reviewer #2 (Public review):

Miyazaki et al. established three distinct BMD mouse models by deleting different exon regions of the dystrophin gene, observed in human BMD. The authors demonstrated that these models exhibit pathophysiological changes, including variations in body weight, muscle force, muscle degeneration, and levels of fibrosis, alongside underlying molecular alterations such as changes in dystrophin and nNOS levels. Notably, these molecular and pathological changes progress at different rates depending on the specific exon deletions in dystrophin gene. Additionally, the authors conducted extensive fiber typing, revealing a site-specific decline in type IIa fibers in BMD mice, which they suggest may be due to muscle degeneration and reduced capillary formation around these fibers.

Strengths:

The manuscript introduces three novel BMD mouse models with different dystrophin exon deletions, each demonstrating varying rates of disease progression similar to the human BMD phenotype. The authors also conducted extensive fiber typing across different muscles and regions within the muscles, effectively highlighting a site-specific decline in type IIa muscle fibers in BMD mice.

Comments on revisions:

The authors did an excellent job addressing all or most of the concerns I raised in my previous review and have incorporated the necessary changes into the manuscript.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews:

Reviewer #1 (Public review):

Summary:

In this article the authors described mouse models presenting with backer muscular dystrophy, they created three transgenic models carrying three representative exon deletions: ex45-48 del., ex45-47 19 del., and ex45-49 del. This article is well written but needs improvement in some points.

Strengths:

This article is well written. The evidence supporting the authors' claims is robust, though further implementation is necessary. The experiments conducted align with the current state-of-the-art methodologies.

Weaknesses:

This article does not analyze atrophy in the various mouse models. Implementing this point would improve the impact of the work

We thank the reviewer for their constructive suggestions and comments on this work. Muscle hypertrophy is shown with growth in dystrophin-deficient skeletal muscle in mdx mice; thus, we did not pay attention to the factors associated with muscle atrophy in BMD mice. As the reviewer suggested, the examination of the association between type IIa fiber reduction and muscle atrophy is important, and the result is considered to be helpful in resolving the cause of type IIa fiber reduction in BMD mice.

In response, we reviewed the following.

(1) The cross-sectional areas (CSAs) of muscles. We confirmed that the CSAs in BMD and mdx mice were rather high at 3 months, in accordance with muscle hypertrophy, compared with those of WT mice. The data is presented in Fig. 4–figure supplement 1B.

(2) The mRNA expression levels of Murf1 and atrogin-1. We confirmed that these muscle atrophy inducing factors did not differ among WT, BMD, and mdx mice. The data is presented in Fig. 4–figure supplements 1C and 1D.

Reviewer #2 (Public review):

Summary:

Miyazaki et al. established three distinct BMD mouse models by deleting different exon regions of the dystrophin gene, observed in human BMD. The authors demonstrated that these models exhibit pathophysiological changes, including variations in body weight, muscle force, muscle degeneration, and levels of fibrosis, alongside underlying molecular alterations such as changes in dystrophin and nNOS levels. Notably, these molecular and pathological changes progress at different rates depending on the specific exon deletions in the dystrophin gene. Additionally, the authors conducted extensive fiber typing, revealing a site-specific decline in type IIa fibers in BMD mice, which they suggest may be due to muscle degeneration and reduced capillary formation around these fibers.

Strengths:

The manuscript introduces three novel BMD mouse models with different dystrophin exon deletions, each demonstrating varying rates of disease progression similar to the human BMD phenotype. The authors also conducted extensive fiber typing across different muscles and regions within the muscles, effectively highlighting a site-specific decline in type IIa muscle fibers in BMD mice.

Weaknesses:

The authors have inadequate experiments to support their hypothesis that the decay of type IIa muscle fibers is likely due to muscle degeneration and reduced capillary formation. Further investigation into capillary density and histopathological changes across different muscle fibers is needed, which could clarify the mechanisms behind these observations.

We thank the reviewer for these positive comments and the very important suggestion about type IIa fiber reduction and capillary change around muscle fibers in BMD mice. From the results of the cardiotoxin-induced muscle degeneration and regeneration model, type IIa and IIx fibers showed delayed recovery compared with that of type-IIb fibers. However, this delayed recovery of type IIa and IIx could not explain the cause of the selective muscle fiber reduction limited to type IIa fibers in BMD mice. Therefore, we considered vascular dysfunction as the reason for the selective type IIa fiber reduction, and we found morphological capillary changes from a “ring pattern” to a “dot pattern” around type IIa fibers in BMD mice. However, the association between selective type IIa fiber reduction and the capillary change around muscle fibers in BMD mice remains unclear due to the lack of information about capillaries around type IIx and IIb fibers. The reviewer pointed out this insufficient evaluation of capillaries around other muscle fibers (except for type IIa fibers), and this suggestion is very helpful for explaining the association between selective type IIa fiber reduction and vascular dysfunction in BMD mice.

In response, we reviewed the following.

(1) The capillary formation around type IIx, IIb, and I fibers, in addition to that around type IIa fibers. We found that capillaries contacting around type IIx, IIb, and I fibers were poor in WT mice compared with that around type IIa fibers, with ‘incomplete ring-patterns’ around type IIx fibers, and ‘dot-patterns’ around type IIb and I fibers in WT mice. Morphological capillary changes around muscle fibers from WT to d45-49 and mdx mice were ‘incomplete dot-pattern’ to ‘dot-pattern’ around type IIx fibers, and ‘dot-pattern’ to ‘dot-pattern’ around type IIb and I fibers. This was in contrast to those around type IIa fibers: remarkable ‘ring-pattern’ to ‘dot-pattern’. These data are presented in Fig. 6B.

(2) The endothelial area in contact with type IIx, IIb, and I fibers, and additionally that in contact with type IIa fibers. The endothelial area in contact with both type IIa and IIx fibers was less in d45-49 and mdx mice than in WT mice, but the reduction was larger around type IIa fibers than around type IIx fibers, reflecting the difference between the ‘ring-pattern’ around the former and the ‘incomplete ring-pattern’ around the latter in WT mice. These data are presented in Fig. 6C.

(3) Transversely interconnected branches and capillary loops, using longitudinal muscle sections. We confirmed that there were fewer interconnected capillaries in BMD and mdx mice than in WT mice. These data are presented in Fig. 6E.

(4) The mRNA expression levels of neuronal nitric oxide synthase (nNOS). We confirmed that nNOS protein expression levels were decreased in BMD and mdx mice in spite of adequate levels of nNOS mRNA expression. The data on nNOS mRNA expression levels is presented in Fig. 3–figure supplement 1C.

(5) We added a sentence in the Abstract about the potential utility of BMD mice in developing vascular targeted therapies.

Recommendation for the authors:

Reviewer #1 (Recommendation for the authors):

Abstract:

Abstract: more emphasis should be on the pathological implications of Becker muscular dystrophy (BMD). Furthermore, should be emphasized the findings made in this article and the conclusions. Abbreviations such as DMD and MDX should be written in full and only then with the acronym.

We appreciate the reviewers’ comments, and we apologize for the confusion over abbreviations. DMD is the gene name encoding dystrophin, and mdx is the strain name of mouse lacking dystrophin.

In the Abstract and the Figure legends we changed:

(1) DMD to DMD;

(2) mdx mice to mdx mice.

Results:

Line 95: in this line, authors evaluated serum creatinine kinase (CK) levels at 1, 3, 6 and 12 months in WT mice and mdx mice. Why did you decide to study it? This part should be described in more detail. Serum CK is one of the main markers of muscle necrosis; therefore, I would report this data alongside the description of the muscle histology and necrotic fibers.

We thank the reviewers for the important remarks. In this study, serum creatine kinase (CK) levels were two-fold to four-fold higher in BMD mice than in WT mice, but its rate of increase was less than that of mdx mice. We consider that the lesser changes in serum CK levels in BMD mice may be due to the smaller area of muscle degeneration because of focal and uneven muscle degeneration compared with that in mdx mice, which showed diffuse muscle degeneration.

In response, we have moved the description of serum CK levels in the Results, from the section about the establishment of BMD mice to the section about site-specific muscle degeneration in BMD mice.

In addition, we added a description in the Discussion about the possible association between the lesser changes in serum CK levels in BMD mice and its uneven distribution of muscle degeneration.

Line 192-202: In these lines, authors observed a decrease in type IIa fibers after 3 months in BMD mice. I suggest evaluating also atrophy through evaluating cross-sectional areas (CSA) and expression of Murf1 and Atrogin1

We thank the reviewer for the point about the association between type IIa fiber reduction and muscle atrophy. We evaluated the CSAs and the mRNA expression levels of Murf1 and atrogin-1. We confirmed that the CSAs in BMD and mdx mice were rather high at 3 months, in accordance with muscle hypertrophy, compared with those of WT mice, and that Murf1 and atrogin-1 mRNA expression levels did not differ among WT, BMD, and mdx mice. These data are presented in Fig. 4–figure supplements 1B, 1C, and 1D. We added a sentence about the changes in CSA and muscle atrophy inducing factors in the Discussion.

Methods and material

Line 342-348: authors have described animals, but not specified sex and number of mice in each group. This part should be improved.

We apologize for our insufficient information about the sex and number of mice in the Materials and methods.

We added a sentence specifying the sex, number, and evaluation period of each mouse group in the section on the generation of BMD mice.

Line 426-433: authors described qPCR. It is necessary that the authors also describe primer sequences.

We apologize for any lack of information about the primer sequences used in qPCR analysis. Supplemental Table 1 lists the primer sequences.

We also added a sentence about the information in the primer list in the section on RNA isolation and RT-PCR in the Materials and methods.

Reviewer #2 (Recommendation for the authors):

Miyazaki et al. established three distinct BMD mouse models by removing different exon regions of the dystrophin gene. The authors demonstrated that the pathophysiological and molecular changes in these models progress at varying rates. Additionally, they observed a site-specific decline in type IIa fibers in BMD mice, while the proportions of other fiber types, such as type I and type IIx, remained consistent with those in wild-type mice. They proposed that the selective decay of type IIa fibers in BMD mice could be due to two primary factors: 1) muscle degeneration and regeneration, supported by their findings in cardiotoxin-treated mouse models, and 2) reduced capillary formation around type IIa fibers. However, the authors also presented evidence that type IIx fibers exhibited delayed recovery, similar to type IIa fibers, as demonstrated in cardiotoxin-induced regeneration models. Additionally, dot-patterned capillary formations were observed around both type IIa and type IIx fibers. Despite these findings, BMD mice did not show any changes in the proportion of type IIx fibers in inner BMD muscles. The authors should consider adding further analysis to strengthen their hypothesis and to disclose any possible mechanisms that led to these discrepancies.

If the authors hypothesize that reduced capillary density around type IIa fibers contribute to their site-specific decay in BMD mice, they should consider measuring and statistically analyzing the endothelial area around all fiber types. By plotting and comparing these measurements across different fiber types between wild-type, BMD, and mdx mice, the authors could provide more robust evidence to support their hypothesis. This approach would help clarify whether reduced capillary density is a contributing factor to the site-specific decay of type IIa fibers in BMD mice and the more diffuse, non-specific muscle changes observed in mdx mice.

The authors reported in the first part of the manuscript that histopathological changes, including muscle degeneration in BMD mice, are predominantly restricted to the inner part of the muscles. In the second part, they noted a decline in type IIa fibers specifically in the inner muscle region. To strengthen the hypothesis that the decay of type IIa fibers in the inner muscle is linked to muscle degeneration, the authors should consider performing histopathological measurements across different fiber types within the inner muscle. Reporting the correlations between these measurements would provide more compelling evidence to support their hypothesis.

We thank the reviewer for these important suggestions about the association between type IIa fiber reduction and capillary change around muscle fibers in BMD mice. We prepared an additional evaluation about the capillary formation (in Fig. 6B) and endothelial area (in Fig. 6C) around type IIx, IIb, and I fibers. We found that capillaries contacting around type IIx, IIb, and I fibers were poor in WT mice compared with those around type IIa fibers, and showed an ‘incomplete ring-pattern’ around type IIx fibers and a ‘dot-pattern’ around type IIb and I fibers in WT mice, in contrast with type IIa fibers, which showed remarkable ‘ring-pattern’ capillaries. Reflecting this, the changes in endothelial area around type IIx, IIb, and I fibers between WT and BMD mice were less than those around type IIa fibers. These results suggest that type IIa fibers may require numerous capillaries and maintained blood flow compared with type IIx, IIb, and I fibers, and this high requirement for blood flow might be associated with the type IIa fiber-specific decay in BMD mice.

We added the following.

(1) Sentences in the Results about the capillary changes around type IIx, IIb, and I fibers in WT, d45-49, and mdx mice.

(2) Sentences in the Results about the changes in endothelial area around type IIx, IIb, and I fibers in WT, d45-49, and mdx mice.

(3) Sentences in the Discussion about the association between the type IIa fiber-specific decay in BMD mice and the differences in capillary changes of each muscle fiber from WT to BMD mice.

We changed a sentence in the Discussion about the delayed recovery of type IIa and IIx fibers after CTX injection, to make it clear that the recovery of type IIx fibers was slower than that of type IIa fibers after CTX injection, and that therefore the type IIa fiber-specific decay in BMD mice might not be explained by this vulnerability and delayed recovery during muscle degeneration and regeneration.

Minor Issues:

Line 103: The word "mice" is duplicated and should be corrected.

We apologize that “mice” was duplicated. We have corrected it.

Line 120: Revise for clarity: "The proportion of opaque fibers is significantly different between d45-48 mice and WT at 3 months, with an increased tendency observed only in 1-month-old mice."

We apologize for the confusion about the proportion of opaque fibers. We revised this sentence as follows.

“Opaque fibers, which are thought to be precursors of necrotic fibers, increased at an earlier age of 1 month in d45–49 mice compared with WT mice; in contrast, the proportion of opaque fibers differs significantly between d45–47 and WT mice at 3 months, with an increased tendency only in 1-month-old mice (Fig. 2C).”

Line 152: Clarify the statement regarding utrophin levels, as it currently contradicts the Western blot data. The sentence reads: "The increased levels of utrophin are 8-fold higher at 1 month and 30-fold higher at 3 months." This should be verified against the data, as the band densities in the Western blots suggest otherwise.

We apologize for the confusion about utrophin expression levels. We revised this sentence as follows.

“By western blot analysis, the utrophin expression levels showed only an increased tendency in all BMD mice at 3 months, whereas there was a significant increase in mdx mice (8-fold at 1 month, and 30-fold at 3 months) compared to WT mice (Figs. 3C and F).”

Line 235: Correct the sentence to accurately reflect the findings: "BMD mice showed reduced muscle weakness."

We apologize for our incorrect wording. We have removed the word “reduced” in this sentence.

eLife Assessment

This valuable work provides solid evidence that a neuronal metallothionein, GIF/MT-3, incorporates metal-persulfide clusters. A variety of well-designed assays support the authors' hypothesis, revealing that sulfane sulfur is released from MT-3. The biological role of the persulfidated form is not yet clearly defined. There are caveats to the findings that limit the study, but the work will nevertheless prompt major follow-up work.

Reviewer #2 (Public review):

Summary:

In this manuscript, the authors reveal that GIF/MT-3 regulates the zinc homeostasis depending on the cellular redox status. The manuscript technically sounds, and their data concretely suggest that the recombinant MTs, not only GIF/MT-3 but also canonical MTs such as MT-1 and MT-2, contain sulfane sulfur atoms for the Zn-binding. The scenario proposed by the authors seems to be reasonable to explain the Zn homeostasis by the cellular redox balance.

Strengths:

The data presented in the manuscript solidly reveal that recombinant GIF/MT-3 contains sulfane sulfur.

Weaknesses:

It remains unclear whether native MTs, in particular induced MTs in vivo contain sulfane sulfur or not.

Comments on revisions:

Although the authors have revealed the sulfane sulfur content in native MT-3, my question, namely, whether canonical MT-1 and MT-2 contained sulfane sulfur after the induction has been left.<br /> The authors argue that the biological significance of sulfane sulfur in MTs lies in its ability to contribute to metal binding affinity, provide a sensing mechanism against oxidative stress, and aid in the regulation of the protein. Due to their biological roles, induced MT-1 and MT-2 could contain sulfane sulfur in their molecules. Thus, I expect the authors to evaluate or explain the sulfane sulfur content in induced MT-1 and MT-2.

Reviewer #3 (Public review):

Summary:

The authors were trying to show that a novel neuronal metallothionein of poorly defined function, GIF/MT3, is actually heavily persulfidated in both the Zn-bound and apo (metal-free) forms of the molecule as purified from a heterologous (bacterial) or native host. Evidence in support of this conclusion is strong, with both spectroscopic and mss spectrometry evidence strongly consistent with this general conclusion. The authors would appear to have achieved their aims.

Strengths:

The analytical data in support of the author's primary conclusions are strong. The authors also provide some modeling evidence that supports the contention that MT3 (and other MTs) can readily accommodate a sulfane sulfur on each of the 20 cysteines in the Zn-bound structure, with little perturbation of the overall structure. This is not the case with Cys trisulfides, which suggests that the persulfide-metallated state is clearly positioned at lower energy relative to the immediately adjacent thiolate- or trisulfidated metal coordination complexes.

Weaknesses:

The biological significance of the findings is not entirely clear. On the one hand, the analytical data are solid (albeit using a protein derived from a bacterial over-expression experiment), and yes, it's true that sulfane S can protect Cys from overoxidation, but everything shown in the summary figure (Fig. 9D) can be done with Zn release from a thiol by ROS, and subsequent reduction by the Trx/TR system. In addition, it's long been known that Zn itself can protect Cys from oxidation. I view this as a minor shortcoming that will motivate follow-up studies.

Impact:

The impact will be high since the finding is potentially disruptive to the MT field for sure. The sulfane sulfur counting experiment (the HPE-IAM electrophile trapping experiment) may well be widely adopted by the field. Those in the metals field always knew that this was a possibility, and it will interesting to see the extent to which metal binding thiolates broadly incorporate sulfane sulfur into their first coordination shells.

Comments on revisions:

The revised manuscript is only slightly changed from the original, with the inclusion of a supplementary figure (Fig. S2) and minor changes in the text. The authors did not choose to carry out the quantitative Zn binding experiment (which I really wanted to see), but given the complexities of the experiment, I'll let it go.

Fig. 9: the authors imply in the mechanistic "redox-switch" figure that Trx/TR can not reduce persulfide linkages. A number of groups have shown this to be the case. I recommend modifying the figure legend or text to make this clear to the reader,

Author response:

The following is the authors’ response to the original reviews.

Reviewer #1 (Public Review):

The manuscript by Dr. Shinkai and colleagues is about the posttranslational modification of a highly important protein, MT3, also known as the growth inhibitory factor. Authors postulate that MT3, or generally all MT isoforms, are sulfane sulfur binding proteins. The presence of sulfane sulfur at each Cys residue has, according to the authors, a critical impact on redox protein properties and almost does not affect zinc binding. They show a model in which 20 Cys residues with sulfane sulfur atoms can still bind seven zinc ions in the same clusters as unmodified protein. They also show that recombinant MT3 (but also MT1 and MT2) protein can react with HPE-IAM, an efficient trapping reagent of persulfides/polysulfides. This reaction performed in a new approach (high temperature and high reagent concentration) resulted in the formation of bis-S-HPE-AM product, which was quantitatively analyzed using LC-MS/MS. This analysis indicated that all Cys residues of MT proteins are modified by sulfane sulfur atoms. The authors performed a series of experiments showing that such protein can bind zinc, which dissociates in the reaction with hydrogen peroxide or SNAP. They also show that oxidized MT3 is reduced by thioredoxin. It gives a story about a new redox-dependent switching mechanism of zinc/persulfide cluster involving the formation of cystine tetrasulfide bridge.

The whole story is hard to follow due to the lack of many essential explanations or full discussion. What needs to be clarified is the conclusion (or its lack) about MT3 modification proven by mass spectrometry. Figure 1B shows the FT-ICR-MALDI-TOF/MS spectrum of recombinant MT3. It clearly shows the presence of unmodified MT3 protein without zinc ions. Ions dissociate in acidic conditions used for MALDI sample preparation. If the protein contained all Cys residues modified, its molecular weight would be significantly higher. Then, they show the MS spectrum (low quality) of oxidized protein (Fig. 1C), in which new signals (besides reduced apo-MT3) are observed. They conclude that new signals come from protein oxidation and modification with one or two sulfur atoms. If the conclusion on Cys residue oxidation is reasonable, how this protein contains sulfur is unclear. What is the origin of the sulfur if apo-MT does not contain it? Oxidized protein was obtained by acidification of the protein, leading to zinc dissociation and subsequent neutralization and air oxidation. Authors should perform a detailed isotope analysis of the isotopic envelope to prove that sulfur is bound to the protein. They say that the +32 mass increase is not due to the appearance of two oxygen donors. They do not provide evidence. This protein is not a sulfane sulfur binding protein, or its minority is modified. Moreover, it is unacceptable to write that during MT3 oxidation are "released nine molecules of H2". How is hydrogen molecule produced? Moreover, zinc is not "released", it dissociates from protein in a chemical process.

Thank you for your comment. According to your suggestion, we have rewritten the corresponding sentences below, together with addition of new Fig.1D.

First, the sentence “which corresponded to the mass of zinc-free apo-GIF/MT3 and indicated that zinc was removed during MS analysis.” was changed to “which corresponded to the mass of zinc-free apo-GIF/MT3 and indicated that zinc dissociates from protein in acidic conditions used for MALDI sample preparation.” in the introduction section. Second, we have added the following sentence “However, FT-ICR-MALDI-TOF/MS analysis failed to detect sulfur modifications in GIF/MT-3 (Fig. 1B), suggesting that sulfur modifications in the protein were dissociated during laser desorption/ionization. Therefore, we postulate that the small amount of sulfur detected in oxidized apo-GIF/MT-3 is derived from the effect of laser desorption/ionization rather than any actual modification of the minority component.” in the discussion section. Third, we have added new Fig. 1D and the corresponding citation in the introduction. Fourth, the sentence “An increase in mass of 32 Da can also result from addition of two oxygen atoms, but we attributed it to one sulfur atom for reasons described later.” was changed to “Note that an increase in mass of 32 Da can also result from addition of two oxygen atoms.”.

Another important point is a new approach to the HPE-IAM application. Zinc-binding MT3 was incubated with 5 mM reagent at 60°C for 36 h. Authors claim that high concentration was required because apoMT3 has stable conformation. Figure 2B shows that product concentration increases with higher temperature, but it is unclear why such a high temperature was used. Figure 1D shows that at 37°C, there is almost no reaction at 5 mM reagent. Changing parameters sounds reasonable only when the reaction is monitored by mass spectrometry. In conclusion, about 20 sulfane sulfur atoms present in MT3 would be clearly visible. Such evidence was not provided. Increased temperature and reagent concentration could cause modification of cysteinyl thiol/thiolates as well, not only persulfides/polysulfides. Therefore, it is highly possible that non-modified MT3 protein could react with HPE-IAM, giving false results. Besides mass spectrometry, which would clearly prove modifications of 20 Cys, authors should use very important control, which could be chemically synthesized beta- or alfa-domain of MT3 reconstituted with zinc (many protocols are present in the literature). Such models are commonly used to test any kind of chemistry of MTs. If a non-modified chemically obtained domain would undergo a reaction with HPE-IAM under such rigorous conditions, then my expectation would be right.

Thank you for your comments. Although we have already confirmed that no false-positive results were observed using this method in Fig. 5 (previously Fig. 4), we have conducted additional experiments by preparing chemically synthesized α- and β-domains of GIF/MT-3, as well as recombinant α- and β-domains of GIF/MT-3. As shown in the new Fig. S2A, the chemically synthesized α- and β-domains of GIF/MT-3 detected almost no sulfane sulfur (less than 1 molecule per protein), whereas the recombinant α- and β-domains detected several molecules of sulfane sulfur (more than 5 molecules per protein) (Fig. S2A). Therefore, I would like to emphasize here that the cysteine residue itself cannot be the source of the bis-S-HPE-AM product (sulfane sulfur derivative).

Accordingly, we have added the following sentence in the results section: “Because this assay was performed at relatively high temperatures (60°C), we also examined the sulfane sulfur levels of several mutant proteins using chemically synthesized α- and β-domains of GIF/MT-3 to eliminate false-positive results. As shown in Fig. S2A, sulfane sulfur (less than 1 molecule per protein) was undetectable in chemically synthesized α- and β-domains of GIF/MT-3, whereas several molecules of sulfane sulfur per protein were detected in recombinant α- and β-domains exhibited (Fig. S2B, left panel). These findings indicated that the sulfane sulfur detected in our assay was derived from biological processes executed during the production of GIF/MT-3 protein. We further analyzed mutant proteins with β-Cys-to-Ala and α-Cys-to-Ala substitutions and found that their sulfane sulfur levels were comparable with those of the α- and β-domains of GIF/MT-3, respectively (Fig. S2B, left panel). Additionally, Ser-to-Ala mutation did not affect the sulfane sulfur levels of GIF/MT-3. The zinc content of each mutant protein was also determined under these conditions (Fig. S2B, right panel).”

- The remaining experiments provided in the manuscript can also be applied for non-modified protein (without sulfane sulfur modification) and do not provide worthwhile evidence. For instance, hydrogen peroxide or SNAP may interact with non-modified MTs. Zinc ions dissociate due to cysteine residue modification, and TCEP may reduce oxidized residue to rescue zinc binding. Again, mass spectrometry would provide nice evidence.

Thank you for your comment. We understand that such experiments can also be applied to non-modified proteins (without sulfane sulfur modification). However, the experiments shown in Fig. 4 and Fig. 6 were conducted to investigate the role of sulfane sulfur under oxidative stress conditions, rather than to examine sulfur modification in the protein itself. As mentioned previously, it is difficult to detect sulfur modifications directly in the protein using MALDI-TOF/MS (Fig. 1), as sulfur modifications appear to dissociate during the laser desorption/ionization process.

- The same is thioredoxin (Fig. 7) and its reaction with oxidized MT3. Nonmodified and oxidized MT3 would react as well.

Thank you for your comment. We understand that such experiments can also be applied to non-modified MT-3 protein. However, to the best of our knowledge, this is the first report demonstrating that apo-MT-3 can serve as a good substrate for the Trx system. In fact, this experiment is not intended to prove that MT-3 is sulfane sulfur-binding protein. Rather, it demonstrates the novel finding that apo-MT3 serves as an excellent substrate for Trx and that the sulfane sulfur (persulfide structure) remains intact throughout the reduction process.

- If HPE-IAM reacts with Cys residues with unmodified MT3, which is more likely the case under used conditions, the protein product of such reaction will not bind zinc. It could be an explanation of the cyanolysis experiment (Fig. 6).

Thank you for your comment. As you pointed out, HPE-IAM reacts with cysteine residues in unmodified MT-3, thereby preventing zinc from binding to the protein. However, we did not use HPE-IAM prior to measuring zinc binding. Instead, HPE-IAM was used solely for determining the sulfane sulfur content in the protein, and thus it cannot explain the results of the cyanolysis experiment.

- Figure 4 shows the reactivity of (pol)sulfides with TCEP and HPE-IAM. What are redox potentials? Do they correlate with the obtained results?

Thank you for your comment. However, we must apologize as we do not fully understand the rationale behind determining redox potentials in this experiment. We believe the data itself to be very clear and presenting convincing results.

- Raman spectroscopy experiments would illustrate the presence of sulfane sulfur in MT3 only if all Cys were modified.

Yes, that is correct. Since approximately 20 sulfane sulfur atoms are detected in the protein with 20 cysteine residues, we believe that nearly all cysteine residues are modified by sulfane sulfur. Therefore, Raman spectroscopy is considered applicable to our current study.

- The modeling presented in this study is very interesting and confirms the flexibility of metallothioneins. MT domains are known to bind various metal ions of different diameters. They adopt in this way to larger size the ions. The same mechanism could be present from the protein site. The presence of 9 or 11 sulfur atoms in the beta or alfa domain would increase the size of the domains without changing the cluster structure.

We truly appreciate your positive evaluation of this work.

- Comment to authors. Apo-MT is not present in the cell. It exists as a partially metallated species. The term "apo-MT" was introduced to explain that MTs are not fully saturated by metals and function as a metal buffer system. Apo-MT comes from old ages when MT was considered to be present only in two forms: apo-form and fully saturated forms.

Thank you for your insightful comments. We find it reasonable to understand that apo-MT exists as a partially metallated species within the cell.

Reviewer #2 (Public Review):

Summary:

In this manuscript, the authors reveal that GIF/MT-3 regulates zinc homeostasis depending on the cellular redox status. The manuscript technically sounds, and their data concretely suggest that the recombinant MTs, not only GIF/MT-3 but also canonical MTs such as MT-1 and MT-2, contain sulfane sulfur atoms for the Zn-binding. The scenario proposed by the authors seems to be reasonable to explain the Zn homeostasis by the cellular redox balance.

Strengths:

The data presented in the manuscript solidly reveal that recombinant GIF/MT-3 contains sulfane sulfur.

Weaknesses:

It is still unclear whether native MTs, in particular, induced MTs in vivo contain sulfane sulfur or not.

Thank you for pointing out the strengths and weaknesses of this manuscript. Based on your suggestions, we have determined the sulfane sulfur content in the native GIF/MT-3 protein, as explained in our response to "Recommendations for the Authors #2."

Reviewer #3 (Public Review):

Summary:

The authors were trying to show that a novel neuronal metallothionein of poorly defined function, GIF/MT3, is actually heavily persulfidated in both the Zn-bound and apo (metal-free) forms of the molecule as purified from a heterologous or native host. Evidence in support of this conclusion is compelling, with both spectroscopic and mass spectrometry evidence strongly consistent with this general conclusion. The authors would appear to have achieved their aims.

Strengths:

The analytical data are compelling in support of the author's primary conclusions are strong. The authors also provide some modeling evidence that strongly supports the contention that MT3 (and other MTs) can readily accommodate sulfane sulfur on each of the 20 cysteines in the Zn-bound structure, with little perturbation of the structure. This is not the case with Cys trisulfides, which suggests that the persulfide-metallated state is clearly positioned at lower energy relative to the immediately adjacent thiolate- or trisulfidated metal coordination complexes.

Weaknesses:

The biological significance of the findings is not entirely clear. On the one hand, the analytical data are clearly solid (albeit using a protein derived from a bacterial over-expression experiment), and yes, it's true that sulfane S can protect Cys from overoxidation, but everything shown in the summary figure (Fig. 8D) can be done with Zn release from a thiol by ROS, and subsequent reduction by the Trx/TR system. In addition, it's long been known that Zn itself can protect Cys from oxidation. I view this as a minor weakness that will motivate follow-up studies. Fig. 1 was incomplete in its discussion and only suggests that a few S atoms may be covalently bound to MT3 as isolated. This is in contrast to the sulfate S "release" experiment, which I find quite compelling.

Impact:

The impact will be high since the finding is potentially disruptive to the metals in the biology field in general and the MT field for sure. The sulfane sulfur counting experiment (the HPE-IAM electrophile trapping experiment) may well be widely adopted by the field. Those of us in the metals field always knew that this was a possibility, and it will interesting to see the extent to which metal-binding thiolates broadly incorporate sulfate sulfur into their first coordination shells.

Thank you for pointing out the strengths and weaknesses of this manuscript. As you noted, the explanations and discussions regarding Fig. 1 were missing. To address this, we have added the following sentences to the discission section: “However, FT-ICR-MALDI-TOF/MS analysis failed to detect sulfur modifications in GIF/MT-3 (Fig. 1B), suggesting that sulfur modifications in the protein were dissociated during laser desorption/ionization. Therefore, we postulate that the small amount of sulfur detected in oxidized apo-GIF/MT-3 is derived from the effect of laser desorption/ionization rather than any actual modification of the minority component.”

Reviewer #1 (Recommendations For The Authors):

Overall, the topic of the study is interesting, but the provided evidence is insufficient to claim that MT3 is a sulfane sulfur-binding protein. Indeed, some recent studies showed that natural and recombinant MT proteins can be modified, but only one or a few cysteine residues were modified. Authors should follow my suggestion and apply mass spectrometry to all performed reactions and, first of all, to freshly obtained protein. I strongly suggest using chemically synthesized and reconstituted domains to test whether the home-developed approach is appropriate. Moreover, native MS and ICP-MS analysis of MT3 would support their claims.

Thank you for your insightful comments. Following your suggestions, we have prepared chemically synthesized proteins of the α- and β-domains of GIF/MT-3 and conducted additional experiments, as explained in response comments to “Public Review #1”. Regarding the MS analysis, we have also added a discussion on the difficulty of detecting sulfur modifications in the protein.

Reviewer #2 (Recommendations For The Authors):

I have some minor points which should be considered by the authors.

(1) Table 1: In the simulation by MOE, the authors speculated 7 atoms of metal bound to GIF/MT-3. Although a total of 7 atoms of Zn or Cd are actually bound to MTs as a divalent ion, the number of Cu and Hg bound to MTs as a monovalent ion is scientifically controversial. Several ideas have been proposed in the literature, however, "7 atoms of Cu or Hg" could be inappropriate as far as I know. The authors should simulate again using a more appropriate number of Cu or Hg in MTs.

Thank you for providing this valuable information. We reviewed several papers by the Stillman group and found that the relative binding constants of Cu4-MT, Cu6-MT, and Cu10-MT were determined after the addition of Cu(I) to apo MT-1A, MT-2, and MT-3 (Melenbacher and Stillman, Metallomics, 2024). However, incorporating these copper numbers into our GIF/MT-3 simulation model proved challenging. Therefore, we decided to omit the score value for copper in Table 1.

On the other hand, some researchers have reported that mercury binds to MT as a divalent ion, and the formation of Hg₇MT is possible (not just other forms). Therefore, we decided to continue using the score value for mercury shown in Table 1.

(2) If possible, native MT samples isolated from an experimental animal should be evaluated for the sulfane sulfur content. Canonical MTs, MT-1 and MT-2, are highly inducible by not only heavy metals but also oxidative stress. Under the oxidative stress condition such as the exposure of hydrogen peroxide, it is questionable whether the induced Zn-MTs contain sulfane sulfur or not.

According to your suggestion, we evaluated the sulfane sulfur content in native GIF/MT-3 samples isolated from mouse brain cytosol (Fig. 10). The measured amount was 3.3 per protein. This suggests that sulfane sulfur in GIF/MT-3 could be consumed under oxidative conditions, as you anticipated. Another possible explanation for the discrepancy between the native form and recombinant protein is likely related to metal binding in the protein. It is generally understood that both zinc and copper bind to GIF/MT-3 in approximately equal proportions in vivo. When we prepared recombinant copper-binding GIF/MT-3 protein, the sulfane sulfur content in the protein was significantly different (approximately 4.0 per protein) compared to the Zn₇GIF/MT-3 form. Further studies are needed to clarify the relationship between sulfane sulfur binding and the types of metals in the future.

(3) The biological significance of sulfane sulfur in MTs is still unclear to me.

Thank you for your comments. To address this question, we have added the following sentence to the discussion section: “The biological significance of sulfane sulfur in MTs lies in its ability to 1) contribute to metal binding affinity, 2) provide a sensing mechanism against oxidative stress, and 3) aid in the regeneration of the protein.”

(4) According to the widely accepted nomenclature of MT, "MT3" should be amended to "MT-3".

According to your suggestion, we have amended from MT3 to MT-3 throughout the manuscript.

Reviewer #3 (Recommendations For The Authors):

Most of my comments are editorial in nature, largely focused on what I perceive as overinterpretation or unnecessary speculation.

The authors state in the abstract that the intersection of sulfane sulfur and Zn enzymes "has been overlooked." This is not actually true - please tone down to "under investigated" or something like this.

Based on your suggestion, we have replaced the term “has been overlooked” with “has been under investigated” in the abstract.

Line 228: The discussion of Fig. 6C involved too much speculation. I cannot see a quantitative experiment that supports this.

Based on your suggestion, we have removed Fig. 6C (currently referred to as Fig. 7C). Additionally, we have revised the sentence from “implying that the sulfane sulfur is an essential zinc ligand in apo-GIF/MT3 and that an asymmetric SSH or SH ligand is insufficient for native zinc binding (Fig. 6C)” to “implying the contribution of sulfane sulfur to zinc binding in GIF/MT-3”.

Line 247 "persulfide in apo-GIF/MT3 seems.." I think the authors mean that the Zn form of the protein is resistant to Trx or TCEP.

Thank you for pointing this out. We realized that the term “persulfide in apo-GIF/MT3” might be confusing. Therefore, we have replaced it with “persulfide formation derived from apo-GIF/MT3” in the corresponding sentence.

Molecular modeling: We need more details- were these structures energy-minimized in any way? Can the authors comment on the plethora of S-S dihedral angles in these structures, and whether they are consistent with expectations of covalent geometry? Please add text to explain or even a table that compiles these data.

Thank you for your comment. Yes, energy minimization calculations for structural optimization were conducted during homology modeling in MOE. In fact, we have already stated in the Methods section that “Refinement of the model with the lowest generalized Born/volume integral (GBVI) score was achieved through energy minimization of outlier residues in Ramachandran plots generated within MOE.” In this model, covalent geometry, including the S-S dihedral angles, is also taken into consideration.

What is a thermostability score? Perhaps a bit more discussion here and what relationship this has to an apparent (or macroscopic) metal affinity constant.

The thermostability score is used to compare the thermal stability between the wild-type and mutant proteins. As shown in Equation (1) in the method section, it is calculated by subtracting the energy of the hypothetical unfolded state from the energy of the folded state. Since obtaining the structure of the unfolded state requires extensive computational effort, MOE employs an empirical formula based on two-dimensional structural features to estimate it. The ΔΔG values represent the difference between ΔGf(WT) and ΔGf(Mut). However, because it is difficult to directly determine ΔGf(Mut) and ΔGf(WT), MOE calculates ΔΔG using the thermodynamic cycle equivalence: ΔΔGs =ΔGsf (WT→Mut) - ΔGsu (WT→Mut), as expressed in Equation (1).

On the other hand, the affinity score represents the interaction energy between the target ligand and the protein. In this study, we calculated the affinity score by selecting metal atoms as the ligands. The interaction energy (E int) is defined as:

E int = E complex − E receptor − E ligand

where each term is as follows:

E complex : Potential energy of the complex.

E receptor : Potential energy of the receptor alone.

E ligand : Potential energy of the ligand alone.

Each potential energy term includes contributions from bonded interactions such as bond lengths and bond angles. However, since there is no structural difference among E receptor, and E ligand, the bonded energy components cancel out. Consequently, E int is determined as:

E int = ΔEele +ΔEvdW +ΔE sol

Here, a negative E int indicates that the complex is more stable, while a positive E int implies that the receptor and ligand are more stable in their dissociated states.

We have revised the sentence "The affinity score was also calculated using MOE software as the difference between the ΔΔGs values of the protein, free zinc, and metal–protein complex” to "The affinity score was also calculated using MOE software as the difference between the potential energy values of the protein, free zinc, and metal–protein complex” to correct the misdescription.

Lines 278-280: The authors state that they observe a "marked enhancement of metal binding affinity, and rearrangement of zinc ions." I don't see support for this rather provocative conclusion. This is the expectation of course. I would love to see actual experimental data on this point, direct binding titrations with metals performed before and after the release of the sulfate sulfur atoms.

Thank you for your comments. Although this statement is based on the 3D modeling simulation, we have also experimentally observed that the diminishment of sulfane sulfur in GIF/MT-3 resulted in a decrease in zinc binding levels, as shown in Fig. 7. However, conducting direct binding titration experiments was difficult for us due to the difficulty in preparing pure GIF/MT-3 protein with or without sulfane sulfur. Therefore, we have revised the sentence "marked enhancement of metal binding affinity, and rearrangement of zinc ions" to simply "enhancement of metal binding affinity" to avoid over-speculation.

Table I- quantitatively lower stability for the Cu complex- the stoichiometry is clearly wrong in this simulation- please redo this simulation with the right stoichiometry or Cu to MT3- consult a Stillman paper.

Thank you for providing this valuable information. We reviewed several papers by the Stillman group and found that the relative binding constants of Cu4-MT, Cu6-MT, and Cu10-MT were determined after the addition of Cu(I) to apo MT-1A, MT-2, and MT-3 (Melenbacher and Stillman, Metallomics, 2024). However, incorporating these copper numbers into our GIF/MT-3 simulation model proved challenging. Therefore, we decided to omit the score value for copper in Table 1.

I like the model for reversible metal release mediated by the thioredoxin system (Fig. 8D)- but you can also do this with thiols- nothing really novel here. Has it been generally established that tetraulfides are better substrates for the Trx/TR system? The data shown in Fig. 7B seems to suggest this, but is this broadly true, from the literature?

There are reports describing that persulfides and polysulfides are reduced by the thioredoxin system. However, it is not well-established that tetraulfides are better substrates for the Trx/TR system. To the best of our knowledge, this is the first report demonstrating that apo-MT-3 can serve as a good substrate for the Trx/TR system. Further research is required to compare the catalytic efficiency between proteins containing disulfide and those with tetraulfide moieties.

Line 380: Many groups have reported that many proteins are per- or polysulfidated in a whole host of cells using mass spectrometry workflows, and that terminal persulfides can be readily reduced by general or specific Trx/TR systems. This work could be better acknowledged in the context of the authors' demonstration of the reduction of the tetrasulfides, which itself would appear to be novel (and exciting!).

We truly appreciate your positive evaluation of this work.

eLife Assessment

This fundamental article significantly advances our understanding of FGF signalling, and in particular, highlights the complex modifications affecting this pathway. The evidence for the authors' claims is convincing, combining state-of-the-art conditional gene deletion in the mouse lens with histological and molecular approaches. This work should be of great interest to molecular and developmental biologists beyond the lens community.

Reviewer #1 (Public review):

Summary:

This manuscript uses the eye lens as a model to investigate basic mechanisms in the Fgf signaling pathway. Understanding Fgf signaling is of broad importance to biologists as it is involved in the regulation of various developmental processes in different tissues/organs and is often misregulated in disease states. The Fgf pathway has been studied in embryonic lens development, namely with regards to its involvement in controlling events such as tissue invagination, vesicle formation, epithelium proliferation and cellular differentiation, thus making the lens a good system to uncover the mechanistic basis of how the modulation of this pathway drives specific outcomes. Previous work has suggested that proteins, other than the ones currently known (e.g., the adaptor protein Frs2), are likely involved in Fgfr signaling. The present study focuses on the role of Shp2 and Shc1 proteins in the recruitment of Grb2 in the events downstream of Fgfr activation.

Strengths:

The findings reveal that the juxtamembrane region of the Fgf receptor is necessary for proper control of downstream events such as facilitating key changes in transcription and cytoskeleton during tissue morphogenesis. The authors conditionally deleted all four Fgfrs in the mouse lens that resulted in molecular and morphological lens defects, most importantly, preventing the upregulation of the lens induction markers Sox2 and Foxe3 and the apical localization of F-actin, thus demonstrating the importance of Fgfrs in early lens development, i.e. during lens induction. They also examined the impact of deleting Fgfr1 and 2, on the following stage, i.e. lens vesicle development, which could be rescued by expressing constitutively active KrasG12D. By using specific mutations (e.g. Fgfr1ΔFrs lacking the Frs2 binding domain and Fgfr2LR harboring mutations that prevent binding of Frs2), it is demonstrated that the Frs2 binding site on Fgfr is necessary for specific events such as morphogenesis of lens vesicle. Further, by studying Shp2 mutations and deletions, the authors present a case for Shp2 protein to function in a context-specific manner in the role of an adaptor protein and a phosphatase enzyme. Finally, the key surprising finding from this study is that downstream of Fgfr signaling, Shc1 is an important alternative pathway - in addition to Shp2 - involved in the recruitment of Grb2 and in the subsequent activation of Ras. The methodologies, namely, mouse genetics and state-of-the-art cell/molecular/biochemical assays are appropriately used to collect the data, which are soundly interpreted to reach these important conclusions. Overall, these findings reveal the flexibility of the Fgf signaling pathway and it downstream mediators in regulating cellular events. This work is expected to be of broad interest to molecular and developmental biologists.

Weaknesses:

A weakness that needs to be discussed is that Le-Cre depends on Pax6 activation, and hence its use in specific gene deletion will not allow evaluation of the requirement of Fgfrs in the expression of Pax6 itself. But since this is the earliest Cre available for deletion in the lens, mentioning this in the discussion would make the readers aware of this issue.

Reviewer #2 (Public review):

Summary

I have reviewed the revised manuscript submitted by Wang et al., which is entitled "Shc1 cooperates with Frs2 and Shp2 to recruit Grb2 in FGF-induced lens development". In this paper, the authors first examined lens phenotypes in mice with Le-Cre-mediated knockdown (KD) of all four FGFR (FGFR1-4), and found that pERK signals, Jag1 and foxe3 expression are absent or drastically reduced, indicating that FGF signaling is essential for lens induction. Next, the authors examined lens phenotypes of FGFR1/2-KD mice and found that lens fiber differentiation is compromised and that proliferative activity and cell survival are also compromised in lens epithelium. Interestingly, Kras activation rescues defects in lens growth and lens fiber differentiation in FGFR1/2-KD mice, indicating that Ras activation is a key step for lens development, downstream of FGF signaling. Next, the authors examined the role of Frs2, Shp2 and Grb2 in FGF signaling for lens development. They confirmed that lens fiber differentiation is compromised in FGFR1/3-KD mice combined with Frs2-dysfunctional FGFR2 mutants, which is similar to lens phenotypes of Grb2-KD mice. However, lens defects are milder in mice with Shp2YF/YF and Shp2CS mutant alleles, indicating that involvement of Shp2 is limited for the Grb2 recruitment for lens fiber differentiation. Lastly, the authors showed new evidence on the possibility that another adapter protein, Shc1, promotes Grb2 recruitment independent of Frs2/Shp2-mediated Grb2 recruitment.

Strength

Overall, the manuscript provides valuable data on how FGFR activation leads to Ras activation through the adapter platform of Frs2/Shp2/Grb2, which advances our understanding on complex modification of FGF signaling pathway. The authors applied a genetic approach using mice, whose methods and results are valid to support the conclusion. The discussion also well summarizes the significance of their findings.

Weakness

The authors found that the new adaptor protein Shc1 is involved in Grb2 recruitments in response to FGF receptor activation. However, the main data on Shc1 are only histological sections and statistical evaluation of lens size. In the revised manuscript, the authors did not answer my major concern that cellular-level data are missing, which is not fully enough to support their main conclusion on the involvement of Shc1 in Grb2 recruitment of FGF signaling for lens development. Since the title of this manuscript is that Shc1 cooperates with Frs2 and Shp2 to recruit Grb2 in FGF-induced lens development, it is important to provide the cellular-level evidence on Shc1.

Reviewer #3 (Public review):

Summary:

The manuscript entitled "Shc1 cooperates with Frs2 and Shp2 to recruit Grb2 in FGF-induced lens development" by Wang et al., investigates the molecular mechanism used by FGFR signaling to support lens development. The lens has long been known to depend on FGFR-signaling for proper development. Previous investigations have demonstrated the FGFR signaling is required for embryonic lens cell survival and for lens fiber cell differentiation. The requirement of FGFR signaling for lens induction has remained more controversial as deletion of both Fgfr1 and Fgfr2 during lens placode formation does not prevent the induction of definitive lens markers such as FOXE3 or αA-crystallin. Here the authors have used the Le-Cre driver to delete all four FGFR genes from the developing lens placode demonstrating a definitive failure of lens induction in the absence of FGFR-signaling. The authors focused on FGFR1 and FGFR2, the two primary FGFRs present during early lens development and demonstrated that lens development could be significantly rescued in lenses lacking both FGFR1 and FGFR2 by expressing a constitutively active allele of KRAS. They also showed that the removal of pro-apoptotic genes Bax and Bak could also lead to a substantial rescue of lens development in lenses lacking both FGFR1 and FGFR2. In both cases, the lens rescue included both increased lens size and the expression of genes characteristic of lens cells.

Significantly the authors concentrated on the juxtamembrane domain, a portion of the FGFRs associated with FRS2. Previous investigations have demonstrated the importance of FRS2 activation for mediating a sustained level of ERK activation. FRS2 is known to associate both with GRB2 and SHP2 to activate RAS. The authors utilized a mutant allele of Fgfr1, lacking the entire juxtamembrane domain (Fgfr1ΔFrs) and an allele of Fgfr2 containing two-point mutations essential for Frs2 binding (Fgfr2LR). When combining three floxed alleles and leaving only one functional allele (Fgfr1ΔFrs or Fgfr2LR) the authors got strikingly different phenotypes. When only the Fgfr1ΔFrs allele was retained, the lens phenotype matched that of deleting both Fgfr1 and Fgfr2. However, when only the Fgfr2LR allele was retained the phenotype was significantly milder, primarily affecting lens fiber cell differentiation, suggesting that something other than FRS2 might be interacting with the juxtamembrane domain to support FGFR signaling in the lens. The authors also deleted Grb2 in the lens and showed that the phenotype was similar to that of the lenses only retaining the Fgfr2LR allele, resulting a failure of lens fiber cell differentiation and decreased lens cell survival. However, mutating the major tyrosine phosphorylation site of GRB2 did not affect lens development. The authors additionally investigated the role of SHP2 in lens development by either deleting SHP2 or by making mutations in the SHP2 catalytic domain. The deletion of the SHP2 phosphatase activity did not affect lens development as severely as total loss of SHP2 protein, suggesting a function for SHP2 outside of its catalytic activity. Although the loss of Shc1 alone has only a slight effect on lens size and pERK activation in the lens, the authors showed that the loss of Shc1 exacerbated the lens phenotype in lenses lacking both Frs2 and Shp2. The authors suggest that SHC1 binds to the FGFR juxtamembrane domain allowing for the recruitment of GRB2 in independently of FRS2.

Strengths:

(1) The authors used a variety of genetic tools to carefully dissect the essential signals downstream of FGFR signaling during lens development.

(2) The authors made a convincing case that something other than FRS2 binding mediates FGFR signaling in the juxtamembrane domain.

(3) The authors demonstrated that despite the requirement of both the adaptor function and phosphatase activity of SHP2 are required for embryonic survival, neither of these activities is absolutely required for lens development.

(4) The authors provide more information as to why FGFR loss has a phenotype much more severe than the loss of FRS2 alone during lens development.

(5) The authors followed up their work analyzing various signaling molecules in the context of lens development with biochemical analyses of FGF-induced phosphorylation in murine embryonic fibroblasts (MEFs).

(6) In general, this manuscript represents a Herculean effort to dissect FGFR signaling in vivo with biochemical backing with cell culture experiments in vitro.

Weaknesses:

(1) The authors demonstrate that the loss of FGFR1 and FGFR2 can be compensated by a constitutive active KRAS allele in the lens and suggest that FGFRs largely support lens development only by driving ERK activation. However, the authors also saw that lens development was substantially rescued by preventing apoptosis through the deletion of BAK and BAX. To my knowledge, the deletion of BAK and BAX should not independently activate ERK. The authors do not show whether ERK activation is restored in the BAK/BAX deficient lenses. Do the authors suggest the FGFR3 and/or FGFR4 provide sufficient RAS and ERK activation for lens development when apoptosis is suppressed? Alternatively, is it the survival function of FGFR-signaling as much as a direct effect on lens differentiation?

(2) Do the authors suggest that GRB2 is required for RAS activation and ultimately ERK activation? If so, do the authors suggest that ERK activation is not required for FGFR-signaling to mediate lens induction? This would follow considering that the GRB2 deficient lenses lack a problem with lens induction.

(3) The increase in p-Shc is only slightly higher in the Cre FGFR1f/f FGFR2r/LR than in the FGFR1f/Δfrs FGFR2f/f. Can the authors provide quantification?

(4) The authors have not shown directly that Shc1 binds to the juxtamembrane region of either Fgfr1 or Fgfr2.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews:

Reviewer #1 (Public review):

Summary:

This manuscript uses the eye lens as a model to investigate basic mechanisms in the Fgf signaling pathway. Understanding Fgf signaling is of broad importance to biologists as it is involved in the regulation of various developmental processes in different tissues/organs and is often misregulated in disease states. The Fgf pathway has been studied in embryonic lens development, namely with regards to its involvement in controlling events such as tissue invagination, vesicle formation, epithelium proliferation, and cellular differentiation, thus making the lens a good system to uncover the mechanistic basis of how the modulation of this pathway drives specific outcomes. Previous work has suggested that proteins, other than the ones currently known (e.g., the adaptor protein Frs2), are likely involved in Fgfr signaling. The present study focuses on the role of Shp2 and Shc1 proteins in the recruitment of Grb2 in the events downstream of Fgfr activation.

Strengths:

The findings reveal that the juxtamembrane region of the Fgf receptor is necessary for proper control of downstream events such as facilitating key changes in transcription and cytoskeleton during tissue morphogenesis. The authors conditionally deleted all four Fgfrs in the mouse lens that resulted in molecular and morphological lens defects, most importantly, preventing the upregulation of the lens induction markers Sox2 and Foxe3 and the apical localization of F-actin, thus demonstrating the importance of Fgfrs in early lens development, i.e. during lens induction. They also examined the impact of deleting Fgfr1 and 2, on the following stage, i.e. lens vesicle development, which could be rescued by expressing constitutively active KrasG12D. By using specific mutations (e.g. Fgfr1ΔFrs lacking the Frs2 binding domain and Fgfr2LR harboring mutations that prevent binding of Frs2), it is demonstrated that the Frs2 binding site on Fgfr is necessary for specific events such as morphogenesis of lens vesicle. Further, by studying Shp2 mutations and deletions, the authors present a case for Shp2 protein to function in a context-specific manner in the role of an adaptor protein and a phosphatase enzyme. Finally, the key surprising finding from this study is that downstream of Fgfr signaling, Shc1 is an important alternative pathway - in addition to Shp2 - involved in the recruitment of Grb2 and in the subsequent activation of Ras. The methodologies, namely, mouse genetics and state-of-the-art cell/molecular/biochemical assays are appropriately used to collect the data, which are soundly interpreted to reach these important conclusions. Overall, these findings reveal the flexibility of the Fgf signaling pathway and its downstream mediators in regulating cellular events. This work is expected to be of broad interest to molecular and developmental biologists.

Weaknesses:

A weakness that needs to be discussed is that Le-Cre depends on Pax6 activation, and hence its use in specific gene deletion will not allow evaluation of the requirement of Fgfrs in the expression of Pax6 itself. But since this is the earliest Cre available for deletion in the lens, mentioning this in the discussion would make the readers aware of this issue. Referring to Jag1 among "lens-specific markers" (page 5) is debatable, suggesting changing to the lines of "the expected upregulation of Jag1 in lens vesicle". The Abstract could be modified to clearly convey the existing knowledge gap and the key findings of the present study. As it stands now, it is a bit all over the place. Some typos in the manuscript need to be fixed, e.g. "...yet its molecular mechanism remains largely resolved" - unresolved? "...in the development lens" - in the developing lens? In Figure 4 legend, "(B) Grb2 mutants Grb2 mutants displayed...", etc.

We thank the reviewer for the thoughtful and constructive feedback. We have added the caveat regarding the Le-Cre dependency on Pax6 expression to the discussion, removed the reference to Jag1 as a “lens-specific marker” and corrected the typographical errors noted by the reviewer.

Reviewer #2 (Public review):

Summary:

I have reviewed a manuscript submitted by Wang et al., which is entitled "Shc1 cooperates with Frs2 and Shp2 to recruit Grb2 in FGF-induced lens development". In this paper, the authors first examined lens phenotypes in mice with Le-Cre-mediated knockdown (KD) of all four FGFR (FGFR1-4), and found that pERK signals, Jag1, and foxe3 expression are absent or drastically reduced, indicating that FGF signaling is essential for lens induction. Next, the authors examined lens phenotypes of FGFR1/2-KD mice and found that lens fiber differentiation is compromised and that proliferative activity and cell survival are also compromised in lens epithelium. Interestingly, Kras activation rescues defects in lens growth and lens fiber differentiation in FGFR1/2-KD mice, indicating that Ras activation is a key step for lens development. Next, the authors examined the role of Frs2, Shp2, and Grb2 in FGF signaling for lens development. They confirmed that lens fiber differentiation is compromised in FGFR1/3-KD mice combined with Frs2-dysfunctional FGFR2 mutants, which is similar to lens phenotypes of Grb2-KD mice. However, lens defects are milder in mice with Shp2YF/YF and Shp2CS mutant alleles, indicating that the involvement of Shp2 is limited for the Grb2 recruitment for lens fiber differentiation. Lastly, the authors showed new evidence on the possibility that another adapter protein, Shc1, promotes Grb2 recruitment independent of Frs2/Shp2-mediated Grb2 recruitment.

Strengths:

Overall, the manuscript provides valuable data on how FGFR activation leads to Ras activation through the adapter platform of Frs2/Shp2/Grb2, which advances our understanding of complex modification of the FGF signaling pathway. The authors applied a genetic approach using mice, whose methods and results are valid to support the conclusion. The discussion also well summarizes the significance of their findings.

Weaknesses:

The authors eventually found that the new adaptor protein Shc1 is involved in Grb2 recruitments in response to FGF receptor activation. however, the main data for Shc1 are histological sections and statistical evaluation of lens size. So, my major concern is that the authors need to provide more detailed data to support the involvement of Shc1 in Grb2 recruitment of FGF signaling for lens development.

We thank the reviewer for the positive comments and valuable suggestions. We have addressed the concerns in detail in the response to the recommendation outlined below.

Reviewer #3 (Public review):

Summary:

The manuscript entitled "Shc1 cooperates with Frs2 and Shp2 to recruit Grb2 in FGF-induced lens development" by Wang et al., investigates the molecular mechanism used by FGFR signaling to support lens development. The lens has long been known to depend on FGFR signaling for proper development. Previous investigations have demonstrated that FGFR signaling is required for embryonic lens cell survival and for lens fiber cell differentiation. The requirement of FGFR signaling for lens induction has remained more controversial as deletion of both Fgfr1 and Fgfr2 during lens placode formation does not prevent the induction of definitive lens markers such as FOXE3 or αA-crystallin. Here the authors have used the Le-Cre driver to delete all four FGFR genes from the developing lens placode demonstrating a definitive failure of lens induction in the absence of FGFR signaling. The authors focused on FGFR1 and FGFR2, the two primary FGFRs present during early lens development, and demonstrated that lens development could be significantly rescued in lenses lacking both FGFR1 and FGFR2 by expressing a constitutively active allele of KRAS. They also showed that the removal of pro-apoptotic genes Bax and Bak could also lead to a substantial rescue of lens development in lenses lacking both FGFR1 and FGFR2. In both cases, the lens rescue included both increased lens size and the expression of genes characteristic of lens cells.

Significantly the authors concentrated on the juxtamembrane domain, a portion of the FGFRs associated with FRS2. Previous investigations have demonstrated the importance of FRS2 activation for mediating a sustained level of ERK activation. FRS2 is known to associate both with GRB2 and SHP2 to activate RAS. The authors utilized a mutant allele of Fgfr1, lacking the entire juxtamembrane domain (Fgfr1ΔFrs), and an allele of Fgfr2 containing two-point mutations essential for Frs2 binding (Fgfr2LR). When combining three floxed alleles and leaving only one functional allele (Fgfr1ΔFrs or Fgfr2LR) the authors got strikingly different phenotypes. When only the Fgfr1ΔFrs allele was retained, the lens phenotype matched that of deleting both Fgfr1 and Fgfr2. However, when only the Fgfr2LR allele was retained the phenotype was significantly milder, primarily affecting lens fiber cell differentiation, suggesting that something other than FRS2 might be interacting with the juxtamembrane domain to support FGFR signaling in the lens. The authors also deleted Grb2 in the lens and showed that the phenotype was similar to that of the lenses only retaining the Fgfr2LR allele, resulting in a failure of lens fiber cell differentiation and decreased lens cell survival. However, mutating the major tyrosine phosphorylation site of GRB2 did not affect lens development. The author additionally investigated the role of SHP2 lens development by making by either deleting SHP2 or by making mutations in the SHP2 catalytic domain. The deletion of the SHP2 phosphatase activity did not affect lens development as severely as the total loss of SHP2 protein, suggesting a function for SHP2 outside of its catalytic activity. Although the loss of Shc1 alone has only a slight effect on lens size and pERK activation in the lens, the authors showed that the loss of Shc1 exacerbated the lens phenotype in lenses lacking both Frs2 and Shp2. The authors suggest that SHC1 binds to the FGFR juxtamembrane domain allowing for the recruitment of GRB2 independently of FRS2.

Strengths:

(1) The authors used a variety of genetic tools to carefully dissect the essential signals downstream of FGFR signaling during lens development.

(2) The authors made a convincing case that something other than FRS2 binding mediates FGFR signaling in the juxtamembrane domain.

(3) The authors demonstrated that despite the requirement of both the adaptor function and phosphatase activity of SHP2 are required for embryonic survival, neither of these activities is absolutely required for lens development.

(4) The authors provide more information as to why FGFR loss has a phenotype much more severe than the loss of FRS2 alone during lens development.

(5) The authors followed up their work analyzing various signaling molecules in the context of lens development with biochemical analyses of FGF-induced phosphorylation in murine embryonic fibroblasts (MEFs).

(6) In general, this manuscript represents a Herculean effort to dissect FGFR signaling in vivo with biochemical backing with cell culture experiments in vitro.

We thank the reviewer for the thorough review of our paper and positive comments.

Weaknesses:

(1) The authors demonstrate that the loss of FGFR1 and FGFR2 can be compensated by a constitutive active KRAS allele in the lens and suggest that FGFRs largely support lens development only by driving ERK activation. However, the authors also saw that lens development was substantially rescued by preventing apoptosis through the deletion of BAK and BAX. To my knowledge, the deletion of BAK and BAX should not independently activate ERK. The authors do not show whether ERK activation is restored in the BAK/BAX deficient lenses. Do the authors suggest the FGFR3 and/or FGFR4 provide sufficient RAS and ERK activation for lens development when apoptosis is suppressed? Alternatively, is it the survival function of FGFR-signaling as much as a direct effect on lens differentiation?

Our interpretation is that at the lens induction stage, where FGFR1 and FGFR2 are crucial, their primary function operates through Ras signaling to promote cell survival. Thus, either constitutively active KRAS or the direct suppression of apoptosis by deleting Bak and Bax is sufficient to rescue lens induction. This rescue enables the subsequent differentiation of lens progenitor cells, a process for which FGFR3 and FGFR4 are sufficient to support.

(2) The authors make the argument that deleting all four FGFRs prevented lens induction but that the deletion of only FGFR1 and FGFR2 did not. Part of this argument is the retention of FOXE3 expression, αA-crystallin expression, and PROX1 expression in the FGFR1/2 double mutants. However, in Figure 1E, and Figure 1F, the staining of the double mutant lens tissue with FOXE3, αA-crystallin, and PROX1 is unconvincing. However, the retention of FOXE3 expression in the FGFR1/FGFR2 double mutants was previously demonstrated in Garcia et al 2011. Also, there needs to be an enlargement or inset to demonstrate the retention of pSMAD in the quadruple FGFR mutants in Figure 1D.

We have updated Figure 1E with a clearer image of FOXE3 staining to better illustrate FOXE3 expression in the FGFR1/2 double mutants. It seems there may have been a misunderstanding regarding our claims about αA-crystallin and PROX1. To clarify, our observation is that both αA-crystallin and PROX1 are lost in the FGFR1/2 double mutants, which we believe is clearly demonstrated in Figure 1F. Additionally, we have added inserts to Figure 1D to highlight the retention of pSMAD.

(3) Do the authors suggest that GRB2 is required for RAS activation and ultimately ERK activation? If so, do the authors suggest that ERK activation is not required for FGFR-signaling to mediate lens induction? This would follow considering that the GRB2 deficient lenses lack a problem with lens induction.

We do believe that GRB2 is required for RAS-ERK signaling activation; however, ERK activation is not absolutely required for lens induction. This conclusion is consistent with our previous study, which showed that deletion of ERK1/2 did not prevent lens induction (Garg et al. eLife 2020;9:e51915), as well as with our current findings demonstrating that the GRB2-deficient mutant is still capable of supporting lens induction.

(4) The increase in p-Shc is only slightly higher in the Cre FGFR1f/f FGFR2r/LR than in the FGFR1f/Δfrs FGFR2f/f. Can the authors provide quantification?

pShc quantification is now provided in Fig. 7B.

(5) The authors have not shown directly that Shc1 binds to the juxtamembrane region of either Fgfr1 or Fgfr2.

It is not yet clear whether Shc1 directly binds to the juxtamembrane region of FGFR1 or FGFR2, as it may also be recruited indirectly. We acknowledge this as an important question that warrants further investigation in future studies.

(6) The authors have used the Le-Cre strain for all of their lens deletion experiments. Previous work has documented that the Le-Cre transgene can cause lens defects independent of any floxed alleles in both homozygous and hemizygous states on some genetic backgrounds (Dora et al., 2014 PLoS One 9:e109193 and Lam et al., Human Genomics 2019 13(1):10. Are the controls used in these experiments Le-Cre hemizygotes?

As stated in the Method section, Le-Cre only or Le-Cre and heterozygous flox mice were used as controls.  

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

Weaknesses

There are only a few minor weaknesses that need to be addressed.

(1) The point could be made in the Discussion that since Le-Cre depends on Pax6 placodal expression, it is challenging to evaluate the impact of deletion of the four Fgfrs on the expression of Pax6 (since Pax6 needs to be activated prior to achieving Fgfr deletion). A different Cre line (e.g. a Cre which is expressed in the surface ectoderm prior to lens placode formation) could help partially address this question, although it may not be able to comment on the requirement of the Fgfrs specifically in the lens ectoderm. Thus, it will be prudent to mention this in the discussion.

We have added the caveat regarding the Le-Cre dependency on Pax6 expression to the discussion.

(2) Referring to Jag1 among "lens-specific markers" (page 5) is debatable, I suggest changing it along the lines of "the expected upregulation of Jag1 in lens vesicle".

The wording has been changed as suggested.  

(3) The Abstract could be modified to clearly convey the existing knowledge gap and the key findings of the present study. As it stands now, it is a bit all over the place.

The abstract has been revised.  

(4) Some typos in the manuscript need to be fixed.

e.g. "...yet its molecular mechanism remains largely resolved" - unresolved?, "...in the development lens" - in the developing lens?, In Fig. 4 legend, "(B) Grb2 mutants Grb2 mutants displayed...", etc.

These typos have been corrected.

Reviewer #2 (Recommendations for the authors):

My specific suggestions are shown below.

(1) The authors need to describe the role of Shc1 in FGF signaling and vertebrate lens development, by citing previous publications in the introduction.

We have detailed previous studies on the role of Shc in FGF signaling in the Introduction and discussed its function in the vertebrate lens in the Discussion section.

(2) Figure 1B bottom panels: Inset images seem to be missing, although frames and arrowheads are there. Please check them.

The inset images were correctly placed.

(3) Results (page 5, line 13): The authors mentioned "Sox2 expression remained at basal levels". Since Figure 1B indicates that Sox2 expression fails to be upregulated in FGFR1/2 mutant lens placode in contrast to Pax6, it is better to clearly mention the failure in upregulation of Sox2 expression in the FGFR1/2 mutants.

This sentence has been rewritten as suggested.  

(4) Results (page 6, line 8): The authors mentioned "we observed .... expression of Foxe3 in ...mutant lens cells (Figure 1E, arrows). However, Foxe3-expressing lens cells are a very small population in Figure 1E. It is important to state the decreased number of Foxe3-expressing lens cells in FGFR1/2 mutants. In addition, I would like to request the authors to show histograms indicating sample size and statistical analysis for marker expression: Foxe3 (Figure 1E), Prox1 and aA-crystallin (Fig. 1F), cyclin D1 and TUNEL (Fig. 1G) and pmTOR and pS6 (Supplementary figure 1B).

We added a statement indicating that the number of Foxe3-expressing cells is reduced in FGFR1/2 mutants, which is now quantified in Fig. 1H. Quantifications for Cyclin D1 and TUNEL are now shown in Fig. 1I and J, respectively. However, we chose not to quantify Prox1, αA-crystallin, pmTOR, and pS6, as the FGFR1/2 mutants showed no staining for these markers.

(5) Results (page 6, line 19- page 7, line 6): The authors showed that inducible expression of constitutive active Kras, KrasG12D, using Le-Cre, recovered lens size to the half level of wild-type control. However, in the lens of mice with Le-Cre; FGFR1/2f/f; LSL-KrasG12D, pERK was detected in the most posterior edge of the lens fiber core, whereas pERK was detected in the broader area of the lens in control. Furthermore, pMEK was detected in the whole lens of mice with Le-Cre; FGFR1/2f/f; and LSL-KrasG12D, whereas pMEK was detected only in the lens epithelial cells at the equator. So, the spatial profile of pERK and pMEK expression was different from those of wild-type, although the authors observed that Prox1 and Crystallin expression are normally induced in the lens of mice with Le-Cre; FGFR1/2f/f; LSL-KrasG12D. I wonder whether the lens normally develops in mice with Le-Cre; LSL-KrasG12D? Is the lens growth enhanced in mice with Le-Cre; LSL-KrasG12D? Please add the panels of mice with Le-Cre; LSL-KrasG12D in Figure 2B and 2C. In addition, I wonder whether apoptosis is suppressed in the lens of mice with Le-Cre; FGFR1/2f/f; LSL-KrasG12D?

As we previously reported (Developmental Biology 355, 2011, 12–20), Le-Cre; LSL-KrasG12D did not lead to enhanced lens growth. While we agree that including images of Le-Cre; LSL-KrasG12D as controls in Fig. 2B and C and evaluating apoptosis in Le-Cre; FGFR1/2f/f; LSL-KrasG12D mutants would be appropriate, we regretfully no longer have these animals available to conduct these experiments.

(6) Results (page 11, line 15): the PCR genotyping image of Fig. 6C seems to be missing.

The PCR genotyping image was correctly placed below Fig. 6B. 

(7) Results (page 11, lines 15-20): there is no citation of Figure 6D in the results section.

The citation for Fig. 6D is added in the results section.

(8) Figures 5H, 6H, and 7A: Western blotting of some of the pERK, ERK lanes is missing.

These western blots all have pERK/ERK overlay images.

(9) Figure 7A, western blotting data on pShc levels are important to suggest the involvement of Shc1 in Frs2-independent Grb2 activation by FGF stimulation. Please provide the histogram for statistical analysis.

pShc quantification is now provided in Fig. 7B.

(10) There is no citation of Figure 7D, E, and F in the results section. Please add them.

These citations have been added.

(11) Figures 7E, and 7F: The authors showed that lens morphology and lens size evaluation in genetic combinations: control, Frs2/Shc1 KD, Frs2/Shp2 KD, and Frs2/Shp2/Shc1 KD. However, I would like to request the authors to show more detailed data in these genetic combinations, for example, pERK, foxe3, Maf, Prox1, Jag1, p57, cyclin D3, g-crystallin, and TUNEL.

Unfortunately, we no longer have these mutant mice to perform these detailed staining.  

Reviewer #3 (Recommendations for the authors):

(1) The figure legend for Figure 2 lists (G) twice. The second (G) should be (H). Also, in Figures 2G and H there is no indication as to what stage lenses were used for the TUNEL and size analyses. I assume that it was E13.5, but it should be explicitly stated.

The figure labeling has been corrected and the stage added to the figure legend.

(2) In Figure 4 A the label should be gamma-crystallin rather than r-crystallin.

The figure labeling has been corrected.

(3) In Figure 6 D, I believe that the immunolabeling for Maf and Foxe3 are reversed. The Maf should be red as it is in the fibers and the Foxe3 should be green as it is epithelial.

The figure labeling has been corrected.

(4) In Figure 6C I believe that the labels for the WT and YF alleles on the western blot are reversed.

The YF PCR band was designed to be larger than WT, so the labeling was correct as is.

(5) In Figure 6F I believe that the labels for WT and CS on the western blot are reversed.

The figure labeling has been corrected.

(6) In Supplemental figure 2 there are no genotype labels for the TUNEL bar graph.

The figure labeling has been added.

eLife Assessment

In this valuable report, the authors investigated the effect of mitochondrial transplantation on post-cardiac arrest myocardial dysfunction (PAMD), which is associated with mitochondrial dysfunction. They convincingly demonstrated that mitochondrial transplantation enhanced cardiac function and increased survival rates after the return of spontaneous circulation (ROSC). They have also shown that myocardial tissues with transplanted mitochondria exhibited increased mitochondrial complex activity, higher ATP levels, reduced cardiomyocyte apoptosis, and lower myocardial oxidative stress post-ROSC.

Reviewer #1 (Public review):

Summary:

In this study, the authors investigate the effect of mitochondrial transplantation on post-cardiac arrest myocardial dysfunction (PAMD), which is associated with mitochondrial dysfunction. The authors demonstrate that mitochondrial transplantation enhances cardiac function and increases survival rates after the return of spontaneous circulation (ROSC). Mechanistically, they found that myocardial tissues with transplanted mitochondria exhibit increased mitochondrial complex activity, higher ATP levels, reduced cardiomyocyte apoptosis, and lower myocardial oxidative stress post-ROSC.

Strengths:

Previous studies have reported that mitochondrial transplantation can improve myocardial recovery after regional ischemia, but its potential for treating myocardial injury following cardiac arrest has not been tested yet. Therefore, the findings are somewhat novel. Remarkably, the increased survival in mitochondria treated group post ROSC is very promising and highlights its translational potential.

Comments on revisions:

My concerns are adequately addressed.

Reviewer #3 (Public review):

In this manuscript titled "Transplantation of exogenous mitochondria mitigates myocardial dysfunction after cardiac arrest", Zhen Wang et al. report that exogenous mitochondrial transplantation can enhance myocardial function and survival rates. It limits mitochondrial morphology impairment, boosts complexes II and IV activity, and increases ATP levels. Additionally, mitochondrial therapy reduces oxidative stress, lessens myocardial injury, and improves PAMD after cardiopulmonary resuscitation. The results of this manuscript clearly demonstrate that mitochondrial transplantation can effectively improve PAMD after cardiopulmonary resuscitation, highlighting its significant scientific and clinical value. The findings shown in this manuscript are interesting to the readers. However, further experiments are needed to confirm this conclusion. In addition, the results should be rewritten to describe and discuss the relevant data in detail.

Major comments from the original round of review:

(1) Can isolated mitochondria be transported to cultured cardiomyocytes, such as H9C2 cells, in vitro?

(2) The description of results in the manuscript is too simple. It lacks detail on the rationale behind the experiments and the significance of the data.

(3) The authors demonstrate that mitochondrial transplantation reduces cardiomyocyte apoptosis. Therefore, Western blot analysis of apoptosis-related caspases could be provided for further confirmation.

(4) Do donor mitochondria fuse with recipient mitochondria? Relevant experiments and data should be provided to address this question.

(5) In Figure 5A, the histograms are not labeled with the specific experimental groups.

Comments on revisions:

The revised manuscript quality has been improved, and most of my concerns were addressed and resolved.

eLife Assessment

These useful findings assigned a novel functional implication of histone acylation, crotonylation. Although the mechanistic insights have been provided in great detail regarding the role of the YEATS2-GCDH axis in modulating EMT in HNC, the strength of evidence for the manuscript is incomplete. The patient cohort is very small, with just 10 patients; to establish a significant result the cohort size should be increased. Furthermore, the functional implication of p300 is also to be looked into.

Reviewer #1 (Public review):

Summary:

This manuscript investigates a mechanism between the histone reader protein YEATS2 and the metabolic enzyme GCDH, particularly in regulating epithelial-to-mesenchymal transition (EMT) in head and neck cancer (HNC).

Strengths:

Great detailing of the mechanistic aspect of the above axis is the primary strength of the manuscript.

Weaknesses:

Several critical points require clarification, including the rationale behind EMT marker selection, the inclusion of metastasis data, the role of key metabolic enzymes like ECHS1, and the molecular mechanisms governing p300 and YEATS2 interactions.

Major Comments:

(1) The title, "Interplay of YEATS2 and GCDH mediates histone crotonylation and drives EMT in head and neck cancer," appears somewhat misleading, as it implies that YEATS2 directly drives histone crotonylation. However, YEATS2 functions as a reader of histone crotonylation rather than a writer or mediator of this modification. It cannot itself mediate the addition of crotonyl groups onto histones. Instead, the enzyme GCDH is the one responsible for generating crotonyl-CoA, which enables histone crotonylation. Therefore, while YEATS2 plays a role in recognizing crotonylation marks and may regulate gene expression through this mechanism, it does not directly catalyse or promote the crotonylation process.

(2) The study suggests a link between YEATS2 and metastasis due to its role in EMT, but the lack of clinical or pre-clinical evidence of metastasis is concerning. Only primary tumor (PT) data is shown, but if the hypothesis is that YEATS2 promotes metastasis via EMT, then evidence from metastatic samples or in vivo models should be included to solidify this claim.

(3) There seems to be some discrepancy in the invasion data with BICR10 control cells (Figure 2C). BICR10 control cells with mock plasmids, specifically shControl and pEGFP-C3 show an unclear distinction between invasion capacities. Normally, we would expect the control cells to invade somewhat similarly, in terms of area covered, within the same time interval (24 hours here). But we clearly see more control cells invading when the invasion is done with KD and fewer control cells invading when the invasion is done with OE. Are these just plasmid-specific significant effects on normal cell invasion? This needs to be addressed.

(4) In Figure 3G, the Western blot shows an unclear band for YEATS2 in shSP1 cells with YEATS2 overexpression condition. The authors need to clearly identify which band corresponds to YEATS2 in this case.

(5) In ChIP assays with SP1, YEATS2 and p300 which promoter regions were selected for the respective genes? Please provide data for all the different promoter regions that must have been analysed, highlighting the region where enrichment/depletion was observed. Including data from negative control regions would improve the validity of the results.

(6) The authors establish a link between H3K27Cr marks and GCDH expression, and this is an already well-known pathway. A critical missing piece is the level of ECSH1 in patient samples. This will clearly delineate if the balance shifted towards crotonylation.

(7) The p300 ChIP data on the SPARC promoter is confusing. The authors report reduced p300 occupancy in YEATS2-silenced cells, on SPARC promoter. However, this is paradoxical, as p300 is a writer, a histone acetyltransferase (HAT). The absence of a reader (YEATS2) shouldn't affect the writer (p300) unless a complex relationship between p300 and YEATS2 is present. The role of p300 should be further clarified in this case. Additionally, transcriptional regulation of SPARC expression in YEATS2 silenced cells could be analysed via downstream events, like Pol-II recruitment. Assays such as Pol-II ChIP-qPCR could help explain this.

(8) The role of GCDH in producing crotonyl-CoA is already well-established in the literature. The authors' hypothesis that GCDH is essential for crotonyl-CoA production has been proven, and it's unclear why this is presented as a novel finding. It has been shown that YEATS2 KD leads to reduced H3K27cr, however, it remains unclear how the reader is affecting crotonylation levels. Are GCDH levels also reduced in the YEATS2 KD condition? Are YEATS2 levels regulating GCDH expression? One possible mechanism is YEATS2 occupancy on GCDH promoter and therefore reduced GCDH levels upon YEATS2 KD. This aspect is crucial to the study's proposed mechanism but is not addressed thoroughly.

(9) The authors should provide IHC analysis of YEATS2, SPARC alongside H3K27cr and GCDH staining in normal vs. tumor tissues from HNC patients.

Reviewer #2 (Public review):

Summary:

The manuscript emphasises the increased invasive potential of histone reader YEATS2 in an SP1-dependent manner. They report that YEATS2 maintains high H3K27cr levels at the promoter of EMT-promoting gene SPARC. These findings assigned a novel functional implication of histone acylation, crotonylation.

Concerns:

(1) The patient cohort is very small with just 10 patients. To establish a significant result the cohort size should be increased.

(2) Figure 4D compares H3K27Cr levels in tumor and normal tissue samples. Figure 1G shows overexpression of YEATS2 in a tumor as compared to normal samples. The loading control is missing in both. Loading control is essential to eliminate any disparity in protein concentration that is loaded.

(3) Figure 4D only mentions 5 patient samples checked for the increased levels of crotonylation and hence forms the basis of their hypothesis (increased crotonylation in a tumor as compared to normal). The sample size should be more and patient details should be mentioned.

(4) YEATS2 maintains H3K27Cr levels at the SPARC promoter. The p300 is reported to be hyper-activated (hyperautoacetylated) in oral cancer. Probably, the activated p300 causes hyper-crotonylation, and other protein factors cause the functional translation of this modification. The authors need to clarify this with a suitable experiment.

(5) I do not entirely agree with using GAPDH as a control in the western blot experiment since GAPDH has been reported to be overexpressed in oral cancer.

(6) The expression of EMT markers has been checked in shControl and shYEATS2 transfected cell lines (Figure 2A). However, their expression should first be checked directly in the patients' normal vs. tumor samples.

(7) In Figure 3G, knockdown of SP1 led to the reduced expression of YEATS2 controlled gene Twist1. Ectopic expression of YEATS2 was able to rescue Twist1 partially. In order to establish that SP1 directly regulates YEATS2, SP1 should also be re-introduced upon the knockdown background along with YEATS2 for complete rescue of Twist1 expression.

(8) In Figure 7G, the expression of EMT genes should also be checked upon rescue of SPARC expression.

Author response:

Public Reviews:

Reviewer #1 (Public review):

Summary:

This manuscript investigates a mechanism between the histone reader protein YEATS2 and the metabolic enzyme GCDH, particularly in regulating epithelial-to-mesenchymal transition (EMT) in head and neck cancer (HNC).

Strengths:

Great detailing of the mechanistic aspect of the above axis is the primary strength of the manuscript.

Weaknesses:

Several critical points require clarification, including the rationale behind EMT marker selection, the inclusion of metastasis data, the role of key metabolic enzymes like ECHS1, and the molecular mechanisms governing p300 and YEATS2 interactions.

We would like to sincerely thank the reviewer for the detailed, in-depth, and positive response. We are committed to implementing constructive revisions to the manuscript to address the reviewer’s concerns effectively.

Major Comments:

(1) The title, "Interplay of YEATS2 and GCDH mediates histone crotonylation and drives EMT in head and neck cancer," appears somewhat misleading, as it implies that YEATS2 directly drives histone crotonylation. However, YEATS2 functions as a reader of histone crotonylation rather than a writer or mediator of this modification. It cannot itself mediate the addition of crotonyl groups onto histones. Instead, the enzyme GCDH is the one responsible for generating crotonyl-CoA, which enables histone crotonylation. Therefore, while YEATS2 plays a role in recognizing crotonylation marks and may regulate gene expression through this mechanism, it does not directly catalyse or promote the crotonylation process.

We thank the reviewer for raising this concern. As stated by the reviewer, YEATS2 functions as a reader protein, capable of recognizing histone crotonylation marks and assisting in the addition of this mark to nearby histone residues, possibly by assisting the recruitment of the writer protein for crotonylation. Our data indicates the involvement of YEATS2 in the recruitment of writer protein p300 on the promoter of the SPARC gene, making YEATS2 a regulatory factor responsible for the addition of crotonyl marks in an indirect manner. Thus, we have decided to make changes in the title by replacing the word “mediates” with “regulates”. Therefore, the updated title can be read as: “Interplay of YEATS2 and GCDH regulates histone crotonylation and drives EMT in head and neck cancer”.

(2) The study suggests a link between YEATS2 and metastasis due to its role in EMT, but the lack of clinical or pre-clinical evidence of metastasis is concerning. Only primary tumor (PT) data is shown, but if the hypothesis is that YEATS2 promotes metastasis via EMT, then evidence from metastatic samples or in vivo models should be included to solidify this claim.

We appreciate the reviewer’s suggestion. Here, we would like to state that the primary aim of this study was to delineate the molecular mechanisms behind the role of YEATS2 in maintaining histone crotonylation at the promoter of genes that favour EMT in head and neck cancer. We have dissected the importance of histone crotonylation in the regulation of gene expression in head and neck cancer in great detail, having investigated the upstream and downstream molecular players involved in this process that promote EMT. Moreover, with the help of multiple phenotypic assays, such as Matrigel invasion, wound healing, and 3D invasion assays, we have shown the functional importance of YEATS2 in promoting EMT in head and neck cancer cells. Since EMT is known to be a prerequisite process for cancer cells undergoing metastasis(1), the evidence of YEATS2 being associated with EMT demonstrates a potential correlation of YEATS2 with metastasis. However, as part of the revision, we will use publicly available patient data to investigate the direct association of YEATS2 with metastasis by checking the expression of YEATS2 between different grades of head and neck cancer, as an increase in tumor grade is often correlated with the incidence of metastasis(2).

(3) There seems to be some discrepancy in the invasion data with BICR10 control cells (Figure 2C). BICR10 control cells with mock plasmids, specifically shControl and pEGFP-C3 show an unclear distinction between invasion capacities. Normally, we would expect the control cells to invade somewhat similarly, in terms of area covered, within the same time interval (24 hours here). But we clearly see more control cells invading when the invasion is done with KD and fewer control cells invading when the invasion is done with OE. Are these just plasmid-specific significant effects on normal cell invasion? This needs to be addressed.

We appreciate the reviewer for the thorough evaluation of the manuscript. The figure panels in question, Figure 2B and 2C, represent two different experiments performed independently, the invasion assay performed after knockdown and overexpression of YEATS2, respectively. We would like to clarify that both panels represent results that are distinct and independent of each other and that the method used to knockdown or overexpress YEATS2 is also different. As stated in the Materials and Methods section, the knockdown is performed using lentivirus-mediated transfection (transduction) of cells, on the other hand, the overexpression is done using standard method of transfection by directly mixing transfection reagent and the respective plasmids, prior to the addition of this mix to the cells. The difference in the experimental conditions in these two experiments might have attributed to the differences seen in the controls as observed previously(3). Hence, we would like to state that the results of figure panels Figure 2B and Figure 2C should be evaluated independently of each other.

(4) In Figure 3G, the Western blot shows an unclear band for YEATS2 in shSP1 cells with YEATS2 overexpression condition. The authors need to clearly identify which band corresponds to YEATS2 in this case.

The two bands seen in the shSP1+pEGFP-C3-YEATS2 condition correspond to the endogenous YEATS2 band (lower band, indicated by * in the shControl lane) and YEATS2-GFP band (upper band, corresponding to overexpressed YEATS2-GFP fusion protein, which has a higher molecular weight). To avoid confusion, the endogenous band will be highlighted (marked by *) in the lane representing the shSP1+pEGFP-C3-YEATS2 condition in the revised version of the manuscript.

(5) In ChIP assays with SP1, YEATS2 and p300 which promoter regions were selected for the respective genes? Please provide data for all the different promoter regions that must have been analysed, highlighting the region where enrichment/depletion was observed. Including data from negative control regions would improve the validity of the results.

Throughout our study, we have performed ChIP-qPCR assays to check the binding of SP1 on YEATS2 and GCDH promoter, and to check YEATS2 and p300 binding on SPARC promoter. Using transcription factor binding prediction tools and luciferase assays, we selected multiple sites on the YEATS2 and GCDH promoter to check for SP1 binding. The results corresponding to the site that showed significant enrichment were provided in the manuscript. The region of SPARC promoter in YEATS2 and p300 ChIP assay was selected on the basis of YEATS2 enrichment found in the YEATS2 ChIP-seq data. We will provide data for all the promoter regions investigated (including negative controls) in the revised version of the manuscript.

(6) The authors establish a link between H3K27Cr marks and GCDH expression, and this is an already well-known pathway. A critical missing piece is the level of ECSH1 in patient samples. This will clearly delineate if the balance shifted towards crotonylation.

We thank the reviewer for their valuable suggestion. To support our claim, we had checked the expression of GCDH and ECHS1 in TCGA HNC RNA-seq data (provided in Figure 4—figure supplement 1A and B) and found that GCDH showed increase while ECHS1 showed decrease in tumor as compared to normal samples. We hypothesized that higher GCDH expression and decreased ECHS1 expression might lead to an increase in the levels of crotonylation in HNC. To further substantiate our claim, we will check the abundance of ECHS1 in HNC patient samples as part of the revision.

(7) The p300 ChIP data on the SPARC promoter is confusing. The authors report reduced p300 occupancy in YEATS2-silenced cells, on SPARC promoter. However, this is paradoxical, as p300 is a writer, a histone acetyltransferase (HAT). The absence of a reader (YEATS2) shouldn't affect the writer (p300) unless a complex relationship between p300 and YEATS2 is present. The role of p300 should be further clarified in this case. Additionally, transcriptional regulation of SPARC expression in YEATS2 silenced cells could be analysed via downstream events, like Pol-II recruitment. Assays such as Pol-II ChIP-qPCR could help explain this.

Using RNA-seq and ChIP-seq analyses, we have shown that YEATS2 affects the expression of several genes by regulating the level of histone crotonylation at gene promoters globally. The histone writer p300 is a promiscuous acyltransferase protein that has been shown to be involved in the addition of several non-acetyl marks on histone residues, including crotonylation(4). Our data provides evidence for the dependency of the writer p300 on YEATS2 in mediating histone crotonylation, as YEATS2 downregulation led to decreased occupancy of p300 on the SPARC promoter (Figure 5F). However, the exact mechanism of cooperativity between YEATS2 and p300 in maintaining histone crotonylation remains to be investigated. To address the reviewer’s concern, we will perform various experiments to delineate the molecular mechanism pertaining to the association of YEATS2 with p300 in regulating histone crotonylation. Following are the experiments that will be performed:

(a) Co-immunoprecipitation experiments to check the physical interaction between YEATS2 and p300.

(b) We will check H3K27cr levels on the SPARC promoter and SPARC expression in p300-depleted HNC cells.

(c) Rescue experiments to check if the decrease in p300 occupancy on the SPARC promoter can be compensated by overexpressing YEATS2.

(d) As suggested by the reviewer, Pol-II ChIP-qPCR at the promoter of SPARC will be performed in YEATS2-silenced cells to explain the mode of transcriptional regulation of SPARC expression by YEATS2.

(8) The role of GCDH in producing crotonyl-CoA is already well-established in the literature. The authors' hypothesis that GCDH is essential for crotonyl-CoA production has been proven, and it's unclear why this is presented as a novel finding. It has been shown that YEATS2 KD leads to reduced H3K27cr, however, it remains unclear how the reader is affecting crotonylation levels. Are GCDH levels also reduced in the YEATS2 KD condition? Are YEATS2 levels regulating GCDH expression? One possible mechanism is YEATS2 occupancy on GCDH promoter and therefore reduced GCDH levels upon YEATS2 KD. This aspect is crucial to the study's proposed mechanism but is not addressed thoroughly.

The source for histone crotonylation, crotonyl-CoA, can be produced by several enzymes in the cell, such as ACSS2, GCDH, ACOX3, etc(5). Since metabolic intermediates produced during several cellular pathways in the cell can act as substrates for epigenetic factors, we wanted to investigate if such an epigenetic-metabolism crosstalk existed in the context of YEATS2. As described in the manuscript, we performed GSEA using publicly available TCGA RNA-seq data and found that patients with higher YEATS2 expression also showed a high correlation with expression levels of genes involved in the lysine degradation pathway, including GCDH. Since the preferential binding of YEATS2 with H3K27cr and the role of GCDH in producing crotonyl-CoA was known(6,7), we hypothesized that higher H3K27cr in HNC could be a result of both YEATS2 and GCDH. We found that the presence of GCDH in the nucleus of HNC cells is correlated to higher H3K27cr abundance, which could be a result of excess levels of crotonyl-CoA produced via GCDH. We also found a correlation between H3K27cr levels and YEATS2 expression, which could arise due to YEATS2-mediated preferential maintenance of crotonylation. This states that although being a reader protein, YEATS2 is affecting the promoter H3K27cr levels, possibly by helping in the recruitment of p300 (as shown in Figure 5F). Thus, YEATS2 and GCDH are both responsible for the regulation of histone crotonylation-mediated gene expression in HNC.

We did not find any evidence of YEATS2 regulating the expression of GCDH in HNC cells. However, we found that YEATS2 downregulation reduced the nuclear pool of GCDH in head and neck cancer cells (Figure 7F). This suggests that YEATS2 not only regulates histone crotonylation by affecting promoter H3K27cr levels (with p300), but also by affecting the nuclear localization of crotonyl-CoA producing GCDH. Also, we observed that the expression of YEATS2 and GCDH are regulated by the same transcription factor SP1 in HNC. We found that the transcription factor SP1 binds to the promoter of both genes, and its downregulation led to a decrease in their expression (Figure 3 and Figure 7).

We would like to state that the relationship between YEATS2 and the nuclear localization of GCDH, as well as the underlying molecular mechanism, remains unexplored and presents an open question for future investigation.

(9) The authors should provide IHC analysis of YEATS2, SPARC alongside H3K27cr and GCDH staining in normal vs. tumor tissues from HNC patients.

We thank the reviewer for their suggestion. We are consulting our clinical collaborators to assess the feasibility of including this IHC analysis in our revision and will make every effort to incorporate it.

Reviewer #2 (Public review):

Summary:

The manuscript emphasises the increased invasive potential of histone reader YEATS2 in an SP1-dependent manner. They report that YEATS2 maintains high H3K27cr levels at the promoter of EMT-promoting gene SPARC. These findings assigned a novel functional implication of histone acylation, crotonylation.

We thank the reviewer for the constructive comments. We are committed to making beneficial changes to the manuscript in order to alleviate the reviewer’s concerns.

Concerns:

(1) The patient cohort is very small with just 10 patients. To establish a significant result the cohort size should be increased.

We thank the reviewer for this suggestion. We will increase the number of patient samples to assess the levels of YEATS2 and H3K27cr in normal vs. tumor samples.

(2) Figure 4D compares H3K27Cr levels in tumor and normal tissue samples. Figure 1G shows overexpression of YEATS2 in a tumor as compared to normal samples. The loading control is missing in both. Loading control is essential to eliminate any disparity in protein concentration that is loaded.

In Figures 1G and 4D, we have used Ponceau S staining as a control for equal loading. Ponceau S staining is frequently used as an alternative for housekeeping genes like GAPDH as a control for protein loading(8). It avoids the potential for variability in housekeeping gene expression. However, it may be less quantitative than using housekeeping proteins. To address the reviewer’s concern, we will probe with an antibody against a house keeping gene as a loading control in the revised figures, provided its expression remains stable across the conditions tested.

(3) Figure 4D only mentions 5 patient samples checked for the increased levels of crotonylation and hence forms the basis of their hypothesis (increased crotonylation in a tumor as compared to normal). The sample size should be more and patient details should be mentioned.

A total of 9 samples were checked for H3K27cr levels (5 of them are included in Figure 4D and rest included in Figure 4—figure supplement 1D). However, as a part of the revision, we will check the H3K27cr levels in more patient samples.

(4) YEATS2 maintains H3K27Cr levels at the SPARC promoter. The p300 is reported to be hyper-activated (hyperautoacetylated) in oral cancer. Probably, the activated p300 causes hyper-crotonylation, and other protein factors cause the functional translation of this modification. The authors need to clarify this with a suitable experiment.

In our study, we have shown that p300 is dependent on YEATS2 for its recruitment on the SPARC promoter. As a part of the revision, we propose the following experiments to further substantiate the role of p300 in YEATS2-mediated gene regulation:

(a) Co-immunoprecipitation experiments to check the physical interaction between YEATS2 and p300.

(b) We will check H3K27cr levels on the SPARC promoter and SPARC expression in p300-depleted HNC cells.

(c) Rescue experiments to check if the decrease in p300 occupancy on the SPARC promoter can be compensated by overexpressing YEATS2.

(d) Pol-II ChIP-qPCR at the promoter of SPARC will be performed in YEATS2-silenced cells to explain the mode of transcriptional regulation of SPARC expression by YEATS2.

(5) I do not entirely agree with using GAPDH as a control in the western blot experiment since GAPDH has been reported to be overexpressed in oral cancer.

We would like to clarify that GAPDH was not used as a loading control for protein expression comparisons between normal and tumor samples. GAPDH was used as a loading control only in experiments using head and neck cancer cell lines where shRNA-mediated knockdown or overexpression was employed. These manipulations specifically target the genes of interest and are not expected to alter GAPDH expression, making it a suitable loading control in these instances.

(6) The expression of EMT markers has been checked in shControl and shYEATS2 transfected cell lines (Figure 2A). However, their expression should first be checked directly in the patients' normal vs. tumor samples.

We thank the reviewer for the suggestion. To address this, we will check the expression of EMT markers alongside YEATS2 expression in normal vs. tumor samples.

(7) In Figure 3G, knockdown of SP1 led to the reduced expression of YEATS2 controlled gene Twist1. Ectopic expression of YEATS2 was able to rescue Twist1 partially. In order to establish that SP1 directly regulates YEATS2, SP1 should also be re-introduced upon the knockdown background along with YEATS2 for complete rescue of Twist1 expression.

To address the reviewer’s concern regarding the partial rescue of Twist1 in SP1 depleted-YEATS2 overexpressed cells, we will perform the experiment as suggested by the reviewer. In brief, we will overexpress both SP1 and YEATS2 in SP1-depleted cells and then assess the expression of Twist1.

(8) In Figure 7G, the expression of EMT genes should also be checked upon rescue of SPARC expression.

We thank the reviewer for the suggestion. We will check the expression of EMT markers on YEATS2/ GCDH rescue and update Figure 7G in the revised version of the manuscript.

References

(1) T. Brabletz, R. Kalluri, M. A. Nieto and R. A. Weinberg, Nat Rev Cancer, 2018, 18, 128–134.

(2) P. Pisani, M. Airoldi, A. Allais, P. Aluffi Valletti, M. Battista, M. Benazzo, R. Briatore, S. Cacciola, S. Cocuzza, A. Colombo, B. Conti, A. Costanzo, L. Della Vecchia, N. Denaro, C. Fantozzi, D. Galizia, M. Garzaro, I. Genta, G. A. Iasi, M. Krengli, V. Landolfo, G. V. Lanza, M. Magnano, M. Mancuso, R. Maroldi, L. Masini, M. C. Merlano, M. Piemonte, S. Pisani, A. Prina-Mello, L. Prioglio, M. G. Rugiu, F. Scasso, A. Serra, G. Valente, M. Zannetti and A. Zigliani, Acta Otorhinolaryngol Ital, 2020, 40, S1–S86.

(3) J. Lin, P. Zhang, W. Liu, G. Liu, J. Zhang, M. Yan, Y. Duan and N. Yang, Elife, 2023, 12, RP87510.

(4) X. Liu, W. Wei, Y. Liu, X. Yang, J. Wu, Y. Zhang, Q. Zhang, T. Shi, J. X. Du, Y. Zhao, M. Lei, J.-Q. Zhou, J. Li and J. Wong, Cell Discov, 2017, 3, 17016.

(5) G. Jiang, C. Li, M. Lu, K. Lu and H. Li, Cell Death Dis, 2021, 12, 703.

(6) D. Zhao, H. Guan, S. Zhao, W. Mi, H. Wen, Y. Li, Y. Zhao, C. D. Allis, X. Shi and H. Li, Cell Res, 2016, 26, 629–632.

(7) H. Yuan, X. Wu, Q. Wu, A. Chatoff, E. Megill, J. Gao, T. Huang, T. Duan, K. Yang, C. Jin, F. Yuan, S. Wang, L. Zhao, P. O. Zinn, K. G. Abdullah, Y. Zhao, N. W. Snyder and J. N. Rich, Nature, 2023, 617, 818–826.

(8) I. Romero-Calvo, B. Ocón, P. Martínez-Moya, M. D. Suárez, A. Zarzuelo, O. Martínez-Augustin and F. S. de Medina, Anal Biochem, 2010, 401, 318–320.

eLife Assessment

In this important study, the authors advance our understanding of copper uptake by chalkophores and their targeted metalloproteins in Mycobacterium tuberculosis. These convincing data demonstrate that chalkophore-acquired copper is solely incorporated into the Mtb bcc:aa3 copper-iron respiratory oxidase under low copper conditions, and that chalkophore-mediated protection of the respiratory chain is critical to Mtb virulence. These findings may be leveraged for drug discovery and will be of broad interest to those studying bacterial pathogenesis.

Reviewer #1 (Public review):

Summary:

It is essential for Mycobacterium tuberculosis (Mtb) to scavenge trace metals from its host to survive. In this study, the authors explore the effects of copper limitation on Mtb. Mtb synthesizes small molecular diisonitrile lipopeptides termed chalkophores, that chelate host copper for import, whereby the copper is incorporated into Mtb metalloproteins. However, the role of chalkophores in Mtb biology and their targeted metalloproteins are unknown. This study investigates Mtb proteins that require chalkophores for copper incorporation and their effect on Mtb virulence. It is known that the nrp operon is induced by copper deprivation and encodes the synthesis of chalkophores. A genetic analysis revealed transcriptional differences for WT and Mtb∆nrp when exposed to the copper chelator tetrathiomolybdate (TTM). The authors found that copper chelation results in upregulation of genes in the chalkophore cluster as well as genes involved in the respiratory chain: specifically, components of the heme-dependent oxidase CytBD and subunits of the bcc:aa3 heme-copper oxidase. Interestingly, treatment of Mtb∆nrp with an inhibitor of the QcrB subunit of the bcc:aa3 oxidase (Q203) resulted in similar transcriptional changes. The bcc:aa3 oxidase and CytBD are functionally redundant, and while both utilize heme as a cofactor, only the first utilizes heme and copper. Utilizing Mtb∆nrp, Mtb∆cydAB and MtbΔnrpΔcydAB along with single gene complementation, the authors showed that copper starvation survival requires diisonitrile chalkophore synthesis and that copper starvation results in dysfunctional bcc:aa3 oxidase. Further genetic analysis combined with inhibitor studies indicate that bcc:aa3 oxidase is the only target impacted by copper starvation. By monitoring oxygen consumption for mutants in combination with inhibitors, the authors show that copper deprivation inhibits respiration through the bcc:aa3 oxidase. Similarly, they show that TTM or Q203 treatment inhibits ATP production in MtbΔnrpΔcydAB, but not in WT, showing that chalkophores maintain oxidative phosphorylation. Lastly, the authors compare the virulence of WT Mtb, Mtb∆nrp and MtbΔnrpΔcydAB strains in mice spleen and lung. The Mtb∆nrp strain showed mild attenuation, but virulence in MtbΔnrpΔcydAB was severely attenuated, and complementation with the chalkophore biosynthetic pathway restored Mtb virulence. These results suggest that chalkophore mediated protection of the respiratory chain is critical to Mtb virulence, and the that redundant respiratory oxidases within Mtb provides respiratory chain flexibility that may promote host adaptation.

Strengths:

Overall, the paper is very clear and well-written, with thorough and well-thought-out experimentation.

The methods are all quite standard, so there are no weaknesses identified with regard to methodology.

Reviewer #2 (Public review):

Summary:

This is a well-written manuscript that clearly demonstrates that the nrp encoded diisonitrile chalkophore is necessary for the function of the bcc-aa3 oxidase supercomplex under low copper conditions. In addition, the study demonstrates that the chlakophore is important early during infection when copper sequestration is employed by the host as a method of nutritional immunity.

Strengths:

The authors use genetic approaches including single and double mutants of chalkophore biosynthesis, and both the Mtb oxidases. They use copper chelators to restrict copper in vitro. A strength of the work was the use of a synthesized a Mtb chalkophore analogue to show chemical complementation of the mutant nrp locus. Oxphos metabolic activity was measuered by oxygen consumption and ATP levels. Importantly, the study demonstrated that chalkophore, especially in a strain lacking the secondary oxidase, was necessary for early infection and ruled out a role for adaptive immunity in the chalkophore lacking Mtb by use of SCID mice. It is interesting that after two weeks of infection and onset of adaptive immunity, the chalkophore is not required, which is consistent with the host environment switching from a copper-restricted to copper overload in phagosomes.

Weaknesses:

Most claims in the manuscript are soundly justified. The one exception is the claim that "maintenance of respiration is the only cellular target of chalkophore mediated copper acquisition." While under the in vitro conditions tested this does appear to be the case; however, it can't be ruled out that the chalkophore is important in other situations. In particular, for maintenance of the periplasmic superoxide dismustase, SodC, which is the other M. tuberculosis enzyme known to require copper.

Reviewer #3 (Public review):

Summary:

In this manuscript, the group of Glickman expands on their previous studies on the function of chalkophores during the growth of and infection by Mycobacterium tuberculosis. Previously, the group had shown that chalkophores, which are metallophores specific for the scavenging of copper, are induced by M. tuberculosis under copper deprivation conditions. Here, they show that chalkophores, under copper limiting conditions, are essential for the uptake of copper and maturation of a terminal oxidase, the heme-copper oxidase, cytochrome bcc:aa3. As M. tuberculosis has two redundant terminal oxidases, growth of and infection by M. tuberculosis is only moderated if both the chalkophores and the second terminal oxidase, cytochrome bd, are inhibited.

Strengths:

A strength of this work is that the lab-culture experiments are expanded upon with mice infection models, providing strong indications that host-inflicted copper deprivation is a condition that M. tuberculosis has adapted to for virulence.

Weaknesses:

Because the phenotype of M. tuberculosis lacking chalkophores is similar, if not identical, to using Q203, an inhibitor of cytochrome bcc:aa3, the authors propose that the copper-containing cytochrome bcc:aa3 is the only recipient of copper-uptake by chalkophores. A minor weakness of the work is that this latter conclusion is not verified under infection conditions and other copper-enzymes might still be functionally required during one or more stages of infection.

Author response:

We thank the reviewers for their careful evaluation of our manuscript and appreciate the suggestions for improvement. We will outline our planned revisions in response to these reviews.

Reviewer 2:

“The one exception is the claim that "maintenance of respiration is the only cellular target of chalkophore mediated copper acquisition." While under the in vitro conditions tested this does appear to be the case; however, it can't be ruled out that the chalkophore is important in other situations. In particular, for maintenance of the periplasmic superoxide dismutase, SodC, which is the other M. tuberculosis enzyme known to require copper.”

And

Reviewer 3:

“Because the phenotype of M. tuberculosis lacking chalkophores is similar, if not identical, to using Q203, an inhibitor of cytochrome bcc:aa3, the authors propose that the copper-containing cytochrome bcc:aa3 is the only recipient of copper-uptake by chalkophores. A minor weakness of the work is that this latter conclusion is not verified under infection conditions and other copper-enzymes might still be functionally required during one or more stages of infection.

Both comments concern the question of whether the bcc:aa3 respiratory oxidase supercomplex is the only target of chalkophore delivered copper. In culture, our experiments suggest that bcc:aa3 is the only target. The evidence for this claim is in Figure 2E and F. In 2E, we show that M. tuberculosis DctaD (a subunit of bcc:aa3) is growth impaired, copper chelation with TTM does not exacerbate that growth defect, and that a DctaDDnrp double mutant is no more sensitive to TTM than DctaD. These data indicate that role of the chalkophore in protecting against copper deprivation is absent when the bcc:aa3 oxidase is missing. Similar results were obtained with Q203 (Figure 2F). Q203 or TTM arrest growth of M. tuberculosis Dnrp, but the combination has no additional effect, indicating that when Q203 is inhibiting the bcc:aa3 oxidase, the chalkophore has no additional role. However, we agree with the reviewers that we cannot exclude the possibility that during infection, there is an additional target of chalkophore mediated Cu acquisition. We will add this caveat to the revised version of this manuscript.

eLife Assessment

This manuscript reports fundamental discoveries on how necrotic cells contribute to organ regeneration through apoptotic signalling to produce cells with non-lethal apoptotic caspase activity that contribute to the regenerated tissue. These findings will be of broad interest to those who study wound repair and tissue regeneration. The strength of the evidence is solid and has been improved in the revised version.

Reviewer #2 (Public review):

In this revised manuscript, Klemm et al., build on top of past published findings (Klemm et al., 2021) to characterize caspase activation in distal cells following necrotic tissue damage within the Drosophila wing imaginal disc. Previously in Klemm et al., 2021, the authors describe necrosis-induced-apoptosis (NiA) following the development of a genetic system to study necrosis that is caused by the expression of a constitutive active GluR1 (Glutamate/Ca2+ channel), and they discovered that the appearance of NiA cells were important for promoting regeneration.

In this manuscript, the authors investigate how tissues regenerate following necrotic cell death. They find that:

(1) the cells of the wing pouch are more likely to have non-autonomous caspase activation than other regions within the wing imaginal disc (hinge and notum),

(2) two signaling pathways that are known to be upregulated during regeneration, Wnt (wingless) and JAK/Stat signaling, act to prevent additional NiA in pouch cells, and may partially explain the region specificity,

(3) the presence of NiA (and/or NiCP) cells promotes regenerative proliferation in the late stages of regeneration,

(4) not all caspase-positive cells are cleared from the epithelium (these cells are then referred to as Necrosis-induced Caspase Positive (NiCP) cells), these NiCP cells continue to live and promote proliferation in adjacent cells,

(5) the initiator caspase Dronc is important for creating NiA/NiCP cells and for these cells to promote proliferation. Animals heterozygous for a Dronc null allele show a decrease in regeneration following necrotic tissue damage. In the revised manuscript, the authors provide improvements through additional data quantifications and text changes to better explain NiA/NiCP lineage tracing methods.

The study has the potential to be broadly interesting due to the insights into how tissues differentially respond to necrosis as compared to apoptosis to promote regeneration. The paper raises many interesting questions for future investigation, including what is the nature of the signaling between the damaged tissue and the NiA/NiCP responsive areas (such as the identity of the DAMPs)? What determines if these cells at a distance undergo apoptosis or remain viable in the tissue as caspase-positive cells? And since the authors have data that indicates that the phenomenon is distinct from 'undead cells', what are the mechanisms by which these cells promote local proliferation?

Reviewer #3 (Public review):

The manuscript "Regeneration following tissue necrosis is mediated by non-apoptotic caspase activity" by Klemm et al. is an exploration of what happens to a group of cells that experience caspase activation after necrosis occurs some distance away from the cells of interest. These experiments have been conducted in the Drosophila wing imaginal disc, which has been used extensively to study the response of a developing epithelium to damage and stress. The authors revise and refine their earlier discovery of apoptosis initiated by necrosis, here showing that many of those presumed apoptotic cells do not complete apoptosis. Thus, the most interesting aspect of the paper is the characterization of a group of cells that experience mild caspase activation in response to an unknown signal, followed by some effector caspase activation and DNA damage, but that then recover from the DNA damage, avoid apoptosis, and proliferate instead.

The authors have addressed the concerns raised, including those about drawing conclusions from RNAi knockdown without evaluating the efficacy of the knockdown, and in doing so they revised their conclusions after ascertaining that the Zfh2 RNAi was not effective.

The authors have added quantification of the imaging data throughout, which strengthens their conclusions.

In addition, the authors have revised some of the text describing the changes in EdU signal and added explanations of reagents such as the caspase sensors to clarify the experimental approaches, results, and interpretation of those results.

The authors have also addressed the minor concerns and questions about the figures and text.

A few questions remain, which the authors may choose to address.

(1) The hh>Stat92ERNAi was assessed by the 10xSTAT-GFP reporter, as shown in Fig 2 Supp1 F. The authors point out the marked reduction in GFP in the ventral part of the hinge but do not comment on the lack of change in GFP in the dorsal part of the hinge. However, the open arrowhead in Figure 2H indicating the lack of cDcp-1 signal in the hinge in the same experiment points to the dorsal hinge, where the reporter suggests no difference in JAK-STAT signaling.

(2) The data used to conclude that DRONC-DN and UAS-DIAP1 do not affect regenerative proliferation were normalized EdU intensities. As discussed in the prior review round, normalized EdU may not be a good comparison across experimental conditions given that the remainder of the disc may also have altered EdU incorporation, so this measurement may not be enough by itself to draw conclusions about regenerative proliferation. To strengthen the conclusion that regenerative proliferation is unaffected under these conditions, the authors may want to consider using a second measure such as adult wing size, PCNA, or quantitate mitoses via anti-phospho histone H3 staining.

Author response:

The following is the authors’ response to the original reviews.

Reviewer #1 (Public Review):

Summary:

In previous work, the authors described necrosis-induced apoptosis (NiA) as a consequence of induced necrosis. Specifically, experimentally induced necrosis in the distal pouch of larval wing imaginal discs triggers NiA in the lateral pouch. In this manuscript, the authors confirmed this observation and found that while necrosis can kill all areas of the disc, NiA is limited to the pouch and to some extent to the notum, but is excluded from the hinge region. Interestingly and unexpectedly, signaling by the Jak/Stat and Wg pathways inhibits NiA. Further characterization of NiA by the authors reveals that NiA also triggers regenerative proliferation which can last up to 64 hours following necrosis induction. This regenerative response to necrosis is significantly stronger compared to discs ablated by apoptosis. Furthermore, the regenerative proliferation induced by necrosis is dependent on the apoptotic pathway because RNAi targeting the RHG genes is sufficient to block proliferation. However, NiA does not promote proliferation through the previously described apoptosis-induced proliferation (AiP) pathway, although cells at the wound edge undergo AiP. Further examination of the caspase levels in NiA cells allowed the authors to group these cells into two clusters: some cells (NiA) undergo apoptosis and are removed, while others referred to as Necrosis-induced Caspase Positive (NiCP) cells survive despite caspase activity. It is the NiCP cells that repair cellular damage including DNA damage and that promote regenerative proliferation. Caspase sensors demonstrate that both groups of cells have initiator caspase activity, while only the NiA cells contain effector caspase activity. Under certain conditions, the authors were also able to visualize effector caspase activity in NiCP cells, but the level was low, likely below the threshold for apoptosis. Finally, the authors found that loss of the initiator caspase Dronc blocks regenerative proliferation, while inhibiting effector caspases by expression of p35 does not, suggesting that Dronc can induce regenerative proliferation following necrosis in a non- apoptotic manner. This last finding is very interesting as it implies that Dronc can induce proliferation in at least two ways in addition to its requirement in AiP.

Strengths:

This is a very interesting manuscript. The authors demonstrate that epithelial tissue that contains a significant number of necrotic cells is able to regenerate. This regenerative response is dependent on the apoptotic pathway which is induced at a distance from the necrotic cells. Although regenerative proliferation following necrosis requires the initiator caspase Dronc, Dronc does not induce a classical AiP response for this type of regenerative response. In future work, it will be very interesting to dissect this regenerative response pathway genetically.

Weaknesses:

No weaknesses were identified.

We thank the reviewer for their positive evaluation and kind words.

Reviewer #2 (Public Review):

Summary / Strengths:

In this manuscript, Klemm et al., build on past published findings (Klemm et al., 2021) to characterize caspase activation in distal cells following necrotic tissue damage within the Drosophila wing imaginal disc. Previously in Klemm et al., 2021, the authors describe necrosis-induced-apoptosis (NiA) following the development of a genetic system to study necrosis that is caused by the expression of a constitutive active GluR1 (Glutamate/Ca2+ channel), and they discovered that the appearance of NiA cells were important for promoting regeneration.

In this manuscript, the authors aim to investigate how tissues regenerate following necrotic cell death. They find that the cells of the wing pouch are more likely to have non-autonomous caspase activation than other regions within the wing imaginal disc (hinge and notum),two signaling pathways that are known to be upregulated during regeneration, Wnt (wingless) and JAK/Stat signaling, act to prevent additional NiA in pouch cells, and may explain the region specificity, the presence of NiA cells promotes regenerative proliferation in late stages of regeneration, not all caspase-positive cells are cleared from the epithelium (these cells are then referred to as Necrosis-induced Caspase Positive (NiCP) cells), these NiCP cells continue to live and promote proliferation in adjacent cells, the caspase Dronc is important for creating NiA/NiCP cells and for these cells to promote proliferation. Animals heterozygous for a Dronc null allele show a decrease in regeneration following necrotic tissue damage.

The study has the potential to be broadly interesting due to the insights into how tissues differentially respond to necrosis as compared to apoptosis to promote regeneration.

Weaknesses:

However, here are some of my current concerns for the manuscript in its current version:

The presence of cells with activated caspase that don't die (NiCP cells) is an interesting biological phenomenon but is not described until Figure 5. How does the existence of NiCP cells impact the earlier findings presented? Is late proliferation due to NiA, NiCP, or both? Does Wg and JAK/STAT signaling act to prevent the formation of both NiA and NiCP cells or only NiA cells? Moreover, the authors are able to specifically manipulate the wound edge (WE) and lateral pouch cells (LP), but don't show how these manipulations within these distinct populations impact regeneration. The authors provide evidence that driving UAS-mir(RHG) throughout the pouch, in the LP or the WE all decrease the amount of NiA/NiCP in Figure 3G-O, but no data on final regenerative outcomes for these manipulations is presented (such as those presented for Dronc-/+ in Fig 7M). The manuscript would be greatly enhanced by quantification of more of the findings, especially in describing if the specific manipulations that impacted NiA /NiCP cells disrupt end-point regeneration phenotypes.

We have added a line to the results to clarify that we believe the finding that some NiA likely persist as NiCP does not affect our conclusions up to this point.

We have added a statement emphasizing the results from our first paper, which demonstrate that LP>miRHG expression reduces the overall capacity to regenerate.

Quantification of the change in posterior NiA number have been added to Figure 2L to strengthen the evidence. Likewise, we have included quantification of the E2F time course presented in Figure 3A (Figure 3 – Figure supplement 1C), and quantification of the change in GC3Ai signal over time has been added to Figure 5 - Figure supplement 1D) to emphasize the perdurance of GC3Ai-positive NiA/NiCP.

How fast does apoptosis take within the wing disc epithelium? How many of the caspase(+) cells are present for the whole 48 hours of regeneration? Are new cells also induced to activate caspase during this time window? The author presented a number of interesting experiments characterizing the NiCP cells. For the caspase sensor GC3Ai experiments in Figure 5, is there a way to differentiate between cells that have maintained fluorescent CG3Ai from cells that have newly activated caspase? What is the timeline for when NiA and NiCP are specified? In addition, what fraction of NiCP cells contribute to the regenerated epithelium? Additional information about the temporal dynamics of NiA and NiCP specification/commitment would be greatly appreciated.

We have included more information concerning the kinetics of apoptotic cell removal, and how this compares to the observations we have made with NiA/NiCP in our GC3Ai experiments. Additionally, we have included a quantification of the percent of the whole wing pouch with GC3Ai signal over time (Figure 5F) as well as the distal wing pouch with GC3Ai signal over time (Figure 5 – Figure supplement 1D) to further support the idea that NiCP persist over time.

We acknowledge that our GC3Ai time course unfortunately cannot confirm whether the increase in GC3Ai signal over time is due to cells with new caspase activity or proliferating NiCP and have included this point in the discussion.

We attempted to track the lineage of NiA/NiCP into the pupal and adult wings with CasExpress and DBS, however the results of these experiments were inconsistent, and therefore we did not feel confident to include these data or draw conclusions in either direction. We are currently designing variations of these lineage trace tools in order to better track the lineage of these cells that we hope to include in a future paper.

The notum also does not express developmental JAK/STAT, yet little NiA was observed within the notum. Do the authors have any additional insights into the differential response between the pouch and notum? What makes the pouch unique? Are NiA/NiCP cells created within other imaginal discs and other tissues? Are they similarly important for regenerative responses in other contexts?

We have added a brief mention of these points to the appropriate results section to avoid further increasing the length of the discussion.

Data on the necrosis of other imaginal discs through FLP/FRT clone formation in haltere and leg discs has been added to Figure 1 Figure supplement 1J, and described in the text.

Reviewer #3 (Public Review):

The manuscript "Regeneration following tissue necrosis is mediated by non- apoptotic caspase activity" by Klemm et al. is an exploration of what happens to a group of cells that experience caspase activation after necrosis occurs some distance away from the cells of interest. These experiments have been conducted in the Drosophila wing imaginal disc, which has been used extensively to study the response of a developing epithelium to damage and stress. The authors revise and refine their earlier discovery of apoptosis initiated by necrosis, here showing that many of those presumed apoptotic cells do not complete apoptosis. Thus, the most interesting aspect of the paper is the characterization of a group of cells that experience mild caspase activation in response to an unknown signal, followed by some effector caspase activation and DNA damage, but that then recover from the DNA damage, avoid apoptosis, and proliferate instead. Many questions remain unanswered, including the signal that stimulates the mild caspase activation, and the mechanism through which this activation stimulates enhanced proliferation.

The authors should consider answering additional questions, clarifying some points, and making some minor corrections:

Major concerns affecting the interpretation of experimental results:

Expression of STAT92E RNAi had no apparent effect on the ability of hinge cells to undergo NiA, leading the authors to conclude that other protective signals must exist. However, the authors have not shown that this STAT92E RNAi is capable of eliminating JAK/STAT signaling in the hinge under these experimental conditions. Using a reporter for JAK/STAT signaling, such as the STAT-GFP, as a readout would confirm the reduction or elimination of signaling. This confirmation would be necessary to support the negative result as presented.

We have included data demonstrating our ability to knock down JAK/STAT activity in the hinge with UAS-Stat92E^RNAi (Figure 2 – Figure supplement 1E and F). Additionally, we have included a quantification of posterior NiA/NiCP with the Stat92E^RNAi (as well as wg^RNAi and Zfh-2^RNAi, Figure 2L) to strengthen our conclusion that JAK/STAT and WNT signaling acts to regulate NiA formation within the pouch.

Similarly, the authors should confirm that the Zfh2 RNAi is reducing or eliminating Zfh2 levels in the hinge under these experimental conditions, before concluding that Zfh2 does not play a role in stopping hinge cells from undergoing NiA.

We have repeated this experiment with a longer knockdown using a GAL4 driver that expresses from early larval stages until our evaluation at L3, but were unable to demonstrate a loss of Zfh-2 with IF labeling. Additionally, we have quantified posterior NiA/NiCP with a Zfh-2RNAi (Figure 2L) and do find a slight increase in NiA/NiCP number, however this change is not significant. We have altered our conclusions to reflect these new data.

EdU incorporation was quantified by measuring the fluorescence intensity of the pouch and normalizing it to the fluorescence intensity of the whole disc. However, the images show that EdU fluorescence intensity of other regions of the disc, especially the notum, varied substantially when comparing the different genetic backgrounds (for example, note the substantially reduced EdU in the notum of Figure 3 B' and B'). Indeed, it has been shown that tissue damage can lead to suppression of proliferation in the notum and elsewhere in the disc, unless the signaling that induces the suppression is altered. Therefore, the normalization may be skewing the results because the notum EdU is not consistent across samples, possibly because the damage-induced suppression of proliferation in the notum is different across the different genetic backgrounds.

To more accurately reflect the observations that we have made with the EdU assay, we have changed our terminology to indicate that the EdU signal is more localized to the damaged tissue in ablated discs, thus taking into account the relative changes across the disc, rather than referring to it as an increase in the pouch. To further strengthen our observation that damage results in a localized proliferation, we have included a quantification of the E2F time course presented in Figure 3A (Figure 3 – Figure supplement 1C), which underscores the trend observed in our EdU experiments.

The authors expressed p35 to attempt to generate "undead cells". They take an absence of mitogen secretion or increased proliferation as evidence that undead cells were not generated. However, there could be undead cells that do not stimulate proliferation non-autonomously, which could be detected by the persistence of caspase activity in cells that do not complete apoptosis. Indeed, expressing p35 and observing sustained effector caspase activation could help answer the later question of what percentage of this cell population would otherwise complete apoptosis (NiA, rescued by p35) vs reverse course and proliferate (NiCP, unaffected by p35).

In our previous work, we showed that P35 expression impairs our ability to detect effector caspases with IF-based tools. This can also be seen in Figure 4 of this work (Figure 4C and F). Given that P35 expression precludes our ability to label and assay effector caspase activity visually, and thus address the concerns outlined above, we relied on other tools such as reporters of AiP mitogens (wg-lacZ & dpp-lacZ) to assay whether NiA participate in AiP. As a functional readout, we also paired P35 expression with the EdU assay to test whether proliferation was altered by the presence of undead cells. The results discussed in Figure 4 lead us to conclude that NiA likely do not participate in the canonical AiP feedforward loop, although it is possible that these experiments generate another type of undead cell – one that utilizes a different mechanism to promote proliferation.

It is unclear if the authors' model is that the NiCP cells lead to autonomous or non-autonomous cell proliferation, or both. Could the lineage-tracing experiments and/or the experiments marking mitosis relative to caspase activity answer this question?

We have added further details to the discussion on the potential for NiA/NiCP to induce cell autonomous/non-autonomous proliferation.

Many of the conclusions rely on single images. Quantification of many samples should be included wherever possible.

We have added quantification to strengthen the results of Figures 2, 3 and 5.

Why does the reduction of Dronc appear to affect regenerative growth in females but not males?

We have repeated this regeneration scoring experiments and have increased the N for control versus droncI29 mutant males, however the results of the analysis for male wing size remain not significant, although the general trend that droncI29 wings are slightly smaller. While there could be sex-specific differences in the capacity to regenerate that contribute to this observation, it is unclear what the underlying mechanism could be.

Reviewer #1 (Recommendations for the authors):

The work in this paper is already very complete and very well worked out. The conclusions are well supported by the data in this manuscript. I do not have any experimental requests, only a few minor and formal requests/questions.

(1) Why does Diap1 overexpression not affect regenerative proliferation, whereas mir(RHG) and dronc[I29] do, given that Diap1 acts between RHG and Dronc?

We speculate on this point in the discussion section but have adjusted some of the phrasing for clarity.

(2) I assume that the authors used the cleaved Dcp-1 antibody from Cell Signaling Technologies. I recommend that the authors refer to this antibody as cDcp-1 in text and figures as this antibody specifically detects the cleaved, and thus activated form of Dcp-1, and not the uncleaved, inactive form of Dcp-1 which has a uniform expression in the discs.

Changed to cDcp-1.

(3) Line 299: Hay et al. 1994 did not show that p35 inhibits Drice and Dcp-1 (in fact, both genes were not even cloned yet). This was shown by Meier et al. 2000 and Hawkins et al. 2000. Please correct references.

Corrected.

(4) Line 574/575. Meier et al. 2000 did not show that Dronc is mono-ubiquitylated. This was shown by Kamber-Kaya et al., 2017. Please correct.

Corrected.

Reviewer #2 (Recommendations for the authors):

(1) Does domeless knockdown cause apoptosis without tissue ablation (Figures 2C-E)? Currently, the non-ablation control is not shown.

Domeless knockdown does not cause apoptosis in the absence of ablation (Added Figure 2 – Figure supplement 1A).

(2) The supplemental experiment with zfh2-RNAi is hard to interpret because there is no evidence of RNAi knockdown based on the staining with the anti-Zfh2 antibody.

As noted above, a longer zfh-2 knockdown does not appear to alter Zfh-2 protein levels. A quantification of posterior NiA/NiCP following knockdown shows a slight (non-significant) increase in posterior NiA/NiCP. Considering these new results, we have altered our interpretation within the appropriate results and discussion sections.

(3) The authors should consider adding a diagram showing where mir(RHG) and DIAP1 are in the apoptotic/caspase activation pathway (Figure 7N).

Completed, Figure 7N and 7O.

Reviewer #3 (Recommendations for the authors):

(1) Figure 2 I -The purported increase in NiA should be quantitated relative to the NiA in G across many discs.

Completed (Figure 2L)

(2) Figure 2 M - contrary to the conclusion drawn, the posterior Dcp1 does not appear different from that in the control (K). This conclusion that the NiA does not occur in the margin could be better supported with more images/quantification.

We have exchanged the image for a representative one that more clearly shows the lack of margin NiA and highlighted with an arrowhead (Figure 2K)

(3) Figure 2 supp 1 E - the "slight increase" in NiA in the pouch is relative to which control? Can this conclusion be supported by quantification?

Figure 2L now quantifies this change.

(4) Figure 2 Supp 1 D, E - these discs supposedly have Zfh2 RNAi expressed, but there appears to be no reduction in Zfh2.

We were unable to demonstrate a reduction of Zfh2, even with a longer knockdown. Considering these new data, we have altered our conclusions from the Zfh2 experiments.

(5) Figure 2 Supp 1 I - please quantitate the Dcp-1 across many discs to support the conclusion.

This is the UAS-wg experiment, which we decided to remove from the quantification given the non-specific increase in cDcp-1 throughout the disc (likely as a result from ectopic Wg expression).

(6) Figure 4 legend M - The authors conclude that the experiment indicates that "NiA promote proliferation independent of AiP". It would be more precise to say that NiA cells do not secrete AiP mitogens and do not increase the proliferation of surrounding cells when prevented from completing apoptosis. To say that the NiA-induced proliferation does not require AiP would require eliminating AiP, perhaps through reaper hid grim knockdown or mitogen knockdown.

Corrected.

Minor concerns and clarification needed:

(7) Line 61 - consider the distinction between a feed-forward loop and a positive feedback loop.

Corrected.

(8) Line 338 - it would be helpful to have a brief explanation of what the GC3Ai consists of and how it reports caspase activity.

Corrected.

(9) Line 343 - the authors should clarify by what they mean when they state GC3Ai-positive cells are "associated with" mitotic cells. Are the GC3Ai cells undergoing mitosis? Or is the increase in mitosis non-autonomous?

Adjusted. “associated with adjacent proliferative cells”.

(10) Lines 392-394 - the authors should add brief descriptions of how the Drice-Based sensor and the CasExpress function, so the readers can better understand the distinctions between these sensors and the previously mentioned sensors (anti-Dcp1 and GC3Ai). In addition, please clarify how the Gal80ts modulates the sensitivity of the CasExpress.

Descriptions of DBS and CasExpress and additional clarification provided.

(11) Line 413: How does Gal80ts suppress the background developmental caspase signal, and how does this suppression lead to NiCP cells expressing GFP?

This section has been reworded to clarify.

(12) Line 417 - which GFP label is referred to here?

This section has been reworded to clarify.

(13) Line 445 is the first mention of the CARD domain - it could be introduced more fully and explained why the DroncDN's lack of effect on proliferation excludes the CARD domain as being important.

Clarified. See also the discussion for the significance of the CARD domain as dispensable for regenerative proliferation following necrosis.

(14) Line 452 - "As mentioned" - the manuscript has not previously mentioned DIAP1 modification of the CARD domain and what that modification does. Perhaps the previous explanatory text was inadvertently removed?

Corrected.

(15) The Discussion is a lengthy list of experiments that the authors did not do or observations they were unable to make. This section could benefit from a more in-depth discussion of necrosis and the possibility that NiCP cells contribute to repair after injury across contexts and species.

We have made several changes to the discussion that elaborate on some of the points listed in the public reviews.

(16) All figures: Consider making single-channel panels grayscale to aid visualization. Also consider using color combinations that can be distinguished by color-blind readers.

We appreciate these suggestions and will consider them for future manuscripts.

(17) All figure legends - are error bars SD or SEM?

Standard deviation. Added to appropriate legends.

(18) Figure 1A,C - it would be helpful in the diagrams to note when the necrosis occurs/completes.

The endpoint of necrosis is not well defined, given the simultaneous changes that occur with regeneration. Thus, we opted to not include an indicator of when necrotic ablation ends.

(19) Figure 1B - it would be helpful to name the GAL4 drivers whose expression domain is depicted to correlate with the terms used in the text.

Completed.

(20) Figure 1 legend- what do the different colors of the arrowheads denote? The dotted lines are in R' and S', not N' and O'.

Completed.

(21) Figure 2G - the yellow dashed line is not in the same place in the two images.

Corrected.

(22) Figure 2I - what is the open arrowhead?

Completed (Figure 2I legend).

(23) Figure 3 legend - please describe what the time course is observing (EdU).

Completed.

(24) Figure 4 - please include the yellow boxes in the Dcp-1 channels.

Completed.

(25) Figure 5 F' - add the arrowheads to all the panels. The yellow arrowhead appears to be pointing to nothing.

Completed.

(27) Figure 5 legend - what is a "cytoplasmic undisturbed cell"? What is the arrowhead in G? J and J' should show the same view at different time points or different views at the same time point.

Figure legend has been corrected.

(28) Figure 5 Supp 1 would be especially helped by having more single-channel panels in grayscale.

For clarity and consistency, we chose to maintain the different color channels.

(29) Figure 5 Supp 1 D and E - It would be helpful to have higher magnification and arrows pointing to the cells of interest. Why are there TUNEL+ cells that do not have caspase activation (green)?

We have added arrowheads as suggested. We believe the disparity in TUNEL and GC3Ai signals are a result of the different sensitivities of the IF staining and the TUNEL assay.

(30) Figure 5 Supp 1 F - perhaps the arrowheads should be in all panels - they point to empty spaces with no H2Av staining in the final panel. Perhaps a higher magnification image would make the "strong overlap" of the two signals more apparent?

We have added arrowheads where appropriate.

(31) Figure 6 D-E - does the widespread GFP lineage tracing signal suggest that most cells in the repaired tissue originated from cells that once had caspases activity?

Possibly, however given that CasExpress leads to significant developmental labeling, we were unable to determine to what extent the signal in this experiment comes from NiA/NiCP activity versus developmental labeling. Note that tubGAL80ts is not present in this experiment.

(32) Writing corrections:

Line 343 "positive" is misspelled.

Completed

Line 429 - a word may be missing.

Completed

Line 639 - the word "day" may be missing.

Completed

Line 658 - what temperature was the recovery?

Completed

Lines 706-708 - were the discs incubated in 55 mL and 65 mL of liquid, or a smaller volume?

Completed

eLife Assessment

This manuscript establishes a mathematical model to estimate the key parameters that control the repopulation of planarian stem cells after sublethal irradiation as they undergo fate-switching as part of their differentiation and self-renewal process. The findings are valuable for future investigation of stem cell division in planarians. The methods are solid, integrating modeling with perturbations of key transcription factors known to be critical for cell fate decisions, but the authors have only shown that this is the case for a small number of stem cell types.

Reviewer #1 (Public review):

Summary:

This is a very creative study using modeling and measurement of neoblast dynamics to gain insight into the mechanism that allows these highly potent cells to undergo fate-switching as part of their differentiation and self-renewal process. The authors estimate growth equation parameters for expanding neoblast clones based on new and prior experimental observations. These results indicate neoblast likely undergo much more symmetric self-amplifying division than loss of the population through symmetric differentiation, in the case of clone expansion assays after sublethal irradiation. Neoblasts take on multiple distinct transcriptional fates related to their terminally differentiated cell types, and prior work indicated neoblasts have a high plasticity to switch fates in a way linked to cell cycle progression and possibly through a random process. Here, the authors explore the impact of inhibition of key transcription factors defining such states (ie "fate specifying transcription factors", FSTFs) plus measurement and modeling in the clone expansion assay, to find that inhibition of factors like zfp1 likely cause otherwise zfp1-fated neoblasts to fail to proliferate and differentiation without causing compensatory gains in other lineages. A mathematical model of this process assuming that neoblasts do not retain a memory of prior states while they proliferate, and transition across specified states can mimic the experimentally determined decreased sizes of clones following inhibition of zfp1. Complementary approaches to inhibit more than one lineage (muscle plus intestine) supports the idea that this is a more general process in planarian stem cells. These results provide an important advance for understanding the fate-switching process and its relationship to neoblast growth.

Overall I find the evidence very well presented and the study compelling. It offers an important new perspective on the key properties of neoblasts. I do have some comments to clarify the presentation and significance of the work.

Reviewer #2 (Public review):

Summary:

Cell cycle duration and cell fate choice are critical to understanding the cellular plasticity of neoblasts in planarians. In this study, Tamar et al. integrated experimental and computational approaches to simulate a model for neoblast behaviors during colony expansion.

Strengths:

The finding that "arresting differentiation into specific lineages disrupts neoblast proliferative capacities without inducing compensatory expression of other lineages" is particularly intriguing. This concept could inspire further studies on pluripotent stem cells and their application for regenerative biology.

Weaknesses:

However, the absence of a cell-cell feedback mechanism during colony growth and the likelihood of the difference needs to be clarified. Is there any difference in interpreting the results if this mechanism is considered? More explanation and discussion should be included to distinguish the stages controlled by the one-step model from those discussed in this study. Although hnf-4 and foxF have been silenced together to validate the model, a deeper understanding of the tgs-1+ cell type and the non-significant reduction of tgs-1+ neoblasts in zfp-1 RNAi colonies is necessary, considering a high neural lineage frequency.

Author response:

Reviewer #1:

Overall I find the evidence very well presented and the study compelling. It offers an important new perspective on the key properties of neoblasts. I do have some comments to clarify the presentation and significance of the work.

We thank the reviewer for the positive feedback and plan to improve the presentation of the work.

Reviewer #2:

However, the absence of a cell-cell feedback mechanism during colony growth and the likelihood of the difference needs to be clarified. Is there any difference in interpreting the results if this mechanism is considered?

We will improve the description of the model assumptions and the interpretation of the data on the basis of these assumptions.

Although hnf-4 and foxF have been silenced together to validate the model, a deeper understanding of the tgs-1+ cell type and the non-significant reduction of tgs-1+ neoblasts in zfp-1 RNAi colonies is necessary, considering a high neural lineage frequency.

We will improve the analysis of this result in light of the experimentally determined frequency of the tgs-1+ neoblast population.

eLife Assessment

This study provides evidence that single-cell multi-omics profiling can reveal key regulators of HIV-1 persistence and early immune dysregulation, particularly implicating KLF2 and Th17 cells as major players in viral reservoir dynamics. The findings are solid, supported by rigorous integration of scRNA-seq and scATAC-seq data, but are limited by sample size and lack of validation with external datasets. Overall, this work makes a valuable contribution to understanding HIV-1 immune evasion and highlights potential therapeutic targets for reservoir eradication.

Reviewer #1 (Public review):

Summary:

The authors aimed to elucidate the molecular mechanisms underlying HIV-1 persistence and host immune dysfunction in CD4+ T cells during early infection (<6 months). Using single-cell multi-omics technologies-including scRNA-seq, scATAC-seq, and single-cell multiome analyses-they characterized the transcriptional and epigenomic landscapes of HIV-1-infected CD4+ T cells. They identified key transcription factors (TFs), signaling pathways, and T cell subtypes involved in HIV-1 persistence, particularly highlighting KLF2 and Th17 cells as critical regulators of immune suppression. The study provides new insights into immune dysregulation during early HIV-1 infection and reveals potential epigenetic regulatory mechanisms in HIV-1-infected T cells.

Strengths:

The study excels through its innovative integration of single-cell multi-omics technologies, enabling detailed analysis of gene regulatory networks in HIV-1-infected cells. Focusing on early infection stages, it fills a crucial knowledge gap in understanding initial immune responses and viral reservoir establishment. The identification of KLF2 as a key transcription factor and Th17 cells as major viral reservoirs, supported by comprehensive bioinformatics analyses, provides robust evidence for the study's conclusions. These findings have immediate clinical relevance by identifying potential therapeutic targets for HIV-1 reservoir eradication.

Weaknesses:

Despite its strengths, the study has several limitations. By focusing exclusively on CD4+ T cells, the study overlooks other relevant immune cells such as CD14+ monocytes, NK cells, and B cells. Additionally, while the authors generated their own single-cell datasets, they need to validate their findings using other publicly available single-cell data from HIV-1-infected PBMCs.

Reviewer #2 (Public review):

Summary:

The authors observed gene ontologies associated with upregulated KLF2 target genes in HIV-1 RNA+ CD4 T Cells using scRNA-seq and scATAC-seq datasets from the PBMCs of early HIV-1-infected patients, showing immune responses contributing to HIV pathogenesis and novel targets for viral elimination.

Strengths:<br /> The authors carried out detailed transcriptomics profiling with scRNA-seq and scATAC-seq datasets to conclude upregulated KLF2 target genes in HIV-1 RNA+ CD4 T Cells.

Weaknesses:

This key observation of up-regulation KLF2 associated genes family might be important in the HIV field for early diagnosis and viral clearance. However, with the limited sample size and in-vivo study model, it will be hard to conclude. I highly recommend increasing the sample size of early HIV-1-infected patients.

Reviewer #3 (Public review):

Summary:

This manuscript studies intracellular changes and immune processes during early HIV-1 infection with an additional focus on the small CD4+ T cell subsets. The authors used single-cell omics to achieve high resolution of transcriptomic and epigenomic data on the infected cells which were verified by viral RNA expression. The results add to understanding of transcriptional regulation which may allow progression or HIV latency later in infected cells. The biosamples were derived from early HIV infection cases, providing particularly valuable data for the HIV research field.

Strengths:

The authors examined the heterogeneity of infected cells within CD4 T cell populations, identified a significant and unexpected difference between naive and effector CD4 T cells, and highlighted the differences in Th2 and Th17 cells. Multiple methods were used to show the role of the increased KLF2 factor in infected cells. This is a valuable finding of a new role for the major transcription factor in further disease progression and/or persistence.

The methods employed by the authors are robust. Single-cell RNA-Seq from PBMC samples was followed by a comprehensive annotation of immune cell subsets, 16 in total. This manuscript presents to the scientific community a valuable multi-omics dataset of good quality, which could be further analyzed in the context of larger studies.

Weaknesses: