Joint Public Review:

Roget et al. build on their previous work developing a simple theoretical model to examine whether ageing can be under natural selection, challenging the mainstream view that ageing is merely a byproduct of other biological and evolutionary processes. The authors propose an agent-based model to evaluate the adaptive dynamics of a haploid asexual population with two independent traits: fertility timespan and mortality onset. Through computational simulations, their model demonstrates that ageing can give populations an evolutionary advantage. Notably, this observation arises from the model without invoking any explicit energy tradeoffs, commonly used to explain this relationship.

Additionally, the theoretical model developed here indicates that mortality onset is generally selected to start before the loss of fertility, irrespective of the initial values in the population. The selected relationship between the fertility timespan and mortality onset depends on the strength of fertility and mortality effects, with larger effects resulting in the loss of fertility and mortality onset being closer together. By allowing for a trans-generational effect on ageing in the model, the authors show that this can be advantageous as well, lowering the risk of collapse in the population despite an apparent fitness disadvantage in individuals. Upon closer examination, the authors reveal that this unexpected outcome is a consequence of the trans-generational effect on ageing increasing the evolvability of the population (i.e., allowing a more effective exploration of the parameter landscape), reaching the optimum state faster.

The simplicity of the proposed theoretical model represents both the major strength and weakness of this work. On one hand, with an original and rigorous methodology, the logic of their conclusions can be easily grasped and generalised, yielding surprising results. Using just a handful of parameters and relying on direct competition simulations, the model qualitatively recapitulates the negative correlation between lifespan and fertility without requiring energy tradeoffs. This alone makes this work an important milestone for the rapidly growing field of adaptive dynamics, opening many new avenues of research, both theoretically and empirically.

On the other hand, the simplicity of the model also makes its relationship with living organisms difficult to gauge, leaving open questions about how much the model represents the reality of actual evolution in a natural context. In particular, a more explicit discussion of how the specifics of the model can impact the results and their interpretation is needed. For example, the lack of mechanistic details on the trans-generational effect on ageing makes the results difficult to interpret. Even if analytical results are obtained, most of the observations appear derived from simulations as they are currently presented. Also, the choice of parameters for the simulations shown in the paper and how they relate to our biological knowledge are not fully addressed by the authors. Finally, the conclusions of evolvability are insufficiently supported, as the authors do not show if the wider genotypic variability in populations with the ageing trans-generational effect is, in fact, selected.

eLife assessment

The fundamental study presents a two-domain thermodynamic model for TetR which accurately predicts in vivo phenotype changes brought about as a result of various mutations. The evidence provided is solid and features the first innovative observations with a computational model that captures the structural behavior, much more than the current single-domain models.

Reviewer #1 (Public Review):

Summary:

The authors' earlier deep mutational scanning work observed that allosteric mutations in TetR (the tetracycline repressor) and its homologous transcriptional factors are distributed across the structure instead of along the presumed allosteric pathways as commonly expected. Especially, in addition, the loss of the allosteric communications promoted by those mutations, was rescued by additional distributed mutations. Now the authors develop a two-domain thermodynamic model for TetR that explains these compelling data. The model is consistent with the in vivo phenotypes of the mutants with changes in parameters, which permits quantification. Taken together their work connects intra- and inter-domain allosteric regulation that correlate with structural features. This leads the authors to suggest broader applicability to other multidomain allosteric proteins.

Here the authors follow their first innovative observations with a computational model that captures the structural behavior, aiming to make it broadly applicable to multidomain proteins. Altogether, an innovative and potentially useful contribution.

Weaknesses:

None that I see, except that I hope that in the future, if possible, the authors would follow with additional proteins to further substantiate the model and show its broad applicability. I realize however the extensive work that this would entail.

Reviewer #2 (Public Review):

Summary:

This combined experimental-theoretical paper introduces a novel two-domain statistical thermodynamic model (primarily Equation 1) to study allostery in generic systems but focusing here on the tetracycline repressor (TetR) family of transcription factors. This model, building on a function-centric approach, accurately captures induction data, maps mutants with precision, and reveals insights into epistasis between mutations.

Strengths:

The study contributes innovative modeling, successful data fitting, and valuable insights into the interconnectivity of allosteric networks, establishing a flexible and detailed framework for investigating TetR allostery. The manuscript is generally well-structured and communicates key findings effectively.

Weaknesses:

The only minor weakness I found was that I still don't have a better sense into (a) intuition and (b) mathematical derivation of Equation 1, which is so central to the work. I would recommend that the authors provide this early on in the main text.

Reviewer #3 (Public Review):

Summary:

Allosteric regulations are complicated in multi-domain proteins and many large-scale mutational data cannot be explained by current theoretical models,, especially for those that are neither in the functional/allosteric sites nor on the allosteric pathways. This work provides a statistical thermodynamic model for a two-domain protein, in which one domain contains an effector binding site and the other domain contains a functional site. The authors build the model to explain the mutational experimental data of TetR, a transcriptional repress protein that contains a ligand and a DNA-binding domain. They incorporate three basic parameters, the energy change of the ligand and DNA binding domains before and after binding, and the coupling between the two domains to explain the free energy landscape of TetR's conformational and binding states. They go further to quantitatively explain the in vivo expression level of the TetR-regulated gene by fitting into the induction curves of TetR mutants. The effects of most of the mutants studied could be well explained by the model. This approach can be extended to understand the allosteric regulation of other two-domain proteins, especially to explain the effects of widespread mutants not on the allosteric pathways.

Strengths:

The effects of mutations that are neither in the functional or allosteric sites nor in the allosteric pathways are difficult to explain and quantify. This work develops a statistical thermodynamic model to explain these complicated effects. For simple two-domain proteins, the model is quite clean and theoretically solid. For the real TetR protein that forms a dimeric structure containing two chains with each of them composed of two domains, the model can explain many of the experimental observations. The model separates intra and inter-domain influences that provide a novel angle to analyse allosteric effects in multi-domain proteins.

Weaknesses:

As mentioned above, the TetR protein is not a simple two-main protein, but forms a dimeric structure in which the DNA binding domain in each chain forms contacts with the ligand-binding domain in the other chain. In addition, the two ligand-binding domains have strong interactions. Without considering these interactions, especially those mutants that are on these interfaces, the model may be oversimplified for TetR.

Author Response

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

eLife assessment

The findings of this study are valuable as they provide new insights into the role of acetylcholine in modulating sensory processing in the auditory cortex. This paper reports a systematic measurement of cell activity in the auditory cortex before and after applying ACh during an oddball and cascade sequence of auditory stimuli in anesthetized rats. The results presented are solid given the rigorous experimental design and statistical analysis. The conclusions are provocative and will interest researchers in auditory neuroscience and neuromodulation, as well as clinicians and individuals with auditory processing disorders. However, the findings support multiple interpretations, beyond that offered by the authors.

Our reply: First and foremost, we would like to thank the editors and reviewers for their constructive criticisms, as well as their thoughtful and thorough evaluations of our manuscript. We greatly appreciate their assessment about the novelty and general significance in our study and have revised the manuscript according to their recommendations. In the following we include detailed responses and revisions based on the reviewer’s recommendations.

Public Reviews:

Reviewer #1 (Public Review):

Summary:

This study examined the impact of exogenous microapplication of acetylcholine (Ach) on metrics of novelty detection in the anesthetized rat auditory cortex. The authors found that the majority of units showed some degree of modulation of novelty detection, with roughly similar numbers showing enhanced novelty detection, suppressed novelty detection, or no change. Enhanced novelty responses were driven by increases in repetition suppression. Suppressed novelty responses were driven by deviance suppression. There were no compelling differences seen between auditory cortical subfields or layers, though there was heterogeneity in the Ach effects within subfields. Overall, these findings are important because they suggest that fluctuations in cortical Ach, which are known to occur during changes in arousal or attentional states, will likely influence the capacity of individual auditory cortical neurons to respond to novel stimuli.

Strengths:

The work addresses an important problem in auditory neuroscience. The main strengths of the study are that the work was systematically done with appropriate controls (cascaded stimuli) and utilizes a classical approach that ensures that drug application is isolated to the micro-environment of the recorded neuron. In addition, the authors do not isolate their study to only the primary auditory cortex, but examine the impact of Ach across all known auditory cortical subfields.

Our reply: Thank you very much for these supportive comments and the appreciation of our work.

Weaknesses:

1. As acknowledged by the authors, this study explicitly examines a phenomenon of high relevance to active listening but is done in anesthetized animals, limiting its applicability to the waking state.

Our reply: We agree; and indeed, this weakness was already recognized in the original manuscript but is now emphasized in the discussion.

1. The authors do not make any attempt to determine, by spike shape/duration, if their units are excitatory or inhibitory, which may explain some of the variance of the data.

Our reply: This is a very interesting question, and in fact, we have previously estimated whether neurons are excitatory or inhibitory based on the spike shape (Pérez-Gonzalez et al., 2021). Originally, we sought to implement a similar analysis here and tried to estimate if the recorded units were excitatory or inhibitory based on the spike shapes. But when we tried to perform this analysis, we found that in many cases the recordings had captured occasional spikes from other neurons. This caveat had introduced alterations in the average spike shape, and thus precluded an accurate categorization. Therefore, we decided to discard this analysis for the sake of correctness. This weakness is further commented on in the discussion.

1. The application of exogenous Ach, potentially in supra-physiological amounts, makes this study hard to extrapolate to a behaving animal. A more compelling design would be to block Ach, particularly at particular receptor types, to determine the effect of endogenous Ach.

Our reply: We agree again with the reviewer; this weakness was already acknowledged, but this is now further highlighted in discussion where we comment that future studies should analyze the effect of muscarinic- and nicotinic- receptors and blockade them to potentially observe more physiologically-comparable effects. Moreover, this issue is also related to a comment raised by reviewer#2 on a possible ‘dose-response relationship’ issue.

Reviewer #2 (Public Review):

Summary:

In this study, the authors investigate the effect of ACh on neuronal responses in the auditory cortex of anesthetized rats during an auditory oddball task. The paradigm consisted of two pure tones (selected from the frequency responses at each recording site) presented in a pseudo-random sequence. One tone was presented frequently (the "standard" tone) and the other infrequently (the "deviant" tone). The authors found that ACh enhances the detection of unexpected stimuli in the auditory environment by increasing or decreasing the neuronal responses to deviant and standard tones.

Strengths:

The study includes the use of appropriate and validated methodology in line with the current state-of-the-art, rigorous statistical analysis, and the demonstration of the effects of acetylcholine on auditory processing.

Our reply: Thank you very much for these supportive comments and the appreciation of our work.

Weaknesses:

The study was conducted in anesthetized rats, and further research is needed to determine the behavioral relevance of these findings.

Our reply: We agree; and indeed, this weakness was already recognized but is now emphasized in discussion.

Reviewer #1 (Recommendations For The Authors):

As outlined above, breaking out the units into those that are putative excitatory or inhibitory cells would be helpful, if possible. Other critiques are minor:

1. "Acetylcholine", "ACh" and "Ach" are used throughout the manuscript. Please define the chosen abbreviation at first use, and be consistent.

2. Line 116, remove comma after "ACh".

3. Line 123, I would add "in the rat at the end of the first sentence since the species was not mentioned up to this point.

4. Fig 2 - it would be useful in the Figure (not just in the text) to label red as being the deviant tone and blue as being the standard.

5. In many Figures (e.g., Fig 5), the term "effect" is found in the legend rather than "ACh". It would seem more intuitive to label these as "ACh".

6. The AUC and MI interpretations are not clear. Both are metrics that quantify similarity but the authors state that when these values decrease the neurons are less able to discriminate between them (i.e., they are more similar). Some clarifying text would be useful.

7. L276 - should "SI increase" be "SI decrease"?

8. L285 - would replace "solely" with "primarily".

9. Fig 7 - the authors may consider indicating with a label what the difference is between A and C compared to B and D.

10. L634 - why were only females used?

11. L646 - "bran" should be "brain".

12. L649 - "homoeothermic" should be "homeothermic".

13. L661 - "allowed to generate" should be "allowed the generation of".

14. L670 - no need for both "about" and "approximately".

15. L681 - please state what the search stimuli were.

16. L688 - should be "closed-field".

17. L754 - add a hyphen to "time-consuming".

Our reply: Thanks so much for the detailed proofreading of the manuscript and suggestions. All them have been clarified or implemented and corrected in the text.

Reviewer #2 (Recommendations For The Authors):

The authors could investigate the effects of different doses of ACh on auditory processing to determine if there is a dose-response relationship.

Our reply: We agree that this is an interesting question also relate to a matter raised by Reviewer#1 that could be linked to the issue of ‘exogenous Ach’.

The study only investigated the effects of ACh on neuronal responses during an auditory oddball task. It would be interesting to investigate the effects of ACh on other aspects of auditory processing, such as sound localization or the discrimination of tones.

Our reply: We agree that, while these aspects of auditory processing are very fascinating, they were outside the scope of the study, and not directly related to predictive coding and precision, so each one of these characteristics would be a full, future project in itself.

The authors could provide more context on the significance of their findings for individuals with auditory processing disorders.

Our reply: Thanks for the suggestion. It remains unclear how abnormal brainstem and cortical processing associated with auditory processing disorders arises (Moore, 2006, 2012). While we are not aware of any known direct connection between auditory processing disorders and acetylcholine, individuals with auditory processing disorders do have difficulties with auditory selective attention, so perhaps one could speculate that ACh, by modulating SSA/prediction error, could have some impact on encoding salient events, and if disrupted could lead to problems with selective attention. Moore (2012) speculated that auditory processing disorders may arise from unbalanced processing in bottom-up and top-down contributions.

Since ACh has been implicated in some neurogenerative diseases and neurodevelopmental disorders, we have also added in the Discussion dialogue about a possible relationship between the modulatory effect of ACh on predictive coding (which involves bottom-up and top-down contributions) and auditory processing disorders. We also cite the recent work by Felix and colleagues (2019) which is the only study we have found on the effects of ACh on auditory processing disorders where they analyzed altered temporal processing at the level of the brainstem in α7-subunit of the nicotinic acetylcholine receptor (α7-nAChR)-deficient mice. After studying α7-nAChR knockout mice of both sexes and wild-type colony controls, they concluded that the malfunction of the CHRNA7 gene that encodes the α7-nAChR may contribute to degraded spike timing in the midbrain, which may underlie the observed timing delay in the ABR signals. These authors propose that their findings are consistent with a role for the α7-nAChR in types of neurodevelopmental and auditory processing disorders. There is also evidence on cholinergic system disfunction being related to the pathophysiology of Alzheimer’s disease (Pérez-González et al., 2022). For instance, disfunction of the synapses of cholinergic neurons in the hippocampus and nucleus basalis of Meynert, as well as decreased choline acetyltransferase activity, is associated to memory disorders in Alzheimer’s disease (Hampel et al., 2018). Also, A Alzheimer’s disease D patients show reduced amounts of the vesicular ACh transporter in some brain areas (Aghourian et al., 2017). Finally, cholinesterase inhibitors seem to have some favorable effect in the treatment of Alzheimer’s disease patients (Sharma, 2019).

Aghourian M, Legault-Denis C, Soucy J-P, Rosa-Neto P, Gauthier S, Kostikov A, et al. 2017. Quantification of brain cholinergic denervation in Alzheimer’s disease using PET imaging with [18F]-FEOBV. Mol. Psychiatry 22:1531–1538. doi: 10.1038/mp.2017.183

Felix RA 2nd, Chavez VA, Novicio DM, Morley BJ, Portfors CV. 2019. Nicotinic acetylcholine receptor subunit α7-knockout mice exhibit degraded auditory temporal processing. J Neurophysiol. 122(2):451-465. doi: 10.1152/jn.00170.2019.

Hampel H, Mesulam M-M, Cuello AC, Khachaturian AS, Vergallo A, Farlow MR, et al. 2018. Revisiting the Cholinergic Hypothesis in Alzheimer’s Disease: emerging Evidence from Translational and Clinical Research. J. Prev. Alzheimers Dis. 6:1–14. doi:10.14283/jpad.2018.43

Moore DR. 2006. Auditory processing disorder (APD)-potential contribution of mouse research. Brain Res. 1091:200–206.

Moore DR. 2012. Listening difficulties in children: bottom-up and top-down contributions. J Commun Disord. ;45:411–418.

Pérez-González D, Parras GG, Morado-Díaz CJ, Aedo-Sánchez C, Carbajal GV, Malmierca MS. 2021. Deviance detection in physiologically identified cell types in the rat auditory cortex. Hear Res. 2021 Jan;399:107997. doi: 10.1016/j.heares.2020.107997.

Pérez-González D, Schreiner TG, Llano DA and Malmierca MS. 2022. Alzheimer’s Disease, Hearing Loss, and Deviance Detection. Front. Neurosci. 16:879480. doi: 10.3389/fnins.2022.879480

Sharma K. 2019. Cholinesterase inhibitors as Alzheimer’s therapeutics. Mol. Med. Rep. 20:1479–1487. doi:10.3892/mmr.2019.1 0374

eLife assessment

The findings of this study are valuable as they provide new insights into the role of acetylcholine in modulating sensory processing in the auditory cortex. This paper reports a systematic measurement of cell activity in the auditory cortex before and after the microiontophoretic application of Ach during an oddball and cascade sequence of auditory stimuli. The evidence presented is convincing, as the study used a rigorous experimental design and statistical analysis. The manuscript will interest researchers in auditory neuroscience and neuromodulation, as well as clinicians and individuals with auditory processing disorders.

Reviewer #1 (Public Review):

Summary:<br /> This study examined the impact of exogenous microapplication of acetylcholine (Ach) on metrics of novelty detection in the anesthetized rat auditory cortex. The authors found that the majority of units showed some degree of modulation of novelty detection, with roughly similar numbers showing enhanced novelty detection, suppressed novelty detection or no change. Enhanced novelty responses were driven by increases in repetition suppression. Suppressed novelty responses were driven by deviance suppression. There were no compelling differences seen between auditory cortical subfields or layers, though there was heterogeneity in the Ach effects within subfields. Overall, these findings are important because they suggest that fluctations in cortical Ach, which are known to occur during changes in arousal or attentional states, will likely influence the capacity for individual auditory cortical neurons to respond to novel stimuli.

Strengths:<br /> The work addresses an important problem in auditory neuroscience. The main strengths of the study are that the work appears to be systematically done with appropriate controls (cascaded stimuli) and utilizes a classical approach that ensures that drug application is isolated to the micro-environment of the recorded neuron. In addition, the authors do not isolate their study to only primary auditory cortex, but examine the impact of Ach across all known auditory cortical subfields.

Weaknesses:<br /> 1. As acknowledged by the authors, this study explicitly examines a phenomenon of high relevance to active listening, but is done in anesthetized animals, limiting its applicability to the waking state.<br /> 2. The authors do not make any attempt to determine, by spike shape/duration, if their units are excitatory or inhibitory, which may explain some of the variance of the data.<br /> 3. The application of exogenous Ach, potentially in supra-physiological amounts, makes this study hard to extrapolate to a behaving animal. A more compelling design would be to block Ach, particularly at particular receptor types, to determine the effect of endogenous Ach

Reviewer #2 (Public Review):

Summary:

In this study, the authors investigate the effect of ACh on neuronal responses in the auditory cortex of anesthetized rats during an auditory oddball task. The paradigm consisted of two pure tones (selected from the frequency responses at each recording site) presented in a pseudo-random sequence. One tone was presented frequently (the "standard" tone) and the other infrequently (the "deviant" tone). The authors found that ACh enhances the detection of unexpected stimuli in the auditory environment by increasing or decreasing the neuronal responses to deviant and standard tones.

Strengths:

The study includes the use of appropriate and validated methodology in line with the current state-of-the-art, rigorous statistical analysis and the demonstration of the effects of acetylcholine on auditory processing.

Weaknesses:

The study was conducted in anesthetized rats, and further research is needed to determine the behavioral relevance of these findings.

Author Response

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

The manuscript by Heyndrickx et al describes protein crystal formation and function that bears similarity to Charcot-Leyden crystals made of galectin 10, found in humans under similar conditions. Therefore, the authors set out to investigate CLP crystal formation and their immunological effects in the lung. The authors reveal the crystal structure of both Ym1 and Ym2 and show that Ym1 crystals trigger innate immunity, activated dendritic cells in the lymph node, enhancing antigen uptake and migration to the lung, ultimately leading to induction of type 2 immunity.

Strengths:

We know a lot about expression levels of CLPs in various settings in the mouse but still know very little about the functions of these proteins, especially in light of their ability to form crystal structures. As such data presented in this paper is a major advance to the field.

Resolving the crystal structure of Ym2 and the comparison between native and recombinant CLP crystals is a strength of this manuscript that will be a very powerful tool for further evaluation and understanding of receptor, binding partner studies including the ability to aid mutant protein generation.

The ability to recombinantly generate CLP crystals and study their function in vivo and ex vivo has provided a robust dataset whereby CLPs can activate innate immune responses, aid activation and trafficking of antigen presenting cells from the lymph node to the lung and further enhances type 2 immunity. By demonstrating these effects the authors directly address the aims for the study. A key point of this study is the generation of a model in which crystal formation/function an important feature of human eosinophilic diseases, can be studied utilising mouse models. Excitingly, using crystal structures combined with understanding the biochemistry of these proteins will provide a potential avenue whereby inhibitors could be used to dissolve or prevent crystal formation in vivo.

The data presented flows logically and formulates a well constructed overall picture of exactly what CLP crystals could be doing in an inflammatory setting in vivo. This leaves open a clear and exciting future avenue (currently beyond the scope of this work) for determining whether targeting crystal formation in vivo could limit pathology.

Weaknesses:

Although resolving the crystal structure of Ym2 in particular is a strength of the authors work, the weaknesses are that further work or even discussion of Ym2 versus Ym1 has not been directly demonstrated. The authors suggest Ym2 crystals will likely function the same as Ym1, but there is insufficient discussion (or data) beyond sequence similarity as to why this is the case. If Ym1 and Ym2 crystals function the same way, from an evolutionary point, why do mice express two very similar proteins that are expressed under similar conditions that can both crystalise and as the authors suggest act in a similar way. Some discussion around these points would add further value.

We agree with reviewer. We have further elaborated the discussion section including these points, stating clearly that more research needs to be done using Ym2 crystals before we can draw parallels in vivo.

Additionally, the crystal structure for Ym1 has been previously resolved (Tsai et al 2004, PMID 15522777) and it is unclear whether the data from the authors represents an advance in the 3D structure from what is previously known.

The crystal structure of Ym1 has indeed been previously solved, and we refer to that paper. In addition, we also provide the crystal structure of in vitro grown Ym1, ashowing biosimilarity. This, for the field of crystallography is a major finding, since it validates the concept that crystal structures generated in vitro can reflect in vivo grown structures. Moreover, the in vivo crystallization of Ym2 was unknown prior to this work, and is now clear as revealed by the ex vivo X-ray crystallography. The strength of our story is that we can now compare Ym1 and Ym2 crystals structures in detail.

Whilst also generating a model to understand Charcot-Leyden crystals (CLCs), the authors fail to discuss whether crystal shape may be an important feature of crystal function. CLCs are typically needle like, and previous publications have shown using histology and TEM that Ym1 crystals are also needle like. However, the crystals presented in this paper show only formation of plate like structures. It is unclear whether these differences represent different methodologies (ie histology is 2D slides), or differences in CLP crystals that are intracellular versus extracellular. These findings highlight a key question over whether crystal shape could be important for function and has not been addressed by the authors.

In contrast to the bipyramidal, needle-like CLC crystals formed by human galectin-10 protein (hexagonal space group P6522), the in vivo grown Ym1 and Ym2 crystals we were able to isolate for X-ray diffraction experiments had a plate-like morphology with identical crystallographic parameters as recombinant Ym1/Ym2 crystals (space group P21). We note that depending on the viewing orientation of the thin plate-like Ym1 crystals, they may appear needle-like in histology and TEM images. In addition, we can fully not exclude that both Ym1 or Ym2 may crystallize in vivo in different space groups (which could result in different crystal morphologies for Ym1/Ym2) but we have no data to support this. It is finally also a possibility that plate like structures can break up in vivo along a long axis as a result of mechanical forces, and end up as rod-or needle like shapes.

Ym1/Ym2 crystals are often observed in conditions where strong eosinophilic inflammation is present. However, soluble Ym1 delivery in naïve mice shows crystal formation in the absence of a strong immune response. There is no clear discussion as to the conditions in which crystal formation occurs in vivo and how results presented in the paper in terms of priming or exacerbating an immune response align with what is known about situations where Ym1 and Ym2 crystals have been observed.

Although Ym1 and Ym2 crystals are often observed in mice at sites of eosinophilic inflammation, they are not made by eosinophils, but mainly by macrophages and epithelial cells, respectively. In vitro, protein crystallization typically starts from supersaturated solutions that support crystal nucleation. Several factors such as temperature and pH can affect the solubility of Ym1 and Ym2 in vivo and thus affect the nucleation and crystallization process. For Ym1 and Ym2 we noticed in vitro that a small drop in pH facilitates the crystallization process. Although the physiological pH is 7.4, during inflammation, there is a drop in pH. This drop in pH is the result of the infiltration and activation of inflammatory cells in the tissue, which leads to an increased energy and oxygen demand, accelerated glucose consumption via glycolysis and thus increased lactic acid secretion. In addition, we cannot exclude that in vivo, the nucleation process for Ym1/Ym2 is facilitated by interaction with ligands in the extracellular space (e.g. polysaccharide ligands or other – yet to be identified – specific ligands to Ym1/Ym2).

Reviewer #2 (Public Review):

Summary:

This interesting study addresses the ability of Ym1 protein crystals to promote pulmonary type 2 inflammation in vivo, in mice.

Strengths:

The data are extremely high quality, clearly presented, significantly extending previous work from this group on the type 2 immunogenicity of protein crystals.

Weaknesses:

There are no major weaknesses in this study. It would be interesting to see if Ym2 crystals behave similarly to Ym1 crystals in vivo. Some additional text in the Introduction and Discussion would enrich those sections.

We agree that this would be interesting to investigate, however, we choose to not include recombinant Ym2 crystal data in this report. However, we have further elaborated the discussion section including this point.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

Suggestions for improved experiments and to strengthen findings:

I think additional data on the ability of Ym2 crystals to induce an immune response would be advantageous. I'm not by any means suggesting the authors repeat all the experiments with Ym2 crystals, but even just the ability to show that Ym2 could promote type 2 immunity in the acute OVA model, would help to strengthen the argument that these crystals in general function in a similar way. Alternatively, a discussion on whether these protein crystals may function in different scenarios/tissues or conditions could help in light of additional data

We agree that this is an interesting point to investigate, however, we choose to not include recombinant Ym2 crystal data in this report. However, we have further elaborated the discussion section including this point.

Measuring IL-33 in lung tissue is difficult to interpret as cells will express intracellular IL-33 that is not active and may explain why the results in Fig 2D are not overly convincing. It could just be that Ym1 crystals are changing the number of cells expressing IL-33 (e.g macrophages, or type 2 pneumocytes) Did the authors also measure active IL-33 release in the BAL fluid which may give a better indication of Ym1's ability to activate DAMPs?

We also measured active IL-33 release in the BAL fluid, but due to the limited sample availability we could only measure this in one of the two repeat experiments, resulting in non-significant results for the BAL fluid. However, certainly for the 6h timepoint we saw a similar trend in the BAL fluid as in the lung tissue, meaning higher levels of IL-33 in the Ym1 crystal group compared to the PBS and soluble Ym1 group.

Crystals in Fig 2F staining with Ym1 appear to be brighter in the soluble Ym1 group. Is this related to increased packing of Ym1 in the crystals formed in vivo as opposed to those formed in vitro? Aside from reduced amount of crystals that form when you give soluble Ym1, could the type of crystal also be influencing the ability of soluble Ym1 crystals to generate an immune response?

Our X-ray diffraction experiments show that the packing of Ym1 is identical for in vivo and in vitro grown crystals. Possibly the apparent difference in brightness is caused by stochastic staining by the antibody. In this regard we note that the crystals formed from soluble Ym1 after 24h also can appear as less bright in a similar fashion as recombinant Ym1 crystals.

Overall, the data and writing of the manuscript is presented to a very high standard

A few minor points:

• Fig 2F - a little unsure what the number in the left top corner of the images represented.

These numbers represent the picture numbers generated by the software, but as they don’t have any added value for the story, we removed these numbers from the images.

• Not clear why two different expression vectors were used - one for Ym1 and one for Ym2?

Because we observed that recombinant Ym2 is more poorly secreted in the mammalian cell culture supernatant as compared to recombinant Ym1, we produced Ym2 with an N-terminal hexahistidine-tag followed by a Tobacco Etch Virus (TEV)-protease cleavage site to facilitate its purification.

Reviewer #2 (Recommendations For The Authors):

The authors briefly outline in their Introduction potential Sources of Ym1/2 in vivo, highlighting monocytes, M2 macrophages, alveolar macrophages, neutrophils and epithelial cells. Do DCs also make detectable/meaningful amounts of Ym1/2 in vivo, particularly in type 2 settings?

In the introduction we only highlighted the main cellular sources of Ym1 and Ym2, but there is literature available stating/showing that Ym1/2 is not only expressed by macrophages, neutrophils, monocytes and epithelial cells, but can also be induced in DCs and mast cells. We added the word ‘mainly’ to this sentence in the introduction, to make clear that macrophages, neutrophils and monocytes are not the only sources of Ym1.

Given the nicely demonstrated similarity of recombinant Ym1 and Ym2 crystals, I think it is important for the authors to include at least initial data on the outcome of recombinant Ym2 crystal admin to mice, in comparison to their Ym1 data.

We agree that this is an interesting point to investigate, however, we choose to not include recombinant Ym2 crystal data in this report. However, we have further elaborated the discussion section including this point.

Given the generation of crystals following in vivo administration of soluble Ym1, albeit at a lower level than when crystals were administered, it would be interesting to see if increased concentrations of soluble material show a dose dependent increase in lung inflammation readouts.

We agree that this would be an interesting point to investigate. Alongside this we could also titrate down the crystal dose, to see if there is a dose dependent decrease in lung inflammation readouts. However, at this time, we choose to not investigate this further.

I couldn't easily follow the authors' Discussion about potential ability of anti Ym-1/2 Abs to dissolve Ym1/2 crystals (similar to what they have demonstrated for Abs vs Gal10 crystals). Have they addressed this possibility experimentally? If so, addition of such data to the manuscript would be extremely interesting, given the obvious potential Ym1/2 crystal dissolving Abs for investigation of the role of these in a range of different murine models of type 2 inflammation.

We agree that the phrasing of this part of the discussion can be unclear/confusing. We rephrased this part to make it clearer. However, we did not address the possibility of Ym1/2 crystal dissolving antibodies experimentally.

In the Results section, the authors briefly comment on the pro-type 2 nature of Ym1 crystals in relation to their previous work with uric acid and Gal10 crystals, proposing that the pulmonary type 2 response may be a 'generic response to crystals of different chemical composition'. The Discussion would be enriched by deeper exploration of this comment.

We have further elaborated the discussion section including this point.

eLife assessment

This is an important and interesting account of the ability of Ym1 crystals to promote type 2 immunity in vivo, in mice. The data presented are compelling, building on and significantly advancing evidence this group has previously published on the type 2 immunogenicity of other protein crystals. The work will be of relevant interest to immunologists and researchers working on type 2 inflammatory disease, in the lung and in others tissues.

Reviewer #1 (Public Review):

Summary:

The manuscript by Heyndrickx et al describes protein crystal formation and function that bears similarity to Charcot-Leyden crystals made of galectin 10, found in humans under similar conditions. Therefore, the authors set out to investigate CLP crystal formation and their immunological effects in the lung. The authors reveal the crystal structure of both Ym1 and Ym2 and show that Ym1 crystals trigger innate immunity, activated dendritic cells in the lymph node, enhancing antigen uptake and migration to the lung, ultimately leading to induction of type 2 immunity.

Strengths:

We know a lot about expression levels of CLPs in various settings in the mouse, but still know very little about the functions of these proteins, especially in light of their ability to form crystal structures. As such data presented in this paper is a major advance to the field.

Resolving the crystal structure of Ym2 and the comparison between native and recombinant CLP crystals is a strength of this manuscript that will be a very powerful tool for further evaluation and understanding of receptor, binding partner studies including ability to aid mutant protein generation.

The ability to recombinantly generate CLP crystals and study their function in vivo and ex vivo has provided a robust dataset whereby CLPs can activate innate immune responses, aid activation and trafficking of antigen presenting cells from the lymph node to the lung and further enhances type 2 immunity. By demonstrating these effects the authors directly address the aims for the study. A key apoint of this study is the generation of a model in which crystal formation/function an important feature of human eosinophilic diseases, can be studied utilising mouse models. Excitingly, using crystal structures combined with understanding the biochemistry of these proteins will provide a potential avenue whereby inhibitors could be used to dissolve or prevent crystal formation in vivo.

Generation of the crystal structure for Ym2 is a particular strength of the authors work and highlights the similarities between Ym1 and Ym2. Whilst the authors did not specifically examine Ym2 function, they have provided a discussion on this and speculate that Ym2 will function in a similar manner to Ym1.

The data presented flows logically and formulates a well constructed overall picture of exactly what CLP crystals could be doing in an inflammatory setting in vivo. Leaves open a clear and exciting future avenue (currently beyond the scope of this work) for determining whether targeting crystal formation in vivo could limit pathology.

Weaknesses:

It would have been nice for the authors to confirm whether Ym2 has similar functions to Ym1 using the in vivo and in vitro systems. However, they have discussed these points and raised it as a potential for future studies.

Reviewer #2 (Public Review):

Summary:

This interesting study addresses the ability of Ym1 protein crystals to promote pulmonary type 2 inflammation in vivo, in mice.

Strengths:

The data are extremely high quality, clearly presented, significantly extending previous work from this group on the type 2 immunogenicity of protein crystals.

Weaknesses:

There are no major weaknesses in this study. It would be interesting to see if Ym2 crystals behave similarly to Ym1 crystals in vivo. Some additional text in the Introduction and Discussion would enrich those sections.

Author Response

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

We thank the reviewers for their thorough reading of the manuscript and insightful comments. We have responded to both the “public review” and the “recommendations” and feel that the manuscript is now significantly strengthened.

Public Review comments

Reviewer #1:

Weaknesses:

1. The abstract does not discuss the reduction of E-gel consumption that occurs after multiple days of exposure to the THC formulation, but rather implies that a new model for chronic oral self-administration has been developed. Given that only two days of consumption was assessed, it is not clear if the model will be useful to determine THC effects beyond the acute measures presented here. The abstract should clarify that there was evidence of reduced consumption/aversive effects with repeated exposures.

Thank you for your observation. We have added language to address this in the manuscript and the abstract. The model developed in the manuscript is an acute exposure model, with the intention of further chronic exposure adaptations to be developed separately (page 2, line 29).

1. In the results section, the authors sometimes describe effects in terms of the concentration of gel as opposed to the dose consumed in mg/kg, which can make interpretation difficult. For example, the text describing Figure 1i states that significant effects on body temperature were achieved at 4 mg CTR-gel and 5 mg THC-gel, but were essentially equivalent doses consumed? It would be helpful to describe what average dose of THC produced effects given that consumption varied within each group of mice assigned to a particular concentration.

We thank the reviewer for this comment and have edited our text to clarify our results. For example, this point is further emphasized by the correlation of the data in Figure1l-n showing the relationship between individual consumption and behavioral readouts (page 11, line 225-226).

1. The description of the PK data in Figure 3 did not specify if sex differences were examined. Prior studies have found that males and females can exhibit stark differences in brain and plasma levels of THC and metabolites, even when behavioral effects are similar. However, this does depend on species, route, timing of tissue collection. It would be helpful to describe the PK profile of males and females separately.

We did compare sex dependent effects and found no significant effects after THC E-gel consumption. We’ve added additional language to address this point in the discussion (Supplementary tables T1 and T2).

1. In Figure 5, it is unclear how the predicted i.p. THC dose could be 30 mg/kg when 30 mg/kg was not tested by the i.p. route according to the figure, and if it had been it would have likely been almost zero acoustic startle, not the increased startle that was observed in the 2 hr gel group. It seems more likely that it would be equivalent to 3 mg/kg i.p. Could there be an error in the modeling, or was it based on the model used for the triad effects? This should be clarified.

We apologize for the confusion created by that data, and it has now been updated for clarity. The original ~30mg/kg was not a predicted dose consumed, but rather an expected dose consumed based on individual male v. female consumption data in Supplemental Figure S1b. For clarity on the figure, we’ve instead placed dashed lines that draw attention only to the predicted startle response expected from our THC-E-gel model. We have also updated the text which hopefully makes this clearer.

Reviewer #2:

Weaknesses:

Certainly, more THC mediated behavioral outcomes could have been tested, but the work presents a proof-of-concept study to investigate acute THC treatment.

It would have been interesting if this application form is also possible for chronic treatment regimen

We agree that a chronic treatment regimen and additional behavioral outcomes is the next, most exciting step for expanding this oral THC-E-gel consumption model, and something we are actively pursuing.

Reviewer #3:

Weaknesses:

The main weaknesses of the manuscript revolve around clarification of the Methods section. All of these weaknesses are described in the "Recommendations to authors" section. Revising the manuscript would account for many of these weaknesses.

Thank you for carefully reading through our methodology. We have made edits according to everything brought up in the recommendation section of reviewer comments.

Recommendations for Authors

Reviewer #1:

Minor edits to the text:

Abstract: "intraperitoneal contingent" should be "intraperitoneal noncontingent".

Line 221, this sentence needs editing for clarity.

Lines 249-250, incomplete sentence.

Line 284, the word "activity" is missing from "locomotor between mice".

Lines 299-301, incomplete sentence.

Thank you for finding these mistakes. All these recommendations have been incorporated into the final publication.

Reviewer #2:

1. The typical THC tetrad includes catalepsy. Why was this behavioral outcome not monitored?

We felt that locomotion, analgesia, and body temperature were robust behavioral readouts for monitoring cannabimimetic responses and that acoustic startle served as an additional, novel means of understanding THC-E-gel effects.

1. Please specify the exact substrain of C57BL/6 (i.e., J or N or some other)

C57BL/6J mice were used for the publication. This clarification has been made in the methods section.

1. Figure S3 is not mentioned in the result part, but only in the discussion.

Figure S3 is now referenced in the main body of the Results section.

1. It might be interesting to follow up the issue that the individual THC consumption is considerable, as depicted in Fig. 1e (at high dose). This will presumably also lead to different behavioral responses. Or is there individual metabolism, also difference male vs. female?

Thank you for the suggestion. We agree that the distribution of THC doses consumed (calculation based on weight) would be worth further investigating and have now included language about this (page 20, line 436). Please note that we did not find a sex difference (Supplemental Figure S1b), but it would be exciting to discover some biologically relevant cause such as individual absorption or metabolism

Reviewer #3:

Major

1. Methods: Were the observers of experiments blinded to animal treatment? Why or why not?

Multiple investigators performed the behavioral measurements and were not blinded to mouse treatments, but the dose consumed by each mouse remained blind. Thus, because animals consumed THC gelatin of their own volition while having ad libitum access, we performed the correlational analysis presented in Figure 1 l-n.

1. Methods: The authors could consider relating their study design to the ARRIVE guidelines and providing a statement as to whether their study adheres to these guidelines. Related to this, were mice provided with any environmental enrichment during the study?

We followed the ARRIVE guidelines with exception to investigator blinding (described above). Please note that mice were not provided with additional environmental enrichment during the study, a point that we specified in our methods (page 5, line 91).

1. Methods / Results: In the Methods it is stated that the triad of cannabimimetic behaviors was measured 1 h post-injection or immediately after gelatin exposure. Why were these timepoints chosen? Perhaps this wording should be revised because measurements of cannabimimetic effects were taken several times after drug exposure. Peak i.p. drug may occur earlier than 1 h whereas peak oral drug effect is likely to occur over a longer time period (i.e., not immediately after) due to delays of absorption and first pass metabolism. Is it possible that the authors have underestimated oral drug effects by selecting these timepoints? Please discuss.

We observed a reduction in locomotion activity starting 1 h following the beginning of exposure to the gelatin (Figure 2), suggesting initial cannabimimetic changes. Based on this observable response we chose to measure all cannabimimetic behaviors immediately following gelatin exposure. The exposure timeline for i.p. injection (1 h post-injection) was selected based on a standard published protocol (Metna-Laurent et al, 2017).

a. Pharmacodynamics: Related to this and because the aim of this paper is to establish a rodent oral dose model, could the authors discuss the need for better characterization of the time course of drug effects? For example, how might anti-nociception or locomotor activity vary following THC E-gel consumption? This is somewhat addressed in the locomotion time course in Figure 2G but could be elaborated on or discussed in more detail.

We agree that future studies should include additional time points measuring behavioral changes. This important point is now emphasized in the discussion (page 21, line 455).

b. Pharmacokinetics: Related to this point above, have the authors considered collecting blood or tissue samples from their i.p.-injected animals to assess drug pharmacokinetics as they relate to drug effect and as compared to oral THC consumption? I am not suggesting the authors conduct a completely new study for this manuscript; however, this could be raised as a future study and/or as a weakness of the current study.

We did not measure blood and tissue concentrations after i.p. administration due to the number of studies reporting these values by our co-author, Dr. Daniele Piomelli, that established these pharmacokinetic measures. Thus, we chose to reference these studies. Please note that repeating such measurements would be labor intensive, unnecessary use federal NIH resources and animals, while being very redundant to the existing literature.

c. Minor, but related to these points: In the results, page 14 line 299: the first sentence of this paragraph is confusing as written. The Reviewer recognizes that the authors are relating the pharmacokinetic work to previously published findings, but still thinks that measuring and comparing THC levels from their cohort of i.p.-injected animals would have benefitted the present study.

Thank you, this edit has been made in the manuscript.

1. Methods, Histology: The methods as described do not contain sufficient detail regarding THC and THC metabolite quantification. In addition, it is not clear from this section what Histology was performed and how (no histology results appear in the manuscript). Please add more detail to this section of the Methods.

We apologize for this typo and have corrected it in the methods section of the manuscript.

1. Methods / Results: The statistics section requires additional detail regarding the rationale for tests being performed on different datasets. In addition, a description of the curve fitting used for data in figures 1H-J, 4B-D, and S4 would be helpful to the reader.

Thank you, we have updated and provided more information regarding the curve fitting that was used in the methods and results section for the respective figure panels (page 9, line 183-184).

Minor

1. Throughout: The use of the phrase "high" dose is somewhat arbitrary and not defined relative to other doses of the THC formulation throughout the manuscript. The Reviewer suggests simply stating that THC was used, specifying the dose, or justifying in the Abstract and/or Introduction the classification of "high" based on relevant literature.

Thank you for the observation. We have removed this ambiguity by specifically mentioning the dose that was consumed (e.g., abstract page 2, line 20).

1. Abstract: define "CB1" in the abstract. Although this is a common abbreviation within the field, its use should be defined.

We have added this definition in the abstract for clarification.

1. Figure 2: why are the consumption panels B, C, and D given separate labels but the locomotor data are all labeled together as panel G?

Thank you for the observation, we have adjusted the labeling, so it is equal for both sets of panels.

Reviewer #1 (Public Review):

Summary:

This manuscript describes the development of an oral THC consumption model in mice where THC is added to a chocolate flavored gelatin. The authors compared the effects of THC consumed in this highly palatable gelatin (termed E-gel) to THC dissolved in a less palatable gelatin (CTR-gel), and to i.p. injections of multiple doses of THC, on the classic triad of CB1R dependent behaviors (hypolocomotion, antinociception, and body temperature).

The authors found that they could achieve consumption of higher concentrations of THC in the E-gel than the CTR-gel, and that this led to larger total dose exposure and decreases in locomotor activity, antinociception, and body temperature reductions similar to 3-4 mg/kg THC when tested after 2 hour consumption and roughly 10 mg/kg if tested immediately after 1 hour consumption. The majority of THC E-gel consumption was found to occur in the first hour on the first exposure day. THC E-gel consumption was lower than VEH E-gel consumption and this persisted on a subsequent consumption day, suggesting that the animals may form a taste aversion and that THC at the dose consumed likely has aversive properties, consistent with the literature on i.p. dosing. The authors also report the pharmacokinetics in brain and plasma of THC and metabolites after 1 or 2 hour consumption, finding high levels of THC in the brain that begins to dissipate at 2.5 hours is gone 24 hours later. Finally, the authors tested THC effects on the acoustic startle response and found an inverted dose response that was more pronounced in males than females after i.p. dosing and a greater startle response in males after E-gel dosing.

Overall, the authors find that voluntary oral consumption of THC can achieve levels of intake that are consistent with the present and prior reported literature on i.p. dosing.

Strengths:

The strengths of the article include a direct comparison of voluntary oral THC consumption to noncontingent i.p. administration, the use of multiple THC doses and oral THC formulations, the inclusion of multiple assays of cannabinoid agonist effects, and the inclusion of males and females. Additional strengths include monitoring intake over 10 minute intervals and validating that effects are CB1R dependent via antagonist studies.

Weaknesses:

1. The abstract does not discuss the reduction of E-gel consumption that occurs after multiple days of exposure to the THC formulation, but rather implies that a new model for chronic oral self-administration has been developed. Given that only two days of consumption was assessed, it is not clear if the model will be useful to determine THC effects beyond the acute measures presented here. The abstract should clarify that there was evidence of reduced consumption/aversive effects with repeated exposures.<br /> 2. In the results section, the authors sometimes describe effects in terms of the concentration of gel as opposed to the dose consumed in mg/kg, which can make interpretation difficult. For example, the text describing Figure 1i states that significant effects on body temperature were achieved at 4 mg CTR-gel and 5 mg THC-gel, but were essentially equivalent doses consumed? It would be helpful to describe what average dose of THC produced effects given that consumption varied within each group of mice assigned to a particular concentration.<br /> 3. The description of the PK data in Figure 3 did not specify if sex differences were examined. Prior studies have found that males and females can exhibit stark differences in brain and plasma levels of THC and metabolites, even when behavioral effects are similar. However, this does depend on species, route, timing of tissue collection. It would be helpful to describe the PK profile of males and females separately.<br /> 4. In Figure 5, it is unclear how the predicted i.p. THC dose could be 30 mg/kg when 30 mg/kg was not tested by the i.p. route according to the figure, and if it had been it would have likely been almost zero acoustic startle, not the increased startle that was observed in the 2 hr gel group. It seems more likely that it would be equivalent to 3 mg/kg i.p. Could there be an error in the modeling, or was it based on the model used for the triad effects? This should be clarified.

Reviewer #2 (Public Review):

Summary:<br /> The work fruitfully adds to the tools to study cannabinoid action and pharmacology specifically, but also this method is applicable to other drugs, in particular, if lipophilic in nature.

Strengths:<br /> The addition of chocolate flavor overcomes aversive reactions which are often experienced in pharmacological treatments, leading to possible caveats in the interpretation of the behavioral outcomes.

Weaknesses:<br /> Certainly, more THC mediated behavioral outcomes could have been tested, but the work presents a proof-of-concept study to investigate acute THC treatment.<br /> It would have been interesting if this application form is also possible for chronic treatment regimen.

Reviewer #3 (Public Review):

Summary: This manuscript explores the development of a rodent voluntary oral THC consumption model. The authors use the model to demonstrate that similar effect levels of THC can be observed to what has previously been described for i.p. THC administration.

Strengths: Overall this is an interesting study with compelling data presented. There is a growing need within the field of cannabinoid research to explore more 'realistic' routes of cannabinoid administration, such as oral consumption or inhalation. The evidence presented here shows the utility of this oral administration model.

Weaknesses: The main weaknesses of the manuscript revolve around clarification of the Methods section. All of these weaknesses are described in the "Recommendations to authors" section. Revising the manuscript would account for many of these weaknesses.

Author Response

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

Thank you very much for forwarding these two important reviews on our paper. Please find hereby our point-by-point responses addressing the ideas, arguments and points of concern raised by the reviewers. We provide explanation of how these points have been incorporated in the paper.

We feel the review process has been a useful exercise and that the paper has greatly benefited in terms of clarity and accessibility. It is our hope that our findings may ignite renewed interest on unexplored and “unexpected” aspects of great ape vocal communication, inspire novel research, and invite bold new advances on the long-standing puzzle of language origins and evolution. In several relevant sections, we have also sought to explicitly address the point of doubt raised in eLife’s editorial assessment, published alongside the reviewed preprint of our paper. The editorial assessment stated that “…However the evidence provided to support the major claims of the paper is currently incomplete. Specifically, it is not yet clear how the rhythmic structuring found in these long calls is more similar to human language recursion per se rather than isochrony as a broader, more common phenomenon.” To directly clarify this point, we provide now various examples of how recursion is distinct from repetition, using everyday objects for an intuitive understanding (e.g., lines 43-51). We have also expanded the discussion to better contextualise and clarify the implications of our findings on language evolution theory. We hope this will help addressing the implicit request for clarification in the previous editorial assessment.

Thank you very much for your kind and dedicated attention in the processing of our study.

Public Reviews:

Reviewer #1 (Public Review):

This study investigates the structuring of long calls in orangutans. The authors demonstrate long calls are structured around full pulses, repeated following a regular tempo (isochronic rhythm). These full pulses are themselves structured around different sub-pulses, themselves repeated following an isochronic rhythm. The authors argue this patterning is evidence for self-embedded, recursive structuring in orangutang long calls.

The analyses conducted are robust and compelling and they support the rhythmicity the authors argue is present in the long calls. Furthermore, the authors went above and beyond and confirmed acoustically the sub-categories identified were accurate.

We thank the reviewer for this important support regarding our methods and findings.

However, I believe the manuscript would benefit from a formal analysis of the specific recursive patterning occurring in the long call. Indeed, as of now, it is difficult for the reader to identify what the authors argue to be recursion and distinguish it from simple repetitions of motifs, which is essential.

We agree with the reviewer that the distinction between repetition and recursion is very important for the adequate interpretation of our findings. Following the reviewer’s point (and the Editorial Assessement), we have now rephrased several passages in the initial paragraph of the paper for added clarity, where recursion is introduced and explained. We now also provide various new examples of recursion in everyday life and popular culture to better illustrate in an easy and accessible way the fundamental nature of recursion. We then use two of these common examples (computer folders and Russian dolls) to specifically distinguish repetition from recursion.

Although the authors already discuss briefly why linear patterning is unlikely, the reader would benefit from expanding on this discussion section and clarifying the argument here (a lay terminology might help).

Corrected accordingly.

I believe an illustration here might help. In the same logic, I believe a tree similar to the trees used in linguistics to illustrate hierarchical structuring would help the reader understand the recursive patterning in place here. This would also help get the "big picture", as Fig 1A is depicting a frustratingly small portion of the long call.

We completely understand the reviewer’s concern here. As proposed by the reviewer, and in addition to changes in the Introduction (see above) and Discussion (see below), we have now added a new figure in the Discussion to help the reader get the “big picture” of our findings.

We have also made revisions throughout the Introduction and Discussion to simplify the text, clarify our exposition and facilitate the reader better and intuitively understand the nature and relevance of our results.

Notwithstanding these comments, this paper would provide crucial evidence for recursion in the vocal production of a non-human ape species. The implication it would have would represent a key shift in the field of language evolution. The study is very elegant and well-constructed. The paper is extremely well written, and the point of view adopted is original, well-argued and compelling.

We are humbled by the reviewer’s words, and we thank the reviewer for attributing these qualities to our paper. This feedback reassures us of the disruptive potential that these and similar future findings may have on our understanding of language evolution.

Reviewer #2 (Public Review):

I am not qualified to judge the narrow claim that certain units of the long calls are isochronous at various levels of the pulse hierarchy. I will assume that the modelling was done properly. I can however say that the broad claims that (i) this constitutes evidence for recursion in non-human primates, (ii) this sheds light on the evolution of recursion and/or language in humans are, when not made trivially true by a semantic shift, unsupported by the narrow claims. In addition, this paper contains errors in the interpretation of previous literature.

We report the first confirmed case of “vocal sequences within vocal sequences” in a wild nonhuman primate, namely a great ape. The currently prevailing models of language evolution often rest on the (purely theorical) premise that such structures do not exist in any animal bar humans. We find the discovery of such structures in a wild great ape exciting, remarkable, and promising. We regret that the reviewer does not share this sentiment with us. We feel that the statement that these findings are trivial and narrow is unfounded.

In order to clarify and better communicate the significance of our findings, we now explain in more detail in the Introduction and Discussion how the discovery of nested isochrony in wild orangutans promises to stimulate new series of studies in nature and captivity. Our findings dovetail nicely with previous captive studies that have shown that animals can learn how to recognise recursive patterns and invite new research efforts for the investigation of recursive abilities in the wild and in the absence of human priming and in nonhuman primates.

The main difficulty when making claims about recursion is to understand precisely what is meant by "recursion" (arguably a broader problem with the literature that the authors engage with). The authors offer some characterization of the concept which is vague enough that it can include anything from "celestial and planetary movement to the splitting of tree branches and river deltas, and the morphology of bacteria colonies". With this appropriately broad understanding, the authors are able to show "recursion" in orangutans' long calls. But they are, in fact, able to find it everywhere.

The reviewer is correct in highlighting that recursion is ubiquitous in nature and this is something that we explicitly state in the paper. This only makes it the more surprising that, when it comes to vocal combinatorics, recursion has only been described in human language and music, but in no other animals. If studies providing such evidence are known to reviewer, we kindly request their corresponding references.

In the new revised version, we have paid attention to this aspect raised by the reviewer, and we have sought to disambiguate that our observations pertain to temporal recursion. This clarification will hopefully allow a better understanding of our results.

The sound of a plucked guitar string, which is a sum of self-similar periodic patterns, count as recursive under their definition as well.

The example pointed out here by reviewer is factually correct; sound harmonics represent a recursive pattern of a fundamental frequency. (In fact, we explain this phenomenon in the Discussion.) The reviewer’s comment seems to offer an analogy to oscillatory phenomena in the physiology of the vocal folds, and so, it is misplaced with regards to our present study, which focused vocal sequences. Admittedly, this misinterpretation may have been implicitly caused by our wording and we apologise for this. We now refer to “vocal combinatorics” instead of “vocal production” throughout the paper to avoid the reader considering that our findings pertain to the physiology of the vocal folds.

One can only pick one's definition of recursion, within the context of the question of interest: evolution of language in humans. One must try to name a property which is somewhat specific to human language, and not a ubiquitous feature of the universe we live in, like self-similarity. Only after having carved out a sufficiently distinctive feature of human language, can we start the work of trying to find it in a related species and tracing its evolutionary history. When linguists speak of recursion, they speak of in principle unbounded nested structure (as in e.g., "the doctor's mother's mother's mother's mother ..."). The author seems to acknowledge this in the first line of the introduction: "the capacity to iterate a signal within a self-similar signal" (emphasis added). In formal language theory, which provides a formal and precise definition of one notion of recursivity appropriate for human language, unbounded iteration makes a critical difference: bounded "nested structures" are regular (can be parsed and generated using finite-state machines), unbounded ones are (often) context-free (require more sophisticated automaton). The hierarchy of pulses and sub-pulses only has a fixed amount of layers, moreover the same in all productions; it does not "iterate".

The reviewer explains here how recursion, in its fully fledged form in modern language(s), is defined by linguistics. We fully agree and do not contest such descriptions and definitions in any way. These descriptions and definitions aim to describe how recursion operates today, not how it evolved. Nor do these descriptions and definitions generate data-driven, testable predictions about precursors or proto-states of recursion as used by modern language-able humans. This is scientifically problematic and heuristically unsatisfying regarding the open question of language evolution.

Following human-specific definitions for recursion, as proposed by the reviewer, cannot per se be used to undertake a comparative approach to evolution because they leave nothing to compare recursion with in other (wild) species. Using human-specific definitions unavoidably leads to black-and-white notions that language is always absolutely present in humans and always absolutely absent in other animals, regardless of their degree of relatedness to humans. It is unpreventable that these descriptions flout foundational principles of evolution, such as descent with modification and shared ancestry.

This conceptual problem is not new. Less than a century ago, it was believed that humans were the only tool-user (thousands of examples are known today in nonhuman animals, including fish and invertebrates), and later, that humans were the only cultural animal (today it is known that migrating caribou and fruit flies can establish traditions based on social learning). We must follow in the footsteps of those who have helped redefine human nature in the past. As famously stated by Louis Leakey when presented with evidence for chimpanzee tool-use collected by Jane Goodall, “Now we must redefine tool, redefine man, or accept chimpanzees as human”. Therefore, as a matter of course, we must redefine recursion, embracing empirically (other than purely theoretically) definitions that allow recursion to take on forms and functions different from that of modern language-able humans.

Another point is that the authors don't show that the constraints that govern the shape of orangutans long calls are due to cognitive processes.

The reviewer is indeed correct. This does not, however, refute our findings. We do not directly show that cognitive processes govern recursion in orangutan long calls. Instead, we show that the observed patterns cannot be explained by simple bodily or motoric processes, excluding therefore low-level explanations. With more than 50 years of accumulated field experience in primatology, this was the only possible way that our team found to go about conducting research and analyses on natural behaviour, in the wild, with a critically endangered primate. We would be very interested in learning from the reviewer what ethical and non-invasive methods, specific locations in the wild, and type of behavioural or socio-ecological data could be otherwise viably used to demonstrate what the reviewer requests. If other scientists believe that the patterns observed in wild orangutan long calls – three independent, but simultaneously-occurring recursive motifs – can be generated based on low-level physiological mechanisms alone, the burden of proof resides with them.

Any oscillating system will, by definition, exhibit isochrony.

We disagree with this statement. The example provided above by the reviewer him/her-self disproves the statement: a guitar string when struck is an oscillating system but it is not isochronic nor is it combinatorial. Isochrony cannot be established with single events, only with event sequences (in practice, ideally >3).

For instance, human trills produce isochronouns or near isochronous pulses. No cognitive process is needed to explain this; this is merely the physics of the articulators. Do we know that the rhythm of the pulses and sub-pulses in orangutans is dictated by cognition as opposed to the physics of the articulators?

The reviewer seems to misinterpret our results here. Our focus is on vocal combinatorics, not vocal fold oscillation (see previous response). We have now reworded all instances where the text could be unclear.

Even granting the authors' unjustified conclusion that wild orangutans have "recursive" structures and that these are the result of cognition, the conclusions drawn by the authors are too often fantastic leaps of induction. Here is a cherry-picked list of some of the far-fetched conclusions: - "our findings indicate that ancient vocal patterns organized across nested structural strata were likely present in ancestral hominids". Does finding "vocal patterns organized across nested structural strata" in wild orangutans suggest that the same were present in ancestral hominids?

Following the reviewer’s comment, we have now rephrased and toned down this passage, stating that such structures “may have been present” in ancestral hominids. We are grateful to the reviewer for this comment.

• "given that isochrony universally governs music and that recursion is a feature of music, findings (sic.) suggest a possible evolutionary link between great ape loud calls and vocal music". Isochrony is also a feature of the noise produced by cicadas. Does this suggest an evolutionary link between vocal music and the noise of cicadas?

We apologise, but it is unclear what the reviewer is exactly suggesting or proposing here. It seems as though it is believed that cicadas are as phylogenetically related to humans as great apes are. Our last common ancestor with great apes diverged about 10mya, but with cicadas 600mya. The last common ancestor with great apes was a great ape (or hominid). The human-cicada last common ancestor would have looked like a worm (it is probable it would already have a nervous bulge at the head, or “brain”). In order to avoid similar misinterpretations, we have now clarified in several instances that our study and interpretation of results are based on shared ancestry within the Hominid family.

It seems that the reviewer may be also misinterpreting our findings. We do not simply report isochrony in a wild great ape (multiple references for isochronous calls in primate are provided in the Discussion). We report isochrony within isochrony in three non-exclusive rhythmic arrangements. In case the reviewer knows of a study on cicadas, or any non-human species, showing recursive sound combinatorics of this nature, we kindly request the citation. We can only hope that such new cases may be gradually unveiled in wild animals to help propel our general understanding of possible ways of how insipient recursive vocal combinatorics in ancient hominids could have given rise to recursion as used today by language-able modern humans.

Finally, some passages also reveal quite glaring misunderstandings of the cited literature. For instance:

• "Therefore, the search for recursion can be made in the absence of meaning-base operations, such as Merge, and more generally, semantics and syntax". It is precisely Chomsky's (disputable) opinion that the main operation that govern syntax, Merge, has nothing to do with semantics. The latter is dealt within a putative conceptual-intentional performance system (in Chomsky's terminology), which is governed by different operations.

Following the reviewer’s comment, we have now removed “meaning-base operations, such as Merge, and more generally” from the target sentence in order to avoid confusion. Thank you.

• "Namely, experimental stimuli have consisted of artificial recursive signal sequences organized along a single temporal scale (though not structurally linear), similarly with how Merge and syntax operate". The minimalist view advocated by Chomsky assumes that mapping a hierarchichal structure to a linear order (a process called linearizarion) is part of the articulatory-perceptual system. This system is likewise not governed by Merge and is not part of "syntax" as conceived by the Chomskyan minimalists.

Following the reviewer’s comment, we have not omitted the target sentence for added clarity.

Reviewer #1 (Recommendations For The Authors):

L55-67: I feel there is a step missing in the logic of the argumentation here. The studies cited by the authors here are mostly about syntactic-like structuring but not recursion. Hence when the authors mention in the next sentence that these studies investigate the perception of recursive signalling, it seems incorrect. I agree with the logic, but the references do not seem appropriate. I would further suggest that if there are no other references, that would make the introduction of the study here even easier: there is very little work investigating this capacity in non-human animals, let alone on a production perspective, therefore, the study conducted here is paramount and fills this important gap in the literature.

We are grateful to Reviewer #1 for these comments, and we are honoured to hear that our findings are filling a literature gap. We have now carefully revised the manuscript, hopefully, streamlining our line of reasoning and improving the paper’s overall readability. We agree that there is very little work investigating the spontaneous “production” of recursion in nonhuman animals. We decided to better detail the logic of our paper by clarifying the difference between recursion and repetition and clarifying that the motifs that we identify in wild orangutan represent a case of "temporal recursion".

L59: Johan J should be removed (same in discussion).

Removed, thanks.

L60: For example is repeated twice, here and L55.

We have rephrased this part of the manuscript, thanks.

L72-73: If we consider the Watson et al., 2020 study an example of recursive perception (which I do not think is true), this was conducted using a passive design - i.e. with no active training.

We have rephrased this part of the manuscript, thanks.

L240-241: Again, non-adjacent dependency processing does not equal recursion.

We agree that non-adjacent dependency processing does not equal recursion. We have now clarified this section accordingly.

L269: one of the most.

Corrected, thanks.

L296: add space after settings.

Corrected, thanks.

Reviewer #2 (Recommendations For The Authors):

In addition to the public portion of the review, I advise the authors' to substantially alter their style of writing. The language used is not accurate and the intended meaning is often not clear. This makes it hard for any reader to follow the authors' reasoning fully. Below I list only a few of the egregious examples but the examples abound:

• "this hints at a neuro-cognitive or neuro-computational transformation in the human brain" what meaning do the author assign to "neuro-cognitive" and "neuro-computational" ? what difference do they place between the two (so that they would be disjoined.) ? What "transformation" are we talking about ? From what to what ?

• " However, recursive signal structures can also unfold in other manners, such as across nested temporal scales and in the absence of semantics (Fitch, 2017a), as in music." what is meant here by nested temporal scales ?

• "The simultaneous occurrence of non-exclusive recursive patterns excludes the likelihood that orangutans concatenate long calls and their subunits in linear structure without any recursive processes": isn't there a more straightforward way to say "excludes the likelihood"? What is meant by "non-exclusive recursive patterns"?

It seems that Reviewer #2 does not share our writing style. Nonetheless, we have tried to meet the reviewer halfway, clarifying throughout the new revised version our definitions, our line of argument, our motivations, our results, the context of our findings in what is known about recursion in animals, and the implication of our discovery for language evolution theory.

eLife assessment

The paper represents a novel application of recursion theory to the long call vocalisations of orangutans to demonstrate repetitive, rhythmic sub-structuring. The authors use detailed acoustic analyses to show compelling evidence for self-embedded and nested isochronic motifs. These fundamental results have the potential to significantly advance current approaches used to compare nonhuman communication systems with human language.

Reviewer #1 (Public Review):

This study investigates the structuring of long calls in orangutans. The authors demonstrate long calls are structured around full pulses, repeated following a regular tempo (isochronic rhythm). These full pulses are themselves structured around different sub-pulses, themselves repeated following an isochronic rhythm. The authors argue this patterning is evidence for self-embedded, recursive structuring in orangutang long calls.

The analyses conducted are robust and compelling and they support the rhythmicity the authors argue is present in the long calls. Furthermore, the authors went above and beyond and confirmed acoustically the sub-categories identified were accurate.

Reviewer #3 (Public Review):

Summary: This paper presents evidence of recursive self-embedding in the vocalization structure of orangutans, using fine-grained acoustical analysis. It proves the existence of isochrony nested in isochrony in the motifs produced by a nonhuman vocal system.

Strengths: Very clear written, clear analysis, excellent responses to the Reviewers.

Weaknesses: Jargonous language may be reduced. A video showing the sound as it unfolds and the spectrogram (as in Fig 1A) of the long call could be useful to best exemplify the results.

eLife assessment

The work is a valuable contribution to understanding the mechanism of nuclear export of tRNA in budding yeast. The authors present solid evidence that Dbp5 functions in parallel with Los1 and Msn5 in tRNA export, in a manner dependent on Gle1 for activation of its ATPase activity but independently of Mex67, Dbp5's partner in mRNA export. It further presents biochemical evidence that Dbp5 can bind tRNA but that Gle1 and InsP6 are required for activating ATP hydrolysis by the Dbp5-tRNA complex, suggesting a possible mechanism for tRNA export by Dbp5.

Author Response

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

We would like to thank the Editors and Reviewers for their additional comments and constructive feedback on our manuscript. We have made minor adjustments to the figures and texts based on their suggestions, including improved images in Figure 1 and correction of figure labels.

Reviewer #1 (Public Review):

In their previous paper (Lari et al, 2019; Azra Lari Arvind Arul Nambi Rajan Rima Sandhu Taylor Reiter Rachel Montpetit Barry P Young Chris JR Loewen Ben Montpetit (2019) A nuclear role for the DEAD-box protein Dbp5 in tRNA export eLife 8:e48410.) as well as in the current manuscript the authors states that Dbp5 is involved in the export of tRNA that is independent of and parallel to Los1. They state that Dbp5 binds to the tRNA independent of known tRNA export proteins. The obtained conclusion is both intriguing and innovative, since it suggests that there are other variables, beyond the ones previously identified as tRNA factors, that might interact with Dbp5 to facilitate the export process. In order to find out additional factors aiding this process the authors may employ total RNA-associated protein purification (TRAPP) experiments ( Shchepachevto et al., 2019; Shchepachev V, Bresson S, Spanos C, Petfalski E, Fischer L, Rappsilber J, Tollervey D. Defining the RNA interactome by total RNA-associated protein purification. Mol Syst Biol. 2019 Apr 8;15(4):e8689. doi: 10.15252/msb.20188689. PMID: 30962360; PMCID: PMC6452921) to identify extra factors involved in conjunction with Dbp5. The process elucidates hitherto uninvestigated tRNA export components that function in conjunction with Dbp5.

Author Response: We greatly appreciate this suggestion and agree with the reviewer that identification of the composition of the export competent Dbp5 containing tRNA complex is a critical next step for understanding the mechanism of Dbp5 mediated tRNA export, which will form the foundation of a future investigation in the laboratory and warrants its own study.

Reviewer #1 (Public Review):

Various reports suggest that eukaryotic translation elongation factor 1 eEF1A is involved tRNA export Bohnsack et al., 2002 (Bohnsack MT, Regener K, Schwappach B, Saffrich R, Paraskeva E, Hartmann E, Görlich D. Exp5 exports eEF1A via tRNA from nuclei and synergizes with other transport pathways to confine translation to the cytoplasm. EMBO J. 2002 Nov 15;21(22):620515. doi: 10.1093/emboj/cdf613. PMID: 12426392; PMCID: PMC137205), Grosshans etal., 2002; Grosshans H, Hurt E, Simos G. An aminoacylation-dependent nuclear tRNA export pathway in yeast. Genes Dev. 2000 Apr 1;14(7):830-40. PMID: 10766739; PMCID: PMC316491). The presence of mutations in eEF1A has been seen to hinder the nuclear export process of all transfer RNAs (tRNAs). eEF1A has been shown to interact with Los1 aiding in tRNA export. The authors can also explore the crosstalk between Dbp5 and eEF1A in this study. Additionally, suppressor screening analysis in dbp5R423A , los1∆dbp5R423A los1∆msn∆dbp5R423A could shed more light on this.

Author Response: Thank you for this suggestion and raising an important possible role for Dbp5 in eEF1A mediated tRNA export. Based on more recent investigation of eEF1A function in tRNA export (PMID: 25838545), it is likely that eEF1A functions in re-export of charged tRNAs specifically (likely in conjunction with Msn5). The current manuscript has largely focused on the role of Dbp5 in pre-tRNA export, but a more careful mechanistic characterization of Dbp5 and re-export will be conducted in follow-up studies given the physical interaction between Dbp5 and spliced tRNAs we previously reported. Similarly, suppressor screens with the Dbp5 and los1Δmsn5Δ mutants will likely be a useful tool in identifying additional tRNA export factors and we thank the reviewer for this suggestion.

Reviewer #1 (Public Review):

The addition of Gle1 is potentially novel but it's unclear why the authors didn't address the potential involvement of IP6.

Author Response: The text has been revised to highlight the importance of InsP6 in Gle1 mediated activation of Dbp5. This includes referencing InsP6 throughout the manuscript during discussions of Gle1 activation of Dbp5 and lines 401-404 discussing the potential role for the small molecule in regulating mRNA and tRNA export in different cellular contexts (e.g., stress and disease).

Reviewer #1 (Public Review):

This study focuses on the defining cellular pathways critical for tRNA export from the nucleus. While a number of these pathways have been identified, the observation that the primary transport receptors identified thus far (Los1 and Msn5) are not essential and that cells are viable even when both the genes are deleted supports the idea that there are as yet unidentified mediators of tRNA export from the nucleus. This study implicates the helicase Dbp5 in one of these parallel pathways arguing that Dbp5 works in a pathway that is independent of Los1 and/or Msn5. The authors present genetic data to support this conclusion. At least one results suggests that the idea of these pathways in parallel may be too simplistic as deletion of the LOS1 gene, which is not essential decreases the interaction of tRNA export substrate with Dbp5 (Figure 2A). If the two pathways were working in parallel, one might have expected removing one pathways to lead to an increase in the use of the other pathway and hence the interaction with a receptor in that pathway. The authors provide solid evidence that Dbp5 interacts with tRNA directly and that addition of the factor Gle1 together with the previously identified co-factor InsP6 can trigger helicase activity and release of tRNA. The combination of in vivo studies and biochemistry provide evidence to consider how Dbp5 contributes to export of tRNA and more broadly adds to the conversation about how coding and non-coding RNA export from the nucleus might be coordinated to control cell physiology.

Reviewer #2 (Public Review):

In the manuscript by Rajan et al., the authors have highlighted the direct interaction between Dbp5 and tRNA, wherein Dbp5 serves as a mediator for tRNA export. This export process is subject to spatial regulation, as Dbp5 ATPase activation occurs specifically at nuclear pore complexes. Notably, this regulation is independent of the Los1-mediated pre-tRNA export route and instead relies on Gle1. The manuscript is well constructed and nicely written.

eLife assessment

This study presents valuable findings on the contrasting responses of two bacteria to the phytoplankton-derived compound azelaic acid. Metabolomics and transcriptomics evidence convincingly shows the assimilation pathway in one marine bacterium and a stress response in a second bacterium. The study provides evidence that azelaic acid can alter marine microbial community structure in mesocosm experiments, though the mechanisms underlying this shift in community structure remain to be explored in future studies.

Author Response

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

Shibl et al., studied the possible role of dicarboxylate metabolite azelaic acid (Aze) in modulating the response of different bacteria, it was used as a carbon source by Phycobacter and possibly toxic for Alteromonas. The experiments were well conducted using transcriptomics, transcriptional factor coexpression networks, uptake experiments, and chemical methods to unravel the uptake, catabolism, and toxicity of Aze on these two bacteria. They identified a putative Aze TRAP transporter in bacteria and showed that Aze is assimilated through fatty acid degradation in Phycobacter. Meanwhile, in Alteromonas it is suggested that Aze inhibits the ribosome and/or protein synthesis, and that efflux pumps shuttles Aze outside the cytoplasm. Further on, they demonstrate that seawater amended with Aze selects for microbes that can catabolize Aze.

Major strengths:

The manuscript is well written and very clear. Through the combination of gene expression, transcriptional factor co-expression networks, uptake experiments, and chemical methods Shibl et al., showed that Aze has a different response in two bacteria.

Major weakness:

There is no confirmation of the Aze TRAP transporters through mutagenesis.

Impact on the field:

Metabolites exert a significant influence on microbial communities in the ocean, playing a crucial role in their composition, dynamics, and biogeochemical cycles. This research highlights the intriguing capacity of a single metabolite to induce contrasting responses in distinct bacterial species, underscoring its role in shaping microbial interactions and ecosystem functions.

We thank the reviewer for their comments on the paper and we appreciate their suggestion to confirm the activity of Aze TRAP transporters through mutagenesis. We agree that this would be a valuable addition to the study, and we mention in the text that “Despite numerous attempts, our efforts to knock-out azeTSL in Phycobacter failed.”

The success rate of mutagenesis experiments is often low and time-consuming. There are a few reasons why our knock-out experiments with Phycobacter have not been successful. Despite using several modified protocols for electroporation, no Phycobacter colonies grew on the antibiotic plate. We then tried the homologous recombination approach for conjugation but were not successful in selecting for Phycobacter cells, even when grown in high salinity conditions that favor Phycobacter and disfavor the carrier, E. coli . While we would love to include a mutagen to confirm the function of this cluster, the task seems to be unattainable at the moment .

Reviewer #2 (Public Review):

This study explores the breadth of effects of one important metabolite, azelaic acid, on marine microbes, and reveals in-depth its pathway of uptake and catabolism in one model bacterial strain. This compound is known to be widely produced by phytoplankton and plants, and to have complex effects on associated microbiomes.

This work uses transcriptomics to assay the response of two strains that show contrasting responses to the metabolite: one catabolizes the compound and assimilates the carbon, while the other shows growth inhibition and stress response. A highly induced TRAP transporter, adjacent to a previously identified regulator, is inferred to be the specific uptake system for azelaic acid. However the transport function was not directly tested via genetic or biochemical methods. Nevertheless, this is a significant finding that will be useful for exploring the distribution of azelaic acid uptake capability across metagenomes and other bacteria.

The authors use pulse-chase style metabolomics experiments to beautifully demonstrate the fate of azelaic acid through catabolic pathways. They also measure an assimilation rate per cell, though it remains unclear how this measured rate relates to natural systems. The metabolomics approach is an elegant way to show carbon flux through cells, and could serve as a model for future studies.

The study seeks to extend the results from two model strains to complex communities, using seawater mesocosm experiments and soil/Arabidopsis experiments. The seawater experiments show a community shift in mesocosms with added azelaic acid. However, the mechanisms for the shift were not determined; further work is necessary to demonstrate which community members are directly assimilating the compound vs. benefitting indirectly or experiencing inhibition. In my opinion the soil and Arabidopsis experiments are quite preliminary. I appreciate the authors' desire to broaden the scope beyond marine systems, but I believe any conclusions regarding different modes of action in aquatic vs terrestrial microbial communities are speculative at this stage.

This work is a nice illustration of how we can begin to tease apart the effects of chemical currencies on marine ecosystems. A key strength of this work is the combination of transcriptomics and metabolomics methods, along with assaying the impacts of the metabolite on both model strains of bacteria and whole communities. Given the sheer number of compounds that probably play critical roles in community interactions, a key challenge for the field will be navigating the tradeoffs between breadth and depth in future studies of metabolite impacts. This study offers a good compromise and will be a useful model for future studies.

We thank the reviewer for their thoughtful comments on the manuscript. We appreciate their feedback on the breadth of effects of Aze on marine microbes, and their insights into the strengths and limitations of our study.

We agree that the specific mechanisms underlying community-level shifts in seawater mesocosm experiments with added Aze are not yet fully understood and we believe such work is beyond the scope of this paper and warrants an in-depth study of its own. This can perhaps be conducted at a larger scale by using a combination of meta-omics and targeted enrichment to identify the community members directly assimilating Aze, as well as those that are benefitting indirectly or experiencing inhibition.

We also agree that the soil and Arabidopsis experiments are exploratory. However, we believe that these experiments are a valuable first step in highlighting the potential for Aze to have different modes of action in aquatic versus terrestrial microbial communities. Our interest in contrasting bacterial molecular responses in terrestrial plant rhizospheres and marine algal phycospheres stems from the fact that both environments share similar molecules and related bacteria, yet exhibit significantly different evolutionary histories and fluid dynamic profiles (Seymour et al 2017, Nature Microbiol ). Although more is known about Aze in Arabidopsis than phytoplankton, there are still gaps in this knowledge. For example, recent work has shown that Aze and derivatives can be secreted into soil (Korenblum et al 2020, PNAS ), but whether Aze directly influences microbial communities in soil as we have shown in seawater has not been explored. Thus, we feel our preliminary experiments in soil are important to provide such a distinction with seawater. Additional studies in these systems to further investigate the importance of Aze, which were beyond the scope of this current work, would be quite beneficial.

Reviewer #1 (Recommendations For The Authors):

General comments:

A complete supplemental file of differentially expressed genes should be provided in the supplemental. Please add tables with the entire DESeq output for Aze additions in the genomes of Phycobacter (0.5 and 8 h) and Alteromonas (0.5 h). While it makes sense to focus the paper on Aze related genes, the full dataset should be made available in a more curated form than just the raw reads in the SRA.

We thank the reviewer for this suggestion. We have included three more sheets in Supplementary Table 1 file where readers can find the entire DESeq outputs of Phycobacter (0.5 and 8 h) and Alteromonas (0.5 h) experiments.

Specific comments:

• L82 indicates the TRAP transporter for Aze. Looking at the table for gene expression of Phycobacter there are 26 significantly enriched transport genes at 0.5 h other than the putative Aze TRAP transporter. Even though the TRAP transporter is likely transporting Aze, it would be good to let the readers know that other transporters showed transcript enrichment.

Thank you for this helpful comment. We modified the sentence accordingly to read as follows: “Among 26 enriched transporter genes in our dataset, a C 4 - dicarboxylate tripartite ATP-independent periplasmic (TRAP) transporter substrate-binding protein (INS80_RS11065) was the most and the third most upregulated gene in Phycobacter grown on Aze at 0.5 and 8 hours, respectively.”

• Figure 1: There are many genes enriched from -1 to 1. Is there a cut off, p-val (can you add it to the caption)? It would be good to have a dashed line or something that indicates the -1 and 1 log2 fold change in the figure.

We thank the reviewer for this suggestion. We added the following sentence to the legend of Fig. 1: “Genes were considered DE with a p -adjusted value of < 0.05 and a log2 fold-change of ≥ ±0.50.”

• Supplementary tables: Add a title on all the supplementary tables. It's hard to tell what each one of the tables means without looking at the text and content of each tables.

A short descriptive title is now added to all supplementary tables.

• Not sure if it matters, though Table S1 was not available in the attached files, though it is in the complete pdf.

Table S1 is now in the attached files and the DESeq output has been added to it as suggested in the general comment above.

Reviewer #2 (Recommendations For The Authors):

Here I offer some more specific suggestions and comments on the methods and presentation.

I recommend being careful throughout with the language regarding conclusions. For instance, the study does not directly demonstrate the activity of the TRAP transporter (as mentioned above), and does not directly demonstrate that the bacteria that increase in abundance in the mesocosm experiments are actually assimilating azelaic acid.

We thank the reviewer for this comment. We agree that further studies are required to get definitive answers regarding the direct activity of the transporter genes and direct assimilation of Aze by bacteria in the mesocosm. These complex experiments would require establishing a reproducible workflow for knocking out genes and further isotope labeling experiments to track Aze assimilation in a natural setting. To that end, we were keen on using language throughout the manuscript indicating that transporter activity is putative. We went through the manuscript again to make sure it was clear that the transporter activity is putative at this time and is not confirmed. For the mesocosms, we cannot rule out that the changes in community structure is not due to other factors besides Aze. We have added this sentence in the discussion of the mesocosm experiments to indicate that the observed changes in microbial community cannot be directly attributed to Aze activity and may be a byproduct of other mechanisms.

Additionally, I find the soil and plant experiments to be very preliminary, and would personally recommend removing them from the manuscript. This is of course the authors' choice, but I find they detract from an otherwise more solid story. I wonder whether 16 hours was sufficient to see community changes and whether adding azelaic acid directly into the plant is necessary or relevant. The study does not measure any plant immune responses so I caution against drawing conclusions about the mechanism. It seems the connection to plant immunity was already shown in the literature, in which case I'm not sure whether these experiments presented here really add anything new to the paper.

We thank the reviewer for these comments. Our 16-hour sampling time point (similar to the seawater experiment) represents an overnight incubation period that should allow sufficient change in the natural microbial composition yet avoids the long-term succession of microbes with high metabolic capacities that may outcompete the rest of the community at long incubation periods. Deciding on this length of incubation was also informed by the uptake rate of Aze and its influence on either bacteria assimilating it as a carbon source or being inhibited by it.

Since no significant changes were observed in the soil, it was necessary to test the hypothesis that the plant host might be indirectly influencing the rhizosphere microbial communities by infiltrating A. thaliana leaves with Aze. As the reviewer mentions, the association between Aze and plant immunity was previously shown; however, the overall influence on the microbial community has not been fully explored yet. The soil and plant experiments were meant to serve an exploratory purpose and we find them necessary to keep in the manuscript as a first step in comparing the mode of action of Aze within marine and terrestrial ecosystems. They are by no means the answer to what role Aze plays in soil systems, but rather they are the starting point. We hope that our results encourage some readers to investigate similar common metabolites to further elucidate the molecular underpinnings of microbial modulation in both environments.

Regarding the transcriptomics data, I am not clear on why the "expression ratio" -- i.e. the fraction of pathway genes that were differentially abundant -- was used. I would not expect all transcripts in a pathway to behave the same way in response to a perturbation, due to variation in half-life/stability, post-transcriptional and post-translational regulation, etc. I recommend removing the expression ratio (right panel) from Figure 1. The left panel shows the data more clearly and more directly.

We thank the reviewer for their insight and we agree that not all transcripts in a pathway behave the same way. However, we find the expression ratio panel visually informative to highlight the importance of a pathway in response to Aze, taking into consideration the total number of key genes involved in a pathway. For example, despite the larger number of DE genes associated with the Amino Acid Metabolism & Degradation pathway compared to the Fatty Acid Degradation pathway, the expression ratio for the former in each transcriptome is lower than its Fatty Acid Degradation counterpart, indicating that the response of key fatty acid degradation genes to Aze is more pronounced. We have qualified the reasons for including expression ratios in Figure 1 legend.

Overall I enjoyed reading the manuscript and applaud the authors on a nice contribution to this important field.

Reviewer #1 (Public Review):

Summary:<br /> Shibl et al., studied the possible role of dicarboxylate metabolite azelaic acid (Aze) in modulating the response of different bacteria, it was used as a carbon source by Phycobacter and possibly toxic for Alteromonas. The experiments were well conducted using transcriptomics, transcriptional factor coexpression networks, uptake experiments, and chemical methods to unravel the uptake, catabolism, and toxicity of Aze on these two bacteria. They identified a putative Aze TRAP transporter in bacteria and showed that Aze is assimilated through fatty acid degradation in Phycobacter. Meanwhile, in Alteromonas it is suggested that Aze inhibits the ribosome and/or protein synthesis, and that efflux pumps shuttles Aze outside the cytoplasm. Further on, they demonstrate that seawater amended with Aze selects for microbes that can catabolize Aze.

Major strengths:<br /> The manuscript is well written and very clear. Through the combination of gene expression, transcriptional factor co-expression networks, uptake experiments, and chemical methods Shibl et al., showed that Aze has a different response in two bacteria.

Major weakness:<br /> There is no phenotype confirmation of the Aze TRAP transporters through mutagenesis.

Impact on the field:<br /> Metabolites exert a significant influence on microbial communities in the ocean, playing a crucial role in their composition, dynamics, and biogeochemical cycles. This research highlights the intriguing capacity of a single metabolite to induce contrasting responses in distinct bacterial species, underscoring its role in shaping microbial interactions and ecosystem functions.

Reviewer #2 (Public Review):

This study explores the breadth of effects of one important metabolite, azelaic acid, on marine microbes, and reveals in-depth its pathway of uptake and catabolism in one model bacterial strain. This compound is known to be widely produced by phytoplankton and plants, and to have complex effects on associated microbiomes.

This work uses transcriptomics to assay the response of two strains that show contrasting responses to the metabolite: one catabolizes the compound and assimilates the carbon, while the other shows growth inhibition and stress response. A highly induced TRAP transporter, adjacent to a previously identified regulator, is inferred to be the specific uptake system for azelaic acid, though this function was not directly tested via genetic or biochemical methods. Nevertheless, this is a significant finding that will be useful for exploring the distribution of azelaic acid uptake capability across metagenomes and other bacteria.

The authors use pulse-chase style metabolomics experiments to beautifully demonstrate the fate of azelaic acid through catabolic pathways. They also measure an assimilation rate per cell, though it remains unclear how this measured rate relates to natural systems. The metabolomics approach is an elegant way to show carbon flux through cells, and could serve as a model for future studies.

The study seeks to extend the results from two model strains to complex communities, using seawater mesocosm experiments and soil/Arabidopsis experiments. The seawater experiments show a community shift in mesocosms with added azelaic acid. The mechanisms for the shift were not determined in this study; further work is necessary to demonstrate which community members are directly assimilating the compound, benefitting indirectly, or experiencing inhibition. The authors also took the unusual and creative step of performing similar experiments in a soil - Arabidopsis system. I admire the authors' desire to identify unifying themes across ecosystems. The parallels are intriguing, and future experiments could determine the different modes of action in aquatic vs terrestrial microbial communities.

This work is a nice illustration of how we can begin to tease apart the effects of chemical currencies on marine ecosystems. A key strength of this work is the combination of transcriptomics and metabolomics methods, along with assaying the impacts of the metabolite on model strains of bacteria and whole communities. Given the sheer number of compounds that probably play critical roles in community interactions, a key challenge for the field will be navigating the tradeoffs between breadth and depth in future studies of metabolite impacts. This study offers a good compromise and will be a useful model for future studies.

Author Response

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

We agree with the reviewer that the statistics are buried in a dense excel file without a read-me page. We will address this by making a summary excel page for p-values during the production process.

---

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

eLife assessment

This important study uses genomically-engineered glypican alleles to demonstrate convincingly that Dally (not Dally-like protein [Dlp]) is the key contributor to formation of the Dpp/BMP morphogen gradient in the wing disc of Drosophila. The authors provide solid genetic evidence that, surprisingly, the core domain of Dally appears to suffice to trap Dpp at the cell surface. They conclude with a model according to which Dally modulates the range of Dpp signaling by interfering with Dpp's internalization by the Dpp receptor Thickveins.

Public Reviews:

Reviewer #1 (Public Review):

How morphogens spread within tissues remains an important question in developmental biology. Here the authors revisit the role of glypicans in the formation of the Dpp gradient in wing imaginal discs of Drosophila. They first use sophisticated genome engineering to demonstrate that the two glypicans of Drosophila are not equivalent despite being redundant for viability. They show that Dally is the relevant glypican for Dpp gradient formation. They then provide genetic evidence that, surprisingly, the core domain of Dally suffices to trap Dpp at the cell surface (suggesting a minor role for GAGs). They conclude with a model that Dally modulates the range of Dpp signaling by interfering with Dpp's degradation by Tkv. These are important conclusions, but more independent (biochemical/cell biological) evidence is needed.

As indicated above, the genetic evidence for the predominant role of Dally in Dpp protein/signalling gradient formation is strong. In passing, the authors could discuss why overexpressed Dlp has a negative effect on signaling, especially in the anterior compartment. The authors then move on to determine the role of GAG (=HS) chains of Dally. They find that in an overexpression assay, Dally lacking GAGs traps Dpp at the cell surface and, counterintuitively, suppresses signaling (fig 4 C, F). Both findings are unexpected and therefore require further validation and clarification, as outlined in a and b below.

a. In loss of function experiments (dallyDeltaHS replacing endogenous dally), Dpp protein is markedly reduced (fig 4R), as much as in the KO (panel Q), suggesting that GAG chains do contribute to trapping Dpp at the cell surface. This is all the more significant that, according to the overexpression essays, DallyDeltaHS seems more stable than WT Dally (by the way, this difference should also be assessed in the knock-ins, which is possible since they are YFP-tagged). The authors acknowledge that HS chains of Dally are critical for Dpp distribution (and signaling) under physiological conditions. If this is true, one can wonder why overexpressed dally core 'binds' Dpp and whether this is a physiologically relevant activity.

According to the overexpression assay, DallyDeltaHS seems more stable than WT Dally (Fig. 4B’, E’, 5A’, B’). As the reviewer suggested, we addressed the difference using the two knock-in alleles and found that DallyDeltaHS is more stable than WT Dally (Fig.4 L, M inset), further emphasizing the insufficient role of core protein of Dally for extracellular Dpp distribution.

In summary, we showed that, although Dally interacts with Dpp mainly through its core protein from the overexpression assay (Fig. 4E, I), HS chains are essential for extracellular Dpp distribution (Fig. 4R). Thus, the core protein of Dally alone is not sufficient for extracellular Dpp distribution under physiological conditions. These results raise a question about whether the interaction of core protein of Dally with Dpp is physiologically relevant. Since the increase of HS upon dally expression but not upon dlp expression resulted in the accumulation of extracellular Dpp (Fig. 2) and this accumulation was mainly through the core protein of Dally (Fig. 4E, I), we speculate that the interaction of the core protein of Dally with Dpp gives ligand specificity to Dally under physiological conditions.

To understand the importance of the interaction of core protein of Dally with Dpp under physiological conditions, it is important to identify a region responsible for the interaction. Our preliminary results overexpressing a dally mutant lacking the majority of core protein (but keeping the HS modified region intact) showed that HS chains modification was also lost. Although this is consistent with our results that enzymes adding HS chains also interact with the core protein of Dally (Fig. 4D), the dally mutant allele lacking the core protein would hamper us from distinguishing the role of core protein of Dally from HS chains.

Nevertheless, we can infer the importance of the interaction of core protein of Dally with Dpp using dally[3xHA-dlp, attP] allele, where dlp is expressed in dally expressing cells. Since Dally-like is modified by HS chains but does not interact with Dpp (Fig. 2, 4), dally[3xHA-dlp, attP] allele mimics a dally allele where HS chains are properly added but interaction of core protein with Dpp is lost. As we showed in Fig.3O, S, the allele could not rescue dallyKO phenotypes, consistent with the idea that interaction of core protein of Dally with Dpp is essential for Dpp distribution and signaling and HS chain alone is not sufficient for Dpp distribution.

b. Although the authors' inference that dallycore (at least if overexpressed) can bind Dpp. This assertion needs independent validation by a biochemical assay, ideally with surface plasmon resonance or similar so that an affinity can be estimated. I understand that this will require a method that is outside the authors' core expertise but there is no reason why they could not approach a collaborator for such a common technique. In vitro binding data is, in my view, essential.

We agree with the reviewer that a biochemical assay such as SPR helps us characterize the interaction of core protein of Dally and Dpp (if the interaction is direct), although the biochemical assay also would not demonstrate the interaction under the physiological conditions.

However, SPR has never been applied in the case of Dpp, probably because purifying functional refolded Dpp dimer from bacteria has previously been found to be stable only in low pH and be precipitated in normal pH buffer (Groppe J, et al., 1998)(Matsuda et al., 2021). As the reviewer suggests, collaborating with experts is an important step in the future.

Nevertheless, SPR was applied for the interaction between BMP4 and Dally (Kirkpatrick et al., 2006), probably because BMP4 is more stable in the normal buffer. Although the binding affinity was not calculated, SPR showed that BMP4 directly binds to Dally and this interaction was only partially inhibited by molar excess of exogenous HS, suggesting that BMP4 can interact with core protein of Dally as well as its HS chains. In addition, the same study applied Co-IP experiments using lysis of S2 cells and showed that Dpp and core protein of Dally are co-immunoprecipitated, although it does not demonstrate if the interaction is direct.

In a subsequent set of experiments, the authors assess the activity of a form of Dpp that is expected not to bind GAGs (DppDeltaN). Overexpression assays show that this protein is trapped by DallyWT but not dallyDeltaHS. This is a good first step validation of the deltaN mutation, although, as before, an invitro binding assay would be preferable.

Our overexpression assays actually showed that DppDeltaN is trapped by DallyWT and by dallyDeltaHS at similar levels (Fig. 5C), indicating that interaction of DppDeltaN and HS chains of Dally is largely lost but DppDeltaN can still interact with core protein of Dally.

We thank the reviewer for the suggesting the in vitro experiment. Although we decided not to develop biophysical experiments such as SPR for Dpp in this study due to the reasons discussed above, we would like to point out that our result is consistent with a previous Co-IP experiment using S2 cells showing that DppDeltaN loses interaction with heparin (Akiyama2008).

However, in contrast to our results, the same study also proposed by Co-IP experiments using S2 cells that DppDeltaN loses interaction with Dally (Akiyama2008). Although it is hard to conclude since western blotting was too saturated without loading controls and normalization (Fig. 1C in Akiyama 2008), and negative in vitro experiments do not necessarily demonstrate the lack of interaction in vivo. One explanation why the interaction was missed in the previous study is that some factors required for the interaction of DppDeltaN with core protein of Dally are missing in S2 cells. In this case, in vivo interaction assay we used in this study has an advantage to robustly detect the interaction.

Nevertheless, the authors show that DppDeltaN is surprisingly active in a knock-in strain. At face value (assuming that DeltaN fully abrogates binding to GAGs), this suggests that interaction of Dpp with the GAG chains of Dally is not required for signaling activity. This leads to authors to suggest (as shown in their final model) that GAG chains could be involved in mediating the interactions of Dally with Tkv (and not with Dpp. This is an interesting idea, which would need to be reconciled with the observation that the distribution of Dpp is affected in dallyDeltaHS knock-ins (item a above). It would also be strengthened by biochemical data (although more technically challenging than the experiments suggested above). In an attempt to determine the role of Dally (GAGs in particular) in the signaling gradient, the paper next addresses its relation to Tkv. They first show that reducing Tkv leads to Dpp accumulation at the cell surface, a clear indication that Tkv normally contributes to the degradation of Dpp. From this they suggest that Tkv could be required for Dpp internalisation although this is not shown directly. The authors then show that a Dpp gradient still forms upon double knockdown (Dally and Tkv). This intriguing observation shows that Dally is not strictly required for the spread of Dpp, an important conclusion that is compatible with early work by Lander suggesting that Dpp spreads by free diffusion. These result show that Dally is required for gradient formation only when Tkv is present. They suggest therefore that Dally prevents Tkv-mediated internalisation of Dpp. Although this is a reasonable inference, internalisation assays (e.g. with anti-Ollas or anti-HA Ab) would strengthen the authors' conclusions especially because they contradict a recent paper from the Gonzalez-Gaitan lab.

Thanks for suggesting the internalization assay. As we discussed in the discussion, our results suggest that extracellular Dpp distribution is severely reduced in dally mutants due to Tkv mediated internalization of Dpp (Fig. 6). Thus, extracellular Dpp available for labelling with nanobody is severely reduced in dally mutants, which can explain the reduced internalization of Dpp in dally mutants in the internalization assay. Therefore, we think that the nanobody internalization assay would not distinguish the two contradicting possibilities.

The paper ends with a model suggesting that HS chains have a dual function of suppressing Tkv internalisation and stimulating signaling. This constitutes a novel view of a glypican's mode of action and possibly an important contribution of this paper. As indicated above, further experiments could considerably strengthen the conclusion. Speculation on how the authors imagine that GAG chains have these activities would also be warranted.

Thank you very much!

Reviewer #2 (Public Review):

The authors are trying to distinguish between four models of the role of glypicans (HSPGs) on the Dpp/BMP gradient in the Drosophila wing, schematized in Fig. 1: (1) "Restricted diffusion" (HSPGs transport Dpp via repetitive interaction of HS chains with Dpp); (2) "Hindered diffusion" (HSPGs hinder Dpp spreading via reversible interaction of HS chains with Dpp); (3) "Stabilization" (HSPGs stabilize Dpp on the cell surface via reversible interaction of HS chains with Dpp that antagonizes Tkv-mediated Dpp internalization); and (4) "Recycling" (HSPGs internalize and recycle Dpp).

To distinguish between these models, the authors generate new alleles for the glypicans Dally and Dally-like protein (Dlp) and for Dpp: a Dally knock-out allele, a Dally YFP-tagged allele, a Dally knock-out allele with 3HA-Dlp, a Dlp knock-out allele, a Dlp allele containing 3-HA tags, and a Dpp lacking the HS-interacting domain. Additionally, they use an OLLAS-tag Dpp (OLLAS being an epitope tag against which extremely high affinity antibodies exist). They examine OLLAS-Dpp or HA-Dpp distribution, phospho-Mad staining, adult wing size.

They find that over-expressed Dally - but not Dlp - expands Dpp distribution in the larval wing disc. They find that the Dally[KO] allele behaves like a Dally strong hypomorph Dally[MH32]. The Dally[KO] - but not the Dlp[KO] - caused reduced pMad in both anterior and posterior domains and reduced adult wing size (particularly in the Anterior-Posterior axis). These defects can be substantially corrected by supplying an endogenously tagged YFP-tagged Dally. By contrast, they were not rescued when a 3xHA Dlp was inserted in the Dally locus. These results support their conclusion that Dpp interacts with Dally but not Dlp.

They next wanted to determine the relative contributions of the Dally core or the HS chains to the Dpp distribution. To test this, they over-expressed UAS-Dally or UAS-Dally[deltaHS] (lacking the HS chains) in the dorsal wing. Dally[deltaHS] over-expression increased the distribution of OLLAS-Dpp but caused a reduction in pMad. Then they write that after they normalize for expression levels, they find that Dally[deltaHS] only mildly reduces pMad and this result indicates a major contribution of the Dally core protein to Dpp stability.

Thanks for the comments. We actually showed that compared with Dally overexpression, Dally[deltaHS] overexpression only mildly reduces extracellular Dpp accumulation (Fig. 4I). This indicates a major contribution of the Dally core protein to interaction with Dpp, although the interaction is not sufficient to sustain extracellular Dpp distribution and signaling gradient.

The "normalization" is a key part of this model and is not mentioned how the normalization was done. When they do the critical experiment, making the Dally[deltaHS] allele, they find that loss of the HS chains is nearly as severe as total loss of Dally (i.e., Dally[KO]). Additionally, experimental approaches are needed here to prove the role of the Dally core.

Since the expression level of Dally[deltaHS] is higher than Dally when overexpressed, we normalized extracellular Dpp distribution (a-Ollas staining) against GFP fluorescent signal (Dally or Dally[deltaHS]). To do this, we first extracted both signal along the A-P axis from the same ROI in the previous version. The ratio was calculated by dividing the intensity of a-Ollas staining with the intensity of GFP fluorescent signal at a given position x. The average profile from each normalized profile was generated and plotted using the script described in the method (wingdisc_comparison.py) as other pMad or extracellular staining profiles.

Although this analysis provides normalized extracellular Dpp accumulation at different positions along the A-P axis, we are more interested in the total amount of Dpp or DppDeltaN accumulation upon Dally or dallyDeltaHS expression. Therefore, in the revised ms, we decided to normalize total amount of extracellular Dpp against the level of Dally or Dally[deltaHS] by dividing total signal intensity of extracellular Dpp staining (ExOllas staining) by total GFP fluorescent signal (Dally or Dally[deltaHS]) around the Dpp producing cells in each wing disc. Statistical analysis showed that accumulation of extracellular Dpp is only slightly reduced without HS chains (Fig.4I), indicating that Dally interacts with Dpp mainly through its core protein.

We agree with the reviewer that additional experimental approaches are needed to address the role of the core protein of Dally. As we discussed in the response to the reviewer1, to understand the importance of the interaction of core protein of Dally with Dpp, it is important to identify a region responsible for the interaction. Our preliminary results overexpressing a dally mutant lacking the majority of core protein (but keeping the HS modified region intact) showed that HS chains modification was also lost. Although this is consistent with our results that enzymes adding HS chains also interact with the core protein of Dally (Fig. 4D), the dally mutant allele lacking the core protein would hamper us from distinguishing the role of the core protein of Dally from HS chains.

Nevertheless, we can infer the importance of the interaction of core protein of Dally with Dpp using dally[3xHA-dlp, attP] allele, where dlp is expressed in dally expressing cells. Since Dally-like is modified by HS chains but does not interact with Dpp (Fig. 2, 4), dally[3xHA-dlp, attP] allele mimics a dally allele where HS chains are properly added but interaction of core protein with Dpp is lost. As we showed in Fig.3O, S, the allele could not rescue dallyKO phenotypes, consistent with the idea that interaction of core protein of Dally with Dpp is essential for Dpp distribution and signaling.

Prior work has shown that a stretch of 7 amino acids in the Dpp N-terminal domain is required to interact with heparin but not with Dpp receptors (Akiyama, 2008). The authors generated an HA-tagged Dpp allele lacking these residues (HA-dpp[deltaN]). It is an embryonic lethal allele, but they can get some animals to survive to larval stages if they also supply a transgene called “JAX” containing dpp regulatory sequences. In the JAX; HA-dpp[deltaN] mutant background, they find that the distribution and signaling of this Dpp molecule is largely normal. While over-expressed Dally can increase the distribution of HA-dpp[deltaN], over-expression of Dally[deltaHS] cannot. These latter results support the model that the HS chains in Dally are required for Dpp function but not because of a direct interaction with Dpp.

Our overexpression assays actually showed that both Dally and Dally[deltaHS] can accumulate Dpp upon overexpression and the accumulation of Dpp is comparable after normalization (Fig. 5C), consistent with the idea that interaction of DppdeltaN and HS chains are largely lost. As the reviewer pointed out, these results support the model that the HS chains in Dally are required for Dpp function but not because of a direct interaction with Dpp.

In the last part of the results, they attempt to determine if the Dpp receptor Thickveins (Tkv) is required for Dally-HS chains interaction. The 2008 (Akiyama) model posits that Tkv activates pMad downstream of Dpp and also internalizes and degrades Dpp. A 2022 (Romanova-Michaelides) model proposes that Dally (not Tkv) internalizes Dpp.

To distinguish between these models, the authors deplete Tkv from the dorsal compartment of the wing disc and found that extracellular Dpp increased and expanded in that domain. These results support the model that Tkv is required to internalize Dpp.

They then tested the model that Dally antagonizes Tkv-mediated Dpp internalization by determining whether the defective extracellular Dpp distribution in Dally[KO] mutants could be rescued by depleting Tkv. Extracellular Dpp did increase in the D vs V compartment, potentially providing some support for their model. However, there are no statistics performed, which is needed for full confidence in the results. The lack of statistics is particularly problematic (1) when they state that extracellular Dpp does not rise in ap>tkv RNAi vs ap>tkv RNAi, dally[KO] wing discs (Fig. 6E) or (2) when they state that extracellular Dpp gradient expanded in the dorsal compartment when tkv was dorsally depleted in dally[deltaHS] mutants (Fig. 6I). These last two experiments are important for their model but the differences are assessed only visually. In fact, extracellular Dpp in ap>tkv RNAi, dally[KO] (Fig. 6B) appears to be lower than extracellular Dpp in ap>tkv RNAi (Fig. 6A) and the histogram of Dpp in ap>tkv RNAi, dally[KO] is actually a bit lower than Dpp in ap>tkv RNAi, But the author claim that there is no difference between the two. Their conclusion would be strengthened by statistical analyses of the two lines.

We provided statistics for all the quantifications for pMad and extracellular Dpp distribution as supplementary data. In the previous version, we argued that extracellular Dpp level in ap>tkvRNAi, dallyKO (Fig.6B) does not increase compared with that in ap>tkvRNAi (Fig.6A). Statistical analysis (t-test) showed that the extracellular Dpp level in Fig. 6B is similar to or lower than that in Fig. 6A (Fig. 6E), confirming our conclusion. Statistical analysis (t-test) also confirmed that extracellular Dpp distribution expanded when tkv was knocked down in dallyHS mutants (Fig. 6I).

Strengths:

1. New genomically-engineered alleles

A considerable strength of the study is the generation and characterization of new Dally, Dlp and Dpp alleles. These reagents will be of great use to the field.

Thanks. We hope that these resources are indeed useful to the field.

1. Surveying multiple phenotypes

The authors survey numerous parameters (Dpp distribution, Dpp signaling (pMad) and adult wing phenotypes) which provides many points of analysis.

Thanks!

Weaknesses:

1. Confusing discussion regarding the Dally core vs HS in Dpp stability. They don't provide any measurements or information on how they "normalize" for the level of Dally vs Dally[deltaHS]? This is important part of their model that currently is not supported by any measurements.

We explained how we normalized in the above section and updated the method section in the revised ms.

1. Lacking quantifications and statistical analyses:

a. Why are statistical significance for histograms (pMad and Dpp distribution) not supplied? These histograms provide the key results supporting the authors' conclusions but no statistical tests/results are presented. This is a pervasive shortcoming in the current study.

Thanks. We provided t-test analyses together with the raw data as supplementary data.

b. dpp[deltaN] with JAX transgene - it would strengthen the study to supply quantitative data on the percent survival/lethal stage of dpp[deltaN] mutants with or without the JAK transgene

In this study, we are interested in the role of dpp[deltaN] during the wing disc development. Therefore, we decided not to perform the detailed analysis on the percent survival/lethal stage of dpp[deltaN] mutants with or without the JAX transgene in the current study. Nevertheless, the fact that dpp[deltaN] allele is maintained with a balanced stock and JAX;dpp[deltaN] allele can be maintained as homozygous stock indicates that the lethality of dpp[deltaN] allele comes from the early stages. Indeed, our preliminary results showed that pMad signal is severely lost in the dpp[deltaN] embryo without JAX (data not shown), indicating that the allele is lethal at early embryonic stages.

c. The graphs on wing size etc should start at zero.

Thanks. We corrected this in the current ms.

d. The sizes of histograms and graphs in each figure should be increased so that the reader can properly assess them. Currently, they are very small.

Thanks. We changed the sizes in the current ms.

The authors' model is that Dally (not Dlp) is required for Dpp distribution and signaling but that this is not due to a direct interaction with Dpp. Rather, they posit that Dally-HS antagonize Tkv-mediated Dpp internalization. Currently the results of the experiments could be considered consistent with their model, but as noted above, the lack of statistical analyses of some parameters is a weakness.

Thanks. We now performed and provided the statistical analyses in the revised ms.

One problematic part of their result for me is the role of the Dally core protein (Fig. 7B). There is a mis-match between the over-expression results and Dally allele lacking HS (but containing the core). Finally, their results support the idea that one or more as-yet unidentified proteins interact with Dally-HS chains to control Dpp distribution and signaling in the wing disc.

Our results simply suggest that Dpp can interact with Dally mainly through core protein but this interaction is not sufficient to sustain extracellular Dpp gradient formation under physiological conditions (dallyDeltaHS) (Fig. 4Q). We find that the mis-match is not problematic if the role of Dally is not simply mediated through interaction with Dpp. We speculate that interaction of Dpp and core protein of Dally is transient and not sufficient to sustain the Dpp gradient without HS chains of Dally stabilizing extracellular Dpp distribution by blocking Tkv-mediated Dpp internalization.

There is much debate and controversy in the Dpp morphogen field. The generation of new, high quality alleles in this study will be useful to Drosophila community, and the results of this study support the concept that Tkv but not Dally regulate Dpp internalization. Thus the work could be impactful and fuel new debates among morphogen researchers.

Thanks.

The manuscript is currently written in a manner that really is only accessible to researchers who work on the Dpp gradient. It would be very helpful for the authors to re-write the manuscript and carefully explain in each section of the results (1) the exact question that will be asked, (2) the prior work on the topic, (3) the precise experiment that will be done, and (4) the predicted results. This would make the study more accessible to developmental biologists outside of the morphogen gradient and Drosophila communities.

Thanks. We modified texts and changed the order of Fig.5. We hope that the changes make this study more accessible to developmental biologists outside of the field.

Joint Public Review:

The authors are trying to distinguish between four models of the role of glypicans (HSPGs) on the Dpp/BMP gradient in the Drosophila wing, schematized in Fig. 1: (1) "Restricted diffusion" (HSPGs transport Dpp via repetitive interaction of HS chains with Dpp); (2) "Hindered diffusion" (HSPGs hinder Dpp spreading via reversible interaction of HS chains with Dpp); (3) "Stabilization" (HSPGs stabilize Dpp on the cell surface via reversible interaction of HS chains with Dpp that antagonizes Tkv-mediated Dpp internalization); and (4) "Recycling" (HSPGs internalize and recycle Dpp).

To distinguish between these models, the authors generate new alleles for the glypicans Dally and Dally-like protein (Dlp) and for Dpp: a Dally knock-out allele, a Dally YFP-tagged allele, a Dally knock-out allele with 3HA-Dlp, a Dlp knock-out allele, a Dlp allele containing 3-HA tags, and a Dpp lacking the HS-interacting domain. Additionally, they use an OLLAS-tag Dpp (OLLAS being an epitope tag against which extremely high affinity antibodies exist). They examine OLLAS-Dpp or HA-Dpp distribution, phospho-Mad staining, adult wing size.

They find that over-expressed Dally - but not Dlp - expands Dpp distribution in the larval wing disc. They find that the Dally[KO] allele behaves like a Dally strong hypomorph Dally[MH32]. The Dally[KO] - but not the Dlp[KO] - caused reduced pMad in both anterior and posterior domains and reduced adult wing size (particularly in the Anterior-Posterior axis). These defects can be substantially corrected by supplying an endogenously tagged YFP-tagged Dally. By contrast, they were not rescued when a 3xHA Dlp was inserted in the Dally locus. These results support their conclusion that Dpp interacts with Dally but not Dlp.

They next wanted to determine the relative contributions of the Dally core or the HS chains to the Dpp distribution. To test this, they over-expressed UAS-Dally or UAS-Dally[deltaHS] (lacking the HS chains) in the dorsal wing. Dally[deltaHS] over-expression increased the distribution of OLLAS-Dpp but caused a reduction in pMad. They do a critical experiment, making the Dally[deltaHS] allele, they find that loss of the HS chains is nearly as severe as total loss of Dally (i.e., Dally[KO]). These results indicate that the HS are critical for Dally's role in Dpp distribution and signaling.

Prior work has shown that a stretch of 7 amino acids in the Dpp N-terminal domain is required to interact with heparin but not with Dpp receptors (Akiyama, 2008). The authors generated an HA-tagged Dpp allele lacking these residues (HA-dpp[deltaN]). It is an embryonic lethal allele, but they can get some animals to survive to larval stages if they also supply a transgene called "JAK" containing dpp regulatory sequences. In the JAK; HA-dpp[deltaN] mutant background, they find that the distribution and signaling of this Dpp molecule is largely normal. While over-expressed Dally can increase the distribution of HA-dpp[deltaN], over-expression of Dally[deltaHS] cannot. These latter results support the model that the HS chains in Dally are required for Dpp function but not because of a direct interaction with Dpp.

In the last part of the results, they attempt to determine if the Dpp receptor Thickveins (Tkv) is required for Dally-HS chains interaction. The 2008 (Akiyama) model posits that Tkv activates pMad downstream of Dpp and also internalizes and degrades Dpp. A 2022 (Romanova-Michaelides) model proposes that Dally (not Tkv) internalizes Dpp. To distinguish between these models, the authors deplete Tkv from the dorsal compartment of the wing disc and found that extracellular Dpp increased and expanded in that domain. These results support the model that Tkv is required to internalize Dpp. They then tested the model that Dally antagonizes Tkv-mediated Dpp internalization by determining whether the defective extracellular Dpp distribution in Dally[KO] mutants could be rescued by depleting Tkv. Extracellular Dpp did increase in the D vs V compartment, potentially providing some support for their model. The results are statistically significant but the statistics are buried in an excel file without a read-me page. The code for the statistics is available from Github. These p values should be made more readily accessible and/or intelligible to the reader.

Strengthens:<br /> 1. New genomically-engineered alleles<br /> A considerable strength of the study is the generation and characterization of new Dally, Dlp and Dpp alleles. These reagents will be of great use to the field.

2. Surveying multiple phenotypes<br /> The authors survey numerous parameters (Dpp distribution, Dpp signaling (pMad) and adult wing phenotypes) which provides many points of analysis.

Weaknesses (minor):<br /> 1. The results are statistically significant but the statistics are buried in a dense excel file without a read-me page. The code for the statistics is available from Github. These p values should be made more readily accessible to the reader.

An appraisal of whether the authors achieved their aims, and whether the results support their conclusions.<br /> The authors' model is that Dally (not Dlp) is required for Dpp distribution and signaling but that this is not due to a direct interaction with Dpp. Rather, they posit that Dally-HS antagonize Tkv-mediated Dpp internalization. Currently the results of the experiments could be considered consistent with their model. Finally, their results support the idea that one or more as-yet unidentified proteins interact with Dally-HS chains to control Dpp distribution and signaling in the wing disc.

There is much debate and controversy in the Dpp morphogen field. The generation of new, high quality alleles in this study will be useful to Drosophila community, and the results of this study support the concept that Tkv but not Dally regulate Dpp internalization. Thus the work could be impactful and fuel new debates among the morphogen researchers.

eLife assessment

This valuable study considers empirical macroecological patterns in microbiome data across multiple taxonomic scales. The work convincingly shows that the Stochastic Logistic Growth model is a more appropriate choice of null model than the neutral theory of biodiversity. The work will be of particular interest to microbial ecologists.

Reviewer #1 (Public Review):

Shoemaker and Grilli analyze publicly available sequencing data to quantify how the microbial diversity of ecosystems changes with the taxonomic scale considered (e.g., diversity of genera vs diversity of families). This study builds directly on Grilli's 2020 paper which used this data to show that for many different microbial species, the distribution of abundances of the species across sampling sites belongs to a simple one-parameter family of gamma distributions. In this work, they show that the gamma distribution also describes the distribution of abundances of higher taxonomic levels. The distribution now requires two parameters, but the second parameter can be approximately derived by treating the distributions of lower-level taxonomic units as being independent. The difference between the species-level result and the result at higher taxonomic levels suggests that in some sense microbial species are ecologically meaningful units.

While the higher-level taxon abundance distributions can be well-approximated assuming independence of the constituent species, this approach substantially underestimates variation in community richness and diversity among sampling sites. Much of this extra variability appears to be driven by variability in sample size across sites. It is not clear to me how much this variation in sample size is itself due to variation in sampling effort versus variation in overall microbial densities. This variation in sample size also produces correlations between taxon richness at lower and higher taxonomic levels. For instance, sites with large samples are likely to have both many species within a genus and many genera. The authors also consider taxon diversity (Shannon index, i.e. entropy), which is constructed from frequencies and is therefore less sensitive to sample size. In this case, correlations between diversity across taxonomic scales instead appear to depend on the idiosyncratic correlations among species abundances.

This paper's results are presented in a fairly terse manner, even when they are describing summary statistics that require a lot of thought to interpret. I don't think it would make sense to try to understand it without having first worked through the 2020 paper. But everyone interested in a general understanding of microbial ecology should read the 2020 paper, and once one has done that, this paper is worth reading as well simply for seeing how the major pattern in that paper shifts as one moves up in taxonomic scale.

Reviewer #3 (Public Review):

Summary<br /> In this research advance, the authors purport to show that the unified neutral theory of biodiversity (UNTB) is not a suitable null model for exploring the relationship between macroecological quantities, and additionally that the stochastic logistic growth model (SLM) is a viable replacement. They do this by citing other studies where UNTB was unable to capture individual macroecological quantities, and then demonstrating SLM's strength at predicting the same diversity metrics. They extend this analysis to show SLM's modeling capability at multiple scales of coarse graining, in addition to its failures at predicting these metrics' variances. Finally, authors conduct a similar analysis to Madi et al. (2020) by investigating the relationship between diversity measures within a group and across coarse-grained groups (e.g. genera diversity in one family compared to diversity of families). The authors show that choosing SLM as a null model reveals some previously reported relationships to be no longer "novel", in the sense that the patterns can be adequately captured by the null model. Authors also show that relationships not captured by the null model can be recovered by adding correlations, suggesting interactions are the driving force behind them.

Strengths<br /> 1. Authors make a strong argument that UNTB is not a good null model of macroecological observables and especially relationships between them. Authors convincingly argue that a SLM is a better null since the gamma distribution it predicts is a better description of the empirical Abundance Fluctuation Distributions (AFD).<br /> 2. Authors show that the gamma distribution predicted by SLM is a good fit for the AFD's at many different scales of coarse graining, not just the OTU level as was previously demonstrated. Authors show the same distribution predicted the mean diversity and richness at all scales of coarse graining.<br /> 3. Authors convincingly demonstrate how SLM can be used to test the relevance of interactions to macroecological relationships.

Weaknesses<br /> This reviewer's concerns were convincingly addressed by the revisions.

Overall Impact<br /> The authors present a convincing argument for the use of SLM as a better non-interacting null model for macroecological quantities and relationships.

eLife assessment

This study presents useful observations about how the human brain uses long-term priors (acquired during our lifetime of listening) to make predictions about expected sounds - an open question in the field of predictive processing. However, the evidence as currently presented is incomplete. Both the theoretical background and analysis approach should be strengthened.

Reviewer #1 (Public Review):

Summary:<br /> In this work, the authors study whether the human brain uses long-term priors (acquired during our lifetime) regarding the statistics of auditory stimuli to make predictions respecting auditory stimuli. This is an important open question in the field of predictive processing.

To address this question, the authors cleverly profit from the naturally existing differences between two linguistic groups. While speakers of Spanish use phrases in which function words (short words like articles and prepositions) are followed by content words (longer words like nouns, adjectives, and verbs), speakers of Basque use phrases in the opposite order. Because of this, speakers of Spanish usually hear phrases in which short words are followed by longer words, and speakers of Basque experience the opposite. This difference in the order of short and longer words is hypothesized to result in a long-term duration prior that is used to make predictions regarding the likely durations of incoming sounds, even if they are not linguistic in nature.

To test this, the authors used MEG to measure the mismatch responses (MMN) elicited by the omission of short and long tones that were presented in alternation. The authors report an interaction between the language background of the participants (Spanish, Basque) and the type of omission MMN (short, long), which goes in line with their predictions. They supplement these results with a source-level analysis.

Unfortunately, serious concerns regarding the predictions put forward by the authors, and the interaction effect found, make the interpretation of these results difficult.

Strengths:<br /> This work has many strengths. To test the main question, the authors profit from naturally occurring differences in the everyday auditory experiences of two linguistic groups, which allows them to test the effect of putative auditory priors consolidated over years. This is a direct way of testing the effect of long-term priors.

The fact that the priors in question are linguistic, and that the experiment was conducted using non-linguistic stimuli (i.e. simple tones), allows for testing of whether these long-term priors generalize across auditory domains.

The experimental design is elegant and the analysis pipeline is appropriate. This work is very well written. In particular, the introduction and discussion sections are clear and engaging. The literature review is complete.

Weaknesses:<br /> There are two main issues in this work. The first one pertains to the predictions put forward by the researchers, and the second with the interaction effect reported.

1) With respect to the predictions, the authors propose that the subjects, depending on their linguistic background and the length of the tone in a trial, can put forward one or two predictions. The first is a short-term prediction based on the statistics of the previous stimuli and identical for both groups (i.e. short tones are expected after long tones and vice versa). The second is a long-term prediction based on their linguistic background. According to the authors, after a short tone, Basque speakers will predict the beginning of a new phrasal chunk, and Spanish speakers will predict it after a long tone.

In this way, when a short tone is omitted, Basque speakers would experience the violation of only one prediction (i.e. the short-term prediction), but Spanish speakers will experience the violation of two predictions (i.e. the short-term and long-term predictions), resulting in a higher amplitude MMN. The opposite would occur when a long tone is omitted. So, to recap, the authors propose that subjects will predict the alternation of tone durations (short-term predictions) and the beginning of new phrasal chunks (long-term predictions).

The problem with this is that subjects are also likely to predict the completion of the current phrasal chunk. In speech, phrases are seldom left incomplete. In Spanish is very unlikely to hear a function-word that is not followed by a content-word (and the opposite happens in Basque). On the contrary, after the completion of a phrasal chunk, a speaker might stop talking and a silence might follow, instead of the beginning of a new phrasal chunk.

Considering that the completion of a phrasal chunk is more likely than the beginning of a new one, the prior endowed to the participants by their linguistic background should make us expect a pattern of results actually opposite to the one reported here.

2) The authors report an interaction effect that modulates the amplitude of the omission response, but caveats make the interpretation of this effect somewhat uncertain. The authors report a widespread omission response, which resembles the classical mismatch response (in MEG) with strong activations in sensors over temporal regions. Instead, the interaction found is circumscribed to four sensors that do not overlap with the peaks of activation of the omission response. Furthermore, the boxplot in Figure 2E suggests that part of the interaction effect might be due to the presence of two outliers (if removed, the effect is no longer significant). Overall, it is possible that the reported interaction is driven by a main effect of omission type which the authors report, and find consistently only in the Basque group (showing a higher amplitude omission response for long tones than for short tones).

Because of these points, it is difficult to interpret this interaction as a modulation of the omission response. It should also be noted that in the source analysis, the interaction only showed a trend in the left auditory cortex, but in its current version the manuscript does not report the statistics of such a trend.

Reviewer #2 (Public Review):

Summary:<br /> Morucci et al. tested the influence of linguistic prosody long-term priors in forming predictions about simple acoustic rhythmic tone sequences composed of alternating tone duration, by violating context-dependent short-term priors formed during sequence listening. Spanish and Basque participants were selected due to the different rhythmic prosody of the two languages (functor-initial vs. Functor final, respectively), despite a common cultural background. The authors found that neuromagnetic responses to casual tone omissions reflected the linguistic prosody pattern of the participant's dominant language: in Spanish speakers, omission responses were larger to short tones, whereas in Basque speakers, omission responses were larger to long tones. Source localization of these responses revealed this interaction pattern in the left auditory cortex, which the authors interpret as reflecting a perceptual bias due to acoustic cues (inherent linguistic rhythms, rather than linguistic content). Importantly, this pattern was not found when the rhythmic sequence entailed pitch, rather than duration, cues. To my knowledge, this is the first study providing neural signatures of a known behavioral effect linking ambiguous rhythmic tone sequence perceptual organization to linguistic experience.

The conclusions of the study are well supported by the data, albeit weakly by the source analysis, but I have the impression that the rationale of the study and the analyses performed may be missing an important aspect of rhythmic sequence perception, namely the involvement of entrained oscillatory activity to the perceived rhythm, particularly phase alignment to pattern onsets. This view would not change the impact of the results but add depth to their interpretation.

Strengths:<br /> 1) The choice of participants. The bilingual population of the Basque country is perfect for performing studies that need to control for cultural and socio-economic background while having profound linguistic differences. In this sense, having dominant Basque speakers as a sample equates that in Molnar et al. (2016), and thus overcomes the lack of direct behavioral evidence for a difference in rhythmic grouping across linguistic groups. Molnar et al. (2016)'s evidence on the behavioral effect is compelling, and the evidence on neural signatures provided by the present study aligns with it.

2) The experimental paradigm. It is a well-designed acoustic sequence, that considers aspects such as gap length insertion, to be able to analyze omission responses free from subsequent stimulus-driven responses, and which includes a control sequence that uses pitch instead of duration as a cue to rhythmic grouping, which provides a stronger case for the differences found between groups to be due to prosodic duration cues.

3) Data analyses. Sound, state-of-the-art methodology in the event-related field analyses at the sensor level.

Weaknesses:<br /> 1) Despite the evidence provided on neural responses, the main conclusion of the study reflects a known behavioral effect on rhythmic sequence perceptual organization driven by linguistic background (Molnar et al. 2016, particularly). Also, the authors themselves provide a good review of the literature that evidences the influence of long-term priors in neural responses related to predictive activity. Thus, in my opinion, the strength of the statements the authors make on the novelty of the findings may be a bit far-fetched in some instances.

2) Albeit the paradigm is well designed, I fail to see the grounding of the hypotheses laid by the authors as framed under the predictive coding perspective. The study assumes that responses to an omission at the beginning of a perceptual rhythmic pattern will be stronger than at the end. I feel this is unjustified. If anything, omission responses should be larger when the gap occurs at the end of the pattern, as that would be where stronger expectations are placed: if in my language a short sound occurs after a long one, and I perceptually group tone sequences of alternating tone duration accordingly, when I hear a short sound I will expect a long one following; but after a long one, I don't necessarily need to expect a short one, as something else might occur.

3) In this regard, it is my opinion that what is reflected in the data may be better accounted for (or at least, additionally) by a different neural response to an omission depending on the phase of an underlying attentional rhythm (in terms of Large and Jones rhythmic attention theory, for instance) and putative underlying entrained oscillatory neural activity (in terms of Lakatos' studies, for instance). Certainly, the fact that the aligned phase may differ depending on linguistic background is very interesting and would reflect the known behavioral effect.

4) Source localization is performed on sensor-level significant data. The lack of source-level statistics weakens the conclusions that can be extracted. Furthermore, only the source reflecting the interaction pattern is taken into account in detail as supporting their hypotheses, overlooking other sources. Also, the right IFG source activity is not depicted, but looking at whole brain maps seems even stronger than the left. To sum up, source localization data, as informative as it could be, does not strongly support the author's claims in its current state.

Reviewer #3 (Public Review):

Summary:<br /> The paper investigates the effects of long-term linguistic experience on early auditory processing, a subject that has been relatively less studied compared to short-term influences. Using MEG, the study examines brain responses to auditory stimuli in speakers of Spanish and Basque, whose syntactic rules provide different degrees of exposure to durational patterns (long-short vs short-long). The findings suggest that both long-term language experience, as well as short-term transitional probabilities, can shape auditory predictive coding for non-linguistic sound sequences, evidenced by differences in mismatch negativity amplitudes localised to the left auditory cortex.

Strengths:<br /> The study integrates linguistics and auditory neuroscience in an interesting interdisciplinary way that may interest linguists as well as neuroscientists. The fact that long-term language experience affects early auditory predictive coding is important for understanding group and individual differences in domain-general auditory perception. It has importance for neurocognitive models of auditory perception (e.g. inclusion of long-term priors), and will be of interest to researchers in linguistics, auditory neuroscience, and the relationship between language and perception. The inclusion of a control condition based on pitch is also a strength.

Weaknesses:<br /> The main weaknesses are the strength of the effects and generalisability. The sample size is also relatively small by today's standards, with N=20 in each group. Furthermore, the crucial effects are all mostly in the .01>P<.05 range, such as the crucial interaction P=.03. It would be nice to see it replicated in the future, with more participants and other languages. It would also have been nice to see behavioural data that could be correlated with neural data to better understand the real-world consequences of the effect.

Author Response

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

We thank the reviewers for their feedback. Our response and a summary of the changes made to the manuscript are shown below. In addition to the changes made in response to the reviewer’s comments, we made the following changes to improve the manuscript:

• We updated figures 8 and 9 using data with improved preprocessing and source reconstruction. We now also include graphical network plots. This helps in the cross method (figure 8 vs 9) and cross dataset (figure 9 vs 10) comparison.

• We added funding acknowledgments and a credit author statement.

Reviewer #1 (Public Review):

Summary:

These types of analyses use many underlying assumptions about the data, which are not easy to verify. Hence, one way to test how the algorithm is performing in a task is to study its performance on synthetic data in which the properties of the variable of interest can be apriori fixed. For example, for burst detection, synthetic data can be generated by injected bursts of known durations, and checking if the algorithm is able to pick it up. Burst detection is difficult in the spectral domain since direct spectral estimators have high variance (see Subhash Chandran et al., 2018, J Neurophysiol). Therefore, detected burst lengths are typically much lower than injected burst lengths (see Figure 3). This problem can be solved by doing burst estimation in the time domain itself, for example, using Matching Pursuit (MP). I think the approach presented in this paper would also work since this model is also trained on data in the time domain. Indeed, the synthetic data can be made more "challenging" by injecting multiple oscillatory bursts that are overlapping in time, for which a greedy approach like MP may fail. It would be very interesting to test whether this method can "keep up" as the data is made more challenging. While showing results from brain signals directly (e.g., Figure 7) is nice, it will be even more impactful if it is backed up with results obtained from synthetic data with known properties.

We completely agree with the reviewer that testing the methods using synthetic data is an important part of validating such an approach. Each of the original papers that apply these methods to a particular application do this. The focus of this manuscript is to present a toolbox for applying these methods rather than to introduce/validate the methods themselves. For a detailed validation of the methods, the reader should see the citations. For example, the following paper introduces the HMM as a method for oscillatory burst detection:

• A.J. Quinn, et al. “Unpacking transient event dynamics in electrophysiological power spectra”. Brain topography 32.6 (2019): 1020-1034. See figures 2 and 3 for an evaluation of the HMM’s performance in detecting single-channel bursts using synthetic data.

We have added text to paragraph 2 in section 2.5 to clarify this burst detection method has been validated using simulated data and added references.

I was wondering about what kind of "synthetic data" could be used for the results shown in Figure 8-12 but could not come up with a good answer. Perhaps data in which different sensory systems are activated (visual versus auditory) or sensory versus movement epochs are compared to see if the activation maps change as expected. We see similarities between states across multiple runs (reproducibility analysis) and across tasks (e.g. Figure 8 vs 9) and even methods (Figure 8 vs 10), which is great. However, we should also expect the emergence of new modes specific to sensory activation (say auditory cortex for an auditory task). This will allow us to independently check the performance of this method.

The following papers study the performance of the HMM and DyNeMo in detecting networks using synthetic data:

• D. Vidaurre, et al. “Spectrally resolved fast transient brain states in electrophysiological data”. Neuroimage 126 (2016): 81-95. See figure 3 in this paper for an evaluation of the HMM’s performance in detecting oscillatory networks using simulation data.

• C. Gohil, et al. “Mixtures of large-scale dynamic functional brain network modes”. Neuroimage 263 (2022): 119595. See figures 4 and 5 for an evaluation of DyNeMo performance in detecting overlapping networks and long-range temporal structure in the data.

We have added text to paragraph 2 in section 2.5 to clarify these methods have been well tested on simulated data and added references.

The authors should explain the reproducibility results (variational free energy and best run analysis) in the Results section itself, to better orient the reader on what to look for.

Considering the second reviewer’s comments, we moved the reproducibility results to the supplementary information (SI). This means the reproducibility results are no longer part of the main figures/text. However, we have added some text to help the reader understand what aspects indicate the results are reproducible in section 2 of the SI.

Page 15: the comparison across subjects is interesting, but it is not clear why sensory-motor areas show a difference and the mean lifetime of the visual network decreases. Can you please explain this better? The promised discussion in section 3.5 can be expanded as well.

It is well known that the frequency and amplitude of neuronal oscillations changes with age. E.g. see the following review: Ishii, Ryouhei, et al. "Healthy and pathological brain aging: from the perspective of oscillations, functional connectivity, and signal complexity." Neuropsychobiology 75.4 (2018): 151-161. We observe older people have more beta activity and less alpha activity. These changes are seen in time-averaged calculations, i.e. the amplitude of oscillations are calculated using the entire time series for each subject.

The dynamic analysis presented in the paper provides further insight into how changes in the time-averaged quantities can occur through changes in the dynamics of frequency-specific networks. The sensorimotor network, which is a network with high beta activity, has a higher fractional occupancy. This indicates the change we observe in time-average beta power may be due to a longer amount of time spent in the sensorimotor network. The visual network, which is a network with high alpha activity, shows reduced lifetimes, which can explain the reduced time-averaged alpha activity seen with ageing.

We hope the improved text in the last paragraph of section 3.5 clarifies this. It should also be taken into account that the focus of this manuscript is the tools rather than an in-depth analysis of ageing. We use the age effect as an example of the potential analysis this toolbox enables.

Reviewer #2 (Public Review):

Summary:

The authors have developed a comprehensive set of tools to describe dynamics within a single time-series or across multiple time-series. The motivation is to better understand interacting networks within the human brain. The time-series used here are from direct estimates of the brain's electrical activity; however, the tools have been used with other metrics of brain function and would be applicable to many other fields.

Strengths:

The methods described are principled, and based on generative probabilistic models.

This makes them compact descriptors of the complex time-frequency data.

Few initial assumptions are necessary in order to reveal this compact description.

The methods are well described and demonstrated within multiple peer-reviewed articles.

This toolbox will be a great asset to the brain imaging community.

Weaknesses:

The only question I had was how to objectively/quantitatively compare different network models. This is possibly easily addressed by the authors.

We thank the reviewer for his/her comments. We address the weaknesses in our response in the “Recommendations For The Authors” section.

Reviewer #1 (Recommendations For The Authors):

Figure 2 legend: Please add the acronym for LCMV also.

We have now done this.

Section 2.5.1 page 8: the pipeline is shown in Figure 4, not 3.

This has been fixed.

Reviewer #2 (Recommendations For The Authors):

This is a great paper outlining a resource that can be applied to many different fields. I have relatively minor comments apart from one.

How does one quantitatively compare network descriptors (from DyNeMo and TDE-HMM for example)? At the moment the word 'cleaner' (P17) is used, but is there any non-subjective way? (eg Free energy/ cross validation etc). At the moment it is useful that one method gives a larger effect size (in a comparison between groups).. but could the authors say something about the use of these methods as more/less faithful descriptors of the data? Or in other words, do all methods generate datasets (from the latent space) that can be quantitatively compared with the original data?

In principle, the variational free energy could be used to compare models. However, because we use an approximate variational free energy (an exact measure is not attainable) for DyNeMo and an exact free energy for the HMM, it is possible that any differences we see in the variational free energy between the HMM and Dynemo are caused by the errors in its approximation. This makes it unreliable for comparing across models. That said, we can still use the variational free energy to compare within models. Indeed, we use the variational free energy for quantitative model comparisons when we select the best run to analyse from a set of 10.

One viable approach for comparing models is to assess their performance on downstream tasks. In this manuscript, examples of downstream tasks are the evoked network response and the young vs old group difference. We argue a better performance in the downstream task indicates a more useful model within that context. This performance is a quantitative measure. Note, there is no straightforward answer to which is the best model. It is likely different models will be useful for different downstream tasks.

In terms of which model provides a more faithful description of the data. The more flexible generative model for DyNeMo means it will generate more realistic data. However, this doesn’t necessarily mean it’s the best model (for a particular downstream task). Both the HMM and DyNeMo provide complementary descriptions that can be useful.

We have clarified the above in paragraph 5 of section 4.

Other comments:

• Footnote 6 - training on concatenated group data seems to be important. It could be more useful in the main manuscript where the limitations of this could be discussed.

By concatenating the data across subjects, we learn a group-level model. By doing this, we pool information across all subjects to estimate the networks. This can lead to more robust estimates. We have moved this footnote to the main text in paragraph 1 of section 2.5 and added further information.

• In the TDE burst detection section- please expand on why/how a specific number of states was chosen.

As with the HMM dynamic network analysis, the number of states must be pre-specified. For burst detection, we are often interested in an on/off type segmentation, which can be achieved with a 2 state HMM. However, if there are multiple burst types, these will all be combined into a single ‘on’ state. Therefore, we might want to increase the number of states to model multiple burst types. 3 was chosen as a trade-off to stay close to the on/off description but allow the model to learn more than 1 burst type. We have added text discussing this in paragraph 4 of section 4.

• Normally the value of free energy is just a function of the data - and only relative magnitude is important. I think figures (eg 7c) would be clearer if the offset could be removed.

We agree only the relative magnitude is important. We added text clarifying this in section 2 of the SI. We think it would still be worthwhile to include the offset so that future users can be sure they have correctly trained a model and calculated the free energy.

• Related to the above- there are large differences in model evidence shown between sets. Yet all sets are the same data, and all parameter estimates are more or less the same. Could the authors account for this please (i.e. is there some other parameter that differentiates the best model in one set from the other sets, or is the free energy estimate a bit variable).

We would like to clarify only the model parameters for the best run are shown in the group-level analysis. This is the run with the lowest variational free energy, which is highlighted in red. We have now clarified this in the caption of each figure. The difference in free energy for the best runs (across sets) is relatively small compared to the variation across runs within a set. If we were to plot the model parameters for each of the 10 runs in a set, we would see more variability. We have now clarified this in section 2 of the SI.

Also note, the group analysis usually involves taking an average. Small differences in the variational free energy could reflect small differences in subject-specific model parameters, which are then averaged out, giving virtually identical group effects.

• And related once again, if the data are always the same, I wonder if the free-energy plots and identical parameter estimates could be removed to free up space in figures?

The reproducibility results have now been moved to the supplementary information (SI).

• When citing p-values please specify how they are corrected (and over what please eg over states, nodes, etc?). This would be useful didactically as I imagine most users will follow the format of the presentation in this paper.

We now include in the caption further details of how the permutation significance testing was done.

• Not sure of the value of tiny power maps in 9C. Would consider making it larger or removing it?

The scale of these power maps is identical to part (A.I). We have moved the reproducibility analysis to the SI, enlarged the figure and added colour bars. We hope the values are now legible.

• Figure 3. I think the embedding in the caption doesn't match the figure (+-5 vs +-7 lags). Would be useful to add in the units of covariance (cii).

The number of embeddings in the caption has been fixed. Regarding the units for the covariances, as this is simulated data there aren’t really any units. Note, there is already a colour bar to indicate the values of each element.

• Minimize variational free energy - it may be confusing for some readers that other groups maximize the negative free energy. Maybe a footnote?

We thank the reviewer for their suggestion. We have added a footnote (1).

• Final question- and related to the Magnetoencephalography (MEG) data presented. These data are projected into source space using a beamformer algorithm (with its own implicit assumptions and vulnerabilities). Would be interested in the authors' opinion on what is standing between this work and a complete generative model of the MEG data - i.e. starting with cortical electrical current sources with interactions modeled and a dynamic environmental noise model (i.e. packing all assumptions into one model)?

In principle, there is nothing preventing us from including the forward model in the generative model and training on sensor level MEG data. This would be a generative model starting from the dipoles inside the brain to the MEG sensors. This is under active research. If the reviewer is referring to a biophysical model for brain activity, the main barrier for this is the inference of model parameters. However, note that the new inference framework presented in the DyNeMo paper (Gohil, et al. 2022) actually makes this more feasible. Given the scope of this manuscript is to present a toolbox for studying dynamics with existing methods, we leave this topic as future work.

eLife assessment

The authors present a comprehensive set of tools to compactly characterize the time-frequency interactions across a network. The utility of the toolbox is compelling and demonstrated through a series of exemplar brain imaging datasets. This fundamental work adds to the repertoire of techniques that can be used to study high-dimensional data.

Reviewer #1 (Public Review):

In their revised manuscript, the authors have addressed all the concerns raised earlier (written below for completeness).

Summary:

These types of analyses use many underlying assumptions about the data, which are not easy to verify. Hence, one way to test how the algorithm is performing in a task is to study its performance on synthetic data in which the properties of the variable of interest can be apriori fixed. For example, for burst detection, synthetic data can be generated by injected bursts of known durations, and checking if the algorithm can pick it up. Burst detection is difficult in the spectral domain since direct spectral estimators have high variance (see Subhash Chandran et al., 2018, J Neurophysiol). Therefore, detected burst lengths are typically much lower than injected burst lengths (see their Figure 3). This problem can be solved by doing burst estimation in the time domain itself, for example, using Matching Pursuit (MP). I think the approach presented in this paper would also work since this model is also trained on data in the time domain. Indeed, the synthetic data can be made more "challenging" by injecting multiple oscillatory bursts that are overlapping in time, for which a greedy approach like MP may fail. It would be very interesting to test whether this method can "keep up" as the data is made more challenging. While showing results from brain signals directly (e.g., Figure 7) is nice, it will be even more impactful if it is backed up with results obtained from synthetic data with known properties.

I was wondering about what kind of "synthetic data" could be used for the results shown in Figure 8-12 but could not come up with a good answer. Perhaps data in which different sensory systems are activated (visual versus auditory) or sensory versus movement epochs are compared to see if the activation maps change as expected? We see similarities between states across multiple runs (reproducibility analysis) and across tasks (e.g. Figure 8 vs 9) and even methods (Figure 8 vs 10), which is great. However, we should also expect emergence of new modes specific to sensory activation (say auditory cortex for an auditory task). This will allow us to independently check the performance of this method.

The authors should explain the reproducibility results (variational free energy and best run analysis) in the Results section itself, to better orient the reader on what to look for.

Page 15: the comparison across subjects is interesting, but it is not clear why sensory-motor areas show a difference and the mean lifetime of the visual network decreases. Can you please explain this better? The promised discussion in section 3.5 can be expanded as well.

Reviewer #2 (Public Review):

Summary:<br /> The authors have developed a comprehensive set of tools to describe dynamics within a single time-series or across multiple time-series. The motivation is to better understand interacting networks within the human brain. The time-series used here are from direct estimates of the brain's electrical activity; however the tools have been used with other metrics of brain function and would be applicable to many other fields.

Strengths:<br /> The methods described are principled, based on generative probabilistic models.<br /> This makes them compact descriptors of the complex time-frequency data.<br /> Few initial assumptions are necessary in order to reveal this compact description.<br /> The methods are well described and demonstrated within multiple peer reviewed articles.<br /> This toolbox will be a great asset to the brain imaging community.

Weaknesses:<br /> The only question I had (originally) was how to objectively/quantitatively compare different network models. This has now been addressed by the authors in the latest revision.

Author Response

We are delighted that the reviewers found our work to have merit and we are thankful for their careful reviews and suggestions for experiments and changes to the text to further improve this study.

eLife assessment

This work uses an interdisciplinary approach combining microfluidics, structural biology, and genetic analyses to provide valuable findings that show that pathogenic enteric bacteria exhibit taxis toward human serum. The data are solid and show that the behavior utilizes the bacterial chemotaxis system and the chemoreceptor Tsr, which senses the amino acid L-serine. The work provides an ecological context for the role of serine as a bacterial chemoattractant and could have clinical implications for bacterial bloodstream invasion during episodes of gastrointestinal bleeding.

Reviewer #1 (Public Review):

Summary:

Glenn et al. present solid evidence that both lab and clinical Enterobacteriaceae strains rapidly migrate towards human serum using an exciting approach that combines microfluidics, structural biology, and genotypic analysis. The authors succeed in bringing to light a novel context for the role of serine as a bacterial chemoattractant as well as documenting what is likely to be a key step in bloodstream entry for some of the main sepsis-associated pathogens during gastrointestinal bleeding. They aim to expand their conclusions from a single lab serovar of Salmonella enterica to a range of clinical serovars and other species within the Enterobacteriaceae family. This is a powerful approach that greatly increases the scope of their findings but I find that some of their conclusions here are not always supported by strong evidence.

I would also like to note that, while I enjoyed the interdisciplinary scope of this study, I am personally not well positioned to review the protein structural aspects of this work.

Strengths:<br /> - The authors first characterise migration towards serum in detail using a well-characterised lab strain but one of the main strengths of this study is that they then expand their scope to observe equivalent behaviours in several clinical serovars. This strongly supports the clinical relevance of the key behaviours that they document.

- The interdisciplinary nature of this study is a real strength and greatly increases the scope of the conclusions presented. Working from a structural understanding of chemoreceptor-ligand binding through to a larger-scale genetic analysis of chemoreceptor phylogeny allows the authors to draw important (although I find not always definitive) conclusions about bacterial migration towards human serum across a wide range of bacterial species (including many important pathogens). This is a very exciting approach and I was particularly interested to see the authors follow this up with observations of migration in C. koseri (a clinical isolate with little known about its chemotactic capabilities).

- The authors use experiments that compete migrating strains against each other and these offer an exciting glimpse into how bacterial movement and navigation could play out in multi-species environments like the human gut.

- The authors successfully identify a single component of human serum (i.e. serine) as one specific attractant driving the bacterial migration response seen here. Teasing apart the response of bacteria to complex stimuli like human serum is an important step here.

Weaknesses:<br /> There are several issues that I would personally like to see addressed in this study:

1) The authors refer to human serum as a chemoattractant numerous times throughout the study (including in the title). As the authors acknowledge, human serum is a complex mixture and different components of it may act as chemoattractants, chemo-repellents (particularly those with bactericidal activities), or may elicit other changes in motility (e.g. chemokinesis). The authors present convincing evidence that cells are attracted to serine within human serum - which is already a well-known bacterial chemoattractant. Indeed, their ability to elucidate specific elements of serum that influence bacterial motility is a real strength of the study. However, human serum itself is not a chemoattractant and this claim should be re-phrased - bacteria migrate towards human serum, driven at least in part by chemotaxis towards serine.

2) Linked to the previous point, several bacterial species (including E. coli - one of the bacterial species investigated here) are capable of osmotaxis (moving up or down gradients in osmolality). Whilst chemotaxis to serine is important here, could movement up the osmotic gradient generated by serum injection play a more general role? It could be interesting to measure the osmolality of the injected serum and test whether other solutions with similar osmolality elicit a similar migratory response. Another important control here would be to treat human serum with serine racemase and observe how this impacts bacterial migration.

3) The inference of the authors' genetic analysis combined with the migratory response of E. coli and C. koseri to human serum shown in Fig. 6 is that Tsr drives movement towards human serum across a range of Enterobacteriaceae species. The evidence for the importance of Tsr here is currently correlative - more causal evidence could be presented by either studying the response of tsr mutants in these two species (certainly these should be readily available for E. coli) or by studying the response of these two species to serine gradients.

4) The migratory response of E. coli looks striking when quantified (Fig. 6C), but is really unclear from looking at Panel B - it would be more convincing if an explanation was offered for why these images look so much less striking than analogous images for other species (E.g. Fig. 6A).

5) It is unclear why the fold-change in bacterial distribution shows an approximately Gaussian shape with a peak at a radial distance of between 50 -100 um from the source (see for example Fig. 2H). Initially, I thought that maybe this was due to the presence of the microcapillary needle at the source, but the CheY distribution looks completely flat (Fig. 3I). Is this an artifact of how the fold-change is being calculated? Certainly, it doesn't seem to support the authors' claim that cells increase in density to a point of saturation at the source. Furthermore, it also seems inappropriate to apply a linear fit to these non-linear distributions (as is done in Fig. 2H and in the many analogous figures throughout the manuscript).

6) The authors present several experiments where strains/ serovars competed against each other in these chemotaxis assays. As mentioned, these are a real strength of the study - however, their utility is not always clear. These experiments are useful for studying the effects of competition between bacteria with different abilities to climb gradients. However, to meaningfully interpret these effects, it is first necessary to understand how the different bacteria climb gradients in monoculture. As such, it would be instructive to provide monoculture data alongside these co-culture competition experiments.

7) Linked to the above point, it would be especially instructive to test a tsr mutant's response in monoculture. Comparing the bottom row of Fig. 3G to Fig. 3I suggests that when in co-culture with a cheY mutant, the tsr mutant shows a higher fold-change in radial distribution than the WT strain. Fig. 4G shows that a tsr mutant can chemotax towards aspartate at a similar, but reduced rate to WT. This could imply that (like the trg mutant), a tsr mutant has a more general motility defect (e.g. a speed defect), which could explain why it loses out when in competition with the WT in gradients of human serum, but actually seems to migrate strongly to human serum when in co-culture with a cheY mutant. This should be resolved by studying the response of a tsr mutant in monoculture.

8) In Fig. 4, the response of the three clinical serovars to serine gradients appears stronger than the lab serovar, whilst in Fig. 1, the response to human serum gradients shows the opposite trend with the lab serovar apparently showing the strongest response. Can the authors offer a possible explanation for these slightly confusing trends?

9) In Fig. S2, it seems important to present quantification of the effect of serine racemase and the reported lack of response to NE and DHMA - the single time-point images shown here are not easy to interpret.

10) Importantly, the authors detail how they controlled for the effects of pH and fluid flow (Line 133-136). Did the authors carry out similar controls for the dual-species experiments where fluorescent imaging could have significantly heated the fluid droplet driving stronger flow forces?

Reviewer #2 (Public Review):

Summary:<br /> This manuscript characterizes a chemoattractant response to human serum by pathogenic bacteria, focusing on pathogenic strains of Salmonella enterica (Se). The researchers conducted the chemotaxis assays using a micropipette injection method that allows real-time tracking of bacterial population densities. They found that clinical isolates of several Se strains present a chemoattractant response to human serum. The specific chemoattractant within the serum is identified as L-serine, a highly characterized and ubiquitous chemoattractant, that is sensed by the Tsr receptor. They further show that chemoattraction to serum is impaired with a mutant strain devoid of Tsr. X-ray crystallography is then used to determine the structure of L-serine in the Se Tsr ligand binding domain, which differs slightly from a previously determined structure of a homologous domain. They went on to identify other pathogens that have a Tsr domain through a bioinformatics approach and show that these identified species also present a chemoattractant response to serum.

Strengths and Weaknesses:<br /> This study is well executed and the experiments are clearly presented. These novel chemotaxis assays provide advantages in terms of temporal resolution and the ability to detect responses from small concentrations. That said, it is perhaps not surprising these bacteria respond to serum as it is known to contain high levels of known chemoattractants, serine certainly, but also aspartate. In fact, the bacteria are shown to respond to aspartate and the tsr mutant is still chemotactic. The authors do not adequately support their decision to focus exclusively on the Tsr receptor. Tsr is one of the chemoreceptors responsible for observed attraction to serum, but perhaps, not the receptor. Furthermore, the verification of chemotaxis to serum is a useful finding, but the work does not establish the physiological relevance of the behavior or associate it with any type of disease progression. I would expect that a majority of chemotactic bacteria would be attracted to it under some conditions. Hence the impact of this finding on the chemotaxis or medical fields is uncertain.

The authors also state that "Our inability to substantiate a structure-function relationship for NE/DHMA signaling indicates these neurotransmitters are not ligands of Tsr." Both norepinephrine (NE) and DHMA have been shown previously by other groups to be strong chemoattractants for E. coli (Ec), and this behavior was mediated by Tsr (e.g. single residue changes in the Tsr binding pocket block the response). Given the 82% sequence identity between the Se and Ec Tsr, this finding is unexpected (and potentially quite interesting). To validate this contradictory result the authors should test E. coli chemotaxis to DHMA in their assay. It may be possible that Ec responds to NE and DHMA and Se doesn't. However, currently, the data is not strong enough to rule out Tsr as a receptor to these ligands in all cases. At the very least the supporting data for Tsr being a receptor for NE/DHMA needs to be discussed.

The authors also determine a crystal structure of the Se Tsr periplasmic ligand binding domain bound to L-Ser and note that the orientation of the ligand is different than that modeled in a previously determined structure of lower resolution. I agree that the SeTsr ligand binding mode in the new structure is well-defined and unambiguous, but I think it is too strong to imply that the pose of the ligand in the previous structure is wrong. The two conformations are in fact quite similar to one another and the resolution of the older structure, is, in my view, insufficient to distinguish them. It is possible that there are real differences between the two structures. The domains do have different sequences and, moreover, the crystal forms and cryo-cooling conditions are different in each case. It's become increasingly apparent that temperature, as manifested in differential cooling conditions here, can affect ligand binding modes. It's also notable that full-length MCPs show negative cooperativity in binding ligands, which is typically lost in the isolated periplasmic domains. Hence ligand binding is sensitive to the environment of a given domain. In short, the current data is not convincing enough to say that a previous "misconception" is being corrected.

Author Response

We would like to express our thorough gratitude to the editors and reviewers, for the helpful comments and valuable suggestions, which provided us an opportunity to further address our research. Prior to submitting our final revision, here we provide our preliminary responses for the comments. Please find our detailed responses to the reviewers’ recommendations below.

Reviewer #1 (Public Review):

Summary:

This study examines the spatial and temporal patterns of occurrence and the interspecific associations within a terrestrial mammalian community along human disturbance gradients. They conclude that human activity leads to a higher incidence of positive associations.

Strengths:

The theoretical framework of the study is brilliantly introduced. Solid data and sound methodology. This study is based on an extensive series of camera trap data. Good review of the literature on this topic.

Weaknesses:

The authors use the terms associations and interactions interchangeably.

Response: This is not the case. In fact, we state specifically that "... interspecific associations should not be directly interpreted as a signal of biotic interactions between pairs of species…" However, co-occurrence can be an important predictor of likely interactions, such as competition and predation. We stand by our original text.

It is not clear what the authors mean by "associations". A brief clarification would be helpful.

Response: Our specific definition of what is meant here by spatial association can be found in the Methods section. To clarify, the calculation of the index of associations is based on the covariance for the two species of the residuals (epsilon) after consideration of all species-specific response to known environmental covariates. These covariances are modelled to allow them to vary with the level of human disturbance, measured as human presence and human modification. After normalization, the final index of association is a correlation value that varies between -1 (complete disassociation) and +1 (complete positive association).

Also, the authors do not delve into the different types of association found in the study. A more ecological perspective explaining why certain species tend to exhibit negative associations and why others show the opposite pattern (and thus, can be used as indicator species) is missing.

Response: Suggesting the ecological underpinnings of the associations observed here would mainly be speculation at this point, but the associations demonstrated in this analysis do suggest promising areas for the more detailed research suggested.

Also, the authors do not distinguish between significant (true) non-random associations and random associations. In my opinion, associations are those in which two species co-occur more or less than expected by chance. This is not well addressed in the present version of the manuscript.

Response: Results were considered to be non-random if correlation coefficients (for spatial association) or overlap (for temporal association) fell outside of 95% Confidence Intervals. This is now stated clearly in the Methods section. In Supplementary Figures S2 and S3, p<0.01 levels are also presented.

The obtained results support the conclusions of the study.

Anthropogenic pressures can shape species associations by increasing spatial and temporal co-occurrence, but above a certain threshold, the positive influence of human activity in terms of species associations could be reverted. This study can stimulate further work in this direction.

Reviewer #2 (Public Review):

Summary:

This study analyses camera trapping information on the occurrence of forest mammals along a gradient of human modification of the environment. The key hypotheses are that human disturbance squeezes wildlife into a smaller area or their activity into only part of the day, leading to increased co-occurrence under modification. The method used is joint species distribution modelling (JSDM).

Strengths:

The data source seems to be very nice, although since very little information is presented, this is hard to be sure of. Also, the JSDM approach is, in principle, a nice way of simultaneously analysing the data.

Weaknesses:

The manuscript suffers from a mismatch of hypotheses and methods at two different levels.

1. At the lower level, we first need to understand what the individual species do and "like" (their environmental niche). That information is not presented, and the methods suggest that the representation of each species in the JSDM is likely to be extremely poor.

Response: The response of each species to the environmental covariates provides a window into their environmental niche, encapsulated in the beta coefficients for each environmental covariate. This information is presented in Figure 2.

1. The hypothesis clearly asks for an analysis of the statistical interaction between human disturbance and co-occurrence. Yet, the model is not set up this way, and the authors thus do a lot of indirect exploration, rather than direct hypothesis testing.

Response: Our JSDM model is set up specifically to examine the effect of human disturbance on co-occurrence, after controlling for shared responses to environmental variables. It directly tests the first hypothesis, since, if increase in indices of human disturbance had not tended to increase the measured spatial correlations between species as detected by the model, we would have rejected our stated hypothesis that human modification of habitats results in increased positive spatial associations between species.

Even when the focus is not the individual species, but rather their association, we need to formulate what the expectation is. The hypotheses point towards presenting the spatial and the temporal niche, and how it changes, species for species, under human disturbance. To this, one can then add the layer of interspecific associations.

Response: Examining each species one by one and how each one responds to human disturbance would miss the effects of any meaningful interactions between species. The analysis presented provides a means to highlight associations that would have been overlooked. Future research could go on to analyze the strongest associations in the community and the strongest effects of human disturbance so as to uncover the underlying interactions that give rise to them and the mechanisms of human impact. We believe that this will prove to be a much more productive approach than trying to tackle this problem species by species and pair by pair.

The change in activity and space use can be analysed much simpler, by looking at the activity times and spatial distribution directly. It remains unclear what the contribution of the JSDM is, unless it is able to represent this activity and spatial information, and put it in a testable interaction with human disturbance.

The topic is actually rather complicated. If biotic interactions change along the disturbance gradient, then observed data are already the outcome of such changed interactions. We thus cannot use the data to infer them! But we can show, for each species, that the habitat preferences change along the disturbance gradient - or not, as the case may be.

Then, in the next step, one would have to formulate specific hypotheses about which species are likely to change their associations more, and which less (based e.g. on predator-prey or competitive interactions). The data and analyses presented do not answer any of these issues.

Response: We suggest that the so-called “simpler” approach described above is anything but simple, and this is precisely what the Joint Species Distribution Model improves upon. As pointed out in the Introduction, simply examining spatial overlap is not enough to detect a signal of meaningful biotic interaction, since overlap could be the result of similar responses to environmental variables. With the JSDM approach, this would not be considered a positive association and would then not imply the possible existence of meaningful interaction.

Another more substantial point is that, according to my understanding of the methods, the per-species models are very inappropriate: the predictors are only linear, and there are no statistical interactions (L374). There is no conceivable species in the world whose niche would be described by such an oversimplified model.

Response: While interaction terms can be included in the JSDM, this would considerably increase the complexity of the models. In previous work, we have found no strong evidence for the importance of interaction terms and they do not improve the performance of the models.

We have no idea of even the most basic characteristics of the per-species models: prevalences, coefficient estimates, D2 of the model, and analysis of the temporal and spatial autocorrelation of the residuals, although they form the basis for the association analysis!

Response: The coefficient estimates for response to environmental variables used in the JSDM are provided in Figure 2.

Why are times of day and day of the year not included as predictors IN INTERACTION with niche predictors and human disturbance, since they represent the temporal dimension on which niches are hypothesised to change?

Also, all correlations among species should be shown for the raw data and for the model residuals: how much does that actually change and can thus be explained by the niche models?

The discussion has little to add to the results. The complexity of the challenge (understanding a community-level response after accounting for species-level responses) is not met, and instead substantial room is given to general statements of how important this line of research is. I failed to see any advance in ecological understanding at the community level.

Response: We agree that the community-level response to human disturbance is a complex topic, and we believe it is also a very important one. This research and its support of the spatial compression hypothesis, while not providing definitive answers to detailed mechanisms, opens up new lines of inquiry that makes it an important advance. For example, the strong effects of human disturbance on certain associations that were detected here could now be examined with the kind of detailed species by species and pair by pair analysis that this reviewer appears to demand.

eLife assessment

In this study, camera trapping and species distribution models are used to show that human disturbance in mountain forests in the eastern Himalayas pushes medium-sized and large mammal species into narrower habitat space, thus increasing their co-occurrence. While the collected data provide a useful basis for further work, the study presents incomplete evidence to support the claim that increased co-occurrence may indicate positive interactions between species.

Reviewer #1 (Public Review):

Summary:<br /> This study examines the spatial and temporal patterns of occurrence and the interspecific associations within a terrestrial mammalian community along human disturbance gradients. They conclude that human activity leads to a higher incidence of positive associations.

Strengths:<br /> The theoretical framework of the study is brilliantly introduced. Solid data and sound methodology. This study is based on an extensive series of camera trap data. Good review of the literature on this topic.

Weaknesses:<br /> The authors use the terms associations and interactions interchangeably. It is not clear what the authors mean by "associations". A brief clarification would be helpful. Also, the authors do not delve into the different types of association found in the study. A more ecological perspective explaining why certain species tend to exhibit negative associations and why others show the opposite pattern (and thus, can be used as indicator species) is missing. Also, the authors do not distinguish between significant (true) non-random associations and random associations. In my opinion, associations are those in which two species co-occur more or less than expected by chance. This is not well addressed in the present version of the manuscript.

The obtained results support the conclusions of the study.

Anthropogenic pressures can shape species associations by increasing spatial and temporal co-occurrence, but above a certain threshold, the positive influence of human activity in terms of species associations could be reverted. This study can stimulate further work in this direction.

Reviewer #2 (Public Review):

Summary:<br /> This study analyses camera trapping information on the occurrence of forest mammals along a gradient of human modification of the environment. The key hypotheses are that human disturbance squeezes wildlife into a smaller area or their activity into only part of the day, leading to increased co-occurrence under modification. The method used is joint species distribution modelling (JSDM).

Strengths:<br /> The data source seems to be very nice, although since very little information is presented, this is hard to be sure of. Also, the JSDM approach is, in principle, a nice way of simultaneously analysing the data.

Weaknesses:<br /> The manuscript suffers from a mismatch of hypotheses and methods at two different levels.

1) At the lower level, we first need to understand what the individual species do and "like" (their environmental niche). That information is not presented, and the methods suggest that the representation of each species in the JSDM is likely to be extremely poor.

2) The hypothesis clearly asks for an analysis of the statistical interaction between human disturbance and co-occurrence. Yet, the model is not set up this way, and the authors thus do a lot of indirect exploration, rather than direct hypothesis testing.

Even when the focus is not the individual species, but rather their association, we need to formulate what the expectation is. The hypotheses point towards presenting the spatial and the temporal niche, and how it changes, species for species, under human disturbance. To this, one can then add the layer of interspecific associations.

The change in activity and space use can be analysed much simpler, by looking at the activity times and spatial distribution directly. It remains unclear what the contribution of the JSDM is, unless it is able to represent this activity and spatial information, and put it in a testable interaction with human disturbance.

The topic is actually rather complicated. If biotic interactions change along the disturbance gradient, then observed data are already the outcome of such changed interactions. We thus cannot use the data to infer them! But we can show, for each species, that the habitat preferences change along the disturbance gradient - or not, as the case may be.

Then, in the next step, one would have to formulate specific hypotheses about which species are likely to change their associations more, and which less (based e.g. on predator-prey or competitive interactions). The data and analyses presented do not answer any of these issues.

Another more substantial point is that, according to my understanding of the methods, the per-species models are very inappropriate: the predictors are only linear, and there are no statistical interactions (L374). There is no conceivable species in the world whose niche would be described by such an oversimplified model.

We have no idea of even the most basic characteristics of the per-species models: prevalences, coefficient estimates, D2 of the model, and analysis of the temporal and spatial autocorrelation of the residuals, although they form the basis for the association analysis! Why are times of day and day of the year not included as predictors IN INTERACTION with niche predictors and human disturbance, since they represent the temporal dimension on which niches are hypothesised to change?

Also, all correlations among species should be shown for the raw data and for the model residuals: how much does that actually change and can thus be explained by the niche models?

The discussion has little to add to the results. The complexity of the challenge (understanding a community-level response after accounting for species-level responses) is not met, and instead substantial room is given to general statements of how important this line of research is. I failed to see any advance in ecological understanding at the community level.

Author Response

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

Reviewer #1 (Public Review):

The manuscript has helped address a long-standing mystery in splicing regulation: whether splicing occurs co- or post-transcriptionally. Specifically, the authors (1) uniquely combined smFISH, expansion microscopy, and live cell imaging; (2) revealed the ordering and spatial distribution of splicing steps; and (3) discovered that nascent, not-yet-spliced transcripts move more slowly around the transcription site and undergo splicing as they move through the clouds. Based on the experimental results, the authors suggest that the observation of co-transcriptional splicing in previous literature could be due to the limitation of imaging resolution, meaning that the observed co-transcriptional splicing might actually be post-transcriptional splicing occurring in proximity to the transcription site. Overall, the work presented here clearly provides a comprehensive picture of splicing regulation.

Major points:

1. Linearity of expansion microscopy. For Figure 2B, it would be helpful to display the same sample before and after expansion, just like Supplementary Figure 3, but with a transcription site and "cloud". In the current version, the transcription site looks quite different in the not-expanded (more green dots on the left) and expanded image (more green dots on the top).

We thank the reviewer for this comment on linearity of expansion. Based on our prior manuscript (Chen et al 2015 Nature Methods. PMID: 27376770), we expect expansion microscopy to yield isotropic expansion. Indeed, as shown in Supplemental Figure 3, we confirmed that expansion of nuclei (3B, top) and transcripts (3B, bottom) is isotropic. Additionally, before splicing inhibition, we demonstrated the linearity of expansion for a transcription site (3B, left), shown at standard resolution with intron stain. The images shown in Figure 2B are meant solely to illustrate the change in resolution upon expansion, and are not meant to imply spatial matching between the expanded and unexpanded image. We apologize for the confusion and have clarified this in the figure legend for Figure 2.

We also point the reader towards Supplemental Figure 4, in which we validate the use of expansion microscopy in these findings. We show that transcription sites in expanded samples were the same size as those imaged using stochastic optical reconstruction microscopy (STORM), demonstrating that expansion did not significantly alter the morphology of the site.

1. FISH dot colocalization. What is the colocalization rate of FISH dots in general under experimental conditions? In addition, in Figures 2C and 2G, why do some 3'exon dots not have co-localized 5'exon dots?

We thank the reviewer for asking for these important clarifications. Under standard (non-expanded) conditions, our colocalization of 3’ and 5’ spots varies by gene, but more than 75% of intron spots colocalize with exon spots for the vast majority of transcripts we evaluated. The percentage of colocalization for each gene and intron can be found in column 4 of Table 1.

Regarding the second point—these individual images may not reflect the actual quantitative number of spot counts at the site, as these transcription sites have a sizable Z dimension that is difficult to capture in one image, and certain dyes are more easily visually distinguished in contrasted images than others. These factors may cause some 3’ spots to appear without a corresponding co-localized 5’ spot in these images. We refer the reviewer to Supplemental Figure 4C for quantitative spot counting of an expanded transcription site, for which there are a similar number of 3’ end and 5’ end spots within the entire Z-stacked image. Importantly, these transcription site clouds contain longer, unspliced transcripts, potentially leading to further separation between the 5’ and 3’ ends of a single transcript when compared to a cytoplasmic, spliced transcript (quantified in Figure 2I).

1. It would be helpful if the authors uploaded a few examples of live cell imaging movies.

Certainly! Please refer to the new Supplementary Movies 1-3 for representative examples of live cell imaging data.

1. It is recommended to double-check the text for errors.

We apologize for errors in the original manuscript, and have made the appropriate corrections.

Reviewer #2 (Public Review):

Allison Coté et al. investigated the ordering and spatial distribution of nascent transcripts in several cells using smFISH, expansion microscopy, and live-cell imaging. They find that pre-mRNA splicing occurs post-transcriptionally at the clouds around the transcription start site, termed the transcription site proximal zone. They show that pre-mRNA may undergo continuous splicing when they pass through the zone after transcription. These data suggest a unifying model for explaining previously reported co-transcriptional splicing events and provide a direction for further study of the nature of the slow-moving zone around the transcription start site.

This paper is well-written. The findings are very important, and the data supports the conclusions well. However, some aspects of the image and description need to be clarified and revised.

The authors describe Figure 4E and 4F results in the main text as that "we performed RNA FISH simultaneously with immunofluorescence for SC35, a component of speckles, and saw that this compartmentalized pre-mRNA did indeed appear near nuclear speckles both before (Supplementary Figure 6C) and after (Figure 4E) splicing inhibition." However, no SC35 staining is shown in the Figure 4E. A similar situation happened in describing Figure 4F.

We thank the reviewer for noting this error. We mistakenly called in text for Figure 4E, when we meant to refer to Figure 4G, which shows combined RNA FISH and SC35 immunofluorescence show compartmentalization within nuclear speckles. Figures 4E and 4F do not show SC35 immunofluorescence. We have altered the text and figure captions accordingly. Recommendations for the authors: please note that you control which revisions to undertake from the public reviews and recommendations for the authors Reviewer #1 (Recommendations For The Authors):

Minor points:

1. For Figures, it would be better to mark co-transcriptional and proximal post-transcriptional splicing in a clearer way. Like in Figure 1A, the simulated RNA FISH signals are almost identical across two conditions, which is a bit confusing. Overlapping and close proximity shall be better illustrated in related figures.

We thank the reviewer for these suggestions. We have iterated these figures through multiple revisions and have found that these diagrams tend to resonate the most, so we have elected to keep them as is, but we do appreciate the suggestion.

1. May include some details of expansion microscopy in the last paragraph of the Introduction. For example, why introduce expansion microscopy? To what level it can help overcome the diffraction limit?

We thank the reviewer for this comment, and have added additional text to this paragraph to further set up the use of expansion microscopy.

1. Double-check the formatting. Some sub-titles are in Bold, some in Italic.

We apologize for any formatting errors, and have made the appropriate corrections.

1. Please double-check the writing. I find many incompatible parts across the manuscript. For example, as described in the Figure 1D caption, there aren't "first" and "second" graphs in the figure. Moreover, some writings require additional refinement. For instance, in the Introduction part, the paragraph discussing RNA imaging, various techniques (such as FISH and live imaging), and concerns (such as microscopy resolution, chromatin fraction, and limitations related to reporter genes) are intertwined without clear indexing or logical structuring. Similar cases in other paragraphs too. Last but not least, I can even find repetitive sentences across the manuscript. For instance, I believe that the authors forgot to delete "By distinguishing the separate fluorescent signals from probes bound to exons and introns, we could visualize splicing intermediates (represented by colocalized intron and exon spots) relative to the site of transcription (represented by bright colocalized intron and exon spots) and fully spliced products (represented by exon spots alone)." in the first paragraph of the Results part, as the exact same sentence re-occurs right after. I've only listed a few examples here. Please refine the manuscript.

We apologize for any errors in the original manuscript, and have made the appropriate corrections.

Reviewer #2 (Recommendations For The Authors):

1. The sentence "By distinguishing the separate fluorescent signals from probes bound to exons and introns, we could visualize splicing intermediates (represented by colocalized intron and exon spots) relative to the site of transcription (represented by bright colocalized intron and exon spots) and fully spliced products (represented by exon spots alone)." is accidentally repeated twice, one of them should be deleted.

We apologize for this duplication, and have made the appropriate correction.

eLife assessment

This fundamental study addresses a long-standing mystery in splicing regulation: does splicing occur co- or post-transcriptionally? The authors provide compelling evidence demonstrating that splicing can occur post-transcriptionally at a transcription site proximal zone, changing the way we think about splicing.

Reviewer #1 (Public Review):

The manuscript has helped address a long-standing mystery in splicing regulation: whether splicing occurs co- or post-transcriptionally. Specifically, the authors (1) uniquely combined smFISH, expansion microscopy, and live cell imaging; (2) revealed the ordering and spatial distribution of splicing steps; and (3) discovered that nascent, not-yet-spliced transcripts move more slowly around the transcription site and undergo splicing as they move through the clouds. Based on the experimental results, the authors suggest that the observation of co-transcriptional splicing in previous literature could be due to the limitation of imaging resolution, meaning that the observed co-transcriptional splicing might actually be post-transcriptional splicing occurring in proximity to the transcription site. Overall, the work presented here clearly provides a comprehensive picture of splicing regulation.

Reviewer #2 (Public Review):

Allison Coté et al. investigated the ordering and spatial distribution of nascent transcripts in several cells using smFISH, expansion microscopy, and live-cell imaging. They find that pre-mRNA splicing occurs post-transcriptionally at the clouds around the transcription start site, termed the transcription site proximal zone. They show that pre-mRNA may undergo continuous splicing when they pass through the zone after transcription. These data suggest a unifying model for explaining previously reported co-transcriptional splicing events and provide a direction for further study of the nature of the slow-moving zone around the transcription start site.

This paper is well-written. The findings are very important, and the data supports the conclusions well. However, some aspects of the image and description need to be clarified and revised.

1) The sentence "By distinguishing the separate fluorescent signals from probes bound to exons and introns, we could visualize splicing intermediates (represented by colocalized intron and exon spots) relative to the site of transcription (represented by bright colocalized intron and exon spots) and fully spliced products (represented by exon spots alone)." is accidentally repeated twice, one of them should be deleted.<br /> 2) The authors describe Figure 4E and 4F results in the main text as that "we performed RNA FISH simultaneously with immunofluorescence for SC35, a component of speckles, and saw that these compartmentalized pre-mRNA did indeed appear near nuclear speckles both before (Supplementary Figure 6C) and after (Figure 4E) splicing inhibition." However, no SC35 staining is shown in the Figure 4E. A similar situation happened in describing Figure 4F.

eLife assessment

Overall, this is a significant study, and it is able to highlight mast cells in amphibians and their putative capability to respond to and combat fungal infections. Therefore, this study is important for the field. However, the manuscript is incomplete from the standpoint that there is functional data lacking on how these mast cells are activated and their precise functional properties. Such experiments would add substantial impact and rigor and fully support the conclusions.

Reviewer #1 (Public Review):

Summary:<br /> The global decline of amphibians is primarily attributed to deadly disease outbreaks caused by the chytrid fungus, Batrachochytrium dendrobatidis (Bd). It is unclear whether and how skin-resident immune cells defend against Bd. Although it is well known that mammalian mast cells are crucial immune sentinels in the skin and play a pivotal role in the immune recognition of pathogens and orchestrating subsequent immune responses, the roles of amphibian mast cells during Bd infections are largely unknown. The current study developed a novel way to enrich X. laevis skin mast cells by injecting the skin with recombinant stem cell factor (SCF), a KIT ligand required for mast cell differentiation and survival. The investigators found an enrichment of skin mast cells provides X. laevis substantial protection against Bd and mitigates the inflammation-related skin damage resulting from Bd infection. Additionally, the augmentation of mast cells leads to increased mucin content within cutaneous mucus glands and shields frogs from the alterations to their skin microbiomes caused by Bd.

Strengths:<br /> This study underscores the significance of amphibian skin-resident immune cells in defenses against Bd and introduces a novel approach to examining interactions between amphibian hosts and fungal pathogens.

Weaknesses:<br /> The main weakness of the study is the lack of functional analysis of X. laevis mast cells. Upon activation, mast cells have the characteristic feature of degranulation to release histamine, serotonin, proteases, cytokines, and chemokines, etc. The study should determine whether X. laevis mast cells can be degranulated by two commonly used mast cell activators IgE and compound 48/80 for IgE-dependent and independent pathways. This can be easily done in vitro. It is also important to assess whether in vivo these mast cells are degranulated upon Bd infection using avidin staining to visualize vesicle releases from mast cells. Figure 3 only showed rSCF injection caused an increase in mast cells in naïve skin. They need to present whether Bd infection can induce mast cell increase and rSCF injection under Bd infection causes a mast cell increase in the skin. In addition, it is unclear how the enrichment of mast cells provides protection against Bd infection and alternations to skin microbiomes after infection. It is important to determine whether skin mast cells release any contents mentioned above.

Reviewer #2 (Public Review):

Summary:<br /> In this study, Hauser et al investigate the role of amphibian (Xenopus laevis) mast cells in cutaneous immune responses to the ecologically important pathogen Batrachochytrium dendrobatidis (Bd) using novel methods of in vitro differentiation of bone marrow-derived mast cells and in vivo expansion of skin mast cell populations. They find that bone marrow-derived myeloid precursors cultured in the presence of recombinant X. laevis Stem Cell Factor (rSCF) differentiate into cells that display hallmark characteristics of mast cells. They inject their novel (r)SCF reagent into the skin of X. laevis and find that this stimulates the expansion of cutaneous mast cell populations in vivo. They then apply this model of cutaneous mast cell expansion in the setting of Bd infection and find that mast cell expansion attenuates the skin burden of Bd zoospores and pathologic features including epithelial thickness and improves protective mucus production and transcriptional markers of barrier function. Utilizing their prior expertise with expanding neutrophil populations in X. laevis, the authors compare mast cell expansion using (r)SCF to neutrophil expansion using recombinant colony-stimulating factor 3 (rCSF3) and find that neutrophil expansion in Bd infection leads to greater burden of zoospores and worse skin pathology.

Strengths:<br /> The authors report a novel method of expanding amphibian mast cells utilizing their custom-made rSCF reagent. They rigorously characterize expanded mast cells in vitro and in vivo using histologic, morphologic, transcriptional, and functional assays. This establishes solid footing with which to then study the role of rSCF-stimulated mast cell expansion in the Bd infection model. This appears to be the first demonstration of the exogenous use of rSCF in amphibians to expand mast cell populations and may set a foundation for future mechanistic studies of mast cells in the X. laevis model organism.

Weaknesses:<br /> The conclusions regarding the role of mast cell expansion in controlling Bd infection would be stronger with a more rigorous evaluation of the model, as there are some key gaps and remaining questions regarding the data. For example:

1. Granulocyte expansion is carefully quantified in the initial time courses of rSCF and rCSF3 injections, but similar quantification is not provided in the disease models (Figures 3E, 4G, 5D-G). A key implication of the opposing effects of mast cell vs neutrophil expansion is that mast cells may suppress neutrophil recruitment or function. Alternatively, mast cells also express notable levels of csfr3 (Figure 2) and previous work from this group (Hauser et al, Facets 2020) showed rG-CSF-stimulated peritoneal granulocytes express mast cell markers including kit and tpsab1, raising the question of what effect rCSF3 might have on mast cell populations in the skin. Considering these points, it would be helpful if both mast cells and neutrophils were quantified histologically (based on Figure 1, they can be readily distinguished by SE or Giemsa stain) in the Bd infection models.

2. Epithelial thickness and inflammation in Bd infection are reported to be reduced by rSCF treatment (Figure 3E, 5A-B) or increased by rCSF3 treatment (Figure 4G) but quantification of these critical readouts is not shown.

3. Critical time points in the Bd model are incompletely characterized. Mast cell expansion decreases zoospore burden at 21 dpi, while there is no difference at 7 dpi (Figure 3E). Conversely, neutrophil expansion increases zoospore burden at 7 dpi, but no corresponding 21 dpi data is shown for comparison (Figure 4G). Microbiota analysis is performed at a third time point,10 dpi (Figure 5D-G), making it difficult to compare with the data from the 7 dpi and 21 dpi time points. Reporting consistent readouts at these three time points is important to draw solid conclusions about the relationship of mast cell expansion to Bd infection and shifts in microbiota.

4. Although the effect of rSCF treatment on Bd zoospores is significant at 21 dpi (Figure 3E), bacterial microbiota changes at 21 dpi are not (Figure S3B-C). This discrepancy, how it relates to the bacterial microbiota changes at 10 dpi, and why 7, 10, and 21 dpi time points were chosen for these different readouts (Figure 5F-G), is not discussed.

5. The time course of rSCF or rCSF3 treatments relative to Bd infection in the experiments is not clear. Were the treatments given 12 hours prior to the final analysis point to maximize the effect? For example, in Figure 3E, were rSCF injections given at 6.5 dpi and 20.5 dpi? Or were treatments administered on day 0 of the infection model? If the latter, how do the authors explain the effects at 7 dpi or 21 dpi given mast cell and neutrophil numbers return to baseline within 24 hours after rSCF or rCSF3 treatment, respectively?

The title of the manuscript may be mildly overstated. Although Bd infection can indeed be deadly, mortality was not a readout in this study, and it is not clear from the data reported that expanding skin mast cells would ultimately prevent progression to death in Bd infections.

Reviewer #3 (Public Review):

Summary:<br /> Hauser et al. provide an exceptional study describing the role of resident mast cells in amphibian epidermis that produce anti-inflammatory cytokines that prevent Batrachochytrium dendrobatidis (Bd) infection from causing harmful inflammation, and also protect frogs from changes in skin microbiomes and loss of mucin in glands and loss of mucus integrity that otherwise cause changes to their skin microbiomes. Neutrophils, in contrast, were not protective against Bd infection. Beyond the beautiful cytology and transcriptional profiling, the authors utilized elegant cell enrichment experiments to enrich mast cells by recombinant stem cell factor, or to enrich neutrophils by recombinant colony-stimulating factor-3, and examined respective infection outcomes in Xenopus.

Strengths:<br /> Through the use of recombinant IL4, the authors were able to test and eliminate the hypothesis that mast cell production of IL4 was the mechanism of host protection from Bd infection. Instead, impacts on the mucus glands and interaction with the skin microbiome are implicated as the protective mechanism. These results will press disease ecologists to examine the relative importance of this immune defense among species, the influence of mast cells on the skin microbiome and mucosal function, and open the potential for modulating mucosal defense.

Weaknesses:<br /> A reduction of bacterial diversity upon infection, as described at the end of the results section, may not always be an "adverse effect," particularly given that anti-Bd function of the microbiome increased. Some authors (see Letourneau et al. 2022 ISME, or Woodhams et al. 2023 DCI) consider these short-term alterations as encoding ecological memory, such that continued exposure to a pathogen would encounter an enriched microbial defense. Regardless, mast cell-initiated protection of the mucus layer may negate the need for this microbial memory defense.

While the description of the mast cell location in the epidermal skin layer in amphibians is novel, it is not known how representative these results are across species ranging in chytridiomycosis susceptibility. No management applications are provided such as methods to increase this defense without the use of recombinant stem cell factor, and more discussion is needed on how the mast cell component (abundance, distribution in the skin) of the epidermis develops or is regulated.

Author Response:

Reviewer #1 (Public Review):

Summary:<br /> The global decline of amphibians is primarily attributed to deadly disease outbreaks caused by the chytrid fungus, Batrachochytrium dendrobatidis (Bd). It is unclear whether and how skin-resident immune cells defend against Bd. Although it is well known that mammalian mast cells are crucial immune sentinels in the skin and play a pivotal role in the immune recognition of pathogens and orchestrating subsequent immune responses, the roles of amphibian mast cells during Bd infections are largely unknown. The current study developed a novel way to enrich X. laevis skin mast cells by injecting the skin with recombinant stem cell factor (SCF), a KIT ligand required for mast cell differentiation and survival. The investigators found an enrichment of skin mast cells provides X. laevis substantial protection against Bd and mitigates the inflammation-related skin damage resulting from Bd infection. Additionally, the augmentation of mast cells leads to increased mucin content within cutaneous mucus glands and shields frogs from the alterations to their skin microbiomes caused by Bd.

Strengths:<br /> This study underscores the significance of amphibian skin-resident immune cells in defenses against Bd and introduces a novel approach to examining interactions between amphibian hosts and fungal pathogens.

Weaknesses:<br /> The main weakness of the study is the lack of functional analysis of X. laevis mast cells. Upon activation, mast cells have the characteristic feature of degranulation to release histamine, serotonin, proteases, cytokines, and chemokines, etc. The study should determine whether X. laevis mast cells can be degranulated by two commonly used mast cell activators IgE and compound 48/80 for IgE-dependent and independent pathways. This can be easily done in vitro. It is also important to assess whether in vivo these mast cells are degranulated upon Bd infection using avidin staining to visualize vesicle releases from mast cells. Figure 3 only showed rSCF injection caused an increase in mast cells in naïve skin. They need to present whether Bd infection can induce mast cell increase and rSCF injection under Bd infection causes a mast cell increase in the skin. In addition, it is unclear how the enrichment of mast cells provides protection against Bd infection and alternations to skin microbiomes after infection. It is important to determine whether skin mast cells release any contents mentioned above.

We would like to thank the reviewer for taking the time to review our work and for providing us with valuable feedback.

Please note that amphibians do not possess the IgE antibody isotype1.

To our knowledge there have been no published studies using approaches for studying mammalian mast cell degranulation to examine amphibian mast cells. Notably, several studies suggest that amphibian mast cells lack histamine2, 3, 4, 5 and serotonin2, 6. While there are commercially available kits and reagents for examining mammalian mast cell granule content, most of these reagents may not cross-react with their amphibian counterparts. This is especially true of cytokines and chemokines, which diverged quickly with evolution and thus do not share substantial protein sequence identity across species as divergent as frogs and mammals. Respectfully, while following up on these findings is possible, it would involve considerable additional work to find reagents that would detect amphibian mast cell contents.

We would also like to respectfully point out that while mast cell degranulation is a feature most associated with mammalian mast cells, this is not the only means by which mammalian mast cells confer their immunological effects. While we agree that defining the biology of amphibian mast cell degranulation is important, we anticipate that since the anti-Bd protection conferred by enriching frog mast cells is seen after 21 days of enrichment, it is quite possible that degranulation may not be the central mechanism by which the mast cells are mediating this protection.

As noted in our manuscript, frog mast cells upregulate their expression of interleukin-4 (IL4), which is a hallmark cytokine associated with mammalian mast cells7. We are presently exploring the role of the frog IL4 in the observed mast cell anti-Bd protection. Should we generate meaningful findings in this regard, we will add them to the revised version of this manuscript.

We are also exploring the heparin content of frog mast cells and capacities of these cells to degranulate in vitro in response to compound 48/80. In addition, we are exploring in vivo mast cell degranulation via histology and avidin-staining. Should these studies generate significant findings, we will include them in the revised version of this manuscript.

Per the reviewer’s suggestion, in our revised manuscript we also plan to include data showing whether Bd infections affect skin mast cell numbers and how rSCF injection impacts skin mast cell numbers in the context of Bd infections.

In regard to how mast cells impact Bd infections and skin microbiomes, our data indicate that mast cells are augmenting skin integrity during Bd infections and promoting mucus production, as indicated by the findings presented in Figure 4A-C and Figure 5A-C, respectively. There are several mammalian mast cell products that elicit mucus production. In mammals, this mucus production is mediated by goblet cells while the molecular control of amphibian skin mucus gland content remains incompletely understood. Interleukin-13 (IL13) is the major cytokine associated with mammalian mucus production8, while to our knowledge this cytokine is either not encoded by amphibians or else has yet to be identified and annotated in these animals’ genomes. IL4 signaling also results in mucus production9 and we are presently exploring the possible contribution of the X. laevis IL4 to skin mucus gland filling. Any significant findings on this front will be included in the revised manuscript. Histamine release contributes to mast cell-mediated mucus production10, but as we outline above, several studies indicate that amphibian mast cells may lack histamine2, 3, 4, 5. Mammalian mast cell-produced lipid mediators also play a critical role in eliciting mucus secretion11 and our transcriptomic analysis indicates that frog mast cells express several enzymes associated with production of such mediators. We will highlight this observation in our revised manuscript.

We anticipate that X. laevis mast cells influence skin integrity, microbial composition and Bd susceptibility in a myriad of ways. Considering the substantial differences between amphibian and mammalian evolutionary histories and physiologies, we anticipate that many of the mechanisms by which X. laevis mast cells confer anti-Bd protection will prove to be specific to amphibians and some even unique to X. laevis. We are most interested in deciphering what these mechanisms are but foresee that they will not necessarily reflect what one would expect based on what we know about mammalian mast cells in the context of mammalian physiologies.

Reviewer #2 (Public Review):

Summary:<br /> In this study, Hauser et al investigate the role of amphibian (Xenopus laevis) mast cells in cutaneous immune responses to the ecologically important pathogen Batrachochytrium dendrobatidis (Bd) using novel methods of in vitro differentiation of bone marrow-derived mast cells and in vivo expansion of skin mast cell populations. They find that bone marrow-derived myeloid precursors cultured in the presence of recombinant X. laevis Stem Cell Factor (rSCF) differentiate into cells that display hallmark characteristics of mast cells. They inject their novel (r)SCF reagent into the skin of X. laevis and find that this stimulates the expansion of cutaneous mast cell populations in vivo. They then apply this model of cutaneous mast cell expansion in the setting of Bd infection and find that mast cell expansion attenuates the skin burden of Bd zoospores and pathologic features including epithelial thickness and improves protective mucus production and transcriptional markers of barrier function. Utilizing their prior expertise with expanding neutrophil populations in X. laevis, the authors compare mast cell expansion using (r)SCF to neutrophil expansion using recombinant colony-stimulating factor 3 (rCSF3) and find that neutrophil expansion in Bd infection leads to greater burden of zoospores and worse skin pathology.

Strengths: <br /> The authors report a novel method of expanding amphibian mast cells utilizing their custom-made rSCF reagent. They rigorously characterize expanded mast cells in vitro and in vivo using histologic, morphologic, transcriptional, and functional assays. This establishes solid footing with which to then study the role of rSCF-stimulated mast cell expansion in the Bd infection model. This appears to be the first demonstration of the exogenous use of rSCF in amphibians to expand mast cell populations and may set a foundation for future mechanistic studies of mast cells in the X. laevis model organism. 

We thank the reviewer for recognizing the breadth and extent of the undertaking that culminated in this manuscript. Indeed, this manuscript would not have been possible without considerable reagent development and adaptation of techniques that had previously not been used for amphibian immunity research. In line with the reviewer’s sentiment, to our knowledge this is the first report of using molecular approaches to augment amphibian mast cells, which we hope will pave the way for new areas of research within the fields of comparative immunology and amphibian disease biology.

Weaknesses:<br /> The conclusions regarding the role of mast cell expansion in controlling Bd infection would be stronger with a more rigorous evaluation of the model, as there are some key gaps and remaining questions regarding the data. For example:

1. Granulocyte expansion is carefully quantified in the initial time courses of rSCF and rCSF3 injections, but similar quantification is not provided in the disease models (Figures 3E, 4G, 5D-G). A key implication of the opposing effects of mast cell vs neutrophil expansion is that mast cells may suppress neutrophil recruitment or function. Alternatively, mast cells also express notable levels of csfr3 (Figure 2) and previous work from this group (Hauser et al, Facets 2020) showed rG-CSF-stimulated peritoneal granulocytes express mast cell markers including kit and tpsab1, raising the question of what effect rCSF3 might have on mast cell populations in the skin. Considering these points, it would be helpful if both mast cells and neutrophils were quantified histologically (based on Figure 1, they can be readily distinguished by SE or Giemsa stain) in the Bd infection models.

We thank the reviewer for this insightful suggestion. We are performing a further examination of skin granulocyte content during Bd infections and plan on including any significant findings in our revised manuscript.

We predict that rSCF administration results in the accumulation of mast cells that are polarized such that they ablate the inflammatory response elicited by Bd infection. Mammalian mast cells, including peritonea-resident mast cells, express csf3r12, 13. Although the X. laevis animal model does not permit nearly the degree of immune cell resolution afforded by mammalian animal models, we do know that the adult X. laevis peritonea contain heterogenous leukocyte populations. We anticipate that the high kit expression reported by Hauser et al., 2020 in the rCSF3-recruited peritoneal leukocytes reflects the presence of mast cells therein. As such and in acknowledgement of the reviewer’s suggestion, we also think that the cells recruited by rCSF3 into the skin may include not only neutrophils but also mast cells. Possibly, these mast cells have distinct polarization states from those enriched by rSCF. While the lack of antibodies against frog neutrophils or mast cells has limited our capacity to address this question, we will attempt to reexamine by histology the proportions of skin neutrophils and mast cells in the skins of frogs under the conditions described in our manuscript. Any new findings in this regard will be included in the revised version of this work.

2. Epithelial thickness and inflammation in Bd infection are reported to be reduced by rSCF treatment (Figure 3E, 5A-B) or increased by rCSF3 treatment (Figure 4G) but quantification of these critical readouts is not shown.

We thank the reviewer for this suggestion. We will score epithelial thickness under the distinct conditions described in our manuscript and present the quantified data in the revised paper.

3. Critical time points in the Bd model are incompletely characterized. Mast cell expansion decreases zoospore burden at 21 dpi, while there is no difference at 7 dpi (Figure 3E). Conversely, neutrophil expansion increases zoospore burden at 7 dpi, but no corresponding 21 dpi data is shown for comparison (Figure 4G). Microbiota analysis is performed at a third time point,10 dpi (Figure 5D-G), making it difficult to compare with the data from the 7 dpi and 21 dpi time points. Reporting consistent readouts at these three time points is important to draw solid conclusions about the relationship of mast cell expansion to Bd infection and shifts in microbiota.

Because there were no significant effects of mast cell enrichment at 7 days post Bd infection, we chose to look at the microbiome composition in a subsequent experiment at 10 days and 21 days post Bd infection, with 10 days being a bit more of a midway point between the initial exposure and day 21, when we see the effect on Bd loads. We will clarify this rationale in the revised manuscript.

The enrichment of neutrophils in frog skins resulted in prompt (12 hours post enrichment) skin thickening (in absence of Bd infection) and increased frog Bd susceptibility by 7 days of infection. Conversely, mast cell enrichment stabilized skin mucosal and symbiotic microbial environment, presumably accounting at least in part for the lack of further Bd growth on mast cell-enriched animals by 21 days of infection. Our question regarding the roles of inflammatory granulocytes/neutrophils during Bd infections was that of ‘how’ rather ‘when’ these cells affect Bd infections. Because the central focus of this work was mast cells and not other granulocyte subsets, when we saw that rCSF3-recruited granulocytes adversely affected Bd infections at 7 days post infection, we did not pursue the kinetics of these responses further. We plan to explore the roles of inflammatory mediators and disparate frog immune cell subsets during the course of Bd infections, but we feel that these future studies are more peripheral to the central thesis of the present manuscript regarding the roles of frog mast cells during Bd infections.

4. Although the effect of rSCF treatment on Bd zoospores is significant at 21 dpi (Figure 3E), bacterial microbiota changes at 21 dpi are not (Figure S3B-C). This discrepancy, how it relates to the bacterial microbiota changes at 10 dpi, and why 7, 10, and 21 dpi time points were chosen for these different readouts (Figure 5F-G), is not discussed.

Our results indicate that after 10 days of Bd infection, control Bd-challenged animals exhibited reduced microbial richness, while skin mast cell-enriched Bd-infected frogs were protected from this disruption of their microbiome. The amphibian microbiome serves as a major barrier to these fungal infections14, and we anticipate that Bd-mediated disruption of microbial richness and composition facilitates host skin colonization by this pathogen. Control and mast cell-enriched animals had similar skin Bd loads at 10 days post infection. However, by 21 days of Bd infection the mast cells-enriched animals maintained their Bd loads to levels observed at 10 days post infection, whereas the control animals had significantly greater Bd loads. Thus, we anticipate that frog mast cells are conferring the observed anti-Bd protection in part by preventing microbial disassembly and thus interfering with optimal Bd colonization and growth on frog skins. In other words, maintained microbial composition at 10 days of infection may be preventing additional Bd colonization/growth, as seen when comparing skins of control and mast cell-enriched frogs at 21 days post infection. By 21 days of infection, control animals rebounded from the Bd-mediated reduction in bacterial richness seen at 10 days. Considering that after 21 days of infection control animals also had significantly greater Bd loads than mast-cell enriched animals suggests that there may be a critical earlier window during which microbial composition is able to counteract _Bd_growth. 

While the current draft of our manuscript has a paragraph to this effect (see below), we appreciate the reviewer conveying to us that our perspective on the relationship between skin mast cells and the kinetics of microbial composition and _Bd_loads could be better emphasized. We plan to revise our manuscript to include the above discussion points. 

Bd infections caused major reductions in bacterial taxa richness, changes in composition and substantial increases in the relative abundance of Bd-inhibitory bacteria early in the infection. Similar changes to microbiome structure occur during experimental Bd infections of red-backed salamanders and mountain yellow-legged frogs15, 16. In turn, progressing Bd_infections corresponded with a return to baseline levels of _Bd-inhibitory bacteria abundance and rebounding microbial richness, albeit with dissimilar communities to those seen in control animals. These temporal changes indicate that amphibian microbiomes are dynamic, as are the effects of Bd infections on them. Indeed, Bd infections may have long-lasting impacts on amphibian microbiomes15. While Bd infections manifested in these considerable changes to frog skin microbiome structure, mast cell enrichment appeared to counteract these deleterious effects to their microbial composition. Presumably, the greater skin mucosal integrity and mucus production observed after mast cell enrichment served to stabilize the cutaneous environment during Bd infections, thereby ameliorating the Bd-mediated microbiome changes. While this work explored the changes in established antifungal flora, we anticipate the mast cell-mediated inhibition of Bd may be due to additional, yet unidentified bacterial or fungal taxa. Intriguingly, while mammalian skin mast cell functionality depends on microbiome elicited SCF production by keratinocytes17, our results indicate that frog skin mast cells in turn impact skin microbiome structure and likely their function. It will be interesting to further explore the interdependent nature of amphibian skin microbiomes and resident mast cells.

5. The time course of rSCF or rCSF3 treatments relative to Bd infection in the experiments is not clear. Were the treatments given 12 hours prior to the final analysis point to maximize the effect? For example, in Figure 3E, were rSCF injections given at 6.5 dpi and 20.5 dpi? Or were treatments administered on day 0 of the infection model? If the latter, how do the authors explain the effects at 7 dpi or 21 dpi given mast cell and neutrophil numbers return to baseline within 24 hours after rSCF or rCSF3 treatment, respectively?

Please find the schematic of the immune manipulation, Bd infection, and sample collection times below. We will include a figure like this in our revised manuscript.

The title of the manuscript may be mildly overstated. Although Bd infection can indeed be deadly, mortality was not a readout in this study, and it is not clear from the data reported that expanding skin mast cells would ultimately prevent progression to death in Bd infections.

We acknowledge this point. The revised manuscript will be titled: “Amphibian mast cells: barriers to chytrid fungus infections”.

Reviewer #3 (Public Review):

Summary:<br /> Hauser et al. provide an exceptional study describing the role of resident mast cells in amphibian epidermis that produce anti-inflammatory cytokines that prevent Batrachochytrium dendrobatidis (Bd) infection from causing harmful inflammation, and also protect frogs from changes in skin microbiomes and loss of mucin in glands and loss of mucus integrity that otherwise cause changes to their skin microbiomes. Neutrophils, in contrast, were not protective against Bd infection. Beyond the beautiful cytology and transcriptional profiling, the authors utilized elegant cell enrichment experiments to enrich mast cells by recombinant stem cell factor, or to enrich neutrophils by recombinant colony-stimulating factor-3, and examined respective infection outcomes in Xenopus.

Strengths:<br /> Through the use of recombinant IL4, the authors were able to test and eliminate the hypothesis that mast cell production of IL4 was the mechanism of host protection from Bd infection. Instead, impacts on the mucus glands and interaction with the skin microbiome are implicated as the protective mechanism. These results will press disease ecologists to examine the relative importance of this immune defense among species, the influence of mast cells on the skin microbiome and mucosal function, and open the potential for modulating mucosal defense.

We thank the reviewer for recognizing the significance and utility of the findings presented in our manuscript.

Weaknesses:<br /> A reduction of bacterial diversity upon infection, as described at the end of the results section, may not always be an "adverse effect," particularly given that anti-Bd function of the microbiome increased. Some authors (see Letourneau et al. 2022 ISME, or Woodhams et al. 2023 DCI) consider these short-term alterations as encoding ecological memory, such that continued exposure to a pathogen would encounter an enriched microbial defense. Regardless, mast cell-initiated protection of the mucus layer may negate the need for this microbial memory defense.

We thank the reviewer their insightful comment. We will revise our discussion to include this possible interpretation.

While the description of the mast cell location in the epidermal skin layer in amphibians is novel, it is not known how representative these results are across species ranging in chytridiomycosis susceptibility. No management applications are provided such as methods to increase this defense without the use of recombinant stem cell factor, and more discussion is needed on how the mast cell component (abundance, distribution in the skin) of the epidermis develops or is regulated.

We appreciate the reviewer’s comment and would like to point out that the work presented in our manuscript was driven by comparative immunology questions more than by conservation biology.

We thank the reviewer for suggesting expanding our discussion to include potential management applications and potential mechanisms for regulating frog skin mast cells. While any content to these effects would be highly speculative, we agree that it may spark new interest and pave new avenues for research. To this end, our revised manuscript will include a paragraph to this effect.

References:

1.         Flajnik, M.F. A cold-blooded view of adaptive immunity. Nat Rev Immunol 18, 438-453 (2018).

2.         Mulero, I., Sepulcre, M.P., Meseguer, J., Garcia-Ayala, A. & Mulero, V. Histamine is stored in mast cells of most evolutionarily advanced fish and regulates the fish inflammatory response. Proc Natl Acad Sci U S A 104, 19434-19439 (2007).

3.         Reite, O.B. A phylogenetical approach to the functional significance of tissue mast cell histamine. Nature 206, 1334-1336 (1965).

4.         Reite, O.B. Comparative physiology of histamine. Physiol Rev 52, 778-819 (1972).

5.         Takaya, K., Fujita, T. & Endo, K. Mast cells free of histamine in Rana catasbiana. Nature 215, 776-777 (1967).

6.         Galli, S.J. New insights into "the riddle of the mast cells": microenvironmental regulation of mast cell development and phenotypic heterogeneity. Lab Invest 62, 5-33 (1990).

7.         Babina, M., Guhl, S., Artuc, M. & Zuberbier, T. IL-4 and human skin mast cells revisited: reinforcement of a pro-allergic phenotype upon prolonged exposure. Archives of dermatological research 308, 665-670 (2016).

8.         Lai, H. & Rogers, D.F. New pharmacotherapy for airway mucus hypersecretion in asthma and COPD: targeting intracellular signaling pathways. J Aerosol Med Pulm Drug Deliv 23, 219-231 (2010).

9.         Rankin, J.A. et al. Phenotypic and physiologic characterization of transgenic mice expressing interleukin 4 in the lung: lymphocytic and eosinophilic inflammation without airway hyperreactivity. Proc Natl Acad Sci U S A 93, 7821-7825 (1996).

10.       Church, M.K. Allergy, Histamine and Antihistamines. Handb Exp Pharmacol 241, 321-331 (2017).

11.       Nakamura, T. The roles of lipid mediators in type I hypersensitivity. J Pharmacol Sci 147, 126-131 (2021).

12.       Aponte-Lopez, A., Enciso, J., Munoz-Cruz, S. & Fuentes-Panana, E.M. An In Vitro Model of Mast Cell Recruitment and Activation by Breast Cancer Cells Supports Anti-Tumoral Responses. Int J Mol Sci 21 (2020).

13.       Jamur, M.C. et al. Mast cell repopulation of the peritoneal cavity: contribution of mast cell progenitors versus bone marrow derived committed mast cell precursors. BMC Immunol 11, 32 (2010).

14.       Walke, J.B. & Belden, L.K. Harnessing the Microbiome to Prevent Fungal Infections: Lessons from Amphibians. PLoS Pathog 12, e1005796 (2016).

15.       Jani, A.J. et al. The amphibian microbiome exhibits poor resilience following pathogen-induced disturbance. ISME J 15, 1628-1640 (2021).

16.       Muletz-Wolz, C.R., Fleischer, R.C. & Lips, K.R. Fungal disease and temperature alter skin microbiome structure in an experimental salamander system. Mol Ecol 28, 2917-2931 (2019).

17.       Wang, Z. et al. Skin microbiome promotes mast cell maturation by triggering stem cell factor production in keratinocytes. J Allergy Clin Immunol 139, 1205-1216 e1206 (2017).

Author Response

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

We thank the two reviewers very much for their careful review and valuable comments. Upon these comments, the following revisions have been made. First, we have performed a new analysis on human accelerated regions (HARs) recently reported by the Zoonomia Project. Second, we have presented more data on experimentally detected and computationally predicted DBSs of MALAT1, NEAT1, and MEG3. Third, we have added details on the RNA-seq data processing and subsequent differential expression testing to the Materials and Methods section. Fourth, we have clarified some details on the human ancestor sequence and the use of parameters and thresholds. Six new citations are added. In addition, we have also carefully polished the main text. We hope these revisions, together with the Responses-to-Reviewers, would help the reader better get the information from the paper.

eLife assessment

In this valuable manuscript, the authors attempt to examine the role of long non-coding RNAs (lncRNAs) in human evolution, through a set of population genetics and functional genomics analyses that leverage existing datasets and tools. Although the methods are at times inadequate - for example, suitable methods and/or relevant controls are lacking at many points, and selection is inferred sometimes too quickly - the results nonetheless point towards a possible contribution of long non-coding RNAs to the evolution of human biology and they suggest clear directions for future, more rigorous study.

Public Reviews:

Reviewer #1 (Public Review):

Summary

While DNA sequence divergence, differential expression, and differential methylation analysis have been conducted between humans and the great apes to study changes that "make us human", the role of lncRNAs and their impact on the human genome and biology has not been fully explored. In this study, the authors computationally predict HSlncRNAs as well as their DNA Binding sites using a method they have developed previously and then examine these predicted regions with different types of enrichment analyses. Broadly, the analysis is straightforward and after identifying these regions/HSlncRNAs the authors examined their effects using different external datasets.

Strengths/weaknesses

By and large, the analysis performed is dependent on their ability to identify HSlncRNAs and their DBS. I think that they have done a good job of showing the performance metrics of their methods in previous publications. Thereafter, they perform a series of enrichment-type analyses that have been used in the field for quite a while now to look at tissue-specific enrichment, or region-specific enrichment, or functional enrichment, and I think these have been carried out well. The authors achieved the aims of their work. I think one of the biggest contributions that this paper brings to the field is their annotation of these HSlncRNAs. Thus a major revisionary effort could be spent on applying their method to the latest genomes that have been released so that the community could get a clean annotation of newly identified HSlncRNAs (see comment 2).

Comments

1. Though some of their results about certain HSlncRNAs having DBSs in all genes is rather surprising/suspicious, I think that broadly their process to identify and validate DBSs is robust, they have multiple lines of checks to identify such regions, including functional validation. These predictions are bound to have some level of false positive/negative rate and it might be nice to restate those here and on what experiment/validation data these were conducted. However, the rest of their analysis comprises different types of enrichment analysis which shouldn't be affected by outlier HSlncRNAs if indeed their FPR/FNR are low.

2. There are now several new genomes available as part of the Zoonomia consortium and 240 Primate consortium papers released. These papers have re-examined some annotations such as Human Accelerated Regions (HARs) and found with a larger dataset as well as better reference genomes, that a large fraction of HARs were actually incorrectly annotated - that is that they were also seen in other lineages outside of just the great apes. If these papers have not already examined HSlncRNAs, the authors should try and re-run the computational predictions with this updated set and then identify HSlncRNAs there. This might help to clarify their signal and remove lncRNAs that might be present in other primates but are somehow missing in the great apes. This might also help to mitigate some results that they see in section 3 of their paper in comparing DBS distances between archaics and humans.

Responses:

(1) Thanks for the good suggestion. We have checked the Zoonomia reported genomes and found that new primate genomes are monkeys and lemurs but not apes (Zoonomia Consortium. Nature 2023. https://doi.org/10.1038/s41586-020-2876-6), and the phylogenetic relationships between monkeys and humans are much more remote than those between apes and humans. In addition, the Zoonomia project did target identifying new lncRNA genes.

(2) We have examined the Zoonomia-reported HARs (Keough et al. Science 2023. DOI: 10.1126/science.abm1696). Of the 312 HARs reported by Keough et al, 8 overlap 26 DBSs of 14 HS lncRNAs; moreover, DBSs greatly outnumber HARs, suggesting that HAR and DBS are different sequences with different functions.

(3) In the revised manuscript, a new paragraph (the second one) has been added to the section “HS lncRNAs regulate diverse genes and transcripts” to describe the HAR analysis result.

1. The differences between the archaic hominins in their DBS distances to modern humans are a bit concerning. At some level, we expect these to be roughly similar when examining African modern humans and perhaps the Denisovan being larger when examining Europeans and Asians, but they seem to have distances that aren't expected given the demography. In addition, from their text for section 3, they begin by stating that they are computing two types of distances but then I lost track of which distance they were discussing in paragraph 3 of section 3. Explicitly stating which of the two distances in the text would be helpful for the reader.

Responses:

(1) Upon the archaic human genomes, the genomic distances from the three modern humans are shorter to Denisovan than to Altai Neanderthal; however, upon the related studies we cite, the phylogenetic relationship between the three modern humans is more remote to Denisovan than to Altai Neanderthal. Thus, the finding that 2514 and 1256 DBSs have distances >0.034 in Denisovans and Altai Neanderthals is not unreasonable. The numbers of DBSs, of course, depend on the cutoff of 0.034, which is somewhat subjective but not unreasonable.

(2) The second paragraph is added to the Discussion, discussing parameters and cutoffs.

(3) Regarding the two types of distance, the distances computed in the first way were not further analyzed because, as we note, “This anomaly may be caused by that the human ancestor was built using six primates without archaic humans”.

1. Isn't the correct control to examine whether eQTLs are more enriched in HSlncRNA DBSs a set of transcription factor binding sites? I don't think using just promoter regions is a reasonable control here. This does not take away from the broader point however that eQTLs are found in DBSs and I think they can perform this alternate test.

Responses:

Indeed, TFBSs are more comparable to DBSs than promoters. However, many more methods have been developed to predict TFBSs than to predict DBSs, making us concerned about TFBS prediction's reliability. Since most QTLs in DBSs are mQTLs (Supplementary Table 13), but many QTLs in TFBSs are eQTLs (Flynn et al. PLoS Genetics 2021. DOI: 10.1371/journal.pgen.1009719), it is pretty safe to conclude that DBSs are enriched in mQTLs.

1. In the Discussion, they highlight the evolution of sugar intake, which I'm not sure is appropriate. This comes not from GO enrichment but rather from a few genes that are found at the tail of their distribution. While these signals may be real, the evolution of traits is often highly polygenic and they don't see this signal in their functional enrichment. I suggest removing that line. Moreover, HSlncRNAs are ones that are unique across a much longer time frame than the transition to agriculture which is when sugar intake rose greatly. Thus, it's unlikely to see enrichment for something that arose in the past 6000-7000 years would in the annotation that is designed to detect human-chimp or human-neanderthal level divergence.

Responses:

(1) The Discussion on human adaptation to high sugar intake is based on both enriched GO terms (Supplementary Table 4, 7) and a set of genes in modern humans with the most SNP-rich DBSs (Table 2). These glucose-related GO terms are not at the tail of the list because, of the 614 enriched GO terms (enriched in genes with strongest DBSs), glucose metabolism-related ones are ranked 208, 212, 246, 264, 504, 522, 591, and of the 409 enriched GO terms (enriched in the top 1256 genes in Altai Neanderthals), glucose metabolism-related ones are ranked 152 and 217.

(2) Indeed, there are other top-ranked enriched GO terms; some (e.g., neuron projection development (GO:0031175) and cell projection morphogenesis (GO:0048858)) have known impact on human evolution, but the impact of others (e.g., cell junction organization (GO:0034330)) remain unclear. We specifically report human adaptation to high sugar intake because the DBSs in related genes show differences in modern humans (Table 2).

Reviewer #2 (Public Review):

Lin et al attempt to examine the role of lncRNAs in human evolution in this manuscript. They apply a suite of population genetics and functional genomics analyses that leverage existing data sets and public tools, some of which were previously built by the authors, who clearly have experience with lncRNA binding prediction. However, I worry that there is a lack of suitable methods and/or relevant controls at many points and that the interpretation is too quick to infer selection. While I don't doubt that lnc RNAs contribute to the evolution of modern humans, and certainly agree that this is a question worth asking, I think this paper would benefit from a more rigorous approach to tackling it.

At this point, my suggestions are mostly focused on tightening and strengthening the methods; it is hard for me to predict the consequence of these changes on the results or their interpretation, but as a general rule I also encourage the authors to not over-interpret their conclusions in terms of what phenotype was selected for when as they do at certain points (eg glucose metabolism).

Responses:

(1) Now, we use more cautious wording to describe the results.

(2) A paragraph (the second one) is added to Discussion to explain parameters and cutoffs.

(3) We make the caution at the end of the third paragraph that “We note that these are findings instead of conclusions, and they indicate, suggest, or support something revealing the primary question of what genomic differences critically determine the phenotypic differences between humans and apes and between modern and archaic humans”.

I note some specific points that I think would benefit from more rigorous approaches, and suggest possible ways forward for these.

1. Much of this work is focused on comparing DNA binding domains in human-unique long-noncoding RNAs and DNA binding sites across the promoters of genes in the human genome, and I think the authors can afford to be a bit more methodical/selective in their processing and filtering steps here. The article begins by searching for orthologues of human lncRNAs to arrive at a set of 66 human-specific lncRNAs, which are then characterised further through the rest of the manuscript. Line 99 describes a binding affinity metric used to separate strong DBS from weak DBS; the methods (line 432) describe this as being the product of the DBS or lncRNA length times the average Identity of the underlying TTSs. This multiplication, in fact, undoes the standardising value of averaging and introduces a clear relationship between the length of a region being tested and its overall score, which in turn is likely to bias all downstream inference, since a long lncRNA with poor average affinity can end up with a higher score than a short one with higher average affinity, and it's not quite clear to me what the biological interpretation of that should be. Why was this metric defined in this way?

Responses:

(1) Binding affinity and length of all DBSs of HS lncRNAs are given in Supplementary Table 2 and 3. Since a triplex (say, 100 bp in length) may have 50% or 70% of nucleotides bound, it is necessary to differentiate binding affinity and length, and the two measures can differentiate DBSs of the same length but with different binding affinity and DBSs with the same binding affinity but different length.

(2) Differentiating DBSs into strong and weak ones is somewhat subjective, accurately differentiating them demands experimental data that are currently unavailable, and it is advisable to separately analyze strong and weak DBSs because they may likely influence different aspects of human evolution.

1. There is also a strong assumption that identified sites will always be bound (line 100), which I disagree is well-supported by additional evidence (lines 109-125). The authors show that predicted NEAT1 and MALAT1 DBS overlap experimentally validated sites for NEAT1, MALAT1, and MEG3, but this is not done systematically, or genome-wide, so it's hard to know if the examples shown are representative, or a best-case scenario.

Responses:

(1) We do not assume/think that identified sites will always be bound. Instead, lncRNA/DBS binding is highly context-dependent (including tissue-specific).

(2) An extra supplementary table (Supplementary Table 15) is added to show what predicted DBSs overlap experimentally detected DBSs for NEAT1, MALAT1, and MEG3. By the way, it is more accurate to say “experimentally detected” than “experimentally validated”, because experimental data have true/false positives and true/false negatives, and different sequencing protocols (for detecting lncRNA/DNA binding) may generate somewhat different results.

It's also not quite clear how overlapping promoters or TSS are treated - are these collapsed into a single instance when calculating genome-wide significance? If, eg, a gene has five isoforms, and these differ in the 3' UTR but their promoter region contains a DBS, is this counted five times, or one? Since the interaction between the lncRNA and the DBS happens at the DNA level, it seems like not correcting for this uneven distribution of transcripts is likely to skew results, especially when testing against genome-wide distributions, eg in the results presented in sections 5 and 6. I do not think that comparing genes and transcripts putatively bound by the 40 HS lncRNAs to a random draw of 10,000 lncRNA/gene pairs drawn from the remaining ~13500 lncRNAs that are not HS is a fair comparison. Rather, it would be better to do many draws of 40 non-HS lncRNAs and determine an empirical null distribution that way, if possible actively controlling for the overall number of transcripts (also see the following point).

Responses:

(1) We analyzed each and every GENCODE-annotated transcript (Supplementary Table 2). For example, if a gene has N TSS and N transcripts, DBSs are predicted in N promoter regions. When analyzing gene expression in tissues, each and every transcript is analyzed.

(2) Ideally, it would be better to do many draws, but statistically, a huge number is needed due to the number of total genes in the human genome.

(3) We feel that doing many draws of 40 non-HS lncRNAs and determining an empirical null distribution is not as straightforward as comparing HS lncRNA-target transcript pairs (45% show significant expression correlation) with random lncRNA-random transcript pairs (2.3% show significant expression correlation).

1. Thresholds for statistical testing are not consistent, or always well justified. For instance, in line 142 GO testing is performed on the top 2000 genes (according to different rankings), but there's no description of the background regions used as controls anywhere, or of why 2000 genes were chosen as a good number to test? Why not 1000, or 500? Are the results overall robust to these (and other) thresholds? Then line 190 the threshold for downstream testing is now the top 20% of genes, etc. I am not opposed to different thresholds in principle, but they should be justified.

Responses:

(1) The over-representation analysis using g:Profiler was applied to the top and bottom 2000 genes with the whole genome as the background. The number “2000” was chosen somewhat subjectively. If more or fewer genes were chosen, more or fewer enriched GO terms would be identified, but GO terms with adjusted P-values <0.05 would be quite stable.

(2) A paragraph (the second one) is added to the Discussion to explain parameters and cutoffs.

Likewise, comparing Tajima's D values near promoters to genome-wide values is unfair, because promoters are known to be under strong evolutionary constraints relative to background regions; as such it is not surprising that the results of this comparison are significant. A fairer comparison would attempt to better match controls (eg to promoters without HS lncRNA DBS, which I realise may be nearly impossible), or generate empirical p-values via permutation or simulation.

Responses:

We examined Tajima’s D in DBSs (Supplementary Figure 9) and in HS lncRNA genes (Supplementary Figure 18). We compared the Tajima’s D values with the genome-wide background in both cases.

1. There are huge differences in the comparisons between the Vindija and Altai Neanderthal genomes that to me suggest some sort of technical bias or the such is at play here. e.g. line 190 reports 1256 genes to have a high distance between the Altai Neanderthal and modern humans, but only 134 Vindija genes reach the same cutoff of 0.034. The temporal separation between the two specimens does not seem sufficient to explain this difference, nor the difference between the Altai Denisovan and Neanderthal results (2514 genes for Denisovan), which makes me wonder if it is a technical artefact relating to the quality of the genome builds? It would be worth checking.

Responses:

(1) The cutoff of 0.034 was chosen upon that DBSs in the top 20% (4248) genes in chimpanzees have distances larger than this cutoff, and accordingly, 4248, 1256, 2514, and 134 genes have DBSs distances >0.034 in chimpanzees, Altai Neanderthals, Denisovans, and Vindija Neanderthals. These numbers of genes qualitatively agree with the phylogenetic distances from chimpanzees, archaic humans to modern humans. If a percentage larger or smaller than 20% (e.g., 10% or 30%) is chosen, and so is a cutoff X, the numbers of genes with DBSs distance >X would not be 4248, 1256, 2514, and 134, but could still qualitatively agree with the phylogenetic distances from chimpanzees, archaic humans to modern humans.

(2) The second paragraph in the Discussion now explains the parameters and cutoffs.

1. Inferring evolution: There are some points of the manuscript where the authors are quick to infer positive selection. I would caution that GTEx contains a lot of different brain tissues, thus finding a brain eQTL is a lot easier than finding a liver eQTL, just because there are more opportunities for it. Likewise, claims in the text and in Tables 1 and 2 about the evolutionary pressures underlying specific genes should be more carefully stated. The same is true when the authors observe high Fst between groups (line 515), which is only one possible cause of high Fst - population differentiation and drift are just as capable of giving rise to it, especially at small sample sizes.

Responses:

(1) We analyzed brain tissues separately instead of taking the whole brain as a tissue, see Supplementary Table 12 and Figure 3.

(2) We make the caution at the end of the third paragraph that “We note that these are findings instead of conclusions, and they indicate, suggest, or support something revealing the primary question of what genomic differences critically determine the phenotypic differences between humans and apes and between modern and archaic humans”.

Reviewer #1 (Recommendations For The Authors):

Some figures are impossible to see/read so I wasn't able to evaluate them - Fig, 1B, 1E, 1F are small and blurry.

Responses:

High-quality figures are provided.

Typo in line 178: in these archaic humans, the distances of HS lncRNAs are smaller than the distances of DBSs.

Responses:

This is not a typo. We use “distance per base” to measure whether HS lncRNAs or their DBSs have evolved more from archaic humans to modern humans. See also Supplementary Note 4 and 5.

Reviewer #2 (Recommendations For The Authors):

1. There's some inconsistency in the genome builds and the database versions used, eg, sometimes panTro4 is used and sometimes panTro5 (line 456). Likewise, the version of GENCODE used is very old (18), the current version is 43. The current version contains 19928 lncRNAs, which is a big difference relative to what is being tested!

Responses:

(1) panTro4 was used to search orthologues of human lncRNAs; this time-consuming work started several years ago when the version of GENCODE was V18 (see Lin et al., 2019).

(2) Regarding “the version of GENCODE used is very old (V18)”, we have later examined the 4396 human lncRNAs reported in GENCODE V36 and found that the set of 66 HS lncRNAs remains the same.

(3) The counterparts of HS lncRNAs’ DBSs in chimpanzees were predicted recently using panTro5.

1. Table 1: What does 'mostly' mean in this context? I understand that it refers to sequence differences between humans and the other genomes, but what is the actual threshold, and how is it defined?

Responses:

The title of Table 1 is “Genes with strongest DBSs and mostly changed sequence distances from modern humans to archaic humans and chimpanzees”. Instead of using two cutoffs, choosing genes with the two features seems easy and sensible.

1. Line 117: The methods do not include information on the RNA-seq data processing and subsequent DE testing.

Responses:

The details are added to the section “Experimentally validating DBS predictiom” (The reads were aligned to the human GRCh38 genome using Hiasat2 (Kim et al., 2019), and the resulting sam files were converted to bam files using Samtools (Li et al., 2009). Stringtie was used to quantify gene expression level (Pertea et al., 2015). Fold change of gene expression was computed using the edgeR package (Robinson et al., 2010), and significant up- and down-regulation of target genes after DBD knockout was determined upon |log2(fold change)| > 1 with FDR < 0.1).

1. Line 180: I looked at the EPO alignment and it's not clear to me what 'human ancestor' means, but it may well explain the issues the authors have with calculating distances (I agree those numbers are weird). Is it the reconstructed ancestral state of humans at around 300-200,000 years ago (coalescence of most human uniparental lineages), or the inferred sequence of the human-chimpanzee most recent common ancestor? If it's the former, it's not surprising it skews results towards shorter distances for modern humans, since the tree distance from that point to archaic hominins is significantly larger than to modern humans.

Responses:

The “human ancestor” is constructed by the EBI team upon the genomes of six primates in the Ensembl website. We find that the reconstructed ancestral state of humans may be unlikely around 300,000-200,000 years, and may be much earlier. We also find that many DNA sequences of the “human ancestor” are low-confidence calls (i.e., the ancestral states are supported by only one primate’s sequence).

1. Line 221: SNP-rich DBS: Is this claim controlled for the length of the DBS?

Responses:

No. Long DBSs tend to have more SNPs. When comparing the same DBS in modern humans, archaic humans, and chimpanzees, both the length and SNP number reflect evolution, so it is not necessary to control for the length.

1. Given that GTEx is primarily built off short-read data and it is impossible to link binding of a lncRNA to a DBS with its impact with a specific transcript

Responses:

As written in the section “Examining the tissue-specific impact of HS lncRNA-regulated gene expression”, we calculated the pairwise Spearman's correlation coefficient between the expression of an HS lncRNA (the representative transcript, median TPM value > 0.1) and the expression of each of its target transcripts (median TPM value > 0.1) using the scipy.stats.spearmanr program in the scipy package. The expression of an HS lncRNA gene and a target transcript was considered to be significantly correlated if the |Spearman's rho| > 0.3, with Benjamini-Hochberg FDR < 0.05.

1. Line 429: should TTO be TFO?

Responses:

Here TTO should be TFO; the typo is corrected.

1. Methods, section 7: Some of the text in this section should perhaps be moved to the results section?

Responses:

Each of the two paragraphs in Methods’ section 7 is quite large, and some contents in Supplementary Notes are also very relevant. Thus, moving them to the Results section could make the Results too lengthy and specific.

1. Line 587: GTEx is built from samples of primarily European ancestry and has poor representation of African ancestry and negligible representation of Asian ancestry (see the GTEx v8 paper supplement). This means that it is basically impossible to find a non-European population-specific eQTL in GTEx, which in turn impacts these results.

Responses:

(1) Indeed, this is a serious issue of data analysis, and this issue cannot be solved until more Africans are sequenced.

(2) Anyway, one can still find considerable African-specific eQTLs in GTEx, such as rs28540058 (with frequency of 0, 0, 0.13 in CEU, CHB, YRI) and rs58772997 (with frequency of 0, 0, 0.12 in CEU, CHB, YRI (see Supplementary Table12 and Supplementary Figure 22).

Reviewer #1 (Public Review):

Summary<br /> While DNA sequence divergence, differential expression, and differential methylation analysis have been conducted between humans and the great apes to study changes that "make us human", the role of lncRNAs and their impact on the human genome and biology has not been fully explored. In this study, the authors computationally predict HSlncRNAs as well as their DNA Binding sites using a method they have developed previously and then examine these predicted regions with different types of enrichment analyses. Broadly, the analysis is straightforward and after identifying these regions/HSlncRNAs the authors examined their effects using different external datasets.

I no longer have any concerns about the manuscript as the authors have addressed my comments in the first round of review.

Reviewer #2 (Public Review):

Lin et al attempt to examine the role of lncRNAs in human evolution in this manuscript. They apply a suite of population genetics and functional genomics analyses that leverage existing data sets and public tools, some of which were previously built by the authors, who clearly have experience with lncRNA binding prediction. However, I worry that there is a lack of suitable methods and/or relevant controls at many points and that the interpretation is too quick to infer selection. While I don't doubt that lnc RNAs contribute to the evolution of modern humans, and certainly agree that this is a question worth asking, I think this paper would benefit from a more rigorous approach to tackling it.

I thank the authors for their revisions to the manuscript; however, I find that the bulk of my comments have not been addressed to my satisfaction. As such, I am afraid I cannot say much more than what I said last time, emphasising some of my concerns with regards to the robustness of some of the analyses presented. I appreciate the new data generated to address some questions, but think it could be better incorporated into the text - not in the discussion, but in the results.

Author Response

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

eLife assessment

The finding that Fusicoccin (FC-A) promotes locomotor recovery after spinal cord injury is useful, and the idea of harnessing small molecules that may affect protein-protein interactions to promote axon regeneration is interesting and worthy of study. However, the main methods, data, and analyses are inadequate to support the primary claim of the manuscript that a 14-3-3-Spastin complex is necessary for the observed FC-A effects.

Response: We appreciate the eLife editorial and review team for consideration and evaluation of our manuscript. In light of the feedback from the editors and reviewers, we recognize that certain aspects of the title and key conclusions require further refinement. We have shown that 14-3-3, through its interaction with phosphorylated spastin, inhibits the degradation of spastin. Also, we have demonstrated that 14-3-3 can enhance spastin's microtubule-severing ability in cell lines. Furthermore, our work has illustrated the significant roles of 14-3-3 and spastin in the repair process of spinal cord injury. However, there is currently insufficient direct evidence to confirm the cooperation between 14-3-3 and spastin during axon regeneration and the recovery of spinal cord injury. Moreover, we have not provided conclusive evidence of their simultaneous action in injured axons, mediating changes in microtubule dynamics. Consequently, we have re-evaluated the manuscript's title and primary conclusions, and have made relevant modifications. For more detailed information, please refer to the reviewer's comments.

Public Reviews:

Reviewer #1 (Public Review):

The present work establishes 14-3-3 proteins as binding partners of spastin and suggests that this binding is positively regulated by phosphorylation of spastin. The authors show evidence that 14-3-3 - spastin binding prevents spastin ubiquitination and final proteasomal degradation, thus increasing the availability of spastin. The authors measured microtubule severing activity in cell lines and axon regeneration and outgrowth as a prompt to spastin activity. By using drugs and peptides that separately inhibit 14-3-3 binding or spastin activity, they show that both proteins are necessary for axon regeneration in cell culture and in vivo models in rats.

The following is an account of the major strengths and weaknesses of the methods and results.

Major strengths

-The authors performed pulldown assays on spinal cord lysates using GST-spastin, then analyzed pulldowns via mass spectrometry and found 3 peptides common to various forms of 14-3-3 proteins. In co-expression experiments in cell lines, recombinant spastin co-precipitated with all 6 forms of 14-3-3 tested. The authors could also co-immunoprecipitate spastin-14-3-3 complexes from spinal cord samples and from primary neuronal cultures.

-By protein truncation experiments they found that the Microtubule Binding Domain of spastin contained the binding capability to 14-3-3. This domain contained a putative phosphorylation site, and substitutions that cannot be phosphorylated cannot bind to 14-3-3.

-Overexpression of GFP-spastin shows a turn-over of about 12 hours when protein synthesis is inhibited by cycloheximide. When 14-3-3 is co-overexpressed, GFP-spastin does not show a decrease by 12 hours. When S233A is expressed, a turn-over of 9 hours is observed, suggesting that phosphorylation increases the stability of the protein. In support of that notion, the phospho-mimetic S233D makes it more stable, lasting as much as the over-expression of 14-3-3.

-By combining FCA with Spastazoline, authors claim that FCA increased regeneration is due to increased spastin activity in various models of neurite outgrowth and regeneration in cell culture and in vivo, the authors show impressive results on the positive effect of FCA in regeneration, and that this is abolished when spastin is inhibited.

Major weaknesses

1. The present manuscript suggests that 14-3-3 and spastin work in the same pathway to promote regeneration. Although the manuscript contains valuable evidence in support for a role of 14-3-3 and spasting in regeneration, the conclusive evidence is difficult to generate, and is missing in the present manuscript. For example, there are simpler explanations for the combined effect of FC-A and spastazoline. The FC-A mechanism of action can be very broad, since it will increase the binding of all 14-3-3 proteins with presumably all their substrates, hence the pathways affected can rise to the hundreds. The fact that spastazoline abolishes FC-A effect, may not be because of their direct interaction, but because spastin is a necessary component of the execution of the regeneration machinery further downstream, in line with the fact that spastazoline alone prevented outgrowth and regeneration, and in agreement with previous work showing that normal spastin activity is necessary for regeneration.

With this in mind, I consider the title and most major conclusions of the manuscript related to these two proteins acting together for the observed effects are overstated.

Response: We appreciate and acknowledge the reviewers' considerations. Our results demonstrated that the spastin inhibitor, spastazolin, almost completely inhibited axon regeneration and the spinal cord injury repair process. This, in turn, leads to the disappearance of any promoting effect on spinal cord injury repair when spastin function is compromised. While we have provided evidence that the expression levels of spastin are moderately increased at the injury site in mice after treatment with FC-A following spinal cord injury, the conclusion that FC-A promotes spinal cord injury repair through the direct interaction between 14-3-3 and spastin still lacks direct evidence. Therefore, we have made appropriate modifications to the manuscript's title and main conclusions.

1. Authors show that S233D increases MT severing activity, and explain that it is related to increased binding to 14-3-3. An alternative explanation is that phosphorylation at S233 by itself could increase MT severing activity. The authors could test if purified spastin S233D alone could have more potent enzymatic activity.

Response: We appreciate the considerations of the reviewer. We believe that supplementing in vitro experiments to assess whether S233D affects spastin's microtubule severing function can more intuitively demonstrate whether phosphorylation of spastin at S233 affects its microtubule severing function; however, spastin forms hexamers through its AAA domain to exert ATPase activity and cut microtubules. Current research has reported that mutation sites leading to changes in microtubule severing function are mainly located within spastin's AAA domain (affecting spastin's ATPase activity, amino acids 342-599), such as E356A, G370R, N386K, K388R, E442Q, K427R and R562Q. Furthermore, studies have shown that mutating 11 phosphorylation sites in spastin's MIT and MTBD regions to alanine does not affect spastin's microtubule severing function, including human S268 (Rat Ser233) (Phosphorylation mutation impairs the promoting effect of spastin on neurite outgrowth without affecting its microtubule severing ability. Eur J Histochem. doi: 10.4081/ejh.2023.3594). Additionally, we also provided supplementary experiments in cell lines which showed that both spastin S233A and S233D could effectively sever microtubules (Fig.S2).

1. The interpretation of the authors cannot explain how Spastin can engage in MT severing while bound to 14-3-3 using its Microtubule Binding Domain.

Response: We appreciate the considerations of the expert reviewer. The IP experiments with truncated fragments suggest that the binding region of 14-3-3 with spastin is located within the region (215-336 amino acids) in spastin. Furthermore, experiments involving site-directed mutagenesis confirm that the actual binding site of 14-3-3 with spastin is the S233 site, rather than its MTBD region (270-328). Therefore, we have made corrections in the manuscript. We also indicate that 14-3-3 enhances spastin's protein levels by binding to the S233 site, which may be due to 14-3-3 masking the ubiquitination sites near spastin S233 (K206 or K254). Our further experiments also demonstrate that 14-3-3 inhibits the ubiquitination degradation pathway of phosphorylated spastin.

1. Also, the term "microtubule dynamics", which is present in the title and in other major conclusions, is overstated. Although authors show, in cell lines, changes in microtubule content, it is far from evidence for changes in "MT dynamics" in the settings of interest (i.e. injured axons).

Response: We appreciate and acknowledge the rigorous feedback. While our manuscript demonstrated the regulatory role of 14-3-3 and spastin in microtubule dynamics in cell lines, we lack direct evidence of these changes in microtubule dynamics within injured axons. Therefore, we have made appropriate modifications to the title, main conclusions, and related statements in our manuscript.

1. In the same lines, the manuscript lacks evidence for the changes of MT content and/dynamics as a function of the proposed 14-3-3 - Spastin pathway.

Response: We appreciate and concur with the opinions of the expert reviewer. The observed changes in microtubule dynamics in spinal cord injury were related to the overall alterations in microtubule dynamics within the spinal cord injury site. We still lack direct evidence that 14-3-3, in conjunction with spastin, alters the microtubule dynamics within axons during the process of regeneration. Therefore, we have made modifications to the manuscript.

Reviewer #2 (Public Review):

Summary:

The idea of harnessing small molecules that may affect protein-protein interactions to promote axon regeneration is interesting and worthy of study. In this manuscript Liu et al. explore a 14-3-3-Spastin complex and its role in axon regeneration.

Strengths:

Some of the effects of FC-A on locomotor recovery after spinal cord contusion look interesting

Weaknesses:

The manuscript falls short of establishing that a 14-3-3-Spastin complex is important for any FC-A-dependent effects and there are several issues with data quality that make it difficult to interpret the results. Importantly, the effects of the spastin inhibitor has a major impact on neurite outgrowth suggesting that cells simply cannot grow in the presence of the inhibitor and raising serious questions about any selectivity for FC-A - dependent growth. Aspects of the histology following spinal cord injury were not convincing.

Response: We appreciate the rigorous review by the expert reviewers. In response to the feedback from reviewer 1, we lack direct evidence to demonstrate that the reparative effect of FC-A on spinal cord injury is mediated by the combined action of 14-3-3 and spastin. We have accordingly made the necessary changes to our manuscript. Additionally, due to upload limitations, the resolution of our tissue slices related to spinal cord injury in the manuscript is relatively low. To address this, we have supplemented relevant images which was enlarged in the supplementary materials (Fig. S7-9), Also, the original confocal files and images were uploaded.

Furthermore, our manuscript does not suggest that the reparative effect of FC-A in spinal cord injury selectively impacts the interaction between 14-3-3 and spastin. Therefore, we have modified our claims (title and conclusions) to ensure a more precise statement. Despite the fact that our axonal markers do not fully align, our evidence still strongly supports the role of FC-A in promoting nerve regeneration after spinal cord injury. Additionally, we will further optimize our immunohistochemistry methods.

Reviewer #3 (Public Review):

Summary:

The current manuscript shows that 14-3-3 are binding partners of spastin, preventing its degradation. It is additionally shown, using complementary methods, that both 14-3-3 and spastin are necessary for axon regeneration in vitro and in vivo. While interesting in vitro and vivo data is provided, some of the claims of the authors are not convincingly supported.

Major strengths:

Very interesting effect of FC-A in functional recovery after spinal cord injury.

Major Weaknesses:

Some of the in vitro data, including colocalizations, and analysis of microtubule severing fall short to support the claims of the authors.

The in vivo selectivity of FC-A towards spastin is not adequately supported by the data presented. There are aspects of the spinal cord injury site histology that are unclear.

Response: Reviewer 3's comments align with those of Reviewers 1 and 2.

Reviewer #1 (Recommendations For The Authors):

-The new blots presented in Fig. 3N lacks corresponding labels as for antibodies used for IP and IB and molecular weight markers.

Response: We appreciate the reviewer's feedback. We have made the corresponding modifications in the figure.

Reviewer #2 (Recommendations For The Authors):

The authors have addressed many of the specific concerns shared with the authors in the first round of review but several issues remain with the manuscript.

1. Fig. 1D - the interpretation that spastin co-localizes with 14-3-3 proteins in hippocampal neurons is still tenuous since 14-3-3 uniformly labels the cell.

Response: We appreciate the reviewer's consideration. Upon re-examining the source files, we found that the predominant reason for 14-3-3 showing a ubiquitous cellular distribution was excessive brightness and insufficient contrast. After appropriate adjustments, we discovered that 14-3-3 exhibits characteristic distribution in axons, including aggregation at growth cone and specific locations in the axon shaft. We have made the relevant changes in the revised version.

1. Line 336. The meaning of the following statement is unclear "To further identify which isoform of 14-3-3 interacts with spastin, we generated six 14-3-3 isoforms in rats (β、γ、ε、ζ、η、θ ), then purified GST fusion 14-3-3 proteins (Figure 1G).

Response: Sorry for any confusing statement. We obtained gene fragments of six 14-3-3 isoforms from rat brain cDNA and inserted these fragments into the pEGX-5X-3 vector. Subsequently, GST 14-3-3 fusion proteins were expressed and purified in vitro. We have made the corresponding revisions in the revised version.

1. Line 341. The authors still fall short of showing that spastin and 14-3-3 interact directly thus it may be more accurate to say that they form a complex.

Response: Thank you for the reviewer's advice. We have made the corresponding corrections in the manuscript.

1. Line 388. Please clarify 2th and the meaning of "moderately" - "S233D) was moderately expressed in primary hippocampal neurons at 2th DIV." While it is specified that the transfection dosage and duration were meticulously controlled - it is unclear what the criteria was for establishing the appropriate moderate dosage.

Response: Sorry about the mistake, it should be "2nd" instead of "2th". In order to establish a model for overexpressing spastin to promote neuronal neurite growth, we transfected 0.2 µg of plasmid into 1 well (1×104 cells/cm2, 24-well plate), with a transfection duration controlled at 24 hours.

1. Line 395 - It is unclear if S233D is toxic as there seem to be no measurements of cell survival.

Response: We have supplemented relevant experiments (See comment 6) based on comment 6 and found that Spastin S233D can promote neuronal neurite growth. The corresponding descriptions have been revised.

1. The pro-growth effects of S233A still does not seem to fit the narrative and the results would have been more convincing if dosage was better controlled to establish any differences between WT and S233A Spastin.

Response: We appreciate the constructive comments from the reviewer. In order to better illustrate the role of spastin S233 in neuronal growth, we have made appropriate adjustments to our experimental conditions based on previous experiments. Cells were transfected with plasmids expressing non-fused GFP and spastin and the relevant S233 mutants at a transfection dose of 0.2 µg into 1 well (1×104 cells/cm2, 24-well plate), duration was controlled at 12 hours. Due to the low expression state of the overexpressed protein, GFP (ab290 antibody for IF) was then stained to trace neuronal morphology. The experimental results demonstrate that spastin promotes neuronal neurite growth, and the dephosphorylation mutant of spastin (spastin S233A) significantly attenuates its neurite-promoting effect compared to wild-type spastin. Conversely, the phosphorylation mutant spastin S233D further enhances the promotion of neuronal neurite growth. We have also made corrections to the relevant statements in the manuscript.

1. The reason for examining protection in response to glutamate is not well rationalized based on known spastin functions. The interpretation of this experiment is unclear with respect to effects on protection vs repair.

Response: Thank you for the reviewer's consideration. We suppose that spastin may be involved in both protective and repair processes. Existing studies suggest that spastin can control store-operated calcium entry (SOCE) by altering endoplasmic reticulum morphology (doi: 10.1093/brain/awac122, doi: 10.3389/fphys.2019.01544), which may indicate its role in regulating calcium overload. Additionally, due to the critical role of spastin in axon growth, it is also essential for neuronal repair after injury. Therefore, we have not strictly distinguished between these two concepts here.

1. It is unclear if Spastazoline simply blocks any type of growth and it is thus difficult to conclude that FC-A functions through a 14-3-3-spastin effect based on the current data.

Response: We have re-evaluated and modified the title and main conclusions of the manuscript based on the reviewer's comments and the existing evidence, as responded to in reviewer 1's comments.

1. The access of FC-A to the CNS with the current protocol has not been clearly established and the effects of FC_A on spastin expression seem to mirror the profile of the control condition.

Response: We agree with the reviewer's comments. The expression trend of spastin after FC-A treatment is consistent with that of the control group, with a slight increase in its expression level compared to the control group.

1. The NF and 5-HT staining is not convincing labelling fibres.

Response: We appreciate the reviewer's comments. We believe that the reason for the incomplete axon staining is closely related to the thickness of the tissue sections. In our future research, we will further optimize our axon labeling methods.

Reviewer #3 (Recommendations For The Authors):

Figure 1D: Both spastin and 14-3-3 label the entire neuron which is rather unusual. Conditions of immunfluorescence should be improved. As it is, this image should not be used to claim colocalization.

Response: We appreciate the reviewer's consideration. In response to comment 1 from the expert reviewer 2, we have re-examined the source files and identified that the primary reason for the overall cell-wide distribution of 14-3-3 and spastin is due to excessive brightness and a lack of sufficient contrast. After making appropriate adjustments, we found that 14-3-3 and spastin exhibit characteristic localization within the axon (concentrated in a particular region of the axon shaft and the growth cone). We have made corresponding revisions in the revised version of the manuscript.

Figure S2: The experimental setup and data provided is not adequate to infer microtubule severing.

Response: We appreciate the reviewer's guidance. We have improved the relevant experiments and used a 100X objective lens to observe the microtubule structures more clearly.

Figure 2 I-K: The functional effect of spastin S233A and S233D on neurite outgrowth does not correlate with a function of 14-3-3 and thus does not support the central hypothesis of the manuscript. Minor: The images selected as representative show differences in neurite length and branching that are not portrayed in the graphs.

Response: Thank you for the reviewer’s comment. Similar to the response to the reviewer 2's comment 6, in order to better illustrate the role of spastin S233 in neurite outgrowth, we made corresponding adjustments to our experimental conditions. Cells were transfected with plasmids expressing non-fused GFP and spastin and the relevant S233 mutants at a transfection dose of 0.2 µg into 1 well (1×104 cells/cm2, 24-well plate), duration was controlled at 12 hours. Due to the low expression state of the overexpressed protein, GFP (ab290 antibody for IF) was then stained to trace neuronal morphology. The experimental results demonstrate that spastin promotes neuronal neurite growth, and the dephosphorylation mutant of spastin (spastin S233A) significantly attenuates its neurite-promoting effect compared to wild-type spastin. Conversely, the phosphorylation mutant spastin S233D further enhances the promotion of neuronal neurite growth. We have also made corrections to the relevant statements in the manuscript.

Figure 5 J and L: The quality, resolution and size of the images is insufficient to support the claims of the authors. As it is, one cannot interpret the data. It is very hard to envisage, even considering the explanation provided by the authors, that spinal cords where spastazoline was used correspond to contusion as a complete discontinuity between the rostral and caudal spinal cord tissue is present.

Response: Due to limitations in file uploads, we encountered issues with the resolution of the tissue slices related to spinal cord injury. To address this, we have adjusted the size and resolution of the corresponding images in the supplementary materials (Fig.S7-S9 ) and included the original confocal files and images.

Additionally, it's important to note that the tissue slices we presented do not represent all layers of the spinal cord, and not all layers exhibit discontinuity. Our slices are taken longitudinally at the dorsal site of the lesion area. The dorsal slices represent areas closer to the injury site, while deeper slices correspond to areas distant from the injury site. Therefore, we selected areas closer to the injury site to reflect the repair process following injury.

Figure 7B: Similar comment to spianl cord images provided in Figure 5. NF and MBP are not supposed to colocalize as they label different cell types...

Response: We appreciate the comments from the expert reviewer, and we agree with their suggestions. We will further optimize our axon labeling methods. The excessive brightness and lack of contrast primarily led to the non-specific labeling of other cell types with the MBP antibody. In fact, our primary goal was to highlight the injured areas by enhancing the fluorescence intensity of the images, which inadvertently resulted in neglecting the exclusion of non-specific staining. Therefore, we have made appropriate adjustments to the images to better visualize the distribution of myelin sheaths.

Joint Public Review:

The present work establishes 14-3-3 proteins as binding partners of spastin and suggests that this binding is positively regulated by phosphorylation of spastin. The authors show evidence that 14-3-3 - spastin binding prevents spastin ubiquitination and final proteasomal degradation. By using drugs and peptides that separately inhibit 14-3-3 binding or spastin activity, they show that both proteins are necessary for axon regeneration in cell culture and in vivo models in rats.

Major strengths<br /> -The data establishing 14-3-3 and spastin as binding partners is convincing, as is its regulation by phosphorylation and its impact on protein levels related to the activity of the ubiquitin-proteasome system.

-The effects of FC-A on locomotor recovery after spinal cord contusion is very interesting.

Major weaknesses<br /> -Given that spastazoline has a major impact on neurite outgrowth suggests that cells simply cannot grow in the presence of the inhibitor and raises serious questions about any selectivity for the concomitant effect FC-A - dependent growth.

-The histological data and analyses following spinal cord injury are not convincing. For example, the colabeling of NF and 5-HT is not convincingly labeling fibers. Also, the quality, resolution and size of the images is insufficient to support the quantitative data and it is hence difficult to interpret the data. Reviewers recognize that during the review process, efforts were made to improve the quality of the images.

-Reviewers also observed that the data to infer Spastin actions on Microtubules across different experimental models is weak and that claims about "MT severing" and "microtubules dynamics" were wrongly used given the provided evidence.

-The manuscript lacks direct evidence that a 14-3-3 and spastin function as a complex in the same pathway to promote regeneration. It is recognized, however, that the authors had made changes in the manuscript title and claims not to imply that the current evidence is sufficient in that matter.

Author Response

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

Reviewer #1 (Public Review):

The importance of the role of sexual behavior, specifically ejaculation rates, is worth emphasizing for the formation of pair bonds in prairie voles. It suggests that the role of sexual behavior in contributing to the strength of pair bonds should be explored more. It is also important to add that males and females in the study were screened for sexual receptivity. It would therefore be important to identify characteristics of animals that did not mate under the laboratory conditions used that may add depth and complexity to what was identified in the current study. The identification of brain regions for pair bond maintenance centered around the amygdala was also intriguing.

Thank you for pointing some interpretations of our findings that can be emphasized in the Discussion. We added the following sentences to the Discussion:

“Our findings, along with this previous work, support the hypothesis that sexual behavior plays a key role in driving pair-bond strength. However, the current study focused on animals that were screened for sexual receptivity, which may have limited variation in sexual behavior across pairs. An intriguing direction for future research will be to test how this variation contributes to bond strength.”

We also emphasized amygdala in relation to pair bond maintenance. We added the following sentence to the Discussion:

“These brain regions, and especially amygdala, will be important candidates for future research on neural regulation of pair-bond maintenance.”

The issue of the lack of a strong presence of the reward circuitry (nucleus accumbens) in the final models is also worth more discussion. Perhaps it has been overly emphasized in the past, but there are strong results from other studies pointing to the importance of reward circuitry.

Thank you for this suggestion. There is a section in the Results that analyses accumbens in more detail than other brain regions. Accumbens did not survive our corrections for multiple statistical tests, however it was significant at early timepoint without these corrections. This Results paragraph states the following:

“Although the nucleus accumbens did not survive multiple test corrections in our ROI analysis (q=0.17), it was significant in univariate analysis (p=0.03), particularly when focused on the 2.5 and 6h timepoints (two sample t-test: t=2.53, p=0.01, Video 2). Furthermore, voxel-level comparisons revealed significant sites within the ventral striatum and the posterior nucleus accumbens (Figure 2A, Figure Supplement 1b-c, Video 2).”

We added Supplementary File 4, which contains model comparison results for accumbens and all other ROIs. We also added more detail on nucleus accumbens to the Discussion:

“Pairing drove increased c-Fos expression in the ventral pallidum, a major node in reward circuity, as well as in the paraventricular nucleus and the medial preoptic area, modulators of reward. This is consistent with a large body of work implicating neuropeptide actions on reward circuits in the formation of bonds (Walum & Young, 2018; Young & Wang, 2004). Conspicuously missing from our list, however, is significant pairing induced c-Fos induction in the nucleus accumbens. One possibility is that an absence of significant accumbens IEG induction reflects the limitations of using c-Fos and other immediate early genes as indicators of neural activity. It is known that some neuronal populations can be active without expressing c-Fos (Sheng & Greenberg, 1990). Indeed, although a variety of studies implicate the accumbens in bond formation (Amadei et al., 2017; Aragona et al., 2006; Scribner et al., 2020), previous work finds only weak c-Fos induction in the prairie vole accumbens during bonding (Curtis & Wang, 2003). Another possibility is that there was heterogeneous activation in the accumbens that was not captured by the precision of our atlas. Consistent with this interpretation, found that the accumbens was significant in univariate tests, as well as in voxel-level analyses. Overall, our results do not conflict with pharmacological, electrophysiological, and calcium-imaging data on the role of the nucleus accumbens in prairie vole bonding (Amadei et al., 2017; Aragona et al., 2006; Scribner et al., 2020). Instead, the absence of significant effects at the level of the entire nucleus accumbens together with the presence of anatomically restricted voxel-level significance suggests substantial anatomical heterogeneity in the contributions of the nucleus accumbens to bond formation.”

Please discuss the consequences of creating the behavioral data for pair bond formation by subtracting same-sex pairs interactions from the opposite-sex interactions. What sources of information are removed by using this approach?

One limitation of our study’s approach is that we are unable to fully separate information related to social novelty from mating experience. Thank you for pointing out that we should touch on this sort of caveat in the paper. We added several sentences to the Discussion:

“It seems likely that sensory and motor areas were important for social processes related to both pair-bonding and reunion with same-sex cagemates, such as investigation and recognition. Our study design, however, highlights differences between treatments, and in order to detect such effects, it might be necessary to compare mating and bonding pairs to animals left in complete isolation.”

We reiterate the point in a new paragraph we added to the Discussion to explicitly provide caveats regarding our data:

“Before offering a synthesis of our findings, it would be useful to acknowledge a few caveats. First, as noted above, IEG induction does not capture all relevant neural activity. Second, the design of our experiment, which controlled for social interaction, likely excluded many circuits important to both pair bonding and sibling social interactions. Third, c-Fos activity within a given brain region may nevertheless rely on distinct cell types, and so the absence of sex differences in c-Fos immunoreactivity does not definitively rule out the sexually dimorphic circuits hypothesized in the “dual function hypothesis” (de Vries, 2004). Lastly, the current study focused on animals that were screened for sexual receptivity, which may have limited the variation in sexual behavior across opposite-sex pairs.”

Time 0 is when the barrier is removed after a two-hour exposure. Please speculate on what is going on during the two-hour exposure. Time zero is potentially more than the time of mating. Is it possible that aggression is being decreased during this timepoint that represents mating? Could it also be a measure of the outcome of an initial compatibility assessment by the male and female?

Thank you for this interesting observation. While the opaque divider prevented physical social interactions, it is possible that animals picked up on auditory or olfactory cues. We did not detect group differences in movement patterns and vocalization rates from the 0 h timepoint group (Figure 2). These findings suggest that potential partner detection and assessment occurred in a similar way for both experiment groups. It is unlikely that this period represents a decrease in aggression, since unbonded prairie voles are not known to be aggressive towards conspecifics. However, the idea that animals may potentially use olfactory or auditory cues to assess each other is an interesting idea, and one that we cannot rule out. We added a brief statement to the Methods “Experiment Design” section about the possibility that the two hours prior to divider removal (0 h timepoint) could represent more than an acclimation period:

“It is important to note that the opaque divider in the acclimation period prevented physical interactions, but it is possible that animal pairs may have detected each other through olfactory or auditory cues.”

We also mention this in the revised Discussion in the context of the PFC cluster, which not only differed between mating and non-mating groups, but also showed differences between isolated (0h) and socially interacting animals (sibs and mates, 2.5h-22h):

“A fourth cluster (“PFC,” green) is composed of prelimbic, infralimbic and olfactory cortex; activity in the vole prefrontal cortex is known to be modulated by hypothalamic oxytocin, and to shape bonding through projections to the nucleus accumbens (Amadei et al., 2017; Burkett et al., 2016; Horie et al., 2020). The pattern of activity in this cluster, however, indicates that it was due in part to differences between the isolated animals (0h) and other time points (Figure 4—figure supplement 1 and Figure 4—figure supplement 2). Because animals in the isolated condition were in a compartment adjacent to either an opposite sexed individual or a familiar former cagemate, we cannot rule out that olfactory or auditory cues may have made animals aware of the presence of a potential social partner. Indeed, we interpret this dimension as capturing appetitive aspects of behaviors associated with investigation of the animal isolated from the subject by the barrier.”

Reviewer #2 (Public Review):

An important caveat to this study not mentioned by the authors is that c-fos provides a snapshot of neural activity and that important populations of neurons could be active and not express c-fos. Thus observed correlations are likely to be robust, but the absence of differences (in say accumbens) may just reflect the limits of c-fos estimation of neural activity. Similarly, highly coordinated neural activity between males and females might still be driven by different mechanisms if different cell types were activated within a specific region.

We now discuss limitations of c-Fos in the Discussion paragraph that focuses on accumbens:

“The absence of significant accumbens IEG induction may reflect the limitations of using c-Fos and other immediate early genes as indicators of neural activity. It is known that some neuronal populations can be active without expressing c-Fos (Sheng & Greenberg, 1990). Indeed, although a variety of studies implicate the accumbens in bond formation (Amadei et al., 2017; Aragona et al., 2006; Scribner et al., 2020), previous work finds only weak c-Fos induction in the prairie vole accumbens during bonding (Curtis & Wang, 2003).”

We also include the following sentence in a new Discussion paragraph that focuses on caveats to our findings:

“Before offering a synthesis of our findings, it would be useful to acknowledge or reiterate a few caveats. First, as noted above, IEG induction does not capture all relevant neural activity (Sheng & Greenberg 1990). Second, the design of our experiment, which controlled for social interaction, likely excluded many circuits important to both pair bonding and sibling social interactions. Third, c-Fos activity within a given brain region may nevertheless rely on distinct cell types, and so the absence of sex differences in c-Fos immunoreactivity does not definitively rule out the sexually dimorphic circuits hypothesized in the “dual function hypothesis” (de Vries, 2004). Lastly, the current study focused on animals that were screened for sexual receptivity, which may have limited the variation in sexual behavior across opposite-sex pairs.”

Recommendations for the authors:

It appears as if df is missing from some statistical reporting.

Thank you for pointing this out. We went through the manuscript and added in sample sizes to statistical reporting.

Reviewer #1 (Recommendations for the authors):

It is surprising that the cortex was not more extensively identified as being involved in pair bonding, but perhaps this is because the emphasis for choosing brain areas in the cortical region is biased towards olfactory regions. Please discuss. It may also be worth noting that brain regions associated with perception may be important in all of these processes, but selected out because of the design.

Thank you for this observation. We agree that some cortical regions may not have been identified due to the study design. For example, social processes related to both pair bonding and cagemate recognition likely rely on overlapping circuits. It is also important to note here that our analysis approach identified the “most” significant regions. This means that several candidate regions did not survive the statistical threshold used to select regions. We now discuss the cortex in more detail in the Discussion, where we also identify the regions that approached significance but did not survive multiple test corrections:

“Although the PFC and other olfactory cortical areas formed a cluster, we did not find widespread c-Fos induction throughout the cortex in response to pairing. It seems likely that sensory and motor areas were important for social processes related to both pair-bonding and reunion with same-sex cagemates, such as investigation and recognition. Our study design, however, highlights differences between treatments, and in order to detect such effects, it might be necessary to compare mating and bonding pairs to animals left in complete isolation. Moreover, several cortical regions that did not survive corrections for multiple tests may have been identified in a less stringent analysis. Several subregions within the isocortex, hippocampal formation, and cortical subplate had statistical models that approached significance (i.e., p-values < 0.1) prior to multiple test corrections. These subregions were found within primary somatosensory area, primary auditory area, dorsal and ventral auditory areas, primary visual area, anteromedial visual area, agranular insular area, temporal association areas, ectorhinal area, postsubiculum, and basomedial amygdala. Frontal cortex subregions were within the agranular insular area and orbital area, as well as additional subregions in prelimbic and infralimbic areas of the PFC.”

Same-sex siblings were isolated for 4-5 days and then repaired. This is a creative way of dealing with this, but was any aggression displayed in the same-sex pairs? Are there bonds or preferences among same-sex individuals? Could the isolation have set the stage for neural changes associated with migrating from the natal group? 4-5 days of isolation is not trivial.

Thank you for these questions. We did not witness aggression between same-sex pairs. We had recorded ‘aggression’ events (lunges and chases) during the 1 h behavioral observation epochs and found that these rates were nearly zero for all sibling timepoint groups (events/h per focal animal in mean ± sd: 2.5 h group = 0.58 ± 1.53, 6 h group = 0.17 ± 0.48, 22 h group = 0.25 ± 0.44).

The question about peer relationships is a good one. Previous literature does suggest that prairie voles can develop preferences for familiar same-sex individuals (e.g., Beery et al. 2018 Front. Behav. Neuro., Lee et al. 2019 Front. Behav. Neuro). Thus, we want to reiterate here that our study design tests for differences between these baseline levels of affiliation with pair bonding in a reproductive context.

It is possible that the period of isolation prior to experiments may have set the stage for neural changes associated with migration from the natal group. Testing this possibility is outside the scope of the current study. We want to point out here that animals were separated from their natal groups several weeks prior to the experiment. Animals were weaned at 21 days and put into same-sex cages, and then experiments occurred between 8-12 weeks of age. All experiment groups went through the same weaning and co-housing conditions.

Pg 26, Line 655: "better" is listed twice in the sentence and only one is needed

Thank you for catching this typo. This is fixed.

Reviewer #2 (Recommendations for the authors):

Why was it necessary to bring voles into estrus when they are induced ovulators? The authors need to state how voles were brought into estrus.

Thank you for this suggestion. We explained estrus induction in the Methods, but this explanation could be missed because it was within the “Behavioral procedures” section. We put the paragraph about estrus induction into a new section called “Estrus induction and animal selection”. We also elaborated on the final sentences of this paragraph to provide a clearer rationale:

“We used this mating assay to restrict study subjects to voles that showed lordosis (females) or mounting behavior (males). By selecting voles who showed sexual behavior, we could control the estrus state and timing of mating across the 0, 2.5, 6 and 22 h study groups. This selection process also ensured that animals assigned to the same-sex sibling pair and opposite-sex mating pair groups had similar sexual motivation and experience.”

I assume in the final manuscript the authors will release the availability of the atlas? Making the atlas public seems to be in the spirit of the eLife publishing model.

The prairie vole reference brain, atlas, and atlas annotation labels, are now included on the Figshare repository site. We updated the Data and code availability section to clarify this.

Reviewer #3 (Recommendations for the authors):

Please clarify in the Methods if same-sex sibling females were also estrogen primed. If not, could the estrogen exposure cause Fos differences?

Thank you for this suggestion. All females were estrogen primed. We refined the Methods section “Estrus induction and animal selection” to make this part of the study design clearer. We edited one of the sentences to say “During this isolation period, all females were induced into estrus[...]” We also added a couple sentences at the end of this paragraph:

“By selecting voles who showed sexual behavior, we could control the estrus state and timing of mating across the 0, 2.5, 6 and 22 h study groups. This selection process also ensured that animals assigned to the same-sex sibling pair and opposite-sex mating pair groups had similar sexual motivation and experience.”

eLife assessment

This is an important study using 3D mapping of neuronal activation throughout the brain after pair-bonding in the monogamous vole, which can be broadly applied to other species and behaviors. The authors provide compelling evidence that there is some synchrony between male and female partners that have formed a pair bond, the strength of which is based on the number of ejaculations received by the female. Same-sex pairs also form a pair bond and were found to have activation in the same brain regions as mixed sex couples. An overall low level of sex differences in the degree and location of brain activation was observed, which was unexpected. This work will be of interest to those interested in social behavior and its neural mechanisms, or brain systems or behavior more broadly.

Reviewer #1 (Public Review):

While the approach used in this study cannot identify cause and effect, the whole brain approach identified clusters representing circuits of potential importance and a series of new hypotheses to explore. The importance of the role of sexual behavior, specifically ejaculations rates, is worth emphasizing for the formation of pair bonds. It suggests that the role of sexual behavior in contributing to the strength of pair bonds should be explored more. It is also important to add that males and females in the study were screened for sexual receptivity. The identification of brain regions for pair bond maintenance centered around the amygdala was also intriguing.

Reviewer #2 (Public Review):

In this manuscript the authors generate an annotated brain atlas for the prairie vole, which is a widely studied organism. This species has a suite of social behaviors that are difficult or impossible to study in conventional rodents, and has attracted a large community of researchers. The atlas is impressive and will be a fantastic resource. The authors use this atlas to examine brain-wide c-fos expression in prairie voles that were paired with same sex or opposite sex vole across multiple timepoints. In some sense the design resembles PET studies done in primates that take whole brain scans after an important behavioral experience. The authors observed increased c-fos expression across a network of brain regions that largely corresponds with the previous literature. The study design captured several novel observations including that c-fos expression in some regions correlate strongly between males and females during pair bond formation and mating, suggesting synchrony in neural activity. The authors address an important caveat that c-fos provides a snapshot of neural activity and that important populations of neurons could be active and not express c-fos. Thus observed correlations are likely to be robust, but that the absence of differences (in say accumbens) may just reflect the limits of c-fos estimation of neural activity. Similarly, highly coordinated neural activity between males and females might still be driven by different mechanisms if different cell types were activated within a specific region. The creation of this resource and it's use in a well designed study is an important accomplishment.

Reviewer #3 (Public Review):

In this manuscript, Gustison et al., describe the development of an automated whole-brain mapping pipeline, including the first 3D histological atlas of the prairie vole, and then use that pipeline to quantify Fos immunohistochemistry as a measure of neural activity during mating and pair bonding in male and female prairie voles. Prairie voles have become a useful animal model for examining the neural bases of social bonding due to their socially monogamous mating strategy. Prior studies have focused on identifying the role of a few neuromodulators (oxytocin, vasopressin, dopamine) acting in limited number of brain regions. The authors use this unbiased approach to determine which areas become activated during mating, cohabitation, and pair bonding in both sexes to identify 68 brain regions clustered in seven brain-wide neuronal circuits that are activated over the course of pair bonding. This is an important study because i) it generates a valuable tool and analysis pipeline for other investigators in the prairie vole research community and ii) it highlights potential involvement of many brain regions in regulating sexual behavior, social engagement and pair bonding that have not been previously investigated.

Strengths of the study include the unbiased assessment of neural activity using the automated whole brain activity mapped onto the 3D histological atlas. The design of the behavioral aspect of study is also a strength. Brains were collected at baseline and 2.5, 6 and 22 hrs after cohabitation with either a sibling or opposite sex partner. These times were strategically chosen to correspond to milestones in pair bond development. Behavior was also quantified during epochs over the 22 hr period providing useful information on the progression of behaviors (e.g. mating) during pair bonding and relating Fos activation to specific behaviors (e.g. sex vs bonding). The sibling co-housed group provided an important control, enabling identification of areas specifically activated by sex and bond formation. The analyses of the data were rigorous, resulting in convincing conclusions. While there was nothing particularly surprising in terms of the structures that were identified to be active during the mating and cohabitation, the statistical analysis revealed interesting relationships in terms of interactions of the various clusters, and also some level of synchrony in brain activation between partners. Furthermore, ejaculation was found to be the strongest predictor of Fos activation in both males and females. The sex differences identified in the study was subtle and less than the authors expected, which is interesting.

While the study provides a potentially useful tool and approach that may be general use to the prairie vole community and identifies in an anatomically precise manner areas that may be important for mating or pair bond formation, there are some weaknesses as well. The study is largely descriptive. It is impossible to determine whether the activated areas are simply involved in sex or in the pair bond process itself. In other words, the authors did not use the Fos data to inform functional testing of circuits in pair bonding or mating behaviors. However, that is likely beyond the scope of this paper in which the goal was more to describe the automated, unbiased approach. This weakness is offset by the value of the comprehensive and detailed analysis of the Fos activation data providing temporal and precise anatomical relationships between brain clusters and in relation to behavior. The manuscript concludes with some speculative interpretations of the data, but these speculations may be valuable for guiding future investigations.

Author Response:

We thank the reviewers for their careful comments. We sincerely agree with the comments from both reviewers, and noticed the word “cell transplantation”, throughout the manuscript including the title, was confusing. We will revise the manuscript to clarify the aim of the study, and to express the conclusion more straightforwardly.

Response to reviewers:

We interpret the data of the present study as the color of each RPE cell is a temporal condition which does not necessarily represent the quality (e.g. for cell transplantation) of the cells. We consider this may be applicable not only in vitro but also in vivo, although we do not know whether RPE shows heterogeneous level of pigmentation in vivo.

As our concern for iPSC-RPE is always about their quality for cell transplantation, maybe we haven’t fairly evaluated the scientific significance obtained from the present study.

Another thing we noticed was, although we used the term “cell transplantation” to explain what we meant by “quality” of the cells, we agree this was confusing. The aim of the study was not to show how the pigmentation level of transplant-RPE affects the result of cell transplantation, but to show the heterogeneous gene expression of iPSC-derived RPE cells, and the less correlation of the heterogeneity with the pigmentation level. We went through the manuscript, including the title, to more straightforwardly lead this conclusion: the degree of pigmentation had some but weak correlation with the expression levels of functional genes, and the reason for the weakness of the correlation may be because the color is a temporal condition (as we interpreted from the data) that is different from more stable characteristics of the cells.

We agree that “cell transplantation” in the title (and other parts) was misleading. So, we will change the title, and removed the phrase that led as if the aim of the study was to show something about cell transplantation or in vivo results.

Also, to face scientifically significant results obtained from the present study appropriately, we will discuss more about the correlation of the pigmentation level with some functional genes, and brought this as one of the conclusions of the manuscript.

eLife assessment

This useful study presents a novel method to analyze the correlation between the degree of pigmentation, and the gene expression profile of human-induced pluripotent stem cell-derived Retinal Pigmented Epithelial (iPSC-RPE) cells at the single cell level, trying to establish if this parameter might be of use as a guide for transplantation. The presented evidence is incomplete: pigmentation was not an indicator of functionality or maturity and the data, although obtained from a pertinent approach, is limited to in vitro conditions; further, this approach is not complete since no attempts were made to graft iPSC-RPE.

Reviewer #1 (Public Review):

Summary:<br /> In this paper, the authors show that the degree of pigmentation for RPE cells is not correlated with a level of maturation and function. They suggest that this status could be different in vitro than in vivo but do not provide proper experiments to validate this hypothesis. However, it is the first time that the absence of a correlation between pigmentation and function has been studied.

Strengths:<br /> The methods are good and the experiments are very rigorous.

Weaknesses:<br /> Demonstration of in vitro but no in vivo data.

Reviewer #2 (Public Review):

Summary:<br /> Nakai-Futatsugi et al. present a novel method to analyze the correlation between the degree of pigmentation and the gene expression profile of human-induced pluripotent stem cell-derived RPE (iPSC-RPE) cells at the single cell level. This was achieved with the use of ALPS (Automated Live imaging and cell Picking System), an invention developed by the same authors. Briefly, it allows one to choose and photograph a specific cell from a culture dish and proceed to single-cell digital RNA-seq. The authors identify clusters of cells that present differential gene expression, but this shows no association with the degree of pigmentation of the cells. Further data analysis allowed the authors to correlate the degree of pigmentation to some degree with the expression of complement and lysosome-related genes.

Strengths:<br /> An important amount of data related to gene expression and heterogeneity of the iPSC-RPE population has been generated in this work.

Weaknesses:<br /> However, the justification of the analysis, and the physiological relevance of the hypothesis and the findings could be strengthened.

Importantly, I fail to grasp from the introduction what is the previous evidence that leads to the hypothesis. Why would color intensity be related to the quality of cell transplantation? In fact, cell transplantation is not evaluated at all in this work. The authors mention "quality metrics for clinical use", but this concept is not further explained. Neither is the concept of "sufficient degree of pigmentation" explained.

On the other hand, the positive correlation of cluster formation with complement and lysosome-related genes is not discussed.

As a consequence, it is very difficult to evaluate the impact of these findings on the field.

eLife assessment

This valuable study applies voltage clamp fluorometry to provide new information about the function of serotonin-gated ion channels 5-HT3AR. The authors convincingly investigate structural changes inside and outside the orthosteric site elicited by agonists, partial agonists, and antagonists, helping to annotate existing cryo-EM structures. This work confirms that the activation of 5-HT3 receptors is similar to other members of this well-studied receptor superfamily. The work will be of interest to scientists working on channel biophysics but also drug development targeting ligand-gated ion channels.

Reviewer #1 (Public Review):

Summary: This study brings new information about the function of serotonin-gated ion channels 5-HT3AR, by describing the conformational changes undergoing during ligands binding. These results can be potentially extrapolated to other members of the Cys-loop ligand-gated ion channels. By combining fluorescence microscopy with electrophysiological recordings, the authors investigate structural changes inside and outside the orthosteric site elicited by agonists, partial agonists, and antagonists. The results are convincing and correlate well with the observations from cryo-EM structures. The work will be of important significance and broad interest to scientists working on channel biophysics but also drug development targeting ligand-gated ion channels.

Strengths: The authors present an elegant and well-designed study to investigate the conformational changes on 5-HT3AR where they combine electrophysiological and fluorometry recordings. They determined four positions suitable to act as sensors for the conformational changes of the receptor: two inside and two outside the agonist binding site. They make a strong point showing how antagonists produce conformational changes inside the orthosteric site similarly as agonists do but they failed to spread to the lower part of the ECD, in agreement with previous studies and Cryo-EM structures. They also show how some loss-of-function mutant receptors elicit conformational changes (changes in fluorescence) after partial agonist binding but failed to produce measurable ionic currents, pointing to intermediate states that are stabilized in these conditions. The four fluorescence sensors developed in this study may be good tools for further studies on characterizing drugs targeting the 5-HT3R.

Weaknesses: Although the major conclusions of the manuscript seem well justified, some of the comparison with the structural data may be vague. The claim that monitoring these silent conformational changes can offer insights into the allosteric mechanisms contributing to signal transduction is not unique to this study and has been previously demonstrated by using similar techniques with other ion channels.

Reviewer #2 (Public Review):

Summary:<br /> This study focuses on the 5-HT3 serotonin receptor, a pentameric ligand-gated ion channel important in chemical neurotransmission. There are many cryo-EM structures of this receptor with diverse ligands bound, however assignment of functional states to the structures remains incomplete. The team applies voltage-clamp fluorometry to measure, at once, both changes in ion channel activity, and changes in fluorescence. Four cysteine mutants were selected for fluorophore labeling, two near the neurotransmitter site, one in the ECD vestibule, and one at the ECD-TMD junction. Agonists, partial agonists, and antagonists were all found to yield similar changes in fluorescence, a proxy for conformational change, near the neurotransmitter site. The strength of the agonist correlated to a degree with propagation of this fluorescence change beyond the local site of neurotransmitter binding. Antagonists failed to elicit a change in fluorescence in the vestibular the ECD-TMD junction sites. The VCF results further turned up evidence supporting intermediate (likely pre-active) states.

Strengths:<br /> The experiments appear rigorous, the problem the team tackles is timely and important, the writing and the figures are for the most part very clear. We sorely need approaches orthogonal to structural biology to annotate conformational states and observe conformational transitions in real membranes- this approach, and this study, get right to the heart of what is missing.

Weaknesses:<br /> The weaknesses in the study itself are overall minor, I only suggest improvements geared toward clarity. What we are still missing is application of an approach like this to annotate the conformation of the part of the receptor buried in the membrane; there is important debate about which structure represents which state, and that is not addressed in the current study.

Reviewer #3 (Public Review):

Summary:<br /> The authors have examined the 5-HT3 receptor using voltage clamp fluorometry, which enables them to detect structural changes at the same time as the state of receptor activation. These are ensemble measurements, but they enable a picture of the action of different agonists and antagonists to be built up.

Strengths:<br /> The combination of rigorously tested fluorescence reporters with oocyte electrophysiology is a solid development for this receptor class.

Weaknesses:<br /> The interpretation of the data is solid but relevant foundational work is ignored. Although the data represent a new way of examining the 5-HT3 receptor, nothing that is found is original in the context of the superfamily. Quantitative information is discussed but not presented.

Author Response:

Reviewer #1 (Public Review):

[...] Weaknesses:

1. I feel the authors need to justify why flow-crushing helps localization specificity. There is an entire family of recent papers that aim to achieve higher localization specificity by doing the exact opposite. Namely, MT or ABC fRMRI aims to increase the localization specificity by highlighting the intravascular BOLD by means of suppressing non-flowing tissue. To name a few:

Priovoulos, N., de Oliveira, I.A.F., Poser, B.A., Norris, D.G., van der Zwaag, W., 2023. Combining arterial blood contrast with BOLD increases fMRI intracortical contrast. Human Brain Mapping hbm.26227. https://doi.org/10.1002/hbm.26227.

Pfaffenrot, V., Koopmans, P.J., 2022. Magnetization Transfer weighted laminar fMRI with multi-echo FLASH. NeuroImage 119725. https://doi.org/10.1016/j.neuroimage.2022.119725

Schulz, J., Fazal, Z., Metere, R., Marques, J.P., Norris, D.G., 2020. Arterial blood contrast ( ABC ) enabled by magnetization transfer ( MT ): a novel MRI technique for enhancing the measurement of brain activation changes. bioRxiv. https://doi.org/10.1101/2020.05.20.106666

Based on this literature, it seems that the proposed method will make the vein problem worse, not better. The authors could make it clearer how they reason that making GE-BOLD signals more extra-vascular weighted should help to reduce large vein effects.

The empirical evidence for the claim that flow crushing helps with the localization specificity should be made clearer. The response magnitude with and without flow crushing looks pretty much identical to me (see Fig, 6d). It's unclear to me what to look for in Fig. 5. I cannot discern any layer patterns in these maps. It's too noisy. The two maps of TE=43ms look like identical copies from each other. Maybe an editorial error?

The authors discuss bipolar crushing with respect to SE-BOLD where it has been previously applied. For SE-BOLD at UHF, a substantial portion of the vein signal comes from the intravascular compartment. So I agree that for SE-BOLD, it makes sense to crush the intravascular signal. For GE-BOLD however, this reasoning does not hold. For GE-BOLD (even at 3T), most of the vein signal comes from extravascular dephasing around large unspecific veins, and the bipolar crushing is not expected to help with this.

The authors would like to clarify that the velocity-nulling gradient is NOT designed to suppress all the contributions from intravascular blood. Instead, we tried to find a balance so that the VN gradient maximally suppressed the macrovascular signal in unspecific veins but minimally attenuated the microvascular signal in specific capillary bed. We acknowledge the reviewer's concern regarding the potential extravascular contributions from large, non-specific vessels. This aspect will be thoroughly evaluated and addressed in the revised manuscript. Additionally, we will make clarifications in other parts that may have cause the reviewer’s misunderstandings.

1. The bipolar crushing is limited to one single direction of flow. This introduces a lot of artificial variance across the cortical folding pattern. This is not mentioned in the manuscript. There is an entire family of papers that perform layer-fmri with black-blood imaging that solves this with a 3D contrast preparation (VAPER) that is applied across a longer time period, thus killing the blood signal while it flows across all directions of the vascular tree. Here, the signal cruising is happening with a 2D readout as a "snap-shot" crushing. This does not allow the blood to flow in multiple directions. VAPER also accounts for BOLD contaminations of larger draining veins by means of a tag-control sampling. The proposed approach here does not account for this contamination.

Chai, Y., Li, L., Huber, L., Poser, B.A., Bandettini, P.A., 2020. Integrated VASO and perfusion contrast: A new tool for laminar functional MRI. NeuroImage 207, 116358. https://doi.org/10.1016/j.neuroimage.2019.116358

Chai, Y., Liu, T.T., Marrett, S., Li, L., Khojandi, A., Handwerker, D.A., Alink, A., Muckli, L., Bandettini, P.A., 2021. Topographical and laminar distribution of audiovisual processing within human planum temporale. Progress in Neurobiology 102121. https://doi.org/10.1016/j.pneurobio.2021.102121

If I would recommend anyone to perform layer-fMRI with blood crushing, it seems that VAPER is the superior approach. The authors could make it clearer why users might want to use the unidirectional crushing instead.

We acknowledge that the degree of velocity nulling varies across the cortical folding pattern. We intend to discuss potential solutions to address this variance, and these may be implemented in the revised manuscript as appropriate. Furthermore, we will provide a comprehensive discussion on the advantages and disadvantages of both CBV-based and BOLD-based approaches.

1. The comparison with VASO is misleading. The authors claim that previous VASO approaches were limited by TRs of 8.2s. The authors might be advised to check the latest literature of the last years. Koiso et al. performed whole brain layer-fMRI VASO at 0.8mm at 3.9 seconds (with reliable activation), 2.7 seconds (with unconvincing activation pattern, though), and 2.3 (without activation). Also, whole brain layer-fMRI BOLD at 0.5mm and 0.7mm has been previously performed by the Juelich group at TRs of 3.5s (their TR definition is 'fishy' though).

Koiso, K., Müller, A.K., Akamatsu, K., Dresbach, S., Gulban, O.F., Goebel, R., Miyawaki, Y., Poser, B.A., Huber, L., 2023. Acquisition and processing methods of whole-brain layer-fMRI VASO and BOLD: The Kenshu dataset. Aperture Neuro 34. https://doi.org/10.1101/2022.08.19.504502

Yun, S.D., Pais‐Roldán, P., Palomero‐Gallagher, N., Shah, N.J., 2022. Mapping of whole‐cerebrum resting‐state networks using ultra‐high resolution acquisition protocols. Human Brain Mapping. https://doi.org/10.1002/hbm.25855

Pais-Roldan, P., Yun, S.D., Palomero-Gallagher, N., Shah, N.J., 2023. Cortical depth-dependent human fMRI of resting-state networks using EPIK. Front. Neurosci. 17, 1151544. https://doi.org/10.3389/fnins.2023.1151544

The authors are correct that VASO is not advised as a turn-key method for lower brain areas, incl. Hippocampus and subcortex. However, the authors use this word of caution that is intended for inexperienced "users" as a statement that this cannot be performed. This statement is taken out of context. This statement is not from the academic literature. It's advice for the 40+ user base that wants to perform layer-fMRI as a plug-and-play routine tool in neuroscience usage. In fact, sub-millimeter VASO is routinely being performed by MRI-physicists across all brain areas (including deep brain structures, hippocampus etc). E.g. see Koiso et al. and an overview lecture from a layer-fMRI workshop that I had recently attended: https://youtu.be/kzh-nWXd54s?si=hoIJjLLIxFUJ4g20&t=2401

Thus, the authors could embed this phrasing into the context of their own method that they are proposing in the manuscript. E.g. the authors could state whether they think that their sequence has the potential to be disseminated across sites, considering that it requires slow offline reconstruction in Matlab? Do the authors think that the results shown in Fig. 6c are suggesting turn-key acquisition of a routine mapping tool? In my humble opinion, it looks like random noise, with most of the activation outside the ROI (in white matter).

Those literatures will be included and discussed in the revised manuscript. Furthermore, we are considering the exclusion of the LGN results presented in Figure 6, as they may divert attention from the primary focus of the study.

We are enthusiastic about sharing our imaging sequence, provided its usefulness is conclusively established. However, it's important to note that without an online reconstruction capability, such as the ICE, the practical utility of the sequence may be limited. Unfortunately, we currently don’t have the manpower to implement the online reconstruction. Nevertheless, we are more than willing to share the offline reconstruction codes upon request.

1. The repeatability of the results is questionable. The authors perform experiments about the robustness of the method (line 620). The corresponding results are not suggesting any robustness to me. In fact, the layer profiles in Fig. 4c vs. Fig 4d are completely opposite. The location of peaks turns into locations of dips and vice versa. The methods are not described in enough detail to reproduce these results. The authors mention that their image reconstruction is done "using in-house MATLAB code" (line 634). They do not post a link to github, nor do they say if they share this code.

It is not trivial to get good phase data for fMRI. The authors do not mention how they perform the respective coil-combination. No data are shared for reproduction of the analysis.

There may have been a misunderstanding regarding the presentation in Figure 4, which illustrates the impact of TEs and the VN gradient. To enhance clarity and avoid further confusion, we will redesign this figure for improved comprehension.

The authors are open to sharing the MATLAB codes associated with our study. However, we were limited by manpower for refining and enhancing the readability of these codes for broader use.

Regarding the coil combination, we utilized an adaptive coil combination approach as described in the paper by Walsh DO, Gmitro AF, and Marcellin MW, titled 'Adaptive reconstruction of phased array MR imagery' (Magnetic Resonance in Medicine 2000; 43:682-690). The MATLAB code for this method was implemented by Dr. Diego Hernando. We will include a link for downloading this code in the revised manuscript for the convenience of interested readers.

1. The application of NODRIC is not validated. Previous applications of NORDIC at 3T layer-fMRI have resulted in mixed success. When not adjusted for the right SNR regime it can result in artifactual reductions of beta scores, depending on the SNR across layers. The authors could validate their application of NORDIC and confirm that the average layer-profiles are unaffected by the application of NORDIC. Also, the NORDIC version should be explicitly mentioned in the manuscript.

Akbari, A., Gati, J.S., Zeman, P., Liem, B., Menon, R.S., 2023. Layer Dependence of Monocular and Binocular Responses in Human Ocular Dominance Columns at 7T using VASO and BOLD (preprint). Neuroscience. https://doi.org/10.1101/2023.04.06.535924

Knudsen, L., Guo, F., Huang, J., Blicher, J.U., Lund, T.E., Zhou, Y., Zhang, P., Yang, Y., 2023. The laminar pattern of proprioceptive activation in human primary motor cortex. bioRxiv. https://doi.org/10.1101/2023.10.29.564658

During our internal testing, we observed that the NORDIC denoising process did not alter the activation patterns. These findings will be incorporated into the revised manuscript. The details of NORDIC will be provided as well.

Reviewer #2 (Public Review):

[...] The well-known double peak feature in M1 during finger tapping was used as a test-bed to evaluate the spatial specificity. They were indeed able to demonstrate two distinct peaks in group-level laminar profiles extracted from M1 during finger tapping, which was largely free from superficial bias. This is rather intriguing as, even at 7T, clear peaks are usually only seen with spatially specific non-BOLD sequences. This is in line with their simple simulations, which nicely illustrated that, in theory, intravascular macrovascular signals should be suppressible with only minimal suppression of microvasculature when small b-values of the VN gradients are employed. However, the authors do not state how ROIs were defined making the validity of this finding unclear; were they defined from independent criteria or were they selected based on the region mostly expressing the double peak, which would clearly be circular? In any case, results are based on a very small sub-region of M1 in a single slice - it would be useful to see the generalizability of superficial-bias-free BOLD responses across a larger portion of M1.

Given the individual variations in the location of the M1 region, we opted for manual selection of the ROI. In the revised manuscript, we plan to explore and implement an independent criterion for ROI selection to enhance the objectivity and reproducibility of our methodology.

As repeatedly mentioned by the authors, a laminar fMRI setup must demonstrate adequate functional sensitivity to detect (in this case) BOLD responses. The sensitivity evaluation is unfortunately quite weak. It is mainly based on the argument that significant activation was found in a challenging sub-cortical region (LGN). However, it was a single participant, the activation map was not very convincing, and the demonstration of significant activation after considerable voxel-averaging is inadequate evidence to claim sufficient BOLD sensitivity. How well sensitivity is retained in the presence of VN gradients, high acceleration factors, etc., is therefore unclear. The ability of the setup to obtain meaningful functional connectivity results is reassuring, yet, more elaborate comparison with e.g., the conventional BOLD setup (no VN gradients) is warranted, for example by comparison of tSNR, quantification and comparison of CNR, illustration of unmasked-full-slice activation maps to compare noise-levels, comparison of the across-trial variance in each subject, etc. Furthermore, as NORDIC appears to be a cornerstone to enable submillimeter resolution in this setup at 3T, it is critical to evaluate its impact on the data through comparison with non-denoised data, which is currently lacking.

We appreciate the reviewer’s comments. Those issues will be addressed carefully.

Reviewer #3 (Public Review):

[...] Weaknesses: - Although the VASO acquisition is discussed in the introduction section, the VN-sequence seems closer to diffusion-weighted functional MRI. The authors should make it more clear to the reader what the differences are, and how results are expected to differ. Generally, it is not so clear why the introduction is so focused on the VASO acquisition (which, curiously, lacks a reference to Lu et al 2013). There are many more alternatives to BOLD-weighted imaging for fMRI. CBF-weighted ASL and GRASE have been around for a while, ABC and double-SE have been proposed more recently.

The principal distinction between DW-fMRI and our methodology lies in the level of the b-value employed. DW-fMRI typically measures cellular swelling by utilizing a b-value greater than 1000 s/mm^2 (e.g. 1800). Conversely, our Velocity Nulling functional MRI (VN-fMRI) approach continues to assess hemodynamic responses, utilizing a smaller b-value specifically for the suppression of signals from draining veins. In addition, other layer-fMRI methods will be discussed.

• The comparison in Figure 2 for different b-values shows % signal changes. However, as the baseline signal changes dramatically with added diffusion weighting, this is rather uninformative. A plot of t-values against cortical depth would be much more insightful.
• Surprisingly, the %-signal change for a b-value of 0 is not significantly different from 0 in the gray matter. This raises some doubts about the task or ROI definition. A finger-tapping task should reliably engage the primary motor cortex, even at 3T, and even in a single participant.
• The BOLD weighted images in Figure 3 show a very clear double-peak pattern. This contradicts the results in Figure 2 and is unexpected given the existing literature on BOLD responses as a function of cortical depth.

In our study, the TE in Figure 2 is shorter than that in Figure 3 (33 ms versus 43 ms). It has been reported in the literature that BOLD fMRI with a shorter TE tends to include a greater intravascular contribution. Acknowledging this, we plan to repeat the experiments with a controlled TE to ensure consistency in our results.

• Given that data from Figures 2, 3, and 4 are derived from a single participant each, order and attention affects might have dramatically affected the observed patterns. Especially for Figure 4, neither BOLD nor VN profiles are really different from 0, and without statistical values or inter-subject averaging, these cannot be used to draw conclusions from.

The order of the experiments were randomized to ensure unbiased results.

It is important to note that the error bars presented in Figures 2, 3, and 4 do not represent the standard deviation of the residual fitting error. Instead, they illustrate the variation across voxels within a specific layer. This approach may lead to the error bars being influenced by the selection of the Region of Interest (ROI). In light of this, we intend to refine our statistical methodologies in the revised manuscript to address this issue.

• In Figure 5, a phase regression is added to the data presented in Figure 4. However, for a phase regression to work, there has to be a (macrovascular) response to start with. As none of the responses in Figure 4 are significant for the single participant dataset, phase regression should probably not have been undertaken. In this case, the functional 'responses' appear to increase with phase regression, which is contra-intuitive and deserves an explanation.
• Consistency of responses is indeed expected to increase by a removal of the more variable vascular component. However, the microvascular component is always expected to be smaller than the combination of microvascular + macrovascular responses. Note that the use of %signal changes may obscure this effect somewhat because of the modified baseline. Another expected feature of BOLD profiles containing both micro- and microvasculature is the draining towards the cortical surface. In the profiles shown in Figure 7, this is completely absent. In the group data, no significant responses to the task are shown anywhere in the cortical ribbon.
• Although I'd like to applaud the authors for their ambition with the connectivity analysis, I feel that acquisitions that are so SNR starved as to fail to show a significant response to a motor task should not be used for brain wide directed connectivity analysis.

We agree that exploring brain-wide directed functional connectivity may be overly ambitious at this stage, particularly before the VN-fMRI technique has been comprehensively evaluated and validated. In the revised manuscript, we will focus more on examining the characteristics of the layer-dependent BOLD signal rather than delving into layer-dependent functional connectivity.

Reviewer #1 (Public Review):

Batra, Cabrera, Spence et al. present a model which integrates histone posttranslational modification (PTM) data across cell models to predict gene expression with the goal of using this model to better understand epigenetic editing. This gene expression prediction model approach is useful if a) it predicts gene expression in specific cell lines b) it predicts expression values rather than a rank or bin, c) it helps us to better understand the biology of gene expression, or d) it helps us to understand epigenome editing activity. Problematically for points a) and b) it is easier to directly measure gene expression than to measure multiple PTMs and so the real usefulness of this approach mostly relates to c) and d).

Other approaches have been published that use histone PTM to predict expression (e.g. 27587684, 36588793). Is this model better in some way? No comparisons are made. The paper does not seem to have substantial novel insights into understanding the biology of gene expression. The approach of using this model to predict epigenetic editor activity on transcription is interesting and to my knowledge novel but I doubt given the variability of the predictions (Figures 6 and S7&8) that many people will be interested in using this in a practical sense. As the authors point out, the interpretation of the epigenetic editing data is convoluted by things like sgRNA activity scoring and to fully understand the results likely would require histone PTM profiling and maybe dCas9 ChIP-seq for each sgRNA which would be a substantial amount of work.

Furthermore from the model evaluation of H3K9me3 it seems the model is not performing well for epigenetic or transcriptional editing- e.g. we know for the best studied transcriptional editor which is CRISPRi (dCas9-KRAB) that recruitment to a locus is associated with robust gene repression across the genome and is associated with H3K9me3 deposition by recruitment of KAP1/HP1/SETDB1 (PMID: 35688146, 31980609, 27980086, 26501517). However, it seems from Figures 2&4 that the model wouldn't be able to evaluate or predict this.

The model seems to predict gene expression for endogenous genes quite well although the authors sometimes use expression and sometimes use rank (e.g. Figure 6) - being clearer with how the model predicts expression rather than using rank or fold change would be very useful.

One concern overall with this approach is that dCas9-p300 has been observed to induce sgRNA-independent off-target H3K27Ac (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8349887/ see Figure S5D) which could convolute interpretation of this type of experiment for the model.

Figure 2<br /> It seems this figure presents known rather than novel findings from the authors' description. Please comment on whether there are any new findings in this figure. Please comment on differences in patterns of repressive and activating histone PTMs between cell lines (e.g. H1-Esc H3K27me3 green 25-50% is more enriched than red 0-25%).

Figure 3&4<br /> There are a number of approaches including DeepChrome and TransferChrome that predict endogenous gene expression from histone PTMs. I appreciate that the authors have not used the histone PTM data to predict gene expression levels of an "average cell" but rather that they are predicting expression within specific cell types or for unseen cell types. But from what is presented it isn't clear that the author's model is better or enabling beyond other approaches. The authors should show their model is better than other approaches or make clear why this is a significant advance that will be enabling for the field. For example is it that in this approach they are actually predicting expression levels whereas previous approaches have only predicted expressed or not expressed or a rank order or bin-based ranking?

Figure 5<br /> From the methods, it seems gene activation is measured by qpcr in hek293 transfected with individual sgRNAs and dCas9-p300. The cells aren't selected or sorted before qPCR so how are we sure that some of the variability isn't due to transfection efficiency associated with variable DNA quality or with variable transfection efficiency?

Figure 6<br /> The use of rank in 6D and 6E is confusing. In 6D a higher rank is associated with higher expression while in 6E a higher rank seems to mean a lower fold change e.g. CYP17A1 has a low predicted fold-change rank and qPCR fold-change rank but in Figure 5 a very high qPCR fold change. Labeling this more clearly or explaining it in the text further would be useful.

Reviewer #2 (Public Review):

Summary:<br /> The authors build a gene expression model based on histone post-translational modifications and find that H3K27ac is correlated with gene expression. They proceed to perturb H3K27ac at 8 gene promoters, and measure gene expression changes to test their model.

Strengths:<br /> The combination of multiple methods to model expression, along with utilizing 6 histone datasets in 13 cell types allowed the authors to build a model that correlates between 0.7-0.79 with gene expression. This group also utilized a tool they are experts in, dCas9-p300 fusions to perturb H3K27ac and monitor gene expression to test their model. Ranked correlations showed some support for the predictions after the perturbation of H3K27ac.

Weaknesses:<br /> The perturbation of only 8 genes, and the only readout being qPCR-based gene expression, as opposed to including H3K27ac, weakened their validation of the computational model. Likewise, the use of six genes that were not expressed being most activated by dCas9-p300 might weaken the correlations vs. looking at a broad range of different gene expressions as the original model was trained on.

Reviewer #1 (Public Review):

Strengths:<br /> The authors first perform several important controls to show that the expressed mutant actin is properly folded, and then show that the Arp2/3 complex behaves similarly with WT and mutant actin via a TIRF microscopy assay as well as a bulk pyrene-actin assay. A TIRF assay showed a small but significant reduction in the rate of elongation of the mutant actin suggesting only a mild polymerization defect.

Based on in silico analysis of the close location of the actin point mutation and bound cofilin, cofilin was chosen for further investigation. Faster de novo nucleation by cofilin was observed with mutant actin. In contrast, the mutant actin was more slowly severed. Both effects favor the retention of filamentous mutant actin. In solution, the effect of cofilin concentration and pH was assessed for both WT and mutant actin filaments, with a more limited repertoire of conditions in a TIRF assay that directly showed slower severing of mutant actin.

Lastly, the mutated residue in actin is predicted to interact with the cardiomyopathy loop in myosin and thus a standard in vitro motility assay with immobilized motors was used to show that non-muscle myosin 2A moved mutant actin more slowly, explained in part by a reduced affinity for the filament deduced from transient kinetic assays. By the same motility assay, myosin 5A also showed impaired interaction with the mutant filaments.

The Discussion is interesting and concludes that the mutant actin will co-exist with WT actin in filaments, and will contribute to altered actin dynamics and poor interaction with relevant myosin motors in the cellular context. While not an exhaustive list of possible defects, this is a solid start to understanding how this mutation might trigger a disease phenotype.

Weaknesses:<br /> Potential assembly defects of the mutant actin could be more thoroughly investigated if the same experiment shown in Fig. 2 was repeated as a function of actin concentration, which would allow the rate of disassembly and the critical concentration to also be determined.

The more direct TIRF assay for cofilin severing was only performed at high cofilin concentration (100 nM). Lower concentrations of cofilin would also be informative, as well as directly examining by the TIRF assay the effect of cofilin on filaments composed of a 50:50 mixture of WT:mutant actin, the more relevant case for the cell.

The more appropriate assay to determine the effect of the actin point mutation on class 5 myosin would be the inverted assay where myosin walks along single actin filaments adhered to a coverslip. This would allow an evaluation of class 5 myosin processivity on WT versus mutant actin that more closely reflects how Myo5 acts in cells, instead of the ensemble assay used appropriately for myosin 2.

Reviewer #2 (Public Review):

Greve et al. investigated the effects of a disease-associated gamma-actin mutation (E334Q) on actin filament polymerization, association of selected actin-binding proteins, and myosin activity. Recombinant wildtype and mutant proteins expressed in sf9 cells were found to be folded and stable, and the presence of the mutation altered a number of activities. Given the location of the mutation, it is not surprising that there are changes in polymerization and interactions with actin binding proteins. Nevertheless, it is important to quantify the effects of the mutation to better understand disease etiology. Some weaknesses were identified in the paper as discussed below.

Throughout the paper, the authors report average values and the standard-error-of-the-mean (SEM) for groups of three experiments. Reporting the SEM is not appropriate or useful for so few points, as it does not reflect the distribution of the data points. When only three points are available, it would be better to just show the three different points. Otherwise, plot the average and the range of the three points.

The description and characterization of the recombinant actin is incomplete. Please show gels of purified proteins. This is especially important with this preparation since the chymotrypsin step could result in internally cleaved proteins and altered properties, as shown by Ceron et al (2022). The authors should also comment on N-terminal acetylation of actin.

The authors do not use the best technique to assess actin polymerization parameters. Although the TIRF assay is excellent for some measurements, it is not as good as the standard pyrene-actin assays that provide critical concentration, nucleation, and polymerization parameters. The authors use pyrene-actin in other parts of the paper, so it is not clear why they don't do the assays that are the standard in the actin field.

The authors' data suggest that, while the binding of cofilin-1 to both the WT and mutant actins remains similar, the major defect of the E334Q actin is that it is not as readily severed/disassembled by cofilin. What is missing is a direct measurement of the severing rate (number of breaks per second) as measured in TIRF.

Figure 4 shows that the E334Q mutation increases rather than decreases the number of filaments that spontaneously assemble in the TIRF assay, but it is unclear how reduced severing would lead to increased filament numbers, rather, the opposite would be expected. A more straightforward approach would be to perform experiments where severing leads to more nuclei and therefore enhances the net bulk assembly rate.

Figure 5 A: in the pyrene disassembly assay, where actin is diluted below its critical concentration, cofilin enhances the rate of depolymerization by generating more free ends. The E334Q mutation leads to decreased cofilin-induced severing and therefore lower depolymerization. While these data seem convincing, it would be better to present them as an XY plot and fit the data to lines for comparison of the slopes.

Figure 5 B and C: the cosedimentation data do not seem to help elucidate the underlying mechanism. While the authors report statistical significance, differences are small, especially for gel densitometry measurements where the error is high, which suggests that there may be little biological significance. Importantly, example gels from these experiments should be shown, if not the complete set included in the supplement. In B, the higher cofilin concentrations would be expected to stabilize the filaments and thus the curve should be U-shaped.

Figure 5 D: these data show that the binding of cofilin to WT and E334Q actin is approximately the same, with the mutant binding slightly more weakly. It would be clearer if the two plots were normalized to their respective plateaus since the difference in arbitrary units distracts from the conclusion of the figure. If the difference in the plateaus is meaningful, please explain.

Figure 6: It is assumed that the authors are trying to show in this figure that cofilin binds both actins approximately the same but does not sever as readily for E334Q actin. The numerous parameters measured do not directly address what the authors are actually trying to show, which presumably is that the rate of severing is lower for E334Q than WT. It is therefore puzzling why no measurement of severing events per second per micron of actin in TIRF is made, which would give a more precise account of the underlying mechanism.

Actin-activated steady-state ATPase data of the NM2A with mutant and WT actin would have been extremely useful and informative. The authors show the ability to make these types of measurements in the paper (NADH assay), and it is surprising that they are not included for assessing the myosin activity. It may be because of limited actin quantities. If this is the case, it should be indicated.

(line 310) The authors state that they "noticed increased rapid dissociation and association events for E334Q filaments" in the motility assay. This observation motivates the authors to assess actin affinities of NM2A-HMM. Although differences in rigor and AM.ADP affinities are found between mutant and wt actins, the actin attachment lifetimes (many minutes) are unlikely to be related to the rapid association and dissociation event seen in the motility assay. Rather, this jiggling is more likely to be related to a lower duty ratio of the myosins, which appears to be the conclusion reached for the myosin-V data. These points should be clarified in the text.

(line 327) The authors report that the 1/K1 value is unchanged. There are no descriptions of this experiment in the paper. I am assuming the authors measured the ATP-induced dissociation of actomyosin and determined ATP affinity (K1) from this experiment. If this is the case, they should describe the experiment and show the data, provide a second-order rate constate for ATP binding, and report the max rate of dissociation (k2). This is a kinetic experiment done frequently by this group, so the absence of these details is surprising.

eLife assessment

This important study contributes insights into the regulatory mechanisms of a protein governing cell migration at the membrane. The integration of approaches revealing protein structure and dynamics provides convincing data for a model of regulation and suggests a new allosteric role for a solubilized phospholipid headgroup. The work will be interesting to researchers focusing on signaling mechanisms, cell motility, and cancer metathesis.

Reviewer #1 (Public Review):

Summary:<br /> The authors perform a multidisciplinary approach to describe the conformational plasticity of P-Rex1 in various states (autoinhibited, IP4 bound, and PIP3 bound). Hydrogen-deuterium exchange (HDX) is used to reveal how IP4 and PIP3 binding affect intramolecular interactions. While IP4 is found to stabilize autoinhibitory interactions, PIP3 does the opposite, leading to deprotection of autoinhibitory sites. Cryo-EM of IP4 bound P-Rex1 reveals a structure in the autoinhibited conformation, very similar to the unliganded structure reported previously (Chang et al. 2022). Mutations at observed autoinhibitory interfaces result in a more open structure (as shown by SAXS), reduced thermal stability, and increased GEF activity in biochemical and cellular assays. Together their work portrays a dynamic enzyme that undergoes long-range conformational changes upon activation on PIP3 membranes. The results are technically sound (apart from a few points mentioned below) and the conclusions are justified. The main drawback is the limited novelty due to the recently published structure of unliganded P-Rex1, which is virtually identical to the IP4 bound structure presented here. Novel aspects suggest a regulatory role for IP4, but the exact significance and mechanism of this regulation have not been explored.

Strengths:<br /> The authors use a multitude of techniques to describe the dynamic nature and conformational changes of P-Rex1 upon binding to IP4 and PIP3 membranes. The different approaches together fit well with the overall conclusion that IP4 binding negatively regulates P-Rex1, while binding to PIP3 membranes leads to conformational opening and catalytic activation. The experiments are performed very thoroughly and are technically sound (apart from a few comments mentioned below). The results are clear and support the conclusions.

Weaknesses:<br /> 1) The novelty of the study is compromised due to the recently published structure of unliganded P-Rex1 (Chang et al. 2022). The unliganded and IP4-bound structure of P-Rex1 appear virtually identical, however, no clear comparison is presented in the manuscript. In the same paper, a very similar model of P-Rex1 activation upon binding to PIP3 membranes and Gbeta/gamma is presented.

2) The authors demonstrate that IP4 binding to P-Rex1 results in catalytic inhibition and increased protection of autoinhibitory interfaces, as judged by HDX. The relevance of this in a cellular setting is not clear and is not experimentally demonstrated. Further, mechanistically, it is not clear whether the biochemical inhibition by IP4 of PIP3 activated P-Rex1 is due to competition of IP4 with activating PIP3 binding to the PH domain of P-Rex1, or due to stabilizing the autoinhibited conformation, or both.

3) It is difficult to judge the error in the HDX experiments presented in Sup. data 1 and 2. In the method section, it is stated that the results represent the average from two samples. How is the SD error calculated in Fig.1B-C?

Reviewer #2 (Public Review):

Summary:<br /> In this new paper, the authors used biochemical, structural, and biophysical methods to elucidate the mechanisms by which IP4, the PIP3 headgroup, can induce an autoinhibit form of P-Rex1 and propose a model of how PIP3 can trigger long-range conformational changes of P-Rex1 to relieve this autoinhibition. The main findings of this study are that a new P-Rex1 autoinhibition is driven by an IP4-induced binding of the PH domain to the DH domain active site and that this autoinhibited form is stabilized by two key interactions between DEP1 and DH and between PH and IP4P 4-helix bundle (4HB) subdomain. Moreover, they found that the binding of phospholipid PIP3 to the PH domain can disrupt these interactions to relieve P-Rex1 autoinhibition.

Strengths:<br /> The study provides good evidence that binding of IP4 to the P-Rex1 PH domain can make the two long-range interactions between the catalytic DH domain and the first DEP domain and between the PH domain and the C-terminal IP4P 4HB subdomain that generate a novel P-Rex1 autoinhibition mechanism. This valuable finding adds an extra layer of P-Rex1 regulation (perhaps in the cytoplasm) to the synergistic activation by phospholipid PIP3 and the heterotrimeric Gbeta/gamma subunits at the plasma membrane. Overall, this manuscript's goal sounds interesting, the experimental data were carried out carefully and reliably.

Reviewer #3 (Public Review):

Summary:<br /> In this report, Ravala et al demonstrate that IP4, the soluble head-group of phosphatiylinositol 3,4,5 - trisphosphate (PIP3), is an inhibitor of pREX-1, a guanine nucleotide exchange factor (GEF) for Rac1 and related small G proteins that regulate cell migration. This finding is perhaps unexpected since pREX-1 activity is PIP3-dependent. By way of Cryo-EM (revealing the structure of the p-REX-1/IP4 complex at 4.2Å resolution), hydrogen-deuterium mass spectrometry, and small angle X-ray scattering, they deduce a mechanism for IP4 activation, and conduct mutagenic and cell-based signaling assays that support it. The major finding is that IP4 stabilizes two interdomain interfaces that block access to the DH domain, which conveys GEF activity towards small G protein substrates. One of these is the interface between the PH domain that binds to IP4 and a 4-helix bundle extension of the IP4 Phosphatase domain and the DEP1 domain. The two interfaces are connected by a long helix that extends from PH to DEP1. Although the structure of fully activated pREX-1 has not been determined, the authors propose a "jackknife" mechanism, similar to that described earlier by Chang et al (2022) (referenced in the author's manuscript) in which binding of IP3 relieves a kink in a helix that links the PH/DH modules and allows the DH-PH-DEP triad to assume an extended conformation in which the DH domain is accessible. While the structure of the activated pREX-1 has not been determined, cysteine mutagenesis that enforces the proposed kink is consistent with this hypothesis. SAXS and HDX-MS experiments suggest that IP4 acts by stiffening the inhibitory interfaces, rather than by reorganizing them. Indeed, the cryo-EM structure of ligand-free pREX-1 shows that interdomain contacts are largely retained in the absence of IP4.

Strengths:<br /> The manuscript thus describes a novel regulatory role for IP4 and is thus of considerable significance to our understanding of regulatory mechanisms that control cell migration, particularly in immune cell populations. Specifically, they show how the inositol polyphosphate IP4 controls the activity of pREX-1, a guanine nucleotide exchange factor that controls the activity of small G proteins Rac and CDC42 . In their clearly written discussion, the authors explain how PIP3, the cell membrane, and the Gbeta-gamma subunits of heterotrimeric membranes together localize pREX-1 at the membrane and induce activation. The quality of experimental data is high and both in vitro and cell-based assays of site-directed mutants designed to test the author's hypotheses are confirmatory. The results strongly support the conclusions. The combination of cryo-EM data, that describe the static (if heterogeneous) structures with experiments (small angle x-ray scattering and hydrogen-deuterium exchange-mass spectrometry) that report on dynamics are well employed by the authors

Weaknesses:<br /> There are a few weaknesses. While the resolution of the cryo-EM structure is modest, it is sufficient to identify the domain-domain interactions that are mechanistically important, since higher-resolution structures of various pREX-1 modules are available.

eLife assessment

In this valuable manuscript, Yao et al. describe new methods for assessing the intracellular itinerary of Botulinum neurotoxin A (BoNT/A), a potent toxin used in clinical and cosmetic applications. The current manuscript challenges previously held views on how the catalytic portion of the toxin makes its way from the endocytic compartment to the cytosol, to meet its substrates. The approach taken is deemed innovative and the experiments are carefully performed, however, they are somewhat incomplete with respect to the drawn conclusions, as it is possible that the scope of their findings could be restricted to the specific neuron model and molecular tools that were used. This paper could be of interest to both cell biologists and physicians.

Reviewer #1 (Public Review):

Summary:<br /> During the last decades, extensive studies (mostly neglected by the authors), using in vitro and in vivo models, have elucidated the five-step mechanism of intoxication of botulinum neurotoxins (BoNTs). The binding domain (H chain) of all serotypes of BoNTs binds polysialogangliosides and the luminal domain of a synaptic vesicle protein (which varies among serotypes). When bound to the synaptic membrane of neurons, BoNTs are rapidly internalized by synaptic vesicles (SVs) via endocytosis. Subsequently, the catalytic domain (L chain) translocates, a process triggered by the acidification of these organelles. Following translocation, the disulfide bridge connecting the H chain with the L chain is reduced by the thioredoxin reductase/thioredoxin system, and it is refolded by the chaperone Hsp90 on SV's surface. Once released into the cytosol, the L chains of different serotypes cleave distinct peptide bonds of specific SNARE proteins, thereby disrupting neurotransmission.

In this study, Yeo et al. extensively revise the neuronal intoxication model, suggesting that BoNT/A follows a more complex intracellular route than previously thought. The authors propose that upon internalization, BoNT/A-containing endosomes are retro-axonally trafficked to the soma. At the level of the neuronal soma, this serotype then traffics to the endoplasmic reticulum (ER) via the Golgi apparatus. The ER SEC61 translocon complex facilitates the translocation of BoNT/A's LC from the ER lumen into the cytosol, where the thioredoxin reductase/thioredoxin system and HSP complexes release and refold the catalytic L chain. Subsequently, the L chain diffuses and cleaves SNAP25 first in the soma before reaching neurites and synapses.

Strengths:<br /> I appreciate the authors' efforts to confirm that the newly established methods somehow recapitulate aspects of the BoNTs mechanism of action, such as toxin binding and uptake occurring at the level of active synapses. Furthermore, even though I consider the SNAPR approach inadequate, the genome-wide RNAi screen has been well executed and thoroughly analyzed. It includes well-established positive and negative controls, making it a comprehensive resource not only for scientists working in the field of botulinum neurotoxins but also for cell biologists studying endocytosis more broadly.

Weaknesses:<br /> I have several concerns about the authors' main conclusions, primarily due to the lack of essential controls and validation for the newly developed methods used to assess toxin cleavage and trafficking into neurons. Furthermore, there is a significant discrepancy between the proposed intoxication model and existing studies conducted in more physiological settings. In my opinion, the authors have omitted over 20 years of work done in several labs worldwide (Montecucco, Montal, Schiavo, Rummel, Binz, etc.). I want to emphasize that I support changes in biological dogma only when these changes are supported by compelling experimental evidence, which I could not find in the present manuscript.

Reviewer #2 (Public Review):

Summary:<br /> The study by Yeo and co-authors addresses a long-lasting issue about botulinum neurotoxin (BoNT) intoxication. The current view is that the toxin binds to its receptors at the axon terminus by its HCc domain and is internalized in recycled neuromediator vesicles just after the release of the neuromediators. Then, the HCn domain assists the translocation of the catalytic light chain (LC) of the toxin through the membrane of these endocytic vesicles into the cytosol of the axon terminus. There, the LC cleaves its SNARE substrate and blocks neurosecretion. However, other views involving kinetic aspects of intoxication suggest that the toxin follows the retrograde axonal transport up to the nerve cell body and then back to the nerve terminus before cleaving its substrate.

In the current study, the authors claim that the BoNT/A (isotype A of BoNT) not only progresses to the cell body but once there, follows the retrograde transport trafficking pathway in a retromer-dependent fashion, through the Golgi apparatus, until reaching the endoplasmic reticulum. Next, the LC dissociates from the HC (a process not studied here) and uses the translocon Sec61 machinery to retro-translocate into the cytosol. Only then, does the LC traffic back to the nerve terminus following the anterograde axonal transport. Once there, LC cleaves its SNARE substrate (SNAP25 in the case of BoTN/A) and blocks neurosecretion.

To reach their conclusion, Yeo and co-authors use a combination of engineered tools: a cell line able to differentiate into neurons (ReNcell VN), a reporter dual fluorescent protein derived from SNAP25, the substrate of BoNT/A (called SNAPR), the use of either native BoNT/A or a toxin to which three fragment 11 of the reporter fluorescent protein Neon Green (mNG) are fused to the N-terminus of the LC (BoNT/A-mNG11x3), and finally ReNcell VN transfected with mNG1-10 (a protein consisting of the first 10 beta strands of the mNG).

SNAPR is stably expressed all over in the ReNcell VN. SNAPR is yellow (red and green) when intact and becomes red only when cleaved by BoNT/A LC, the green tip being degraded by the cell. When the LC of BoNT/A-mNG11x3 reaches the cytosol in ReNcell VN transfected by mNG1-10, the complete mNG is reconstituted and emits a green fluorescence.

In the first experiment, the authors show that the catalytic activity of the LC appears first in the cell body of neurons where SNAPR is cleaved first. This phenomenon starts 24 hours after intoxication and progresses along the axon towards the nerve terminus during an additional 24 hours. In a second experiment, the authors intoxicate the ReNcell VN transfected by mNG1-10 using the BoNT/A-mNG11x3. The fluorescence appears also first in the soma of neurons, then diffuses in the neurites in 48 hours. The conclusion of these two experiments is that translocation occurs first in the cell body and that the LC diffuses in the cytosol of the axon in an anterograde fashion.

In the second part of the study, the authors perform a siRNA screen to identify regulators of BoNT/A intoxication. Their aim is to identify genes involved in intracellular trafficking of the toxin and translocation of the LC. Interestingly, they found positive and negative regulators of intoxication. Regulators could be regrouped according to the sequential events of intoxication. Genes affecting binding to the cell-surface receptor (SV2) and internalization. Genes involved in intracellular trafficking. Genes involved in translocation such as reduction of the disulfide bond linking the LC to the HC and refolding in the cytosol. Genes involved in signaling such as tyrosine kinases and phosphatases. All these groups of genes may be consistent with the current view of BoNT intoxication within the nerve terminus. However, two sets of genes were particularly significant to reach the main conclusion of the work and definitely constitute an original finding important to the field. One set of genes consists of those of the retromer, and the other relates to the Sec61 translocon. This should indicate that once endocytosed, the BoNT traffics from the endosomes to the Golgi apparatus, and then to the ER. Ultimately, the LC should translocate from the ER lumen to the cytosol using the Sec61 translocon. The authors further control that the SV2 receptor for the BoNT/A traffics along the axon in a retromer-dependent fashion and that BoNT/A-mNG11x3 traverses the Golgi apparatus by fusing the mNG1-10 to a Golgi resident protein.

Strengths:<br /> The findings in this work are convincing. The experiments are carefully done and are properly controlled. In the first part of the study, both the activity of the LC is monitored together with the physical presence of the toxin. In the second part of the work, the most relevant genes that came out of the siRNA screen are checked individually in the ReNcell VN / BoNT/A reporter system to confirm their role in BoNT/A trafficking and retro-translocation.

These findings are important to the fields of toxinology and medical treatment of neuromuscular diseases by BoNTs. They may explain some aspects of intoxication such as slow symptom onset, aggravation, and appearance of central effects.

Weaknesses:<br /> The findings antagonize the current view of the intoxication pathway that is sustained by a vast amount of observations. The findings are certainly valid, but their generalization as the sole mechanism of BoNT intoxication should be tempered. These observations are restricted to one particular neuronal model and engineered protein tools. Other models such as isolated nerve/muscle preparations display nerve terminus paralysis within minutes rather than days. Also, the tetanus neurotoxin (TeNT), whose mechanism of action involving axonal transport to the posterior ganglia in the spinal cord is well described, takes between 5 and 15 days. It is thus possible that different intoxication mechanisms co-exist for BoNTs or even vary depending on the type of neurons.

Although the siRNA experiments are convincing, it would be nice to reach the same observations with drugs affecting the endocytic to Golgi to ER transport (such as Retro-2, golgicide or brefeldin A) and the Sec61 retrotranslocation (such as mycolactone). Then, it would be nice to check other neuronal systems for the same observations.

Reviewer #3 (Public Review):

Summary:<br /> The manuscript by Yao et al. investigates the intracellular trafficking of Botulinum neurotoxin A (BoNT/A), a potent toxin used in clinical and cosmetic applications. Contrary to the prevailing understanding of BoNT/A translocation into the cytosol, the study suggests a retrograde migration from the synapse to the soma-localized Golgi in neurons. Using a genome-wide siRNA screen in genetically engineered neurons, the researchers identified over three hundred genes involved in this process. The study employs organelle-specific split-mNG complementation, revealing that BoNT/A traffics through the Golgi in a retromer-dependent manner before moving to the endoplasmic reticulum (ER). The Sec61 complex is implicated in the retro-translocation of BoNT/A from the ER to the cytosol. Overall, the research challenges the conventional model of BoNT/A translocation, uncovering a complex route from synapse to cytosol for efficient intoxication. The findings are based on a comprehensive approach, including the introduction of a fluorescent reporter for BoNT/A catalytic activity and genetic manipulations in neuronal cell lines. The conclusions highlight the importance of retrograde trafficking and the involvement of specific genes and cellular processes in BoNT/A intoxication.

Strengths:<br /> The major part of the experiments are convincing. They are well-controlled and the interpretation of their results is balanced and sensitive.

Weaknesses:<br /> To my opinion, the main weakness of the paper is in the interpretation of the data equating loss of tGFP signal (when using the Red SNAPR assay) with proteolytic cleavage by the toxin. Indeed, the first step for loss of tGFP signal by degradation of the cleaved part is the actual cleavage. However, this needs to be degraded (by the proteasome, I presume), a process that could in principle be affected (in speed or extent) by the toxin.

eLife assessment

This important study reports on an improved deep-learning-based method for predicting TCR specificity. The evidence supporting the overall method is compelling, although the inclusion of real-world applications and clear comparisons with the previous version would have further strengthened the study. This work will be of broad interest to immunologists and computational biologists.

Reviewer #1 (Public Review):

In this article, different machine learning models (pan-specific, peptide-specific, pre-trained, and ensemble models) are tested to predict TCR-specificity from a paired-chain peptide-TCR dataset. The data consists of 6,358 positive observations across 26 peptides (as compared to six peptides in NetTCR version 2.1) after several pre-processing steps (filtering and redundancy reduction). For each positive sample, five negative samples were generated by swapping TCRs of a given peptide with TCRs binding to other peptides. The weighted loss function is used to deal with the imbalanced dataset in pan-specific models.

The results demonstrate that the redundant data introduced during training did not lead to performance gain; rather, a decrease in performance was observed for the pan-specific model. The removal of outliers leads to better performance.<br /> <br /> To further improve the peptide-specific model performance, an architecture is created to combine pan-specific and peptide-specific models, where the pan-specific model is trained on pan-specific data while keeping the peptide-specific part of the model frozen, and the peptide-specific model is trained on a peptide-specific dataset while keeping the pan-specific part of the model frozen. This model surpassed the performance of individual pan-specific and peptide-specific models. Finally, sequence similarity-based predictions of TCRbase are integrated into the pre-trained CNN model, which further improved the model performance (mostly due to the better discrimination of binders and non-binders).<br /> <br /> The prediction for unseen peptides is still low in a pan-specific model; however, an improvement in prediction is observed for peptides with high similarity to the ones in the training dataset. Furthermore, it is shown that 15 observations show satisfactory performance as compared to the ~150 recommended in the literature.<br /> <br /> Models are evaluated on the external dataset (IMMREP benchmark). Peptide-specific models performed competitively with the best models in the benchmark. The pre-trained model performed worst, which the authors suggested could be because of positive and negative sample swapping across training and testing sets. To resolve this issue, they applied the redundancy removal technique to the IMMER dataset. The results agreed with the earlier conclusion that the pre-trained models surpassed peptide-specific models and the integration of similarity-based methods leads to performance boost. It highlights the need for the creation of a new benchmark without data redundancy or leakage problems.

The manuscript is well-written, clear, and easy to understand. The data is effectively presented. The results validate the drawn conclusions.

Reviewer #2 (Public Review):

Summary:<br /> The authors describe a novel ML approach to predict binding between MHC-bound peptides and T-Cell receptors. Such approaches are particularly useful for predicting the binding of peptide sequences with low similarity when compared to existing data sets. The authors focus on improving dataset quality and optimizing model architecture to achieve a pan-specific predictive model in hopes of achieving a high-precision model for novel peptide sequences.

Strengths:<br /> Since assuring the quality of training datasets is the first major step in any ML training project, the extensive human curation, computational analysis, and enhancements made in this manuscript represent a major contribution to the field. Moreover, the systematic approach to testing redundancy reduction and data augmentation is exemplary, and will significantly help future research in the field.

The authors also highlight how their model can identify outliers and how that can be used to improve the model around known sequences, which can help the creation and optimization of future datasets for peptide binding.

The new models presented here are novel and built using paired α/β TCR sequence data to predict peptide-specific TCR binding, and have been extensively and rigorously tested.

Weaknesses:<br /> Achieving an accurate pan-specific model is an ambitious goal, and the authors have significant difficulties when trying to achieve non-random performance for the prediction of TCR binding to novel peptides. This is the most challenging task for this kind of model, but also the most desirable when applying such models to biotechnological and bioengineering projects.

The manuscript is a highly technical and extremely detailed computational work, which can make the achievements and impact of the work hard to parse for application-oriented researchers.<br /> The authors briefly mention real-world use cases for TCR specificity predictions, but do not contextualize the work into possible applications.

Joint Public Review:

Here, the authors compare how different operationalizations of adverse childhood experience exposure related to patterns of skin conductance response during a fear conditioning task. They use a large dataset to definitively understand a phenomenon that, to date, has been addressed using a range of different definitions and methods, typically with insufficient statistical power. Specifically, the authors compared the following operationalizations: dichotomization of the sample into "exposed" and "non-exposed" categories, cumulative adversity exposure, specificity of adversity exposure, and dimensional (threat versus deprivation) adversity exposure. The paper is thoughtfully framed and provides clear descriptions and rationale for procedures, as well as package version information and code. The authors' overall aim of translating theoretical models of adversity into statistical models, and comparing the explanatory power of each model, respectively, is an important and helpful addition to the literature. However, the analysis would be strengthened by employing more sophisticated modelling techniques that account for between-subjects covariates and the presentation of the data needs to be streamlined to make it clearer for the broad audience for which it is intended.

Strengths<br /> Several outstanding strengths of this paper are the large sample size and its primary aim of statistically comparing leading theoretical models of adversity exposure in the context of skin conductance response. This paper also helpfully reports Cohen's d effect sizes, which aid in interpreting the magnitude of the findings. The methods and results are generally thorough.

Weaknesses<br /> The largest concern is that the paper primarily relies on ANOVAs and pairwise testing for its analyses and does not include between-subjects covariates. Employing mixed-effects models instead of ANOVAs would allow more sophisticated control over sources of random variance in the sample (especially important for samples from multi-site studies such as the present study), and further allow the inclusion of potentially relevant between-subjects covariates such as age (e.g. Eisenstein et al., 1990) and gender identity or sex assigned at birth (e.g. Kopacz II & Smith, 1971) (perhaps especially relevant due to possible to gender or sex-related differences in ACE exposure; e.g. Kendler et al., 2001). Also, proxies for socioeconomic status (e.g. income, education) can be linked with ACE exposure (e.g. Maholmes & King, 2012) and warrant consideration as covariates, especially if they differ across adversity-exposed and unexposed groups. On a related methodological note, the authors mention that scores representing threat and deprivation were not problematically collinear due to VIFs being <10; however, some sources indicate that VIFs should be <5 (e.g. Akinwande et al., 2015).

Additionally, the paper reports that higher trait anxiety and depression symptoms were observed in individuals exposed to ACEs, but it would be helpful to report whether patterns of SCR were in turn associated with these symptom measures and whether the different operationalizations of ACE exposure displayed differential associations with symptoms. Given the paper's framing of SCR as a potential mechanistic link between adversity and mental health problems, reporting these associations would be a helpful addition. These results could also have implications for the resilience interpretation in the discussion (lines 481-485), which is a particularly important and interesting interpretation.

Given that the manuscript criticizes the different operationalizations of childhood adversity, there should be greater justification of the rationale for choosing the model for the main analyses. Why not the 'cumulative risk' or 'specificity' model? Related to this, there should also be a stronger justification for selecting the 'moderate' approach for the main analysis. Why choose to cut off at moderate? Why not severe, or low? Related to this, why did they choose to cut off at all? Surely one could address this with the continuous variable, as they criticize cut-offs in Table 2.

In the Introduction, the authors predict less discrimination between signals of danger (CS+) and safety (CS-) in trauma-exposed individuals driven by reduced responses to the CS+. Given the potential impact of their findings for a larger audience, it is important to give greater theoretical context as to why CS discrimination is relevant here, and especially what a reduction in response specifically to danger cues would mean (e.g. in comparison to anxiety, where safety learning is impacted).

Author Response

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

The authors developed computational models that capture the electrical and Ca2+ signaling behavior in mesenteric arterial cells from male and female mice. A baseline model was first formulated with eleven transmembrane currents and three calcium compartments. Sex-specific differences in the L-type calcium channel and two voltage-gated potassium channels were then tuned based on experimental measurements. To incorporate the stochastic ion channel openings seen in smooth muscle cells under physiological conditions, noise was added to the membrane potential and the sarcoplasmic Ca2+ concentration equations. Finally, the models were assembled into 1D vessel representations and used to investigate the tissue-level electrical response to an L-type calcium channel blocker.

Strengths:

A major strength of the paper is that the modeling studies were performed on three different scales: individual ionic currents, whole-cell, and 1D tissue. This comprehensive computational framework can help provide mechanistic insight into arterial myocyte function that might be difficult to achieve through traditional experimental methods.

The authors aimed to develop sex-specific computational models of mesenteric arterial myocytes and demonstrate their use in drug-testing applications. Throughout the paper, model behavior was both validated by experimental recordings and supported by previously published data. The main findings from the models suggested that sex-specific differences in membrane potential and Ca2+ handling are attributable to variability in the gating of a small number of voltage-gated potassium channels and L-type calcium channels. This variability contributes to a higher Ca2+ channel blocker sensitivity in female arterial vessels. Overall, the study successfully met the aims of the paper.

Thank you for your insightful review and for recognizing the strengths of our study. We appreciate your encouraging comment regarding our multi-scale approach. Indeed, we believe that by systematically connecting these scales—individual ionic currents, whole-cell, and 1D tissue—we can integrate and reconcile experimental and clinical data. We anticipate that this approach will not only provide mechanistic insights into arterial myocyte function that may not be easy to glean from traditional experimental methods but will also facilitate the translation of this information into the development of therapeutic interventions.

Weaknesses:

A main weakness of the paper, as addressed by the authors, is the simplicity of the 1D vessel model; it does not take into account various signaling pathways or interactions with other cell types which could impact smooth muscle electrophysiology.

Thank you for highlighting areas for improvement in our study. The strength of computational modeling lies in its iterative nature, allowing us to introduce and examine variables in a systematic manner. While our current model is simplified and does not contain all details, the modular nature of the build will allow continuous expansion to add the important elements described by the reviewer. We are enthusiastic about progressively enriching the model in subsequent studies, introducing signaling pathways in a step-by-step manner, and ensuring their validation with rigorous experimental data.

Another potential shortcoming is the use of mouse data for optimizing the model, as there could be discrepancies in signaling behavior that limit the translatability to human myocyte predictions.

We appreciate this important comment. Our model was parametrized using data from mouse mesenteric artery smooth muscle cells as initial proof of concept. Mouse arteries are a good representation of human arteries, as they have similar intravascular pressure-myogenic tone relationships, resting membrane potentials, and express similar ionic channels (e.g., CaV1.2, BK channels, RyRs, etc) (PMID: 28119464, PMID: 29070899, PMID: 23232643). In response to the reviewer, we have modified the discussion section of the manuscript to specifically note the mouse is not identical to the human but does share some common important features that make mice a good approximate model.

Reviewer #2 (Public Review):

In this study, Hernandez-Hernandez et al developed a gender-dependent mathematical model of arterial myocytes based on a previous model and new experimental data. The ionic currents of the model and its sex difference were formulated based on patch-clamp experimental data, and the model properties were compared with single-cell and tissue scale experimental results. This is a study that is of importance for the modeling field as well as for experimental physiology.

Thank you for the comment. In fact, we developed a model that incorporates sex-dependent differences that allowed for male and female models. It’s an important distinction as sex is a biological variable and gender is a self-ascribed characteristic.

Reviewer #3 (Public Review):

Summary:

This hybrid experimental/computational study by Hernandez-Hernandez sheds new light on sex-specific differences between male and female arterial myocytes from resistance arteries. The authors conduct careful experiments in isolated myocytes from male and female mice to obtain the data needed to parameterize sex-specific models of two important ionic currents (i.e., those mediated by CaV1.2 and KV2.1). Available experimental data suggest that KV1.5 channel currents from male and female myocytes are similar, but simulations conducted in the novel Hernandez-Hernandez sex-specific models provide a more nuanced view. This gives rise to the first of the authors' three key scientific claims: (1) In males, KV1.5 is the dominant current regulating membrane potential; whereas, in females, KV2.1 plays a primary role in voltage regulation. They further show that this (2) the latter distinction drives drive sex-specific differences in intracellular Ca2+ and cellular excitability. Finally, working with one-dimensional models comprising several copies of the male/female myocyte models linked by resistive junctions, they use simulations to (3) predict that the sensitivity of arterial smooth muscle to Ca2+ channel-blocking drugs commonly used to treat hypertension is heightened in female compared to male cells.

Strengths:

The Methodology is described in exquisite detail in straightforward language that will be easy to understand for most if not all peer groups working in computational physiology. The authors have deployed standard protocols (e.g., parameter fitting as described by Kernik et al., sensitivity analysis as described by Sobie et al.) and appropriate brief explanations of these techniques are provided. The manoeuvre used to represent stochastic effects on voltage dynamics is particularly clever and something I have not personally encountered before. Collectively, these strengthen the credibility of the model and greatly enrich the manuscript.

We appreciate your comment highlighting the robustness of our methodology. Your acknowledgment of our approach to represent stochastic effects on voltage dynamics is especially encouraging. Indeed, noise is a fundamental component of physiological systems, including in vascular myocytes

Broadly speaking, the Results section describes findings that robustly support the three key scientific claims outlined in my summary. While there is certainly room for further discussion of some nuanced points as outlined below, it is evident these experiments were carefully designed and carried out with care and intentionality. In the present version of the manuscript, there are a few figures in which experimental data is shown side-by-side with outputs from the corresponding models. These are an excellent illustration of the power of the authors' novel sex-specific computational simulation platform. I think these figures will benefit from some modest additional quantitative analysis to substantiate the similarities between experimental and computational data, but there is already clear evidence of a good match.

We sincerely appreciate your constructive feedback on the Results section. We have included additional quantitative analysis to substantiate the similarities between experimental and computational data. We agree with the reviewer that the suggestion on the potential value of a more quantitative assessment. As such we have updated the figure to include an in-depth analysis that provides greater insights and solidifies the power of our simulation predictions when compared to experimental results. A detailed analysis of the male and female data as well as the male and female simulations are summarized in the text as follows:

Baseline membrane potential is -40 mV in male myocytes compared to -30 mV. The frequency of hyperpolarization transients (THs) is 1 Hz in male and 2.5 Hz in female cells for the specific baseline membrane potential shown in Figure 5 A-B. In the range of membrane potentials from -50 mV to -30 mV the frequency increases from 1-2.8Hz which is identical to the experimental frequency range.

Areas for Improvement:

The authors used experimental data from a prior publication to calibrate their model of the BKCa current. As indicated in the manuscript, these data are for channel activity measured in a heterologous expression system (Xenopus oocytes). A similar principle applies to other major ion channels/pumps/etc. Is it possible there might be relevant sex-specific differences in these players as well? In the context of the present work, this feels like an important potential caveat to highlight, in case male/female differences in the activity of BKCa or other currents might influence model-predicted differences (e.g., the relative importance of KV1.5 and KV2.1). This should be discussed, and, if possible, related to the elegant sensitivity analysis presented in Fig. 5C (which shows, for example, that the models are relatively insensitive to variation in GBK).

We fully agree with the reviewer - an important caveat to highlight is the unknown sex-specific differences in all the other players regulating membrane potential and calcium signaling. While our initial assessments indicated that the contribution of BKCa channels to the total voltage-gated K+ current (IKvTOT) was small within the physiological range of -50 mV to -30 mV, further analysis of spontaneous transient outward currents revealed sex-specific variations. We have investigations underway to explore if BKCa channel expression and organization may be also sex-dependent.

The authors state that their model can be expanded to 2D/3D applications, "transitioning seamlessly from single-cell to tissue-level simulations". I would like to see more discussion of this. For example, given the modest complexity of the cell-scale model, how considerable would the computational burden be to implement a large network model of a subset of the human female or male arterial system? Are there sex-specific differences in vessel and/or network macro-structure that would need to be considered? How would this influence feasibility? Rather than a 1D cable as implemented here, I imagine a multi-scale implementation would involve the representation of myocytes wrapped around vessels. How would the behavior of such a system differ from the authors' presented work using a 1D representation of 100 myocytes coupled end-to-end? Could these differences partially explain why the traces in Fig. 8D are smoother than those in Fig. 8C? From my standpoint, discussing these points would enrich the paper.

We appreciate the reviewer’s thoughtful and forward-looking ideas! Indeed, we are very interested to extend the model to incorporate a number of these important items.

Our choice for the 1D cable model was driven by its anatomical relevance to the structure of third and fourth-order mesenteric arteries. These arteries possess a singular layer of vascular myocytes encircling the lumen in a cylindrical arrangement. When we conceptualize this structure as unrolled or viewed laterally, it aligns with a flat, rectangular form, closely paralleling our 1D cable implementation. One option is to expand this into a 2D representation by connecting multiple 1D cables together. Another option would be to connect the 1D cable end-to-end to create a ring to represent a cross section. While these approaches would appear to be different geometries, in either case, the dynamics will remain consistent because the cells comprising the tissue are the same. There is no propagating impulse (for example – although even then in a 2D homogenous tissue, a planar wave is identical in 1D), and the only effect will be an increase in electrotonic load (sink) from neighboring cells, which can readily be approximated in 1D by increasing coupling or modification of the boundary conditions.

We totally agree that future investigation should include exploration into the potential sex-specific differences in vessel and/or network macro-structure, as these factors may critically impact predictions and indeed the difference in traces observed between Fig. 8D and Fig. 8C may well involve “insulating” effects of vessel layers and interaction between various cell types and other structural factors. In particular, the contribution of endothelial cells in modulating membrane potential in vascular myocytes might be one such influential factor. In future studies, we are also keen to investigate blood flow regulation where a 3D configuration might become necessary.

The nifedipine data presented in Fig. 9 are quite compelling, and a nice demonstration of the potential power of the new models. How does this relate to what is known about the clinical male/female responses to nifedipine? Are there sex differences in drug efficacy?

Thank you for your comment regarding Fig. 9.

It is well known that sex-specific differences in pharmacokinetics and pharmacodynamics influence antihypertensive drug responses [PMID: 8651122., PMID: 22089536]. Previous studies, notably by Kloner et al., have illustrated this point quantitatively, highlighting a more pronounced diastolic BP response in women (91.4%) compared to men (83%) when treated with dihydropyridine-type channel blockers, such as amlodipine/nifedipine. Importantly, this distinction persisted even after adjusting for confounding factors such as baseline BP, age, weight, and dosage per kilogram [PMID: 8651122]. An interesting observation from Kajiwara et al. emphasizes that vasodilation-related adverse symptoms occur significantly more frequently in younger women (<50 years) compared to their male counterparts, suggesting a heightened sensitivity to dihydropyridine-type calcium channel blockers [PMID: 24728902].

While our findings resonate with clinical observations, a word of caution is in order. Our data suggest that, in the mouse model, nifedipine elicits distinct sex-specific effects. Importantly, future research should test the direct translatability and implications of these observations in human subjects.

Reviewer #1 (Recommendations For The Authors):

1. Cellular simulations with noise: It might be useful to also include in this section how noise was introduced specifically into the [Ca]SR equations.

We agree. The manuscript now includes an expanded explanation of how noise was incorporated into the model. This includes the addition of Equation 6 into section 2.4 "Cellular simulations with noise" to describe how noise was specifically integrated into the [Ca]SR equations. Please see LINE 355.

1. For equation 14, the description might be confusing. RCG and Ri are not explicitly included.

Thank you – this has been corrected.

1. In the paragraph starting with, "Having explored the regulation of graded membrane potential..." , the references to Figure 7C-D do not seem to match the content of the text. Namely, the figures show female versus male responses to nifedipine, which is not introduced until the next paragraph. Additionally, the graphs in 7C-D do not have the panels titled and the y-axes labeled.

We apologize for the error. We have modified the text and figures to address these issues.

1. Perhaps give more detail on how the effects of nifedipine were mathematically simulated at the ionic current level.

Good suggestion. Briefly, previous studies [PMID: 1329564] have shown that at the therapeutic dose of nifedipine (i.e., about 0.1 μM) L-type Cav1.2 channel currents are reduced by about 70%. Accordingly, we decreased ICaL in our mathematical simulations by the same extent. It is known that dihydropyridine-type channel blockers exhibit a voltage-dependent behavior, predominantly binding to the inactivated state. In smooth muscle cells, these blockers initiate inhibition quickly within a voltage range of -60 to -40 mV. This range aligns with the membrane potential baseline of vascular muscle cells (PMID: 8388295), ensuring the blockers are effective without the need of inducing significant depolarization. Therefore, the voltage dependency of dihydropyridine-type channel blockers can be neglected.

1. For the simulations with 400 uncoupled myocytes, the methods stated that the "gap junctional resistance [was set] to zero". Did the authors mean to use "conductivity" or am I misunderstanding?

Thank you for bringing up this issue with the term "gap junctional resistance." We now state that the "gap junctional conductivity" was set to zero to indicate no electrical communication/coupling.

1. Address whether there are differences-such as in cell geometry, degree of sex-based ionic current changes, and frequency of spontaneous hyperpolarization-between mice and human smooth muscle myocytes that could limit the predictive capability of the model.

Excellent point. Our model was parametrized using data from mouse mesenteric artery smooth muscle cells as initial proof of concept. In general terms, mouse arteries are a good animal model for human arteries, as they have similar intravascular pressure-myogenic tone relationships, resting membrane potentials, and express similar ionic channel (e.g., CaV1.2, BK channels, RyRs, etc) (PMID: 28119464, PMID: 29070899). Unfortunately, these studies have largely been done in male arteries and myocytes. Thus, while we recognize that the physiological distinctions between mice and humans could introduce variances in the model's outcomes. Our model offers valuable insights into the sex-specific mechanisms of KV2.1 and CaV1.2 channels in controlling membrane potential and Ca2+ dynamics in mice. It has been shown that sex-specific differences in pharmacokinetics and pharmacodynamics influence antihypertensive drug responses [[PMID: 8651122., PMID: 22089536]. Previous studies, notably by Kloner et al., have illustrated this point quantitatively, highlighting a more pronounced diastolic BP response in women (91.4%) compared to men (83%) when treated with dihydropyridine-type channel blockers, such as amlodipine/nifedipine. Importantly, this distinction persisted even after adjusting for confounding factors such as baseline BP, age, weight, and dosage per kilogram [PMID: 8651122]. An interesting observation from Kajiwara et al. emphasizes that vasodilation-related adverse symptoms occur significantly more frequently in younger women (<50 years) compared to their male counterparts, suggesting a heightened sensitivity to dihydropyridine-type calcium channel blockers [PMID: 24728902].

While our findings resonate with clinical observations, a word of caution is in order. Our data suggest that, in the mouse model, nifedipine elicits distinct sex-specific effects. Importantly, future research should test the direct translatability and implications of these observations in human subjects.

1. "A virtual drug-screening system that can model drug-channel interactions" (pg 32) sounds very novel.

Thank you for highlighting this. We recognize the typo in our manuscript and have made the necessary corrections to ensure clarity and accuracy.

Reviewer #2 (Recommendations For The Authors):

The manuscript is well written. I only have some minor comments:

1. In the patch clamp experiments, there is no information on the recovery of the ionic currents. Is recovery important or not in arterial myocytes? This question is related to the results shown in Figs 5-7. In Fig.5, is the oscillation caused by noise alone or a spontaneous oscillation (such as the oscillation in Fis.6-7) modulated by noise? In general, recovery is an important parameter for the frequency of spontaneous oscillations. It seems to me that the spontaneous oscillations in Fig.8 are mainly noise-driven since they disappear after the cells are coupled through gap junctions.

One important aspect of the oscillatory behavior of the smooth muscle cells is the very long timescales, with fluctuations occurring on the order of seconds. But the majority of ion channels are operating and recovering on the order of milliseconds, so a reasonable approximation is that most ion channels in the cell are operating at steady state at low voltages.

Oscillations in Fig.5: Both the intrinsic oscillations and the noise play key roles in shaping in the oscillations.

The intrinsic deterministic dynamics of the model cells are oscillatory (as seen in Figures 6-7), but the noise can trigger sparks early or delay them, which leads to substantial fluctuations in the inter-spark intervals. Therefore, the spontaneous oscillations are technically modulated by the noise rather than driven by the noise. Nevertheless, in both cases, recovery dynamics play an essential role in shaping the oscillations and determining their frequency

Note however that, when an excitable system is around the bifurcation for oscillations and noise is included, the "firing" statistics in the oscillatory state and the non-oscillatory state are indistinguishable for moderate to high levels of noise.

Noise Exclusion in Figures 6-7: To offer a clear and undistracted interpretation of the results, noise was intentionally omitted from Figures 6-7. This was done to ensure that the primary phenomena under investigation were not obscured. While we recognize the significance of incorporating all elements, including noise, in simulating biological systems, in this case we prioritized a clear point to be made in this context.

Oscillations in Fig.8: Your observation regarding Fig.8 is insightful. Here, uncoupled cells indeed display a spontaneous oscillatory behavior. As documented in previous research, this behavior is not an artifact resulting from cell isolation from the vessel but represents an intrinsic characteristic vital for maintaining electrical signals. The noise in the cells leads to substantial fluctuations in the inter-spike intervals. Because the noise in each cell is uncorrelated, it acts to desynchronize the activity of the cells. Therefore, instead of synchronizing the activity of the cells, the gap junction coupling quenches the large-scale oscillations (the spikes), creating lower amplitude irregular oscillations.

1. The calcium level is much higher in women than in men as shown in Figs.7 and 9. Do women have higher arterial pressure than men?

We thank the reviewer for the observation regarding the calcium levels in Figs.7 and 9. All data presented comes from both male and female C57BL/6J animal models, forming the foundation of our experimental framework.

From earlier studies by the Santana lab (PMID: 32015129), distinct sex-specific differences were found between male and female vascular mesenteric vessels. When the endothelium was removed from small arteriole segments and these segments were subsequently pressurized within a range of 20–120 mmHg, the female arterioles exhibited a pronounced myogenic response in comparison to the male ones. This brings to the forefront the marked sex-based differences, especially in the context of vascular smooth muscle activity.

Yet, when examining the behavior of whole, intact vessels, a different picture emerges. Despite clear sex-specific differences in conditions with the endothelium removed, these distinctions become less pronounced in whole, intact vessels. In essence, both male and female mice exhibit analogous arterial pressure patterns. This suggests possible compensatory mechanisms related to the caliber and structure of the small vessels.

To address the core issue: Despite our data showing higher calcium levels in female samples, it doesn't necessarily imply females consistently exhibit higher arterial pressure across all physiological scenarios.

1. In Fig.9, where is the intravascular pressure (a variable or a parameter) in the mathematical model?

In our model, the intravascular pressure effects are implicitly introduced by modulating the conductance of the non-selective cation currents (INSCC). Specifically, the increase in INSCC is our way of simulating the effects of pressure-induced membrane depolarization. This approach allows us to capture the physiological response to intravascular pressure changes without explicitly introducing it as a separate parameter in the model. We have modified the manuscript to ensure that this rationale is clarified.

1. In Eq.14, the given units of Rmyo (Ohmcm) and Rg (Ohmcmcm) are different, but Eq.14 implies they should have the same unit.

We sincerely appreciate the reviewer's meticulous observation regarding the units discrepancy in Eq.14. We have revised the manuscript to correct the error.

Reviewer #3 (Recommendations For The Authors):

Suggestions for improved or additional experiments, data, or analyses:

Fig. 5 A-B: This is a beautiful qualitative comparison between experimental and simulation data! I think it would be even more impactful if the authors carried out some quantitative analysis of the similarity between male/female experimental/simulation data. For example, the "resting" Vm levels (approx. -30 mV and -40 mV for females and males, respectively) and the peak levels of Vm hyperpolarization could be compared, as well as the frequency of transient hyperpolarization events. It seems like the female model is much more prone to intervals of relative quiescence (i.e., absence of transient hyperpolarization events - e.g., from ~5-6.5 s). Is this consistent with the duration of such ranges in the experimental data (e.g., from 0 to 2.5 s in Fig. 5A).

Thank you for your positive remarks concerning the qualitative comparison in Fig. 5 A-B. We are indeed enthusiastic about the parallels we've identified between experimental and simulation outcomes. We agree with the reviewer that the suggestion on the potential value of a more quantitative assessment. As such we have updated the figure to include an in-depth analysis that provides greater insights and solidifies the power of our simulation predictions when compared to experimental results. A detailed analysis of the male and female data as well as the male and female simulations are summarized in the text as follows:

Baseline membrane potential is -40 mV in male myocytes compared to -30 mV. The frequency of hyperpolarization transients (THs) is 1 Hz in male and 2.5 Hz in female cells for the specific baseline membrane potential shown in Figure 5 A-B. In the range of membrane potentials from -50 mV to -30 mV the frequency increases from 1-2.8Hz which is identical to the experimental frequency range.

• Fig. 7 C-D: Likewise, it would be helpful to quantitatively characterize male/female differences in the model's response to simulated Ca channel blockade (e.g., rate of transient hyperpolarization events, relative levels of ICa and [Ca]i).

Thank you for the constructive feedback on Fig. 7 C-D. We appreciate the emphasis on a quantitative approach to solidify our understanding and have modified the results as follows:

Next, we simulated the effects of calcium channel blocker nifedipine on ICa at a steady membrane potential of -40 mV in male and female simulations. Briefly, previous studies70 have shown that at the therapeutic dose of nifedipine (i.e., about 0.1 μM) L-type Cav1.2 channel currents are reduced by about 70%. Accordingly, we decreased ICa in our mathematical simulations by the same extent. In Figure 7C-D, we show the predicted male (gray) and female (pink) time course of membrane voltage at -40 mV (top panel), ICa (middle panel), and [Ca2+]i (lower panel). First, we observed that in both male and females 0.1 μM nifedipine modifies the frequency of oscillation in the membrane potential, by causing a reduction in oscillation frequency. Second, both male and female simulations (middle panels) show that 0.1 μM nifedipine caused a reduction of ICa to levels that are very similar in male and female myocytes following treatment. Consequently, the reduction of ICa causes both male and female simulations to reach a very similar baseline [Ca2+]i of about 85 nM (lower panels). As a result, simulations provide evidence supporting the idea that CaV1.2 channels are the predominant regulators of intracellular [Ca2+] entry in the physiological range from -40 mV to -20 mV. Importantly, these predictions also suggest that clinically relevant concentrations of nifedipine cause larger overall reductions in Ca2+ influx in female than in male arterial myocytes.

Recommendations for improving the writing and presentation:

When I accessed the GitHub repository linked in section 2.7 (Aug 17, 13:30 PT) it only contained a LICENSE file and none of the described codes and model equations appeared to be publicly available. I would like to access and examine these files. Based on the Clancy lab's excellent track record for making their work publicly available, I have no doubt that the published files will be complete, thoroughly documented, and ready for implementation in studies to reproduce or extend the work described in this manuscript.

https://github.com/ClancyLabUCD/sex-specific-responses-to-calcium-channel-blockers-in-mesenteric-vascular-smooth-muscle

We sincerely apologize for the omission regarding the GitHub repository. It was never our intention to omit the crucial files that should accompany our manuscript. We deeply regret any inconvenience this may have caused in your review process.

We deeply value transparency and the importance of making our work accessible to fellow researchers and the wider community. As you rightly pointed out, the Clancy lab has always been committed to ensuring that our work is available publicly, and this instance is no exception. Please find all codes and documentation here:

Minor corrections to the text and figures:

The introduction is somewhat lengthy, and some of the material contained therein might be more suitable to be merged into the Discussion instead (e.g., paragraphs on negative feedback regulation and the recent study by O'Dwyer et al.).

Thank you – we have updated the introduction but left some foundational work descriptions intact.

• Page 6, section 1.1: There is a missing word (mice?) in the first sentence.

• Page 11, under Eqn. 7: Luo is misspelled as Lou. (Also twice on Page 20.)

Thank you – these have been corrected.

Figs. 2-3: As a colorblind person, it was somewhat challenging for me to differentiate between the red and black lines. Choosing a higher-contrast colour pairing would be beneficial. For some reason, this is not so much of an issue for other figures that use the red/black scheme later in the manuscript (e.g., Figs. 5, 7-8).

We truly appreciate your feedback on the color contrast used in our figures. Accessibility and clarity are crucial to us, and we regret any difficulty you encountered due to the color choices. Based on your valuable feedback, we have included different color pairings in our visual representations to ensure they are comprehensible to all readers, including those who are colorblind.

Fig. 2-3: I am also confused about the use of symbols to indicate significant differences in these plots. In Fig. 2, ** is defined in the legend but not used in the figure. In both figures, the symbols are placed above/below specific sets of points, but it is unclear whether large differences for other x-axis values are statistically significant (e.g., -20 mV in Fig. 3B, +40 mV in Fig. 2C, etc.) This should be clarified.

Thank you – we now have included all the significant differences in the data discussed in the manuscript.

Page 22: The authors state that they "introduced noise into the [Ca]SR..." but the specifics of this approach are not described. As with other aspects of the Methods section, it would be suitable to provide a brief description of the technique used in ref. 40, perhaps added to section 2.4.

Thank you – it has been corrected.

Fig.7 C-D: Axis labels and units are missing. Even though the labels and units will be inferred by most readers, it would be helpful to include them here (at least in C).

Thank you for pointing out the inconsistency between the textual references and Figure 7C-D. We have added the corrected figure.

Page 32: "...the first step toward the development of a virtual drug-screaming system..." I think the authors mean drug-screening. As a side note, this is immediately in the running for the best typo I've ever seen as a peer reviewer.

<good laugh> Thank you for pointing out this error, and we sincerely appreciate your sense of humor about it. You are indeed correct; the intended word is "drug-screening." We have corrected this typo in the manuscript. We're grateful for your thorough review and the light-hearted way you brought this to our attention.

eLife assessment

The study is of importance for the cardiac modeling field by developing a novel mathematical model with sex difference. The data are compelling, and the model is helpful for mechanistic understanding, and thus is also important for experimental physiology. The model is based on experimental data and validated against some experimental data.

Reviewer #1 (Public Review):

The authors developed computational models that capture the electrical and Ca2+ signaling behavior in mesenteric arterial cells from male and female mice. Sex-specific differences in the L-type calcium channel and two voltage-gated potassium channels were carefully tuned based on experimental measurements. To incorporate the stochasticity of ion channel openings seen in smooth muscle cells under physiological conditions, noise was added to the membrane potential and the sarcoplasmic Ca2+ concentration equations. Finally, the models were assembled into 1D vessel representations and used to investigate the tissue-level electrical response to an L-type calcium channel blocker. This comprehensive computational framework helped provide nuanced insight into arterial myocyte function difficult to achieve through traditional experimental methods and can be further expanded into tissue-level studies that incorporate signaling pathways for blood pressure control.

Throughout the paper, model behavior was both validated by experimental recordings and well supported by previously published data. The main findings from the models suggested that sex-specific differences in membrane potential regulation and Ca2+ handling are attributable to variability in the gating of a small number of voltage-gated potassium channels and L-type calcium channels. This variability contributes to a higher Ca2+ channel blocker sensitivity in female arterial vessels. Overall, the study successfully presented novel sex-specific computational models of mesenteric arterial myocytes and demonstrated their use in drug-testing applications.

Reviewer #2 (Public Review):

In this study, Hernandez-Hernandez et al developed a gender-dependent mathematical model of arterial myocytes based on a previous model and new experimental data. The ionic currents of the model and its sex difference were formulated based on patch clamp experimental data, and the model properties were compared with single cell and tissue scale experimental results. This is a study that is of importance for the modeling field as well as for experimental physiology.

Reviewer #3 (Public Review):

Summary:

This hybrid experimental/computational study by Hernandez-Hernandez sheds new light on sex-specific differences between male and female arterial myocytes from resistance arteries. The authors conduct careful experiments in isolated myocytes from male and female mice to obtain the data needed to parameterized sex-specific models of two important ionic currents (i.e., those mediated by CaV1.2 and KV2.1). Available experimental data suggest that KV1.5 channel currents from male and female myocytes are similar, but simulations conducted in the novel Hernandez-Hernandez sex-specific models provide a more nuanced view. This gives rise to the first of the authors' three key scientific claims: (1) In males, KV1.5 is the dominant current regulating membrane potential; whereas, in females, KV2.1 plays a primary role in voltage regulation. They further show that this (2) the latter distinction drives drive sex-specific differences in intracellular Ca2+ and cellular excitability. Finally, working with one-dimensional models comprising several copies of the male/female myocyte models linked by resistive junctions, they use simulations to (3) predict that sensitivity of arterial smooth muscle to Ca2+ channel-blocking drugs commonly used to treat hypertension is heightened in female compared to male cells.

In my opinion, the following strengths of the work are particularly notable:

• The Methodology is described in exquisite detail in straightforward language that will be easy to understand for most if not all peer groups working in computational physiology. The authors have deployed standard protocols (e.g., parameter fitting as described by Kernik et al., sensitivity analysis as described by Sobie et al.) and appropriate brief explanations of these techniques are provided. The manoeuvre used to represent stochastic effects on voltage dynamics is particularly clever and something I have not personally encountered before. Collectively, these strengthen the credibility of the model and greatly enrich the manuscript.<br /> • The Results section describes findings that robustly support the three key scientific claims outlined in my Summary. It is evident these experiments were carefully designed and carried out with care and intentionality. Several figures show experimental data side-by-side with outputs from the corresponding models. These are an excellent illustration of the power of the authors' novel sex-specific computational simulation platform.

Author Response

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

We would like to thank the reviewers for their strong interest in our studies and their excellent suggestions for improvement.

Reviewer #1:

Weaknesses:

Comment 1. The authors identified NPR-15 and ASJ neurons that are involved in both molecular and behavioral responses to pathogen attack. This finding, by itself, is significant. However, how the NPR-15/ASJ circuit regulates the interplay between the two defense strategies was not explored. Therefore, emphasizing the interplay in the title and the abstract is misleading.

Response to comment 1. We have removed the word “interplay.”

Comment 2. Although the discovery of a single GPCR regulating both immunity and avoidance behavior is significant and novel, NPR-15 is not the first GPCR identified with these functions. Previously, the same lab reported that the GPCR OCTR-1 also regulates immunity and avoidance behavior through ASH and ASI neurons respectively (PMID: 29117551). This point was not mentioned in the current manuscript.

Response to Comment 2. We’d like to clarify that it remains unclear whether OCTR-1 itself controls both immunity and behavior (PMID: 29117551). The reference study showed that OCTR-1-expressing neurons ASH and ASI control immunity and behavior, respectively. We modified the manuscript to make this point clearer: “While OCTR-1-expressing neurons ASI play a role in avoidance (34), the specific role of OCTR-1 in ASH and ASI neurons remains unclear. “

Comment 3. The authors discovered that NPR-15 regulates avoidance behavior via the TRPM gene, GON-2. Only two factors (GON-2 and GTL-2) were examined in this study, and GON-2 happens to function through the intestine.

Response to comment 3. We studied GON-2 and GTL-2 because a recent screen of intestinal TRPM genes showed that they are the only two involved in the control of pathogen avoidance. We modified the manuscript to make this rationale clearer: “Because transient receptor potential melastatin (TRPM) ion channels, GON-2 and GTL-2, are required for pathogen avoidance (32), we studied whether they may be part of the NPR-15 pathway that controls pathogen avoidance”

Comment 3b. It is possible that NPR-15 may broadly regulate multiple effectors in multiple tissues. Confining the regulation to the amphid sensory neuron-intestinal axis, as stated in the title and elsewhere in the manuscript, is not accurate.

Response to comment 3b. We agree that NPR-15 may broadly regulate multiple effectors in different tissues. Indeed, we have shown that the transcriptional activity of ELT-2, HLH-30, DAF-16, and PMK-1 is higher in npr-15 than in WT animals. We found that expression of NPR-15 only in ASJ cells rescues both the survival and behavioral phenotypes of npr-15 animals (Figs. 4F and 5C).

Comment 4. The C. elegans nervous system is simple, and hermaphrodites only have 302 neurons. Individual neurons possessing multiple regulatory functions is expected. Whether this is conserved in mammals and other vertebrates is unknown, because in higher animals, neurons and neuronal circuits could be more specialized.

Response to Comment 4. We agreed. We have removed the statements discussing conservation in that manner.

Comment 5. A key question, that is, why would NPR-15 suppress immunity (which is bad for defense) but enhance avoidance behavior (which is good for defense), is not addressed or explained. This could be due to temporal regulation, for example, upon pathogen exposure, NPR-15 could regulate behavior to avoid the pathogen, but after infection, NPR-15 could suppress excessive immune responses or quench the responses for the resolution of infection.

Response to comment 5. We found that NPR-15 controls the expression of immune genes in the absence of an infection. Without further experiments, we think it would be too speculative to discuss the possibility of a temporal regulation. However, we modified the manuscript to address the control of both molecular and behavioral immunity by NPR-15. The revised discussion reads: “Our findings shed light on the role of NPR-15 in the control of the immune response. NPR-15 seems to suppress specific immune genes while activating pathogen avoidance behavior to minimize potential tissue damage and the metabolic energy cost associated with activating the molecular immune response against pathogen infections. Overall, the control of immune activation is essential for maintaining homeostasis and preventing excessive tissue damage caused by an overly aggressive and energy-costly response against pathogens (60-63).”

Comment 6. Discussion appears timid in scope and contains some repetitive statements. Point 5 can be addressed in the Discussion.

Response to comment 6. We have removed repetitive concepts and modified the discussion as mentioned in the response to point 5.

Comment 7. Overall, the authors presented an impactful study that identified specific molecules and neuronal cells that regulate both molecular and behavioral immune responses to pathogen attack. Most conclusions are supported by solid evidence. However, some statements are overreaching, for example, regulation of the interplay between molecular and behavioral immune responses was emphasized but not explored. Nonetheless, this study reported a significant and novel discovery and has laid a foundation for investigating such an interplay in the future.

Response to comment 7: We removed the statements that may have appeared to be overreaching and addressed the weakness raised by the reviewer. The revised discussion reads “Our findings shed light on the role of NPR-15 in the control of the immune response. NPR-15 seems to suppress specific immune genes while activating pathogen avoidance behavior to minimize potential tissue damage and the metabolic energy cost associated with activating the molecular immune response against pathogen infections. Overall, the control of immune activation is essential for maintaining homeostasis and preventing excessive tissue damage caused by an overly aggressive and energy-costly response against pathogens (60-63).”

Recommendations for the authors:

Recommendations 1. The title, abstract and some statements in the main text need to be re-written to reflect the fact that regulation of the interplay between molecular and behavioral immune responses was not explored in this study.

Response to recommendations 1. We modified the title and abstract accordingly.

Recommendations 2. It should be mentioned in the manuscript that OCTR-1 is the first GPCR that was identified to regulate both immunity and avoidance behavior.

Response to recommendation 2. We addressed this issue as discussed in the response to comment 2.

Recommendations 3. Repetitive statements should be removed from Discussion.

Response to recommendations 3. The statements were removed.

Recommendations 4. It is surprising to see that pmk-1 RNAi did not affect the survival of npr-15(tm12539) animals against S. aureus because PMK-1 has a general role in defense against S. aureus infection.

Response to recommendations 4. We agree. However, the RNAi studies were validated using mutants (Fig. S3B).

Recommendations 4b. Also, the rationale for using skn-1 RNAi as a control was not given. These need to be explained adequately in the manuscript.

Response to recommendations 4b. There’s no need to include skn-1 RNAi and we removed the data.

Recommendations 5. The conclusion that the lack of avoidance behavior by NPR-15 loss-of-function is independent of immunity and neuropeptide genes was drawn entirely based on experiments with RNAi of individual genes. Functional redundancy among genes could render RNAi of individual genes ineffective, thus masking the dependence of avoidance behavior on these genes. More experiments are needed to support this conclusion, or the wording of the conclusion need to be changed.

Response to recommendations 5. We modified the conclusion to address this issue: “Given the possibility of functional redundancy among these genes, we cannot rule out the possibility that different combinations may play a role in controlling avoidance behavior.”

Recommendations 6. What is representation factor in Fig. 2B and 2C?

Response to recommendations 5. Figure 2B shows significantly enriched terms with a Q value < 0.1, sorted by P values. Figure 2 C shows the representation factor that is calculated using a tool, http://nemates.org/MA/progs/overlap_stats.html. The calculation is based on the number of genes in set 1, the number of genes in set 2, and the Overlap between set 1 and set 2, as well as the number of genes in the genome.

We corrected the Figure legends and included the corresponding information in Material and Methods.

Recommendations 7. The legend of Fig. 6 was wrong and should be changed to 'GPCR/NPR-15 suppressed immune response and enhanced avoidance behavior via sensory neurons'.

Response to recommendations 7. Thank you for pointing this out. We changed the legend.

Reviewer #2:

Comments 1. There is some variance in lawn occupancy of wt strains between the different trials in WT animals (e.g. in Fig. 1: 25 for wt vs 60% for npr mutant; S1c 5% for wt and 60% for npr mutant).

Response to comment 1. We appreciate the observation. We did notice some variation in both the WT and npr-15(tm12539) animals during our study. Notably, the variation appeared to be more in the WT compared to the npr-15(tm12539) animals. However, it's important to note that these variations did not significantly affect the outcome of our findings. We calculated the means, standard deviation, and standard error across different experimental trials that are presented in the manuscript (Table S2) (new Table). It's worth noting that these variations did not significantly impact the observed differences in lawn occupancy between the wild-type (WT) and npr-15 mutant strains.

We addressed this issue in the revised manuscript: “Interestingly, we noticed that the variation in lawn occupancy is greater in WT than in npr-15(tm12539) animals across experiments (Table S2), which suggests that the strong lack of avoidance of npr-15(tm12539) somehow counteracts the experimental variation”

Comment 2. Does this reflect rates of migration or re-occupancy in WT?

Response to comment 2. We did not observe any re-occupancy in either the WT or npr-15 animals at 24-hour time points (which we mostly use in this study) or beyond. To address the comment, we performed a new experiment and found that the re-occupancy of npr-15 mutants is comparable to that of WT animals at 4 hours post-exposure (Figure S1B).

Comment 3. Does pathogen avoidance persist and/or the rate of avoidance differ in npr mutant worms?

Response to comment 3. As illustrated in new Figure S1B, the avoidance behavior in response to pathogens remained consistent even when we extended our observations up to 48 hours (Figure S1B).

Comment 4. if animals were exposed then re-exposed, could the authors to determine whether a learned avoidance was similarly affected by this mutation by assessing rate changes?

Response to comment 4. We conducted the proposed experiment and observed that the WT animals learned to avoid the pathogen but not npr-15(tm12539) mutants (Figure S1C). The revised manuscript reads: “We also found that npr-15(tm12539) exhibited reduced learned avoidance compared to WT animals (Figure S1C).”

Comment 5: Is there any difference in gene expression of animals that have migrated off the lawn to those remaining on the lawn (e.g. in partial lawn experiments?).

Response to comment 5. This is an interesting question that has not been addressed in the field yet. While we think the study is exciting, we believe that it is outside the scope of our work. All the gene expression studies performed here are in non-avoiding conditions.

Comment 6. No concerns but the P values in the legends are a pain to read. Why not put them in figures as in above figures.

Response to comment 6. We included the P values as suggested.

Recommendations for the authors:

Recommendation 1. Fig. 1/S1. Comments: There is some variance in lawn occupancy of wt strains between the different trials in WT animals (e.g. in Fig. 1: 25 for wt vs 60% for npr mutant; S1c 5% for wt and 60% for npr mutant).

Response to recommendation 1. We addressed this issue as discussed in the response to comment 1.

Recommendation 2. Fig. 1/S1. Comments. Does this reflect rates of migration or re-occupancy in WT?

Response to recommendation 2. We have responded to this issue in comment 2.

Recommendations 3. Fig. 1/S1. Comments. Does pathogen avoidance persist and/or the rate of avoidance differ in npr mutant worms.

Response to recommendation 3. We have responded to this issue in comment 3.

Recommendation 4. Fig. 1/S1. Comments B. and if animals were exposed then re- exposed, could the authors to determine whether a learned avoidance was similarly affected by this mutation by assessing rate changes?

Response to recommendation 4: We have responded to this issue in comment 4 above.

Recommendation 5. Fig. 2/S2. Comment: Is there any difference in gene expression of animals that have migrated off the lawn to those remaining on the lawn (e.g. in partial lawn expts?).

Response to recommendation 5. We have responded to this issue in comment 5 above.

Recommendation 6. Fig. 3/S3. Comment. No concerns but the P values in the legends are a pain to read. Why not put them in figures as in above figures.

Response to recommendation 6. We included the P values.

Recommendation 7. Fig. 5. Comments: The authors suggest that the ASJ/NPR15 effect to limit avoidance acts via inhibition of GON-2 in the intestine. The observation that GON-2 inhibition effects on pathogen avoidance occur independently of neurons could suggest that it is a redundant way of accomplishing the same thing, which then makes one wonder if or what the connection is exists between the neuron and the gut. The effect of ASJ via NPR on pathogen avoidance is not neuropeptide dependent, which they show. So how the neuronal-gut communication works. Specific Transmitters... perhaps.

Response to Recommendation 7 Fig. 5. Thanks for this observation. To address the recommendation, we modified the discussion: “Our research additionally indicates that the regulation of NPR-15-mediated avoidance is not influenced by intestinal immune and neuropeptide genes. Given the potential for functional redundancy and our focus on genes upregulated in the absence of NPR-15, we cannot entirely rule out the possibility that unexamined immune effectors or neuropeptides, not transcriptionally controlled by NPR-15, might be involved. Different intestinal signals may also participate in the NPR-15 pathway that controls pathogen avoidance.”

Recommendation 8. Comment. Since ASJ neurons control entry into dauer, perhaps isn't surprising that DAF-16 showed up as an NPR-15. induced factor (and dauer worms are resistant to a lot of stressors); that said dauer hormones might be involved as well. Is there any evidence that DAF-16 down-regulates GON-2 expression (see Murphy, Kenyon et al. 2005), and along these lines would GON-2 RNAi work in a DAF-16 mutant? I think addressing these issues are the subject of future studies.

Response to recommendation 8. We checked the data in the study by Murphy, Kenyon et al., and found that the gon-2 gene was not downregulated.

Recommendation 9. Minor: Regarding the description to Fig. 5. "Consistently with our previous findings, we found that only " The adverb form of consistent should not be used here.

Response to recommendation 9. Thank you for pointing this out. The description of Figure 5 was corrected.

eLife assessment

The important work by Aballay et al. significantly advances our understanding of how G protein-coupled receptors (GPCRs) regulate immunity and pathogen avoidance. The authors provide convincing evidence for the GPCR NPR-15 to mediate immunity by altering the activity of several key transcription factors. This work will be of broad interest to immunologists.

Reviewer #1 (Public Review):

Summary:

Otarigho et al. presented a convincing study revealing that in C. elegans, the neuropeptide Y receptor GPCR/NPR-15 mediates both molecular and behavioral immune responses to pathogen attack. Previously, three npr genes were found to be involved in worm defense. In this study, the authors screened mutants in the remaining npr genes against P. aeruginosa-mediated killing and found that npr-15 loss-of-function improved worm survival. npr-15 mutants also exhibited enhanced resistance to other pathogenic bacteria but displayed significantly reduced avoidance to S. aureus, independent of aerotaxis, pathogen intake and defecation. The enhanced resistance in npr-15 mutant worms was attributed to upregulation of immune and neuropeptide genes, many of which were controlled by the transcription factors ELT-2 and HLH-30. The authors found that NPR-15 regulates avoidance behavior via the TRPM gene, GON-2, which has a known role in modulating avoidance behavior through the intestine. The authors further showed that both NPR-15-dependent immune and behavioral responses to pathogen attack were mediated by the NPR-15-expressing neurons ASJ. Overall, the authors discovered that the NPR-15/ASJ neural circuit may regulate distinct defense mechanisms against pathogens under different circumstances. This study provides novel and useful information to researchers in the fields of neuroimmunology and C. elegans research.

Strengths:

1. This study uncovered specific molecules and neuronal cells that regulate both molecular immune defense and behavior defense against pathogen attack and indicate that the same neural circuit may regulate distinct defense mechanisms under different circumstances. This discovery is significant because it not only reveals regulatory mechanisms of different defense strategies but also suggests how C. elegans utilize its limited neural resources to accomplish complex regulatory tasks.

2. The conclusions in this study are supported by solid evidence, which are often derived from multiple approaches and/or experiments. Multiple pathogenic bacteria were tested to examine the effect of NPR-15 loss-of-function on immunity; the impacts of pharyngeal pumping and defecation on bacterial accumulation were ruled out when evaluating defense; RNA-seq and qPCR were used to measure gene expression; gene inactivation was done in multiple strains to assess gene function.

3. Gene differential expression, gene ontology and pathway analyses were performed to demonstrate that NPR-15 controls immunity through regulating immune pathways.

4. Elegant approaches were employed to examine avoidance behavior (partial lawn, full lawn, and lawn occupancy) and the involvement of neurons in regulating immunity and avoidance (the use of a diverse array of mutant strains).

5. Statistical analyses were appropriate and adequate.

Reviewer #2 (Public Review):

Summary:<br /> The authors are studying the behavioral response to pathogen exposure. They and others have previously describe the role that the G-protein coupled receptors in the nervous system plays in detecting pathogens, and initiating behavioral patterns (e.g. avoidance/learned avoidance) that minimize contact. The authors study this problem in C. elegans, which is amenable to genetic and cellular manipulations and allow the authors to define cellular and signaling mechanisms. This paper extends the original idea to now implicate signaling and transcriptional pathways within a particular neuron (ASJ) and the gut in mediating avoidance behaviour.

Strengths:<br /> The work is rigorous and elegant and the data are convincing. The authors make superb use of mutant strains in C. elegans, as well tissue specific gene inactivation and expression and genetic methods of cell ablation. to demonstrate how a gene, NPR15 controls behavioral changes in pathogen infection. The results suggest that ASJ neurons and the gut mediate such effects. I expect the paper will constitute an important contribution to our understanding of how the nervous system coordinates immune and behavioral responses to infection.

Author Response

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

Reviewer #1:

A weakness of the paper is that the power of the model is illustrated for only one specific set of parameters, added in a stepwise manner and the comparison to one specific empirical TGM, assumed to be prototypical; And that this comparison remains descriptive. (That is could a different selection of parameters lead to similar results and is there TGM data which matches these settings less well.)

The fact that the comparisons in the paper are descriptive is a central point of criticism from both reviewers. As mentioned in my preliminary response, I intentionally did not optimise the model to a specific TGM or show an explicit metric of fitness. As I now explicitly mention in the new experimental section of the paper:

“The previous analyses were descriptive in the sense that they did not quantify how much the generated TGMs resembled a specific empirical TGM. This was deliberate, because empirical TGMs vary across subjects and experiments, and I aimed at characterising them as generally as possible by looking at some characteristic features in broad terms. For example, while TGMs typically have a strong diagonal and horizontal/vertical bars of high accuracy, questions such as when these effects emerge and for how long are highly dependent on the experimental paradigm. For the same reason, I did not optimise the model hyperparameters, limiting myself to observing the behaviour of the model across some characteristic configurations”

And, in the Discussion:

“The demonstrations here are not meant to be tailored to a specific data set, and are, for the most part, intentionally qualitative. TGMs do vary across experiments and subjects; and the hyperparameters of the model can be explicitly optimised to specific scientific questions, data sets, and even individuals. In order to explore the space of configurations effectively, an automatic optimisation of the hyperparameter space using, for instance, Bayesian optimisation (Lorenz, et al., 2017) could be advantageous. This may lead to the identification of very specific (spatial, spectral and temporal) features in the data that may be neurobiologically interpreted.”

Nonetheless, it is possible to fit the model to a specific TGMs by using a explicit metric of fitness. For illustration, this is what I did in the new experimental section Fitting and empirical TGM, where I used correlation with an empirical TGM to optimise two temporal parameters: the rise slope and the fall slope. As can be seen in the Figure 8, the correlation with the empirical TGM was as high as 0.7, even though I did not fit the other parameters of the model. As mentioned in the paragraph above, more sophisticated techniques such as Bayesian optimisation might be necessary for a more exhaustive exploration, but this would be beyond the scope of the current paper.

I would also like to point out that fitting the parameters in a step-wise manner was a necessity for interpretation. I suggest to think of the way we use F-tests in regression analyses as a comparison: if we want to know how important a feature is, we compare the model with and without this feature and see how much we loss.

It further remained unclear to me, which implications may be drawn from the generative model, following from the capacities to mimic this specific TGM (i) for more complex cases, such as the comparison between experimental conditions, and (ii) about the complex nature of neural processes involved.

Following on the previous points, the object of this paper (besides presenting the model and the associated toolbox) was not to mimic a specific TGM, but to characterise the main features that we generally see across studies in the field. To clarify this, I have added Figure 2 (previously a Supplemental Information figure), and added the following to the Results section:

“Figure 2 shows a TGM for an example subject, where some archetypal characteristics are highlighted. In the experiments below, specifically, I focus on the strong narrow diagonal at the beginning of the trial, the broadening of accuracy later in the trial, and the vertical/horizontal bars of higher-than-chance accuracy. Importantly, this specific example in Figure 2 is only meant as a reference, and therefore I did not optimise the model hyperparameters to this TGM (except in the last subsection), or showed any quantitative metric of similarity.”

I mention the possibility of using the model to explore more complex cases in the Introduction, although doing so here would be out of scope:

“Other experimental paradigms, including motor tasks and decision making, can be investigated with genephys”

Towards this end, I would appreciate (i) a more profound explanation of the conclusions that can be drawn from this specific showcase, including potential limitations, as well as wider considerations of how scientists may empower the generative model to (ii) understand their experimental data better and (iii) which added value the model may have in understanding the nature of underlying brain mechanism (rather than a mere technical characterization of sensor data).

To better illustrate how to use genephys to explore a specific data set, I have added a section (Fitting an empirical TGM) where I show how to fit specific hyperparameters to an empirical TGM in a simple manner.

In the Introduction, I briefly mentioned:

“This (not exhaustive) list of effects was considered given previous literature (Shah, et al., 2004; Mazaheri & Jensen, 2006; Makeig, et al., 2002; Vidaurre, et al., 2021), and each effect may be underpinned by distinct neural mechanisms. For example, it is not completely clear the extent to which stimulus processing is sustained by oscillations, and disentangling these effects can help resolving this question”

In the Discussion, I have further commented:

“Genephys has different available types of effect, including phase resets, additive damped oscillations, amplitude modulations, and non-oscillatory responses. All of these elements, which may relate to distinct neurobiological mechanisms, are configurable and can be combined to generate a plethora of TGMs that, in turn, can be contrasted to specific empirical TGMs. This way, we can gain insight on what mechanisms might be at play in a given task.

The demonstrations here are not meant to be tailored to a specific data set, and are, for the most part, intentionally qualitative. TGMs do vary across experiments and subjects; and the hyperparameters of the model can be explicitly optimised to specific scientific questions, data sets, and even individuals. In order to explore the space of configurations effectively, an automatic optimisation of the hyperparameter space using, for instance, Bayesian optimisation (Lorenz, et al., 2017) could be advantageous. This may lead to the identification of very specific (spatial, spectral and temporal) features in the data that may be neurobiologically interpreted. “

On p. 15 "Having a diversity of frequencies but not of latencies produces another regular pattern consisting of alternating, parallel bands of higher/lower than baseline accuracy. This, shown in the bottom left panel, is not what we see in real data either. Having a diversity of latencies but not of frequencies gets us closer to a realistic pattern, as we see in the top right panel." The terms frequency and latency seem to be confused.

The Reviewer is right. I have corrected this now. Thank you.

Reviewer #2:

The results of comparisons between simulations and real data are not always clear for an inexperienced reader. For example, the comparisons are qualitative rather than quantitative, making it hard to draw firm conclusions. Relatedly, it is unclear whether the chosen parameterizations are the only/best ones to generate the observed patterns or whether others are possible. In the case of the latter, it is unclear what we can actually conclude about underlying signal generators. It would have been different if the model was directly fitted to empirical data, maybe of different cognitive conditions. Finally, the neurobiological interpretation of different signal properties is not discussed. Therefore, taken together, in its currently presented form, it is unclear how this method could be used exactly to further our understanding of the brain.

This critique coincides with that of Reviewer 1. In the current version, I made more clear the fact that I am not fitting a specific empirical TGM and why, and that, instead, I am referring to general features that appear broadly throughout the literature. See more detailed changes below.

Regarding whether the chosen parameterizations are the only/best ones to generate the observed patterns, the Discussion reflects this limitation:

“Also importantly, I have shown that standard decoding analysis can differentiate between these explanations only to some extent. For example, the effects induced by phase-resetting and the use of additive oscillatory components are not enormously different in terms of the resulting TGMs. In future work, alternatives to standard decoding analysis and TGMs might be used to disentangle these sources of variation (Vidaurre, et al., 2019). ”

And

“Importantly, the list of effects that I have explored here is not exhaustive …”

Of course, since the list of signal features I have explored is not exhaustive, it cannot be claimed without a doubt that these features are the ones generating the properties we observe in real TGMs. The model, however, is a step forward in that direction, as it provides us with a tool to at least rule out some causes.

Firstly, it was not entirely clear to me from the introduction what gap exactly the model is supposed to fill: is it about variance in neural responses in general, about which signal properties are responsible for decoding, or about capturing stability of signals? It seems like it does all of these, but this needs to be made clearer in the introduction. It would be helpful to emphasize exactly what insights the model can provide that are unable to be obtained with the current methods.

I have now made this explicit in in the Introduction, as suggested:

“To gain insight into what aspects of the signal underpin decoding accuracy, and therefore the most stable aspects of stimulus processing, I introduce a generative model”

To help illustrating what insights the model can provide, I have added the following sentence as an example:

“For example, it is not completely clear the extent to which stimulus processing is sustained by oscillations, and disentangling these effects can help resolving this question.”

Furthermore, I was unclear on why these specific properties were chosen (lines 71 to 78). Is there evidence from neuroscience to suggest that these signal properties are especially important for neural processing? Or, if the logic has more to do with signal processing, why are these specific properties the most important to include?

To clarify this the text now reads:

“In the model, when a channel responds, it can do it in different ways: (i) by phase-resetting the ongoing oscillation to a given target phase and then entraining to a given frequency, (ii) by an additive oscillatory response independent of the ongoing oscillation, (iii) by modulating the amplitude of the stimulus-relevant oscillations, or (iv) by an additive non-oscillatory (slower) response. This (not exhaustive) list of effects was considered given previous literature (Shah, et al., 2004; Mazaheri & Jensen, 2006; Makeig, et al., 2002; Vidaurre, et al., 2021), and each effect may be underpinned by distinct neural mechanisms”

The general narrative and focus of the paper could also be improved. It might help to start off with an outline of what the goal is at the start of the paper and then explicitly discuss how each of the steps works toward that goal. For example, I got the idea that the goal was to capture specific properties of an empirical TGM. If this was the case, the empirical TGM could be placed in the main body of the text as a reference picture for all simulated TGMs. For each simulation step, it could be emphasized more clearly exactly which features of the TGM is captured and what that means for interpreting these features in real data.

Thank you. To clarify the purpose of the paper better, I have brought Figure 2 to the front (before a Supplementary Figure), and in the first part of Results I have now added:

“Figure 2 shows a TGM for an example subject, where some archetypal characteristics are highlighted. In the experiments below, specifically, I focus on the strong narrow diagonal at the beginning of the trial, the broadening of accuracy later in the trial, and the vertical/horizontal bars of higher-than-chance accuracy. Importantly, this specific example in Figure 2 is only meant as a reference, and therefore I did not optimise the model hyperparameters to this TGM (except in the last subsection), or showed any quantitative metric of similarity. ”

I have enunciated the goals more clearly in the Introduction:

“To gain insight into what aspects of the signal underpin decoding accuracy, and therefore the most stable aspects of stimulus processing, …”

Relatedly, it would be good to connect the various signal properties to possible neurobiological mechanisms. I appreciate that the author tries to remain neutral on this in the introduction, but I think it would greatly increase the implications of the analysis if it is made clearer how it could eventually help us understand neural processes.

The Reviewer is right in pointing out that I preferred to remain neutral on this. While I have still kept that tone of neutrality throughout the paper, I have now included the following sentence as an example of a neurobiological question that could be investigated with the model:

“For example, it is not completely clear the extent to which stimulus processing is sustained by oscillations, and disentangling these effects can help resolving this question.”

And, more generally,

“Genephys has different available types of effect, including phase resets, additive damped oscillations, amplitude modulations, and non-oscillatory responses. All of these elements, which may relate to distinct neurobiological mechanisms, are configurable and can be combined to generate a plethora of TGMs that, in turn, can be contrasted to specific empirical TGMs. This way, we can gain insight on what mechanisms might be at play in a given task. ”

Line 57: this sentence is very long, making it hard to follow, could you break up into smaller parts?

Thank you. The sentence is fragmented now.

Please replace angular frequencies with frequencies in Hertz for clarity.

Here I have preferred to stick to angular frequencies because it is more general than if I talk about Hertz, because that would entail having a specific sampling frequency. I think doing so would create confusion precisely of the sorts that I am trying to clarify in this revision: that is, that these results are not specific of one TGM but reflect general features that we see broadly in the literature.

There are quite some types throughout the paper, please recheck

Thank you. I have revised and have made my best to clear them out.

Author Response

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

Reviewer #1:

A weakness of the paper is that the power of the model is illustrated for only one specific set of parameters, added in a stepwise manner and the comparison to one specific empirical TGM, assumed to be prototypical; And that this comparison remains descriptive. (That is could a different selection of parameters lead to similar results and is there TGM data which matches these settings less well.)

The fact that the comparisons in the paper are descriptive is a central point of criticism from both reviewers. As mentioned in my preliminary response, I intentionally did not optimise the model to a specific TGM or show an explicit metric of fitness. As I now explicitly mention in the new experimental section of the paper:

“The previous analyses were descriptive in the sense that they did not quantify how much the generated TGMs resembled a specific empirical TGM. This was deliberate, because empirical TGMs vary across subjects and experiments, and I aimed at characterising them as generally as possible by looking at some characteristic features in broad terms. For example, while TGMs typically have a strong diagonal and horizontal/vertical bars of high accuracy, questions such as when these effects emerge and for how long are highly dependent on the experimental paradigm. For the same reason, I did not optimise the model hyperparameters, limiting myself to observing the behaviour of the model across some characteristic configurations”

And, in the Discussion:

“The demonstrations here are not meant to be tailored to a specific data set, and are, for the most part, intentionally qualitative. TGMs do vary across experiments and subjects; and the hyperparameters of the model can be explicitly optimised to specific scientific questions, data sets, and even individuals. In order to explore the space of configurations effectively, an automatic optimisation of the hyperparameter space using, for instance, Bayesian optimisation (Lorenz, et al., 2017) could be advantageous. This may lead to the identification of very specific (spatial, spectral and temporal) features in the data that may be neurobiologically interpreted.”

Nonetheless, it is possible to fit the model to a specific TGMs by using a explicit metric of fitness. For illustration, this is what I did in the new experimental section Fitting and empirical TGM, where I used correlation with an empirical TGM to optimise two temporal parameters: the rise slope and the fall slope. As can be seen in the Figure 8, the correlation with the empirical TGM was as high as 0.7, even though I did not fit the other parameters of the model. As mentioned in the paragraph above, more sophisticated techniques such as Bayesian optimisation might be necessary for a more exhaustive exploration, but this would be beyond the scope of the current paper.

I would also like to point out that fitting the parameters in a step-wise manner was a necessity for interpretation. I suggest to think of the way we use F-tests in regression analyses as a comparison: if we want to know how important a feature is, we compare the model with and without this feature and see how much we loss.

It further remained unclear to me, which implications may be drawn from the generative model, following from the capacities to mimic this specific TGM (i) for more complex cases, such as the comparison between experimental conditions, and (ii) about the complex nature of neural processes involved.

Following on the previous points, the object of this paper (besides presenting the model and the associated toolbox) was not to mimic a specific TGM, but to characterise the main features that we generally see across studies in the field. To clarify this, I have added Figure 2 (previously a Supplemental Information figure), and added the following to the Results section:

“Figure 2 shows a TGM for an example subject, where some archetypal characteristics are highlighted. In the experiments below, specifically, I focus on the strong narrow diagonal at the beginning of the trial, the broadening of accuracy later in the trial, and the vertical/horizontal bars of higher-than-chance accuracy. Importantly, this specific example in Figure 2 is only meant as a reference, and therefore I did not optimise the model hyperparameters to this TGM (except in the last subsection), or showed any quantitative metric of similarity.”

I mention the possibility of using the model to explore more complex cases in the Introduction, although doing so here would be out of scope:

“Other experimental paradigms, including motor tasks and decision making, can be investigated with genephys”

Towards this end, I would appreciate (i) a more profound explanation of the conclusions that can be drawn from this specific showcase, including potential limitations, as well as wider considerations of how scientists may empower the generative model to (ii) understand their experimental data better and (iii) which added value the model may have in understanding the nature of underlying brain mechanism (rather than a mere technical characterization of sensor data).

To better illustrate how to use genephys to explore a specific data set, I have added a section (Fitting an empirical TGM) where I show how to fit specific hyperparameters to an empirical TGM in a simple manner.

In the Introduction, I briefly mentioned:

“This (not exhaustive) list of effects was considered given previous literature (Shah, et al., 2004; Mazaheri & Jensen, 2006; Makeig, et al., 2002; Vidaurre, et al., 2021), and each effect may be underpinned by distinct neural mechanisms. For example, it is not completely clear the extent to which stimulus processing is sustained by oscillations, and disentangling these effects can help resolving this question”

In the Discussion, I have further commented:

“Genephys has different available types of effect, including phase resets, additive damped oscillations, amplitude modulations, and non-oscillatory responses. All of these elements, which may relate to distinct neurobiological mechanisms, are configurable and can be combined to generate a plethora of TGMs that, in turn, can be contrasted to specific empirical TGMs. This way, we can gain insight on what mechanisms might be at play in a given task.

The demonstrations here are not meant to be tailored to a specific data set, and are, for the most part, intentionally qualitative. TGMs do vary across experiments and subjects; and the hyperparameters of the model can be explicitly optimised to specific scientific questions, data sets, and even individuals. In order to explore the space of configurations effectively, an automatic optimisation of the hyperparameter space using, for instance, Bayesian optimisation (Lorenz, et al., 2017) could be advantageous. This may lead to the identification of very specific (spatial, spectral and temporal) features in the data that may be neurobiologically interpreted. “

On p. 15 "Having a diversity of frequencies but not of latencies produces another regular pattern consisting of alternating, parallel bands of higher/lower than baseline accuracy. This, shown in the bottom left panel, is not what we see in real data either. Having a diversity of latencies but not of frequencies gets us closer to a realistic pattern, as we see in the top right panel." The terms frequency and latency seem to be confused.

The Reviewer is right. I have corrected this now. Thank you.

Reviewer #2:

The results of comparisons between simulations and real data are not always clear for an inexperienced reader. For example, the comparisons are qualitative rather than quantitative, making it hard to draw firm conclusions. Relatedly, it is unclear whether the chosen parameterizations are the only/best ones to generate the observed patterns or whether others are possible. In the case of the latter, it is unclear what we can actually conclude about underlying signal generators. It would have been different if the model was directly fitted to empirical data, maybe of different cognitive conditions. Finally, the neurobiological interpretation of different signal properties is not discussed. Therefore, taken together, in its currently presented form, it is unclear how this method could be used exactly to further our understanding of the brain.

This critique coincides with that of Reviewer 1. In the current version, I made more clear the fact that I am not fitting a specific empirical TGM and why, and that, instead, I am referring to general features that appear broadly throughout the literature. See more detailed changes below.

Regarding whether the chosen parameterizations are the only/best ones to generate the observed patterns, the Discussion reflects this limitation:

“Also importantly, I have shown that standard decoding analysis can differentiate between these explanations only to some extent. For example, the effects induced by phase-resetting and the use of additive oscillatory components are not enormously different in terms of the resulting TGMs. In future work, alternatives to standard decoding analysis and TGMs might be used to disentangle these sources of variation (Vidaurre, et al., 2019). ”

And

“Importantly, the list of effects that I have explored here is not exhaustive …”

Of course, since the list of signal features I have explored is not exhaustive, it cannot be claimed without a doubt that these features are the ones generating the properties we observe in real TGMs. The model, however, is a step forward in that direction, as it provides us with a tool to at least rule out some causes.

Firstly, it was not entirely clear to me from the introduction what gap exactly the model is supposed to fill: is it about variance in neural responses in general, about which signal properties are responsible for decoding, or about capturing stability of signals? It seems like it does all of these, but this needs to be made clearer in the introduction. It would be helpful to emphasize exactly what insights the model can provide that are unable to be obtained with the current methods.

I have now made this explicit in in the Introduction, as suggested:

“To gain insight into what aspects of the signal underpin decoding accuracy, and therefore the most stable aspects of stimulus processing, I introduce a generative model”

To help illustrating what insights the model can provide, I have added the following sentence as an example:

“For example, it is not completely clear the extent to which stimulus processing is sustained by oscillations, and disentangling these effects can help resolving this question.”

Furthermore, I was unclear on why these specific properties were chosen (lines 71 to 78). Is there evidence from neuroscience to suggest that these signal properties are especially important for neural processing? Or, if the logic has more to do with signal processing, why are these specific properties the most important to include?

To clarify this the text now reads:

“In the model, when a channel responds, it can do it in different ways: (i) by phase-resetting the ongoing oscillation to a given target phase and then entraining to a given frequency, (ii) by an additive oscillatory response independent of the ongoing oscillation, (iii) by modulating the amplitude of the stimulus-relevant oscillations, or (iv) by an additive non-oscillatory (slower) response. This (not exhaustive) list of effects was considered given previous literature (Shah, et al., 2004; Mazaheri & Jensen, 2006; Makeig, et al., 2002; Vidaurre, et al., 2021), and each effect may be underpinned by distinct neural mechanisms”

The general narrative and focus of the paper could also be improved. It might help to start off with an outline of what the goal is at the start of the paper and then explicitly discuss how each of the steps works toward that goal. For example, I got the idea that the goal was to capture specific properties of an empirical TGM. If this was the case, the empirical TGM could be placed in the main body of the text as a reference picture for all simulated TGMs. For each simulation step, it could be emphasized more clearly exactly which features of the TGM is captured and what that means for interpreting these features in real data.

Thank you. To clarify the purpose of the paper better, I have brought Figure 2 to the front (before a Supplementary Figure), and in the first part of Results I have now added:

“Figure 2 shows a TGM for an example subject, where some archetypal characteristics are highlighted. In the experiments below, specifically, I focus on the strong narrow diagonal at the beginning of the trial, the broadening of accuracy later in the trial, and the vertical/horizontal bars of higher-than-chance accuracy. Importantly, this specific example in Figure 2 is only meant as a reference, and therefore I did not optimise the model hyperparameters to this TGM (except in the last subsection), or showed any quantitative metric of similarity. ”

I have enunciated the goals more clearly in the Introduction:

“To gain insight into what aspects of the signal underpin decoding accuracy, and therefore the most stable aspects of stimulus processing, …”

Relatedly, it would be good to connect the various signal properties to possible neurobiological mechanisms. I appreciate that the author tries to remain neutral on this in the introduction, but I think it would greatly increase the implications of the analysis if it is made clearer how it could eventually help us understand neural processes.

The Reviewer is right in pointing out that I preferred to remain neutral on this. While I have still kept that tone of neutrality throughout the paper, I have now included the following sentence as an example of a neurobiological question that could be investigated with the model:

“For example, it is not completely clear the extent to which stimulus processing is sustained by oscillations, and disentangling these effects can help resolving this question.”

And, more generally,

“Genephys has different available types of effect, including phase resets, additive damped oscillations, amplitude modulations, and non-oscillatory responses. All of these elements, which may relate to distinct neurobiological mechanisms, are configurable and can be combined to generate a plethora of TGMs that, in turn, can be contrasted to specific empirical TGMs. This way, we can gain insight on what mechanisms might be at play in a given task. ”

Line 57: this sentence is very long, making it hard to follow, could you break up into smaller parts?

Thank you. The sentence is fragmented now.

Please replace angular frequencies with frequencies in Hertz for clarity.

Here I have preferred to stick to angular frequencies because it is more general than if I talk about Hertz, because that would entail having a specific sampling frequency. I think doing so would create confusion precisely of the sorts that I am trying to clarify in this revision: that is, that these results are not specific of one TGM but reflect general features that we see broadly in the literature.

There are quite some types throughout the paper, please recheck

Thank you. I have revised and have made my best to clear them out.

eLife assessment

This study presents a valuable finding on developing a state-of-the-art generative model of brain electrophysiological signals to explain temporal decoding matrices widely used in cognitive neuroscience. The evidence supporting the authors' claims is convincing. The results will be strengthened by providing more clear mappings between neurobiological mechanisms and signal generators in the model. The work will be of interest to cognitive neuroscientists using electrophysiological recordings.

Reviewer #1 (Public Review):

With genephys, the author provides a generative model of brain responses to stimulation. This generative model allows to mimic specific parameters of a brain response at the sensor level, to test the impact of those parameters on critical analytic methods utilized on real M/EEG data. Specifically, they compare the decoding output for differently set parameters to the decoding pattern observed in a classical passive viewing study in terms of the resulting temporal generalization matrix (TGM). They identify that the correspondence between the mimicked and the experimental TGM to depend on an oscillatory component that spans multiple channels, frequencies, and latencies of response; and an additive, slower response with a specific (cross-frequency) relation to the phase of the oscillatory, faster component.

A strength of the article is that it considers the complexity of neural data that contribute to the findings obtained in stimulation experiments. An additional strength is the provision of a Python package that allows scientists to explore the potential contribution of different aspects of neural signals to obtained experimental data and thereby to potentially test their theoretical assumptions critical parameters that contribute to their experimental data.

A weakness of the paper is that the power of the model is illustrated for only one specific set of parameters, added in a stepwise manner and the comparison to on specific empirical TGM, assumed to be prototypical; And that this comparison remains descriptive. (That is could a different selection of parameters lead to similar results and is there TGM data which matches these settings less well.) It further remained unclear to me, which implications may be drawn from the generative model, following from the capacities to mimic this specific TGM (i) for more complex cases, such as the comparison between experimental conditions, and (ii) about the complex nature of neural processes involved.

Towards this end I would appreciate (i) a more profound explanation of the conclusions that can be drawn from this specific showcase, including potential limitations, as well as wider considerations of how scientists may empower the generative model to (ii) understand their experimental data better and (iii) which added value the model may have in understanding the nature of underlaying brain mechanism (rather than a mere technical characterization of sensor data).

Reviewer #2 (Public Review):

This paper introduces a new model that aims to explain the generators of temporal decoding matrices (TGMs) in terms of underlying signal properties. This is important because TGMs are regularly used to investigate neural mechanisms underlying cognitive processes, but their interpretation in terms of underlying signals often remains unclear. Furthermore, neural signals are often variant over different instances of stimulation despite behaviour being relatively stable. The author aims to tackle these concerns by developing a generative model of electrophysiological data and then showing how different parameterizations can explain different features of TGMs. The developed technique is able to capture empirical observations in terms of fundamental signal properties. Specifically, the model shows that complexity is necessary in terms of spatial configuration, frequencies and latencies to obtain a TGM that is comparable to empirical data.

The major strength of the paper is that the novel technique has the potential to further our understanding of the generators of electrophysiological signals which are an important way to understand brain function. The paper clearly outlines how the method can be used to capture empirical data. Furthermore, the used techniques are state-of-the-art and the developed model is publicly shared in open source code.

On the other hand, there is no unambiguous mapping between neurobiological mechanisms and different signal generators, making it hard to draw firm conclusions about neural underpinnings based on this analysis.

eLife assessment

Studies of synaptic development and plasticity in the nematode C. elegans have been limited by the difficulty of rapid, accurate assessments of synaptic structure. Here, with a series of convincing studies, the authors introduce and validate a valuable computational pipeline, "WormPsyQi," that allows rapid, reproducible quantitation of fluorescent synaptic puncta while minimizing human error and bias. The authors also describe a new set of strains carrying synaptic markers. Together, these tools should provide groups studying this model system with the ability to quantitatively characterize chemical and electrical synapses, even in densely packed regions in 3D space such as the nerve ring.

Reviewer #1 (Public Review):

Summary:

The paper by Majeed et al has a valuable and worthwhile aim: to provide a set of tools to standardize the quantification of synapses using fluorescent markers in the nematode C. elegans. Using current approaches, the identification of synapses using fluorescent markers is tedious and subject to significant inter-experimenter variability. Majeed et al successfully develop and validate a computational pipeline called "WormPsyQi" that overcomes some of these obstacles and will be a powerful resource for many C. elegans neurobiologists.

Strengths:

The computational pipeline is rigorously validated and shown to accurately quantitate fluorescent puncta, at least as well as human experimenters. The inclusion of a mask - a region of interest defined by a cytoplasmic marker - is a powerful and useful approach. Users can take advantage of one of four pre-trained neural networks, or train their own. The software is freely available and appears to be user-friendly. A series of rigorous experiments demonstrates the utility of the pipeline for measuring differences in the number of synaptic puncta between sexes and across developmental stages. Neuron-to-neuron heterogeneity in patterns of synaptic growth during development are convincingly demonstrated. Weaknesses and caveats are realistically discussed.

Reviewer #2 (Public Review):

Summary:

This paper nicely introduces WormPsyQi, an imaging analysis pipeline that effectively quantifies synaptically localized fluorescent signals in C. elegans through high-throughput automation. This toolkit is particularly valuable for the analysis of densely packed regions in 3D space, such as the nerve ring. The authors applied WormPsyQi to various aspects, including the examination of sexually dimorphic synaptic connectivity, presynaptic markers in eight head neurons, five GRASP reporters, electrical synapses, the enteric nervous system, and developmental synapse comparisons. Furthermore, they validated WormPsyQi's accuracy by comparing its results to manual analysis.

Strengths:

Overall, the experiments are well done, and their toolkit demonstrates significant potential and offers a valuable resource to the C. elegans community. This will expand the range of possibilities for studying synapses in C. elegans.

Reviewer #3 (Public Review):

Summary:

In this manuscript, the authors present a new automated image analysis pipeline named WormPsyQi which allows researchers to quantify various parameters of synapses in C. elegans. Using a collection of newly generated transgenic strains in which synaptic proteins are tagged with fluorescent proteins, the authors showed that WormPsyQi can reliably detect puncta of synaptic proteins, and measure several parameters, including puncta number, location, and size.

Strengths:

The image analysis of fluorescently labeled synaptic (or other types of) puncta pattern requires extensive experience such that one can tell which puncta likely represent bona fide synapse or background noise. The authors showed that WormPsyQi nicely reproduced the quantifications done manually for most of the marker strains they tested. Many researchers conducting such types of quantifications would receive significant benefits in saving their time by utilizing the pipeline developed by the authors. The collection of new markers would also help researchers examine synapse patterning in different neuron types which may have unique mechanisms in synapse assembly and specificity. The authors describe the limitations of the use of toolkits and potential solutions users can take.

Author Response

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

We are grateful for the comments and suggestions from the reviewers and have followed the recommendation in producing our revised manuscript. We have modified the text and performed additional statistical analysis as detailed below, which we believe has improved the overall manuscript.

Reviewer #1 (Public Review):

Establishing direct links between the neuronal connectivity information of connectomics datasets with circuit physiology and behavior and exciting current research area in neurobiology. Until recently, studies of aggression in Drosophila had been conducted largely in males, and many of the neurons involved in this behavior are male-specific clusters. Since the currently available fly brain connectomes come from female brains, their applicability for the study of the circuitry underlying aggressive behavior is very limited.

The authors have previously used the Janelia hemibrain connectome paired with behavior analysis to show that activating either the aIPg or pC1d cell types can induce short-term aggression in females, while activation of other PC1 clusters (a-c and e) does not. Here they expand on those findings, showing that optogenetic stimulation of aIPg neurons was sufficient to promote an aggressive internal state lasting at least 10 minutes following a 30-second activation. In addition, the authors show that while stimulation of PC1d alone is not sufficient to induce this persistent aggressive state, simultaneous activation of PC1d + PC1e is, suggesting a synergistic effect. Connectomics analysis performed in the authors' previous study had shown that PC1d and aIPg are interconnected. However, silencing pC1d neuronal activity did not reduce aIPg-evoked persistent aggression, indicating that the aggressive state did not depend on pC1d-aIPg recurrent connectivity.

The conclusions are well supported by the data, and the results presented in this manuscript represent an important contribution to our understanding of the neuronal circuitry underlying female aggression.

Reviewer #1 (Recommendations For The Authors):

1. Previously, the authors have shown that the activation of PC1e alone does not induce female aggression. In this study, they investigate the role of aIPg, PC1d, or PC1d+e on aggression persistence, but they do not explore the effect of activation of PC1e alone. It is possible that PC1e activation may not produce an immediate short-term effect but could lead to a gradual increase in aggression over time, potentially explaining at least in part the observed effect upon PC1d+e activation. Incorporating an examination of the long-term impact of PC1e activation on aggression could provide valuable information.

We did perform mixed pair experiments with the pC1e-SS1 line from the Schretter et al. (2020) paper and did not find any significant changes in aggression over time in this setup as well. We have now added a reference to these experiments in the revised submission in lines 135 to 136.

1. Some important controls are missing: flies with the genetic combinations employed in the activation experiments shown in Figure 2 but in the absence of activation and under the exact same conditions and for a similar observation period.

For Figure 2, we used an empty split-Gal4 driver as a genetic control for our activation paradigms. As these flies contain the same number of copies of mini-white while not labeling the targeted cell types, we believe that they provide an appropriate control for these experiments. The control information is specified in all figure legends as well.

1. The quantification shown in Fig 3- Supplementary Figure 1 shows no effect during stimulation (13 s + 15s), but based on the plots of Figure 3, there may be an effect of silencing PC1d on aIPg-induced aggression during the initial 13 second period. Those two time periods (13 s vs 15 s) could be quantified separately to determine if this is the case.

We examined the two stimulation periods separately and did not find any significant differences in either period (13s period, p = 0.2978; 15s period, p = 0.6650). We have now added this into the figure legend for Figure 3 and Figure 3 supplement 1.

1. Expression of Kir2.1 in pC1d neurons while aIPg neurons were activated did not suppress aggression after aIPg stimulation, suggesting that connections from pC1d neurons are not necessary for the persistent aggressive state promoted by aIPg. Since previously the authors have shown that TNT-mediated inhibition of aIPg reduces aggression, the reciprocal experiment would be informative: determining if stimulation of PC1d+e no longer produces persistent aggression when aIPg neurons are silenced.

In this manuscript, we were primarily testing if the connections from aIPg to pC1d were necessary for the persistent aggressive state induced by aIPg activation. Therefore, we believe the suggested experiment is beyond the scope of the current manuscript.

1. How many times was each experiment repeated? This is important information and should be in the methods section for each type of experiment or in each figure legend.

We have now added this information in the appropriate figure legends.

1. Determining the effect on persistent aggression of silencing sNPF (for example via RNAi or Crispr-Cas9 mediated mutagenesis) in aIPG neurons would be an important addition to the manuscript. If peptidergic signaling is underlying the persistence phenotype of aIPg neurons, that would explain why the recurrent connectivity found between those cells and the PC1 cluster does not play a role.

We agree with the reviewer that this would be a logical next step in extending this work.

Reviewer #2 (Public Review):

The mechanisms that mediate female aggression remain poorly understood. Chiu, Schretter, and colleagues, employed circuit dissection techniques to tease apart the specific roles of particular doublesex and fruitless expressing neurons in the fly Drosophila in generating a persistent aggressive state. They find that activating the fruitless positive alPg neurons, generated an aggressive state that persisted for >10min after the stimulation ended. Similarly, activating the doublesex positive pC1de neurons also generated a persistent state. Activating pC1d or pC1e individually did not induce a persistent state. Interestingly, while neural activation of alPGs and pC1d+e neurons induced persistent behavioural states it did not induce persistent activity in the neurons being activated.

The conclusions of this paper are well supported by the data, there were only a few points where clarification might help:

1. Figure 3 is a little confusing. This is a circuit behavioural epistasis experiment where the authors activate alPg with CsChrimson while inhibiting pC1d with Kir2.1. In Fig. 2 flies were separated for 10 min following stimulation which allowed for identification of a persistent state. However, in Fig 3 it appears as if flies were allowed to freely interact during and immediately post-stimulation. It is unclear why flies were not separated as in Fig. 2, which makes it difficult to compare the two results. Some discussion of this point would help. Also, from the rasters it appears as if inhibition of pC1d reduced aggression induced by alPg during the stimulation period. Is this true?

We thank the reviewer for pointing out the need for clarification and we have modified the legend in Figure 3 to address the points raised. The flies were allowed to freely interact during the experiments shown in Figure 3 and we have added this information to the figure legend. To obtain a high level of aggressive behavior that would make it easier to observe a suppression of aggression, the epistasis experiments were performed with freely moving same-genotype pairs. The level of aggression triggered by the generation 1 LexA line labeling aIPg was lower than that observed when using with the aIPg-SS GAL4 line. The experiment was performed as in Schretter et al. (2020) where we found that aIPg activation induced persistent fighting in same genotype pairs. We have added a brief explanation in lines 152 to 155.

Inhibition of pC1d does not significantly reduce the overall aggression induced by aIPg stimulation in the 13s + 15s period. We also examined the differences within the two stimulation periods and did not find any significant differences (13s period, p = 0.2978; 15s period, p = 0.6650). We have now added this information to the figure legends for Figure 3 and Figure 3 supplement 1.

1. pC1e neurons also have recurrent connectivity with alPg neurons. It might help to also discuss the potential role of this arm of the microcircuit.

We thank the review for this suggestion. The number of synapses that aIPg sends back to pC1e is a very low proportion of its total output (0.177%). However, based on the experiments that we have performed, we cannot rule out that this microcircuit might contribute to maintaining persistence. We have added this point into the discussion in lines 210 to 211.

Reviewer #2 (Recommendations For The Authors):

1. Line 129-130: A citation for group-housed flies showing lower aggression would be helpful.

We have now added in the reference to Chiu et al. (2021), as they showed this effect for females, in line 130.

1. Figure 2 - figure supplement 1: In the legend, change "when pC1d neurons were stimulation" to "when pC1d neurons were stimulated".

We thank the reviewer for finding this error and have now corrected this.

Reviewer #3 (Public Review):

Two studies published in 2020 independently identified the alPg, pC1d, and pC1e neurons to be involved in initiating and maintaining a state of aggression in female Drosophila. Both studies combined behavioural analyses, optogenitic manipulation of neurons, and connectomics. One of these studies proposed that the extensive interconnections seen between the alPg and pC1d+e neurons might represent a recurrent motif known to support persistent behvioural states in other systems. In this manuscript, the authors test this idea and report that their data do not support it. Specifically, they report that alPg or pC1d+e (but not pC1d alone) can initiate a persistent state of aggression. But they find that the persistent aggressive state is maintained even when the pC1d neurons are inactivated. Finally, they show that neither of these neurons themselves sustains neuronal activity upon stimulation, nor do either of them induce a persistent activity in the other. Together, their data suggest that the recurrent connection between alPg and pC1d is not what supports the persistent state. The data underlying these claims are convincing. A possibility to explore before ruling out recurrent motifs (at this circuit level) in maintaining aggression is that the connections between alPg and pC1e can compensate for the loss of pC1e. Overall, the study is important and will be of interest to those who study the circuit basis of persistent behavioural states, but also to neuroscientists in general.

Reviewer #3 (Recommendations For The Authors):

I enjoyed reading this manuscript for its clarity in writing and data presentation.

I would like the authors to comment on the possibility that pC1e can compensate for the loss of pC1d. It is possible that if they silence both pC1d+e in the context of alPg activation, the persistent aggression is lost?

We agree with the reviewer that this is an intriguing hypothesis. In order to examine if pC1e does compensate for pC1d, we would need to also activate pC1e while inhibiting pC1d. However, such an experiment is not currently possible as we do not have a LexA line that specifically labels either pC1d or pC1e alone.

For the pC1d+e silencing experiments, we were primarily testing to see if the most prominent recurrent connection, which is between pC1d and aIPg, was responsible for the behavioral persistence. We agree with the reviewer that this would be a logical follow up experiment to be performed in the future.

Have the authors looked for activity in the pC1e neuron upon simulation of alPg? (Deutsch et al 2020 observed many regions in the brain that maintained sustained activity upon pC1d+e stimulation.)

We have not examined this activity. We agree that this would be a good follow up experiment; however, we believe it is beyond the scope of the current work.

Would the more appropriate experiment in Figure 4c be the co-stimulation of pC1d+e while imaging from alPg?

For these experiments, we were testing to see if the most prominent recurrent connection, which is between pC1d and aIPg, was responsible for the behavioral persistence. We agree with the reviewer that this would be a good follow up experiment

eLife assessment

This study by Chiu and colleagues is a valuable contribution to the study of the circuitry of aggressive behaviours and of mechanisms that generate persistent behavioural states. The authors find that activation of two interconnected sets of neurons results in an increase in female aggression. The data ruling out recurrent connectivity between these clusters underlying this persistent state are convincing.

Reviewer #1 (Public Review):

Establishing direct links between the neuronal connectivity information of connectomics datasets with circuit physiology and behavior and exciting current research area in neurobiology. Until recently, studies of aggression in Drosophila had been conducted largely in males, and many of the neurons involved in this behavior are male-specific clusters. Since the currently available fly brain connectomes come from female brains, their applicability for the study of the circuitry underlying aggressive behavior is very limited.

The authors have previously used the Janelia hemibrain connectome paired with behavior analysis to show that activating either the aIPg or pC1d cell types can induce short term aggression in females, while activation of other PC1 clusters (a-c and e) does not. Here they expand on those findings, showing that optogenetic stimulation of aIPg neurons was sufficient to promote an aggressive internal state lasting at least 10 minutes following a 30 second activation. In addition, authors show that while stimulation of PC1d alone is not sufficient to induce this persistent aggressive state, simultaneous activation of PC1d + PC1e is, suggesting a synergistic effect. Connectomics analysis performed in the authors' previous study had shown that PC1d and aIPg are interconnected. However, silencing pC1d neuronal activity did not reduce aIPg-evoked persistent aggression, indicating that the aggressive state did not depend on pC1d-aIPg recurrent connectivity.

The conclusions are well supported by the data, and the results presented in this manuscript represent an important contribution to our understanding of the neuronal circuitry underlying female aggression.

Reviewer #2 (Public Review):

The mechanisms that mediate female aggression remain poorly understood. Chiu, Schretter, and colleagues, employed circuit dissection techniques to tease apart the specific roles of particular doublesex and fruitless expressing neurons in the fly Drosophila in generating a persistent aggressive state. They find that activating the fruitless positive alPg neurons, generated an aggressive state that persisted for >10min after the stimulation ended. Similarly, activating the doublesex positive pC1de neurons also generated a peristent state. Activating pC1d or pC1e individually did not induce a persistent state. Interestingly, while neural activation of alPGs and pC1d+e neurons induced a persistent behavioural states it did not induce persistent activity in the neurons being activated.

The authors have revised the manuscript in accordance with comments of the reviewers. The conclusions of this paper are by and large well supported by the data. These data will be a useful addition to the literature on the circuit basis of female aggression, and open up intriguing avenues for further studies to explore.

Author Response

Reviewer #1 (Public Review):

Summary:

This study uses whole genome sequencing to characterise the population structure and genetic diversity of a collection of 58 isolates of E. coli associated with neonatal meningitis (NMEC) from seven countries, including 52 isolates that the authors sequenced themselves and a further 6 publicly available genome sequences. Additionally, the study used sequencing to investigate three case studies of apparent relapse. The data show that in all three cases, the relapse was caused by the same NMEC strain as the initial infection. In two cases they also found evidence for gut persistence of the NMEC strain, which may act as a reservoir for persistence and reinfection in neonates. This finding is of clinical importance as it suggests that decolonisation of the gut could be helpful in preventing relapse of meningitis in NMEC patients.

Strengths:

The study presents complete genome sequences for n=18 diverse isolates, which will serve as useful references for future studies of NMEC. The genomic analyses are high quality, the population genomic analyses are comprehensive and the case study investigations are convincing.

We agree

Weaknesses:

The NMEC collection described in the study includes isolates from just seven countries. The majority (n=51/58, 88%) are from high-income countries in Europe, Australia, or North America; the rest are from Cambodia (n=7, 12%). Therefore it is not clear how well the results reflect the global diversity of NMEC, nor the populations of NMEC affecting the most populous regions.

The virulence factors section highlights several potentially interesting genes that are present at apparently high frequency in the NMEC genomes; however, without knowing their frequency in the broader E. coli population it is hard to know the significance of this.

We acknowledged the limitations of our NMEC collection in the Discussion. We agree the prevalence of virulence factors in our collection is interesting. The limited size of our collection prevented further evaluation of the prevalence of these virulence factors in a broader E. coli population.

Reviewer #2 (Public Review):

Summary:

In this work, the authors present a robust genomic dataset profiling 58 isolates of neonatal meningitis-causing E. coli (NMEC), the largest such cohort to be profiled to date. The authors provide genomic information on virulence and antibiotic resistance genomic markers, as well as serotype and capsule information. They go on to probe three cases in which infants presented with recurrent febrile infection and meningitis and provide evidence indicating that the original isolate is likely causing the second infection and that an asymptomatic reservoir exists in the gut. Accompanying these results, the authors demonstrate that gut dysbiosis coincides with the meningitis.

Strengths:

The genomics work is meticulously done, utilizing long-read sequencing.

The cohort of isolates is the largest to be sampled to date.

The findings are significant, illuminating the presence of a gut reservoir in infants with repeating infection.

We agree

Weaknesses:

Although the cohort of isolates is large, there is no global representation, entirely omitting Africa and the Americas. This is acknowledged by the group in the discussion, however, it would make the study much more compelling if there was global representation.

We agree. In the Discussion we state this is likely a reflection of the difficulty in acquiring isolates causing neonatal meningitis, in particular from countries with limited microbiology and pathology resources.

Reviewer #3 (Public Review):

Summary:

In this manuscript, Schembri et al performed a molecular analysis by WGS of 52 E. coli strains identified as "causing neonatal meningitis" from several countries and isolated from 1974 to 2020. Sequence types, virulence genes content as well as antibiotic-resistant genes are depicted. In the second part, they also described three cases of relapse and analysed their respective strains as well as the microbiome of three neonates during their relapse. For one patient the same E. coli strain was found in blood and stool (this patient had no meningitis). For two patients microbiome analysis revealed a severe dysbiosis.

Major comments:

Although the authors announce in their title that they study E. coli that cause neonatal meningitis and in methods stipulate that they had a collection of 52 NMEC, we found in Supplementary Table 1, 29 strains (therefore most of the strains) isolated from blood and not CSF. This is a major limitation since only strains isolated from CSF can be designated with certainty as NMEC even if a pleiocytose is observed in the CSF. A very troubling data is the description of patient two with a relapse infection. As stated in the text line 225, CSF microscopy was normal and culture was negative for this patient! Therefore it is clear that patient without meningitis has been included in this study.

We have reviewed the clinical data for our 52 NMEC isolates, noting that for some of the older Finish isolates we relied on previous publications. This data is shown in Table S1. To address the Reviewer’s comment, we have added the following text to the methods section (new text underlined).

‘The collection comprised 42 isolates from confirmed meningitis cases (29 cultured from CSF and 13 cultured from blood) and 10 isolates from clinically diagnosed meningitis cases (all cultured from blood).’

Patient 2 was initially diagnosed with meningitis based on a positive blood culture in the presence of CSF pleocytosis (>300 WBCs, >95% polymorphs). We understand there may be some confusion with reference to a relapsed infection, which we now more accurately describe as recrudescent invasive infection in the revised manuscript.

Another major limitation (not stated in the discussion) is the absence of clinical information on neonates especially the weeks of gestation. It is well known that the risk of infection is dramatically increased in preterm neonates due to their immature immunity. Therefore E. coli causing infection in preterm neonates are not comparable to those causing infection in term neonates notably in their virulence gene content. Indeed, it is mentioned that at least eight strains did not possess a capsule, we can speculate that neonates were preterm, but this information is lacking. The ages of neonates are also lacking. The possible source of infection is not mentioned, notably urinary tract infection. This may have also an impact on the content of VF.

We agree. In the Discussion we now note the following (new text underlined):

‘… we did not have clinical data on the weeks of gestation for all patients, and thus could not compare virulence factors from NMEC isolated from preterm versus term infants.’

Submission to Medrxiv, a requirement for review of our manuscript at eLife, necessitated the removal of some patient identifying information, including precise age and detailed medical history.

Sequence analysis reveals the predominance of ST95 and ST1193 in this collection. The high incidence of ST95 is not surprising and well previously described, therefore, the concluding sentence line 132 indicating that ST95 E. coli should exhibit specific virulence features associated with their capacity to cause NM does not add anything. On the contrary, the high incidence of ST1193 is of interest and should have been discussed more in detail. Which specific virulence factors do they harbor? Any hypothesis explaining their emergence in neonates?

We compared the virulence factors of ST95 and ST1193 and summarized this information in Figure 4. We also discussed how the K1 polysialic acid capsule in ST95 and ST1193 could contribute to the emergence of these STs in NM. Specifically, we stated the following: ‘We speculate this is due to the prevailing K1 polysialic acid capsule serotype found in ST95 and the newly emerged ST1193 clone [22, 37] in combination with other virulence factors [15, 28, 29] (Figure 4) and the immature immune system of preterm infants.’

In the paragraph depicted the VF it is only stated that ST95 contained significantly more VF than the ST1193 strains. And so what? By the way "significantly" is not documented: n=?, p=?

We compared the prevalence of known virulence factors between ST95 and ST1193, and showed that ST95 strains in our collection contained significantly more virulence factors than the ST1193 strains. The P-value and the statistical test used were included in Supplementary Figure 3. To address the reviewers concern, we have now also added this to the main manuscript text as follows (new text underlined):

‘Direct comparison of virulence factors between ST95 and ST1193, the two most dominant NMEC STs, revealed that the ST95 isolates (n = 20) contained significantly more virulence factors than the ST1193 isolates (n=9), p-value < 0.001, Mann-Whitney two-tailed unpaired test (Supplementary Table 1, Supplementary Figure 3).’

The complete sequence of 18 strains is not clear. Results of Supplementary Table 2 are presented in the text and are not discussed.

NMEC isolates that were completely sequenced in this study are indicated in bold and marked with an asterisk in Figure 1. This information is indicated in the figure legend and was provided in the original submission. All information regarding genomic island composition and location, virulence genes and plasmid and prophage diversity is included in Supplementary Table 2. This information is highly descriptive and thus we elected not to include it as text in the main manuscript.

46 years is a very long time for such a small number of strains, making it difficult to put forward epidemiological or evolutionary theories. In the analysis of antibiotic resistance, there are no ESBLs. However, Ding's article (reference 34) and other authors showed that ESBLs are emerging in E. coli neonatal infection. These strains are a major threat that should be studied, unfortunately, the authors haven't had the opportunity to characterize such strains in their manuscript.

We agree 46 years is a long time-span. The study by Ding et al examined 56 isolates comprised of 25 different STs isolated in China from 2009-2015, with ST1193 (n=12) and ST95 (n=10) the most common. Our study examined 58 isolates comprised of 22 different STs isolated in seven different geographic regions from 1974-2020, with ST1193 (n=9) and ST95 (n=20) the most common. Thus, despite differences in the geographic regions from which isolates in the two studies were sourced, there are similarities in the most common STs identified. The fact that we observed less antibiotic resistance, including a lack of ESBL genes, in ST1193 is likely due to the different regions from which the isolates were sourced. We acknowledged and discussed the potential of ST1193 harbouring multidrug resistance including ESBLs in our manuscript as follows:

‘Concerningly, the ST1193 strains examined here carry genes encoding several aminoglycoside-modifying enzymes, generating a resistance profile that may lead to the clinical failure of empiric regimens such as ampicillin and gentamicin, a therapeutic combination used in many settings to treat NM and early-onset sepsis [35, 36]. This, in combination with reports of co-resistance to third-generation cephalosporins for some ST1193 strains [22, 34], would limit the choice of antibiotic treatment.’

Second part of the manuscript:

The three patients who relapsed had a late neonatal infection (> 3 days) with respective ages of 6 days, 7 weeks, and 3 weeks. We do not know whether they are former preterm newborns (no term specified) or whether they have received antibiotics in the meantime.

As noted above, patient ages were not disclosed to comply with submission to Medrxiv, a requirement for review of our manuscript at eLife.

Patient 1: Although this patient had a pleiocytose in CSF, the culture was negative which is surprising and no explanation is provided. Therefore, the diagnosis of meningitis is not certain. Pleiocytose without meningitis has been previously described in neonates with severe sepsis. Line 215: no immunological abnormalities were identified (no details are given).

We respectfully disagree with the reviewer. The diagnosis of meningitis is made unequivocally by the presence of a clearly abnormal CSF microscopy (2430 WBCs) and an invasive E. coli from blood culture. This does not seem controversial to the authors. We had believed it unnecessary to include this corroborative evidence, but have added the following to support our assertion:

‘The child was diagnosed with meningitis based on a cerebrospinal fluid (CSF) pleocytosis (>2000 white blood cells; WBCs, low glucose, elevated protein), positive CSF E. coli PCR and a positive blood culture for E. coli (MS21522).’

On the contrary, the authors are surprised by the statement that CSF pleocytosis occurs in neonatal sepsis ‘without meningitis’ and do not know of any definitions of neonatal meningitis that are not tied to the presence of a CSF pleocytosis. Furthermore, the later isolation of E. coli from the CSF during the relapsed infection re-enforces the initial diagnosis.

Patient 2: This patient had a recurrence of bacteremia without meningitis (line 225: CSF microscopy was normal and culture negative!). This case should be deleted.

In a similar vein to the previous comment, we respectfully assert that this patient has clear evidence of meningitis (330 WBCs in the CSF, taken 24h after initiation of antibiotic treatment). In this case, molecular testing was not performed as, under the principle of diagnostic stewardship, it was not considered necessary by the clinical microbiologists and treating clinicians following the culture of E. coli in the bloodstream. We agree that this is not a case of recurrent meningitis, but our intention was to highlight the recrudescence of an invasive infection (urinary sepsis requiring admission to hospital and intravenous antibiotics) which we hypothesise has arisen from the intestinal reservoir. We did not state that all patients suffered from relapsed meningitis.

Despite this, to address this reviewers concern, we have changed all reference to ‘relapsed infection’ to now read ‘recrudescent invasive infection’ in the revised manuscript.

Patient 3: This patient had two relapses which is exceptional and may suggest the existence of a congenital malformation or a neurological complication such as abscess or empyema therefore, "imaging studies" should be detailed.

This patient underwent extensive imaging investigation to rule out a hidden source. This included repeated MRI imaging of head and spine, CT imaging of head and chest, USS imaging of abdomen and pelvis and nuclear medicine imaging to detect a subtle meningeal defect and CSF leak. All tests were normal, and no abscess or empyema found.

We have modified the text to include this information:

Text in original submission: ‘Imaging studies and immunological work-up were normal.’

New text in revised manuscript (underlined): ‘Extensive imaging studies including repeated MRI imaging of the head and spine, CT imaging of the head and chest, ultrasound imaging of abdomen and pelvis, and nuclear medicine imaging did not show a congenital malformation or abscess. Immunological work-up did not show a known primary immunodeficiency. At two years of age, speech delay is reported but no other developmental abnormality.’

The authors suggest a link between intestinal dysbiosis and relapse in three patients. However, the fecal microbiomes of patients without relapse were not analysed, so no comparison is possible. Moreover, dysbiosis after several weeks of antibiotic treatment in a patient hospitalized for a long time is not unexpected. Therefore, it's impossible to make any assumption or draw any conclusion. This part of the manuscript is purely descriptive. Finally, the authors should be more prudent when they state in line 289 "we also provide direct evidence to implicate the gut as a reservoir [...] antibiotic treatment". Indeed the gut colonization of the mothers with the same strain may be also a reservoir (as stated in the discussion line 336). Finally, the authors do not discuss the potential role of ceftriaxone vs cefotaxime in the dysbiosis observed. Ceftriaxone may have a major impact on the microbiota due to its digestive elimination.

We addressed the limitations of our study in the Discussion, including that we did not have access to urine or stool samples from the mother of the infants that suffered recrudescence, and thus cannot rule out mother-to-child transmission as a mechanism of reinfection. We have now added that we did not have clinical data on the weeks of gestation for all patients, and thus could not compare virulence factors from NMEC isolated from preterm versus term infants. The limitations of our study are summarised as follows in the Discussion (new text underlined):

‘This study had several limitations. First, our NMEC strain collection was restricted to seven geographic regions, a reflection of the difficulty in acquiring strains causing this disease. Second, we did not have access to a complete set of stool samples spanning pre- and post-treatment in the patients that suffered NM and recrudescent invasive infection. This impacted our capacity to monitor E. coli persistence and evaluate the effect of antibiotic treatment on changes in the microbiome over time. Third, we did not have access to urine or stool samples from the mother of the infants that suffered recrudescence, and thus cannot rule out mother-to-child transmission as a mechanism of reinfection. Finally, we did not have clinical data on the weeks of gestation for all patients, and thus could not compare virulence factors from NMEC isolated from preterm versus term infants.’

Author Response

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

Public Reviews:

Reviewer #1 (Public Review):

Peng et al develop a computational method to predict/rank transcription factors (TFs) according to their likelihood of being pioneer transcription factors--factors that are capable of binding nucleosomes--using ChIP-seq for 225 human transcription factors, MNase-seq and DNase-seq data from five cell lines. The authors developed relatively straightforward, easy to interpret computational methods that leverage the potential for MNase-seq to enable relatively precise identification of the nucleosome dyad. Using an established smoothing approach and local peak identification methods to estimate positions together with identification of ChIP-seq peaks and motifs within those peaks which they referred to as "ChIP-seq motifs", they were able to quantify "motif profiles" and their density in nucleosome regions (NRs) and nucleosome depleted regions (NDRs) relative to their estimated nucleosome dyad positions. Using these profiles, they arrived at an odd-ratio based motif enrichment score along with a Fisher's exact test to assess the odds and significance that a given transcription factor's ChIP-seq motifs are enriched in NRs compared to NDRs, hence, its potential to be a pioneer transcription factor. They showed that known pioneer transcription factors had among the highest enrichment scores, and they could identify a number of relatively novel pioneer TFs with high enrichment scores and relatively high expression in their corresponding cell line. They used multiple validation approaches including (1) calculating the ROC-AUC and Matthews correlation coefficient (MCC) and generating ROC and precision-recall curves associated with their enrichment score based on 32 known pioneer TFs among their 225 TFs which they used as positives and the remaining TFs (among the 225) as negatives; (2) use of the literature to note that known pioneer TFs that acted as key regulators of embryonic stem cell differentiation had a highest enrichment scores; (3) comparison of their enrichment scores to three classes of TFs defined by protein microarray and electromobility shift assays (1. strong binder to free and nucleosomal DNA, 2. weak binder to free and nucleosomal DNA, 3. strong binding to free but not nucleosomal DNA); and (4) correlation between their calculated TF motif nucleosome end/dyad binding ratio and relevant data from an NCAP-SELEX experiment. They also characterize the spatial distribution of TF motif binding relative to the dyad by (1) correlating TF motif density and nucleosome occupancy and (2) clustering TF motif binding profiles relative to their distance from the dyad and identifying 6 clusters.

The strengths of this paper are the use of MNase-seq data to define relatively precise dyad positions and ChIP-seq data together with motif analysis to arrive at relatively accurate TF binding profiles relative to dyad positions in NRs as well as in NDRs. This allowed them to use a relatively simple odds ratio based enrichment score which performs well in identifying known pioneer TFs. Moreover, their validation approaches either produced highly significant or reasonable, trending results.

The weaknesses of the paper are relatively minor, and the authors do a good job describing the limitations of the data and approach.

Reviewer #2 (Public Review):

In this study, the authors utilize a compendium of public genomic data to identify transcription factors (TF) that can identify their DNA binding motifs in the presence of nuclosome-wrapped chromatin and convert the chromatin to open chromatin. This class of TFs are termed Pioneer TFs (PTFs). A major strength of the study is the concept, whose premise is that motifs bound by PTFs (assessed by ChIP-seq for the respective TFs) should be present in both "closed" nucleosome wrapped DNA regions (measured by MNase-seq) as well as open regions (measured by DNAseI-seq) because the PTFs are able to open the chromatin. Use of multiple ENCODE cell lines, including the H1 stem cell line, enabled the authors to assess if binding at motifs changes from closed to open. Typical, non-PTF TFs are expected to only bind motifs in open chromatin regions (measured by DNaseI-seq) and not in regions closed in any cell type. This study contributes to the field a validation of PTFs that are already known to have pioneering activity and presents an interesting approach to quantify PTF activity.

For this reviewer, there were a few notable limitations. One was the uncertainty regarding whether expression of the respective TFs across cell types was taken into account. This would help inform if a TF would be able to open chromatin. Another limitation was the cell types used. While understandable that these cell types were used, because of their deep epigenetic phenotyping and public availability, they are mostly transformed and do not bear close similarity to lineages in a healthy organism. Next, the methods used to identify PTFs were not made available in an easy-to-use tool for other researchers who may seek to identify PTFs in their cell type(s) of interest. Lastly, some terms used were not define explicitly (e.g., meaning of dyads) and the language in the manuscript was often difficult to follow and contained improper English grammar.

Reviewer #3 (Public Review):

Peng et al. designed a computational framework for identifying pioneer factors using epigenomic data from five cell types. The identification of pioneer factors is important for our understanding of the epigenetic and transcriptional regulation of cells. A computational approach toward this goal can significantly reduce the burden of labor-intensive experimental validation.

The authors have addressed my previous comments.

The main issue identified in this re-review is based on the authors' additional experiments to investigate the reproducibility of the pioneer factors identified in the previously analysis that anchored on H1 ESCs.

The additional analysis that uses the other four cell types (HepG2, HeLa-S3, MCF-7, and K562) as anchors reveals the low reproducibility/concordance and high dependence on the selection of anchor cell type in the computational framework. In particular, now several stem cell related TFs (e.g. ESRRB, POU5F1) are ranked markedly higher when H1 ESC is not used as the anchor cell type as shown in Supplementary Figure 5.

Of note, the authors have now removed the shape labels that denote Yamanaka factors in Figure 2c (revised manuscript) that was presented in the main Figure 2a in the initial submission. The NFYs and ESRRB labels in Supplementary 4a are also removed and the boxplot comparing NFYs and ESRRB with other TF are also removed in this figure. Removing these results effectively hides the issues of the computational framework we identified in this revision. Please justify why this was done.

In summary, these new results reveal significant limitations of the proposed computational framework for identifying pioneer factors. The current identifications appear to be highly dependent on the choice of cell types.

Response: We thank all reviewers for their thoughtful and constructive comments and suggestions, which helped us to strengthen our paper. Following the suggestions, we have further addressed the reviewer’s comments and the detailed responses are itemized below.

Reviewer #1 (Recommendations For The Authors):

The following few minor mistakes/discrepancies/omissions should be addressed:

1. In Figure 3, the Nucleosome Occupancy curves and legend are orange and the Binding Motif Profiles are blue; however, the y-axis label for Nucleosome occupancy profile is blue, and the y-axis label of Binding motif profile is orange. The colors seem to be switched, or I'm missing something.

Response: We thank the reviewer for pointing it out. We have changed the colors to make it consistent.

1. The text at the bottom of p. 11 of the main manuscript describing Supplementary Fig. 5 states: "If we repeat our anaysis by redefining differentially open regions as those closed in differentiated cell lines and open in H1 embryonic cell line, then ESSRB and Yamanaka pioneer transcription factor POU5F1 (OCT4) showed significantly higher enrichment scores (Supplementary Figure 5)." However, Supplementary Fig. 5 legend states: "Enrichment analysis of different TFs using the differentially open from one cell line (shown in the title) and conserved open regions from other four cell lines.". These two descriptions of the differential chromatin criteria used in the analysis don't appear to match. The description in the text is the one that makes much more sense to me. The legend should be written a little more clearly and reflect the statement in the main text. One can see from the cut and paste the "analysis" is also misspelled.

Response: We have rewritten the legend of Supplementary Figure 5 to make it clear and consistent. The misspelling has also been corrected.

1. It might be helpful to add that a random classifier would yield a constant precision recall (PR) curve (as a function of Recall) with the Precision = P/(P+N) or the fraction of positives for all plotted PR curves which in the case of Fig. 2a is 32/225 = 0.142, for example.

Response: We thank the reviewer for the suggestions. We have added the fraction of positives for Figure 2.

1. On p. 17 line 513, the authors refer to "Supplementary 7, 9 and 13". I'm assuming it's "Supplementary Tables 7, 9 and 13".

Response: It has been corrected.

1. On p. 18 line 539, "essays" should be "assays".

Response: It has been corrected.

Reviewer #2 (Recommendations For The Authors):

We are satisfied with the revisions in this version of the manuscript.

Reviewer #3 (Public Review):

The main issue identified in this re-review is based on the authors' additional experiments to investigate the reproducibility of the pioneer factors identified in the previously analysis that anchored on H1 ESCs.

The additional analysis that uses the other four cell types (HepG2, HeLa-S3, MCF-7, and K562) as anchors reveals the low reproducibility/concordance and high dependence on the selection of anchor cell type in the computational framework. In particular, now several stem cell related TFs (e.g. ESRRB, POU5F1) are ranked markedly higher when H1 ESC is not used as the anchor cell type as shown in Supplementary Figure 5.

Of note, the authors have now removed the shape labels that denote Yamanaka factors in Figure 2c (revised manuscript) that was presented in the main Figure 2a in the initial submission. The NFYs and ESRRB labels in Supplementary 4a are also removed and the boxplot comparing NFYs and ESRRB with other TF are also removed in this figure. Removing these results effectively hides the issues of the computational framework we identified in this revision. Please justify why this was done.

In summary, these new results reveal significant limitations of the proposed computational framework for identifying pioneer factors. The current identifications appear to be highly dependent on the choice of cell types.

Response: We would like to clarify that our enrichment score used for TF classification, defined by Equation 3, is expected to be cell-type specific. The value of the enrichment score is modulated by a number of factors beyond the property of a TF to act as a PTF, such as the abundance of a given TF in a given cell line, cell type-specific nucleosome binding maps and interactions with other TFs. Thus, it is expected that the enrichment scores calculated for the same TF in different cell lines should be quantitatively different. Following the initial suggestion of Reviewer 3, we have diversified our analysis by using different cell lines as anchors. This analysis showed that most PTFs that we identified could be confirmed based on different cell lines, when comparing the relative enrichment scores within each cell line. On the other hand, it is not expected that the values of enrichment scores of a given TF should be similar across different cell lines.

Regarding a specific comment about ESRRB and POU5F1, these TFs are known pioneer factors with roles in reprogramming of somatic cells into induced pluripotent stem cells and suppressing cell differentiation. They have the ability to open closed chromatin regions in the differentiated cell lines. Therefore, if we redefine the differentially open regions as those closed in differentiated cell lines and open in H1 embryonic cell line, these pioneer factors are expected to have high enrichment scores. Indeed, our new results validated the roles of these PTFs in cell reprogramming. As mentioned above, their enrichment scores in different cell lines are not expected to be the same.

We also would like to clarify that no results were removed during the update of the figures, and all modifications of the manuscript following the suggestions of the reviewers were only made to improve the figures and make them clearer and the message more straightforward.

Reviewer #1 (Public Review):

Peng et al develop a computational method to predict/rank transcription factors (TFs) according to their likelihood of being pioneer transcription factors--factors that are capable of binding nucleosomes--using ChIP-seq for 225 human transcription factors, MNase-seq and DNase-seq data from five cell lines. The authors developed relatively straightforward, easy to interpret computational methods that leverage the potential for MNase-seq to enable relatively precise identification of the nucleosome dyad. Using an established smoothing approach and local peak identification methods to estimate positions together with identification of ChIP-seq peaks and motifs within those peaks which they referred to as "ChIP-seq motifs", they were able to quantify "motif profiles" and their density in nucleosome regions (NRs) and nucleosome depleted regions (NDRs) relative to their estimated nucleosome dyad positions. Using these profiles, they arrived at an odd-ratio based motif enrichment score along with a Fisher's exact test to assess the odds and significance that a given transcription factor's ChIP-seq motifs are enriched in NRs compared to NDRs, hence, its potential to be a pioneer transcription factor. They showed that known pioneer transcription factors had among the highest enrichment scores, and they could identify a number of relatively novel pioneer TFs with high enrichment scores and relatively high expression in their corresponding cell line. They used multiple validation approaches including (1) calculating the ROC-AUC and Matthews correlation coefficient (MCC) and generating ROC and precision-recall curves associated with their enrichment score based on 32 known pioneer TFs among their 225 TFs which they used as positives and the remaining TFs (among the 225) as negatives; (2) use of the literature to note that known pioneer TFs that acted as key regulators of embryonic stem cell differentiation had a highest enrichment scores; (3) comparison of their enrichment scores to three classes of TFs defined by protein microarray and electromobility shift assays (1. strong binder to free and nucleosomal DNA, 2. weak binder to free and nucleosomal DNA, 3. strong binding to free but not nucleosomal DNA); and (4) correlation between their calculated TF motif nucleosome end/dyad binding ratio and relevant data from an NCAP-SELEX experiment. They also characterize the spatial distribution of TF motif binding relative to the dyad by (1) correlating TF motif density and nucleosome occupancy and (2) clustering TF motif binding profiles relative to their distance from the dyad and identifying 6 clusters.

The strengths of this paper are the use of MNase-seq data to define relatively precise dyad positions and ChIP-seq data together with motif analysis to arrive at relatively accurate TF binding profiles relative to dyad positions in NRs as well as in NDRs. This allowed them to use a relatively simple odds ratio based enrichment score which performs well in identifying known pioneer TFs. Moreover, their validation approaches either produced highly significant or reasonable, trending results.

The weaknesses of the paper are relatively minor, and the authors do a good job of describing the limitations of the data and approach.

Reviewer #2 (Public Review):

In this study, the authors utilize a compendium of public genomic data to identify transcription factors (TF) that can identify their DNA binding motifs in the presence of nuclosome-wrapped chromatin and convert the chromatin to open chromatin. This class of TFs are termed Pioneer TFs (PTFs). A major strength of the study is the concept, whose premise is that motifs bound by PTFs (assessed by ChIP-seq for the respective TFs) should be present in both "closed" nucleosome wrapped DNA regions (measured by MNase-seq) as well as open regions (measured by DNAseI-seq) because the PTFs are able to open the chromatin. Use of multiple ENCODE cell lines, including the H1 stem cell line, enabled the authors to assess if binding at motifs changes from closed to open. Typical, non-PTF TFs are expected to only bind motifs in open chromatin regions (measured by DNaseI-seq) and not in regions closed in any cell type. This study contributes to the field a validation of PTFs that are already known to have pioneering activity and presents an interesting approach to quantify PTF activity.

For this reviewer, there were a few notable limitations. One was the uncertainty regarding whether expression of the respective TFs across cell types was taken into account. This would help inform if a TF would be able to open chromatin. Another limitation was the cell types used. While understandable that these cell types were used, because of their deep epigenetic phenotyping and public availability, they are mostly transformed and do not bear close similarity to lineages in a healthy organism. Next, the methods used to identify PTFs were not made available in an easy-to-use tool for other researchers who may seek to identify PTFs in their cell type(s) of interest. Lastly, some terms used were not defined explicitly (e.g., meaning of dyads) and the language in the manuscript was often difficult to follow and contained improper English grammar.

Reviewer #3 (Public Review):

Peng et al. designed a computational framework for identifying pioneer factors using epigenomic data from five cell types. The identification of pioneer factors is important for our understanding of the epigenetic and transcriptional regulation of cells. A computational approach toward this goal can significantly reduce the burden of labor-intensive experimental validation.

The authors have addressed my previous comments.

The main issue identified in this re-review is based on the authors' additional experiments to investigate the reproducibility of the pioneer factors identified in the previous analysis that anchored on H1 ESCs.

The additional analysis that uses the other four cell types (HepG2, HeLa-S3, MCF-7, and K562) as anchors reveals the low reproducibility/concordance and high dependence on the selection of anchor cell type in the computational framework. In particular, now several stem cell related TFs (e.g. ESRRB, POU5F1) are ranked markedly higher when H1 ESC is not used as the anchor cell type as shown in Supplementary Figure 5.

Of note, the authors have now removed the shape labels that denote Yamanaka factors in Figure 2c (revised manuscript) that was presented in the main Figure 2a in the initial submission. The NFYs and ESRRB labels in Supplementary 4a are also removed and the boxplot comparing NFYs and ESRRB with other TF are also removed in this figure. Removing these results effectively hides the issues of the computational framework we identified in this revision. Please justify why this was done.

In summary, these new results reveal significant limitations of the proposed computational framework for identifying pioneer factors. The current identifications appear to be highly dependent on the choice of cell types.

Author Response

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

Public Reviews:

Reviewer #1 (Public Review):

Weaknesses:

Reviewer comment: Here, the activity of SWIFT molecules was assessed in single cell types with or without BKlotho expression. Ultimately, the ability of the SWIFT molecules to activate Wnt signaling in a cell type-specific manner should be tested in the context of many different cellular identities that express BKlotho to different extents. It would be good to demonstrate that Wnt activation by SWIFT correlates with BKlotho expression level in multiple cell types - such data would strengthen the claim of cell-type specificity.

Response: We agree with the reviewer’s comment, it would be interesting to correlate the signaling level to the expression levels of βKlotho. The tools to carry out such an experiment are not currently available, as this would require a culture system that allows efficient growth of different cell types, and the reagents to detect both the receptor protein levels of βKlotho (as well as FZD/LRP) and signaling levels. We did perform an additional experiment to further support this targeting approach using a 2-layered (transwell) cell culture system. In this culture system, one cell type is put into the top well and the other cell type is put into the bottom well. Molecules to be tested were added to the media which is shared and freely diffuse across the two cell types. In this 2-layer cell system, the results again demonstrate the ability of the SWIFT molecules to specifically induce signaling only in βKlotho expressing hepatoma Huh7 cells and not in non-targeting HEK293 cells. This new data is included as Fig. 3H in the revised manuscript.

Reviewer comment: The study does not address whether the targeted cells express FGFR1c/2c/3c and whether the FGF21 full-length moiety or the 39F7 IgG moiety of SWIFT molecules could unintentionally activate FGF signaling in these cells.

Response: We agree with the reviewer’s comment. The receptor βKlotho and its binders (FGF21 and 39F7) were used to test the BRAID/SWIFT concept, the effects on FGF signaling were not the focus of the current study. This comment has now been added to the revised manuscript in the discussion. Inclusion of αGFP controls in the study also suggests the observed reporter activity in the targeted scenario is unlikely caused indirectly by any unexpected FGF signaling.

Reviewer #2 (Public Review):

Weaknesses:

Reviewer comment: The study shows the SWIFT approach works in vitro using cell lines, primary human hepatocytes, and human intestinal organoids, but it lacks an in vivo animal model or clinical validation. The applicability of this approach to therapy is still unknown.

Response: The βKlotho binder, 39F7, is specific to the human receptor and does not cross react with mouse. Unfortunately, we are not able to test these SWIFTs in a mouse model.

Reviewer comment: The success of SWIFT depends on the presence and expression of the bridging receptor (βKlotho) on target cells. The approach may fail if the target receptor is not expressed or available.

Response: We agree with the reviewer, the SWIFT molecules should not induce signaling on cells where bridging receptor is not expressed, therefore, achieving target cell specificity. As pointed out by the reviewer, finding the right bridging receptor on the target cell is critical.

Reviewer #1 (Recommendations For The Authors):

Reviewer comment 1: One way to further validate the specificity of SWIFT molecules would be to apply them to a mix of different cell types and quantify BKlotho level and Wnt reporter activity at the single cell level, potentially through imaging, FACS, or transcriptomics.

Response: We agree with the reviewer’s comment, it would be interesting to correlate the signaling level to the expression levels of βKlotho. The tools to carry out such an experiment are not currently available, as this would require a culture system that allows efficient growth of different cell types, and the reagents to detect both the receptor protein levels of βKlotho (as well as FZD/LRP) and signaling levels. We did perform an additional experiment to further support this targeting approach using a 2-layered (transwell) cell culture system. In this culture system, one cell type is put into the top well and the other cell type is put into the bottom well. Molecules to be tested were added to the media which is shared and freely diffuse across the two cell types. In this 2-layer cell system, the results again demonstrate the ability of the SWIFT molecules to specifically induce signaling only in βKlotho expressing hepatoma Huh7 cells and not in non-targeting HEK293 cells. This new data is included as Fig. 3H in the revised manuscript.

Reviewer comment 2: The experiments presented demonstrate activation of one signaling pathway in cells specifically expressing a target receptor rather than demonstrating "the feasibility of combining different signaling pathways" as stated in the abstract.

Response: We thank the reviewer for pointing this out and have adjusted the sentence accordingly.

Reviewer comment 3: What are the biological consequences of activating Wnt signaling in cells expressing BKlotho and why is that of interest? Could these biological outcomes be used as an additional, perhaps more consequential, readout for SWIFT activity?

Response: βKlotho is expressed on several different cell types that include hepatocytes, WAT, BAT, and certain regions in CNS. Our studies here focused on the WNT signaling pathway, and βKlotho/FGF21/39F7 receptor ligand system was used to illustrate the BRAID/SWIFT cell targeting concept. Whether these molecules may additional modulate endocrine FGF signaling and metabolic homeostasis, and whether there is any interaction between βKlotho and Wnt signaling pathways could be the subject of future studies. This is now added to the revised manuscript.

Reviewer comment 4: The manuscript would benefit from a careful review to improve wording and address grammatical errors.

Response: We thank the reviewer for this suggestion, and we have now had another round of language editing by a professional service.

Reviewer #2 (Recommendations For The Authors):

Reviewer comment 1. The expression of KLB in Fig 3G and 4B seems way too low and may not represent the amount on the cell surface. Did the authors validate the expression on the cell surface?

Response: In both figures we have displayed the expression level normalized to housekeeping gene ACTB. Housekeeping genes such as ACTB can be among the most abundant transcripts in a cell. The observation that KLB mRNA detection is below ACTB mRNA levels is expected and we would argue not too low. The average real-time PCR cycle threshold (Ct) for KLB in Huh7 and primary hepatocytes was 18 and 24 respectively. To avoid any confusion, we have now displayed the expression data normalized to HEK293 and intestinal organoids as a fold difference in a new Figure 3G and 4B.

Comment 2. Fig 3G needs statistical significance.

Response: We thank the reviewer for highlighting this, we have now included the statistical analysis in an updated Figure 3G.

eLife assessment

This important study presents a new way to selectively activate a cell signaling pathway in a specific cell type by designer ligands that link signaling co-receptors to a marker specific to the target cells. Convincing experimental results demonstrate that the agonist molecules activate Wnt signaling in target cells expressing the marker as intended. More broadly, this concept could be used to induce Wnt signaling or another pathway initiated by co-receptor association in a cell type-specific manner. In vitro results in this study could be further strengthened by assessing the biological consequences of Wnt activation in target cells.

Reviewer #1 (Public Review):

Summary: The goal of this study was to develop and validate novel molecules to selectively activate a cell signaling pathway, the Wnt pathway in this case, in target cells expressing a specific receptor. This was achieved through a two-component system that the authors call BRAID, where each component simultaneously binds the target cell-specific marker BKlotho and a Wnt co-receptor. These components, called SWIFT molecules, bring together the Wnt co-receptors LRP and FZD, activating the pathway specifically in cells that express BKlotho.

Results presented in the study demonstrate the desired activity of SWIFT molecules; the binding assays support simultaneous association of SWIFT with BKlotho and a Wnt co-receptor, and the Wnt reporter and qPCR assays support pathway activation in cell lines and primary cells in a BKlotho-dependent manner. In the future, the BRAID approach could be applied to activate Wnt signaling or another pathway initiated by a co-receptor complex in a cell type-specific manner, and/or in a FZD subtype-specific manner to activate distinct branches of Wnt signaling.

Strengths:

• This study successfully demonstrates a novel way to activate Wnt signaling in target cells expressing a specific marker. Given the role of the Wnt signaling pathway in key processes such as cell proliferation and tissue renewal and the value of modulating cell signaling in a cell type-specific manner, the cell targeting system developed here holds great therapeutic and research potential. It will be curious to see whether the BRAID design can be applied to other cell surface markers for Wnt activation, or for activation of other signaling pathways that require co-receptor association.

• Octet assay results show simultaneous binding of SWIFT molecules to both the Wnt co-receptor FZD/LRP and BKlotho, while negative control molecules without the FZD/LRP or BKlotho-binding module show neither receptor binding nor Wnt pathway activation. These results indicate that SWIFT molecules function through the intended mechanism.

• Exposure of two cell types simultaneously exposed to the SWIFT molecules in 2-layer cell culture demonstrated the ability of the molecules to activate Wnt signaling in a cell type- and BKlotho expression-specific manner.

Weaknesses:

• The study does not address whether the targeted cells express FGFR1c/2c/3c and whether the FGF21 full length moiety or the 39F7 IgG moiety of SWIFT molecules could unintentionally activate FGF signaling in these cells.

Reviewer #2 (Public Review):

Summary:

The study introduces BRAID, a novel approach for targeting drugs to specific cell types, addressing the challenges of pleiotropic drug actions. Unlike existing methods, this one involves breaking a protein drug molecule into inactive parts that are then put back together using a bridging receptor on the target cell. The individual components of this assembly are not required to be together, thereby affording it a degree of flexibility. The authors applied this idea to the WNT/-catenin signaling pathway by splitting a WNT mimic into two parts with FZD and LRP binding domains and bridging receptors. This combined method, which is called SWIFT, showed that WNT signaling was turned on in target cells, showing that cell-specific targeting is. The technique shows promise for the development of therapeutics, as it provides a way to more precisely target signaling pathways.

The authors have effectively elucidated their strategy through visually appealing diagrams, providing clear and thorough visual aids that facilitate comprehension of the concept. In addition, the authors have provided convincing evidence that the C-terminal region of FGF21 is essential for the binding process. Their meticulous and thorough presentation of experimental results emphasizes the significance of this specific binding domain and validates their findings.

Strengths:

BRAID, a novel cell targeting method, divides an active drug molecule into inactive components formed by a bridging receptor. This novel approach to cell-specific drug action may reduce systemic toxicity.

The SWIFT approach successfully targets cells in the WNT/β-catenin signaling pathway. The approach activates WNT signaling only in target cells (hepatocytes), proving its specificity.

The study indicates that the BRAID approach can target various signaling systems beyond WNT/β-catenin, indicating its versatility. Therapeutic development may benefit from this adaptability.

Weaknesses:

The study shows the SWIFT approach works in vitro using cell lines, primary human hepatocytes, and human intestinal organoids, but it lacks in vivo animal model or clinical validation. I believe future studies will determine this aspect.

The success of SWIFT depends on the presence and expression of the bridging receptor (βKlotho) on target cells. The approach may fail if the target receptor is not expressed.

Author Response

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

Public Reviews:

Reviewer #1 (Public Review):

In the manuscript there is not much comparison between the crystal and cryoEM structures provided, and on inspection they appear to be very similar. The crystal structures also reveal parts of the CC domains in Las1, which is not present in the cryoEM structures. It is interesting the CC domains in Sc and Cj are quite different as illustrated in Figure 4B. They also seem to be somewhat disconnected from the rest of the complex (more so for Cj), even though that's not apparent in Figures 2-4. Despite this, it would be very useful to show the cryoEM densities when describing the catalytic site and C-terminal domain interactions, for example, as this can be very useful to increase confidence in the model and proposed mechanisms.

We thank the reviewer for this suggestion. We have added a figure (Figure 5- Figure supplement 3) to show cryo-EM and crystal densities of key amino acids, when describing the catalytic site and C-terminal domain interactions. In analyzing the interaction between Las1 and Grc3, we have also provided additional comparisons of the crystal structure and the cryo-EM structure (Figure 5, Figure 5-figure supplement 1, 2 and 3, Figure 6, Figure 6-figure supplement 1).

The description of the complex as a butterfly is engaging, and from a certain angle it can be made to look as such; this was also described previously in (Pillon et al., 2019, NSMB) for the same complex from a different organism (Ct). However, it is a bit misleading, because the complex is actually C2 symmetric. Under this symmetry, the 'body' would consist of two 'heads' one pointing up, one down facing towards the back, and one wing would have its back toward the viewer, the other the front. The structures presented here in Sc and Cj seem quite similar to the previous structure of the same complex in Ct, though the latter was only solved with cryoEM, and was also lacking the structure of the CC domain in Las1.

We thank the reviewer for pointing out this issue. We have re-wrote these sentences and changed the butterfly description of Las1-Grc3 complex in the revised manuscript.

For the model suggested in Figure 8, perhaps in the 'weak activity' state, the LCT in Las1 could still be connected to Grc3, via the LCT, rather than disconnected as shown. This could facilitate faster assembly of the 'high activity' state. The complex is described as 'compact and stable', but from the structure and this image, it appears more dynamic, which would serve its purpose and the illustrated model better. The two copies of HEPN appear to have more connective area, meaning they are indeed more likely to remain assembled in the 'weak activity' state. On the other hand, HEPN in one protein appears to have less binding surface with PNK in Grc3, and even less so with the CTD (both PNK and CTD being from the other associated protein), meaning these bindings could release easily to form the 'weak activity' state.

There is also the potential to speculate that the GCT is bound to HEPN near the catalytic area in the 'weak activity' state. The reduced activity when the GCT residues are replaced by Alanine could then be explained by the complex not being able to assemble as quickly upon binding of the substrate, as it could if the GCT remained bound, rather than by a conformational change that it induces upon binding. The conformational change is also likely to be influenced by the combined binding of PNK and CTD in the assembled state, which also contact HEPN, rather than by GCT alone.

We thank the reviewer for this suggestion. We have revised our model in the new Figure 8 of our revised manuscript. We apologize for the un-clarity description of the 'weak activity' state in our model. In fact, we believe that Las1 is in a "weakly activity" state before binding to Grc3 and is in a "highly activity" state when it forms a complex with Grc3. We strongly agree that the Las1-Grc3 complex is more dynamic than compact and stable, so it is easy to change its active state. We have changed our description and revised our model in the revised manuscript.

When comparing the structure of the HEPN domain in the lone Las1 protein to the structure of Las1-HEPN in the Las1-Grc3 complex, it is mentioned that 'large conformational changes are observed'. These could be described a bit better. The conformational change is ~3-4Å C-alpha RMSD across all ~150 residues in the domain (~90 residues forming a stable core that only changes by ~1Å). There is also a shift in the associated HEPN domain in Las1B domain compared to the bound HEPN in the Las1-Grc3 complex, as shown in Figure 7D: ~1Å shift and ~12degrees rotation. This does point to the conformation of HEPN changing upon complex formation, as does the relative positions of the HEPN domains in Las1A and Las1B. The conformational change and relative shift could indeed by key for the catalysis of the substrate as mentioned.

We thank the reviewer for this great suggestion. We have replaced the sentence describing the conformational changes in our revised manuscript.

Overall, the structures presented should be very useful in further study of this system, even though the exact dynamics and how the substrate is bound are aspects that are perhaps not fully clear yet. The addition of the structures of the CC domain in two different organisms and the Las1 HEPN domain (not in complex with Grc3) as new structural information should allow for increasing our understanding of the overall complex and its mechanism.

We thank this reviewer for these encouraging comments, which helped us with greatly improving our manuscript.

Reviewer #2 (Public Review):

In this manuscript, Chen et al. determined the structural basis for pre-RNA processing by Las1-Grc3 endoribonuclease and polynucleotide kinase complexes from S. cerevisiae (Sc) and C. jadinii (Cj). Using a robust set of biochemical assays, the authors identify that the sc- and CjLas1-Grc3 complexes can cleave the ITS2 sequence in two specific locations, including a novel C2' location. The authors then determined X-ray crystallography and cryo-EM structures of the ScLas1-Grc3 and CjLas1-Grc3 complexes, providing structural insight that is complimentary to previously reported Las1-Grc3 structures from C. thermophilum (Pillon et al., 2019, NSMB). The authors further explore the importance of multiple Las1 and Grc3 domains and interaction interfaces for RNA binding, RNA cleavage activity, and Las1-Grc3 complex formation. Finally, evidence is presented that suggests Las1 undergoes a conformational change upon Grc3 binding that stabilizes the Las1 HEPN active site, providing a possible rationale for the stimulation of Las1 cleavage by Grc3.

Several of the conclusions in this manuscript are supported by the data provided, particularly the identification and validation of the second cleavage site in the ITS2. However, several aspects of the structural analysis and complimentary biochemical assays would need to be addressed to fully support the conclusions drawn by the authors.

We thank the reviewer for the positive comments.

• There is a lack of clarity regarding the number of replicates performed for the biochemical experiments throughout the manuscript. This information is critical for establishing the rigor of these biochemical experiments.

We apologize for not providing the detailed information on the number of replicates of biochemical experiments. All the biochemical experiments were repeated three times. We have provided this information in the figure legends.

• The authors conclude that Rat1-Rai1 can degrade the phosphorylated P1 and P2 products of ITS2 (lines 160-162, Figure 1H). However, the data in Fig. 1H shows complete degradation of 5'Phos-P2 and 5'Phos-P4 of ITS2, while the P1 and 5'Phos-P3 fragments remain in-tact. Additional clarification for this discrepancy should be provided.

We thank the reviewer for pointing out this issue. “phosphorylated P1 and P2 products” should be “phosphorylated P2 and P4 products”. We have corrected this clerical error. In addition, we have also provided an explanation for why phosphorylated P3 product show only partial degradation. We suspect that P3 product may be too short to completely degrade.

• The authors determined X-ray crystal structures of the ScLas1-Grc3 (PDB:7Y18) and CjLas1-Grc3 (PDB:7Y17) complexes, which represents the bulk of the manuscript. However, there are major concerns with the structural models for ScLas1-Grc3 (PDB:7Y18) and CjLas1-Grc3 (PDB:7Y17). These structures have extremely high clashscores (>100) as well as a significant number of RSRZ outliers, sidechain rotamer outliers, bond angle outliers, and bond length outliers. Moreover, both structures have extensive regions that have been modeled without corresponding electron density, and other regions where the model clearly does not fit the experimental density. These concerns make it difficult to determine whether the structural data fully support several of the conclusions in the manuscript. A more careful and thorough reevaluation of the models is important for providing confidence in these structural conclusions.

We thank the reviewer for pointing out this issue. We have used the cryo-EM datasets to further validate our conclusions of the manuscript. We analyzed the active site of Las1-Grc3 complex and the interactions between Las1 and Grc3 using the cyro-EM structures and presented new figures (Figure 5- Figure supplement 1, Figure 5- Figure supplement 2, Figure 5- Figure supplement 3, Figure 6- Figure supplement 1) in our revised manuscript. Both the refinement and validation statistical parameters of the cryo-EM datasets are within a reasonable range (Table 2), which will provide confidence for our structure conclusions. The X-ray crystal structures of ScLas1-Grc3 (PDB:7Y18) and CjLas1-Grc3 (PDB:7Y17) complexes has high calshscores and many outliers, which is mainly due to the great flexibility of Las1-Grc3 complex, especially the CC domain of Las1. We have improved our crystal structure models with better refinement and validation of statistical parameters. The clashscores of ScLas1-Grc3 complex and CjLas1-Grc3 complex are 25 and 45, respectively. There are no rotamer outliers and C-beta outliers to report for both ScLas1-Grc3 complex and CjLas1-Grc3 complex.

• The presentation of the cryo-EM datasets is underdeveloped in the results section drawing and the contribution of these structures towards supporting the main conclusions of the manuscript are unclear. An in-depth comparison of the structures generated from X-ray crystallography and cryo-EM would have greatly strengthened the structural conclusions made for the ScLas1-Grc3 and CjLas1-Grc3 complexes.

We thank the reviewer for this suggestion. We have performed structural comparisons between X-ray crystal structure and cyro-EM structure in analyzing the active site of Las1-Grc3 complex and the interactions between Las1 and Grc3 (Figure 5- Figure supplement 1, Figure 5- Figure supplement 2, Figure 6- Figure supplement 1). We have also added a figure (Figure 5- Figure supplement 3) to show cryo-EM and crystal densities of the Las1 active site as well as the key amino acids for Las1 and Grc3 interactions. These comparisons and densities have greatly strengthened our structural conclusions.

• The authors conclude that truncation of the CC-domain contributes to Las1 IRS2 binding and cleavage (lines 220-222, Fig. 4C). However, these assays show that internal deletion of the CC-domain alone has minimal effect on cleavage (Fig 4C, sample 3). The loss in ITS2 cleavage activity is only seen when truncating the LCT and LCT+CC-domain (Fig 4C, sample 2 and 4, respectively). Consistently, the authors later show that Las1 is unable to interact with Grc3 when the LCT domain is deleted (Fig. 6 and Fig. 6-figure supplement 2). These data indicate the LCT plays a critical role in Las1-Grc3 complex formation and subsequent Las1 cleavage activity. However, it is unclear how this data supports the stated conclusion that the CC-domain is important for LasI cleavage.

Our EMSA data shows that the CC domain contributes to the binding of ITS2 RNA (Figure 4D), suggesting that the CC domain may play a role of ITS2 RNA stabilization in the Las1 cutting reaction. The in vitro RNA cleavage assays (Figure 4C) indicate that the LCT is important for Las1 cleavage because it plays a critical role in the formation of the Las1-Grc3 complex. Compared with LCT, the CC domain, although not particularly important for Las1 cutting ITS2, still has some influence (Fig 4C, sample 1 and 3, sample 2 and 4,). Therefore, we conclude that the CC domain may mainly play a role in the stabilization of ITS2 RNA, thereby enhancing ITS2 RNA cleavage.

• The authors conclude that the HEPN domains undergo a conformational change upon Grc3 binding, which is important for stabilization of the Las1 active site and Grc3-mediated activation of Las1. This conclusion is based on structural comparison of the HEPN domains from the CjLas1-Grc3 complex (PDB:7Y17) and the structure of the isolated HEPN domain dimer (PDB:7Y16). However, it is also possible that the conformational changes observed in the HEPN domain are due to truncation of the Las1 CC and CGT domains. A rationale for excluding this possibility would have strengthened this section of the manuscript.

We thank the reviewer for pointing out this issue. We agree that the complete Las1 structure information is helpful in illuminating the conformational activation of the Las1 by Grc3. We screened about 1200 crystallization conditions with full-length Las1 proteins, but ultimately did not obtain any crystals, probably due to flexibility. The CC domain exhibits a certain degree of flexibility, which has not been observed in the structure obtained from electron microscopy. The LCT is involved in binding to the CTD domain of Grc3. The coordination of the active center of HEPN domains by LCT and CC domains is unlikely due to the limited nuclease activity observed in full-length Las1. The conformational changes of the active center are essential for HEPN nuclease activation. Our structure shows that the GCTs of Grc3 interact with the active residues of Las1 HEPN domains, which probably induce conformational changes in the active center of the HEPN domain to activate Las1. Of course, we cannot exclude the possibility that truncation of the Las1 CC and LCT domains will result in little conformational change in the HEPN domains. We have explained this possibility in our revised manuscript.

Reviewer #1 (Recommendations For The Authors):

1) It would be very useful to show the cryoEM densities when describing the catalytic site and C-terminal domain interactions.

The new Figure 5-figure supplement 2 have showed the Cyro-EM densities of the catalytic site of ScLas1 and the C-terminal domain of ScGrc3.

2) "ScLas1 cleaves the 33-nt ITS2 at C2 site to theoretically generate a 10-nt 5′-terminal product and a 23-nt 3′-terminal product (Figure 1A). Our merger data shows that the final 5′-terminal and 3′-terminal product bands are at nearly the same horizontal position on the gel (Figure 1B), indicating that they are similar in size." These two sentences seem to contradict, i.e. 10-nt and 23-nt are similar in size even though they are different lengths?

We apologize for the contradiction in these two sentences mentioned above. We have re-wrote these two sentences in the revised manuscript.

3) We observed four cleavage bands of approximately 23-nt (P2), 14-nt (P3), 10-nt (P1), and 9-nt (P4) in length (Figure 1C). "

Figure 1C. The bands show 23 nt, 22 nt, 21nt, 14 nt, 13nt, and 11nt, so this text does not seem to describe the figure.

We have re-wrote this sentence in the revised manuscript.

4) "We obtained similar cleavage results with a longer 81-nt ITS2 RNA substrate 6 (Figure 1D, E). " Figure 1D,E. The lengths in Figure 1E do not correspond to all bands in Figure 1E, e.g. the 13 nt band, though the others do, e.g. 14 nt, 30nt, 37nt, etc.

In order to better evaluate the size of the cut product, we used an RNA marker as a comparison. The RNA marker will have more bands than the cleavage products. To further confirm the cleavage site of C2′, we also mapped the cleavage sites of the 81-nt ITS2 using reverse transcription coupling sequencing methods (Figure 1F).

5) In Figure 3, domains are colored different but it's hard to know which are different proteins.

We have added a diagram in Figure 3 to show the Las1-Grc3 complex structure, and it is now clear how Las1 and Grc3 are assembled into a tetramer.

6) Line 267. "we screened a lot of crystallization conditions with full-length Las1 proteins" How many? Rough numbers ok, but 'a lot' is not very informative

We have provided the approximate numbers of crystallization conditions in our revised manuscript.

Reviewer #2 (Recommendations For The Authors):

1) The authors missed an excellent opportunity to compare and contrast the ScLas1-Grc3 and CjLas1-Grc3 complex structures presented here with that of the previously determined CtLas1-Grc3 structure (Pillon et al., 2019, NSMB). For example, His130 in the ScLas1-Grc3 complex active site adopts a similar conformation to His142 in the TcLas1-Grc3 complex active site (Pillon et al., 2019, NSMB). Interestingly, the analogous His134 active site residue in the CjLas1-Grc3 adopts an alternative (maybe inactive) conformation. This observation could provide a structural rationale for the activation of scLas1 and TcLas1 by Grc3, while also providing a rationale for the fairly weak activation of CjGrc3 by CjGrc3.

We thank the reviewer for this suggestion. We have performed structural comparisons between ScLas1-Grc3, CjLas1-Grc3 and CtLas1-Grc3 complexes, especially the Las1 nuclease active center. We added two figures (Figure7-figure supplement 3A and 3B) in the revised manuscript to contrast and highlight the conformational differences of active amino acids in active centers between ScLas1-Grc3, CtLas1-Grc3 and CjLas1-Grc3. These structural comparisons provide stronger evidence that further reinforces the conclusions of our manuscript.

2) Can the authors speculate as to whether the structural data can provide any insight into how the Las1-Grc3 may cleave both C2 and C2' positions in the ITS2 RNA? This commentary would further strengthen the discussion section of the manuscript.

We thank the reviewer for this suggestion. We have provided a speculation in the discussion section of the revised manuscript.

We think that the structural data may provide some insight into how Las1-Grc3 complex cleaves ITS2 RNA at both C2 and C2' positions. The Las1-Grc3 tetramer complex has one nuclease active center and two kinase active centers. The nuclease active center consists of two Las1 molecules in a symmetric manner, while the kinase active center consists of only one Grc3 molecule. The ITS2 RNA is predicted to form a stem-loop structure. The symmetrical nuclease active center recognizes the stem region of ITS2 RNA and makes it easy to perform C2 and C2' cleavages on both sides of the stem. C2 and C2' cleavage products are further phosphorylated by two Grc3 kinase active centers, respectively.

3) The method used for the plasmid generation, expression, and purification of the Las1 truncations and the Las1 and Grc3 point mutants should be provided in the methods section.

The method used for the plasmid generation, expression, and purification of the Las1 truncations and the Las1 and Grc3 point mutants have be provided in the methods section.

4) The exact amino acid cutoffs for the truncated forms of Las1 used for the biochemical assays in Fig. 4 should be provided.

We have provided the exact amino acid cutoffs for the truncated forms of Las1 in the figure legend of Figure 4C.

5) The models associated with the cryo-EM datasets should be deposited in the PDB.

The models associated with the Cryo-EM datasets have be deposited in the PDB with the following accession codes: 8J5Y (ScLas1-Grc3 complex), and 8J60 (CjLas1-Grc3 complex).

6) Lines 232-234: Arg129 should be changed to His134.

We have corrected it.

7) Figure 5B: the bottom half of the HEPN active site has been labeled incorrectly. The labels should be Arg129, His130, and His134 (from left to right).

We have corrected it.

8) Line 252: "multitudinous" should be changed to "multiple."

We have corrected it.

eLife assessment

This study represents a valuable mechanistic contribution towards understanding how ribosomal RNA is processed during ribosome biogenesis. The biochemical evidence supporting the major conclusions is convincing. This work will be of interest to cell biologists and biochemists working on ribosome biogenesis.

Reviewer #1 (Public Review):

The findings in this paper can be split into three parts.

1) Processing of ITS2

Firstly, the authors identify two sites on ITS2 which are cleaved by the ScLas1-Grc3 complex, as part of 25S ribosomal RNA maturation.

For a smaller segment of ITS2 (33nt), the two sites separate out 3 parts with sizes of 10nt, 14nt, and 9nt (Figure 1C). However, bands in mass spec. occur at 23nt (14nt+9nt) and 14nt, but not 9nt alone. Additional bands can be seen at 22nt and 21nt. Hence the evidence for these two specific sites seems somewhat uncertain. It is not clear if there is an experimental limitation in terms of accuracy, or that the cleavage is perhaps somewhat approximate at the two sites. The authors may try to clarify these results a bit further.

For a larger ITS2 (81nt), similar support is found for the two cleavage sites, but now the possible fragments are 14nt, 30nt, 37nt, and 44nt (14nt+30nt) (Figure 1E). The observed bands match these fragments at 44nt, 37nt, 30nt, but again there are additional bands at 36nt, 28nt, and 13nt, which are not fully explained. It may be useful for explain or discuss these discrepancies.

Another useful result of these experiments is to confirm that Las1 alone has only weak activity against ITS2, but very strong activity when it is part of the Las1-Grc3 complex.

2) Structure of Las1, and Las1-Grc3 complexes

A second important contribution of this work are X-ray and cryoEM structures of Las1-Grc3 from Sc and Cj. It is interesting that even though the complexes are very similar, CjGrc3 shows weak activation of CjLas1, whereas ScGrc3 more strongly activates ScLas1. The X-ray and cryoEM structures are very similar. However, the X-ray structures also show an additional (CC) domain from Las1 not resolved in the cryoEM map. This difference is significant, because it suggests the CC domain may remain more flexible in solution, but stabilizes in the crystal. Also interestingly, the CC domains have different structures and are in different positions in ScLas1-Grc3 vs CjLas1-Grc3, again hinting that they are more dynamic. Further experiments described by the authors confirm the CC domain is indeed important in RNA binding and activity. Whether they are only implicated in binding RNA or both binding and cleavage is somewhat unclear.

The structure of Las1-Grc3 is described as resembling a butterfly, with Las1 being the body and Grc3 the wings. While this is a useful description, it may be a bit misleading. The complex has C2 symmetry, with one Las1-Grc3 unit related to the other by about ~180 rotation around a vertical axis parallel to the body of the butterfly as proposed. To use the butterfly analogy, one half of the body and one wing faces the opposite way as the other, not a mirror symmetry as a real butterfly would have.

Both Cj and Sc structures show the C-terminal of Grc3 binding to the active pocket of Las1, explaining its effect on activity. Mutation experiments also further show the importance of these residues on activity. Reciprocally, a region in Las1, LCT, inserts into Grc3, forming a stable complex. Again mutating these residues affects activity, strengthening their importance and the evidence for how the stable complex forms.

Finally, an X-ray structure of dimeric Las1 in Cj, without Grc3 is presented. Truncating CC and LCT appeared to be necessary to allow the dimer to crystallize. Superposition with Las1 dimer in Las1-Grc3 shows a conformational difference, and different distances between residues in the active pocket, explaining the change in activity with and without Grc3. Interestingly, the Las1 domains themselves do not change too much, i.e. both domains can be matched with less than 1Å Ca-RMSD, so the difference may be more of a repositioning of the two domains for the active conformation.

One notable strength of this study is the use of both X-ray and cryoEM to obtain structures of the Las1-Grc3 and dimeric Las1 complexes. Typically structures of cryoEM at ~3Å are sufficient for reliable modeling; for example, the backbone and side chains of residues in the active site are well resolved. However, in this case, a cryoEM model of the Las1 dimer was not obtained, so it was important to show first that the Las1-Las1 conformation in the Las1-Grc3 complex is the same in both X-ray and cryoEM models. Otherwise, there may remain doubt whether the X-ray model of Las1 dimer could be compared to the cryoEM map of Las1-Grc3, as crystallization conditions could potentially influence conformation and arrangement. It would be interesting to know whether a cryoEM structure of the Las1 dimer alone was attempted - perhaps it was too small to be reliably seen in micrographs. Having had such a model could avoid the need of X-ray structures, although of course more experimental data are always useful.

3) Mechanism of ITS2 cleavage

The proposed mechanism shown in Figure 8 seems to be well supported by the obtained structures and biochemical experiments. A question that remains is why it is proposed that both C2 and C2' cleavage can be performed upon a single binding of the ITS2 RNA, i.e. seeming to suggest there are two binding sites. This would seem to directly generate 3 fragments, without any other intermediate products. Mass spec. seemed to show the intermediate products, perhaps indicating two binding events for each cleavage process. Perhaps the authors could discuss this more. Also perhaps can be good to discuss whether it would be possible to obtain a structure with the bound RNA, further giving structural information of how the exact cleavage process is performed.

Reviewer #2 (Public Review):

In this manuscript, Chen et al. determined the structural basis for pre-RNA processing by Las1-Grc3 endoribonuclease and polynucleotide kinase complexes from S. cerevisiae (Sc) and C. jadinii (Cj). Using a robust set of biochemical assays, the authors identify that the sc- and CjLas1-Grc3 complexes can cleave the ITS2 sequence in two specific locations, including a novel C2' location. The authors then determined X-ray crystallography and cryo-EM structures of the ScLas1-Grc3 and CjLas1-Grc3 complexes, providing structural insight that is complimentary to previously reported Las1-Grc3 structures from C. thermophilum (Pillon et al., 2019, NSMB). The authors further explore the importance of multiple Las1 and Grc3 domains and interaction interfaces for RNA binding, RNA cleavage activity, and Las1-Grc3 complex formation. Finally, evidence is presented that indicates Las1 undergoes a conformational change upon Grc3 binding that stabilizes the Las1 HEPN active site, providing a possible rationale for the stimulation of Las1 cleavage by Grc3.

In the revised manuscript, the authors have made significant efforts towards addressing initial reviewer comments. This includes further clarification for key biochemical experiments, significant improvement in structural model quality, and additional structural analysis that further strengthens major conclusions in the manuscript. Overall, the authors conclusions are now well supported by the biochemical and structural data provided.

Author Response

We are grateful to the reviewers for their thorough and thoughtful critiques, including their agreement on the significant value of this dataset. We intend to respond to their comments in full with a revision in the near future. However, we would like to make an initial comment at this stage. A key concern raised by the reviewers was that the analyses described do not adequately support the claim that "movie-watching data can identify retinotopic regions" (quoted from R2, similar sentiment expressed by R1). To be clear, we agree with this assessment. Our primary aim was not to identify visual areas with movie-watching data. Rather, our focus was on how movies can reveal fine-grained organization in infant visual cortex, which would support their potential utility for understanding the development of dynamic visual processing. To demonstrate this potential, we tested and found that maps of visual activity generated from movies are significantly similar to those generated by a retinotopy task. Nevertheless, we did not intend to argue that movie-based maps are sufficiently accurate to replace task-based retinotopic maps when defining visual areas, nor did we test this possibility. We accept that this point was unclear in the original manuscript and will make edits to avoid this miscommunication. We also plan to incorporate the reviewers’ many other helpful recommendations, including addressing concerns about the clarity of the presentation and double dipping, as well as adding several new analyses we hope will provide greater confidence in the findings and interpretation.

eLife assessment

This study presents valuable findings on the potential of short-movie viewing fMRI protocol to explore the functional and topographical organization of the visual system in awake infants and toddlers. Although the data are compelling given the difficulty of studying this population, the evidence presented is incomplete and would be strengthened by additional analyses to support the authors' claims. This study will be of interest to cognitive neuroscientists and developmental psychologists, especially those interested in using fMRI to investigate brain organisation in pediatric and clinical populations with limited fMRI tolerance.

Reviewer #1 (Public Review):

Summary:<br /> Ellis et al. investigated the functional and topographical organization of the visual cortex in infants and toddlers, as evidenced by movie-viewing data. They build directly on prior research that revealed topographic maps in infants who completed a retinotopy task, claiming that even a limited amount of rich, naturalistic movie-viewing data is sufficient to reveal this organization, within and across participants. Generating this evidence required methodological innovations to acquire high-quality fMRI data from awake infants (which have been described by this group, and elsewhere) and analytical creativity. The authors provide evidence for structured functional responses in infant visual cortex at multiple levels of analyses; homotopic brain regions (defined based on a retinotopy task) responded more similarly to one another than to other brain regions in visual cortex during movie-viewing; ICA applied to movie-viewing data revealed components that were identifiable as spatial frequency, and to a lesser degree, meridian maps, and shared response modeling analyses suggested that visual cortex responses were similar across infants/toddlers, as well as across infants/toddlers and adults. These results are suggestive of fairly mature functional response profiles in the visual cortex in infants/toddlers and highlight the potential of movie-viewing data for studying finer-grained aspects of functional brain responses, but further evidence is necessary to support their claims and the study motivation needs refining, in light of prior research.

Strengths:<br /> - This study links the authors' prior evidence for retinotopic organization of visual cortex in human infants (Ellis et al., 2021) and research by others using movie-viewing fMRI experiments with adults to reveal retinotopic organization (Knapen, 2021).

- Awake infant fMRI data are rare, time-consuming, and expensive to collect; they are therefore of high value to the community. The raw and preprocessed fMRI and anatomical data analyzed will be made publicly available.

Weaknesses:<br /> - The Methods are at times difficult to understand and in some cases seem inappropriate for the conclusions drawn. For example, I believe that the movie-defined ICA components were validated using independent data from the retinotopy task, but this was a point of confusion among reviewers. In either case: more analyses should be done to support the conclusion that the components identified from the movie reproduce retinotopic maps (for example, by comparing the performance of movie-viewing maps to available alternatives (anatomical ROIs, group-defined ROIs). Also, the ROIs used for the homotopy analyses were defined based on the retinotopic task rather than based on movie-viewing data alone - leaving it unclear whether movie-viewing data alone can be used to recover functionally distinct regions within the visual cortex.

- The authors previously reported on retinotopic organization of the visual cortex in human infants (Ellis et al., 2021) and suggest that the feasibility of using movie-viewing experiments to recover these topographic maps is still in question. They point out that movies may not fully sample the stimulus parameters necessary for revealing topographic maps/areas in the visual cortex, or the time-resolution constraints of fMRI might limit the use of movie stimuli, or the rich, uncontrolled nature of movies might make them inferior to stimuli that are designed for retinotopic mapping, or might lead to variable attention between participants that makes measuring the structure of visual responses across individuals challenging. This motivation doesn't sufficiently highlight the importance or value of testing this question in infants. Further, it's unclear if/how this motivation takes into account prior research using movie-viewing fMRI experiments to reveal retinotopic organization in adults (e.g., Knapen, 2021). Given the evidence for retinotopic organization in infants and evidence for the use of movie-viewing experiments in adults, an alternative framing of the novel contribution of this study is that it tests whether retinotopic organization is measurable using a limited amount of movie-viewing data (i.e., a methodological stress test). The study motivation and discussion could be strengthened by more attention to relevant work with adults and/or more explanation of the importance of testing this question in infants (is the reason to test this question in infants purely methodological - i.e., as a way to negate the need for retinotopic tasks in subsequent research, given the time constraints of scanning human infants?).

Reviewer #2 (Public Review):

Summary:<br /> This manuscript shows evidence from a dataset with awake movie-watching in infants, that the infant brain contains areas with distinct functions, consistent with previous studies using resting state and awake task-based infant fMRI. However, substantial new analyses would be required to support the novel claim that movie-watching data in infants can be used to identify retinotopic areas or to capture within-area functional organization.

Strengths:<br /> The authors have collected a unique dataset: the same individual infants both watched naturalistic animations and a specific retinotopy task. These data position the authors to test their novel claim, that movie-watching data in infants can be used to identify retinotopic areas.

Weaknesses:<br /> To claim that movie-watching data can identify retinotopic regions, the authors should provide evidence for two claims:

- Retinotopic areas defined based only on movie-watching data, predict retinotopic responses in independent retinotopy-task-driven data.

- Defining retinotopic areas based on the infant's own movie-watching response is more accurate than alternative approaches that don't require any movie-watching data, like anatomical parcellations or shared response activation from independent groups of participants.

Both of these analyses are possible, using the (valuable!) data that these authors have collected, but these are not the analyses that the authors have done so far. Instead, the authors report the inverse of (1): regions identified by the retinotopy task can be used to predict responses in the movies. The authors report one part of (2), shared responses from other participants can be used to predict individual infants' responses in the movies, but they do not test whether movie data from the same individual infant can be used to make better predictions of the retinotopy task data, than the shared response maps.

So to be clear, to support the claims of this paper, I recommend that the authors use the retinotopic task responses in each individual infant as the independent "Test" data, and compare the accuracy in predicting those responses, based on:

- The same infant's movie-watching data, analysed with MELODIC, when blind experimenters select components for the SF and meridian boundaries with no access to the ground-truth retinotopy data.<br /> - Anatomical parcellations in the same infant.<br /> - Shared response maps from groups of other infants or adults.<br /> - (If possible, ICA of resting state data, in the same infant, or from independent groups of infants).

Or, possibly, combinations of these techniques.

If the infant's own movie-watching data leads to improved predictions of the infant's retinotopic task-driven response, relative to these existing alternatives that don't require movie-watching data from the same infant, then the authors' main claim will be supported.

The proposed analysis above solves a critical problem with the analyses presented in the current manuscript: the data used to generate maps is identical to the data used to validate those maps. For the task-evoked maps, the same data are used to draw the lines along gradients and then test for gradient organization. For the component maps, the maps are manually selected to show the clearest gradients among many noisy options, and then the same data are tested for gradient organization. This is a double-dipping error. To fix this problem, the data must be split into independent train and test subsets.

Reviewer #3 (Public Review):

The manuscript reports data collected in awake toddlers recording BOLD while watching videos. The authors analyse the BOLD time series using two different statistical approaches, both very complex but do not require any a priori determination of the movie features or contents to be associated with regressors. The two main messages are that 1) toddlers have occipital visual areas very similar to adults, given that an SRM model derived from adult BOLD is consistent with the infant brains as well; 2) the retinotopic organization and the spatial frequency selectivity of the occipital maps derived by applying correlation analysis are consistent with the maps obtained by standard and conventional mapping.

Clearly, the data are important, and the author has achieved important and original results. However, the manuscript is totally unclear and very difficult to follow; the figures are not informative; the reader needs to trust the authors because no data to verify the output of the statistical analysis are presented (localization maps with proper statistics) nor so any validation of the statistical analysis provided. Indeed what I think that manuscript means, or better what I understood, may be very far from what the authors want to present, given how obscure the methods and the result presentation are.

In the present form, this reviewer considers that the manuscript needs to be totally rewritten, the results presented each technique with appropriate validation or comparison that the reader can evaluate.

Reviewer #2 (Public Review):

Summary:<br /> Developing a mechanical model of C. elegans is difficult to do from basic principles because it moves at a low (but not very small) Reynolds number, is itself visco-elastic, and often is measured moving at a solid/liquid interface. The ElegansBot is a good first step at a kinetic model that reproduces a wide range of C. elegans motiliy behavior.

Strengths:<br /> The model is general due to its simplicity and likely useful for various undulatory movements. The model reproduces experimental movement data using realistic physical parameters (e.g. drags, forces, etc). The model is predictive (semi?) as shown in the liquid-to-solid gait transition. The model is straightforward in implementation and so likely is adaptable to modification and addition of control circuits.

Weaknesses:<br /> Since the inputs to the model are the actual shape changes in time, parameterized as angles (or curvature), the ability of the model to reproduce a realistic facsimile of C. elegans motion is not really a huge surprise.

The authors do not include some important physical parameters in the model and should explain in the text these assumptions. 1) The cuticle stiffness is significant and has been measured [1]. 2) The body of C. elegans is under high hydrostatic pressure which adds an additional stiffness [2]. 3) The visco-elasticity of C. elegans body has been measured. [3]

There is only a very brief mention of proprioception. The lack of inclusion of proprioception in the model should be mentioned and referenced in more detail in my opinion.

These are just suggested references. There may be more relevant ones available.

1. Rahimi M, Sohrabi S, Murphy CT. Novel elasticity measurements reveal C. elegans cuticle stiffens with age and in a long-lived mutant. Biophys J. 2022 Feb 15;121(4):515-524. doi: 10.1016/j.bpj.2022.01.013. Epub 2022 Jan 19. PMID: 35065051; PMCID: PMC8874029.

2. Park SJ, Goodman MB, Pruitt BL. Analysis of nematode mechanics by piezoresistive displacement clamp. Proc Natl Acad Sci U S A. 2007 Oct 30;104(44):17376-81. doi: 10.1073/pnas.0702138104. Epub 2007 Oct 25. PMID: 17962419; PMCID: PMC2077264.

3. Backholm M, Ryu WS, Dalnoki-Veress K. Viscoelastic properties of the nematode Caenorhabditis elegans, a self-similar, shear-thinning worm. Proc Natl Acad Sci U S A. 2013 Mar 19;110(12):4528-33. doi: 10.1073/pnas.1219965110. Epub 2013 Mar 4. PMID: 23460699; PMCID: PMC3607018.

eLife assessment

This useful study introduces a simple mechanical model of C. elegans locomotion that captures aspects of the worm's behavioral repertoire beyond forward crawling. While the kinetic model (ElegansBot) provides a useful compromise and starting point to help understand the mechanical components of C. elegans behavior, the claim that this work improves on extant mechanical models is incomplete. In addition, the results of the application of the model to previously unstudied behaviors are primarily qualitative and do not produce new predictions.

Reviewer #1 (Public Review):

Summary:<br /> This work describes a simple mechanical model of worm locomotion, using a series of rigid segments connected by damped torsional springs and immersed in a viscous fluid. It uses this model to simulate forward crawling movement, as well as omega turns.

Strengths:<br /> The primary strength is in applying a biomechanical model to omega-turn behaviors. The biomechanics of nematode turning behaviors are relatively less well described and understood than forward crawling. The model itself may be a useful implementation to other researchers, particularly owing to its simplicity.

Weaknesses:<br /> The strength of the model presented in this work relative to prior approaches is not well supported, and in general, the paper would be improved with a better description of the broader context of existing modeling literature related to undulatory locomotion. This paper claims to improve on previous approaches to taking body shapes as inputs. However, the sole nematode model cited aims to do something different, and arguably more significant, which is to use experimentally derived parameters to model both the neural circuits that induce locomotion as well as the biomechanics and to subsequently compare the model to experimental data. Other modeling approaches do take experimental body kinematics as inputs and use them to produce force fields, however, they are not cited or discussed. Finally, the overall novelty of the approach is questionable. A functionally similar approach was developed in 2012 to describe worm locomotion in lattices (Majmudar, 2012, Roy. Soc. Int.), which is not discussed and would provide an interesting comparison and needed context.

The idea of applying biomechanical models to describe omega turns in C. elegans is a good one, however, the kinematic basis of the model as used in this paper (the authors do note that the control angle could be connected to a neural model, but don't do so in this work) limits the generation of neuromechanical control hypotheses. The model may provide insights into the biomechanics of such behaviors, however, the results described are very minimal and are purely qualitative. Overall, direct comparisons to the experiments are lacking or unclear. Furthermore, the paper claims the value of the model is to produce the force fields from a given body shape, but the force fields from omega turns are only pictured qualitatively. No comparison is made to other behaviors (the force experienced during crawling relative to turning for example might be interesting to consider) and the dependence of the behavior on the model parameters is not explored (for example, how does the omega turn change as the drag coefficients are changed). If the purpose of this paper is to recapitulate the swim-to-crawl transition with a simple model, and then apply the model to new behaviors, a more detailed analysis of the behavior of the model variables and their dependence on the variables would make for a stronger result. In some sense, because the model takes kinematics as an input and uses previously established techniques to model mechanics, it is unsurprising that it can reproduce experimentally observed kinematics, however, the forces calculated and the variation of parameters could be of interest.

Relatedly, a justification of why the drag coefficients had to be changed by a factor of 100 should be explored. Plate conditions are difficult to replicate and the rheology of plates likely depends on a number of factors, but is for example, changes in hydration level likely to produce a 100-fold change in drag? or something more interesting/subtle within the model producing the discrepancy?

Finally, the language used to distinguish different modeling approaches was often unclear. For example, it was unclear in what sense the model presented in Boyle, 2012 was a "kinetic model" and in many situations, it appeared that the term kinematic might have been more appropriate. Other phrases like "frictional forces caused by the tension of its muscles" were unclear at first glance, and might benefit from revision and more canonical usage of terms.

Reviewer #3 (Public Review):

Summary:<br /> A mechanical model is used with input force patterns to generate output curvature patterns, corresponding to a number of different locomotion behaviors in C. elegans

Strengths:<br /> The use of a mechanical model to study a variety of locomotor sequences and the grounding in empirical data are strengths. The matching of speeds (though qualitative and shown only on agar) is a strength.

Weaknesses:<br /> What is the relation between input and output data? How does the input-output relation depend on the parameters of the model? What biological questions are addressed and can significant model predictions be made?

Author Response

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

We would like the reviewers for their positive and useful comments. Below please find our answers to the issues raised.

Reviewer #1 (Public Review):

Overall, the experiments are well-designed and the results of the study are exciting. We have one major concern, as well as a few minor comments that are detailed in the following.

Major:

1) The authors suggest that "Visuomotor experience induces functional and structural plasticity of chandelier cells". One puzzling thing here, however, is that mice constantly experience visuomotor coupling throughout life which is not different from experience in the virtual tunnel. Why do the authors think that the coupled experience in the VR induces stronger experience-dependent changes than the coupled experience in the home cage? Could this be a time-dependent effect (e.g. arousal levels could systematically decrease with the number of head-fixed VR sessions)? The control experiment here would be to have a group of mice that experience similar visual flow without coupling between movement and visual flow feedback.

Either change would be experience-dependent of course, but having the "visuomotor experience dependent" in the title might be a bit strong given the lack of control for that. We would suggest changing the pitch of the manuscript to one of the conclusions the authors can make cleanly (e.g. Figure 4).

Although the plasticity is induced by the visuomotor experience in the tunnel, we agree that we do not know what aspect of the repeated exposure to the virtual tunnel caused the plasticity. We cannot rule out that it was the exposure to the visual stimuli alone that caused it. Therefore, we rephrased sentences that suggested that it was the coupling between visual stimuli and motor behavior that was responsible for the plasticity. We also changed the title to “Experience Shapes Chandelier Cell Function and Structure in the Visual Cortex”.

We do believe that training the mice in the virtual tunnel does significantly increase experience with coupled visuomotor activity, though. In their home cage, mice are mostly active in the dark and there is litle space to run.

Minor:

2). "ChCs shape the communication hierarchy of cortical networks providing visual and contextual information." We are not sure what this means.

We thank the reviewer for helping to raise clarity and we rephrased this sentence to: “…ChCs may establish a hierarchical relationship among cortical networks.”

3) "respond to locomotion and visuomotor mismatch, indicating arousal-related activity" This is not clear. We think we understand what the authors mean but would suggest rephrasing.

Agreed. We rephrased this sentence to: "...respond to events that are known to increase arousal levels, such as locomotion and visuomotor mismatch.”

4) 'based on morphological properties revealed that 87% (287/329) of labeled neurons were ChCs" Please specify the morphological properties used for the classification somewhere in the methods.

We added that the neurons were positioned at the border of L1 and L2 and had a dendrite reaching into layer 1.

5) We may have missed this - in the patch clamp experiment (Fig.1 H-K), please add information about how many mice/slices these experiments were performed in.

We have added the information to the legend of Fig. 1.

6) "These findings suggest that the rabies-labeled L1-4 neurons providing monosynaptic input to ChCs are predominantly inhibitory neurons". We are not sure this conclusion is warranted given the sparse set of neurons labelled and the low number of cells recorded in the paired patch experiment. We would suggest properly testing (e.g. stain for GABA on the rabies data) or rephrasing.

We weakened the statement to: “These findings suggest that the rabies-labeled L1-4 neurons providing monosynaptic input to ChCs may include many inhibitory neurons.”

7) Figure 2E. A direct comparison of dF/F across different cell types can be subject to a problematic interpretation. The transfer function from spikes to calcium can be different from cell type to cell type. Additionally, the two cell populations have been marked with different constructs (despite the fact that it's the same GECI) further reducing the reliability of dF/F comparisons. We would recommend using a different representation here that does not rely on a direct comparison of dF/F responses (e.g. like the "response strength" used in Figure 3B). Assuming calcium dynamics are different in ChCs and PyCs - this similarity in calcium response is likely a coincidence.

We have removed the quantification in this figure.

8) If ChCs are more strongly driven by locomotion and arousal, then it's a bit counterintuitive that at the beginning of the visual corridor when locomotion speed consistently increases, the activity of ChCs consistently decreases. This does not appear to be driven by suppression by visual stimuli as it is present also in the first and last 20cm of the tunnel where there are no visual stimuli. How do the authors explain this?

We do believe that this is suppression driven by visual stimuli. Although on average the strongest visual suppression happens between 20-80 cm, neurons that have their receptive fields toward the center of the visual field will already respond to the stimuli before the mouse reaches the 20 cm location of the tunnel. In addition, although the visual stimuli are the strongest sensory inputs, the background of the visual part of the tunnel has a black and white noise patern, which might already mildly suppress ChC activity. Both arguments are supported by the observation that the visual PyCs (V-PyCs, blue line) in Fig. 4D are already activated at the beginning of the tunnel and that the activity of V-PyCs matches well with the suppression of ChC activity.

9) The authors mention that "ChC responses underwent sensory-evoked plasticity during the repeated visual exposure, even though the visual stimuli were different from those encountered during training in the virtual tunnel". How would this work? And would this mean all visual responses are reduced? What is special about the visual experience in the virtual tunnel? It does not inherently differ from visual experience in the home cage, given that the test stimuli (full field gratings) are different from both.

As mentioned in our answer to point 1, the exposure to visual stimuli is strongly increased since, firstly, they are presented during the dark phase when the mice are most active and, secondly, they do not get these types of visual inputs in their home cage.

10) Just as a point to consider for future experiments: For the open-loop control experiments, the visual flow is constant (20cm/s) - ideally, this would be a replay of the running speed the mouse previously generated to match statistics.

We agree with this point and will implement replay of earlier sessions in future experiments.

11) We would recommend specifying the parameters used for neuropil correction in the methods section.

This is described on page 24, under “preprocessing”. We also refer to the analysis package (Spectral Segmentation - SpecSeg) in which the neuropil correction as used by us here is explained in more detail.

12) If we understand correctly, the F0 used for the dF/F calculation is different from that used for division. Why is this?

We apologize for this mistake, which was based on an older version of the software. We have now corrected this in the revised manuscript.

13) Authors compare neuronal responses using "baseline-corrected average". Please specify the parameters of the baseline correction (i.e. what is used as baseline here).

In the revised version we have now beter explained this in the methods, page 24, under “Passive Sessions”.

Reviewer #2 (Public Review):

Summary:

Seignete et al. investigated the potential roles of axo-axonic (chandelier) cells (ChCs) in a sensory system, namely visual processing. As introduced by the authors, the axo-axonic cell type has remained (and still is) somehow mysterious in its function. Seignete and colleagues leveraged the development of a transgenic mouse line selective for ChC, and applied a very wide range of techniques: transsynaptic rabies tracing, optogenetic input activation, in vitro electrophysiology, 2-photon recording in vivo, behavior and chemogenetic manipulations, to precisely determine the contribution of ChCs to the primary visual cortex network.

The main findings are 1) the identification of synaptic inputs to ChC, with a majority of local, deep layer principal neurons (PN), 2) the demonstration that ChC is strongly and synchronously activated by visual stimuli with low specificity in naive animals, 3) the recruitment of ChC by arousal/visuomotor mismatch, 4) the induction of functional and structural plasticity at the ChC-PN module, and, 5) the weak disinhibition of PNs induced by ChCs silencing. All these findings are strongly supported by experimental data and thoroughly compared to available evidence.

Strengths:

This article reports an impressive range of very demanding experiments, which were well executed and analyzed, and are presented in a very clear and balanced manner. Moreover, the manuscript is well- writen throughout, making it appealing to future readers. It has also been a pleasure to review this article.

In sum, this is an impressive study and an excellent manuscript, that presents no major flaws.

Notably, this study is one of the first studies to report on the activities and potential roles of axo-axonic cells in an active, integrated brain process, beyond locomotion as reported and published in V1. This type of research was much awaited in the fields of interneuron and vision research.

Weaknesses:

There are no fundamental weaknesses; the later mainly concern the presentation of the main results. The main weakness may be that the different sections appear somehow disconnected conceptually.

Additionally, some parts deserve a more in-depth clarification/simplification of concepts and analytic methods for scientists outside the subfield of V1 research. Indeed, this paper will be of key interest to researchers of various backgrounds.

Reviewer #3 (Public Review):

Summary:

The authors set out to characterize the anatomical connectivity profile and the functional responses of chandelier cells (ChCs) in the mouse primary visual cortex. Using retrograde rabies tracing, optogenetics, and in vitro electrophysiology, they found that the primary source of input to ChCs are local layer 5 pyramidal cells, as well as long-range thalamic and cortical connections. ChCs provided input to local layer 2/3 pyramidal neurons, but did not receive reciprocal connections.

With two-photon calcium imaging recordings during passive viewing of drifting gratings, the authors showed that ChCs exhibit weakly selective visual responses, high correlations within their own population, and strong responses during periods of arousal (assessed by locomotion and pupil size). These results were replicated and extended in experiments with natural images and prediction of receptive field structure using a convolutional neural network.

Furthermore, the authors employed a learned visuomotor task in a virtual corridor to show that ChCs exhibit strong responses to mismatches between visual flow and locomotion, locomotion-related activation (similar to what was shown above), and visually-evoked suppression. They also showed the existence of two clusters of pyramidal neurons with functionally different responses - a cluster with "classically visual" responses and a cluster with locomotion- and mismatch-driven responses (the later more correlated with ChCs). Comparing naive and trained mice, the authors found that visual responses of ChCs are suppressed following task learning, accompanied by a shortening of the axon initial segment (AIS) of pyramidal cells and an increase in the proportion of AIS contacted by ChCs. However, additional controls would be required to identify which component(s) of the experimental paradigm led to the functional and anatomical changes observed.

Finally, using a chemogenetic inactivation of ChCs, the authors propose weak connectivity to pyramidal cells (due to small effects in pyramidal cell activity). However, these results are not unequivocally supported, as the baseline activity of ChCs before inactivation is considerably lower, suggesting a potentially confounding homeostatic plasticity mechanism might already be operating.

Strengths:

The authors bring a comprehensive, state-of-the-art methodology to bear, including rabies tracing, in vivo two-photon calcium imaging, in vitro electrophysiology, optogenetics and chemogenetics, and deep neural networks. Their analyses and statistical tests are sound and for the most part, support their claims. Their results are in line with previous findings and extend them to the primary visual cortex.

Weaknesses:

• Some of the results (e.g. arousal-related responses) are not entirely surprising given that similar results exist in other cortical areas.

We agree that previous studies have shown arousal-related responses of ChC cells and our study confirms those findings. However, this is not the main message of the article and we present many findings that are novel.

• Control analyses regarding locomotion paterns before and atier learning the task (Figure 5), and additional control experiments to identify whether functional and anatomical changes following task learning were due to learning, repeated visual exposure, exposure to reward, or visuomotor experience would strengthen the claims made.

In figure 5 we excluded running trials, so locomotion paterns are unlikely to play a major role. We agree that testing what are the factors that contribute to the observed plasticity are important to investigate in future experiments.

• The strength of the results of the chemogenetics experiment is impacted by the lower baseline activity of ChCs that express the KORD receptor. At present, it is not possible to exclude the presence of homeostatic plasticity in the network before the inactivation takes place.

Although we do not know why there is a difference in the baseline df/f (e.g. expression levels), we consider it unlikely that expression of the KORD receptor itself without exposure to the ligand causes reduction of ChC activity. Moreover, we are not sure how homeostatic plasticity in the network would occur selectively in KORD-expressing ChCs. Finally, we do not find evidence for a relationship between lower ChC calcium signals and the effects of ChC silencing on PyC activity. We performed an additional analysis in which we correlated baseline ChC activity (before salvinorin B injection) with the effect of ChC silencing on PyC activity (post – pre) across mice, and found that this correlation was not significant (R = 0.41, p = 0.18).

Reviewer #1 (Recommendations For The Authors):

In the spirit of openness of the scientific discussion, all our feedback and recommendations to the authors are included in the public reviews.

Reviewer #2 (Recommendations For The Authors):

Most of my comments and suggestions concern the presentation of the data, to (hopefully) help and convey as clearly as possible the messages of this important article.

Main

The main weakness of the paper may be that the different sections appear somehow disconnected conceptually. This is particularly true for:

-structural plasticity: how can we link this finding with the rest of the study? Are there ways to correlate this finding with physiological recordings in individual animals, or to directly test whether particular functional types of PNs (visual, non-visual) undergo plasticity at their AIS?

This is a very interesting question that may be addressed in future experiments.

-the indirect finding suggesting that ChC weakly inhibits PNs using chemogenetic silencing of PNs. Do chemogenetic manipulations of ChCs affect PN responses in visual paradigm and/or modify the induction of structural plasticity at the ChC-AIS connection?

This is also a very interesting question for future work.

Additionally, some parts would deserve a more in-depth clarification/simplification of concepts and analytic methods (OSI, DSI, MEI...) for scientists outside the subfield of V1 research. Indeed, this paper will be of key interest to researchers of various backgrounds.

In the revised manuscript we briefly explain what an MEI is when first introduced, and introduce the abbreviations OSI and DSI at the correct location. We believe orientation and direction selectivity are well-known concepts for the audience reading this article.

Minor

These are discussed by order of appearance in the text.

Abstract

The alternative interpretation of error/mismatch negativity to explain ChC activation deserves to appear in the abstract. Arousal consistency in prediction should be in the introduction. "In mice running in a virtual tunnel, ChCs respond strongly to locomotion and halting visual flow, suggesting arousal-related activity."

This comment holds for the end of the introduction and the beginning of the discussion, as well.

"These findings suggest that ChCs provide an arousal-related signal to layer 2/3 pyramidal cells that may modulate their activity". This statement appears to be in contradiction with the weak effect mentioned just before. This comment holds for the end of the introduction.

The full sentence was: “These findings suggest that ChCs provide an arousal-related signal to layer 2/3 pyramidal cells that may modulate their activity and/or gate plasticity of L2/3 PyCs in V1.” Our results show that activity of layer 2/3 pyramidal cells is modulated (albeit weakly) and it is well possible that ChCs regulate plasticity at the AIS. Therefore, we do not believe that this statement contradicts the weak direct effect of ChCs on layer 2/3 pyramidal cell activity. Therefore , we think that this statement does not contradict the weak direct effect of ChCs on layer 2/3 pyramidal cell activity.

We changed the last sentence of the introduction to “Our findings suggest that ChCs predominantly respond to arousal related to locomotion or unexpected events/stimuli, and act to weakly modulate activity and/or gate plasticity of L2/3 PyCs in V1.”

Introduction First paragraph

Coming from a field outside of vision research, it is not obvious to me what has been learned from interneuron classes in the past. An example would be welcome in the introduction.

The literature on the role of different interneuron types in visual processing and plasticity is too large to pick one or two examples. For the sake of conciseness, we have therefore provided some important references and reviews for the interested readers (references 1 to 10).

Interneuron "subtypes" seem to refer to main classes (e.g. PV+): please rephrase accordingly (ChC being a type and PV+ ChC a subtype).

We changed interneuron “subtypes” to “types” and left L2/3 pyramidal cell “subtypes” unchanged.

Second paragraph

Beyond the reversal potential of GABA-ARs at the axon initial segment, GABA may inhibit action potential generation in various conditions (Lipkin et al. 2023, DOI: 10.1523/JNEUROSCI.0605-23.2023 : should be cited).

We added this citation.

Fourth paragraph

"ChCs alter the number of synapses at the AIS based on the activity of their postsynaptic targets": the concept of alteration is too vague to let the reader grasp the concept: could the authors rephrase?

We have rephrased the sentence to:

“…ChCs increase the number of synapses at the AIS if their postsynaptic targets are chemogenetically activated…”

Results 1) ChCs receive input from long-range sources and L5 PyCs in V1 It is not clear how morphological identification of ChC was performed. Did dendrites and/or axons of starter cells occasionally overlap as can be expected, complicating the cell-by-cell morphological classification?

"Most labeled neurons were located on the border between L1 and L2/3 and displayed typical ChC morphology": maybe clarify that this concerns neurons expressing eYFP-TVA?

We assessed the location (at the border of L1 and L2) and spatial distribution of the labeled cells and whether they had a dendrite extending upwards towards into L1. We have now indicated this in the results section and clarified that these neurons express eYFP-TVA.

-Likewise the following would benefit from clarification " This is further supported by the distributed localization of the labeled neurons": it would also help here to remind the reader of the labelling (presumably retrogradely-labeled mCherrry+ neurons).

We have now clarified in the text that these are mCherry+ neurons labeled by the rabies virus

2) Chandelier cells are modulated by arousal and show high correlations

-The authors indicate that the results "(suggest) that ChCs distribute a synchronized signal during high arousal." : it would be stronger to defend this claim by showing a higher ChC-ChC correlation during "arousal" vs. baseline (i.e. analyze high arousal epochs outside of movement). It may be difficult to perform this analysis due to low fluorescence changes outside running episodes, but this should be discussed accordingly. In this respect, the title of the section is more in line with the data presented.

We believe our statement is correct. The activity of ChCs is highly synchronized and their firing rates increase during arousal. We do not state that synchronization increases with arousal.

-A brief explanation of DSI and OSI meaning would be nice for the audience that will definitely extend beyond vision research given the importance of this study.

See above

3) ChCs are weakly selective to visual information

-I may very well miss the point, but the equivalence in response strength among cell classes (Fig3B) seems inconsistent with the wider distribution of high response strength in ChCs (Fig3C). Perhaps a graphical representation taking into account the distribution of single data points in Fig3B would help resolve this discrepancy.

This is because in panel C the response strengths are normalized. We now also state this in the legend to avoid confusion.

-"clearly oriented edge-like paterns with sharp ON and OFF regions": it would help if a representative example was highlighted in Figure 3F.

The majority of L2/3 pyramidal MEIs presented in this panel show this patern.

-It is interesting and surprising that properties of ChCs appear more distinct from those of L5 PNs than from those of L2-3 PNs (Fig 3G-J), given the fact that V1 ChCs were found by the authors to derive their inputs from V1 L5 PNs (please see comments of the discussion for this specific point).

How ChCs respond based on L5 input depends strongly on how the connections between L5 and ChCs are organized. Similarity between responses of L5 and ChC neurons is not required.

4) Locomotion and visuomotor mismatch drive chandelier cell activity in a virtual tunnel This is the least convincing part in terms of presentation.

-It is unclear where/when visuomotor mismatch has been induced in the tunnel: please clarify in the text and in Fig 4B.

We realized that the title of the paragraphs was indeed confusing. In fig. 4A-D and the first paragraph about the virtual tunnel, we do not discuss the visuomotor mismatch. This comes later, when we describe the results in Fig. 4E. The titles have been changed.

-No result on visuomotor mismatch is reported in the text of this section, while this is presented in the subsequent section: this needs to be corrected (merge this section with the next?).

We agree, apologies for the confusion. See above.

-It would be interesting to further analyze responses to CS and US. Regarding the US: is water rewarding in non-water-restricted mice? This should be mentioned.

We realized that we did not mention that the mice were water restricted during behavioral training and during the imaging sessions when mice performed the virtual tunnel task. We have now added this to the methods section. Sorry for the omission.

-Along this line: was water sometimes omited? This would provide a complementary way to test the prediction error theory for ChC activation with an alternative modality.

We never omited the water reward. It would be interesting to test this in a future experiment.

5) ChCs have similar response properties as non-visual PyCs

• It would help to explicitly mention that in Ai65 mice, only Cre and Flp+ cells express tdTomato (here Vipr2 and PV+).

We added the following sentence: “In these mice, tdTomato was only expressed in cells expressing both Vipr2 and PV.”

6) Visuomotor experience in the virtual tunnel induces plasticity of ChC-AIS connectivity

• In relation to the previous section, Jung et al. (doi.org/10.1038/s41593-023-01380-x) recently reported that motor learning reduced ChC-ChC synchrony in M2. Did the author observe a similar change in ChC- ChC synchrony with visual experience/habituation to the task? If available, these data should be reported to help build a clearer picture of ChC functions in the neocortex.

We tested this and also found reduced correlations between ChCs in trained mice vs naïve mice. We added this as text on p14 in the results section.

• The low number of ChC boutons' appositions per AIS may be misleading: "While the average number of ChC boutons per AIS remained constant (~2-3 ChC boutons/AIS)"). It would be helpful to make it clear that these are "virally" labelled boutons, as opposed to absolute numbers, if compared with the detailed quantification of Schneider-Mizell et al, 2021 (7.4 boutons per AIS in average; doi: 10.7554/eLife.73783.).

We added "virally labeled"

• It may be difficult to clearly isolate boutons in light microscopic images of ChC boutons. could the authors comment on this and explain how they solved this issue (in the methods section for instance)?

We elaborated on our definition of a bouton under confocal microscopy conditions. We also added that the analysis was performed under blinded conditions for the experimenter (i.e. the experimenter did not know whether the images came from trained or untrained mice).

• Is there any suggestion for heterogeneity/selectivity for a subset of PNs (the distribution does not seem to show this, though)? It would be interesting to discuss this and try to link this finding to the rest of the study a bit more directly. Future work could also investigate if genetically defined PN types undergo different pre-synaptic plasticity at their AISs (e.g. work cited by the authors by O'Toole et al, 2023 doi: 10.1016/j.neuron.2023.08.015 -this reference can be updated as well, since the work has been published in the meantime).

In our data, we did not find evidence for heterogeneity or selectivity of targeting, also not in the physiology using KORD (see below). We do agree that it is an interesting question and deserves atention in future experiments. We also updated the reference.

7) ChCs weakly inhibit PyC activity independent of locomotion speed

The authors state that "recent work in adult mice has reported hyperpolarizing and shunting effects in prelimbic cortex, S1 and hippocampus (18, 26, 27)": however, to my knowledge studies presented in refs 26 & 27 found reduced activity/firing of PNs upon optogenetic activation of ChCs in vivo, but did not perform intracellular recordings to assess GABA-A reversal potential at the AIS. I would like to kindly ask the authors to correct this sentence.

If the polarity of responses is discussed, they may rather refer to the corresponding literature including Rinetti Vargas et al (doi: 10.1016/j.celrep.2017.06.030), Lipkin et al (doi: 10.1523/JNEUROSCI.0605- 23.2023), and Khirug et al (doi: 10.1523/JNEUROSCI.0908-08.2008.).

We added the reference to Lipkin et al and changed the sentence so that it matches the references..

• In an atempt to link findings from several parts of the article, did the authors investigate whether chemogenetic effects were different in visual vs non-visual PNs? As ChCs are functionally related to visual PNs, one might indeed speculate that these cells are synaptically connected.

We did not find evidence for selectivity in the chemogenetic effect. We compared the chemogenetic effect to locomotion modulation (see text accompanying Fig 7.) – based on our observation that non- visual PyCs were more strongly modulated by locomotion (see Fig. 4) – but did not find any significant correlation.

• " We first looked at the average activity of neurons in both essions.": sessions

Thank you for noticing. We corrected this.

Discussion

Summary of findings

-It would be worthwhile to include in the summary the finding of mismatch-related activity, that appears to explain more convincingly ChC activation than arousal per se (with the data available).

We updated the summary of the discussion accordingly.

-Moreover, the last part of the article (weak inhibition of PNs by ChCs), despite being very important, is not mentioned.

We now mention this in the summary of the discussion (“Finally, ChCs only weakly inhibit PyCs.”)

Discussion of findings

-" Optogenetic activation of cortical feedback": it is not clear what the authors mean by cortical feedback. As RS was retrogradely labeled, this region may rather provide feedforward inhibition to V1 via ChCs.

Retrosplenial cortex is a higher order cortical area and only provides feedback to V1.

-"This means that each ChC receives input from many L5 PyCs, which could explain the low selectivity of ChC responses we observed to natural images compared to those of L2/3 and L5 PyCs". : perhaps state explicitly that the convergence of many PN inputs each carrying different RF/visual properties "averages out" in ChC (as you do a few lines below for MEI).

At this point, we do not know how the connections from L5 to ChCs are organized. Whether this converge results in “average out” is therefore not so certain. We have made an atempt to clarify the situation. (“This convergence of L5 PyC inputs, if not strongly organized, could explain the low selectivity of ChC responses we observed to natural images compared to those of L2/3 and L5 PyCs.”)

-"However, we did not identify neuromodulatory inputs to ChCs in our rabies tracing experiment. Possibly, these inputs act predominantly through extrasynaptic receptors and were therefore not labeled by the transsynaptic rabies approach.": here, the authors should cite the work by Lu et al (doi: 10.1038/nn.4624) which found basal forebrain (diagonal band of Broca) cholinergic inputs to ChC of the PFC in the Nkx2.1CreER mouse model. Moreover, the authors should discuss potential technical differences (?) responsible for this discrepancy. Beyond the extrasynaptic release of neuromodulators, rabies strains may display different tropism profiles for neuron classes.

We have now added a sentence discussing this and added the reference in the revised manuscript.

-The section dedicated to prediction error is particularly interesting and relevant. In my opinion, this interpretation should be further emphasized in the abstract and summary of findings paragraph in the discussion (as already indicated).

Yes, we agree and have added some emphasis.

-" These findings are thus in contrast with the general notion that ChCs exert powerful control over PyC output (28, 78), but consistent with computational simulations predicting a relatively small inhibitory effect of GABAergic innervation of the AIS, possibly involving shunting inhibition (79, 80)." These findings are also consistent with results from PFC and dCA1 studies showing, with electrophysiological recordings combined with optogenetic stimulation of ChCs, that a small proportion of putative PNs was inhibited upon ChC stimulation (doi: 10.1038/nn.4624 doi: 10.1016/j.neuron.2021.09.033).

Perhaps the effect of ChCs is limited in all these experiments by a suboptimal efficiency of ChC targeting. Moreover, inhibition might be restricted to a subset of PNs carrying a specific function. This could be discussed.

We added an explanation for the weak effects of silencing to the discussion and stated that our results are in line with findings in PFC and CA1. (“One explanation for the weak effects we observed is the high variability in the number of GABAergic boutons that PyCs receive at their AISs. Possibly, only a smaller fraction of PyCs with high numbers of AIS synapses are inhibited when ChCs are active. Indeed, we find that only a small fraction of PyCs increased their activity upon chemogenetic silencing of ChCs, in line with findings by others showing that manipulating ChC activity in vivo has relatively weak effects on small populations of PyCs (27, 28).”)

Although we cannot rule out that ChC targeting is suboptimal in our and other experiments, the expression of the KORD receptor as visualized by mCyRFP1 fluorescence appeared very strong. In addition, the common notion in the ChC field is that ChCs exert powerful control over PyC firing. Even suboptimal labeling should in that case show clear inhibitory effects. Similar experiments with PV+ interneurons would show very convincing inhibition, even if labeling is suboptimal. To keep the discussion concise, we prefer to leave this particular point out.

-" ChC activation could prevent homeostatic AIS shortening of L2/3 PyCs if their activity occurs during behaviorally relevant, arousal inducing events": this postulate seems to be very interesting but is not very clear and lacks some mechanistic speculation.

We considered elaborating more on this hypothesis. However – given that it is merely a speculation at this point – we do not wish to lengthen the discussion further on this point.

• A reference to previous studies demonstrating high levels of synchronous ChC activities is missing: the authors may cite Dudok et al., Schneider-Mizell et al., and Jung et al. (and discuss a change in synchrony with learning or habituation in the case of this study; see above).

We have now also referred to these papers in the context of high correlations between ChCs.

Methods

Beyond references to reagents (eg antibodies, viruses), lot numbers should be provided whenever this is possible. Indeed, there might be strong lot-to-lot variations in specificity and efficiency.

Reviewer #3 (Recommendations For The Authors):

Major:

• (Figure 5) Control analysis missing. Mice before and after training in VR will almost definitely exhibit different running paterns when viewing driftng gratings. Since ChCs are strongly modulated by locomotion, assess whether results depend on changes in running.

Although we did not compare locomotion paterns before and after training, we removed all trials in which the mice were running (see methods). Therefore, we can exclude that these results are caused by changes in running behavior.

• (Figure 5 & 6) What would happen with simple passive visual experience, not in a visuomotor task? What if there was no reward? What if there was an open-loop experiment with random reward? To which specific aspect of the experiment are the results atributable?

These are indeed very interesting questions that may be tested in future experiments.

(Figure 7 B, H) The pre-injection ChC activity in the KORD group is less than 50% of that in control mice! Discuss the effect of such a shift in baseline. Plasticity of PyCs even before ChC inactivation?

See answer to the above question in the public section of reviewer 3.

• (Figure 3 H) Contrast tuning results, as far as I understand, come only from the CNN. However, if I understood correctly, during the passive viewing of gratings there were already different contrasts. Why not show contrast tuning there? Do the results disagree?

We did indeed show stimuli at different contrasts during the passive viewing of gratings. Although the results from those recordings were not optimal for defining contrast sensitivity, they also showed that ChC responses were less modulated by contrast than PyCs.

Minor: - (Figure 3) Explain the potential impact of different indicators 8m vs 6f due to different baselines and dynamics.

We believe there is no impact of different indicators, because for the CNN analyses we estimated spikes using CASCADE. This toolbox is specifically designed to generalize across different calcium indicators. Although GCaMP8m was not included in their training set, the wide variety of indicators used provides a solid basis for generalizable spike estimation. Importantly, comparisons between L2/3 PyCs and ChCs also would not be affected by this concern.

• (Figure 4) NV-PyCs. Would you call all of these mismatch-responsive neurons? Discuss the difference in the percentage of neurons (more than 50% of total PyCs here, compared to significantly less - up to 40% in previous studies, as far as I'm aware)

Not all NV-PyCs appeared to be mismatch-responsive neurons.

• (Figure 6 D) No error bars?

This is a representation of the fraction of all contacted AISs, which has no error bars indeed.

• (Figure 6 E-F and H-I) These pairs of panels contain essentially the same information. The first panel of each pair seems redundant.

We prefer to keep both plots in place, as in this case the skewness of the histogram can be helpful, which is less clear in the boxplot (which in itself displays the quantiles beter).

• The equation for direction tuning still has ang_ori, instead of ang_dir which I'm assuming should be there.

Thank you for noticing, we corrected it.

• The response for drifting gratings is calculated from a different interval (0.2-1.2s) compared to natural images (0-0.5s). Why?

Because we used spike probability in the case of the natural images to shorten the signal, and the visual stimuli were presented for 0.5 s (instead of 1 s as with the gratings).

Very minor:

• It would be helpful for equations to have numbers.

Done

• Sparsity equation. Beter to have it as a general equation, with N instead of 40. Then below it can be explained that N is the number of images = 40.

Done

• "The similarity of these MEIs with those we found for ChCs is in line with the idea that ChCs are driven by input from a large number of L5 PyCs (but do not exclude alternative explanations)." - in parenthesis it should be does not exclude.

Corrected.

• "In contrast, the response strength of PyCs was only mildly and non-significantly reduced after training"

• statistically non-significant..

Corrected.

"We first looked at the average activity of neurons in both essions." - sessions

Corrected.

• (Figure 7 C) Explain what points and error bars represent

Done.

Reviewer #3 (Public Review):

Summary:<br /> The authors set out to characterize the anatomical connectivity profile and the functional responses of chandelier cells (ChCs) in the mouse primary visual cortex. Using retrograde rabies tracing, optogenetics, and in vitro electrophysiology, they found that the primary source of input to ChCs are local layer 5 pyramidal cells, as well as long-range thalamic and cortical connections. ChCs provided input to local layer 2/3 pyramidal neurons, but did not receive reciprocal connections.

With two-photon calcium imaging recordings during passive viewing of drifting gratings, the authors showed that ChCs exhibit weakly selective visual responses, high correlations within their own population, and strong responses during periods of arousal (assessed by locomotion and pupil size). These results were replicated and extended in experiments with natural images and prediction of receptive field structure using a convolutional neural network.

Furthermore, the authors employed a learned visuomotor task in a virtual corridor to show that ChCs exhibit strong responses to mismatches between visual flow and locomotion, locomotion-related activation (similar to what was shown above), and visually-evoked suppression. They also showed the existence of two clusters of pyramidal neurons with functionally different responses - a cluster with "classically visual" responses and a cluster with locomotion- and mismatch-driven responses (the latter more correlated with ChCs). Comparing naive and trained mice, the authors found that visual responses of ChCs are suppressed following task learning, accompanied by a shortening of the axon initial segment (AIS) of pyramidal cells and an increase in the proportion of AIS contacted by ChCs. However, additional controls would be required to identify which component(s) of the experimental paradigm led to the functional and anatomical changes observed.

Strengths:<br /> The authors bring a comprehensive, state-of-the-art methodology to bear, including rabies tracing, in vivo two-photon calcium imaging, in vitro electrophysiology, optogenetics and chemogenetics, and deep neural networks. Their analyses and statistical tests are sound and for the most part, support their claims. Their results are in line with previous findings and extend them to the primary visual cortex.

Weaknesses:<br /> - Some of the results (e.g. arousal-related responses) are not entirely surprising given that similar results exist in other cortical areas.

Reviewer #2 (Public Review):

Summary:<br /> Seignette et al. investigated the potential roles of axo-axonic (chandelier) cells (ChCs) in a sensory system, namely visual processing. As introduced by the authors, the axo-axonic cell type has remained (and still is) somehow mysterious in its function. Seignette and colleagues leveraged the development of a transgenic mouse line selective for ChC, and applied a very wide range of techniques: transsynaptic rabies tracing, optogenetic input activation, in vitro electrophysiology, 2-photon recording in vivo, behavior and chemogenetic manipulations, to precisely determine the contribution of ChCs to the primary visual cortex network.

The main findings are 1) the identification of synaptic inputs to ChC, with a majority of local, deep layer principal neurons (PN), 2) the demonstration that ChC is strongly and synchronously activated by visual stimuli with low specificity in naive animals, 3) the recruitment of ChC by arousal/visuomotor mismatch, 4) the induction of functional and structural plasticity at the ChC-PN module, and, 5) the weak disinhibition of PNs induced by ChCs silencing. All these findings are strongly supported by experimental data and thoroughly compared to available evidence.

Strengths:<br /> This article reports an impressive range of very demanding experiments, which were well executed and analyzed, and are presented in a very clear and balanced manner. Moreover, the manuscript is well-written throughout, making it appealing to future readers. It has also been a pleasure to review this article.

In sum, this is an impressive study and an excellent manuscript, that presents no major flaws.

Notably, this study is one of the first studies to report on the activities and potential roles of axo-axonic cells in an active, integrated brain process, beyond locomotion as reported and published in V1. This type of research was much awaited in the fields of interneuron and vision research.

Weaknesses:<br /> There are no fundamental weaknesses; the latter mainly concern the presentation of the main results.

The main weakness may be that the different sections appear somehow disconnected conceptually.

Additionally, some parts deserve a more in-depth clarification/simplification of concepts and analytic methods for scientists outside the subfield of V1 research. Indeed, this paper will be of key interest to researchers of various backgrounds.

eLife assessment

BMP signaling plays a vital role in skeletal tissues, and the importance of its role in microtia prevention is novel and promising. This important study will shed light on the role of BMP signaling in preventing microtia in the ear. Solid data broadly support the claims with only minor weaknesses.

Reviewer #1 (Public Review):

Summary:<br /> In this manuscript, Ruichen Yang et al. investigated the importance of BMP signaling in preventing microtia. The authors showed that Cre recombinase-mediated deletion of Bmpr1a using skeletal stem-specific Cre Prx1Cre leads to microtia in adult and young mice. In these mice, the distal auricle is more affected than the middle and proximal. In these Bmpr1a floxed Prx1Cre mice, auricle chondrocytes start to differentiate into osteoblasts through an increase in PKA signaling. The authors showed human single-cell RNA-Seq data sets where they observed increased PKA signaling in microtia patients which resembles their animal model experiments.

Strengths:<br /> Although the importance of BMP signaling in skeletal tissues has been previously reported, the importance of its role in microtia prevention is novel and very promising to study in detail. The authors satisfied the experimental questions by performing the correct methods and explaining the results in detail.

Weaknesses:<br /> There are minor concerns like typo mistakes and missing control data histology pictures which should be corrected.

Reviewer #2 (Public Review):

This is a nice story about auricular chondrocyte maintenance, its molecular mechanism, and the role of the BMP 1 receptor in microtia disease conditions. A detailed analysis of different parts of the ear, their proliferation, and their differentiation condition with histological and immunofluorescence analysis strengthens the evidence. Further validation with patient sample RNA-Seq also helps the study end with an informative story.

From the public point of view, I want to say that the authors want to explain how auricular chondrocytes differ from growth plates or other chondrocytes. The authors show that Prxx1 is a good marker to differentiate auricular chondrocytes from different types of chondrocytes, which I doubt because other chondrocytes have an expression of Prxx1 at a lower level.

Another thing the authors mention is that microtia conditions develop through reduced size without affecting proliferation and apoptosis. The authors never provide any evidence about how the ablation of Bmpr1a affects the size, protein trafficking, and ECM organization.

Crosstalk between BMP-PKA in auricular chondrocytes and switching the chondrocytes' cell fate in osteoblast cells are not entirely stable by these studies for physiological functions.

eLife assessment

The study has added value to what we have already known in the potential pharmacological immunomodulatory therapies in LPS-induced sepsis, and especially the use of oral leucine might be of great interest to the readers engaged in this field. We believe this study is important and provides solid evidence on the potential use of leucine in sepsis.

Joint Public Review:

Summary:

The major purpose of this manuscript is to examine whether leucine treatment would be a potential strategy to treat cytokine storm syndrome (CSS). CSS is a common symptom in multiple infectious diseases in clinic, gradually leads to multiple organ failure and high mortality. Strategies to treat CSS including pulse steroid therapy normally leads to severe side effects. Therefore, it is still required to develop safe strategy with high efficacy to treat CSS. In clinic, sepsis is well characterized to exhibit CSS and therefore multiple studies utilized LPS-induced sepsis model to evaluate CSS symptom. In this study, the authors examined whether leucine, an essential amino acid that has been absorbed daily in our body, could ameliorate CSS symptom in the LPS-induced sepsis mouse model. They found a potential protective effect of leucine in terms of the survival rate and inflammatory responses.

Strengths:

The study is overall well designed and the results are well analyzed with only minor issues. The methods they utilized is appropriate.

Weaknesses:

The mechanistical insights are not sufficient and could not fully explain the phenotype they found. Considering the importance of this study is to identify the potential protective role of leucine in CSS, the authors could also consider investigator-initiated clinical trials to further expand the significance of this study.

Reviewer #1 (Public Review):

Force sensing and gating mechanisms of the mechanically activated ion channels is an area of broad interest in the field of mechanotransduction. These channels perform important biological functions by converting mechanical force into electrical signals. To understand their underlying physiological processes, it is important to determine gating mechanisms, especially those mediated by lipids. The authors in this manuscript describe a mechanism for mechanically induced activation of TREK-1 (TWIK-related K+ channel. They propose that force induced disruption of ganglioside (GM1) and cholesterol causes relocation of TREK-1 associated with phospholipase D2 (PLD2) to 4,5-bisphosphate (PIP2) clusters, where PLD2 catalytic activity produces phosphatidic acid that can activate the channel. To test their hypothesis, they use dSTORM to measure TREK-1 and PLD2 colocalization with either GM1 or PIP2. They find that shear stress decreases TREK-1/PLD2 colocalization with GM1 and relocates to cluster with PIP2. These movements are affected by TREK-1 C-terminal or PLD2 mutations suggesting that the interaction is important for channel re-location. The authors then draw a correlation to cholesterol suggesting that TREK-1 movement is cholesterol dependent. It is important to note that this is not the only method of channel activation and that one not involving PLD2 also exists. Overall, the authors conclude that force is sensed by ordered lipids and PLD2 associates with TREK-1 to selectively gate the channel.

The proposed mechanism is solid and the authors have revised the manuscript to address previous issues with the first version

Reviewer #2 (Public Review):

This manuscript by Petersen and colleagues investigates the mechanistic underpinnings of activation of the ion channel TREK-1 by mechanical inputs (fluid shear or membrane stretch) applied to cells. Using a combination of super-resolution microscopy, pair correlation analysis and electrophysiology, the authors convincingly show that the application of shear to a cell can lead to changes in the distribution of TREK-1 and the enzyme PhospholipaseD2 (PLD2), relative to lipid domains defined by either GM1 or PIP2. The activation of TREK-1 by mechanical stimuli was shown to be sensitized by the presence of PLD2, but not a catalytically inactive xPLD2 mutant. In addition, the activity of PLD2 is increased when the molecule is more associated with PIP2, rather than GM1 defined lipid domains. The presented data do not exclude direct mechanical activation of TREK-1, rather suggest a modulation of TREK-1 activity, increasing sensitivity to mechanical inputs, through an inherent mechanosensitivity of PLD2 activity. The authors additionally demonstrate that cellular uptake of cholesterol inhibits TREK-1 activation and, in ex vivo studies, that depletion of cholesterol from astrocytes reduces correlation of TREK-1 and G1 lipids in mouse brain slices. In vivo studies, using Drosophila melanogaster behavioural assays, were used to demonstrate that disrupting PLD2 altered behavioural responses to mechanical and electrical inputs. These data demonstrate that manipulation of PLD2 analogue in the fly can alter sensory transduction, suggesting that PLD functions to regulate sensitivity to mechanical force. However, as the authors note, there is no TREK-1 homologue in this organism: thus the identity of the downstream effectors of PLD in D. melanogaster remain unknown. This work will be of interest to the growing community of scientists investigating the myriad mechanisms that can tune mechanical sensitivity of cells, providing valuable insight into the role of functional PLD2 in sensitizing TREK-1 activation in response to mechanical inputs, in some cellular systems.

The authors convincingly demonstrate that, post application of shear, an alteration in the distribution of TREK-1 and mPLD2 (in HEK293T cells) from being correlated with GM1 defined domains (no shear) to increased correlation with PIP2 defined membrane domains (post shear). The association of TREK-1 with PIP2 required functional mPLD2. These data were generated using super-resolution microscopy to visualise, at sub diffraction resolution, the localisation of labelled protein, compared to labelled lipids. The use of super-resolution imaging enabled the authors to visualise changes in cluster association that would not have been achievable with diffraction limited microscopy.

This work provides further evidence of the astounding flexibility of mechanical sensing in cells. By outlining how mechanical activation of TREK-1 can be sensitised by mechanical regulation of PLD2 activity, the authors highlight a mechanism by which TREK-1 sensitivity could be regulated under distinct physiological conditions.

Reviewer #3 (Public Review):

The manuscript "Mechanical activation of TWIK-related potassium channel by nanoscopic movement and second messenger signaling" presents a new mechanism for the activation of TREK-1 channel. The mechanism suggests that TREK1 is activated by phosphatidic acids that are produced via a mechanosensitive motion of PLD2 to PIP2-enriched domains. Overall, I found the topic interesting but several typos and unclarities reduced the readability of the manuscript. Additionally, I have several major concerns on the interpretation of the results. Therefore, the proposed mechanism is not fully supported by the presented data. Lastly, the mechanism is based on several previous studies from the Hansen lab, however, the novelty of the current manuscript is not clearly stated. For example, in the 2nd result section, the authors stated, "fluid shear causes PLD2 to move from cholesterol dependent GM1 clusters to PIP2 clusters and this activated the enzyme". However, this is also presented as a new finding in section 3 "Mechanism of PLD2 activation by shear."<br /> In the revised manuscript, the authors addressed most of my concerns. I still have the following suggestions/confusions.<br /> 1. the reviewer would highly appreciate verification of the cholesterol assay, either by additional experiment or by citations of independent work.

2. The claim on "shear thinning" is still very confusing. First, asymmetric insertion of molecules to one monolayer of the membrane is a main mechanism for membrane bending and curvature formation. Second, why is "shear thinning" equivalent to entropy/order?

eLife assessment

In this important paper, Blin and colleagues develop a high-throughput behavioral assay to test spontaneous swimming and olfactory preference in individual Mexican cavefish larvae. The authors present compelling evidence that the surface and cave morphs of the fish show different olfactory preferences and odor sensitivities and that individual fish show substantial variability in their spontaneous activity that is relevant for olfactory behaviour. The paper will be of interest to neurobiologists working on the evolution of behaviour, olfaction, and the individuality of behaviour.

Reviewer #1 (Public Review):

Summary:<br /> The authors posed a research question about how an animal integrates sensory information to optimize its behavioral outputs and how this process evolved. Their data (behavioral output analysis with detailed categories in response to the different odors in different concentrations by comparing surface and cave populations and their hybrid) partially answer this tough question. They built a new low-disturbance system to answer the question. They also found that the personality of individual fish is a good predictor of behavioral outputs against odor response. They concluded that cavefish evolved to specialize their response to alanine and histidine while surface fish are more general responders, which was supported by their data.

Strengths:<br /> With their new system, the authors could generate clearer results without mechanical disturbances. The authors characterize multiple measurements to score the odor response behaviors, and also brought a new personality analysis. Their conclusion that cavefish evolved as a specialist to sense alanine and histidine among 6 tested amino acids was well supported by their data.

Weaknesses:<br /> The authors posed a big research question: How do animals evolve the processes of sensory integration to optimize their behavioral outputs? I personally feel that, to answer the questions about how sensory integration generates proper (evolved) behavior, the authors at least need to show the ecological relevance of their response. For the alanine/histidine preference in cavefish, they need data for the alanine and other amino acid concentrations in the local cave water and compare them with those of surface water.

Also, as for "personality matters", I read that personality explains a large variation in surface fish. Also, thigmotaxis or wall-following cavefish individuals are exceeded to respond well to odorants compared with circling and random swimming cavefish individuals. However, I failed to understand the authors' point about how much percentages of the odorant-response variations are explained (PVE) by personality. Association (= correlation) was good to show as the authors presented, but showing proper PVE or the effect size of personality to predict the behavioral outputs is important to conclude "personality is matter"; otherwise, the conclusion is not so supported.

From the above, I recommend the authors reconsider the title also their research questions well. At this moment, I feel that the authors' conclusions and their research questions are a little too exaggerated, with less supportive evidence.

Also, for the statistical method, Fisher's exact test is not appropriate for the compositional data (such as Figure 2B). The authors may quickly check it at https://en.wikipedia.org/wiki/Compositional_data or https://www.annualreviews.org/doi/pdf/10.1146/annurev-statistics-042720-124436.

The authors may want to use centered log transformation or other appropriate transformations (R-package could be: https://doi.org/10.1016/j.cageo.2006.11.017). According to changing the statistical tests, the authors' conclusion may not be supported.

Reviewer #2 (Public Review):

In their submitted manuscript, Blin et al. describe differences in the olfactory-driven behaviors of river-dwelling surface forms and cave-dwelling blind forms of the Mexican tetra, Astyanax mexicanus. They provide a dataset of unprecedented detail, that compares not only the behaviors of the two morphs but also that of a significant number of F2 hybrids, therefore also demonstrating that many of the differences observed between the two populations have a clear (and probably relatively simple) genetic underpinning.

To complete the monumental task of behaviorally testing 425 six-week-old Astyanax larvae, the authors created a setup that allows for the simultaneous behavioral monitoring of multiple larvae and the infusion of different odorants without introducing physical perturbations into the system, thus biasing the responses of cavefish that are particularly fine-tuned for this sensory modality. During the optimization of their protocol, the authors also found that for cave-dwelling forms one hour of habituation was insufficient and a full 24 hours were necessary to allow them to revert to their natural behavior. It is also noteworthy that this extremely large dataset can help us see that population averages of different morphs can mask quite significant variations in individual behaviors.

Testing with different amino-acids (applied as relevant food-related odorant cues) shows that cavefish are alanine- and histidine-specialists, while surface fish elicit the strongest behavioral responses to cysteine. It is interesting that the two forms also react differently after odor detection: while cave-dwelling fish decrease their locomotory activity, surface fish increase it. These differences are probably related to different foraging strategies used by the two populations, although, as the observations were made in the dark, it would be also interesting to see if surface fish elicit the same changes in light as well.

Further work will be needed to pinpoint the exact nature of the genetic changes that underlie the differences between the two forms. Such experimental work will also reveal how natural selection acted on existing behavioral variations already present in the SF population.

It will be equally interesting, however, to understand what lies behind the large individual variation of behaviors observed both in the case surface and cave populations. Are these differences purely genetic, or perhaps environmental cues also contribute to their development? Does stochasticity provided by the developmental process has also a role in this? Answering these questions will reveal if the evolvability of Astyanax behavior was an important factor in the repeated successful colonization of underground caves.