Reviewer #3 (Public Review): 

This manuscript presents a comprehensive study of CD4 memory T cell receptor beta repertoire response to hepatitis B vaccination, including repertoire correlates of early, late, and non seroconversion, identification of antigen specific and epitope specific clones, and a statistical classifier to potentially predict early Vs late seroconverters based on their pre-vaccination bulk repertoire. The major strengths are a unified experimental and computational analysis of bulk TCR repertoire data with antigen and epitope specific sorted T cells from the same individuals, allowing them to track personalized dynamics of vaccine specific clones, as well as translate across individuals to predict vaccine-induced seroconversion outcomes from pre-vaccination repertoires. The experimental data and reproducible analysis code are publicly accessible, and represent a useful resource that will likely be of interest beyond this study to other immune repertoire researchers. 

The results seem to support the authors conclusions, however several reported findings based on statistical analysis are less convincing, and would benefit from improved validation, clarification, or reworking. I next detail these aspects ordered by results sections. 

Section beginning line 128: 

The reported finding of this section is that early-converters (and not late-converters) undergo repertoire remodeling by day 60 post vaccination that decreases repertoire clonality. The evidence presented to support this is a computation of Shannon entropy for day 60 Vs day 0 in each individual, and a paired sample statistical test that is nominally significant for early and not for late converters. However, this nominally significant p value 0.042 is quite marginal, and the associated plots (Fig 2a) indicate only a very modest visual difference, and the presence of a distant outlier. The p value for late converters is not shown, however the marginally significant p value for early converters may not be nominally significant (at alpha 0.05) after multiple test correction (two tests). Additionally, the range of possible entropy values depends on the total sample size, so part of this difference may be driven by sample size. It may be more appropriate to use the Shannon equitability index (normalizing by the maximum possible entropy given the sample size, which is the log of the richness). 

Section beginning line 147: 

Ag specific T cells were isolated from day 60 samples and sequenced, allowing the authors to track the dynamics of these clones in the bulk repertoire data across time points. In all vaccinee groups these Ag specific clones are found to increase from day 0 to day 60 in the bulk repertoires. A marginal p value (0.04909) is presented to support early-converters showing more increase in these Ag specific clones. However, statistics comparing early to non or late to non converters are not mentioned (and these would require a multiple test correction on the p value that is discussed). 

A more general difficulty I have with this section is that the null hypothesis isn't made clear, and is probably more subtle and complicated than it appears. Cells are sorted from day 60, then their prevalence is compared between day 0 and day 60. Don't we expect to see more of them in day 60 even if there is no specific expansion for these clonotypes, but just random repertoire churn? My concern here is that double dipping from day 60 is affecting the analysis, since this time point is initially used to define the marker clones in the first place. If you take a random set of day 60 TCRs as null marker clones do you also see they are more prevalent in day 60 Vs day 0, or are you assuming that there should be no difference under the null? 

The last analysis in this section (presented in Fig 2d) does group-level comparisons of Ag specific clone fractions at day 60. I don't follow why normalizing by the number of Ag-specific clones detected in each individual is correct (i.e. would result in no differences under the null). Here again it could be helpful to see if null marker clone sets (the same size as the true Ag specific sets for each individual) indeed show no significant differences between groups. 

Section beginning line 185: 

In this section, a peptide pool approach is used to identify epitope specific TCRs from each individual at day 60, and a classifier is constructed to discriminate between early and late converter bulk repertoires, using a quantity R_hbs that measures the relative fraction of peptide specific TCRs in the repertoire according to Hamming distance similarity to the peptide specific TCRs. Importantly (as stated in the methods) a cross validation procedure is employed where TCRs from a given individual are not used for classification of that same individual. Since d is Hamming distance on CDR3 sequences, presumably comparisons are only made for TCRs with identical CDR3 length differences. This seems like a limitation, since clones with identical V and J gene, and CDR3 that differ by only one in CDR3 length could very well bind the same epitope. A more TCR-specific distance function, such as the TCRdist of Dash et al., may significantly increase classifier performance. 

There is a distance cutoff parameter c required to define R_hbs. How was this parameter chosen? In particular, if it was tuned to produce the best AUROC, then the cross validation procedure is not legitimate (nested cross validation would be needed, or separate held out test set).

Reviewer #1 (Public Review): 

This study examines recurrent neural network models trained on sequence-prediction tasks analogous to those used in human cognitive studies. The results demonstrate that such models lead to highly accurate predictions for challenging sequences in which the statistics are non-stationary and change at random times. This is a novel and remarkable result that opens up new avenues for cognitive modelling. 

Strengths:

- Trained artificial networks are probed using tasks and analysis tools identical to human cognition. 

- The results show that trained recurrent networks exhibit effective learning rates that adapt to sudden shifts in statistics, in a manner similar to optimal Bayesian agents. 

- Thorough analyses demonstrate that the hidden states of the networks encode emergent latent variables such as the precision of the prediction, or current "context".

- Very clear writing style. 

Weaknesses: 

- The manuscript insists on the fact that gating between neural units is a necessary component. A more conservative conclusion is that gating facilitates training. The analysis in the manuscript cannot exclude the possibility that networks without explicit gating could reach an optimal performance with a different training protocol. 

- The text insists on the fact that a very small number of neural units is sufficient (11 in most figures). It is not clear why this is a relevant limit when comparing with biological networks. 

Despite these weaknesses, the results strongly support the conclusions. Most importantly, this study opens up the possibility of using a new class of models for cognitive modelling.

Reviewer #2 (Public Review): 

This manuscript examines the suitability of a specific class of artificial neural network models with 'gated recurrent units (GRUs) for solving the general problem of prediction in a stochastic volatile environment. The authors frame the problem in the context of predicting the next sample in a sequence of 0's and 1's generated from increasingly more complex generative models. They compare the performance of the GRU network to that of an optimal observer as well as various heuristic solutions and reduced variants of the GRU. Results indicate that full GRU networks are closer to optimal than heuristic models and that all their key computational building blocks (gating, lateral connections, and recurrent weight tuning) contribute to their near-optimal performance. 

There has been a longstanding interest in developing normative models of how humans handle latent information in stochastic and volatile environments. The paper is important and timely as it leverages recent advances in machine learning to tackle this question using a different approach that involves task-optimized neural network models. This approach has proven quite fruitful in several behavioral domains including perception, motor behavior, timing, and simple decision-making. The paper builds on this growing body of work to generate hypotheses about the computational building blocks that may underlie quasi-optimal behavior in the presence of stochasticity and volatility. 

Strengths:

The paper is very well written and accessible despite being computationally quite sophisticated. 

The approach is systematic. It uses a well-defined task (binary sequence prediction) that can be adapted to increasingly more complex latent structures without changing the observables. 

The analyses are comprehensive. The behavior of the network models is compared to various optimal and suboptimal observer models as well as reduced versions of the network models. This overall approach is maintained throughout the paper so that one can appreciate the key takeaway points. 

The paper aims to offer a deeper understanding and not just an engineering solution. Many papers of this kind do not take the effort to 'open' the trained networks and provide an algorithmic understanding. This paper does. More impressively, the paper goes beyond simple correlative measures of the network states; it uses an innovative perturbation technique to verify that the inferred latent representations in the network are indeed functional. I really liked this perturbation analysis. It is highly valuable and broadly applicable. 

By comparing the full GRU with its reduced versions, the paper makes a serious attempt to guard against criticisms that the success of the full GRU is due to the degrees of freedom it offers. 

Weaknesses:

The most notable weakness of the paper is that is not clear whether its aim is to develop a neural model that is close to optimal or a neural model that explains how biological brains handle stochasticity and volatility. There is no serious and quantitative comparison to behavior or neural data recorded in humans or animal models. All the comparisons are with other algorithms and reduced GRU networks. One can appreciate these comparisons if the goal is to show that a full GRU network is close to optimal (which as they show, in many cases, it is). But do humans exhibit a similar level of optimality? What I was hoping to see was some sort of analysis that would show that the types of errors the model makes are in some counterintuitive (or even intuitive) way like the types of errors humans make. In some of the papers where certain heuristics were proposed, the entire goal was to explain characteristic sub-optimalities in human behavior. Without such comparisons, I think the results might most effectively drive progress in purely computational circles.

Reviewer #1 (Public Review): 

In the present work the authors have defined several objectives. First, to define an appropriate benchmarking standard for drug repositioning based on signature reversion in silico studies Second, apply the method previously defined to determine the best practice approach for LINCS data-based computational drug repositioning. Last, use the methods defined in the second objective to identify novel drugs that can be used for liver cancer treatment. As results of this approaches the authors defined two independent benchmarking standards (AUC-based and KS statistic-based) that shown a good agreement of the evaluation results. Then, since the use of reference signatures showed a high degree of cell-type specific effect the authors demonstrated that XSum is the optimal method for matching compound and disease signatures. Additionally, the authors also investigated the optimal clinical features that should recapitulate the query signature and demonstrate that a good signature should possess the ability to comprehensively recapitulate the clinical features of corresponding disease, rather than only reflect the disease characteristic from single perspective such as normal vs diseased states or prognosis. Finally, using the methods developed the above findings together with in vitro assays the authors determined that homoharringtonine could be not only a promising anti liver cancer agent but also a treatment for the underlying liver chronic disease. All in all, the data provided a good pipeline for the drug repositioning using a in silico approach, and the authors have discussed most of the strengths and weaknesses of their work. Therefore, the conclusions of this paper are mostly well supported, however some aspects would need to be clarified and validated.

Reviewer #2 (Public Review): 

In this paper, the authors aimed at developing an optimal strategy for LINCS data-based therapeutic discovery. They thoroughly analyzed available drug pertubation expression data and also suggested a new therapeutic for HCC. This paper is of potential interest to a broad audience in cancer biology, cancer drug development and associated research.The conclusions of this paper are well supported by the data presented. The paper presents new benchmarking standards regarding methodology and parameters for quantitatively estimating drug retrieval performance. Furthermore, new therapeutics for liver cancer were suggested.

Joint Public Review: 

The proteasome governs the fate of hundreds of proteins in the cell and is the most prominent destination for ubiquitin-protein conjugates. One of the persistently enigmatic aspects of the proteasome function is how it is coupled to deubiquitination of the substrate. Uch37 is one of three deubiquitinating enzymes, each very different, that effect these steps in substrate processing. Undoubtedly, the specificity of Uch37 for branched ubiquitin chains is a significant issue in the field. But it is a challenging problem technically and linking the debranching activity to specific biology is at an early stage at best. This study explores these issues at a high technical level. There are considerable structural data on Uch37's interaction with ubiquitin, but they do not provide a concrete model for how the branch is recognized. It is an intuitively reasonable model that, to resolve a branch, two ubiquitins must be recognized, and that the "first" recognition site is simply the active site of Uch37. Through exemplary experiments involving among other things NMR, enzyme kinetics, microscale thermophoresis, and tailor-made mutations in specific ubiquitin groups within the three-part branched structure, the authors construct a convincing argument that specific hydrophobic patches within two of the three ubiquitin are recognized by Uch37, the two ubiquitin groups being the substrate-distal groups. Uch37 functions as part of a complex with Rpn13, a ubiquitin receptor, but it is excluded that the ubiquitin recognition site in Rpn13 provides the "second" site. Interestingly, another domain in Rpn13 does contribute to the specificity of Uch37, by helping to position an active site loop within Uch37 so as to suppress hydrolysis of some non-branched chains. 

The latter half of the paper consists mainly of studies in cells that explore the biological significance of debranching. When ubiquitin chain levels were examined (Fig 6A), the strengths of the effects of various mutants varied in arguably surprising ways, as did the relative strengths of the effects when assessed by different anti-ubiquitin antibodies (where the exact same samples were being examined). Thus, the EWI mutant had a modest effect with FK2 antibodies, much less than C88A, but was spot on with C88A with the K11/K48 anti-branch antibody. Should this be interpreted as the K11/K48 antibody reporting on deubiquitination of true Uch37 substrates whereas the FK2 antibody is reporting as well (perhaps mainly) on some more indirect perturbation of proteasome function (where ubiquitin conjugates may be unproductively retained at Uch37, not only not deubiquitinated). It would also help to show the extent of the enzymatic defect in the EWI mutant using the standard in vitro assays in the paper. 

This paper is technically excellent and is a significant contribution to our understanding of deubiquitination, focusing on a very interesting member of this enzyme family. The manuscript follows up on a previous publication that UCH37 catalyzes deubiquitination of branched ubiquitin chains, cleaving the K48 linkage. The authors find preference for K6/K48 branches over K11/K48 and K48/K63 branches and use NMR to identify interactions with the hydrophobic patches of both distal ubiquitins. Amino acid substitutions demonstrate the importance of the K6 distal Ub hydrophobic patch and not the proximal Ub for debranching of the chain. UCH37 is further found to localize to proteasome and K48 ubiquitin containing stress granules, which increase in abundance and K11/K48 ubiquitin chains when UCH37 is knocked out of cells. Altogether this manuscript provides new advances in characterizing UCH37 activity in cells. 

This manuscript reports a careful mechanistic study of the deubiquitylating enzyme UCH37 that was shown to be involved in hydrolyzing K48-linked branched ubiquitin structures. Using a clever combination of chemical biology techniques, the authors synthesized homogenous branched ubiquitin substrates for assays with UCH37. This led to the discovery that K11/K48 linkage is turned over the fastest and that UCH37 binds both ubiquitin proteins attached to the central ubiquitin in K6/K48 linked triubiquitin. Furthermore, the authors demonstrated that mutation of UCH37 leads to the formation of proteasomal foci due to the accumulation of polyubiquitylated species. Thus, the authors have associated the debranching function of UCH37 with efficient protein turnover by 26S proteasomes. As such, the manuscript is an excellent addition to our knowledge of UCH37 from a previous study by Strieter and colleagues in Mol. Cell 2020 that also demonstrated that debranching activity of UCH37 was associated with efficient proteasomal turnover of ubiquitylated proteins. The novel aspect of the current study lies in the demonstration that UCH-37 inactivation leads to enhanced proteasomal foci formation, and these foci are also enriched in the shuttle protein RAD23B. Thus, the authors provide a comprehensive picture of UCH37 function that confirms the Molecular Cell study and extends it to specific binding modes and cellular phenomena such as foci formation. 

Overall, this is an excellent story with well thought out experiments. The diversity of chemistry applied was marvellous.

Reviewer #1 (Public Review): 

The premise for this manuscript is that vitamin D is important for optimal fetal development. The data for this is correlative and there are human and murine mutations in the vitamin D receptor that do not show important abnormalities in the affected fetuses. However, this does not diminish the important findings of active transport of vitamin D metabolites by the placenta and of significant regulation of target genes in the placenta itself. In the system used, most of the 25D is metabolized by the placenta and only ~10% is transferred to the fetal circulation, which raises questions as to how well this model recapitulates the in vivo dynamics. Of interest, most of the active and inactive metabolites end up in the maternal circulation. This raises the question as to whether there is impaired transport of vitamin D and its metabolites into the fetal circulation in this experimental system. It would have been useful to use a control substance that is readily transported into the fetal circulation to demonstrate that the placental unit is intact. Nevertheless, these studies demonstrate novel findings.

Reviewer #2 (Public Review): 

Ashley et al., studied the transfer and metabolism of 25(OH)D3 in maternal and fetal circulations; thereafter utilizing the term human placenta, now considered the "gold standard" among currently available translocation models. Moreover, the authors investigated the effects of vitamin D on placental methylation, transcriptomic, and the proteomic landscape. They demonstrated the active mechanisms of 25(OH)D3 uptake by the placenta and found that placental metabolism of 25(OH)D3 can modify fetal and maternal levels of Vitamin D. Furthermore, the authors have shown that 25(OH) D3 induces placental specific effects on the transcriptome and proteome involving pathways relevant to placental function and fetal development. Moreover, these effects are dependent on the underlying epigenetic landscape. 

This is an important finding because it provides further understanding as to the active role of human placenta in Vitamin D transport and metabolism, with further influence on fetal growth and development. 

Though it is a well-reported descriptive study, transcriptomic and proteomic data require further validation.

Reviewer #2 (Public Review): 

Recent work has provided insight into the neural mechanisms that limit the capacity of working memory. Recordings from non-human primates and functional imaging in humans, has found that increasing the number of items held in working memory reduces information about any individual item and shown that this reduction in information is due to divisive normalization of neural responses. The current manuscript builds on this work, showing that the same mechanisms are seen in crows. In brief, crows were trained on a change-detection task that required them to remember 2-5 objects over a short delay. Extracellular recordings from nidopallium caudolaterale (a prefrontal-like area) revealed neurons that were selective for the identity of the objects during visual presentation and during the memory delay. As with monkeys, the total amount of information about the identity of an object decreased as the animal was required to remember more objects. Finally, the authors provide evidence that this reduction in information is due to a divisive normalization like regularization. 

Overall, this is an interesting manuscript that extends the field in an exciting direction. This is a unique dataset and, by showing that similar mechanisms constrain the working memory capacity of a non-mammalian species, the results support the idea that there is a normative reason for these constraints that is constant across species. My largest concerns are around whether the authors convincingly demonstrate that divisive normalization can explain the changes in neural selectivity with increasing memory loads. In particular, the authors report a subset of neurons increase the amount of information they carry about a stimulus/memory as the number of items in working memory increases. It is not clear how this can be explained with a simple divisive normalization model.

Reviewer #1 (Public Review): 

Sato et al present a very well written manuscript focused on exploring the mechanism underlying brain aging, specifically related to excess iron. While increased iron in the aging brain has previously been demonstrated in both animal models and humans, until the current work, no evidence of mechanism has been found. The current work presents a compelling story of the important finding that localized hepcidin expression leads to iron sequestration by downregulating FPN1. However, the authors do not reconcile how this work can be explained in the context of Lu et al 2017 where no difference in FPN1 expression was observed. Furthermore, recent data on this subject warrants discussion. In addition, a more robust evaluation of TFR2 expression is unregulated in the aging brain and why / whether relatively ineffective erythropoiesis (also associated with aging) may be driving iron loading in the liver, muscle, and brain. Finally, it is not clear whether hepcidin may be able to traverse the BBB at least unidirectionally. Some additional questions and clarifications are delineated below. 

1) Please add to the discussion how to reconcile the current results with those of Lu et al 2017 mentioned in the introduction as showing no change in brain FPN1 expression. 

2) Can the authors present data on erythropoiesis-related parameters in the young vs aged mice? Specifically, if the data is showing that ALAD and Hox1 are increased, these may be related to anemia in this model. As a consequence of an already known association between aging and anemia of unknown significance, an anemia associated with ineffective erythropoiesis, it is conceivable that ERFE is increased and leads to relatively suppressed hepcidin despite iron loading in the liver. Can the authors present data on erythropoiesis in the aged vs young mice, along with spleen size, and bone marrow ERFE expression? 

3) Where in the brain is hepcidin produced? What specific cells express FPN1? These are important questions to reconcile the data presented with prior work in the field. 

4) The authors have too quickly discounted a role for TFR2 in the brain. Several manuscripts exist in both mouse models (Pellegrino Sci Rep 2016) and GBM in humans (Voth J Clin Neurosci 2015) to promote a reason to explore this further. Worthy of further discussion and exploration. 

5) Please edit the discussion to remove general knowledge about iron regulation and focus specifically on brain iron. Several manuscripts have been published recently from the group in Case Western on this subject (Chaudhary J Alz Dis 2021) that warrant discussion.

Reviewer #2 (Public Review): 

Levels of non-heme iron are shown to be elevated in both the mitochondria and the cytoplasm of old mice relative to young mice. Among iron-related genes, hepcidin expression is notably upregulated in aged brain. Hepcidin triggers ubiquitination and degradation of ferroportin, the only known cellular iron exporter. Accordingly, it is shown that ferroportin protein levels are decreased and ubiquitinated ferroportin in increased in aged brains. The study is interesting, but currently of limited impact. 

Major limitations: 

1) Analysis is done only at the whole brain level. The is no cell-level analysis, so it's unclear which cell types might be producing hepcidin, degrading their ferroportin, or accumulating iron. 

2) No mechanistic information explaining hepcidin upregulation is provided. 

Additional limitations: 

1) Mice: it is unclear why the UM-HET3 mouse strain was chosen, and why only female mice were studied. Were the females nulliparous? What form of iron was in the diet, and how much? 

2) It is not stated which part of the brain was dissected for the studies 

3) There is no validation of the purity/enrichment in the mitochondrial fractionation 

4) Aconitase activity is the only measure of oxidative stress employed. Additional markers should be tested for corroboration.

Reviewer #3 (Public Review): 

The manuscript by Sato et al. addresses an important question related to brain iron accumulation with aging. This is a timely question because iron accumulates in the brains of aged individuals and is believed to contribute to neurodegenerative diseases such as AD and PD. 

Major strengths of the study are the question itself, a combination of techniques to measure brain iron, whether upregulation of local hepcidin is the main cause of iron accumulation, and if upregulation of hepcidin gene is responsible for the increase in protein expression. The authors also evaluate the expression of other iron modulating proteins, including Fpn, the downstream effector hepcidin. 

Weaknesses include lack of robust data to support the author's claims. In many instances, the results do not support the conclusions drawn by the authors.

Reviewer #1 (Public Review): 

The authors propose a mechanism for error-correction in enzyme-free templated copying, one that could have plausibly operated in prebiotic processes. In contrast to an energy-consuming enzymatic process that corrects errors as they occur, as in kinetic proofreading, here a non-equilibrium environment preferentially removes erroneous copies. The two key ingredients are i. that errors slow down polymerisation, and ii. that the environment periodically and selectively `leaks' out smaller fragments, such that in a finite time the only copies produced (and retained) are those with few errors. 

As the authors show, these conditions are met by non-enzymatic copying of DNA and RNA in fluctuating environments that, they argue, are plausible prebiotic niches. By a thorough quantitative analysis, the authors establish bounds on the environmental timescales that allow for faithful copying of nucleic acid polymers of different lengths. 

The paper is clearly written and balances an intuitive description of the proposed kinetic filtering mechanism with the experimentally-grounded model of enzyme-free nucleic acid replication. The numerical and analytical results are thorough and lend credence to the plausibility of the mechanism. 

While there has been some recent work on kinetic aspects of proofreading (for instance, Sartori and Pigolotti, PRL 2013 and J Stat Phys 2016), the proposed mechanism is, as far as I can tell, truly distinct. Nonetheless, while the authors focus on its implementation in enzyme-free nucleic acid copying in a prebiotic context, it seems sufficiently generalisable that it could potentially be at play in other situations. 

Overall, the strengths of this paper are a. a demonstration of how non-equilibrium environments can circumvent the error problem in non-enzymatic copying, and b. a novel mechanism of error-correction that is conceptually distinct from kinetic proofreading.

Reviewer #2 (Public Review): 

The paper proposes a mechanism that reduces the replication errors of prebiotic polymerization. The paper assumes a PCR-like situation, that is, template strands, primers, thermal cycles, and copying machines are ready. The authors demonstrated by simulation and theory that a simple kinetic mechanism reduces replication errors while keeping the yield reasonable. The paper uses the following experimental facts and the quantities obtained by experiments: 

- The misincorporation of wrong nucleotides slows down the extension. 

- The misincorporation triggers successive incorporation of errors, generating error clusters, which result in a significant stalling of the extension. 

Hence, the errors can be kinetically discriminated by limiting the polymerization time. They found a time scale where errors are significantly suppressed while a reasonable yield is obtained. Based on this observation, they tried to filter errors out by repeating temperature cycles with tuned cycle duration. A high temperature resets the extension by dissociating the perfect and partial products from the templates. They found that there is a time-scale range where the replication does not suffer the fidelity-speed trade-off. 

Strengths: 

- The replication with fidelity is indispensable for stable autocatalytic systems to emerge in the prebiotic environment. However, the replication is erroneous without sophisticated error-suppression machinery used in modern biology. The hypercycle proposed by Eigen and Schuster in the 1970s is one of the mechanisms to suppress errors without such modern machinery. However, the hypercycle requires the formation of a complex autocatalytic network and is not expected to emerge spontaneously. As well, the hypercycle is not very stable due to multiple problems. The kinetic error filtering proposed in this paper provides a promising mechanism to reduce errors because it does not require complex systems. In addition, the simulations are based on the parameters obtained by experiments. 

Weaknesses: 

- The maximum length that can be copied without errors is 50nt for DNA and 25nt for RNA. It is not clear if sequences with such a short length can have sufficient polymerization activity. We would need more mechanisms to solve the information-keeping problems. However, the kinetic error filtering is simple and could be combined with other mechanisms. Or, some nucleotide analogs may have better parameters generating longer strands. It would be true that the kinetic filtering lowers the hurdle for the emergence of autocatalytic systems in the prebiotic soup. 

- Other things to be considered include the effect of the product-template rebinding. The template concentration is assumed to be very small so that the product does not rebind to the template. However, it would be necessary to consider the effect in more realistic situations. The product-template rebinding is one of the significant issues in the prebiotic autocatalytic system. In the present scenario, if the products with incomplete copying bind to the template, they quickly finish copying during the limited cycle duration. This may weaken the kinetic filtering.

Reviewer #3 (Public Review): 

This manuscript proposes a kinetic error-correction mechanism that does not require enzymes. The authors argue that this mechanism could have played a relevant role in prebiotic environments. The enzyme-free kinetic mechanism proposed in this manuscript is conceptually different from kinetic proofreading, which is a non-equilibrium error-correction mechanism characterizing accurate polymerization assisted by enzymes. 

Strengths:

The results presented in this manuscript are convincing. The authors make a case that the kinetic error filtering mechanism is a plausible scenario for the emergence of low error rates in prebiotic copying of DNA and RNA. I feel that the manuscript is an interesting, timely, and important contribution to the debate on the origin of life and is likely to stimulate experimental efforts to test the proposed hypothesis. 

Weaknesses:

I feel that the results are compelling, but the modeling choice and the comparison with previous models in the literature should be clarified. I feel that this clarification is important both to understand the relevance of modeling details and for the impact of this paper on the theoretical research on error correction in biology.

Reviewer #1 (Public Review): 

The manuscript prepared by Alexandre P. Thiery et al. is focused on enamel knot (EK) formation in the shark teeth. The authors brought here evidence of the existence of an enamel knot-like structure in this non-mammalian species showing its role on regulation of the tooth shape and cusps formation similarly to the mammalian EK. Shark tooth development has been previously used as an alternative model for odontogenesis by authors and there is no doubt that it offers valuable tools for research on the tooth regeneration through the formation of successional generations of teeth as well as on serial organ formation (activation and inhibition processes during tooth development). In silico modelling is properly used by authors in this present work and it serves here as a valuable tool documenting the experimental results and confirming formulated conclusions. 

This manuscript brings very interesting results showing that EK formation is highly conserved in the animal species, and it is not a structure unique only for the mammalian tooth development. The authors mapped the expressions of mammalian EK markers in details in the shark teeth to compare the expressions of genes in both mammalian and non-mammalian signalling centres. Interestingly they documented the expression of Fgf10 gene expressed in several epithelial cells at the tip of the dental germ in the catshark. Since Fgf10 is not expressed in EK in the mouse but it has been shown to be expressed in opossum, this could document the phylogenetic adaptation of EK on the way to the mammalian dentition. The manuscript also showed similarities of the enamel knot-like structure morphology with the mammalian EK (eg. the presence of non-proliferative cells detected by PCNA in this structure). In contrast to mammalian EK no apoptosis has been detected in the enamel-knot-like structure in the shark, what is however not preventing the formation of tooth cusps at all. 

The authors furtherly focused on the important role of Wnt signalling during tooth cusps formation. Using experimental manipulation of the canonical Wnt pathway by small molecule activators and inhibitors they documented the role of Wnt signalling during tooth-crown morphogenesis, concretely in the cusp number determination. Upregulated Wnt signalling caused a higher number of cusps in the shark teeth and downregulation in contrast lead to the lower number of cusps or to unicusped teeth formation. The experiments were modelled using in silico model of tooth development (ToothMaker). The biomodelling confirmed the data obtained using manipulation experiments. 

Based on these approaches the authors concluded that the enamel knot-like structure in the shark shows a role on regulation of the tooth shape and cusps formation similarly to the mammalian EK. According to obtained results the authors have stated that phylogenetically EK seems to serve as a general organizer of the shape (morphology) maintaining its function throughout the species evolution. They also suggest here an introduction of the term "apical enamel knot" as a more general signalling centre present not only in teeth. 

The present manuscript does not have any substantial weaknesses. However, few points need to be addressed and discussed furtherly. 

The authors claimed a difference between early and new generations of teeth in the catshark during development. The early generation develops more superficially in contrast to deeper ingrowing germs of additional generation of teeth. Interestingly, in the lower mouse incisor region it has been documented that the first and initial tooth placode at early stages of the development with its own signalling centre is also formed more superficially in contrast to the subsequently appearing pEK located deeper in the epithelium of the incisor germ. The authors should compare the expressions of especially epithelial markers shown to play a role in the first row of teeth with later appearing teeth in the catshark and discuss this comparison with respect to the knowledge of these two signalling centres with different roles in the mouse. 

The authors used the computational in silico model, where the regulation of the inhibitor and activator of the tooth stems from its EK signalling centre. The authors claim in the manuscript that the tooth size and number of cusps is increasing in the new generations of dentition in the catshark, what they relate to the increasing initial site of the tooth formation and what also corresponded with the stimulated model situation. This could be probably connected to the growth of the jaws which gives more space between single signalling centres. Following this space expansion, the changes of the activation and inhibition (of activators and inhibitors levels defunding between neighbouring teeth in the tooth-row as well as between the tooth generations) could be the cause of this change in size and shape. This should be discussed more with respect to what is known in the mouse. 

The authors documented that Shh, which is a key marker of EK in mammals including mice, is expressed within the apical dental epithelium in the catshark. However, Shh expression was downregulated between cusps within the inter-cusp dental epithelium. According to the Figures E, F the Shh expression seems to be region specific, and it changes antero-posteriorly. It could reflect the antero-posterior appearance of the tooth germs. The posterior ones would be less advanced and thus Shh expression in the anterior part of the tooth-row would be upregulated in fact. This should be discussed.

Reviewer #2 (Public Review): 

The "enamel knot signaling center" is an organizer of tooth morphogenesis first characterized in mammalian teeth. The so-called "primary enamel knot" corresponds to non-cycling epithelial cells sitting at the tip of the first forming cusp. "Secondary enamel knots" are found at the tip of the other cusps in multicuspidated teeth. There is a debate in the community whether homologous structures exist or not in non-mammalian vertebrates. The authors looked for an "enamel knot signaling center" in the developing teeth of catshark. The bauplan of catshark teeth resemble some mammalian teeth, like seal teeth: a primary, more or less central and tall cusp, is arranged on a line with smaller secondary cusps (3-cusps teeth), and sometimes with even smaller tertiary cusps (5-cusps teeth). 

The authors first report the expression pattern in catshark developing teeth for a selection of signaling genes. They emphasize Fgf3 and Fgf10, which they found expressed in the epithelial cells sitting at the tip of the primary and secondary cusps. This reminds the enamel knot expression of Fgf3 in mouse and opossum and Fgf10 in opossum only. BrdU staining showed that this cusp-tip expression coincides with proliferation arrest of epithelial cells from cusp tip down into the valleys, as seen in mammals. Because Wnt signaling is important for enamel knot formation in mammals, they perturbed this pathway with pharmacological treatments. Inhibiting the pathway reduced tooth size and cusp number whereas activating it increased tooth size and cusp number. The authors also played with a model built for seal teeth, in which activation-inhibition mechanisms rule the formation of enamel knots, and morphogenesis and differentiation proceeds tip-down starting from these enamel knot. Catshark teeth have a similar organization like seal teeth, as mentioned above, and a similar development: tip-down, with the primary cusp developing first and the secondary cusp developing in a second time. The model built for seal teeth therefore not surprisingly can form catshark-looking teeth, and playing with activation-inhibition parameters superficially recapitulates the increase or decrease in cusp number seen in experimental perturbations as well as the variability in tooth shape seen at different positions and/or lifetime in catshark. Shark-specific aspects of perturbed teeth are however not recapitulated by the model. The authors conclude that catshark teeth possess a signaling center homologous to the enamel knots in mammals and based on gene expression similarities at the tip of developing dermal denticles (notably of Fgf3), propose a deep homology with an apical signaling center preceding tooth evolution. 

Strengths:

- The authors combined experiments: gene expression pattern for many genes, BrdU incorporation proliferation pattern, functional perturbation of a key pathway, some are tricky in such non-model species. 

- The experiments combining both over-activation and reduction of Wnt signaling pathway clearly demonstrate a role in promoting tooth growth and cusp formation. 

- The effect of treatments was quantified with 2D-outline analysis and statistics. 

- Using a mammalian model of tooth development is interesting, but it has to be dissected and challenged in its principles to make sure its use is correct and not based on superficial similarities. 

Weaknesses:

- Clearly, there is distinctive gene expression at the tip (even though it tends to then progress tip-down for several markers, rather than staying expressed only at the tip), proliferation arrest goes tip-down, and normal Wnt activity is necessary to reach proper cusp number and tooth morphology. But there is no functional demonstration that signaling from the tip controls cusp number, tooth morphology nor tip-down proliferation arrest. As a consequence, the data support the tip-down development of shark teeth with some critical involvement of the Wnt pathway more than the development of shark teeth from an apical signaling center. 

- The model only very superficially recapitulates the experimental perturbation: reducing Wnt activity produces a very short unicuspid tooth, much shorter than the wild type primary tooth; the model produces a single large cusp, larger than in the wild type. Increasing Wnt activity produces a tiny accessory cusp, as predicted by the model but also large cusps, that do not fit at all (figure F). It is thus very unclear if the principles of the model hold true for shark, beyond the general principle of sequential and -down development. Moreover it is also not sufficient to state that the Wnt pathway plays the role of the activator/inhibitor, and this role only. Overall, too many shortcuts are taken, from a superficial resemblance of perturbed teeth to the Wnt pathway as "the activator", and from Wnt signaling as the activator, to a Wnt-centered view of tooth shape variation across tooth position and developmental time. 

- Before looking for homology, it is necessary to define precisely the criteria that should be met, and especially those that can make the difference between homology and homoplasy. Such clear definitions are lacking. Mammalian teeth and shark teeth both develop on a tip-down principle: is it surprising that proliferation arrest goes tip-down, or would this be seen in any structure developing tip-down, including by pure homoplasy instead of shared ancestrality? What would be the minimal specifications to be met by an homologous apical signaling center (in terms of signaling, proliferation state, effect on tissue around...)? What similarities would be unsufficient? This is not clearly stated. 

- When comparing developing organs in different species, it is common to find a mosaic of conserved and divergent patterns. It is therefore important to decide in advance, what are the most relevant genes, and what would be the minimal requirement (quantitative and/or qualitative) to decide that a given, well-defined, developmental process is homologous. 

- Apical signaling centers organizing morphogenesis are found not only in dermal denticles, but are also found in many epithelial organs. It is therefore necessary to identify precise criteria to distinguish deep homology of dermal denticles and teeth from deep homology of developmental principles. 

In conclusion, the authors provide a nice piece of data, including functional experiments not easily done in a non-standard model like catshark. These data suggest that teeth are developing tip-down and that signaling molecules are found at the tip of the primary and secondary cusps. However, it is not always clear how far these molecules are specifically emitted at the tip or are simply expressed by a developmentally more advanced tissue (e.g. with expression later progressing tip-down as differentiation proceeds). Moreover the functional experiments do not directly demonstrate that this putative signaling center, or the mechanisms patterning this signaling center, is/are responsible for cusp patterning and tooth morphogenesis. The link is done through the use of a mammalian model of tooth morphogenesis, which only superficially recapitulates perturbed teeth morphology. This strongly bias interpretations towards homology, and alternative scenarios are not explored. A more precise theoretical and experimental framework would be needed to support the conclusion and reach a broad public.

Reviewer #3 (Public Review): 

Perhaps because teeth are so widely distributed among the vertebrates, there has been considerable interest in the evolutionary origin of teeth. A related question is the regulatory logic, or developmental basis of tooth development. It is well established that tooth development from fish to mammals uses largely the same set of developmental genes. Similarly, early stages of development are quite similar among all the epithelial organs (e.g., teeth hair, glands, scales). In this context the novelty of the new work is in asking whether even the regulatory logic of teeth is a shared, evolutionary ancient feature. Thiery et al. address this question by studying multicusped teeth of sharks. Multicusped, complex tooth morphologies are prevalent in mammals, and their development is thought to require precise regulation by epithelial signaling centers, called the enamel knots. Previously enamel knots have been thought to be restricted to mammals, but Thiery et al. argue that also developing shark teeth have them. Their evidence to support the presence of enamel knots in sharks are 1) gene expression patterns of several genes known to be dynamically expressed in the mammalian enamel knots and teeth, 2) experimental modulation of Wnt-signaling in developing sharks to see the phenotypic changes in tooth shape, and 3) computational modeling to test whether shark tooth development obeys similar regulatory logic as mammalian teeth. My overall inference from the evidence provided is that the basic thesis is quite convincing, and definitely worth publishing. There are, however, quite a few in individual components in the work that require clarification, and some statements are simply over the top.

Reviewer #1 (Public Review): 

Investigators assumed that our mood fluctuations could be predictive of risky versus non-risky choices we make. Hence, they aimed to understand the neural basis for "mood" fluctuations that could explain why there is behavioral variability. The main tenet of the authors' current work is that behavioral variation can be explained by changes in neural signals in different regions of the human brain. Here, they report findings of fluctuations in two areas of the brain that correlate with mood fluctuations in opposite directions. 

A number of methodological issues need to be resolved: 

The documented findings may be explained by the artifact of task design and the way the signals were calculated: The vmPFC was the only ROI for which a positive correlation was found between BGA and mood rating and TML. Instead, most other regions showed negative correlation (inlc da-Insula, dorsolateral prefrontal cortex, the visual cortex, the motor cortex, the dorsomedial premotor cortex, the ventral somatosensory cortex, and the ventral inferior parietal lobule). This can be purely an artifact of task itself: In 25% of mood rating trials, subjects were presented with a question. They had to move the cursor from left (very bad) to the right (very good) along a continuous visual analog scale (100 steps) with left and right-hand response buttons. They even got a warning if they were slow. 

In 75% of trials, subjects saw none of this and the screen was just blank and the subjects rested. First of all it is unclear if the 25% and 75% trials were mixed. I am assuming that they were not mixed as that could represent a fundamental mistake. The manuscript gives me the impression that this was not done (please clarify). Assuming that they were not mixed and we are seeing the data from 75% of trials only. These trials would trigger increased BGA activity in the default mode areas such as the vmPFC, and opposite patterns in the salience, visual and motor areas. Hence the opposite correlations. 

The authors should just plot BGA activity across regions during rest trials and see if this was the case. That would provide a whole different interpretation. 

In addition, it is entirely unclear how the BGA in a given electrode was plotted. How is BGA normalized for each electrode? What is baseline here? Without understanding what baseline was used for this normalization, it is hard to follow the next section about the impact of the intracerebral activity on decision-making. 

Line 237: how was the correction for multiple comparisons done? Subject by subject, ROI by ROI, electrode by electrode? Please clarify.

Reviewer #2 (Public Review): 

This study used intracranial EEG to explore links between broad-band gamma oscillations and mood, and their impact on decisions. The topic is interesting and important. A major strength is the use of intracranial EEG (iEEG) techniques, which allowed the authors to obtain electrical signals directly from deep brain areas involved in decision making. With its precise temporal resolution, iEEG allowed the authors to study activity in specific frequency bands. While the results are potentially interesting, one major concern with the analysis procedure-specifically grouping of all data across all subjects and performing statistics across electrodes instead of across subjects-reduces enthusiasm for these findings. There is also a question about how mood impacts attentional state, which has already been shown to impact baseline (pre-stimulus) broad band gamma. 

Major comments: 

The number of subjects with contacts in vmPFC, daIns, and both vmPFC and daIns should be stated in the manuscript so the reader doesn't have to refer to the supplementary table to find this information. 

Effects shown in figs 2 and 3 are combined across subjects. We don't know the effective sample size for the comparisons being made, and the effects shown could be driven by just a few subjects. If the authors compute trial-wise regressions between mood and BGA for each subject, and then perform the statistics across subjects instead of across electrodes, do these results still pan out? 

Furthermore, how many of the subjects show statistically significant regressions between BGA and mood at any electrode? For example, the error bars in fig 2b are across electrodes. How would this figure look if error bars indicated variance across subjects instead? 

In panel f, we can see that a large number of sites in both ROIs show correlations in the opposite direction to the reported effects. How can this be explained? How do these distributions of effects in electrodes correspond to distributions of effects in individual subjects? 

Baseline (pre-stimulus) gamma amplitudes have been shown to be related to attentional states. Could these effects be driven by attention rather than mood? The relationship between mood and decisions may be more complex than the authors describe, and could impact other cognitive factors such as attention, which have already been shown to impact baseline broad-band gamma. 

The authors used a bipolar montage reference. Would it be possible that effects in low frequencies are dampened because of the bipolar reference instead of common average reference?

Reviewer #3 (Public Review): 

In this interesting paper, Cecchi et al. collected intracerebral EEG data from patients performing decision-making tasks in order to study how patient's trial-by-trial mood fluctuations affect their neural computation underlying risky choices. They found that the broadband gamma activity in vmPFC and dorsal anterior Insula (daIns) are distinctively correlated with the patient's mood and their choice. 

I found the results very interesting. This study certainly will be an important contribution to cognitive and computational neuroscience, especially how the brain may encode mood and associate it to decisions. 

Questions: 

1) The authors showed that the mood is positively correlated in vmPFC on high mood trials alone and negatively correlated daIns in low mood trials alone. This is interesting. But those are the trials in which these regions' activity predict choice (using the residual of choice model fit)? 

2) It would be helpful to see how high-mood trials and low-mood trials are distributed. Are they clustered or more intermixed? 

3) I am not sure how I should reconcile the above finding of the correlation between mood and BGA on high-mood vs. low-mood trials, and the results about how high vs. low bassline BGA predict choice. I may have missed something related to this in the discussion section, but could you clarify?

Reviewer #1 (Public Review): 

The manuscript by Gronberg et al. discloses the crystal structures of the P1B4-type ATPase sCoaT in a late E2P state (E2P*) and a transition state of E2P dephosphorylation (E2.Pi). As P1B-ATPases maintains heavy metal homeostasis in the cell, its inhibitors are the candidate of antibiotics. Two crystal structures with phosphate analogs BeF and AlF reported in this study appear to represent the same conformation of outward-occluded E2-P state. Determined structures of sCoaT define its TM topology and absence of HMBD in its N-terminus. Tight configurations in the TM metal-binding site centered by H657 and C327 in E2P state suggest the role of H657 as a build-in counter ion as is observed in the SsZutA Lys-Asp pair. The ATPase assays address the role of the N-terminus, an inter-domain salt-bridge, as well as residues hypothesized to contribute to substrate binding and release. MD simulation suggests the exoplasmic gate opening to release cargo mental. They also found a salt bridge formation at the cytoplasmic regulatory site that mimic K+-binding hence this enzyme is K+-independent. The authors further explored the specific inhibitors for sCoaT, and two candidate compound shows antibiotic activity.

Reviewer #2 (Public Review): 

In this study, the authors determined the high-resolution structure of the ATP-driven metal transporter (P1B-4 ATPase), characterized unique structural and mechanistic features of this transporter, and developed inhibitors for this transporter. Transition metals such as Zn, Co, and Cu play important roles in many aspects of cell biology as cofactors of enzymes, integral components of protein structure and signaling molecules. The concentration of transition metals is regulated by the P1B-type ATPases, which also serve as important virulence factors in microorganisms. This new structure of the P1B-4 type ATPase with specificity to zinc, cobalt, and several other metals resolves the controversy about this protein topology and add new important information about its transport mechanism. 

The work is carefully executed. Generation of the structural model is accompanied by mutational studies that test the functional significance of several amino-acid residues and by molecular dynamic simulations, which explore the mechanism of metal release. This approach offers solid support to the authors' claims about a distinctness of the P1B-4 ATPases' transport mechanism. The following findings are particularly significant. The authors demonstrated that the Cys-xx-Cys motif, which in other P1B ATPases is found in the N-terminus, in P1B-4 ATPases is the part of the membrane domain. The authors also identified the conserved histidine residue as a potential built-in counter ion and demonstrated independence of the transporter activity of potassium. These new findings expand our understanding of how transition metal transporters work. The authors also made the first step towards identifying specific inhibitors of this transporter, which show efficacy in vivo against Mycobacterium, and thus provided foundation for drug discovery. Overall, the work is an important milestone in studies of the transition metal transporters. 

The structural models were built for two similar conformations that do not have bound metal. This limits the authors ability to discuss the structural basis of metal selectivity, which is an important topic for future studies.

Reviewer #3 (Public Review): 

The authors provide important new insights into a heavy-metal transporting P-type ATPase, sCoaT from Sulfitobacter sp. NAS14-1. The functional and structural data presented are high quality and novel insights are provided. The overall structure, the internal counterion, the potential ion-release pathway, and the A-domain modulatory site are all interesting insights into this subclass of transport proteins. However, despite assertions that the PIB-4 P-type ATPases are distinct with uncertain topologies, the structure of sCoaT is similar to previous structures of PIB-2 P-type ATPases CopA and ZntA. 

The authors describe sCoaT as a zinc transporter, yet the ATPase activity is almost three times higher for cadmium. The authors do not discuss this or why cadmium is not considered a preferred transport ion. 

Two structures of sCoaT are presented in the presence of BeF3- (3.1Å) and AlF4- (3.2Å). The authors claim that these structures represent different transport intermediates (late E2P and E2.Pi, respectively), yet the structures are identical to one another and they appear to be in an E2.Pi conformation. The density surrounding the phosphomimetic ligands are not shown and the densities are quite similar in the electron density maps provided. Stronger evidence or supporting figures need to be provided to show that BeF3- and AlF4- are in fact bound. It is possible that the crystal lattice traps the same conformation of sCoaT despite the use of ligands that should trap E2P and E2.Pi intermediates. The assertion that the two structures represent different, closely related conformational states, is not yet supported by the current presentation.

Reviewer #1 (Public Review): 

Authors introduced new strategy of genetic manipulation in mice to reveal functional development of the retrotrapezoid nucleus (RTN) neurons that is known as an important brainstem region for central chemoreception and the dysfunction is relate to congenital central hypoventilation syndrome (CCHS) neuropathology. They used a conditional mutation of Phox2b within Atoh1-derived cells (Atoh1Cre/Phox2bΔ8 mice) and examined a) respiratory rhythm; b) ventilatory responses to hypercapnia and hypoxia and c) number of RTN-chemosensitive neurons. They found that 1) mice with mutant Phox2b expression showed a suppressed breath activity to hypoxia and hypercapnia in neonates; 2) adult mutant mice presented irregular breathing pattern, partial recovery of the ventilatory response to hypoxia and complete recovery of response to hypercapnia; 3) anatomical data showed reduced number of activated neurons by hypercapnia and Phox2b immunoreactivity in the RTN. They concluded that conditionally expression of Phox2b mutation by Atoh1 affected development of the RTN neurons and suggested that Atoh1/Phox2b system in the RTN was essential for the activation of breathing under hypoxic and hypercapnia condition. They thought that their findings provided new evidence for mechanisms related to CCHS neuropathology. 

The conclusions of this paper are well supported by data, but careful discussion seems to be required for comparison with results of various previous studies performed by different genetic strategies for the RTN development.

Reviewer #2 (Public Review): 

Mutations in the Phox2B gene can lead to congenital central hypoventilation syndrome with variable presentations. Two distinct classes of causative mutations have been found in the human population. The first group consists of mutations that result in trinucleotide, polyalanine repeat expansions, referred to as PARM. The second group are non- polyalanine repeat expansion mutations (NPARM) that includes missense, nonsense, and frameshift mutations. Each group (and even specific mutations) present with differing clinical phenotype severity, with NPARM mutations typically being more severe. As Phox2B is expressed across a multitude of cell types across the life an individual, there remains much to be understood as to the cell specific effects of various Phox2B mutations on phenotype. To add to our understanding, the authors utilized a conditional Phox2bΔ8 allele that, upon recombination, replaces Exon 3 and UTR with a mutated exon and IRES GFP reporter. This approach allows for an inducible NPARM mutation and reporter expression in a targeted cell type. The authors focused on Atoh1 expressing cells using an Atoh1 expressing Cre recombinase line (Atoh1_Cre). Atoh1 has been shown to also be co-expressed in the RTN and in the para and inter-trigeminal regions of the Pons. After inducing the Phox2B mutations in one allele, the authors examined respiratory features in both adults and neonate mice under room air, hypercapnia (7%) and Hypoxia (8%). The Atoh1_Cre; Phox2bΔ8 adult mice showed a significant body weight difference. Under their plethysmography approach neonate mice breathing room air showed few differences with a potential difference in tidal volume. Notably adult mice show irregularity in their breathing. Both adult and neonate mice may show compromised chemosensory deficits. A potential hypercapnic deficit likely resolves in the adult but there may remain a compromised hypoxic reflex in the adult. Notably, Atoh1_Cre; Phox2bΔ8 mice showed reduced cfos expression in the RTN after hypercapnic stimulation and reduced Phox2B immuno-reactivity. 

The premise of the paper is to examine how a distinct mutation in a specific cellular context may contribute to clinical outcomes. The potential phenotypes are interesting and illuminate how differing mutations may drive different phenotypes or phenotype severity. While the RTN is likely a key mediator of the reported phenotypes, the conclusions drawn by the authors cannot be fully supported with the data presented. 

The authors assign all phenotypes to RTN function. However, there are other documented and potential undocumented areas of Atoh1 and Phox2b overlap that could either impact breathing directly or indirectly through metabolism and stress responses (PMID 8184995). As noted above, para trigeminal neurons including those in the ITR also co-express Atoh1 and Phox2B and are captured in the Atoh1_Cre; Phox2bΔ8 mouse model. The inter-trigeminal region is associated with apneic reflexes and jaw opening(PMID: 19914183). Thus, perturbations to this center may underlie the increased irregularity seen in adult life. A potential role in chemosensory function cannot be entirely ruled out either. While Rose et al. assert that the RTN and para- and inter- trigeminal neurons are the only ones co-expressing Atoh1 and Phox2B (using antibodies), the persistent cumulative GFP labeled fate map offered by the Atoh1_Cre; Phox2bΔ8 model would allow the authors to rule in or rule out any other uncharacterized overlapping populations. Such a fate map may also help to inform as to why the adult mice are significantly underweight. The weight phenotype may stem from metabolic dysregulation, changes in behavior, or feeding. Changes in metabolism may drive secondary changes in breathing and chemosensory reflexes that play a role in the reported phenotypes. Ultimately, the relative roles of para-trigeminal and RTN neurons in these phenotypes should be dissected out. 

Both the adult and neonate plethysmography was not collected in line with current best practices. Adult whole body plethysmography is best carried out in a temperature controlled chamber held at thermo-neutrality. This minimizes any thermo-regulatory and metabolic effects on respiratory drive. Concurrent measurement of one or more metabolic parameters such as VO2 or VCO2 is required to determine if baseline breathing and chemosensory reflex phenotypes may be affected by changes metabolism or persistent metabolic imbalances (acidosis or alkalosis). Whole body measurements in neonates are do not allow for accurate assessment of tidal volume. Rather head out or facemark pneumotachography are more accurate, (PMID: 25017785).

Reviewer #3 (Public Review): 

The work by Ferreira and colleagues set to define the functional consequences of a PHOX2B (Phox2bdelta8) mutation, belonging to the group of non-polyalanine repeat expansions, when restricted to Atoh1 expressing cells. In doing so, the authors generated a new mouse model (Atoh1Cre,Phox2bdelta8 mice) for the study of the central respiratory chemoreceptor circuit. Ferreira et al., found that these conditional mutants present with largely unaffected breathing parameters in postnatal life. However, neonatal breathing irregularities, normally observable in control neonates, are not corrected with the maturation of the conditional mutants. Furthermore, the authors found that conditional Atoh1Cre,Phox2bdelta8 neonates fail to display ventilatory responses to hypoxic (low O2 content in air) and hypercapnic (high CO2 content in air) challenges. The authors show that Atoh1Cre,Phox2bdelta8 adult mice appear to "recover" the capacity to response to hypercapnic, but not hypoxic, challenges. Lastly, the authors found reduced numbers of Phox2b+ cells in an "area" where the retrotrapezoid nucleus, a key center in the respiratory chemoreceptor circuit, normally locates. 

Strengths: 

The most exciting aspect of this work is the modelling of the Phox2bdelta8 mutation in one element of the central neuronal circuit mediating respiratory reflexes, that is in the retrotrapezoid nucleus. To date, mutations in the PHOX2B gene are commonly associated with most patients diagnosed with central congenital hypoventilation syndrome (CCHS), a disease characterized by hypoventilation and absence of chemoreflexes, in the neonatal period, which in severe cases can lead to respiratory arrest during sleep. Two distinct types of PHOX2B mutations have been identified in CCHS patients: i) polyalanine repeat expansions, and ii) non-polyalanine repeat expansions. Non-polyalanine repeat expansions tend to be more prevalent in severe cases of CCHS. Thus, the characterization of the Phox2bdelta8 mutation could allow for a better understanding of the etiology behind CCHS. 

Weaknesses: 

Whereas the most exciting part of this work is the modelling of the Phox2bdelta8 mutation in retrotrapezoid neurons using conditional mutagenesis driven by Atoh1 (i.e. Atoh1Cre,Phox2bdelta8 mice), the weakness of this study is the lack of a clear physiological, developmental, and anatomical distinction between this approach and similar studies already reported elsewhere, for instance the use of Atoh1Cre,Phox2bflox/flox and P2b::CreBAC1;Atoh1lox/lox mice (Ruffault et al., 2015, DOI: 10.7554/eLife.07051), Egr2cre;P2b27Alacki (Ramanantsoa et al., 2011, DOI: 10.1523/JNEUROSCI.1721-11.2011), Atoh1Phox2bCKO mice (Huang et al., 2017, DOI: 10.1016/j.neuron.2012.06.027) and Egr2cre;Lbx1FS (Hernandez-Miranda et al., 2018, DOI: 10.1073/pnas.1813520115). 

Several conclusions presented in this work are not directly supported by the provided data. For instance, the reduction in the number of retrotrapezoid neurons in Atoh1Cre,Phox2bdelta8 mice or the reduction of fos+ activated retrotrapezoid neurons after CO2 exposure, as the identity of retrotrapezoid neurons was not thoroughly determined. Furthermore, the authors conclude from their plethysmograph (respiratory recordings) data that Atoh1Cre,Phox2bdelta8 neonatal mice display an impaired ventilatory responses to hypoxia (low O2 in air) and hypercapnia (high CO2 in air), but that these mutant animals recover the capacity to respond to hypercapnia, but not to hypoxia, in the adult life. This is a bit of an overstatement, as their plethysmograph recordings show that adult Atoh1Cre,Phox2bdelta8 mice do respond to low O2 in air, as these mice accelerate respiration, increase tidal volumes and minute ventilation in the same fashion as control mice. However, what the presented data show is that adult Atoh1Cre,Phox2bdelta8 mice do not sustain the ventilatory response as efficient as control mice.

Reviewer #1 (Public Review):

Nasrollahzadeh et al. tried to explore the optimal physiological culture conditions, like the thermo and the mechanical environments for chondrocytes in vitro . For this purpose, an extracorporeal model system is designed to simulating the joint movement generated mechanical stimulation and self-heating. As the authors explained, this model system consists of 2 parts: 1). a fatigue resistant hydrogel functionalized by RGD peptides and loaded with human chondro-progenitor cells to recapitulate cartilage viscoelastic properties; 2). a modular bioreactor to independently control applied mechanical loading, temperature increase as well as gas concentration and humidity levels during stimulation. Whereafter, the individual and cooperative effects of temperature and mechanical loading are investigated based on this system, and chondrogenic related gene expressions, like Col2a, Agc et al. are employed to determine the cartilage regeneration. The results indicated that the loading induced self-heating significantly enhance the chondrogenic responses. Furthermore, the role of Ca2+ signaling in thermo-mechanotransduction process is studied. The TRPV4 ion channels are chosen as model and the involvement is confirmed by the with or without of TRPV4 agonists and inhibitors. In general, the conclusions of this paper are mostly well supported by data, but some aspects of image acquisition and data analysis need to be clarified and extended.

1) The authors functionalized the hydrogel with RGD-peptide to improve the cell adhesion and spreading on the scaffold, so hydroxylation and RGD grafting protocols are performed. Whether these reactions change the surface morphology and further influence the cell morphology ? If more details about the surface morphology, such as SEM images can be provided, it will be enlightening.

2) Although gene expression is one of the direct evidences for inducing chondroblast differentiation, more detection methods, such as GAG expression can be provided to make it more convincing.

3) As the authors hypothesize, calcium signaling is a major mechanism of action in transduction of thermo-mechanical cues sensed by TRPV4 channels. However, the available data on calcium signaling pathways do not support this conclusion strongly.

4) It's better to numbering the Figures in SI section according to the sequence in which they appeared in the manuscript.

Reviewer #2 (Public Review):

This is an interesting and exciting study that addresses a topic in cartilage mechanobiology that has generally been neglected, that is, the effects of mechanical loading on temperature changes in the joint, and how that might affect cell response to loading. I think the work is a major breakthrough. There are some specific questions about the model system that could use further explanation or discussion to improve the impact of the work.

Reviewer #1 (Public Review):

This manuscript provides an interesting analysis of the evolution of the IR75a protein across the Drosophila phylogeny. As reported by the authors previously, IR75a in D melanogaster and several related species preferentially recognize the odor acetic acid, while IR75a in D sechellia instead prefers butyric acid. Here, the authors report that more distant Drosophila species, as well as the putative ancestral IR75a, also prefer butyrate, suggesting that IR75a changed its preference from butyrate to acetate within the melanogaster/obscura group, with reversion to butyrate subsequently occurring in D sechellia. Moreover, the authors identify a key site (position 538) whose identity as Phe or Leu tracks with odor preference. They also identify other secondary lineage-specific mutations that presumably provide structural support to help optimize ligand preference. Interestingly, different solutions for secondary optimization were observed across lineages, suggesting multiple evolutionary and structural paths for tweaking the ligand pocket. These data are generally solid and expertly generated, but I do note that there is substantial speculation based on molecular modeling (which the authors acknowledge) as well as speculation of mutational timeline which should be trimmed or removed. There are many ways to extend these findings, such as linking odor recognition properties to behavior, which would substantially increase the impact of this study.

Reviewer #2 (Public Review):

While the tuning of sensory receptors is thought to have an important role in shaping ecological fitness, how tuning shifts occur through evolution is poorly understood, in part due to the difficulties in comparing receptor sequences and receptive fields in highly diversified families of receptors.<br> This work seeks to identify the molecular trajectory that led to the functional divergence of a previously characterized insect olfactory receptor, Ir75a, providing an unusual opportunity to hone in on individual amino-acid residues that are causative of functional divergence.<br> The authors first compare the receptor responses of 10 orthologous receptors in a Drosophila lineage spanning 40 million years of evolution and find that the response profiles can be classified into two main categories. This functional clustering largely matches the evolutionary relationships between the studied species and allows the authors to hypothesize the evolutionary trajectory that the ancestral receptor followed to give rise to differences in function. Based on the extant receptor sequences in the group they predict two ancestral sequences at key positions in the phylogeny using maximum likelihood estimates, and then 'resurrect' these receptors by synthesizing the genes and expressing them in a well characterized in vivo expression system. They find that these hypothetical ancestral proteins functionally behave as predicted, and, using sequence alignments and previous experimental observations, they narrow down the candidate amino-acid positions that cause functional variation to 3 residues. Mutation of these 3 residues in the Dmel Ir75a into the Dsec variant shifts its tuning to perfectly match the tuning of the Dsec receptor, and the converse experiment yields the analogous result. The authors then synthesize the 8 possible combinations of these 3 mutations (single, double, and triple mutants) and functionally test their responses in vivo. This set of experiments allows them to analyze the phenotypic contribution of each mutation, which they then use to hypothesize on the most likely evolutionary trajectory from the ancestral Dmel-like receptor variant to the Dsec variant.<br> This analysis identifies a single position that has the largest epistatic effect. Across the clade, two residues can be found occupying this position, and the identity of this residue perfectly segregates with the functional phenotype of this receptor in the 10 species under study. Through this analysis the authors arrive to the conclusion that this particular site in this receptor has undergone at least 2 critical retuning mutations in a timespan of 40 million years. Further sequence alignments and chimeric experiments show that the analogous position in a related receptor Ir75b is in part responsible for adaptive change, leading the authors to conclude that this is a 'hotspot' for evolutionary variation, repeatedly tweaked across the lineage to achieve species-specific olfactory capabilities. The authors then use homology models and docking studies to attempt to shed light on the possible role of this amino-acid position in receptor function. Overall, the study is rigorous, detailed, and elegantly designed. The experimental layout is carefully explained, and the conclusions are cautiously presented.

Strengths: A main contribution of this work is to provide a plausible evolutionary path of discrete molecular events that led to the adaptive tuning of a receptor, and substantiate this analysis with an unusual high level of experimental support. Indeed, these studies are often hindered by difficulties in expressing receptors for functional studies as well as a high level of sequence divergence that prevents the identification of potential sites of causal change. In that light, the functional analysis of receptors across a lineage, together with the reconstruction of ancestral variants and dissection of the evolutionary trajectory of this receptor's tuning represents an important contribution to the field. This set of experiments and analyses offers an elegant example of a rigorous way to approach the challenging topic of the genetic basis of sensory adaptation.

Weaknesses: The authors attempt to link the characterized molecular events to the ecological needs that might have played a significant role in the linage's evolution, and to the structural aspects of receptor-ligand interaction. These two aspects are on the more speculative side, which the authors themselves acknowledge as limitations of the study.

1. In terms of the ecological context, this study focuses on a narrow set of ligands that are probed at concentrations that are quite high and thus of unclear physiological relevance. While their results are very exciting, they are restricted to the inverse relationship of the responses to C2-C4 carboxylic acids. Although data in the literature shows that these ligands, in particular acetic acid, are likely relevant, other data support that ligands from the same series may also be relevant. In particular, noni volatiles are dominated by octanoic acid (Auer et al., 2020; Pino et al., 2010), and Ir75a responds very strongly to propionic acid as well as acetic acid (Pietro Godino et al., 2016, Silbering et al., 2011). While the C2-C4 relationship captures most of the variance in PCA (which leads the authors to focus on these ligands in the first place), perhaps at other more physiological concentrations the relationship between other ligands in the series becomes more prominent, which would be interesting to explore.

2. The authors don't discuss whether the proposed polymorphisms are found in population genomic data, which is available at least for Dmel and Dsec. Mining these datasets and looking at intraspecific variation (or lack thereof) has the potential to support their speculations on the evolutionary trajectories of mutations with empirical data and offer complementary insight.

3. A more critical limitation is the use of docking onto homology models. Modeling techniques are incredibly powerful as they can provide solid hypotheses for how protein-ligand interactions might occur. However, much caution should be taken to interpreting modeling results without experimental validation. The reliability of homology models scales substantially with sequence identity, which turns this protein family into rather poor substrates for extracting atomic-scale conclusions from these models. In this case, homology models are combined with docking and very little support is offered. For example, the docking scores presented for the homology models are relatively low, and there is no significant difference between the docking score of acetic and butyric acid onto Dmel Ir75a. Although it is well known that docking results will in general only qualitatively match the behavior of a receptor-ligand pair, in the absence of alternative validation of the modeling procedures, these results fail to convince the reader that the homology models and docking results are reasonably likely. It should be noted that the proposed mode of action is entirely plausible and an interesting possibility, but as it is it appears too speculative and without validation.

These weaknesses do not detract from the overall value of the study and the authors take considerable care in acknowledging some of these caveats, which helps interpret these results and weigh the various aspects of it in full light.

Reviewer #3 (Public Review):

Although genomic information has been increasingly accumulated in closely related insect species, a molecular basis for functional changes in the olfactory receptors and the relationship with a change in food preference during the course of evolution has not been fully resolved. The authors successfully identified a 'hotspot' amino acid in an olfactory receptor, IR75a, that plays a crucial role in the selectivity of acid odors that appears to affect feeding behavior in Drosophila species. The finding provides an insight into how a molecular evolution occurred in an IR in correlation with the biological significance.

Reviewer #1 (Public Review):

The manuscript assessed the role of the anterior cingulate area (ACA) in fear acquisition and expression in a naturalist task using a live cat as the unconditioned stimulus where mice were the experimental subjects. Silencing of the ACA using DREADDs showed that the ACA is important for contextual fear acquisition and expression. Using the retrograde tracer Fluoro Gold in conjunction with Fos immunohistochemistry, the authors provide evidence for the involvement of RSPv, VISam, CL, AMv inputs to the ACA in during cat exposure, and primarily CL input during threat retrieval in the context. The authors then examined the role of some of these input structures to the ACA in the task using continuous 5min optogenetic inhibition. Silencing AM>ACA input during cat exposure (i.e., acquisition) impaired in contextual fear retrieval. Individual ACA>BLA and ACA>PERI pathway inhibition during cat exposure but not during disrupted contextual fear retrieval in line with greater FG+Fos double labeling in the former compared to the latter condition. ACA>POST pathway inhibition was without effect. ACA>PAGdl pathway inhibition during cat exposure had no effect on contextual fear retrieval whereas inhibition of this pathway during test disrupted retrieval, again in line with greater FG+Fos double labeling in the latter compared to the former condition.

One of the greatest strengths of the manuscript are the detailed analyses of the ACA input and particularly output pathways in the task. Further, the manuscript examined the effects at both acquisition and retrieval, providing a thorough examination of the effects of ACA and related structures at key memory phases. The complementary analysis of FG+Fos is most welcome, showing what level of overlap is necessary for a behavioural effect of pathway inhibition.

Some weaknesses of the manuscript design include the prolonged continuous optical inhibition, lack of optical control and the possibility of a state-dependent effect on neural manipulation. It is unclear if the similarity in the effects obtained in some cases are due to similarity in the underlying behavioral process that is disrupted (e.g. failing to acquire an accurate representation of the context vs. associating the context with the threat).

Reviewer #2 (Public Review):

The authors' main goals were to identify key circuits for the acquisition and expression of contextual fear conditioning in a paradigm that uses a live cat as the unconditioned stimulus. They used neuroanatomical tracing, opto-, and chemo-genetic techniques to observe and manipulate activity in anterior cingulate area (ACA) afferents and efferents at either the acquisition or retrieval stages. The strengths include a thorough characterization of multiple circuits and robust behavioral effects. Weaknesses include the confound of the experimenter being in the room for the acquisition, but not retrieval phase, a lack of characterization of escape-like behaviors, and the exclusive use of male animals, which reduces the potential impact.

Reviewer #3 (Public Review):

In this study, de Lima et. al. examined the function of ACA in acquisition and expression of contextual fear memory to predator threat. The authors found that ACA is necessary for both processes. Using optogenetic terminal inactivation, the authors further demonstrate a necessary role of AM input to ACA in the contextual fear acquisition phase. At the output level, the projections from ACA to BLA and PERI are necessary for contextual fear acquisition while the projection from ACA to PAG is essential for contextual fear expression. Overall, the study is interesting and the results are straightforward. The presented data largely support the conclusions. The paper will provide new insight into the neural circuit underlying contextual fear learning and expression to predator threat. Some limitations of the study include the lack of controls to demonstrate how specific is the ACA response to threat and threat paired context is and validation of the terminal inhibition. Further characterization of the projections would also help to understand these results.

Reviewer #1 (Public Review):

In this manuscript the authors introduce a novel operant conditioning paradigm for prairie voles that assesses motivation of individuals to access pair-bonded partners versus unfamiliar social individuals. Prairie voles are an important model species for studying social relationships, and a large group of investigators have settled on standard methodology for partner preference tests to assess pair bonds. As the authors highlight, partner preference tests (a current gold standard in the field) cannot distinguish between motivation to associate with a pair-mate or avoidance of an unfamiliar individual. Operant training for social stimuli has only been done a handful of times, and never used to study social relationships in an animal model. The authors also combined lever pressing data with ethological scoring of behavior when focal animals received access to stimulus animals. Their results show that while female prairie voles lever press more for familiar individuals (both a male mate or same-sex cagemate), males lever press more for access to females regardless of familiarity. Lever pressing for an empty compartment was positively correlated with pressing for an unfamiliar individual, suggesting that lever presses for strangers might be spontaneous. In contrast when measuring huddling behavior (the standard measure in partner-preference tests), both males and females huddled more with familiar individuals. The experiments are rigorously conducted and proper extinction controls are included to demonstrate the effectiveness of the method. The authors' results highlight how partner-preference tests capture only a fraction of the behavioral processes involved in pair bonding in prairie voles and make an important contribution to our understanding of the processes involved in maintenance of social relationships.

Reviewer #2 (Public Review):

The authors used an operant task in voles to assess preferences for social relationships and whether these preferences differed by sex and species. They also correlated some outcomes with oxytocin receptor binding.

A strength of the paper is the use of the vole models which allow comparisons between socially monogamous (prairie voles) vs. promiscuous breeders (meadow voles). Because prairie voles show a stronger preference for peers and mates than many other rodent species, they are a great model to assess selective relationships. The other major strength of this paper is the use of the operant procedure to assess preference. To my knowledge, this is the first time this procedure has been used to assess social reward in a monogamous species, and it has advantages over place preference procedures.

A weakness of the paper is the omission of groups from certain manipulations. Male meadow voles were excluded from the operant procedure and all male data was excluded from oxytocin receptor analysis with little rationale. This limits some of the conclusions that can be made about males in the study. Additionally, some experimental design details were missing. A subset of rats went through extinction. However, it was unclear whether those used for oxytocin receptor analysis went through extinction or not (or were from both conditions). The biochemical analysis also assessed the effect of oxytocin receptor genotype, replicating the effect that C allele carriers have higher oxytocin receptor binding in specific brain regions. However, the analysis of this genotype's effect on behavior was limited due, presumably, to power issues with most measures. Thus, conclusions regarding genotype were limited. Finally, there was a missed opportunity to do a progressive ratio test to better assess motivation for the partner rat.

The operant task and the subsequent behavioral results will be useful for the field, but the design issues somewhat limit impact. However, assessing the formation of selective relationships and using the vole model is innovative.

Reviewer #3 (Public Review):

In this paper, the authors attempted to disentangle social preference and social motivation in two different vole species (monogamous prairie voles and polygynous meadow voles) and in different types of relationships (same-sex and opposite-sex). They also examined the role of sex differences in these questions. Finally, they examined a single nucleotide polymorphism in the oxytocin receptor gene, and its role in predicting oxytocin receptor density and social behavior.

Strengths: The authors have performed very elegant and detailed behavioral tests and analyses. The focus on sex differences and relationship type, and the inclusion of same-sex pairs, is a major strength. The graphs are nicely done and I especially appreciate the individual data points in some graphs and the way that they are connected, so we can see how the same individual performed in different conditions. The authors included a food control in the rewards experiments, showing that their results were specific to social reward.

Weaknesses: At least one part of the study (the oxytocin receptor density) was carried out only in females. It is also not clear what broader significance the role of this particular single nucleotide polymorphism of the oxytocin receptor might have; for example, whether this same polymorphism is found in other species including humans. Some experiments had relatively small sample sizes.

The authors did achieve their aims and the results supported their conclusions.

Studies on same-sex relationships are still very rare in behavioral neuroscience, and in the study of animal behavior in general. It is quite significant that these authors have shown that female prairie voles are just as motivated to gain access to a same-sex partner as an opposite-sex partner. While most of these methods have been published elsewhere already, this paper is a model of attention to detail in the study of behavior.

In addition, the prairie vole vs. meadow vole comparison is now a classic one in social neuroscience, and this paper adds depth to the interpretation of differences between the species.

Reviewer #1 (Public Review): 

In this manuscript, the authors challenge the long-standing conclusion that Orco and IR-dependent olfactory receptor neurons are segregated into subtypes such that Orco and IR expression do not overlap. First, the authors generate new knock-in lines to tag the endogenous loci with an expression reporter system, QF/QUAS. They then compare the observed expression of these knock-ins with the widely used system of enhancer transgenes of the same receptors, namely Orco, IR8a, IR25a, and IR76b. Surprisingly, they observe an expansion of the expression of the individual knock-in reporters as compared to the transgenic reporters in more chemosensory neurons targeting more glomeruli per receptor type than previously reported. They verify the expression of the knock-in reporters with antibody staining, in situ hybridization and by mining RNA sequencing data. 

Finally, they address the question of physiological relevance of such co-expression of receptor systems by combining optogenetic activation with single sensillum recordings and mutant analysis. Their data suggests that IR25a activation can modulate Orco-dependent signaling and activation of olfactory sensory neurons. 

The paper is well written and easy to follow. The data are well presented and very convincing due in part to the combination of complementary methods used to test the same point. Thus, the finding that co-receptors are more broadly and overlappingly expressed than previously thought is very convincing and invites speculation of how this might be relevant for the animal and chemosensory processing in general. In addition, the new method to make knock-ins and the generated knock-ins themselves will be of interest to the fly community. 

The last part of the manuscript, although perhaps the most interesting, is the least developed compared to the other parts. In particular, the following points could be addressed: 

- It would be good to see a few more traces and not just the quantifications. For instance, the trace of ethyl acetate in Fig. 6C, and penthyl acetate for 6G. 

- In Fig. 4D, the authors show the non-retinal fed control, which is great. An additional genetic control fed with retinal would have been nice. 

- It appears that mostly IR25a is strongly co-expressed with other co-receptors. The provided experiments suggest a possible modulation between IR25a and Orco-dependent neuronal activity. However, what does this mean? How could this be relevant? And moreover, is this a feature of Drosophila melanogaster after many generations in laboratories?

Reviewer #2 (Public Review): 

In the present study, the authors: 

1) generated knock-in lines for Orco, Ir8a, Ir25a, and IR7ba, and examined their expression, with a main focus on the adult olfactory organs. <br> 2) confirmed the expression of these receptors using antibody staining. <br> 3) examined the innervation patterns of these knock-in lines in the nervous system. <br> 4) identified a glomerulus, VM6, that is divided into three subdivisions. <br> 5) examined olfactory responses of neurons co-expressing Orco and Ir25a 

The results of the first four sets of experiments are well presented and support the conclusions, but the results of the last set of experiments (the electrophysiology part) need some details. Please find my detailed comments below. 

Major Points:

Line 167-171: I wonder if the authors also compared the Orco-T2A-QF2 knock-in with antibody staining of the antenna. 

Lines 316-319 (Figure 4D): It would be better if the authors compare the responses of Ir25a>CsChrimson to those of Orco>CsChrimson. 

Line 324-326: Why the authors tested control flies not fed all-trans retinal? They should test Ir25a-T2A-QF2>QUAS-CsChrimson not fed all-trans retinal as a control. 

Line 478-500: I wonder if the observed differences between the wildtype and Ir25a2 mutant lines are due to differences in the genetic background between both lines. Did the authors backcross Ir25a2 mutant line with the used wildtype for at least five generations? 

Line 1602-1603: Does the identification of ab3 sensilla using fluorescent-guided SSR apply for ab3 sensilla in Orco mutant flies. How does this ab3 fluorescent-guided SSR work? 

Line 1602-1604: There is no mention of how the authors identified ab9 sensilla. 

Line 1648: what are the set of odorants that were used to identify the different coeloconic sensilla?

Reviewer #1 (Public Review): 

This paper uses microfluidics and Xenopus extracts to show the effects of mitotic feedbacks on the cell cycle period. The authors show that perturbing both the activity of Wee1 and mitotic phosphatases can alter the cell cycle period. These are well-conducted and original experiments. A weakness of the paper is that it is framed very generally as a mechanism of controlling the cell cycle period. However, there are many additional mechanisms that likely contribute to control of cell cycle duration. This paper should be of interest to researchers working on the cell cycle, specifically on the regulation of mitosis. The microfluidic methodology should be generally applicable and significantly extends the ability to analyze multiple extracts at the same time.

Reviewer #2 (Public Review): 

This is a study employing an innovative approach of microencapsulation of Xenopus cycling extract to address the role of the positive feedback branches in the cell cycle oscillation. 

Xenopus extract is an established system to study the mechanism of cell division. At the core of the Xenopus cell cycle is a negative feedback oscillator of cyclinB-CDK-APC/C. Recently, the importance of phosphatases, such as pp2a, in regulating cell cycle was proposed. In addition to the central negative feedback circuit, the role of various positive feedback branches in a broader cell cycle network has been discussed, mostly through simulation. 

The method created in this study is a quantitative and high throughput platform to address mechanistic questions in Xenopus cell cycle. Their study explored the important question of the functions of different feedback modules in the cell cycle. 

I find their results generally interesting and valuable to the cell cycle community, though some appear incomplete.

Reviewer #3 (Public Review): 

The authors developed a microfluidic device and a reporter molecule to study cell cycle progression in micro size droplets of in frog egg extracts. They have investigated how interference with regulators of the G2/M transition or mitotic exit control the oscillations in the activity of the main cell cycle controller, Cdk1. They found that the inhibition of the G2 controller Wee1 did not reduce the responsiveness of the system for changes in Cyclin B levels. On the contrary, inhibition of PP2A leads to diverse responses, all having more severe effect on the cell cycle, some cases a complete block of Cdk1 activity changes. They update a published mathematical model to explain the observations of the diverse responses to PP2A inhibition. 

Technically and methodologically a great paper, where several new tools were developed, but the biological conclusions are somewhat limited, given earlier work in the field.

Reviewer #1 (Public Review):

This study reports the function of ZHX2 as a new oncogene in triple negative breast cancer. Large amount of work was conducted to validate its role in promoting cancer progression, and detailed analysis revealed the mechanism of ZHX2 interfering with HIF and activating downstream transcription. However, experimental in vivo data and the mechanistic exploration could be strengthened.

Reviewer #2 (Public Review):

In this manuscript, authors report that ZHX2 is upregulated in TNBC patients and cell lines. Several aggressive attributes of TNBC cells including proliferation, migration/invasion, and orthotopic tumor growth and spontaneous lung metastasis, require ZHX2 expression. The authors have used integrated ChIP-seq and RNA-seq analysis to show that ZHX2 is co-occupied with HIF1α on transcriptionally active promoter sites. Furthermore, the authors have used molecular modelling and site directed mutagenesis to establish that R491, R581, and R674 residues on ZHX2 are crucial for regulating HIF1 activity and downstream functions. Overall, the manuscript is (mostly) well written and the claims are well supported by the data. The data, however, is heavily reliant on in vitro cell-based models and lacks validation in TNBC mouse models or data from human tumors.

Reviewer #3 (Public Review):

Dysregulated expression of ZHXs has recently been reported in several cancer types, and ZHX2 is involved in carcinogenesis and cancer progression. In this manuscript, the authors perform a study on the molecular mechanisms how ZHX2 regulated TNBC progression. CHIP-Seq and gene expression profiling demonstrated that ZHX2 co-occupied with HIF1a promoters and promote gene expression, the evidences are solid and provide a potential target for breast cancer therapy.

Reviewer #4 (Public Review):

This work provides a new method to generate blood-brain barrier (BBB) endothelial cells (ECs) from human pluripotent stem cells (hPSCs), by activating the Wnt/β-catenin pathway at an early, susceptible time point when EC are still at a naïve state during differentiation. Although physiological inducers of Wnt/β-catenin signaling have been investigated, the work shows best performance for the small molecular GSK3 inhibitor CHIR99021 (CHIR) in the employed in vitro setting. Wnt/β-catenin hPSCs upregulated important BBB signature genes and promoted tightening of an EC monolayer. The experimental model could evolve into a more general technique typically used in BBB research, so it may allow rigorous testing of basic research questions at the BBB in vitro as well as compounds crossing the BBB for drug development.

Strengths:

The work deciphers in detail the schedule of EC differentiation from hPSCs and identified the correct time point for BBB-EC generation by Wnt/β-catenin activation. Hence, beside offering a tool for BBB and drug delivery research, the work opens the possibility to generate human BBB models from human pluripotent stem cells for personalized medicine.

Weaknesses:

Although the paper does have strengths in principle, the weaknesses of the paper are insufficient and in some case inappropriate analyses are performed to fully support the key claims in the manuscript by the data presented. In particular:

It would be important to demonstrate that Wnt7a/7b indeed are functional as recombinant proteins in the described experimental setting. Wnt7 has been shown to be only little diffusible, requiring the direct contact of the sending cell to the receiving cells (Eubelen, M. et al. Science (New York, NY) 361, eaat1178, 2018). The authors do not address in sufficient depth the cellular mechanisms of how naïve cells are susceptible for Wnt/β-catenin activation.

Reviewer #1 (Public Review):

Gastfriend et al present an analysis of the properties and gene expression patterns elicited by pharmacologically activating the Wnt signaling pathway in human pluripotent stem cells (hPSCs) that have been cultured under conditions that favor their differentiation to naïve endothelial cell (EC) progenitors. The result of Wnt activation is partial induction of blood-brain barrier-like properties, as judged by immunostaining for several well characterized marker proteins, trans-epithelial resistance, and RNAseq. Most of the experiments, and the largest effects, were obtained with CHIR99021, a small molecular weight inhibitor of GSK3-beta, the kinase that phosphorylates beta-catenin, leading to its ubiquitinylation and proteosomal degradation. Interestingly, low-passage ("naïve") hPSC-derived ECs were more susceptible to CHIR99021-induced BBB-like conversion than higher passage ("mature") hPSC-derived ECs. The experimental data is of uniformly high quality.

By way of context, several publications describe the use of hPSC or other starting cell sources for the generation of ECs with BBB-like properties, and several have described the BBB-enhancing effects of activating different signaling pathways (retinoic acid, TGF-beta, Wnt). The effect on BBB properties of manipulations that would be predicted to increase Wnt signaling varied among published studies - it was detectable in some studies (e.g. Paolinelli et al., 2013, Laksitorini et al., 2019) and it was undetectable in another (Sabbagh and Nathans, 2020). The present work adds to this literature and presents additional information on this attractive experimental system for dissecting Wnt signaling and potentially other signaling systems that affect the BBB-like differentiation of CNS ECs.

One potential question raised by the present study is the interpretation of the response to CHIR99021. GSK3-beta has many substrates, not just beta-catenin. Could the bioactivity of CHIR99021 for the BBB-like conversion reflect a combination of beta-catenin stabilization and reduced phosphorylation of other GSK3-beta substrates?

A minor point on page 8, lines 180-185. It might be appropriate to note that neural-rosette-conditioned and astrocyte-conditioned media may contain factors in addition to Wnt7a.

Reviewer #2 (Public Review):

The authors optimized the generation of endothelial progenitor cells from hPSCs and describe the effects of different Wnt-pathway activations on establishment of blood-brain barrier (BBB) characteristics.

Specifically they show, that Wnt pathway activation by Wnt 7a/b, neural rosette medium or most strongly the GSK-3 inhibitor CHIR is capable of inducing CNS-like phenotypes measured by expression of glut-1, claudin5 and caveolin-1. They further analyzed the barrier phenotype in CHIR treated cells and show an increase in tight junction protein expression. They also analyzed the timing and found that naïve endothelial progenitors are more responsive to Wnt activation than more mature ECs.

Using RNA sequencing for transcriptome analysis they show that many BBB EC-properties can be induced by Wnt signaling activation through CHIR, however other characteristic gene expressions are still lacking, indicating that other signaling pathways or stimuli are needed for developing an in vivo equivalent of BBB endothelial cells.

The authors succeed in the analysis of the model represented, however the impact for the community is limited by the lack of novelty and the (expected and existing) variation between the different differentiation protocols.

In general many of the claims (e.g. comparing between different effects of WNT ligands) are not supported by statistical analysis, as all conditions were only significant compared to the control.

The use of the different quantification methods (eg, number of Glut-1 expressing cells versus Glut1/Hoechst intensity of expression) is not always optimally adapted to the question, and therefore the drawn conclusions have to be observed with caution.

Overall this represents more a methods paper and will be valuable to the community as such.

Reviewer #3 (Public Review):

Benjamin D. Gastfriend et al. used endothelial progenitors derived from human pluripotent stem cells to decipher the effects of activation of the canonical Wnt pathway in naïve endothelial cells. This is an interesting approach to better understand the role of Wnt pathway in endothelial barrier genesis.

Strength: The experiments are well designed, and the findings are supported by data. A massive amount of data has been generated, notably using RNA sequencing and these data have been well exploited, notably by cross comparison with other datasets to compare the CHIR-regulated expression of genes in ECs to CNS-like ECs characteristics in other dataset. In addition, the dataset has been made available and could be used for further investigations in the field.

Weakness: This is a data-rich manuscript which sometimes could be simplified to facilitate understanding by readers.

Nevertheless, the conclusions of this paper are well supported by data and provide interesting insight in the contribution of Wnt pathway BBB formation.

Reviewer #1 (Public Review):

Manuscript by Li et al. entitled "Clp protease and antisense RNA jointly regulate the global regulator CarD to mediate mycobacterial starvation response" reports two mechanisms that regulate levels of CarD under stress conditions, including starvation. They report that level of CarD was tightly regulated, and there was a dramatic decrease in the levels of CarD when cells switched from the nutrient-rich to the starvation condition. They discovered two synergistic mechanisms that led to this dramatic decrease in the levels of CarD: 1) SigF-dependent induction of antisense RNA of CarD (AscarD), which inhibits CarD translation, and 2) Clp protease-mediated degradation of intracellular CarD.

CarD is an essential global transcription regulator which activates transcription and modulates the expression of about two-third of genes in M. tuberculosis. It binds RNA polymerase and helps in stabilizing the RNAP open complex. Based on the previous reports, the levels of CarD transcripts increase under stress conditions, including the starvation/stationary phase. Since CarD upregulates the expression of rRNA genes, so this was not clear how levels of genes coding for rRNA goes down while CarD transcripts go up under various stress conditions. This is one of the key unaddressed research gaps in understanding how CarD helps cells adapt under stress conditions. The data presented in the manuscript is an attempt to address this question.

As stated above, the authors succeeded in discovering two mechanisms that M. tuberculosis employs to tightly regulate the levels of a global transcription regulator. The experiments were well designed, and the data presented in the manuscript support the major conclusion that levels of CarD decrease under various stress conditions. This discovery now helps explain why levels of genes under the control of CarD goes down under stress conditions. This manuscript further reemphasizes to compare both transcriptomics and proteomics data to understand the role of target genes in biological processes. The study also highlights the importance of non-coding RNAs in regulating gene expression.

Reviewer #2 (Public Review):

Our understanding of bacterial mechanisms of transcriptional regulation has been pushed forward by studies in a few model organisms that were initially amenable to biochemical experiments. More recently, however, it has become clear that many regulatory pathways are distinct from these model organisms. This fact, coupled with continued appearance of drug-resistant strains of pathogenic bacteria motivates the serious study of important pathways directly in the pathogenic bacteria themselves.

The mycobacterial transcription factor CarD has been studied via many approaches and the biophysical mechanisms underlying its effect on transcription initiation are well-described. Curiously though, the biological role of CarD is far less well-understood. The authors have approached this question by uncovering two modes of the regulation of CarD concentration during nutrient starvation which provides new insight into how the molecular activity of CarD is beneficial to the bacteria.

First, they uncover the proteolytic-dependent degradation of CarD in conditions where growth needs to be slowed. Secondly, they describe an anti-sense RNA to the CarD transcript which likely inhibits CarD translation. Combined, these mechanisms reduce the amount of CarD in the cell, which then results in global changes in the gene expression. in particular, a reduction in ribosomal RNA synthesis fits well with the need to slow growth.

The data are very clear and this work will likely have a large impact on the field. It both sets the stage for future work on the response of Mtb to stress and provides an stress response pathway distinct to that of E. coli and other model systems that may help inform on the regulation in other pathogenic species.

Reviewer #3 (Public Review):

This is a study of the regulation of CarD expression by proteolysis and anti sense RNA. CarD is a well-studied RNAP interacting protein with complex roles on RNAP promoter dynamics, as the authors note, but the full mechanisms controlling its cellular levels under various stress conditions are not well defined. One prior report indicated that CarD is subject to proteolysis by the ClpP system. This study demonstrates that CarD protein levels are subject to ClpP proteolysis during stationary phase and that conditional expression of ClpP2 reverses this effect. The authors also identify an antisense RNA at the CarD locus that also negatively regulates CarD protein levels and overexpression of this antisense RNA confers sensitivity to various stresses, phenocopying prior data with CarD depletion.

Overall, the study is carefully done. The genetics are carefully done and in general the conclusions are well supported. The strongest aspects of the paper are the demonstration of the effect of Clp and asCarD on CarD protein levels. Given the substantial prior literature on CarD and its roles in mycobacterial transcription, this study does add substantial complexity to CarD biology.

I think the paper could be stronger in two areas:

1) Although the paper presents a clear picture of these two mechanisms individually on CarD levels, the effects of these mechanisms on cellular phenotypes is less completely explored. For example, for the stationary phase dependent regulation of CarD, does this mechanism effect stationary phase survival? Most of the data shows cell density, but no survival assays are done (as one example). Similarly, what about the the stress conditions examined? Some data is given, but I think more exploration of the cellular effects of these mechanisms would make the paper more appealing to a broad audience. With their ability of tune CarD levels via graded manipulation of asCarD or Clp, the authors could learn much more about the physiologic and global transcriptional effects of these regulatory mechanisms and this would be a major advance for the field.

2) Although the individual effects of Clp and asCarD are documented, the paper would be stronger if it explored the relative importance and interaction of these two mechanisms in different conditions. Are they simply additive as the model figure suggests? Does the relative importance of each mechanism differ depending on the condition studied?

Reviewer #1 (Public Review): 

Chen et al. trained male and female animals on an explore/exploit (2-armed bandit) task. Despite similar levels of accuracy in these animals, authors report higher levels of exploration in males than in females. The patterns of exploration were analyzed in fine-grained detail: males are less likely to stop exploring once exploring is initiated, whereas female mice stop exploring once they learn. Authors find that both learning rate (alpha) and noise parameter (beta) increase in exploration trials in a hidden Markov model (HMM). When reinforcement learning (RL) models were fitted to animal data, they report females had a higher learning rate and over days of testing, suggesting higher meta-learning in females. They also report that of the RL models they fit, the model incorporating a choice kernel updating rule was found to fit both male and female learning. The results do suggest one should pay greater attention to the influence of sex in learning and exploration. Another important takeaway from this study is that similar levels of accuracy do not imply similar strategies. Essential revisions include a request to show more primary behavioral data, to provide a rationale for the different RL models and their parameters, to clarify the difference between learning and 'steady state,' and to qualify how these experiments uniquely identify latent cognitive variables not previously explored with similar methods.

Reviewer #2 (Public Review): 

The authors investigated sex differences in explore-exploit tradeoff using a drifting binary bandit task in rodents. The authors tried to claim that males and females use different means to achieve similar levels of accuracy in making explore-exploit decisions. In particular, they argue that females explore less but learn more quickly during exploration. The topic is very interesting, but I am not yet convinced on the conclusions. 

Here are my major points: 

1) This paper showed that males explore more than females, and through computational modeling, they showed that females have a higher learning rate compared to males. The fact that males explore more and have lower learning rates compare to females, can be an interesting finding as the paper tried to claim, but it can also be that female rats simply learn the task better than male rats in the task used. 

(a) First, from Figure 1B, it looks like p(reward, chance) are similar between sex, but visually the female rats' performances, p(reward, obtained), look slight better than males. It would be nice if the authors could show a bar plot comparison like in Figure 1C and 1E. A non-significant test here only fails to show sex differences in performance, but it cannot be concluded that there are no sex differences in performance here. Further evidence needs to be reported here to help readers see whether there are qualitative differences in performances at all. 

(b) The exploration and exploitation states are defined by fitting a hidden Markov model. In the exploration phase, the agent chooses left and right randomly. From Figure 1E and 1F, it looks like for male rats, they choose completely randomly 70% of the times (around 50% for females). The exploration state here is confounded with the state of pure guessing (poor performance). 

(c) Figure 2 basically says that you can choose randomly for two reasons, to be more "noisy" in your decisions (have a higher temperature term), or to ignore the values more (by having a learning rate of 0, you are just guessing). It would be nice to show a simulation of p(reward, obtained) by learning rate x inverse temperature (like in Figure 2C). From Figure 2B, it looks like higher learning rates means better value learning in this task. It seems to me that it's more likely the male rats simply learn the task more poorly and behave more randomly which show up as more exploration in the HMM model. 

(d) From figure 3E, it looks like female rats learn better across days but male rats do not, but I am not sure. If you plot p(reward, obtained) vs times(days), do you see an improvement in female rats as opposed to males? Figure 4 also showed that females show more win-stay-lose-shift behavior and use past information more, both are indicators of better learning in this task. 

Taken the above together, I am not convinced about the strategic sex differences in exploration, it looks more like that the female rats simply learn better in this task. 

2) I do like how the authors define exploration states vs exploitation states via HMM using choices alone. It would be interesting to see how the sex differences in reaction time are modulated by exploration vs exploitation state. As the authors showed, RT in exploration state is longer. Hence, it would make a conceptual difference whether the sex difference in reaction times is due to different proportions of time spent on exploration vs exploitation across sex.

Reviewer #3 (Public Review): 

In the manuscript 'Sex differences in learning from exploration', Chen and colleagues investigated sex differences in decision making behavior during a two-armed spatial restless bandit task. Sex differences and exploration dysregulation has been observed in various neuropsychiatric disorders. Yet, it has been unclear whether sex differences in exploration and exploitation contributes to sex-linked vulnerabilities in neuropsychiatric disorders. 

Chen and colleagues applied comprehensive modeling (model free Hidden Markov model (HMM), and various reinforcement learning (RL) models) and behavioral analysis (analysis of choice behavior using the latent variables extracted from HMM), to answer this question. They found that male mice explored more than female mice and were more likely to spend an extended period of their time exploring before committing to a favored choice. In contrast, female mice were more likely to show elevated learning during the exploratory period, making exploration more efficient and allowing them to start exploiting a favored choice earlier. 

Overall, I find the question studied in this work interesting, and compelling. Also, the results were convincing and the analysis through. However, assumptions in the proposed HMM is not fully justified and additional analyses are needed to strengthen authors' claims. To be more specific, the effect of obtained reward on state transitions, and biased exploitations should be further explored.

Reviewer #1 (Public Review): 

In a recent study published in e-Life (Salinas et al., 2019), it was shown, in a speeded anti-saccade task, that cognitive control was temporarily impaired immediately after stimulus onset, resulting in many erroneous saccades directed to visual targets. The generality of this phenomenon and its relationship with attention mechanism were still not fully known. In the present study, the author demonstrates an analogous phenomenon in manual responses. Also, it is shown that a similar phenomenon occurs even when the conflict was created by the incongruency between the central and peripheral stimuli in the Eriken's flanker task. Therefore, the temporary reduction in cognitive control after a stimulus onset might be a general phenomenon independent of stimulus and response modality. The results clearly support the main conclusion of the study.

Reviewer #2 (Public Review): 

This study reports the results of two experiments in which human participants had to evaluate visual stimuli under high urgency conditions (meaning that they had little time to make their choices and often had to guess). Participants responded via button presses. In Experiment 1, the target stimulus indicated a movement to the left or to the right and was itself located to the left or to the right of fixation. In Experiment 2, the target, which again instructed a movement to the left or to the right, was presented at the center of the screen and was flanked by distracter stimuli that pointed either to the same or to the opposite direction. In both cases, the key comparison was between the congruent condition, in which all features pointed to the same response, and the incongruent condition, in which the target and the non-relevant features pointed to different responses. Notably, either of these tasks would be quite easy and mundane in the absence of time pressure; the experiment is novel and informative because urgency makes it possible to accurately track the evolution of the participants' choice over time. 

Indeed, the data yielded performance curves of choice accuracy as a function of processing time (cue viewing time), and the main result was that the curves for incongruent trials were shifted to the right relative to the congruent, and also demonstrated an initial dip to below-chance performance indicative of trials in which the irrelevant features captured attention and evoked erroneous responses. The conclusion from these experiments is that, under high urgency conditions, salient visual stimuli can bias impending motor actions to a much higher degree than in the absence of time pressure. In other words, the cognitive filtering mechanisms that normally mediate how we respond to visual stimuli are transiently interrupted under high urgency. 

The main strengths of the study are: 

- Clear, concise exposition. 

- It generalizes the perceptual capture phenomenon beyond the oculomotor system, to button presses. 

- It generalizes the perceptual capture phenomenon beyond spatial localization, to non-spatial visual features (e.g., shape). 

- It demonstrates that so-called 'urgent tasks' do indeed produce changes in internal state that are consistent with variations in physiological markers of arousal and subjective sense of urgency (e.g., pupil dilation). 

- It suggests that the transient, urgency-enabled alteration of the cognitive mechanisms that normally filter and interpret salient visual stimuli is a general, widespread phenomenon. 

No major weaknesses were spotted. The only minor concern of note was that the language used to describe the results was perhaps a bit misleading. That is, saying that "urgency disrupts cognitive control" suggests some sort of failure or anomaly in sensory processing, whereas my sense is that the phenomenon under study is just part of how perceptual circuits work, and that the transient interruption in top-down control is not a bug, but a design feature that is there for a reason (Salinas and Stanford, Sci Rep, 2018). 

Overall, however, this is a novel contribution that provides deeper insight into an intriguing cognitive phenomenon that is perhaps much more widespread than initially thought.

Public Review (Reviewer #1): 

This study reports measurements of ligand binding on- and off-rates for three different conformations of α4β1 as well as α5β1 integrin. These measurements were carried out with a well characterized set of Fabs that arrest the integrin in different conformational states. In line with their previously published work (Li et al. EMBO J, (2017); Li & Springer PNAS, (2017); Li & Springer, J Cell Biol, (2018), the measured binding kinetics appear to follow a conformational selection model of integrin-ligand binding, whereas the on- and off-rates for the active (EO) integrin binding for ligand is surprisingly slow, thereby potentially explaining the integrin activation trajectory on a cell encountering the ECM. These findings are novel and important. The drawback of the study is that the three different conformational states were stabilized with Fabs, which, although well characterised, may interfere with ligand binding and not necessarily work as expected.

Public Review (Reviewer #2):<br> This manuscript describes a detailed measurement and calculation of integrin ligand-binding kinetics, which are very important for the understanding of integrin activation. The data clearly indicated that low-affinity binding states of closed conformation of integrin bind ligand with the mode of "fast on fast off", while the high-affinity binding of the open conformation results from the much slower of the off-rate. The kinetics measurements were well designed and a lot of work was done in this study.

Public Review (Reviewer #3): 

The manuscript by Li et al. entitled "Low affinity integrin states have faster binding kinetics than the high affinity state" addresses the fundamental question of how the different conformational states of an adhesion receptor control its adhesive properties and thus cell adhesion. In particular, the authors studied the binding kinetics of distinct conformational states of integrins, both on and off rate, and the relative abundance of these different conformational states at the cell surface. 

The manuscript by Li et al. builds on previous articles by the same group, which essentially (among other groundbreaking studies on cell adhesion and in particular integrins) established the structural knowledge on integrins, in particular the fact that integrins can exist in 3 conformational states: the low-affinity bent-closed (BC) and extended-closed (EC) conformations and the high-affinity extended-open (EO) conformation. 

The main messages of the manuscript are: 1. low affinity states of integrins (BC, EC) bind faster than the high-affinity state (EO); 2. the higher affinity of the EO state results from a slower off-rate compare to low affinity BC and EC states; 3. Low affinity integrin states are denser compared to the high affinity integrin state at the cell surface. These results could shed new light on the sequence of molecular events leading to integrin activation/adhesion in the cellular context. In particular, on the relative contribution of the outside-in versus inside-out mechanisms to activate integrins. 

The authors used well-characterized conformation-specific Fab combinations to stabilize integrins (α4β1 and α5β1) in different conformational states and measured ligand binding/unbinding kinetics (on rate and off-rate). They used two experimental configurations: 1. Flow cytometry on intact suspended cells (Jurkat cells), to study α4β1 and α5β1 integrins in their cellular environment; 2. bio-layer interferometry to study an ectodomain fragment of α5β1. 

Using the first experimental configuration, to stabilize the EC and EO conformations of α4β1 integrin they used 9EG7 and to stabilize the EO conformation of β1 integrins they used a combination of 9EG7 and HUTS4Fab. Under basal conditions (BC, EC, EO) association to and dissociation from a ligand (FITC-LDVP) were the fastest, association and dissociation were slower for extended conformations (EC, EO) and even slower when only the EO conformation was present. Then found the same results, association and dissociation were slower when only the EO conformation was present, with VCAM D1D2 binding to integrin α4β1, although they could not measure these parameters for the basal condition since the affinity of α4β1 for VCAM D1D2 is too low. 

Then, they studied the binding of Fn39-10 to α5β1 integrin on K562 cells and found the same results, association and dissociation were slower when only the EO conformation was present compared to when the EC and EO were present. Then they quantified the apparent Kon and Koff assuming a 1 vs. 1 Langmuir binding model. Thus, overall, their results show that the ligand associates and dissociates more slowly to/from the EO conformation than to/from the BC and EC conformations. 

Then using bio-layer interferometry they performed the same analysis on soluble α5β1 ectodomain binding to Fn39-10. In that configuration the authors could also quantify the basal ensemble Fn39-10 binding kinetics, by raising the population of the EO state by truncation and favoring high mannose glycoforms posttranslational modification. The results confirmed that the open EO state associates and dissociates more slowly compared to mixture of the close BC and EC states. 

To measure the off-rate of the closed states, the authors first enabled ligand binding to integrins to reach steady state, and then added closure-stabilizing Fab to measure the dissociation kinetics. By using this strategy, they could measure the koff for the mixture of closed conformations (BC + EC) and also for the extended closed conformation only. In these conditions, the dissociation from the basal and extended α4β1 ensembles on Jurkat cells were similar. They found the similar results with Fn39-10 dissociation from basal or extended ensembles of the α5β1 ectodomain. These results confirmed that the Koff for the closed states (BC + EC) are much faster than for the open state EO. 

Finally, assuming that integrin conformational transition kinetics are sufficiently fast and therefore do not influence the measured kinetics, since integrins and ligand-bound integrins can be considered to equilibrate between their conformational states, the authors used a 1 vs. 1 Langmuir binding model to calculate the ligand binding kinetics from the ensemble measurements. 

Overall, the authors found that integrins α4β1 and α5β1 closed states (BC and EC) have a lower affinity compared to their open states, but closed states have higher on-rates than their EO open states. The higher affinities of the open states compared to the closed conformations for α4β1 and α5β1 integrins lie in a much slower off-rate. These results are supported by published structural data showing that the open conformation of the integrin has a tighter ligand binding pocket than the closed conformations, which creates a steric barrier to ligand binding but also stabilises ligand binding (dissociation barrier). 

The fact that the closed conformations (BC and EC) have a faster Kon compared to the open conformation (EO), could explain how integrins can probe the extracellular matrix (ECM) without the need for the EO state which could be stabilized by interaction with intracellular regulators and the actin cytoskeleton (inside-out). In that case integrin binding to its ligand (outside-in) could initiate an adhesive structure mainly in regions where the actin flow generates enough force to stabilize the transition to the EO state (e.g. lamellipodium, focal adhesions). The fact that BC and EC closed conformations possess faster Koff compared to the EO state is also very interesting. In this scenario, there could be a transient period where integrins have the possibility to unbind their ligand by transition to the EC and BC, before full activation by force generated by the actin cytoskeleton. This will also enable reversible binding of integrins once connection with the actin cytoskeleton is lost, which is consistent with integrin turnover or diffusion-immobilization cycles found in adhesive structures. 

The conclusions of the manuscript are convincingly supported by the results. The authors have performed a very comprehensive characterization of the kinetic parameters of α4β1 and α5β1 integrins association to and dissociation from their ligands. These results could provide a better understanding of the mechanisms that control the binding of integrins to their ligands in adherent cells. In particular, the results could shed light on the sequence of molecular events leading to integrins that are simultaneously bound to their extracellular ligand and connected to the intracellular actin cytoskeleton.

Reviewer #1 (Public Review): 

Zhao et al present a very well written manuscript focused on exploring the effect and mechanism of TFR1 loss in osteoclast lineage cells. This work is unique with no precedent in or expectation of whether and how iron deficiency via TFR1 loss affects bone homeostasis. In the current manuscript, the authors focus specifically on osteoclast-lineage cells and demonstrate decreased osteoclast function via impact on mitochondrial ROS-dependent function and respiration leading to a resorption defect and in vivo increase in bone volume in TFR1 fl/fl-Ctsk-Cre mice. The work is compelling in many ways and yet several comments are warranted for clarification. Specifically, it is unclear what the added value of the LysM-cre model is in the current work and appears to only distract from the main focus of the manuscript, that effects at the mature osteoclast stage are most robust. Furthermore, the authors do not reconcile whether effects on the mitochondrial (Fig 9 and 10) are reversed with the addition of hemin. In addition, it is not clear why TFR1-fl-LysM-cre mice are used rather than TFR1-fl-Ctsk-cre in the in vitro assay performed; it remains a possibility that the results would be altered if the TFR1 loss occurred later in differentiation. Finally, the data on Hem1 overexpression is interesting but all appropriate controls are not included to definitively support a conclusion that TFR1-meadiated effects on osteoclasts are a consequence of destabilizing WRC complex.

Reviewer #2 (Public Review): 

The authors investigated the role of the transferrin receptor 1 (Tfr1)-mediated iron uptake in osteoclasts and its impact on bone homeostasis and found that conditional deletion of Tfr1 in osteoclast lineage cells resulted in increased bone mass primarily in the long bones of female mice. In addition, the authors show that genetic disruption of Tfr1 alters mitochondrial metabolism and the osteoclast cytoskeletal organization. The strength of the manuscript is the novelty i.e. the first to assess the physiological contribution of Tfr1 in osteoclasts using two complementary genetically-modified mouse models that conditionally target osteoclast-progenitor cells and mature osteoclasts. The weakness of the study is that the skeletal phenotyping is restricted to a single time point i.e. 10-weeks of age and mechanistic evidence linking Tfr1-mediated iron uptake to regulation of the osteoclast cytoskeleton remains preliminary. Overall, this study offers the first detailed phenotypic assessment of the skeleton of Tfr1-deficient mice and provides new insights into the importance of iron uptake for osteoclast function during bone turnover.

Reviewer #3 (Public Review): <br> Das et al. aimed to investigate the role of Tfr1 for iron uptake in osteoclasts. They have used two in vivo models to knockout Tfr1 in osteoclast precursors (myeloid cells) and mature osteoclasts and show that Tfr1 is an important iron uptake receptor in mature osteoclasts. In particular, iron is important for mitochondrial function and cytoskeletal reorganization. 

The strength of this investigation is the rigorous use of controls. Also, their use of different ages and Tfr1-knockout strains adds confidence to their conclusions. A weakness is the rather superficial characterization of osteoclastogenesis, which could be done more accurately to really pin-down at which stage of differentiation Tfr1 is particularly required. 

Overall, this study adds important information regarding the role of Tfr1 and iron metabolism in osteoclasts and suggests that targeting iron-related pathways may be a way forward to therapeutically treat bone loss.

Reviewer #1 (Public Review): 

In this manuscript, the authors investigate the role of glutamate-oxaloacetate transaminase 2 (GOT2) in the growth of pancreatic ductal adenocarcinoma (PDAC). Building on previous work demonstrating an important role of cytosolic malic enzyme in PDAC progression, the authors tested the importance of GOT2, an enzyme in the malate-aspartate shuttle, as a therapeutic target. The authors find that GOT2 is required for colony formation of PDAC cells in vitro but is entirely dispensable for growth of both xenograft and autochthonous models of PDAC in mice. A major strength of the manuscript is the multitude of cell lines used, demonstrating the generalizability of these findings in vitro, as well as the generation of novel mouse models for GOT2 perturbation. 

In vitro, GOT2 deficiency can be rescued by conditioned medium from cancer associated fibroblasts (CAF) or exogenous pyruvate. Mechanistically, pyruvate serves to alter redox balance that rescues GOT2 loss. In vitro, perturbing pyruvate import or metabolism does not affect tumor growth. The finding that GOT2 deficiency can be rescued by restoring cytosolic redox adds a new layer of understanding to the role of GOT2 in supporting cancer cell growth. The discovery that CAFs contribute to redox balance in cancer cells represents an important conceptual advance. However, whether maintenance of redox balance underlies the lack of phenotype of GOT2-deficient tumors in vitro remains unclear, as interventions to block pyruvate uptake or conversion to lactate have no impact on tumor growth. While technical issues may underlie these negative results, at present there is a lack of evidence supporting the conclusion that CAFs promote redox balance in vivo. Alternative explanations for the ability of GOT2-deficient tumors to sustain growth are not explored. Therefore, the manuscript at present does not resolve whether CAFs support cancer cell redox balance in vivo or, more broadly, whether redox balance is a constraint on cancer cell growth in vivo, is not addressed.

Reviewer #2 (Public Review): 

Cancer cells frequently display changes to cell metabolism, suggesting that impairments in cancer specific metabolic dependencies may represent a viable path to improve cancer therapy. Work from these authors and others has determined that Kras mutant pancreatic cancer cells may alter the metabolism of TCA cycle associated metabolites and cause a particular dependence on malic enzyme 1 (ME1). The substrate of ME1, malate, is in relation to metabolites and metabolic enzymes in the malate-aspartate shuttle, including through the activity of mitochondrial transaminase enzyme GOT2. In this manuscript Kerk et al. investigate the role for GOT2 in support of pancreatic cancer cell proliferation in culture and in murine tumor models. The authors find that, despite GOT2 loss causing robust proliferation impairments in culture, the same alteration has no significant effect on tumor growth in vivo. Mechanistically, the authors determine that fibroblast conditioned media can cause resistance to GOT2 dependent proliferation defects in PDAC cells in culture, likely because of pyruvate released by fibroblasts that functions as an electron acceptor to promote cell proliferation. Regardless, no significant effect on tumor burden is observed in autochthonous mouse models of GOT2 deficient PDAC, nor does impairing pyruvate uptake synergize with GOT2 loss to prevent growth of PDAC xenografts. 

Overall, the manuscript is well done, with rigorous use of pancreatic cancer cell lines and mouse models, comprehensive metabolic profiling, and strong mechanistic investigations of the role of pyruvate supplementation in GOT2 knockdown cells, however, the conclusions are somewhat disappointing since no clear answer is reached as to how these cells proliferate without GOT2 in tumors. From a writing perspective, I commend the authors for being honest about these limitations and describing possible explanations in earnest. The conclusions reached by the authors are therefore justified by the data. That said, the manuscript ultimately serves as a cautionary tale that the metabolic dependencies identified in culture may not translate to the complex microenvironment of tumors. Nonetheless, the hypothesis was well founded, and the results are highly interpretable, making this manuscript of interest for the scientific community.

Reviewer #3 (Public Review): 

This is a nicely done study and well written article by Kerk and colleagues focusing on the role of GOT2 in pancreatic tumor growth that highlights critical differences between in vitro and in vivo dependencies. Such differences are crucial to understand for effective targeting of metabolic pathways in PDA. The authors show that GOT2 silencing drastically impairs PDA cell colony formation in vitro, but that it is not required for PDA tumor growth in vivo. This led authors to hypothesize that tumor microenvironmental factors may facilitate growth in the absence of GOT2 in vivo. Authors go on to find that conditioned medium from fibroblasts is sufficient to rescue colony formation in GOT2 KD cells, that pyruvate accumulates in CM, and that pyruvate is also sufficient to rescue colony formation. It is shown that PDA cells convert pyruvate into lactate and that mitochondrial import of pyruvate is not required for the rescue. These data lead authors to test whether GOT2 loss disrupts redox homeostasis, finding an increase in the NADH/NAD+ ratio. Other manipulations that boost NAD+ production such as LbNox expression or aKB supplementation partially rescue colony formation. MCT1 KO or inhibition also impairs pyruvate or CM rescue in vitro, though does not impact tumor growth in vivo. Finally, authors generate KC-Got2 mice and show that they have no differences in tumor formation. Overall, the study is well done and conclusions appropriate, although additional controls are warranted. While disappointing that Got2 KO and MCT inhibition together had no effect in vivo, the findings highlight the challenges of targeting tumor metabolism and authors do a good job of discussing the relevant factors that may explain this result. I commend them for transparently reporting the negative in vivo data.

Reviewer #1 (Public Review): 

The title of the manuscript is overly broad and does not reflect the specific experiments presented in this manuscript. Specifically, this manuscript narrowly evaluates the ability of the two factors FOXA1 and HNF4A to either independently or cooperatively activate inaccessible regulatory DNA elements in the K562 human cancer cell line when overexpressed at likely non-physiological levels. Given the narrow scope of the experiments presented in this manuscript, it is disingenuous to state that this manuscript is actually "A Test of the Pioneer Factor Hypothesis", which implies a much broader scope. 

Overall, their data supports the notion that in K562 cells, overexpressed FOXA1 acts as a pioneering factor, and overexpressed HNF4A can similarly act as a pioneering factor. The authors rightly state that the HNF4A results are surprising given its prior status as a non-pioneering factor. However, it may be the case that HNF4A acts as a pioneering factor in K562 cells when overexpressed, but may not have pioneering activity in other tissues or samples, or when expressed to a lesser degree. Similarly, it is possible that HNF4A is cooperatively interacting with other TFs that are endogenously expressed in K562 cells. The authors need to discuss the possible limitations of their experimental setup. 

The authors need to attempt to reconcile their findings with prior studies showing HNF4A as a non-Pioneering factor. What is potentially causing these divergent conclusions? 

Similarly, please clarify what specific data is used to draw the conclusion that "... the mode of action of these targets did not conform to the sequential activity predicted by the PFH." If anything, Figure 4 presents data that a substantial proportion of both FOXA1 and HNF4A elements require cooperative occupancy of both of these factors to be opened. 

Pg. 16, line 313-5. "Pioneer activity may best be summarized then by the free energy balance between TFs, nucleosomes and DNA ... rather than as a property of specific classes of TFs." As discussed above, this conclusion seems overly broad given the data presented only applied to two factors in a single cell line.

Reviewer #2 (Public Review): 

How transcription factors access their DNA binding motifs in chromatin and cooperate with other transcription factors in DNA binding remains a contentious question. It is clear that some transcription factors ("pioneer transcription factors") play a dominant role in opening chromatin during development and reprogramming. But it has also been clear that the ability of such transcription factors to do so lies on a spectrum, that pioneer transcription factors may mutually interact with other transcription factors in their pioneering activity (e.g. Swinstead et al. 2016) and that their mode of binding is still poorly understood (see recent crystal structures). 

This study clearly illustrates our lack of understanding and thus provides impetus for an important discussion. I particularly like that the study uses de novo chromatin accessibility in vivo as a readout for pioneering activity. There has been much effort into measuring the ability to bind nucleosomes in vitro or their behavior in imaging assays in vivo, but ultimately the more relevant assay for understanding enhancer function is the assay the authors performed, and this is not done often enough. Having said that, the manuscript's presentation relies on refuting an overly simplified pioneer factor hypothesis. Furthermore, it is doing so by only highlighting the inconsistencies of a single transcription factor, Hnf4a, without providing new evidence for a more consistent hypothesis. There is insufficient genome-wide evidence to support the suggestion that Foxa1 and Hnf4a's pioneering activity at individual sequences is linked to affinity or sequence context. Overall, this is an important topic and a good experimental approach, but the limited analysis and overly simplified interpretation limits the scope of the discussion.

Reviewer #1 (Public Review):

The authors make juxtacellular recordings on awake mice, which should yield clear responses of actions potentials, and employ a number of manipulations to silence pathways. They also record from a "non"-whisker secondary thalamic region, LP, as a null hypothesis to establish if certain effects are related to "behavior" - read arousal or saliency". I have no major qualms.

In light of Petersen's paper (Cell Reports 2014) on cholinergic effects on spike rates in primary whisker somatosensory cortex, I can imagine that the authors considered measuring from cholinergic neurons in nucleus basalis during whisking. I'll assume that this is easier said than done. As such, the current manuscript passes my threshold for publication modulo issues raised below that are related to anatomy.

I provide a figure-by-figure critique:

(1) Recent work from Deschênes et al (Neuron 2016) points to a description of whisking in terms of Angle = Set-point_angle - Whisking-amplitude [1 + cosine(Phase - Phase_0)], where Phase is a rapidly varying, typically rhythmic function of time. Why not use this notation as opposed to yet another descriptive statistic and report the kinetics as the time averaged parameters <Set-point_angle>, i.e., the most forward position, and ,Whisking-amplitude>, i.e., the half-amplitude of the average whisk?<br> A critical issue is to confirm where the recording were made. This the authors should supply at least a typical record of anatomy from their POm as well as VPM and LP recording. The beauty of the juxtacellular technique is that neurons can be labeling after the recording

(2) Did the authors make sure that the mystacial pad is not moving by imaging the pad as opposed to just the shaft of the whiskers? The top view in Figure 1A makes this hard to check. Further, did the authors perform post-hoc anatomy to insure that both the ramus buccolabialis inferior and ramus buccolabialis superior muscles were cut? This is critical; it is also easy to leave the maxillolabialis (external retractor) innervated if the cut is too far rostral.

(3/4) As relevant background, the text should note that whisker primary motor cortex maintains a copy of the envelope of the whisking, i.e., an ill-defined summation of set-point and amplitudes, even if the sensory input (Ahrens & Kleinfeld J Neurophysiol 2004) or motor output (Fee et al. J Neurophysiol 1997) in the periphery are cut.

(6/7) Same comments in (1) in whisking parameters and anatomy.

Reviewer #2 (Public Review):

The authors are attempting to reveal the nature and source of state-dependent activity in a somatosensory nucleus of the mouse. They observe that PoM exhibits activity related to the slow components of whisker movements. Block of reafferent singles from the face, by nerve-cut induced paralysis, optogenetic inhibition of somatosensory cortex, or electrolytic lesion of the superior colliculi are all unable to block the whisker-related movement signal in PoM. The authors find similar state dependent activity in the LP nucleus, which is a higher level visual thalamic nucleus. They suggest that the activity is more related to arousal than movement, per se.

This is a timely study in that many labs have observed state-dependent activity throughout the cortex and thalamus, but the mechanisms of this activity are incompletely understood. Since this activity accounts for a large fraction of the activity of the forebrain in awake mice, it is a priority to understand the mechanisms of its generation. This study brings us closer to revealing the source of this signal.

Unfortunately, the authors do not discover the source of the arousal/movement signal - but give evidence against some of the more popular/likely candidates.

Reviewer #3 (Public Review):

Previous studies in urethane-anesthetized rats (PMID 16605304) proposed that POm cells code whisker movements. This was observed using "artificial whisking" procedures (stimulating the motor nerve to produce a whisking-like movement). It has been clear for some time now that there are substantial (obvious) differences between this procedure and natural whisking. In addition, under urethane-anesthesia animals are in a sleep-like state that is very dissimilar to waking (although some work has tested the effect of network state on artificial whisking responses in both primary thalamus and cortex; see 25505118). In the present study, the authors measured activity in POm cells during whisking in awake (head-fixed) mice to determine if they code whisking movement. However, this seems to have already been done previously. For instance, Moore et al (2015; 26393890) found that coding of whisking in the ascending paralemniscal pathway, including POm, is "relatively poor" (as stated in the abstract), which is the same conclusion reached in the present study. The authors should clarify the main differences observed in whisking coding between their study and previous work.

The authors then focused on the idea that POm codes behavioral state. However, many studies have previously determined that state has a great impact on thalamocortical dynamics; thalamic cells are very sensitive to state including cells in primary whisker thalamic nuclei, such as VPM, and these effects can be produced by neuromodulators (see work by Castro-Alamancos' group, for example, 16306412). There is nothing special about VPM in this regard; other thalamic sensory nuclei are also sensitive to behavioral state and neuromodulators. Therefore, the observation that POm and LP cells are sensitive to state is unsurprising. It is also known that these thalamic state changes have a great impact on the state of the cortex (see 20053845), which seems very relevant to the main conclusion. The POm has to be doing something different than coding behavioral state since most thalamic nuclei do this. The study did not identify the role of POm, which certainly has to be different from LP (otherwise, why would these nuclei be differentiated?). POm is unlikely to be specialized for monitoring state since this is done by most of the thalamus -including VPM, which projects to the same cortical region. Thus, while it is interesting that most of the whisker-related activity in POm is state-dependent, the study does not clarify the role of POm.

The main strength of the study is that it was performed in awake mice with behavioral state monitoring, which contributes to the current understanding of active whisking coding in the complex network of the vibrissa system.

Reviewer #1 (Public Review):

Cytoplasmic germ granules are a common feature of germ cells across species. Great effort has gone into trying to (1) identify proteins that localize to these condensates, (2) gain insights into how these granules assemble, and (3) characterize their functional significance. Pioneering studies on P granules in C. elegans have greatly expanded our understanding of these structures.

In this paper, Price et al. use proximately labeling to identify previously unknown P granule components. They tag several known P granule protein genes with TurboID at their endogenous loci and perform biotinylation reactions in two different ways. Labeled proteins were subsequentially pulled down and identified using mass spectrometry. This approach successfully identified known P granule components and many new potential candidates. The authors focus their efforts on characterizing two related proteins EGGD-1 and EGGD-2 (also known as MIP-1 and MIP-2, based on recently published work). Knockdown of EGGD-1 and EGGD-2 using RNAi results in P granule defects. Cas-9 induced mutations confirm and extend this genetic analysis. The authors also perform structure/function analysis on EGGD-1 and define specific roles for its LOTUS and Intrinsically Disordered Region (IDR) domains in perinuclear P granule formation and function. Epistasis analysis shows that EGGD-1 acts upstream of GLH-1 in P granule assembly, and overexpression of EGGD-1 can drive granule formation outside of germ cells.

Strengths

The P granule proteome data presented here provides a useful resource for the community. The genetic analysis on eggd-1 provides important insights into the function of a new P granule component. The data are convincing, and the experiments are well-controlled.

Weaknesses

There are relatively few weaknesses and the general conclusions are supported by the data.

Reviewer #2 (Public Review):

In this manuscript from Price et al, proximity labeling is used to define the proteome of the C. elegans P granule. This is an ideal method to use for this experiment because of the phase-separated nature of P granules, meaning that a traditional co-immunoprecipitation may not work to identify transiently interacting proteins. While many of the interacting proteins identified were previously known (validating the proximity labeling approach), the authors focused on two previously uncharacterized proteins, EGGD-1 and -2 (also known as MIP-1 and -2) which they find to disrupt PGL-1 and GLH-1 localization by RNAi and knockout mutants. The authors then proceed to do a careful structure-function analysis of EGGD-1 - demonstrating that the Lotus domains are important for recruitment of other P granule proteins while the IDRs are important for EGGD-1 localization to the nuclear periphery. Lastly, the manuscript demonstrates that EGGD-1 is sufficient to form perinuclear granules in somatic cells, and ectopically recruit GLH-1 to these granules. These data suggest that EGGD-1 directly interacts with nuclear pores (or at least nuclear pore associated proteins found in somatic cells) and does not require any other germline-specific P granule proteins for its perinuclear association. This is a nicely designed paper that is thorough and rigorous. There are no major weaknesses.

Reviewer #3 (Public Review):

The manuscript of Price et al. used TURBO-ID of multiple P granule components to identify new factors required for their assembly and function. They identify over 75 shared proteins, including 2 related TUDOR-domain proteins that they analyze in further detail through mutation and localization studies. The two proteins are necessary for the localization of several P granule components to the nuclear periphery, and they show in an ectopic tissue that EGGD-1 is sufficient to localize GLH-1 to the nuclear periphery.

While the paper could go further to test physical interactions between EGGD-1 (and EGGD-2) with GLH-1 and the nuclear pore protein, this is not critical, although the authors should be cautious in their model that they have not proven that these associations are direct.

It would be important to provide evidence that the STOP-IN cassette in eggd-2 is a true null. There may be downstream methionines that can be used. And the phenotype is weaker than eggd-1 so this could be an issue.

All of the TURBO-ID strains have reduced viability. Since there is some concern that P granule composition could be affected in the tagged strains, showing the localization of other known components of P granules in these mutants (GLH-1, PGL-1) would be critical.

Reviewer #1 (Public Review):

Markiewicz et al. describe the principles, history and technological building blocks of the OpenNeuro neuroimaging data sharing platform. In addition, Markiewicz et al., provide data on OpenNeuro growth in terms of data uploaded, data downloaded and publications re-using OpenNeuro data, all of which are extremely impressive. They convincingly argue for the importance of minimizing data sharing restrictions, while rigorously maintaining data safety. They describe the origins of the BIDS standard and provide arguments for its critical importance to data sharing. They conclude with discussing OpenNeuro's limitations, the funding challenges faced in maintaining it and future directions. The article makes it clear that OpenNeuro and the accompanying BIDS standard are critically important for the success of neuroimaging research and that without it, neuroimaging will not, cannot progress. In summary, this article describing OpenNeuro and BIDS provides fuel for the belief that neuroimaging research will emerge from its reproducibility crisis, strengthened, modernized and more open.

Reviewer #2 (Public Review):

• This work sets the standard for data sharing and management resources in neuroimaging and could become the dominant standard resource in the field. Challenges to this aspiration are the high degree of decentralization and poor present organization, large size of these datasets. The authors have profited from Amazon's program of freely sharing not for profit public data sets, but this will scale as OpenNeuro becomes for widely used is less clear.<br> • Comparing with other major imaging databases such as ADNI this resource is clearly the direction the field should go. The authors and developers have taken a truly open science approach.<br> • The BIDS standard is important and as part of the backbone of OpenNeuro perhaps somewhat more description of how it is used in OpenNeuro would be important and interesting to readers, particularly that thousands of datasets even outside of this resource use it.<br> • The description and implementation FAIR principles is well described and clear, and important part of paper.

Reviewer #3 (Public Review):

This is a well-written paper describing the OpenNeuro data archive. OpenNeuro covers diverse mesoscale brain imaging data (fMRI, PET, etc) from many projects and including multiple experimental paradigms. The focus is on human imaging, but primate and rodent data are also represented. The project rests on a standardized and relatively mature Data format (BIDS) for imaging data. This is key to implementing FAIR principles and automated checking for compliance. OpenNeuro is already producing discoveries that could not have been made without meta-analysis across diverse data sets. OpenNeuro is also used to improve data analysis pipelines. By its nature, this kind of paper always reads a bit like an advertisement.

The paper makes a compelling case for domain-specific repositories supported by a modern cloud architecture and sound computer science. I appreciate the discussion of the challenges that have been overcome (e.g. versioning of data sets; privacy and consent) and others that are looming (e.g. long-term maintenance in the absence of obvious commercial drivers).

I would like to see a bit more comparison to other archives, including some mature projects in other fields (e.g. astronomy, IVOA; structural molecular biology; CCDC), as well and more nascent efforts in brain research (e.g. DANDI; BIL).

Although OpenNeuro is a best-in-class archive with huge potential impact, similar archives operating with similar principles have operated for a while in other fields (e.g. astronomy; genomics etc). I thus wonder if the paper is of 'general interest'; i.e. of interest outside of the worlds of neuroimaging and those interested in data archives & data sharing in general. OpenNeuro is a pioneer in brain research and can function as a model for other subfields in neuroscience.

Reviewer #1 (Public Review):

The habenula is a remarkable asymmetrical structure in the vertebrate brain whose integrative functions are diverse and not well understood. The authors have identified an interesting subset of dorsal hanenular neurons expressing lratd2a and localized only on the right side of the body that receive inputs from the olfactory bulb, project to the ventral interpeduncular nucleus (vIPN). By combining sophisticated genetics, calcium imaging and behavior in adult zebrafish, the authors argue that an asymmetric dorsal habenula - ventral interpeduncular nucleus pathway is involved in processing aversive cues and mediates avoidance responses to olfactory cues.

Reviewer #2 (Public Review):

In this manuscript, Choi et al combine new intersectional genetics and CRISPR-mediated knock-in strategies to dissect the role of an interesting population of lecithin retinol 23 acyltransferase domain containing 2a (lratd2a)-expressing cholinergic neurons in the right dorsal habenula. They show that these neurons are responsive to aversive odorant cues such as cadaverine, and that silencing these neurons specifically reduces avoidance behavior to this odorant. The authors also use mutant lines to explore the role of habenular asymmetry in aversive responses, which provide additional supporting evidence for the hemispheric specialization of behaviors. However, given the pleiotropic effects of such mutations, the results from these mutants are more difficult to interpret.

The genetic manipulations and anatomical characterization are excellently done, behavioral results sound, and the manuscript well-written. I do not believe any additional experiments are needed for publication, but will make some recommendations in terms of data analysis and improving clarity in text and figures, to facilitate better comparisons across the different experiments and odorant cues (cadaverine vs alarm substance).

1. Presentation, analysis, and discussion of calcium imaging results

a) As the authors correctly pointed out, having a water control is indeed essential for interpreting calcium imaging results. As such, I would recommend having a water control panel (currently in Figure S3) in the main figure.

b) The current presentation and analysis of calcium imaging data in Figure 2B does not seem appropriate and can be improved - since the dynamics of olfactory responses are likely highly variable across neurons and fish, rather than comparing responses across time, it would be better to compare the summed response over a longer time window (as already done in Figure S2, but also including water flow control data). Do also mention the time window over which the calcium responses were integrated.

c) Discussion Line 393: "From calcium imaging, we validated that the right dHb appears more responsive than the left when larval zebrafish are exposed to aversive odors such as cadaverine or chondroitin sulfate" - this conclusion cannot be drawn from the existing presented data, unless calcium imaging was also performed in the left habenula.

d) It would be good to include in the methods section more detail on how the odor was delivered, volume delivered etc, and whether control experiments were done on the same day / clutch of fish etc.

2. Presentation, analysis, and discussion of c-Fos results and comparison with calcium imaging

a) Figure 2C-D: The difference / overlap between blue and brown are difficult to make out in the images, especially at this resolution and magnification. Is there a way to specifically quantify the % of lratd2a neurons that are activated by c-fos, rather than just neurons in the dorsal habenula as a whole? This would be necessary to support the claim in line 278: "Thus, the lratd2a subpopulation in the right dHb responds to cadaverine in both larvae and adults".

b) In larvae (using calcium imaging), the effects of cadaverine to chondroitin sulfate were compared, whereas in adults (using c-Fos), the comparison is between cadaverine and alarm substance. Is there a reason why the alarm substance was not used in larvae, or chondroitin on adult fish? Perhaps the authors can elaborate on their rationale.

3. Presentation, analysis, and discussion of behavioral results

a) The presentation of the alarm substance behavior results could be improved. The authors could include the words "alarm substance" somewhere in the panels so it is clear to the readers that they are looking at responses to that rather than to cadaverine which is described in the preceding panels. Similarly, to avoid confusion and to facilitate comparison, the same parameters should be presented for Figures 3-5 (currently distance in top is not shown in Figure 3 or 5, onset of fast swim and interval time not shown in Figures 4-5).

b) Does cadaverine induce changes in swim speed and other kinematic parameters? Similarly, does the alarm substance induce avoidance of one side of the tank like cadaverine? It can be difficult for the reader to compare the effects of genetic manipulations on responses to both odorants since different behavioral parameters are being quantified, hence some means of direct comparison could be helpful.

c) The effect of cadaverine on control groups seems to be quite variable. In Figures 3 and 4 the avoidance effect persists the entire duration of the experiment. In Figures 5 and S4 the effect is only significant in 2 time bins. The authors' conclusions are still valid since the correct comparisons are indeed to their respective sibling controls, however it does make it a bit difficult to compare results across genotypes. For example, non-botox-expressing lratd2a:QF2 fish appear to have about the same degree of cadaverine avoidance as lratd2a:QF2, scl5a7a:Cre, QUAS:Botox fish. Similar to point (b), are there other parameters that can be measured that are more consistent in controls across genotypes? Or at the least, some discussion of the behavioral variability in the text.

d) tcf7l2 mutants (like bsx mutants) also have a significantly lower swim speed than controls, this is also worth mentioning / discussing in the text.

e) The link between habenular LR asymmetry and aversive behavior is indeed interesting - in the discussion, one proposal was that this asymmetry could promote directed turning and escape. From the existing data (particularly for the lratd2a:QF2, scl5a7a:Cre, QUAS:Botox fish), is there any evidence of differences in turning behavior (LR asymmetry, or probability of turns in general)?

f) As a related point, it is not clear to me that one would expect an enhancement of cadaverine avoidance in bsx mutants, especially if the argument is that asymmetry is important for aversive behavior. Perhaps the discussion on this point could be framed less as a negative result but as a notable observation.

4. Statistical analyses: Unless data is normally-distributed, non-parametric tests should be used on calcium and behavioral imaging data (e.g. Kruskal-Wallis for time course calcium / behavioral data, Wilcoxon Rank-Sum Test for others)

Reviewer #3 (Public Review):

The authors use genetic approaches to label, monitor and manipulate a specific set of cholinergic neurons in the right dorsal habenula of zebrafish that connect to a territory of the interpeduncular nucleus. The original question and the combination of the approaches represent clear strengths of this study. The manuscript would benefit from a better description of the data, especially the calcium imaging experiment lack a bit of clarity. The conclusions provided by the authors are based on the data provided. I find the discussion session well thought as puts the study in the context of the published literature in a variety of model systems.

Reviewer #1 (Public Review): 

The paper uses a microfluidic-based method of cell volume measurement to examine single cell volume dynamics during cell spreading and osmotic shocks. The paper successfully shows that the cell volume is largely maintained during cell spreading, but small volume changes depend on the rate of cell deformation during spreading, and cell ionic homeostasis. Specifically, the major conclusion that there is a mechano-osmotic coupling between cell shape and cell osmotic regulation, I think, is correct. Moreover, the observation that fast deforming cell has a larger volume change is informative. 

The authors examined a large number of conditions and variables. It's a paper rich in data and general insights. The detailed mathematical model, and specific conclusions regarding the roles of ion channels and cytoskeleton, I believe, could be improved with further considerations. 

Major points of consideration are below. 

1) It would be very helpful if there is a discussion or validation of the FXm method accuracy. During spreading, the cell volume change is at most 10%. Is the method sufficiently accurate to consider 5-10% change? Some discussion about this would be useful for the reader. 

2) The role of cell active contraction (myosin dynamics) is completely neglected. The membrane tether tension results, LatA and Y-compound results all indicate that there is a large influence of myosin contraction during cell spreading. I think most would not be surprised by this. But the model has no contribution from cortical/cytoskeletal active stress. The authors are correct that the osmotic pressure is much larger than hydraulic pressure, which is related to active contraction. But near steady state volume, the osmotic pressure difference must be equal to hydraulic pressure difference, as demanded by thermodynamics. Therefore, near equilibrium they must be close to each other in magnitude. During cell spreading, water dynamics is near equilibrium (given the magnitude of volume change), and therefore is it conceptually correct to neglect myosin active contraction? BTW, 1 solute model does not imply equal osmolarity between cytoplasm and external media. 1 solute model with active contraction was considered before, e.g., ref. 17 and Tao, et al, Biophys. J. 2015, and the steady state solution gives hydraulic pressure difference equal to osmotic pressure difference. 

3) The authors considered the role of Na, K, and Cl in the model, and used pharmacological inhibitors of NHE exchanger. I think this part of the experiments and model are somewhat weak. I am not sure the conclusions drawn are robust. First there are many ion channels/pumps in regulating Na, K and Cl. The most important of which is NaK exchanger. NHE also involves H, and this is not in the model. The ion flux expressions in the model are also problematic. The authors correctly includes voltage and concentration dependences, but used a constant active term S_i in SM eq. 3 for active pumping. I am not sure this is correct. Ion pump fluxes have been studied and proposed expressions based on experimental data exist. A study of Na, K, Cl dynamics, and membrane voltage on cell volume dynamics was published in Yellen et al, Biophys. J. 2018. In that paper, they used different expressions based on previously proposed flux expressions. It might be correct that in small concentration differences, their expressions can be linearized or approximated to achieve similar expressions as here. But this point should be considered more carefully.

Reviewer #2 (Public Review): 

The work by Venkova et al. addresses the role of plasma membrane tension in cell volume regulation. The authors study how different processes that exert mechanical stress on cells affect cell volume regulation, including cell spreading, cell confinement and osmotic shock experiments. They use live cell imaging, FXm (cell volume) and AFM measurements and perform a comparative approach using different cell lines. As a key result the authors find that volume regulation is associated with cell spreading rate rather than absolute spreading area. Pharmacological assays further identified Arp2/3 and NHE1 as molecular regulators of volume loss during cell spreading. The authors present a modified mechano-osmotic pump and leak model (PLM) based on the assumption of a mechanosensitive regulation of ion flux that controls cell volume. 

This work presents interesting data and theoretical modelling that contribute new insight into the mechanisms of cell volume regulation.

Reviewer #3 (Public Review): 

The study by Venkova and co-workers studies the coupling between cell volume and the osmotic balance of the cell. Of course, a lot of work as already been done on this subject, but the main specific contribution of this work is to study the fast dynamics of volume changes after several types of perturbations (osmotic shocks, cell spreading, and cell compression). The combination of volume dynamics at very high time resolution, and the robust fits obtained from an adapted Pump and Leak Model (PLM) makes the article a step-forward in our understanding of how cell volume is regulated during cell deformations. The authors clearly show that: 

-The rate at which cell deforms directly impacts the volume change 

-Below a certain deformation rate (either by cell spreading or external compression), the cells adapt fast enough not to change their volume. The plot dV/dt vs dA/dt shows a clear proportionality relation. 

-The theoretical description of volume change dynamics with the extended PLM makes the overall conclusions very solid. 

Overall the paper is very well written, contains an impressive amount of quantitative data, comparing several cell types and physiological and artificial conditions. 

My main concern about this study is related to the role of membrane tension. In the PLM model, the coupling of cell osmosis to cell deformation is made through the membrane-tension dependent activity of ion channels. While the role of ion channels is extensively tested, it brings some surprising results. Moreover, the tension is measured only at fixed time points, and the comparison to theoretical predictions is not always as convincing as expected: when comparing fig 6I and 6J, I see that predictions shows that EIPA (+ or - Y27), CK-666 (+ or - Y27) and Y27 alone should have lower tension than in the control conditions, and this is clearly not the case in fig 6J. But I would not like to emphasize too much on those discrepancies, as the drugs in the real case must have broad effects that may not be directly comparable to the theory. 

But I wonder if the authors would have a better time showing that the dynamics of tension are as predicted by theory in the first place, as comparing theoretical predictions with experiments using drugs with pleiotropic effects may be hazardous. 

Actually, a recent publication (https://doi.org/10.1101/2021.01.22.427801) shows that tension follows volume changes during osmotic shocks, and overall find the same dynamics of volume changes than in this manuscript. I am thus wondering if the authors could use the same technique than describe in this paper (FLIM of flipper probe) in order to study the dynamics of tension in their system, or at least refer to this paper in order to support their claim that tension is the coupling factor between volume and deformation.

Reviewer #2 (Public Review): 

This manuscript sets out to determine the previously uncharacterized relationship between rotavirus (RV) infection and m6A modification during intestinal infection by using a collection of both in vivo and in vitro models. The authors found m6A modification inversely correlates with resistance to RV. The increased resistance to RV is due to elevated anti-viral IFN responses in a host with diminished m6A. They investigate mechanism by RNASeq and m6A-RIP-Seq and identified IRF7 as the proposed major mediator of this anti-viral response in the m6A deficient setting. During RV infection, the virus up-regulates m6A modifications to limit the IFN response, potentially targeting m6A eraser ALKBH5 for degradation by viral NSP1 protease. 

The primary finding of this manuscript is novel and intriguing, as it broadens the role of this important type of RNA modification. However, there are concerns as to whether the mechanistic conclusions are sufficiently supported. Given the central role of IRF7 in the interferon response, it is perhaps not surprising that removing IRF7 reverses the resistance to RV displayed by mettl3 mutant mice. As the authors mention in their discussion, m6A modification of RV RNA, endogenous retroviral elements, and IFN transcripts can all contribute to the decreased anti-viral response conferred by this modification. IRF7 is an ISG itself and could mediate these other mechanisms.

Reviewer #1 (Public Review): 

This important research supports the idea that adverse experiences during childhood can have lasting impact on people through to adulthood. This study has some important strengths, including the relatively large sample size and population-based sample, although is limited in that the data reported here were collected at only one point in time and relied on the memory of participants for collecting data. The research has important implications for health and social policy in protecting the interests of children, particularly those who are vulnerable and at greater risk of adverse experiences during childhood.

Reviewer #2 (Public Review): 

Hilda Björk Daníelsdóttir et al. demonstrated the relationship between adverse childhood experiences (ACEs) and adult resilience, measured as perceived coping ability and psychiatric resilience. While this was already done for specific single types and subsets of ACEs, no study so far could show these associations for the full spectrum of all ACEs. 

The large dataset of Icelandic women from the ongoing Stress-And-Gene-Analysis (SAGA) cohort gives the opportunity for an in-depth analysis, controlling for sociodemographic variables like age group, income, civil status and additionally, childhood deprivation, perceived social support, sleep quality, binge drinking and traumatic events. Furthermore, the authors could provide information about correlations (if present) between ACEs as well as between perceived coping ability, psychiatric resilience and between perceived coping ability and different tools to obtain the mental health status. 

Overall, the paper is well-structured, only descriptive data and associations of the main outcomes are reported as tables within the paper whereas all further results are presented within the supplement. Additionally, the statistical methods like Chi-square tests, rank order correlations and linear models are well known, which additionally is helping to address the paper to a wider audience. The large dataset allowing control for many variables and their possible correlations as well as for correlation between all ACEs, reducing "noisy" effects of small sample sizes. 

In the end, the authors succeed identifying the relationship between women without ACEs versus women with one or more ACEs as well as the cumulative effects of ACEs on adult resilience. In addition to the importance of the unique results, the paper it is not too "technical" and well written. Therefore, the paper has the potential to communicate the results outside of the "academic sphere" without too many changes in the main text. In this way, it can make an important contribution to raising awareness of the topic in public.

Reviewer #3 (Public Review): 

Daníelsdóttir and colleagues, using data from an ongoing cohort study of Icelandic women with the objective of studying the associations of ACEs with two measures of adult resiliency: an established questionnaire measure of psychological resilience and a measure based on psychiatric outcomes. One of the key findings of this study was that there was that resilience decreased as ACEs accumulated. 

Strengths: The data they used are from nearly 20000 women exposed to at least one traumatic event and complete data on outcomes and exposures. In addition to this large data set, this study had a large number of ACEs. Having a validated measure of psychological resilience is another strength of this study. 

Weaknesses: The authors were not very successful in conceptualising resilience. Although Connor-Davidson scale is well known and can be introduced as such, in this paper there is a new outcome-based measure was introduced, which require laying a strong foundation on the concept of resilience. Not engaging with literature on resilience from the likes of Garmezy, Rutter, Masten, Werner, and others might have contributed to the conceptual weakness. An example might be that the authors refer to Connor- Davidson measure as perceived coping ability while the original authors meant their scale as a measure of psychological resilience in the face of perceived stress. 

Although operationalisation of resilience varies, there is a common understanding that resilience is flourishing despite adversity. In this conceptualisation, establishing the role of an exposure as adversity is of paramount importance. In this paper, the psychiatric resilience is operationalised as reduction of psychiatric morbidity in the presence of trauma. However, the role of childhood events as adversities for the psychiatric morbidity cannot be established in this data where everyone has faced at least one trauma, thus leading to the absence of a counterfactual.

Reviewer #1 (Public Review):

This manuscript presents a study on the effects of dopamine on the interactions between model-based (MB) and model-free (MF) systems. Sixty-two subjects were tested on a variant of the classic two-step task after receiving 150 mg levodopa or placebo, in a double-blind fashion. The results show that dopamine has no effects on MF learning or MB inference. However, dopamine modulates interactions between MB and MF systems, such that MF credit assignment that requires retrospective MB inference is enhanced. Additional analyses using computational modeling confirm these basic behavioral findings. Finally, the authors show that some of the effects are related to working memory.

Strengths

1. The study addresses a timely and interesting question about interactions between MB and MF learning, and how dopamine contributes to this interaction. The experiment is well-designed and rigorous (within-subject, double-blind, placebo-controlled) and the sample is large, leading to robust results.

2. The experimental task is a real strength of the study. It offers a variety of conditions in which the effects of MF credit assignment and MB inference, and the effects of retrospective MB inference on MF credit assignment can be isolated. In particular, the latter effect is evident in a higher probability to repeat inferred selected choice option (GN) after a rewarded informative state, relative to the inferred non-selected option (GR). This difference shows a significant group difference (group by GN vs GN interaction). These findings are further backed up by a computational model, revealing no effect of dopamine on either MB or MF learning, but an effect of MF credit assignment guided by retrospective MB inference.

Weaknesses

1. The authors present evidence that dopamine enhances MF credit assignment guided by retrospective MB inference for rewards at informative locations (Figure 3B), but not for non-informative locations (Figure 3D), at least not in the model-agnostic analysis. It is somewhat unclear why dopamine would modulate MF credit assignment guided by retrospective MB inference only in some conditions (although it should be noted that the computational modeling results show no difference between informative and non-informative locations).

2. The authors show that individual differences in the effect of dopamine on MB inference are negatively correlated with the effects of retrospective MB inference on MF credit assignment (depending on working memory). This in itself is a surprising finding which the authors interpret as competition between these two forms of MB inference. However, they do not show an effect of dopamine on MB inference as such, which is also surprising given previous evidence.

3. The logic of the different sets of the analysis is complex. This makes reading the paper a bit complicated and sometimes hard to understand.

Reviewer #2 (Public Review):

In this work, Deserno, Moran and others have studied the effects of dopamine on the credit assignment problem in reinforcement learning (RL). In a placebo-controlled within-subject design authors have tested the effects of levodopa on credit assignment using a task that has been introduced recently by the authors (Moran et al. 2019). There are two types of trials in this task, standard trials and uncertainty trials. In standard trials, participants perform a two-step decision task (similar to Daw et al, 2011; but with important differences), in which multiple choices might be presented as a pair in the first step. In the uncertainty trials, participants witness the outcome of a "ghost" choice, and their knowledge of task allows them to update value of each task (i.e. credit assignment).

This study addresses an important and overlooked problem in the decision neuroscience literature. The pharmacological design and task are well-designed, the manuscript is well-written and analyses sound rigorous. However, I believe, based on these results, that the interpretation of data that dopamine influences credit assignment through MB inference (hence the title and abstract) is not justified.

I believe that the most important effect in this study is the one presented in Figure 3C and 3D, which authors have called the "Clash condition". In this one, the same chosen pair on the preceding uncertainty trial is presented in a standard trial and subjects are asked to choose between the two choices. This is, I believe, is the ultimate test trials in the study; and there is no significant effect of drug in those trials. Looking at Table S2, it seems to me that authors have done a very good job in increasing the within-subject power for that trial type (mixed-effect df for the clash trials is 4861; the df for the repeat/switch trials is 239 according to Table S2). Related to this point, authors have found no significant effect of DA on credit assignment in their computational modeling analysis.

- I don't think that authors statement of Cools' theory in terms of DA synthesis capacity is correct (page 17). I believe the main prediction, based on that theory, or at least its recent form, is that DA effect is baseline-dependent and therefore it is actually quite consistent with what authors found. Based on this theory, WM-span test is a good "proxy" of baseline dopamine synthesis capacity. I suggest to revise that part of the discussion.

- I believe that a note on how implication of this study to psychiatric disorders that are related to dopamine would be of interest for many readers.

Reviewer #3 (Public Review):

This paper reports that levodopa administration to healthy volunteers enhances the guidance of model-free credit assignment (MFCA) by model-based (MB) inference without altering MF and MB learning per se. The issue addressed is fascinating, timely and clinically relevant, the experimental design and analysis strategy (reported previously) are complex, but sophisticated and clever and the results are tantalizing. They suggest that ldopa boosts model-based instruction about what (unobserved or inferred) state the model-free system might learn about. As such, the paper substantiates the hypothesis that dopamine plays a role specifically in the interaction between distinct model-based and model-free systems. This is really a very valuable contribution, one that my lab and I expect many other labs had already picked up immediately after it appeared as a preprint.

Major strengths include the combination of pharmacology with a substantial sample size, clever theory-driven experimental design and application of advanced computational modeling. The key effect of ldopa on retroactive MF inference is not large, but substantiated by both model-agnostic and model-informed analyses and therefore the primary conclusion is supported by the results.

The paper raises the following questions.

What putative neural mechanism led the authors to predict this selective modulation of the interaction? The introduction states that "Given DA's contribution to both MF and MB systems, we set out to examine whether this aspect of MB-MF cooperation is subject to DA influence." This is vague. For the hypothesis to be plausible, it would need to be grounded in some idea about how the effect would be implemented. Where exactly does dopamine act to elicit an effect on retroactive MB inference, *but not* MB learning per se? If the mechanism is a modulation of working memory and/or replay itself, then shouldn't that lead to boosting of both MB learning as well as MB influences on MF learning? Addressing this involves specification of the mechanistic basis of the hypothesis in the introduction, but the question also pertains to the discussion section. Hippocampal replay is invoked, but can the authors clarify why a prefrontal working memory (retrieval) mechanism invoked in the preceding paragraph would not suffice. In any case, it seems that an effect of dopamine on replay would also alter MB choice/planning?

A second issue is that the critical drug effects seems somewhat marginally significant and the key plots (e.g. Fig3b and Fig 44b,c, but also other plots) do not visualize relevant variability in the drug effect. I would recommend plotting differences between LDopa and placebo, allowing readers to appreciate the relevant individual variability in the drug effects.

Third, I do wonder how to reconcile the lack of a drug x common reward effect (the lack of a dopamine effect on MF learning) as well as the lack of a drug effect on choice generalization with the long literature on dopamine and MF reinforcement and newer literature on dopamine effects on MB learning and inference. The authors mention this in the discussion, but do not provide an account. Can they elaborate on what makes these pure MB and MF metrics here less sensitive than in various other studies, and/or what are the implications of the lack of these effects for our understanding of dopamine's contributions to learning?

Fourth, the correlation with WM and drug effect on preferential MBCA for non-informative but not informative destination is really quite small, and while I understand that WM should be associated with preferential MBCA under placebo, it does not become clear what makes the authors predict specifically that WM predicts a dopa effect on this metric, rather than the metric taken under placebo, for example.

A fifth issue is that I am not quite convinced about the negative link between dopamine's effects on MBCA and on PMFCA. The rationale for including WM, informativeness as well as DA effects on MBCA in the model of DA effects on PMFCA wasn't clear to me. The reported correlation is statistically quite marginal, and given that it was probably not the first one tested and given the multiple factors involved, I am somewhat concerned about the degree to which this reflects overfitting. I also find the pattern of effects rather difficult to make sense of: in high WM individuals, the drug-effects on PMFCA and MBCA are negatively related for informative and non-informative destinations. In low WM individuals, the drug-effects on PMFCA and MBCA are negatively related for informative, but not non-informative destinations. It is unclear to me how this pattern leads to the conclusion that there is a tradeoff between PMFCA and MBCA. And even if so, why would this be the case? It would be relevant to report the simple effects, that is the pattern of correlations under placebo separately from those under ldopa.

More generally I would recommend that the authors refrain from putting too much emphasis on these between-subject correlations. Simple power calculation indicates that the sample size one would need to detect a realistically small to medium between-subject effect (that interacts with all kinds of within-subject factors) is in any case much larger than the sample size in this study.

Another question is how worried should we be that the critical MB guidance of MFCA effect was not observed under placebo (Figure 3b)? I realize that the computational model-based analyses do speak to this issue, but here I had some questions too. Are the results from the model-informed and model-agnostic analyses otherwise consistent? Model-agnostic analyses reveal a greater effect of LDopa on informative destination for the ghost-nominated than the ghost-rejected trials and no effect for noninformative destination. Conversely model-informed analyses reveal a nomination effect of ldopa across informative and noninformative trials. This was not addressed, or am I missing something? In fact, regarding the modeling, I am not the best person to evaluate the details of the model comparison, fitting and recovery procedures, but the question that does rise is, and I would make explicit in the current paper how does this model space, the winning model and the modeling exercise differ (or not) from that in the previous paper by Moran et al without LDopa administration.

Finally, the general story that dopamine boosts model-based instruction about what the model-free system should learn is reminiscent of the previous work showing that prefrontal dopamine alters instruction biasing of reinforcement learning (Doll and Frank) and I would have thought this might deserve a little more attention, earlier on in the intro.

Reviewer #1 (Public Review):

Understanding the mechanisms of energy homeostasis is paramount to uncovering novel treatments for diseases such as obesity and diabetes. The authors found, consistent with previous reports (PMIDs: 29066463 and 31638856) that the renin-angiotensin system (RAS) - which normally controls blood pressure and which the authors studied previously in the context of diabetes - plays a critical role in thermogenesis (the process by which the fat stores in the body expends energy to produce heat in response to stimuli such as cold exposure). The authors find that Ace2 and Mas (the initiator and receptor of the RAS system, respectively) are upregulated in mouse brown adipose tissue (BAT; the fat stores responsible for thermogenesis) and are sensitive to cold-exposure.

The authors then provide convincing evidence that mice lacking Ace2 or Mas activity have decreased energy expenditure, decreased body temperature when exposed to the cold, and functional BAT. The authors also show that Ace2 or Mas deficient mice express less PGC1-a than mice with competent RAS signaling. PCG1-a is a critical regulator of mitochondrial production and mitochondrial uncoupling, two processes required for BAT thermogenesis. This suggests in the absence of RAS there is an ability for BAT to produce necessary proteins for mitochondrial uncoupling and heat production. Further the authors provide evidence that Ace2 deficient mice have less mitochondrial activity, consistent with this inability to produce mitochondria as underlying the thermogenesis defect in RAS deficient mice. The authors also transplant Mas deficient BAT into otherwise normal mice and find that these animals have also have abnormal energy expenditure, suggesting RAS signaling specifically in the BAT is necessary for normal energy expenditure.

The authors then determined if modulating RAS signaling could effect energy expenditure in diabetic mice that also harbor impairments in thermal regulation. The authors show that forced expression of Ace2 through adenoviral expression or treatment with angiotensin1-7 (the product of Ace2 activity) improves energy expenditure, and improves the ability to regulate body temperature when exposed to cold. White adipose tissue (WAT) which is normally not involved in thermogenesis can undergo "browning" and take on thermogenic properties in response to certain stimuli. The authors further find increased browning of WAT upon Ace2 re-expression measured by H&E staining, as well as increased expression of UCP1, PGC1a, and other BAT marker proteins.

The authors then seek to determine how RAS signaling leads to increased thermogenesis in BAT. Based on previous literature and transcriptomics of BAT upon RAS inhibition, the authors hypothesize that AKT and cAMP-PKA signaling could be critical in relaying the RAS signaling to elicit thermogenesis. The authors first find that RAS signaling modulates activating phosphorylation of AKT and PKA, and subsequently show that treatment of isolated adipocytes with AKT and PKA inhibitors reverse expression of mitochondrial and thermogenesis genes that are increased upon RAS signaling. Further, these inhibitors also block increased mitochondrial function upon triggering RAS signaling with angiontensin1-7 in isolated adipocytes. Collectively, these experiments suggest AKT and PKA signaling are critical for RAS regulation of thermogenesis but mechanistic elements remain to be elucidated about how RAS engages AKT and PKA signaling.

Reviewer #2 (Public Review):

Previous reports, including studies from the Yang group, had identified ACE2 as an important mediator of glucose and lipid homeostasis in the context of obesity. These studies hinted that ACE2 and other factors in the RAS system may affect adipocyte biology, but rigorous characterization of the physiological effects of the ACE2 pathway on adipose tissue - particularly that of brown adipose tissue - had not been completed. This manuscript by Cao, X, Shi, T, and Zhang C et al. fills these gaps in knowledge by demonstrating that multiple components of the ACE2 pathway impact thermogenesis and BAT function. The authors utilize a variety of methods, including six independent mouse models as well as isolated primary cells to test the hypothesis that the ACE2 pathway regulates thermogenesis and energy metabolism. Within the abstract, the authors make the following claims, which are supported by the evidence that follows:

1. "that ACE2 is highly expressed in brown adipose tissue (BAT) and that cold stimulation increases ACE2 and Ang-(1-7) levels in BAT and serum."

The authors show that ACE2 is expressed BAT at the protein (Figure 1A) and RNA (Figure 1B) levels. The authors demonstrate that ACE2 is induced by cold exposure for 6 hours (Figure 1C) and 24 hours of cold exposure (Figures 1D and E). The authors also demonstrate increased levels of ACE2 and Ang-(1-7), a cleaved product of ACE2, in serum after 2 or 4 days of cold exposure (Figures 1F and G).

2. "ACE2 knockout mice (ACE2-/y), Mas knockout mice (Mas-/-) and mice transplanted with brown adipose tissue from Mas-/- mice displayed impaired thermogenesis."

The authors show that ACE2-/y mice have decreased body temperature at 22C and 4C for eight hours (Figure 2F) and acutely after transition to 4C (Supplemental figure 1G). They show that Mas-/- knockout mice show similar phenotypes - decreased body temperature at 22C and 4C for eight hours (Supplemental figure 4J) and acutely after transition to 4C (Supplemental figure 4L). The authors then show that wild type C57BL/6 mice transplanted with BAT from Mas-/- mice show metabolic impairments consistent with ACE2-/y and Mas-/- mice, but do not show data regarding their body temperature.

3. "...impaired thermogenesis of db/db obese diabetic mice and high-fat diet-induced obese mice were ameliorated by overexpression of ACE2 or continuous fusion of Ang-(1-7)."

The authors show that db/db obese mice have decreased body temperature at 22C and 4C for eight hours which is rescued (or perhaps partially rescued) by adenoviral expression of ACE2 (Figure 3F). The authors additionally show that subcutaneous infusion of Ang-(1-7) elevates body temperature in db/db obese mice in response to cold challenges at 22C and 4C for eight hours (Figure 4F). The authors also show that infusion of Ang-(1-7) can increase body temperature acutely after cold challenge in mice fed a high fat diet (HFD) (Figure 4H), but they do not test the effects of overexpression of ACE2 on body temperature in the HFD model.

4. "Activation of {the} ACE2 pathway was associated with improvement of metabolic parameters, including blood glucose, lipids, and energy expenditure in multiple animal models."

The authors show that adenoviral overexpression of ACE2 in db/db mice increases their energy expenditure (Figure 3E) and improves glucose tolerance (Supplemental Figure 2B) and serum triglyceride levels (Supplemental figure 2C) relative to control mice. The authors show similar phenotypes for their model infusing Ang-(1-7) into the db/db mouse model, showing increased energy expenditure (Figure 4E), improved glucose tolerance (Supplemental figure 3B), and decreased serum triglycerides (Supplemental Figure 3C).

5. "...[the] ACE2 pathway activated Akt/FOXO1 and PKA pathway<s>, leading to induction of UCP1 and activation of mitochondrial function."

The authors show that primary brown adipocytes have increased basal oxygen consumption and spare respiratory capacity when treated with Ang-(1-7), and that these effects are negated by treatment with an AKT inhibitor (MK2206, Figure 6E) or a PKA inhibitor (HA, Figure 6J), strongly implicating these signaling pathways in increased mitochondrial function (as assayed by oxygen consumption). The authors also show that Ang-(1-7) induces UCP1 expression and that this is blunted by an AKT inhibitor (Figure 1C) or a PKA inhibitor (Figure 6I).

Strengths

This is a strong paper addressing the role of the angiotensin II pathway in obesity-related metabolic parameters and thermogenesis. The authors use multiple orthogonal methods to rigorously characterize the physiology of the RAS pathway in BAT. First, the authors use six independent mouse models to test the effects of the ACE2 pathway on metabolic homeostasis and thermogenesis. This includes three distinct loss of function models, including ACE2-/y mice, Mas-/- mice, and wild type C57BL/6 mice with subcutaneously transplanted Mas-/- BAT mice. In each of these loss of function models the authors found that loss of the ACE2 pathway leads to worsened metabolic parameters, and in most of these models, decreased thermogenesis and BAT function. Collectively, these data strongly suggest that intact ACE2 pathway signaling is required for BAT function, thermogenesis, and subsequent metabolic health in mice. Additionally, the authors characterize three distinct gain of function models, including an adenoviral ACE2 overexpression system in db/db mice, and two continuous infusion models of Ang-(1-7) in db/db mice or mice fed a high fat diet. In each of these models, increasing function of the ACE2 pathway is associated with improvements in systemic metabolism, and, in most models, improved thermogenesis. Finally, the authors utilize primary cells to test the pathways involved in Ang-(1-7) function to identify the AKT and PKA signaling pathways as mediators of the physiological response seen. The authors show that both AKT and PKA signaling are required to elevate mitochondrial function (as assayed by oxygen consumption) in response to Ang-(1-7) treatment. While the precise molecular mechanisms of how the ACE2 pathway activates these signaling pathways and downstream mitochondrial are not fully worked out, these details are well beyond the scope of this study.

In summary, this paper identifies the ACE2 pathway as critical for maintaining thermogenesis and energy expenditure in mice. The authors demonstrate this through multiple independent mouse models as well as primary cells. Their data give important insights on the ACE2 pathway in mediating brown adipose tissue function in metabolism and thermogenesis in obesity.

Weaknesses

There are few major weaknesses in this study. Most noted weaknesses in the manuscript can be resolved without the need for further experimentation (e.g., by clarifying and/or reanalyzing select data or figure panels). One note is that some of the claims put forth in the abstract that have not been directly addressed (as noted above). This can be addressed by either performing experiments to address the claims or by modifying the claims to reflect the current experimentation.

Reviewer #1 (Public Review): 

The manuscript by Reinke and colleagues describes the identification and characterization of a host factor, aaim-1, which promotes microsporidia infection of C. elegans. The authors conduct a genetic screen that results in the identification of aaim-1 and confirm the identity through rescue experiments and the characterization of additional alleles. Evidence for intestinal expression is also provided through tissue-specific expression experiments though endogenous expression is challenging to observe. A potential mechanism is inferred through careful characterization of intestinal lumen uptake and invasion in mutant host animals, and the authors suggest that the aaim-1 promotes the orientation of microsporidia spores that promotes successful invasion of the C. elegans intestine. The authors then show a modest increase in susceptibility to the bacterial pathogen Pseudomonas aeruginosa in aaim-1 mutant animals. Further characterization in competitive advantage assays and coinfection assays are also carried out. 

Little is known regarding the host factors required for microsporidia invasion, and thus the successful forward genetic identification of a previously uncharacterized host factor is noteworthy and represents an elegant use of the experimental system in which microsporidia infect C. elegans. The authors' data point to a functional role for secretion of AAIM-1 into the host lumen. As is often the case in trying to define mechanism of action of previously uncharacterized genes identified from genetic screens, defining a molecular mechanism is challenging. The model in which AAIM-1 may function to promote the proper orientation of spores to the intestinal lumen represent an appealing hypothesis, but the relative low-resolution images and unclear implications of the orientation leave considerable doubt. Perhaps higher-resolution microscopy (and more three-dimensional definition) and further discussion of the evidence of the significance of the angle with regard to successful infection would be helpful. Nevertheless, the authors have conducted a straightforward dissection the possible steps of infection where AAIM-1 may act and have arrived at a plausible hypothesis. 

Whereas the characterization of AAIM-1 as a host factor required for microsporidia invasion is novel and convincing, the characterization of its phenotype in bacterial resistance is less compelling. The magnitude of the effects observed with regard to Pseudomonas are modest at best. Competitive fitness assays do little more than corroborate the resistance phenotypes. The manuscript argues for the novelty and significance of a factor that promotes microsporidia invasion but limits bacterial infection (including emphasis in the title of the manuscript), yet the relative strengths of the phenotypes (microsporidia-STRONG, SOLID; bacteria-WEAK, MARGINAL) are such that this narrative is not compelling. I think the more straightforward message of this work, which focuses on the identification of a novel host factor required for microsporidia invasion out of an elegant C. elegans genetic screen, is well-supported by the data and will be of interest to the community of investigators studying the interaction between animal hosts and eukaryotic pathogens.

Reviewer #2 (Public Review): 

Using a forward genetics screen, the authors discovered a gene, aaim-1, to be a regulator of invasion of C. elegans intestine by microsporidia, a pathogen present in its natural habitat. This gene was known to be induced by infection with pathogenic bacterium P. aeruginosa. The authors go on to show that the phenotype is L1 larva stage-specific although the reason for this specificity is not entirely clear. The study proposes that the orientation of the spore in the intestinal lumen is somewhat altered in the aaim-1 mutant. They go on to show that aaim-1 mutant has better fitness during microsporidia infection but lower fitness during P. aeruginosa infection. The study is incomplete in its present form and a close review of the data indicates that the conclusions are not entirely supported by the data. 

Strengths: The strength of the study lies in the use of a forward genetics approach to uncover a new regulator for microsporidia invasion. I commend the authors for taking the approach. All 4 strains identified in the screen map to the same locus, for aaim-1. The phenotype of these alleles is phenocopied by a deletion allele created using CRISPR. The authors show some interesting evidence for alteration in microsporidia and P. aeruginosa load in aaim-1 mutants suggesting that AAIM-1 may regulate responses to these two classes of pathogens differentially. 

The study appears to be incomplete in its current form. The major weakness lies in the lack of data on the possible functions of AAIM-1. 

The authors have missed taking note of the observation that microsporidia invasion causes larval arrest along with brood size defects (Fig 1). Both these can also result from starvation, caloric restriction, and in the case of certain infections [Van Gilst 2005; Garsin 2003; Dasgupta 2020]. It is quite possible that AAIM-1 mutation relieves the block caused by infection (microsporidia or other) which may impinge upon C. elegans metabolism (specifically lipid metabolism). Several papers in the recent past have indicated that lipid metabolism is altered during infection with both Gram-negative and Gram-positive bacteria. Does the same happen in microsporidia invasion? 

The significance of spore orientation relative to the epithelium is not clear. Is there published literature, from authors or others, to show that spore orientation is causal or correlated with the firing of polar tubes and subsequent invasion? Authors should present evidence for this. I suggest that the authors find some means (perhaps the use of microfluidic channels with adhesive molecules), to show how spore orientation is linked to the presence of AAIM-1 protein. In absence of this information, the conclusion that AAIM-1 regulates spore orientation and firing is misleading. 

An alternate explanation for altered spore orientation is that adhesion molecules expressed on intestinal epithelia in the aaim-1 mutant might be different from those in wild type/reference strain of worms. This can be studied at the level of transcriptome and proteome and specifically for proteins with carbohydrate modifications. 

Microvilli are known to be lost in intestinal epithelial during S. aureus infection [Irazoqui 2010], is there any evidence of whether microvilli are altered in the aaim-1 mutant which may affect adhesion or orientation of spores? 

The effect of aaim-1 mutation on worm's survival during PA14 infection is very mild. Did authors find a known regulator of the innate immune pathway (p38 MAPK, XBP-1, ATF-7, ZIP-2, ELT-2, etc) or their transcriptional targets to be altered?

Along the same lines, one would predict that aaim-1 mutation would affect fitness on other pathogenic bacteria found in the same habitat as microsporidia [Troemel 2008; Schulenberg 2016]. If AAIM-1 provides a trade-off between protection from pathogenic bacteria and invasion by microsporidia, it should become apparent in other cases as well. 

The statement that AAIM-1 is exploited by microsporidia for successful invasion is not satisfactorily supported by data. To support this claim, authors can test whether AAIM-1 binds to or decorates spores which would suggest a possible impact on the binding of spores to worm's intestinal epithelium? Alternatively, authors can examine expression of lectins in aaim-1 mutant (Lectins are altered widely in C. elegans during infection with various pathogens and are known to affect adhesion in various contexts).

Reviewer #1 (Public Review): 

Burlingham and colleagues investigated how a task-related hemodynamic response in visual cortex, measured with fMRI-BOLD, varied with task difficulty and behavioral performance in humans during a visual orientation discrimination task. In prior work from this group (Roth et al. 2020, PLOS) and others (Cardoso et al. 2012, Nat. Neuro; Cardoso et al. 2019, PLOS), this "task-related response" (TRR) was shown to be distinct from a stimulus or attention-evoked response, modulated by reward size, and linked to physiologic arousal markers. Unique from prior work that centered on trial-averaged responses, a major aim of this work was to use a general linear mixed model (GLMM) to estimate TRRs at the single trial level. The authors showed TRRs were higher in amplitude and more precise in timing for more difficult trials and trials where the subject made an error or responded faster, situations consistent with elevated arousal. They propose arousal as an underlying driver for TRRs based on similar trends in physiologic metrics, such as pupil, respiration and heart rate. Additionally, they found evidence that TRRs, and the effects of task difficulty, were spatially diffuse (also present in V2 and V3), but weaker in magnitude higher in the visual hierarchy. 

Strengths: 

The task-related response (TRR) is a poorly understood hemodynamic signature that has been reported in the literature. Given its magnitude is similar to that of the stimulus-evoked hemodynamic response, understanding and accounting for TRRs will be critically useful for the field. This work takes a step forward in characterizing TRRs on a trial-by-trial basis with their use of a GLMM. 

Specifically, TRRs were modeled as the sum of the hemodynamic responses to trial onset, time on task, and button press. This model allowed Burlingham et al. to develop hypotheses to explain complex features of their results. For example, the effect of reaction time on the trial TRR would not have been easily captured by one variable, but rather the summed influence of the time the subject spent on the trial and the time the subject took to respond. Their GLMM allowed them to simulate fMRI activity in response to each input independently to show that the combined inputs best explained the experimental results. This modeling approach was a key strength of this work and future work on TRRs will greatly benefit from utilizing a similar approach. Even studies that do not aim to focus on studying the TRR can make use of this strategy to remove or account for this hemodynamic signature in the data. 

Additionally, this work evaluated changes in the TRR within the framework of arousal level changes related to behavioral performance and task difficulty. Identifying mechanisms by which arousal influences brain activity is of significant interest to the field of cognitive neuroscience. The task paradigm in this work evoked changes in arousal level by manipulating task difficulty and then assessed these changes using behavioral and physiologic markers of arousal. Their results provide additional support to existing literature that TRRs could be a cortical arousal signature. 

Weaknesses: 

While the implications are compelling, a few other controls and analyses would better establish the link between arousal level and the TRR. 

First, it is difficult to link changes in task difficulty to arousal level without demonstrating that the subjects did not change their strategy between easy and difficult task conditions by, for example, looking directly at the more difficult targets instead of maintaining central fixation as the task required. Without this control, the changes reported in TRRs could be attributed to changes in eye movements and the concomitant changes in the the visual field, especially given measurements were made in visual cortex. In the same vein, a more detailed or explicit differentiation between the stimulus or attention-evoked hemodynamic response and the TRR is necessary to help the reader evaluate the TRR without simultaneous eye tracking to remove trials where the visual field may have changed. 

Given arousal is a loosely defined cognitive phenomenon, physiologic arousal markers (ex. pupil, heart rate, respiration) are commonly used to track changes in arousal level, as is the case in this work. The evidence in this work that arousal level changed between task conditions (ex. difficult and easy trials) requires a more detailed analysis to control for the large number of variables and determine the effects that survive. While an accompanying data set showed changes in pupil diameter in a manner consistent with arousal changes during the task, this data was recorded in a separate experiment. This does provide a source of eye movement data for potential control analysis. 

Lastly, the authors speculate about the origin of the TRR by comparing its magnitude and modulation in different task conditions across different levels of the visual cortical hierarchy (V1 vs. V2 vs V3). A direct statistical comparison of these effects would be necessary to convincingly demonstrate differences in the TRR across visual regions.

Reviewer #2 (Public Review): 

The main results for this paper come from an fMRI study in 9 participants. Stimuli were Gabor patches, presented in the right lower visual field. The participants' response identifying the orientation of the Gabor was identified as correct or incorrect with a tone at the time of a button press. Stimuli were presented with a 15 s inter-stimulus interval to allow for examination of the hemodynamic response shape after each stimulus. 

"TRRs" were defined as responses time locked to these stimuli but in the ipsilateral visual field. 

The authors find that TRR magnitude (defined both based on Fourier amplitude and by mean % change) is strong in V1 (and may be localized to the superior bank). These responses are modulated by correctness (Figure 2B, effect for Easy trial only), and by reaction time (both easy and hard correct and incorrect trials) (Figure 2D). Interestingly, the effect of reaction time was opposite in the Hard vs Easy trials. The authors used a GLMM model to show that task onset, button press and time on task all contribute to the signal. 

These TRRs were biggest in V1, and smaller in V2 and V3. 

This scaling with reaction time fit a model incorporating trial onset, button press, as well as time on task (reaction time). <br> The reviewers also did a similar experiment in 5 participants examining physiological responses outside the scanner, with a 3.5 s inter-trial interval. Accuracy and difficulty also had similar effects on these metrics. This suggests a hypothesis that the TRR effect may be driven by something similar to the physiological responses. 

Taken together these results suggest that there is trial-driven hemodynamic signal in V1 that covaries with behavior in a way that is consistent with arousal. The discussion suggests that this relates to a LC-NE arousal process. The connection is suggested by the data, but further work would be needed to cement this idea. The data are interesting, and a good window into further understanding of this effect. 

In the discussion line 330, they suggest that the TRR should be separately modeled and removed from fMRI data in preprocessing. While the authors have convinced this reader that the TRR is likely related to arousal, it is far from clear that this means that this effect should be removed from fMRI data in preprocessing. Many arousal effects exist naturally in fMRI data, and in brain activity in general. Many arousal effects are observable in spiking and LFPs. Since no spiking or LFPs were measured here, we don't know whether this signal is or is not related to spiking or LFPs (though some data from monkeys suggests a similar signal is hemodynamic only, it would take more convincing that the current TRRs arise from the same process as the previously reported primate literature).

Reviewer #3 (Public Review): 

Burlingham et al. extend their previous work on fMRI signals ('task-related responses' or TRRs) that are measured in visual cortex in response to task events, but which are spatially more widespread - in this case, to ipsilateral V1, V2 and V4, which receive input from the visual hemifield where no stimulus is presented. They show that these TRRs covary with a number of task and behavioral factors on a trial-by-trial basis. Similar patterns are observed for psychophysiological measures: pupil dilation, heart rate and breathing. 

This paper characterizes an interesting signal that is especially important to take into account when studying visual fMRI responses. The modeling and quantification is rigorous, and the link to central arousal systems is very promising. However, a weakness of the paper is that the authors do not pursue the computational/functional significance, nor the biological drivers, of TRRs. For instance, linking TRRs to an explicit model of decision-making (beyond showing they covary with RTs and lapses), or further discussing their potential link to widespread arousal and movement variables in rodent calcium imaging and ephys data, would strongly increase the interest from those beyond the visual fMRI community.

Reviewer #1 (Public Review):<br> This paper examines whether judgments of agency (JoA) are best characterized as 1st order measures of internal signals (prediction error type) or as metacognitive reports - i.e. 2nd order measures of lower-order agency signals. The authors make a clear prediction: if JoAs are metacognitive reports, they should treat noise/uncertainty in the same way as other metacognitive reports such as confidence. 

To test this prediction, the authors designed two tasks in which participants were asked to report their (maximum) agency on movements of a virtual hand presented with or without delay, under high or low noise, and/or their confidence in their agency decision. The hypothesis tested is the following: if confidence and agency judgements share the same computational characteristics (2nd order uncertainty monitoring), then they should show similar sensitivity to internal estimates of sensory noise in the task. 

The predictive power of two models is compared. The model that best explains patients' agency ratings does not involve any estimates of sensory noise (no metacognitive monitoring of noise), unlike the model that best explains confidence measures. Based on these results, the authors conclude that JoAs are not meta-cognitive (in the sense that they are not 2nd order reports of agency signals) but rather reflect 1st order measures of internal signals. 

The study features a nice combination of experimental tasks and computational models specifically designed to address the question at hand. The originality of the approach is to take into account the uncertainty associated with the processing of one's agency, which classical experimental work on JoA generally considers as variability of non-interest, and to offer a nice computational characterization of how JoA responds to this uncertainty (i.e. through scaling of subjective ratings). All this is well articulated with the existing literature and previous models on agency - e.g. the comparator model. 

This paper is important. It offers convincing evidence that confidence mesures, but also (pre-reflexive) feeling of agency and (explicit) judgment of one's agency tap into different computational mechanisms and factor in different contextual information, which is relevant for research on action control but also for the science of consciousness, and which could potentially inform methodological choices about how to measure cognition (e.g. implicit or explicit measures).

Reviewer #2 (Public Review): 

This study investigated the influence of sensory noise on judgements of agency (JoA); the subjective feeling that an action is caused by ourselves. The idea is that this influence can reveal whether JoA's are only like metacognitive judgements conceptually, in that they entail cognition about cognition, or also computationally, in that they incorporate the uncertainty of the signal in a similar way that confidence judgments do. An elegant combination of pre-registered hypotheses, psychophysics and computational modelling is used to answer this question. The authors find convincing evidence for a 'rescaling model' in which JoA's are rescaled differently depending on the noise condition. This is different to the computational mechanism underlying confidence judgments, which instead incorporates sensory noise to make Bayesian optimal decisions about confidence. The paper is well written, the experiments are well designed, the analyses are sophisticated and appropriate and the conclusions largely follow from the results. My comments are about conceptual clarification, the interpretation of the two tasks and about whether the results can support one of the main claims. 

It is really great that the authors have pre-registered the research questions, methods and analyses in such detail and also explicitly indicate where they diverge from this pre-registration. I also want to applaud their data and code sharing. This paper is a really great example of open science. The sharing of data and code will also make it easier for other researchers to use these methods in future studies, which is great because the authors have developed very elegant paradigms to study (metacognition of) sense of agency. 

I was a bit confused about the relationship between the confidence task and the judgment of agency task. The confidence task (Fig. 1) measures confidence about a discrimination based on judgments of agency, while the judgement of agency task (Fig. 2) is about directly inferring agency from one stimulus. So, in a way the confidence task by design reflects a higher-order judgement about judgements of agency and, in this task setting, the JoA's are treated as the first-order judgements. I wonder whether this set-up leads to the difference in computations underlying the confidence responses and JoA's found and whether alternative set-ups could show similar effects for confidence and JoA. This does not question the main point of the manuscript, which is about determining which kind of computations underlie JoA's, but it is important in relating the results of these two tasks. 

Finally, the authors state that their results show that judgements of agents are not computationally metacognitive, but I am not sure whether this conclusion fully follows from their results. They found evidence for a rescaling model over a Bayesian model, but if I understood the rescaling model correctly, it still requires participants to estimate whether their signals contain high or low noise, and then rescale their JoA's accordingly (less extreme judgments when there is more noise). This means that participants do take into account the noise to make their judgements of agency, but they do so in a different way than what has been found for confidence judgements. I am not sure whether this is enough to say that JoA's are not metacognitive in nature.

Reviewer #3 (Public Review): 

Constant et al describe a study investigating an important issue - are judgements of agency metacognitive in nature? While this topic has received a lot of theoretical attention, empirically the issue is underexplored, partly due to a lack of appropriate frameworks and tools. Here the authors suggest the issue can be tackled by thinking more precisely about the computations involved in both judging agency over an outcome and in forming a (metacognitive) confidence report. This focus on constituent computations is an important conceptual strength of the paper. 

The authors choose to operationalise metacognitive computations as those where agents have "second order access to sensory noise" and design two similar tasks - a confidence judgement task and an agency judgement task - where observers report their experience of controlling a virtual hand that can move synchronously or be delayed. Crucially, the uncertainty of the incoming sensory signals is varied, and the authors explore whether agency and confidence judgements are influenced by this sensory noise, and which kind of computational processes can best explain how. While the authors find empirically noise has an effect on both kinds of judgements, computational modelling suggests that agency judgements are best explained by a 'rescaling' model which does not include an explicit representation of the noise, whereas confidence judgements are better explained by a 'Bayesian' model which does represent noise. 

There is lots to enjoy about this paper. It is particularly inspired to have an agency and confidence task that are so similar, making them more directly comparable. Indeed, they are compared in the paper with basically identical computational models, something which to my knowledge has never been achieved in this field of work. The models themselves all seem well chosen given certain design assumptions, though I suspect the more general insight of generating explicit computational models of agency-like judgements is one that could inspire other researchers in this field, and charts a route to progress on thorny issues on this and related topics. 

However, while this approach is intriguing, I think the main weakness of this study relates to the core experimental manipulation: introducing temporal delays between actions and outcomes to influence ratings of control. While this is a popular approach in the field, recent authors (e.g., Wen, 2020, Consciousness and Cognition) have suggested that this manipulation may be problematic for a number of reasons. In similar types of paradigm, Wen (2020) notes that agents are able to accurately judge their control over action outcomes that are substantially delayed (e.g., well over 1000 ms) and thus it is possible that 'delay manipulation' designs actually introduce response biases, where participants are somewhat artificially reporting variance in the delays they experience rather than their actual experience/belief about what they can and cannot control. Indeed, in the methods of this present paper, the authors note participants were asked to "focus specifically on the timing of the movement" of the virtual hand, which may make this concern particularly apposite. 

Because of this manipulation, all of the computational modelling (naturally) assumes that agents are engaged in a task where they have to detect the delay and compare this to some criterion value. Indeed, there is nothing else they could be doing in these tasks. The report of "agency" is thus generated directly from this internal variable that encodes "did I detect a delay?", and any confidence report is a metacognitive judgement about that decision. 

This raises an important issue of conceptual validity: is a judgement of agency equivalent to judging whether an outcome was delayed or not? Many results (see review by Wen, 2020) suggest that agents can simultaneously tell an action outcome was delayed, but still judge themselves to be the agent, suggesting that an equivalence along these lines is unlikely. If so, this would mean acknowledging the generalisability of these is conclusions is potentially limited: rather than concluding that agency judgements in general are non-metacognitive, the conclusion would be the sensorimotor delay judgements in particular are non-metacognitive. The latter conclusion is by no means uninteresting, but has a somewhat narrower theoretical significance for the key debate used to frame this paper ("do agency judgements monitor uncertainty in a metacognitive way?") 

A second important issue relates to what exactly makes a computation 'metacognitive'. For example, the authors argue their Bayesian model is a metacognitive one, because it requires the observer to have second-order access to an estimate of their own sensory noise. I am not completely sure this follows: the Bayesian model in this paper clearly incorporates an estimate of the noise/uncertainty in the signal, but not all representations of noise are second-order or metacognitive. For example, Shea (2012) has noted that in precision-weighted Bayesian inference models throughout neuroscience (e.g., Bayesian cue combination, also discussed in this paper) the models contain noise estimates but the models are not metacognitive in nature. For example, when we combine a noisy visual estimate and a noisy auditory estimate, the Bayesian solution requires you account for the noise in the unimodal signals. But - as Shea argues - the precision parameters in these models do not necessarily refer to uncertainty in the agent's perceptions or beliefs, but uncertainty in the outside world. It seems a similar argument is relevant to the Bayesian model of agency offered by the authors in the present paper. It is not clear to me why we should think the uncertainty parameter in the Bayesian model is something metacognitive (e.g., about the agent's internal comparator representations) rather than something about the outside world too (e.g., the sensory environment is noisy). 

References:

Shea (2012) Reward prediction error signals are meta-representational. Nous, DOI: 10.1111/j.1468-0068.2012.00863.x 

Wen (2020). Does delay in feedback diminish sense of agency? A Review. Consciousness and Cognition, DOI: 10.1016/j.concog.2019.05.007

Reviewer #2 (Public Review):

The authors study fast transient amplification of external inputs in nonlinear recurrent networks of excitatory (E) and inhibitory (I) neurons with different forms neural or synaptic adaptation: spike frequency adaptation (SFA) or short-term synaptic plasticity (STP). They seek conditions under which nonlinear transient amplification (NTA) at the onset of the stimulus is strong, and yet activity reaches a stable steady-state with not too large (biologically plausible) activity. The mechanism for the NTA is strong recurrent excitation (E-E connections), while the re-stabilization of activity after amplification is provided by the adaptive mechanisms (SFA or STP). They find that while SFA is unable to re-stabilize post-onset activity,

A) STP - either in the form of short-term depression (STD) of E-E connections, or<br> short-term facilitation (STF) of E-to-I connections-- is capable of doing so.

In addition they demonstrate other features of NTA in networks with STP, which they characterize as follows:

B) NTA requires symmetric recurrent EE

C) unlike in linear transient amplification, NTA happens only when stimulus strength is above a nonzero threshold

D) NTA is orders of magnitude stronger than TA in non-normal linear E/I networks

E) The post-transient steady state is inhibitory stabilized (ISN) and shows the associated paradoxical effect.

F) Stimulus onset responses (i.e. during NTA) provide better pattern completion and "pattern separation" (or stimulus selectivity) compared to steady state responses.

G) E/I co-tuning "broadens the parameter regime of NTA".

While the study provides novel findings and thus has merit, it is problematic that the main findings A, B & C are actually not novel and have been studied in detail before: they were studied and characterized as such in networks with strong recurrent excitation with STD, by Misha Tsodyks' group (main references are Loebel and Tsodyks 2002, and Loebel, Nelken and Tsodyks 2007, which studied rate networks as in the current study, but also Tsodyks, Uziel & Markram 2000 which studied the phenomenon in spiking nets).

In particular, Loebel et al. 2007 provided a model of Auditory cortex (that captured various empirically observed aspects of fast onset auditory cortical responses) with E and I neurons, with exactly the same mechanism for NTA (A & B): strong symmetric E-E recurrence, with re-stabilization to a low activity state provided by STD. In particular, property C was demostrated (and conditions for it analytically derived in Loebel et al. 2002) and was used (in Loebel et al. 2007) to account for the V-shaped Frequency-Intensity tuning curves observed in auditory cortex.

Yet none of these past studies are cited in the current study, and the results A-C are presented as novel.<br> Properties D and E also possibly/probably hold in Loebel et al. 2007's model too, but that paper did not study/characterise these features.

So the findings D-G are indeed novel (to my knowledge), as are the extension of A-C to networks with STF instead of STD (as is the negative result that SFA cannot yield restabilization, I believe).

Thus I think a major rewriting of the manuscript is in order, in which the main focus (in particular the selling points in Abstract/intro/discussion, e.g. lines 74-75) is on these truly novel findings (D-G) instead of on A-C (which of course could/should still be mentioned). Moreover some of these newly characterized features (in particular D & E which are currently only cursorily mentioned, and were not properly quantified, or their parameter dependence was not studied) should be more thoroughly/quantitatively characterized.

Reviewer #1 (Public Review):

In this manuscript, the authors introduce a new mechanism, named nonlinear transient amplification (NTA), to explain how cortical circuits can amplify specific patterns of activity without dynamical slowing of network dynamics. Combining analytical and numerical investigations of networks with NTA, the authors provide an extensive characterization of how properties such as pattern completion and separation evolve in time and depend on the tuning of inhibitory cells. Moreover, they show that the NTA mechanism applies also to networks of spiking neurons.

The idea of combining network dynamics effects, such as the increase of effective coupling between cells with input strength, with short term plasticity is novel and can drive the field to explore how known biophysical properties of neurons and synapses shape brain computations. Two-photon stimulation technologies now allow experimentalists to investigate if and how networks of neurons in the brain implement computation such as pattern completion and separation; new theories like the one introduced in this paper are critical to interpret results obtained in these photostimulation experiments.

Overall, this is a stimulating and timely paper. However, there are important concerns (see list below) that I think should be addressed in order to state that nonlinear transient amplification underlies computations in the brain.

1) Critical features of the model do not seem compatible with experimental results (1a and 1b). The authors should either explain how their model can be reconciled with these experimental findings, or acknowledge the limitations of their results.<br> 1a) The network model is characterized by an input dependent dynamical stability (Figure 2) and, for low inputs, operates in the non-inhibition stabilized regime. However, experiments have shown that cortical circuits, at least in the mouse brain, are in the ISN regime already at baseline, i.e. in the absence of sensory stimuli [Sanzeni et al., eLife 2020].<br> 1b) The model predicts that pattern completion in cortex should be limited to response onset, and should be followed by suppression of cotuned unstimulated cells (Figure 4). Pattern completion in cortex has been recently investigated in optogenetic experiments (e.g. [Carrillo-Reid et al., Science 2016; Marshal et al., Science 2019]); these experiments have shown that stimulation of a subset of cells activates cotuned unstimulated neurons. Contrary to the model prediction, unstimulated neurons remain active throughout the whole stimulation period, which lasts a few seconds, i.e. much longer than the typical time scale of STP.

2) The description of balanced amplification done in the text should be improved. In its original formulation [Murphy and Miller, Neuron 2009], balanced amplification has been shown to selectively amplify specific patterns without slowing of network dynamics. This amplification emerges in networks with different ensembles of excitatory cells, each of which is characterized by symmetric connectivity between excitatory cells, and cotuned inhibitory cells. Therefore, the authors should frame their work as an alternative mechanism to balanced amplification, rather than a solution to an unsolved conundrum.

3) The stability mechanism underlying the results of Fig. 2 should be analyzed in more depth. Fig. 2B suggests that the network is stabilized by STP and not by inhibition, since it does not show runaway activity when inhibition is clamped. Despite this fact, the authors refer to the steady state as inhibition stabilized but it is not clear what evidence supports this claim. The authors use the paradoxical effect as a proxy for inhibitory stabilization, but the implication "paradoxical response->ISN" has been proven only in networks without STP and we do not know if it holds in networks with STP. The "ISN index" defined in Eq. (13) gives information about the stability of the excitatory population but, in networks with STP or STF, does not tell us if the network is stabilized by inhibition or by another mechanism (e.g. STP).

4) Page 7-8. In networks of excitatory and inhibitory neurons without STP (two degrees of freedom), the derivative of the characteristic function F(z) is related to the eigenvalues of the network Jacobian matrix [Kraynyukova and Tchumatchenko, PNAS 2018]; this relation allows to analyze the stability of a network fixed point by studying the slope of F(z). The authors apply this approach in their model models which has three degrees of freedom (e.g. activity of excitatory-inhibitory neurons and STD variable x). To do so, they should prove that the approach of [Kraynyukova and Tchumatchenko, PNAS 2018] can be generalized to their model, e.g. they should show how the derivative of F(z) relates to the three eigenvalues of the network Jacobian matrix and to the network stability.

Reviewer #1 (Public Review): 

The authors here present a useful new technique for investigating human neuronal axons by cryo-electron tomography. Axons are extended from organoids onto EM grids, which are then vitrified and studied by CLEM and cryo-tomography. The authors also present some observational results derived from the cryo-tomography, with some support from light microscopy. 

The technique is an exciting one, with the important benefit of isolating human axons on EM grids, compared to previous cryo-ET methods of studying of whole rodent neurons. This is the first cryo-tomography experiment on human axons to my knowledge, however, once free of the organoid context, the axons cannot really be described as benefiting from a more physiological 3D organoid environment. The technique will be of good use to neuroscientists working on axonal ultrastructure. 

The data presented is chiefly exploratory and observational but is generally experimentally solid and well-presented. In addition to the preliminary methodological characterisations, CLEM and tomograms presented as proof of concept, there is some quantitative work. This data is scientifically sound but is mainly confirmatory of previous findings by other groups in other systems. For example, uniform microtubule polarity in axons, the thinness of axonal ER (e.g see work by Mark Terasaki) and to some extent axonal ribosome paucity are all well-described phenomena. L1CAM-GFP or ESYT1-GFP overexpression were interesting experiments, but the authors report no differences with wild-type without presenting quantitative evidence of this conclusion. However, this data serves well as evidence of the utility of this novel human experimental system.

Reviewer #2 (Public Review): 

The use of cerebral air-liquid interface organoids (i.e., ALI-COs recently introduced by Lancaster Lab) to study growing axons in combination with cryoCLEM/cryoET has been convincingly demonstrated in this manuscript. The authors show filaments, microtubules, mitochondria, vesicles, ER, contact sites and ribosome-like particles in the tomograms obtained. They quantitatively analysed the polarity of individual microtubules, the plasma membrane surface area to deduce the lipid supply and the local ribosome concentration in growing human axon shafts. Although argumentatively sound, most of the results are at the ultrastructural level, some are conjectural, and a comparison to previously known findings obtained by other methods is not present in the present version. 

Strengths: 

The fact that such cerebral organoids can now be prepared and studied at molecular resolution will certainly allow in-depth follow-up studies. It is therefore an original proof-of-concept report that is of general interest for neuroscience and cellular structural biology. The manuscript nicely illustrates the power of tomography to visualize the intracellular landscape of a biological material previously inaccessible to this method and presents the first data on variations found in different human axon regions. 

Weaknesses: 

Although "molecular" is stated several times, the authors have only partially performed a true molecular analysis; most of it is at the ultrastructural level. All "molecular" results are either indirectly derived; lipid supply/concentration across the plasma membrane surface and local "ribosome concentration" based on identification of ribosomes on size, shape and contrast in tomograms. It would have been more compelling to at least have performed a rough subtomogram analysis here to be sure that the putative ribosomes are indeed ribosomes. I'm pretty convinced that the authors picked ribosomes, and I don't want to be hypercritical here, but there's no proof. The FLM data only support the overall impression that there are fewer ribosomes in axon shafts than in dendrites, but of course with limited spatial resolution.

The discussion is therefore somewhat elusive, starting with "molecular features," but most of the findings are at the ultrastructural level and the authors choose " likely," "could reflect/represent," or " may explain" to account for this absence.

Reviewer #3 (Public Review): 

Determining cellular ultrastructure by cryo-electron tomography (cryo-ET) or by correlative light and electron cryo-microscopy (cryo-CLEM) have so far been on isolated cells. These have limited the ability to study cellular ultrastructure in context of tissues or in near-native conditions. For the first time, Hoffman et al. have elegantly utilized cerebral organoid technology to study the ultrastructure of growing axons by cryo-CLEM. In this manuscript the authors describe the procedure of growing cerebral organoids at air-liquid interface, along with uniquely labeling various cellular components like cytoskeleton elements and organelles for studying their distribution over time by live cell imaging and fluorescent microscopy. Although cryo-CLEM studies of axons are not new, but the novelty of this work is the combination of ALI-CO technique with growing axonal extensions from the organoids on EM grids and isolating them for cryo-EM sample vitrification. This eliminates the need for culturing isolated neurons from neuronal tissues for structural studies using cryo-CLEM. By uniquely labeling axonal and dendritic markers and showing their distributions by fluorescent imaging, the authors have supported their claim that the extensions from the organoids are axons. By using this technique, the authors were able to visualize the arrangement microtubules along with determining their polarity, ER morphology, membrane contacts and distribution of ribosomes. These support the claim by the authors that this technique will allow studying neurons in context of tissues or under different physiological conditions. However, the major weakness of the manuscript is that it does not demonstrate how this technique can be universally used for other cell types or tissues. The unique properly of axons to grow as thin projections from CO has made this possible for studying neurons, but how this technique can be used for other tissues is lacking in this study. This study could have been more impactful if the authors would have demonstrated more broader application of the methodology and not limited it to CO or neurons.

Reviewer #1 (Public Review):

Lambey et al. used x-ray crystallography, docking, and MD simulations to generate high-quality models of chemokine receptors interacting with Staphylococcus aureus (SA) leukotoxins (LukE). SA leukotoxins cause severe infections and invade host cells by first interacting with chemokine receptors on the host cell. The structural models emerging from this work identify novel interactions sites that are also highly conserved. This work is significant because, in addition to advancing our mechanistic understanding of the recognition of features exploited by LukE, it identified specific sites that we could be target and design inhibitors that prevent leukotoxins from invading host cells.

Reviewer #2 (Public Review):

The manuscript authored by Lambey et al delves into the interactions of leukotoxins secreted by S.aureus and their interactions with chemokine receptors prior to pore formation. This study uses a wide array of techniques to understand an important step in the host-pathogen interaction that leads to specific targeting of chemokine receptor-containing cells by a staphylococcal leukotoxin component (LukE) which proceeds to cell lysis when the F component for the toxin is available (LukD). The authors use X-ray structures of the LukE subunit (previously determined by Nocadello et al., 2016) in apo state and in complex with the p-cresol sulfate to mimic sulfated tyrosine and N-terminal sulfated tyrosine containing peptides of the CCR2 and ACKR1 receptors to demonstrate the importance of this posttranslational modification (sulfation) in dictating the interactions with the host chemokine receptors. In addition, they perform a very nice FRET experiment to demonstrate the interaction of the LukE to compete the native labeled ligand of the chemokine receptors to demonstrate nanomolar affinity of LukE to ACKR1, CCR2 and CCR5 receptors. They further use docking tools and MD analysis to suggest interaction sites of the LukE toxin subunit with the chemokine receptors.

Reviewer #3 (Public Review):

Manuscript by Lambey et al attempts to understand structural insights how bacterial toxin LukE binds to cytokine receptors. This is an interesting biological problem to understand, which may facilitate development of new antibiotics. Authors determined two crystal structures of LukE at 1.5 and 1.9 A resolutions and three complexes of the toxin with bound p-cresol sulfate or sulfated peptides derived from cytokine receptors. Based on these data, authors provide novel insights into the recognition of sulfated tyrosine residues in the N-terminal region of cytokine receptors. To understand the overall architecture of LukE-receptor complexes, they performed extensive MD simulations supported by previously reported mutagenesis data. Overall, the models of LukE complexes look reasonable; however, these models lack sufficient experimental validation. In addition, more data are needed to map the binding site of N-terminal fragment of cytokine receptors in solution. Binding sites of p-cresol sulfate and peptides with sulfated tyrosines have been obtained from crystal soaking experiments and may be biased by crystal packing. As presented, data are preliminary and need to be strengthened by additional biophysical/structural studies.

Reviewer #1 (Public Review):

Kursel et al. examined the evolution of synaptonemal complex proteins in C.elegans. While the sequence of the SC proteins evolved rapidly analysis of the structure of SC central region proteins from Caenorhabditis, Drosophila and mammalian species revealed that the length and placement of the coiled-coil domains, as well as overall protein length, were highly conserved across species. This conservation in the structure of coiled-coil proteins within the SC led to the proposal that the conserved structural parameters of the SC proteins and their coiled-coil domains could be used to identify central region components of the SC in species where components could not be identified on sequence conservation alone. Kursel et al demonstrated their parameters could be used to identify a transverse filament protein of the SC in the organism Pristionchus pacificus.

Due to high sequence divergence identifying SC proteins in new model systems has been challenging. The identification by Kursel et al. of potential search parameters to identify these diverged proteins will be useful to the those who work on the synaptonemal complex. This approach has the potential to applicable to other types of proteins that show rapid sequence divergence. As the mammalian, fly, and worm SC proteins all displayed different lengths and placements of their coiled-coil domains within their SC proteins this approach is limited by the availability of related identified sequences to the model organism of interest. Additionally, this approach may still yield multiple candidates that fit the structural parameters which will require additional means to ultimately identify the protein of interest. The data in the manuscript supports the authors' claims of structural conservation within SC proteins but only additional applications of their search methods will reveal how useful it is to search for other types of proteins based on structural features.

Reviewer #2 (Public Review):

In this article, Kursel and colleagues sought to identify evolutionary features of components of the SC the are evident in the absence of strict amino-acid conservation. After identifying three joint evolutionary properties of SC proteins - conservation of coiled-coil architecture, conservation of length and significant amino acid divergence - they show that these properties can be used to identify unknown SC proteins in divergent species. Overall, their general conclusion is very well supported and they do an excellent job functionally testing their approach by showing that one identified candidate for a novel SC protein in Pristionchus is in fact a component of the SC. In addition to providing new insight into the evolutionary forces that shape the evolution of SC proteins, this article provides new insight into how one might generally identify functionally similar or homologous proteins despite very deep divergence. Thus, this work has broader relevance to molecular evolution and evolution of protein structure.

There are some places where smaller conclusions need more support. In particular, it is not entirely clear that this triple pattern - conservation of coiled-coil architecture, conservation of length and significant amino acid divergence - is broadly applicable to SC components beyond Dipterans and Nematodes. In particular, the pattern is weaker in Eutherian mammals. Some further investigation is needed to claim that the pattern is similar in mammals. In addition, it is not clear if coiled-coil conservation rather than simply having a coiled-coil domain is important as a mark of SC proteins. A comparison of coiled-coil conservation among proteins that have coiled-coil domains would be needed for this conclusion. Finally, there should be some additional clarification that not all nematode SC proteins have a pattern of insertion and deletion that is limited to regions outside of the coil-coil domains.

Reviewer #3 (Public Review):

The manuscript "Unconventional conservation reveals structure-function relationships in the synaptonemal complex" by Kursel, Cope, and Rog, describes a novel bioinformatics analysis of proteins in the eukaryotic synaptonemal complex (SC). The SC is a highly conserved structure that links paired homologs in prophase of meiosis, and in most organisms is required for the successful completion of interhomolog recombination. An enigmatic feature of SC proteins is that they are highly diverged between organisms, to the point where they are nearly unrecognizable by sequence alone except among closely related organisms. Kursel et al show that within the Caenorhabditis family of nematodes, SC proteins show a reproducible pattern of coiled-coil segments and highly conserved overall length, while their primary sequences are extremely diverged. They use these findings to develop a method to identify new SC candidate proteins in a diverged nematode, Pristionchus pacificus, and confirm that one of these candidates is the main SC transverse filament protein in this organism. Finally, the authors expand their analysis to SC proteins in flies (Drosophila melanogaster and relatives) and eutherian mammals, and show similar findings in these protein families. In the discussion, the authors describe an interesting and compelling theory that the coiled coils of SC proteins directly support phase separation/condensation of these proteins to aid assembly of the SC superstructure.

Overall, this work is well done, the findings are well-supported, and are of interest to meiosis researchers; especially those working directly on the SC. The manuscript is also well put-together: I could barely find a typo. From a broader perspective, however, I'm not convinced that the work provides a new paradigm for thinking about "conservation" in protein families and how to best detect it. Methods that use structural information to detect homology between highly diverged proteins beyond the capabilities of BLAST or even PSI-BLAST are well-developed (e.g. PHYRE2, HHPred, and others). The use of coiled-coil length as a metric for conservation, while it works nicely in the case of SC proteins, is likely to not be generalizable to other protein families. Even within SC proteins, the method does not seem to scale past specific families to, say, allow identification of homology between distantly-related eukaryotic groups (e.g. between Caenorhabditis and Drosophila or Caenorhabditis and eutherian mammals). To be fair, this failure to scale is not because of any limitation with the method; rather, simply that SC proteins diverge quickly through evolution. Overall, however, these limitations seem to limit the application of this method to the specialized case of SC proteins, thus limiting the audience and scope of the work.

Reviewer #3 (Public Review):

Following up on an initial observation that the genome of the black-legged tick encodes an adiponectin-like receptor (ISARL), but lacks an obvious cognate adiponectin homolog, Tang et. al uncover the interesting finding that ISARL is important for colonization of the tick by the Lyme disease agent, Borrelia burgdorferi. Spurred by compelling data that silencing of the ISARL gene significantly attenuates tick acquisition of B. burgdorferi from infected mice, the authors link ISARL production to the differential expression of tick genes involved in metabolism. They show that ISARL mediates regulation of tick phospholipid metabolic pathways and that this phenotype is unique to bloodmeals taken from B. burgdorferi infected mice. Data are presented that support the contention that tick metabolic pathways linked to phosphatidylserine synthase I are critical to spirochete colonization. To investigate potential ligands for ISARL, the authors first examine mammalian adiponectin using knock-out mice. They show that adiponectin regulates glucose metabolism pathways in an ISARL-dependent manner, but has no impact on B. burgdorferi colonization. Instead, a homology search of the Ixodes genome using the C1q globular domain of adiponectin as a query led to the identification of tick C1q-like protein 3 (C1QL3). The authors show that tick C1QL3 regulates ISARL expression and like ISARL is critical for B. burgdorferi colonization. The authors conclude that B. burgdorferi influences tick C1QL3 expression through an undefined mechanism, leading to increased ISARL-mediated signaling effects on metabolic pathways that aid B. burgdorferi colonization of the tick.

Strengths:

This is a well written and carefully designed study that lays the foundation for asking many new questions about the complex interplay between the Lyme disease spirochete, its tick vector, and its vertebrate hosts. I agree with the authors that these findings are also likely relevant to other important arthropod-borne pathogens.

The extensive use of an in vivo system that relies on tick acquisition from the blood of infected mice is a significant strength of the study. By silencing a series of genes in the tick the authors develop a convincing case for the mechanistic relationship of ISARL to B. burgdorferi colonization.

Weaknesses:

1) Potential mechanisms of B. burgdorferi influence on C1QL3 expression are not addressed. While outside the scope of the current work, the manuscript would be improved by some consideration of this issue in the discussion.

2) Adiponectin and C1QL3 are shown to be ISARL ligands that cause differential regulation of tick metabolic pathways. B. burgdorferi infection does not alter adiponectin concentrations in the blood of mice (Fig. 3H). Presumably tick C1QL3 competes with mammalian adiponectin for ISARL-binding, but this is not addressed. Similarly, the homology of murine or human C1QL3 (i.e. CTRP13) is not shown and its potential relevance, along with other C1Q/TNF-related proteins are not discussed in the context of ISARL, but are instead discussed in their known host associated roles only. An improved discussion of how adiponectin, CTRP13, and other C1Q/TNF-related vertebrate proteins may act (or not act) in the tick C1QL3/ISARL pathway should be provided.

3) The authors conclude that mammalian adiponectin/ISARL-mediated glucose metabolism changes have no impact on B. burgdorferi colonization. However, the authors report a significant difference in engorgement weights between the GFP controls and G6p1/2 knock downs. Furthermore, a majority of the samples evaluated for G6P1 exhibited flaB/actin ratios near zero, indicating low colonization, including for the GFP control. The authors should clarify how these factors potentially influenced the claim that glucose metabolism changes (particularly G6p1) did not cause statistically significant differences in B. burgdorferi acquisition.

Reviewer #1 (Public Review):

In this manuscript, the authors present evidence that mouse blood meals containing Lyme disease spirochetes induce upregulation and activation of an adiponectin receptor (ISARL) in the midguts of Ixodes scapularis ticks. Activation of the receptor initiates transcriptional changes, not seen with blood meals from uninfected mice, that give rise to metabolic alterations in the midguts required for replication of spirochetes. Using RNA-seq, they trace these critical alterations to genes encoding enzymes for synthesis of phospholipids. Although mouse adiponectin induced transcriptional changes in engorged tick midguts related to glucose and energy metabolism, it did not influence B. burgdorferi colonization. The authors conclude by presenting evidence that tick complement C1q-like protein (C1QL3), also upregulated in response to a blood meal containing spirochetes, is an ISARL ligand and that knockdown of C1QL3 impairs spirochete colonization.

This work extends our understanding of the complex and intimate physiologic interactions between Borrelia burgdorferi and Ixodes ticks that sustain the spirochete's enzootic cycle. It builds upon prior work by others showing that feeding ticks provide spirochetes with glycerol, an alternative carbohydrate energy source and essential building block for phospholipid biosynthesis. It also appears relevant to previously published work showing that B. burgdorferi can extract lipids from the membranes of eukaryotic cells to which they are attached.

The strength of the study is that it uses state-of-the-art genetic, bioinformatic, and transcriptional approaches to garner novel insights into the unique transcriptional/metabolic changes that occur in ticks when they ingest blood from B. burgdorferi¬-infected mice. It enhances the now well-established, but still far from well understood, viewpoint that ticks are not mere biologic syringes for injection of spirochetes. And it demonstrates, probably more than any preceding study, the extent to which tick midgut interactions with Lyme disease spirochetes re-configure metabolic responses/adaptations to the blood meal. Viewed from these contexts, the major outcomes of this study - identification of ISARL as a midgut metabolic regulator and a tick derived ISARL ligand - are groundbreaking. On the other hand, the main conclusions of the paper, though consistent with the data, are less than definitive. The authors can only infer that spirochetes take up phospholipids produced within the tick midgut following ISARL stimulation, and they stop short of showing that C1QL3-ISARL interactions mediate the transcriptional/metabolic changes involving phospholipid biosynthesis attributed to activation of ISARL during an infection blood meal.

Reviewer #2 (Public Review):

The authors searched for human and murine Adiponectin and Adiponectin receptors homologous sequences in the I. scapularis NCBI database. They found one homologous sequence for Adiponectin receptor 1 and 2, called I. scapularis Adiponectin receptor-like (ISARL) and none for Adiponectin. ISARL showed 71% homology with AdipoR1 and 2 human and murine, 384 amino acids long, and 87% homology with the D. melanogaster ortholog.

Then the authors, characterized ISARL functionally in the tripartite interaction between vector (I. scapularis, deer tick), mammal (mice) and Lyme disease spirochete (B. burgdorferi, bacteria). They used an elegant paradigm by which they intervened the interaction of B. burgdorferi with its vector I. scapularis by injecting siRNAs or adiponectin in the nymphal tick guts to silence or activate ISARL and other proteins of interest. They observed that the blood meal from mice infected with B. burgdorferi increased the expression of ISARL in the tick guts and by silencing ISARL they were able to reduce the colonization of the bacteria in the tick gut without affecting the feeding habits. The silencing of ISARL, however, did not prevent or reduced the ability of the spirochete to infect mice after being biten by the tick.

The authors then, screened for genes related to B. burgdorferi on the tick guts by comparing the RNAseq profile of tick guts when fed from uninfected and infected mice and ISARL were silenced. The comparisons shown in figure 3 were clean and follow a logical line of reasoning. On one hand, the comparison between silenced and non-silenced fed blood meal from uninfected mice showed 17 differentially expressed gene, one of those was the silenced ISARL. On the other hand, the comparison between silenced and non-silenced from infected bloodmeal showed 35 differentially expressed genes, from which one was the silenced ISARL. None of the two sets, showed genes in common except from ISARL. The GO analysis showed that several metabolic pathways were modified by B. burgdorferi. From those the authors chose 18 genes that were robustly represented and confirmed their expression using qPCR 17 of them passed the analysis and four of them changed significantly. They chose Phosphatidylserine synthase 1 (PTDSS1) as a paradigm to silence because it is involved in the synthesis of phospholipids (PE and thus PC) and B. burgdorferi lack the machinery to synthesize them. The silencing of PTDSS1 effectively reduced the content of the spirochete in the tick guts without affecting its feeding behavior and moreover, silencing of PSD another enzyme downstream PTDSS1 also involved in PE synthesis induced the same effect. This was an elegant demonstration that the pathway involved downstream ISARL was the phospholipid synthesis pathway.

Because ISARL resembles AdipoR1 and 2 and bloodmeal may contain its natural ligand adiponectin, the authors investigated the influence of Adiponectin on B. burgdorferi effects on Tick guts. They injected adiponectin or they fed bloodmeal from mice wild type and Adiponectin KO and found in both cases that Adiponectin presence decreases the expression of G6P1 and 2 (2 isomers found in ticks) just as it does in mammalian systems, but the injection of Adiponectin only reduced the expression of two of the three phosphoenolpyruvate carboxykinase PEPCKs found downstream in ticks glycolytic and gluconeogenic pathways (PEPCK2 and 3 but not with 1). On the contrary, Bloodmeal does not reduce any of them. But what it is more important Adiponectin and glucose metabolism does not have any effect in the infectivity or colonization of B. burgdorferi.

Because ISARL respond to ligands, the authors searched for one in the database in the tick using the C1Q motif of human Adiponectin than interacts with the receptor. They found a match of 181 aa (pre-protein) and 157 aa, the protein mature (without the signal peptide). The proposed ligand, called C1QL3, was increased in expression when the tick was fed bloodmeal of infected mice (as it was ISARL), and when silenced feeding behavior did not changed but the content of B. burgdorferi in the tick gut decreased. They demonstrated in a heterologous system (human HEK cells expressing ISARL) that recombinant C1QL3 interacted with ISARL by immunocytochemistry and pull-down assay.

After silencing C1QL3, ISARL expression was decreased and the bloodmeal with infected mice lost the ability to increase the receptor expression level but the tick's gut.

Critique:

Overall, the authors have done an impeccable job in the demonstration of the pathways involving ISARL in the tripartite interaction of mammalian-insect-bacteria system. However, the medical relevance of the interaction portrayed in the present manuscript, although very interesting from the biological-evolutive, point of view remains to be demonstrated and it is an opportunity that was not taken in the discussion. In the discussion the authors just described the systems in the three species which is usually done in the introduction, instead they repeated all the conclusions drawn in results and minus a couple of paragraphs results a waste of space give such a good scientific work made with the results section. I would suggest the authors to concentrate in areas where their work would be elevated challenging the reader with new ideas. For instance, there is literature about the role of Adiponectin receptors in lipids metabolism and uptake that was not mentioned. ADIPOR1 is expressed in the retina and retinal pigment epithelial cells. Mutant of the Adipor1 gene in these cells results in the inability to take up and retain the essential fatty acid family member docosahexaenoic acid (DHA, 22:6,n-3). Therefore, phospholipids in those cells display a selective shortage in DHA, not in arachidonic acid. In addition, the elongation products 32:6 ,n-3 and 34:6,n-3 are depleted. A consequence is photoreceptor cell death and retinal degeneration( Rice DS, Calandria JM, Gordon WC, et al. Adiponectin receptor 1 conserves docosahexaenoic acid and promotes photoreceptor cell survival. Nat Commun. 2015;6:6228-6242.).

ADIPOR1 recently has been shown critical for retinal degenerative diseases for axample a single amino acid mutation of Adipor1 occurs in different forms of retinitis pigmentosa. Also polymorphisms of this receptor have been found in age-related macular degeneration (AMD). AdipoR1 has an adiponectin independent role. This was demonstrated by the fact that adiponectin KO do not change DHA and do not result in retinal degeneration. Therefore it seems that is a regulatory switch of DHA uptake, retention, conservation, and elongation in retinal cells, to sustain photoreceptor cell integrity.

There is also literature regarding PEs and survival pathways involving ceramides, route that was not taken by the authors. It would be specially interesting to analyze this aspect from the point of view of the therapy against Lyme disease. Other aspects would be from the point of view of control of reproduction of B. burgdorferi in tick's population and strategies to control the disease.

Reviewer #1 (Public Review):

This interesting study focused on the roles of mechanical forces generated by actin polymerization and myosin II contractility in Fc receptor -mediated phagocytosis. Through an elegant combination of microparticle traction force microscopy and lattice light-sheet microscopy, the authors identify F-actin 'teeth' that are important for microparticle constriction throughout the phagocytosis process, as well as elucidate the specific roles of Arp2/3 nucleated actin networks, and myosin II -based contractile structures in the process.

The data are of good technical quality and the study provides interesting new insights into Fc receptor -mediated phagocytosis.

Reviewer #2 (Public Review):

In this manuscript, the authors use lattice light sheet microscopy and custom made soft micro-particles to examine the forces generated during phagocytosis and assess the molecular functions and localization of various components of the system. The imaging is truly fantastic and the discovery of phagocytic 'teeth' that exert force normal to the bead surface is a real advance to the field.

However, the functional studies using pharmacological inhibitors are more problematic. Specifically, the authors use pharmacological agents to test the roles of NMII (Blebb), the Arp2/3 complex (CK666) and supposedly formins (SMIFH2). The formin inhibitor is particularly problematic since it has known off target effects such as NMII (Sellers et al) and has never really be validated in terms of specificity or potency. I realize this drug has been used a lot in the literature, but so was BDM before it was finally discredited. It doesn't really give much of an effect and, combined with the fact the two formins checked are not in the cup, this data should just be removed from the paper.

As for the Arp2/3 inhibition, the data showing that this complex is important for generating force at the 'teeth' is convincing, however, the only real straightforward test of whether the complex is required for phagocytosis (Fig. 3g) shows that this drug has no effect on the fraction of engulfed particles. Doesn't this mean that the forces generated by branched actin are irrelevant for the kinetics of phagocytosis? That would be consistent with the published literature showing that genetic deletion of the Arp2/3 complex has only a partial inhibitory effect on FcR phagocytosis of rigid particles (a point the authors avoid discussing). Perhaps the forces generated by Arp2/3-branched actin are important under more challenging conditions such as where sheer flow was affecting the cells/particles, but this part of the paper is problematic.

The blebbistatin data are interesting, but also somewhat contradictory with the literature showing this drug does not affect the uptake of rigid particles. It would be helpful if the authors could compare soft and much more rigid particles with this treatment to test this idea. The localization of myosin at the end stage of phagocytosis is very nice.

Altogether, with some revisions, this work is perfect for the broad readership and will have a significant impact on the field.

Reviewer #3 (Public Review):

Having revealed the role of class 1 myosins myosin 1e and myosin 1f during phagocytosis and having recently developed an innovative method to reveal the forces generated in this process, Daan Vorselen and colleagues studied the molecular mechanisms involved in force production during macrophage phagocytosis. In particular, they documented the involvement in force generation at the phagosome of Arp2/3-based actin protrusions, called « teeth », which assemble into a ring-like superstructure, and of myosin-II, which presumably plays a role in phagosome closure. Finally, they document phagocytosis failures in the form of popping mechanisms. The imaging and mechanical analyses of this article are particularly impressive, and these data allow the authors to propose a new model for the generation and balance of forces during phagocytosis.

This precise work thus paves the way to understanding the generation and transmission of forces during phagocytosis. However, some information could be added to strengthen the impact of the manuscript. In particular, current knowledge about the role of the Arp2/3 complex during phagocytosis could be detailed in the introduction, and what this article adds to the literature on this subject could be discussed better.

Reviewer #1 (Public Review): 

In this paper, the authors recorded in CA1 region of the dorsal hippocampus and medial prefrontal cortex (mPFC) while mice underwent appetitive trace conditioning tasks. The tasks included discrete trials presented every ~40 sec (ITI). In some trials, a neutral auditory stimulus preceded the delivery of reward after a 1-second delay, while in other trials, another stimulus was presented by itself. They reported that a sizable proportion of CA1 and mPFC neurons changed firing rates in response to the stimuli and reward. As ensembles, neurons in both regions differentiated the reward-predictive stimuli from the non-predictive one and between two different stimuli predictive of the same reward. Furthermore, these ensemble firing patterns were reactivated during ITIs, and the timing of the reactivation was locked to sharp-wave ripples (SWR) in CA1. 

The experiments were well designed and yielded exciting, valuable findings. The analyses were sophisticated and generally sound. In particular, direct comparisons of neural selectivity between CA1 and the mPFC in the same non-spatial task is unique and speaks to their potential functional differences. Furthermore, the reactivation of task-related ensemble activity is a relevant extension of the literature on memory trace reactivation, which is built exclusively based on spatial memory.

Reviewer #2 (Public Review): 

This manuscript by Klee et al describes their study on the dynamics of neuronal activities in the hippocampal CA1 and the prefrontal cortex (PFC) in an auditory trace conditioning task. The authors trained mice to respond (with anticipatory licking) to one or two sounds (CS+) associated with water reward delivered after a trace delay period, but not to a control sound (CS-). The authors recorded CA1 and PFC neurons and examined the dynamic changes in the activities of these neurons during a baseline period, the period of stimulus (CS+ or CS-) presentation, the delay period, the reward consumption period, and during the inter-trial interval afterwards. The results are complex. But mainly they found CA1 and PFC neurons responded to and distinguished stimulus identities (between CS+ and CS- or between two CS+ sounds) during the stimulus presentation. During the delay and later reward periods, PFC neurons maintained high level of activities after the CS+ presentation, but CA1 neurons reduced their activities. In both areas, the ensemble activities (states) deviated from the baseline after the stimulus onset and stayed so afterwards. In addition, they found that the CA1 neuronal ensembles reactivated during the inter-trial interval were related to those with decreased activities during the delay period and the activation increased with trials within a session. In contrast, the PFC ensembles were related to those PFC neurons with increased activities during reward consumption and increased slowly with training days. 

Both CA1 and PFC neurons have been extensively studied in spatial navigation tasks. Their activities and dynamics are relatively less examined in classical conditioning tasks, especially in trace conditioning, although important progress has been made. The different dynamics between CA1 and PFC during the delay and inter-trial interval periods identified here are new findings that are insightful. However, these findings are broad and general in nature, and thus are limited in new understanding of how the stimulus identity is encoded during the delay period and how the reactivation during the inter-trial interval contributes to the task.

Reviewer #3 (Public Review): 

Klee et al. investigate prefrontal and hippocampal ensemble activity patterns before and after learning of an appetitive trace conditioning task using auditory cues. They report evoked and sustained firing in both regions during trace periods that retains stimulus identity, with distinct prevalence of trace-period activity suppression in CA1 and enhancement in PFC that emerges after learning for conditioned stimuli. A novel finding reported is the reactivation of task-related neuronal assemblies during awake hippocampal Sharp-Wave Ripples (aSWRs) in inter-trial intervals that shows different learning-dependent changes in the two regions. 

The manuscript adds to our understanding of prefrontal and hippocampal activity patterns in trace conditioning tasks, with the primary novel finding being assembly reactivation during aSWRs. However, there are some key weaknesses in the current form. The findings need to be better grounded in existing literature on prefrontal and hippocampal activity in such tasks. Results are presented separately for the two regions throughout the manuscript, and there is no examination of relationships or coordination across the two regions. Several results will benefit from additional analyses for clarification. 

1) The study uses recordings in both medial prefrontal cortex (PFC) and hippocampal area CA1 to investigate how neuronal activity patterns can encode and maintain stimulus information during delay period of an appetitive auditory trace conditioning task. Animals learned to discriminate between two stimuli, CS+ and CS-, over several days with only CS+ stimulus paired with reward at the end of a trace period, resulting in anticipatory licking behavior. Neural populations in CA1 and PFC showed emergence of distinct response profiles when animals learned the task. Evoked response for CS+ stimuli were selectivity enhanced for CS+ stimuli in both regions. During the trace period, CA1 population activity showed overall suppression of activity compared to baseline firing for CS+ stimuli, whereas PFC showed enhanced trace-period activity. 

This analysis primarily use a subtractive measure (change in firing rate from baseline), and averages across all neurons to infer sustained trace-period suppression in CA1 and enhancement in PFC for CS+ stimuli. Since the population consists of neurons with variable baseline rates and variable levels of responsiveness, the results need to be confirmed using another metric, which can control for this variability and avoid outlier contributions from neurons with high firing rates. 

2) Both regions showed prevalence of Trace-Up and Trace-Down neurons, exhibiting increased and decreased firing during trace periods respectively. In CA1, Trace-Down neurons showed stronger suppression for CS+ stimuli, whereas in PFC, Trace-Up neurons showed stronger enhancement for CS+ stimuli. Population activity in both regions was able to distinguish stimulus identity during trace periods. 

In these analyses, the differing contributions of Trace-up and Trace-down neurons to the overall differential population response in the two regions is not clear. 

3) These results examining CA1 and PFC responses during task periods is presented separately for the two regions. Several previous studies have examined prefrontal and hippocampal activity patterns separately during learning of trace conditioning tasks, and the current results need to be discussed in context of existing findings in literature. Given the interpretation of activity patterns in the CA1-PFC circuit for learning and execution of the task, the manuscript will significantly benefit from examination of potential relationships or coordination between neuronal activity in the two regions. 

4) The key novel finding of the study is that during inter-trial periods, task-related neuronal assemblies were reactivated during awake hippocampal sharp-wave ripples (aSWRs) in both regions, which may play a role in learning of the task. This reactivation of cell assemblies increased significantly from pre- to post- learning sessions for PFC, and within individual pre-learning sessions for CA1. Strongly reactivated assemblies in PFC were related to reward responses, and those in CA1 were related to suppression of activity during trace periods. 

The authors conclude that this reactivation plays a role in learning the task. This claim requires further supporting evidence, including elucidation of relationship of reactivated cell assemblies with activity patterns described for CS+ and CS- stimuli, details about rate and distribution of aSWRs over learning, and examination of relationship of reactivation in the two regions and any potential changes over learning.

Reviewer #1 (Public Review):

The authors present a system that allows the measurement of OCR on diverse tissues. Using two optopes, one before the tissue under examination, and one after, allows the OCR to be measured as the difference between the concentration of O2 in the in-flow gas and the concentration of O2 in the out-flow gas. The system maintains the tissue at a set concentration of dissolved O2 so that experiments can be performed over a long period of time. The authors have provided ample data and full methods and their conclusions are most likely reliable.

Currently, we know that O2 is critical for diverse physiological processes, however it is rarely as well controlled for as well as non-gas solutes such as glucose, as we lack methods to control its delivery and infer its consumption. By addressing this need, the authors contribute something valuable to the field, which will hopefully be built on by others. The authors have already begun to show the utility of their system by exploring the complicated biology of H2S. As delivering this gas in a controlled manner is hard, often people use NaHS instead. In line with previous studies (well cited by the authors), differences are observed.

Specific points

1) The gas control system is used with islets, INS-1 832/12 cells, retinas, and liver tissue, demonstrating its broad applicability.

2) The system as a platform can have diverse extra measurement modalities attached to it, for example visible-wavelength absorbance and fluorescence. Metabolite concentrations in the tissue culture outflow could also be measured.

3) The reduction state of cyt c and cyt c oxidase are measured from the second derivative of absorbance at 550 and 605 nm. Ideally, to reliably decompose these signals full spectra around 550-605 nm would be collected. As the authors are only using cytochrome reduction state as a qualitative measure and appear careful to avoid over-interpretation this method should be fine. However, the authors ought to show a representative time course including the fully oxidised and reduced states demonstrating this approach as making these measurements is demanding and will depend on the exact spectroscopic set-up. Without this information it is hard to judge the reliability of the paper.

Reviewer #2 (Public Review):

The present project is an extension of prior work from this work group in which they describe a technological advancement to their published flow-culture system. Such improvements now incorporate technology that allows for metabolic characterization of mammalian tissues while precisely controlling the concentration of abundant gases (e.g., O2), as well as trace gases (e.g., H2S). The present article demonstrates the utility of this system in the context of hypoxia/re-oxygenation experiments, as well as exposure to H2S. Although the methodology described herein is clearly capable of detecting nuanced metabolic changes in response to variations in O2 or H2S, the lack of a head-to-head comparison with other techniques makes it difficult to discern the potential impact of the technology. In addition, diffusion gradients both in the bath, as well as the tissue itself likely impact the accuracy of the metabolic measurements. This is likely relevant for the liver slices experiments. Following resection, liver tissue can be mechanically permeabilized (PMID: 12054447). In the present experiments, no controls were put in place to discern if the tissue was permeabilized. This could be checked by adding in adenylates and additional carbon substrates and assessing the impact on OCR. Similar controls likely need to be implemented for the islet and retina experiments. Additional comments are detailed below:

- The experiments with H2S are particularly interesting, as this system does seem well suited to investigate the metabolic effects of H2S.

- The authors state the transient rise in O2 consumption was surprising; however, accumulation of succinate during ischemia and rapid oxidation upon reperfusion has been previously demonstrated (PMID: 32863205).

- In the paper, Zaprinast was used to block pyruvate uptake. However, the rationale to use this compound, as opposed to the more specific MPC inhibitor UK5099 is unclear.

- Throughout the paper, the authors list 'COVID-19' as a potential application. It is not clear how this technology could be used in the context of COVID-19.

Reviewer #1 (Public Review):

The study combines theoretical and experimental approaches to probe the laws governing strategy for coping with the control of stepping on uneven terrain. Both congruent results in the anticipatory and reactive adjustments indicate that a simple strategy based on the conservation of energy may be expressed within the neural control pathways. The current version of the manuscript could benefit from the clarification of the study methodology. The development of computational tools supplemented with experimental validation is one of the most effective methods to achieve holistic validity of conclusions and to provide an accurate path to reproducibility and applications.

Reviewer #2 (Public Review):

The authors evaluated whether their previously published model-based predictions of strategies to take an uneven step during walking agree with new experimental observations. Predictions were obtained under the assumption of optimal control based on a simple mechanical model of walking (rimless wheel). The optimality criterion was minimization of mechanical work. Experimental observations supported the following key model outcomes: (1) compensation steps occurring around the uneven step (as opposed to either after or before), (2) pattern of speed fluctuations before and after the uneven step, (3) scaling of speed fluctuations with gait speed. The paper thereby provides additional support for the optimal control hypothesis. The claim that 'humans compensate with an anticipatory pattern of forward speed adjustments, with a criterion of minimizing mechanical energy input' might be somewhat strong, given that the model relied on a set of limiting assumptions (see also below), reasonable alternative modeling assumptions have not been tested, and only a subset of the model predictions published earlier have been evaluated. The conclusions of the paper could be strengthened by:

1) demonstrating that the predictions also hold for a model with variable step length;

2) demonstrating that optimal control also predicts no time lost due to stepping up as compared to walking on even terrain;

3) demonstrating that the number of compensation steps N that minimizes work corresponds to the observed number of compensation steps;

4) demonstrating that minimal work leads to better agreement between simulations and observations than other plausible optimality criteria;

5) demonstrating that the predicted dependence of speed fluctuations on step height is in agreement with experimental observations.

Reviewer #3 (Public Review):

Summary. This is an elegant study on predicting and explaining human locomotor choices via optimality principles. The authors find that the humans exhibit stereotypical speed fluctuations when encountering an up or down step in their path, and they argue that the qualitative aspects of these speed fluctuations can be captured by a simple model minimizing energy over the whole walking task. This article adds to the growing literature on predicting human behavior via energy or other optimality, including transient non-steady state behavior.

Reviewer #1 (Public Review): 

This paper introduces a new statistical framework to study cellular lineages and traits. Several new measures are introduced to infer selection strength from individual lineages. The key observation is that one can simply relate cumulants of a fitness landscape to population growth, and all of this can be simply computed from one generating function, that can be inferred from data. This formalism is then applied to experimental cell lineage data. 

I think this is a very interesting and clever paper. However, in its current form the paper is very hard to read, with very few explanations beyond the mathematical observations/definitions, which makes it almost unreadable for people outside of the field in my opinion. Some more intuitive explanations should be given for a broader audience, on all aspects : definitions of fitness « landscape », selection strength(s), connections between cumulants and other properties (including skewness) etc... There are many new definitions given with names reminiscent of classical concepts in evolutionary theory, but the connection is not always obvious. It would be great to better explain with very simple, intuitive examples, what they mean, beyond maths, possibly with simple examples. Some of this might be obvious to population geneticists, and in fact some explanations made in discussion are more illuminating, but earlier would be much better. I give more specific comments below. 

Major comments : 

- The authors give names to several functions, for instance before equation (1) they mention « fitness landscape », then describe « net fitness » , which allows the authors to define « fitness cumulants ». Later on, a « selection » is defined. Those terms might mean different things for different authors depending on the context, to the point there are sometimes almost confusing. For instance, why is h a « landscape » ? For me, a landscape is kind of like a potential, and I really do not see how this is connected to h. « fitness cumulants » is particularly jargonic. There are also two kinds of selection strengths, which is very confusing. I would recommend that the authors make a glossary of the term, explain intuitively what they mean and maybe connect them to standard definitions. 

- Along the same line, it would be good to give more intuitive explanations of the different functions introduced. For instance I find (2) more intuitive than (1) to define h . I think some more intuition on what the authors call selection strengths would be super useful . In Table 1 selection strengths are related to Kublack Leibler divergence (which does not seem to be defined), it would be good to better explain this. 

- It seems to me the authors implicitly assume that, along a lineage, one would have almost stationary phenotypes (e.g. constant division rate) . However, one could imagine very different situations, for instance the division rates could depend on interactions with other cells in the growing population, and thus change with time along a lineage. One could also have some strong random components of division rate over time . I am wondering how those more complex cases would impact the results and the discussion 

- «  Therefore, in contrast to a common assumption that selection necessarily decreases fitness variance, here we show that under certain conditions selection can increase fitness variance among cellular ». This is a super interesting statement, but there is such a lack of explanations and intuition here that it is obscure to me what actually happens here.

Reviewer #2 (Public Review): 

The paper addresses a fundamental question: how do phenotypic variations among lineages relate to the growth rate of a population. A mathematical framework is presented which focuses on lineage traits, i.e. the value of a quantitative trait averaged over a cell lineage, thus defining a fitness landscape h(x). Several measures of selection strengths are introduced, whose relationships are clarified through the introduction of the cumulant generating function of h(x). These relationships are illustrated in analytical mathematical models and examined in the context of experimental data. It is found that higher than third order cumulants are negligible when cells are in early exponential phase but not when they are regrowing from a stationary phase. 

The framework is elegant and its independence from mechanistic models appealing. The statistical approach is broadly applicable to lineage data, which are becoming increasingly available, and can for instance be used to identify the conditions under which specific traits are subject to selection.

Reviewer #3 (Public Review): 

In this work the authors have constructed a useful mathematical framework to delineate contributions leading to differences in lineages of populations of cells. In principle, the framework is widely applicable to exponentially growing populations. An attractive feature is that the framework is not tailored to particular growth models or environmental conditions. I expect it will be valuable for systems where contributions from phenotypic heterogeneity overwhelm contributions from intrinsic stochasticity in cellular dynamics. 

I am generally very positive about this work. Nevertheless, a few specific concerns: 

1) In here, lineages are considered as fitter if they have more division events. But this consideration neglects inherent stochasticity in division events. Even in a completely homogeneous population, the number of division events for different lineages is different due to intrinsic stochasticity, but applying the methods discussed in this manuscript may lead to falsely assigning different fitness levels to different lineages. The reason why (despite having different number of division events) these lineages ought be assigned the same fitness level is that future generations of these cells will have identical statistics, in contrast with those of cells that are phenotypically different. Extending the idea to heterogeneous populations, the actual difference in fitness levels may be significantly different from what is obtained from the mathematical framework presented here, depending on the level of inherent stochasticity. 

2) In one of the sections the authors mention having performed analytical calculations for a cellular population in which cells divide with gamma distributed uncorrelated interdivision times. It's unclear if 1) within specific sub-populations, cells with the sub-population divide with the same division time, and the distribution of division times is due to the diverse distribution of sub-populations; or 2) if there are no such sub-populations and all cells stochastically choose division time from the same distribution irrespective of their past lineage. If the latter, then I do not see the need for a lineage-based mathematical formulation when the problem can dealt with in much simpler traditional ways which so not keep track of lineages. 

3) The analytical calculations provided seem to be exact only for trajectories of almost infinite duration (or in practice, duration much greater than typical interdivision time). For example, if the observation time is of the order of division time, this would create significant artifacts / artificial bias in the weights of lineages depending on whether the cell was able to divide within the observation time or not. Thus, the results claiming that contributions of higher order cumulants become significant in the regrowth from a late stationary phase are questionable, especially since authors note that 90% of cells showed no divisions within the observation time.

Reviewer #1 (Public Review): 

A role for integrins in lowering the threshold for B cell activation was first observed over 15 years ago, but the mechanism has remained elusive. In this paper, Wang et al. investigate the role of LFA-1:ICAM-1 ligation in B cell synapse formation using live-cell super-resolution fluorescence microscopy in both primary B cells and the A20 B cell line. The use of super-resolution imaging is critical to the investigation as it reveals a level of organisation of the actomyosin network that is not visible with conventional microscopy approaches such as TIRF microscopy. They find that LFA-1:ICAM-1 ligation promotes the formation of actomyosin arcs that regulate various activities in the B cell synapse including BCR signalling, BCR:antigen microcluster transport, and the centralisation of antigen. In agreement with earlier studies, they show that LFA-1:ICAM-1 ligation is required for B cells to centralise antigen that is present at very low density. They also demonstrate that myosin IIa contractility is required for the formation of the actomyosin arcs and promotes the exertion of strong traction forces on the antigen- and ICAM-1-presenting substrate. Using a small molecule inhibitor of formin activity in combination with miRNA knockdown of the formin mDia1, the authors show that the actomyosin arcs originate at the outer edge of the synapse and that their generation is formin dependent. These data provide a much-needed advance to our understanding of the role LFA-1 plays in the earliest events in B cell responses to antigen. 

The conclusions of the paper are mostly well supported by the data, but there are a few points that would need to be clarified. 

1) The requirement for LFA-1:ICAM-1 ligation in the formation of the actomyosin arcs is not clear. The authors observe that ~30% of B cells form actomyosin arcs with anti-IgM stimulation only (Figure 1). Does LFA-1:ICAM-1 ligation simply stabilise the arcs and therefore make their appearance more likely, or does it promote the formation of a distinct actomyosin network with unique functions? The images and videos selected to represent cells stimulated with anti-IgM only (Figure 1; Movies 1A and 1B) seem form a highly branched actin network throughout the synapse, but it would be informative to see cells having the actomyosin arcs for comparison. Since B cells stimulated with anti-IgM alone are capable of signalling and centralising antigen, it would be interesting to know whether and how these two populations (with and without arcs) differ. 

2) The authors propose that the contractile actomyosin network formed in the presence of LFA-1:ICAM-1 interactions promotes B cell activation especially at low antigen concentrations; however, their data focus only on early signalling (pCD79a and pCD19) and it would be helpful to know whether LFA-1:ICAM-1 interactions impact signalling further downstream. 

3) The observation that some GC B cells centralise antigen is very interesting, but there are a few aspects of this investigation that should be expanded upon. The authors show that with LFA-1:ICAM-1 interactions, GC B cells are about equally likely to organise BCR:antigen complexes into peripheral clusters and centralised clusters. It would be informative to have, in the same study (Figure 7), a comparison with GC B cells stimulated with antigen alone. The reason is that other studies investigating GC B cell synapse architecture did not quantify antigen organisation in this way, so it is difficult to make comparisons with previous work. It would also be very useful to see how the actomyosin network is organised in GC B cells exhibiting different synaptic architectures (i.e. peripheral versus central clusters), especially given the critical role of myosin IIa activity in GC B cell antigen affinity discrimination. Additionally, while it is a very interesting observation that LFA-1:ICAM-1 interactions may affect GC B cell synapse organisation, it is not clear whether this has an impact on cellular function. For instance, does antigen and actomyosin organisation in GC B cell synapses contribute to differences in signalling or traction force generation? In the introduction the authors suggest that actomyosin arcs contribute to antibody affinity maturation (line 87-88), but without functional studies to support this claim I think it is too speculative.

Reviewer #2 (Public Review): 

The manuscript utilizes elegant imaging tools to describe the contractile actomyosin arcs, induced by integrin-ligation, and their involvement in antigen gathering in B cells. The findings are important and have the potential to make a considerable impact in the field. The main conclusions are well supported by strong data and the manuscript convincingly brings across the need of integrin-ligation to induce generation of the arc network and the role of this structure in antigen gathering. The methods and the quality of imaging are state-of-the-art and provide an important example for future studies in B cell immune synapse. Some aspects of the study would benefit from clarification and extended experimentation or analysis. 

1) In addition to cultured B cells, the work includes naïve primary B cells as well as isolated germinal center B cells. While the use of primary cells adds value to the study, in most cases the cells are activated first with LPS prior to transfection with F-Tractin constructs. Such a treatment is likely to alter the cytoskeletal features of the naïve B cells and, thus, it would be informative to provide an analysis of this effect. 

2) Various analyses in the manuscript rely on quantifying the different parts of the synapse, dSMAC, pSMAC and cSMAC. The regions are segmented manually and this is likely to be relatively ambiguous at times, especially in cells where the arc network is hampered. Yet, the segmentation has a direct and strong effect on the results, which is why this aspect would benefit from more emphasis. 

3) The example images are beautiful and detail robust arc organization, or lack of it in given conditions. However, it also seems likely that if only 70% of the cells have detectable arcs there are also a lot of cells with very poor arc content raising questions if a wider selection of representative images would be needed to evaluate the data. It would be important to describe the true variability of the cytoskeletal organization in the samples. 

4) The data on mDia1 would be strengthened by rescue-experiments or a scrambled sequence control. 

5) The authors show actomyosin arcs also in germinal center B cells data but while the functional role of arcs in antigen gathering or uptake is suggested it is not tested.

Reviewer #3 (Public Review): 

The work 'A B cell actomyosin arc network couples integrin co-stimulation to mechanical force-dependent immune synapse formation' by Wang et al. describes the importance of integrin mediated B-cell co-stimulation for IS formation in B-cells by fostering the formation of myosin II A driven actin arcs that are essential in the transport of IgM clusters towards the IS center. 

The work presented here, i.e. experiments and analysis, is very thoroughly done and includes tests and controls using different labelling strategies and constructs of myosin II A, multiple cell types including primary cells and a range of chemical inhibitors to rule out artefacts. 

The authors claim that the observation of actin arcs in B-cells co-stimulated by ICAM-1 - LFA-1 interaction is important for the efficient activation of B-cells in the presence of limiting levels of anti-IgM and this is very well supported by the experiments. However, it was a bit surprising that the paper did not draw much of parallels between the observed phenomenon and the reported actin arcs in activated T-cells even though some of the authors were very much involved in such work on T-cells. If there is a good reason to believe there is no ground to draw comparisons, this would then also need to be highlighted by the authors. 

The work on establishing the drivers of actin arc formation and dynamics is well done, but it is important to note that previous work has analyzed actin arc formation in other cell types. Work by Bershadsky has already established many 'ground rules' for the formation of actin arcs and the role of integrin adhesion, formin activity and myosin II in the process (Tee YH, Shemesh T, Thiagarajan V, Hariadi RF, Anderson KL, Page C, Volkmann N, Hanein D, Sivaramakrishnan S, Kozlov MM, Bershadsky AD. 2015. Cellular chirality arising from the self-organization of the actin cytoskeleton. Nat Cell Biol 17:445-457. doi:10.1038/ncb3137). It might be very instructive if the authors could put their findings in relation to this work. 

The formation of actin arcs is also well studied in U2OS cells and the results presented here could highlight interesting general features of this process observed in very different cell types (Tojkander S, Gateva G, Husain A, Krishnan R, Lappalainen P. 2015. Generation of contractile actomyosin bundles depends on mechanosensitive actin filament assembly and disassembly. Elife 4:1-28. doi:10.7554/eLife.06126; Bur-nette DT, Shao L, Ott C, Pasapera AM, Fischer RS, Baird MA, Der Loughian C, Delanoe-Ayari H, Paszek MJ, Davidson MW, Betzig E, Lippincott-Schwartz J. 2014. A contractile and counterbalancing adhesion system controls the 3D shape of crawling cells. J Cell Biol 205:83-96. doi:10.1083/jcb.201311104). 

In this regard, the findings about the importance of myosin II A activity, integrin adhesion and mDia1 in the formation of actin arcs is not that surprising and the authors might rather highlight the important role of these newly studied structures for co-stimulation in B-cells as this is the more novel and insightful bit of the work.

Reviewer #1 (Public Review): 

In this manuscript, Angela Kim et al. use a combination of in vitro and in vivo studies to determine how glucose-control of central AVP release controls pancreatic alpha-cell calcium influx and glucagon secretion to modulate blood glucose homeostasis. The manuscript clearly shows that activation of AVP release from magnocellular AVP neurons stimulates pancreatic islet glucagon secretion. Furthermore, the manuscript finds AVP (measured by circulating Copeptin) is elevated in plasma following insulin induced hypoglycemia, which also activates AVP neuron electrical excitability and calcium entry. To confirm that AVP release stimulates glucagon secretion via islet alpha-cell Avpr1b activation, both Avpr1b antagonists and an Avpr1b-/- mouse model were utilized. Finally, the manuscript looks at plasma AVP in humans undergoing a hypoglycemic clamp; while this results in AVP release in non-diabetic controls, AVP release is blunted following hypoglycemia in type-1 diabetic patients. Based on an extensive amount of high-quality data, the authors conclude that AVP release from magnocellular AVP neurons is involved in regulating glucagon secretion in response to hypoglycemia. The manuscript is well written and easy to follow. As the exact mechanism that controls glucagon secretion is still unknown, this manuscript adds important information for the diabetes research community detailing the importance of CNS control of islet glucagon secretion through glucose regulated AVP release. Overall, this is an excellent manuscript that will be very useful to the diabetes research community.

Reviewer #2 (Public Review): 

The authors cover a lot of ground, physiologically, by expanding from the islet up through multiple regions of the brain, but they do so in a manner that is stepwise and logical. And in the end, their efforts in probing further and further up the pathway results in a clean model of hypoglycemia sensing through to glucagon release. How AVP fits into counterregulation has been unclear, but Briant and colleagues are filling that gap. The paper is well written, the data are of high quality and well presented. 

Specific comments: 

• Lines 137-139 state that reducing glucose from 8 to 4 mM does not stimulate glucagon from ex vivo islets, but the experiment does not appear to show glucose being reduced. Rather, islets were incubated in separate glucose concentrations and the glucagon from the separate wells was then measured. Methods indicate that islets were incubated at 3 mM prior to each treatment, so glucose was actually raised from 3 to 4 mM and separately from 3 to 8 mM. Suggest either changing the wording or show a perfusion secretion experiment demonstrating the drop from 8 to 4 mM.

• Lines 195-196 & Figure 3f: If YM254890 blocks AVP-induced calcium, please indicate with statistics comparing frequency in AVP and AVP + YM

• Adrenergic signaling as a method of physiological glucagon stimulation is dismissed multiple times, yet is not tested/compared with AVP. The known and robust activation of calcium by AVP in alpha cells notwithstanding, epinephrine is a strong activator of alpha cell calcium responses and glucagon secretion. In multiple panels of the paper, blockade or deletion of Avp1rb reduces, but does not prevent hypoglycemia-induced glucagon secretion, which demonstrates that AVP is not the only signal stimulating alpha cells under these conditions.

Reviewer #1 (Public Review):

This study presents new phylogenomic and molecular clock analyses of echinoids. The manuscript is very well written and presented; the figures are informative and clear. The introduction nicely sums up the main questions for readers not acquainted with echinoids, which adds further value to this study that is relevant for a wide readership.

The phylogenomic dataset includes data for several new lineages, which allows the authors to clarify several unsettled relationships in the echinoid tree of life. Molecular clock analyses explore the effect of several key methodological decisions (clock model, complexity of evolutionary model, gene sampling). This is done using a new interesting approach implemented in the tool 'chronospace'. The new approach to exploring the sensitivity of molecular clock analysis is very relevant and the new tool seems very promising for the community. Perhaps I missed the exploration of yet another key decision in molecular clock analyses: the effect of different prior probability distributions for fossil calibrations.

Overall, the data are thoroughly explored and the conclusions reached by the authors are strongly supported by their data. Results are relevant towards better understanding the evolutionary history of echinoids and provide interesting clues for re-evaluating their fossil record. The new tool 'chronospace' is likely to be highly adopted by the phylogenomics community.

Reviewer #2 (Public Review):

In this paper Koch et al. present an updated phylogeny of Echinoids. The novelty here lies on the author's inclusion of 17 newly sequenced genomes. This new data, together with previous molecular data matrices are then used to estimate the tree topology and divergence times under various software, substitution models, priors, etc. The impact of these analysis set-ups is assessed by using multivariate statistics. This approach is original and helps visualise the data (but certainly discarding a lot of information as all PCA-like methods do). The importance of this approach appears overstated.

The work shows attention to detail, is extensive and the figures are well presented. The text appears long in places and could be shortened to make a tighter narrative.

The authors find relaxed rate models (autocorrelated vs independent rates) have a big impact on time estimates. This is an important result. So, which model is right? Perhaps a Bayesian model selection analysis on a subset of the data may shed light on this issue. Then the authors could focus their discussion of time estimates on the sets of results obtained under the best fitting rate model.

Reviewer #1 (Public Review):

The work aims to increase our understanding of the relationship between young transposable elements and repressive mechanisms. Specifically, it is focused on variable methylation patterns observed at a subset of elements called metastable epialleles. First, the authors provide an analysis based on sequence clustering and CpG content of mouse transposable elements. They then extend their analysis to human elements, recapitulating some of their observations. Finally, they showcase a new hypothesis suggesting that CFP1 plays a role in the mechanism of establishment of these methylation patterns.

Strengths

- The way it sometimes presents information about classification of VM-IAPs elements is interesting. Even if the findings have been recently published elsewhere, the ChIP signal overlaid on multiple alignment helps pinpoint the exact regions that are susceptible to be impactful. However it is puzzling why they didn't use the same approach for Figure 2.<br> - The analysis of human transposons showing that those that are known to be variably methylated share some of the same patterns (High CpG and evolutionary recent) as VM-IAP is of interest.<br> - The demonstration that Trim28 haploinsufficiency is reactivating many young transposable elements is informative, although I would have expected more detail about VM-IAPs and a breakdown per clade for IAPs to link it with the rest of the manuscript.

Weaknesses

- A lot of the first half of the article have been described before (in Bertozzi et al, eLife 2021, Elmer et al, Mobile DNA 2020 and Bertozzi et al., PNAS 2020. While this is acknowledged, I feel there could be more new insights presented from an in-depth analysis - for example with a zoom-in on precise motifs / binding sites being different between clades.<br> - The use of public datasets of various strains is a source of potential concern in this specific case - VM-IAPs and their KZFP controllers are sometimes strain specific, and information should be provided about the precise strains for every public dataset used. The same strain should be used whenever possible.<br> - Similarly, the use of cell line as proxies for liver human data when there is primary tissue datasets available (from GTEx for example) is less than ideal.<br> - The absence of DNA methylation when KZFPs are not present, such as in their Tc21 model, is expected and previously described (Wiznerowicz, JBC 2007).<br> - Most importantly, the CPF1 results do not demonstrate a causal relationship as the authors claim - all I see is that unprotected IAPs are accessible to DNA binding proteins, which is not surprising in the slightest - it is the case for CPF1, and I'm sure many dozens of other transcription factors.

Reviewer #2 (Public Review):

ERVs constitute around 10% of the mouse genome and are usually kept under control through host silencing mechanisms. However, a few ERV families are able to escape such control, as IAP elements. In "Mechanisms regulating interindividual epigenetic variability at transposable elements", Costello and colleagues address the establishment and maintenance of IAP epialleles by thoroughly searching IAP sequence variants that match activation signals (lack of silencing complexes and presence of permissive chromatin marks). More precisely, the authors take advantage of a previously published set of IAP loci that are variably methylated between individuals (VM-IAP), to perform multiple sequence alignment and hierarchical clustering of IAP LTRs, and IAP-ez internal regions, and search for VM-IAP enrichment in the obtained sequence clusters. The authors are able to demonstrate that VM-IAPs are enriched in specific clades, and such clades lack KZFB binding, and show decrease KAP1 enrichment. The authors also suggest that VM-IAPs are mainly found within 50Kb of a constitutively expressed gene, and conclude that this proximity might be necessary for variable IAP methylation. Finally, the authors broadly study CpG content and chromatin regulation, and suggest that permissive chromatin marks can be deposited at VM-IAPLTRs thanks to CpG rich regions. The final model suggests endogenous retrovirus that harbor high CpG content and also loss of KZFP binding, are the target of ZF-CxxC proteins, recruiting K4me3 to the IAP copy.

In general, I find the article to be well written and the results robust. The conclusions are of course complicated as they illustrate the biology of IAP regulation: it depends on the sequence, it depends on the chromatin context, it depends on the genomic background. I find the authors could discuss such complexity, specially within the framework of already published research articles on IAP regulation.

The usage of transchromosomic mouse model is elegant to show the equilibrium between KZFP targeting and K4me3 deposition, along with the Trim28 haploinsufficiency experiment. However, one of the main weaknesses relies on the material and methods section, where unfortunately it is difficult to understand how some of the analysis were performed: uniquely/multimapped ChIP-seq, alignment of LTR elements belonging to full-length copies, and VM vs non-VM comparisons.

Reviewer #3 (Public Review):

Building on previous studies that identified ~50 so called metastable alleles driven by transposable elements (TEs) that exhibit variable methylation (VM) between genetically identical mice, this paper aims to assess whether genetic sequence and sequence-specific TE-binding genetic modifiers contribute to VM in mammals. To investigate genetic determinants of metastable alleles, the authors focus primarily on mouse-specific TEs of the IAPLTR1 and 2 class known to constitute the majority of mouse metastable alleles, but also provide functional data on human VM- LTR12C elements.

Strengths:

The paper identifies sequence variants that characterize IAPLTR1 and 2 elements enriched for VM-IAPs and makes use of a broad array of published sequencing data to characterize these sequence variants further. As sequence variants alone are not sufficient to establish VM-IAPs this bioinformatic analysis is important to advance the understanding of metastable alleles. The use of a diverse array of available datasets from different tissues is appropriate as VM-IAPs are generally retained across adult tissues. By looking into the known genetic modifiers of the KZFP family (previously implicated in metastability and TE repression) at IAP-LTR1 elements the authors show that sequence variants enriched for VM-IAPs correlate with diminished sequence-variant-specific KZFP binding and diminished binding of the KZFP co-repressor KAP1. This observation indicates escape of sequences enriched in VM-TEs from silencing by the KZFP/KAP1 system that evolved to repress TEs. The authors convincingly show that these VM-prone IAP-LTR1 variants are also the least conserved across mouse strains, a correlation that also holds true in other mammals, as previously identified human variably methylated loci are also shown to be enriched for young TEs.

As diminished KZFP binding occurs across VM and non-VM IAPLTR1 elements alike, the paper uses annotated regulatory sequences and proximity to constitutively expressed genes as a proxy for a euchromatic context, which they find enriched at VM, but not no-VM-IAPLTR1 elements suggesting that this chromatin context contributes to establishing metastable alleles. Like others before them, the authors find mouse and human VM young TEs to have also a particularly high CpG density. As CpG density is also shared among VM and non-VM TEs in mice the authors assess the binding of the CxxC-domain containing protein TET1, known to bind to CpG rich regions, and the active histone mark H3K4me3, which they find both enriched specifically at VM-prone IAPLTR1 sequence variants. While expected, this correlation is novel in the context of metastable alleles and lets the authors propose the interesting hypothesis that competition between KRAB and CxxC proteins may contribute to establish metastable alleles. Notably, H3K4me3 at ERVs has also been found to be enriched upon KZFP or KAP1 depletion in mouse ESCs, consistent with their model.

To functionally test the model build on these correlations, the authors use an elegant Tc1 mouse model that harbors human chr21 and with it human-specific variably methylated TE-derived CpG islands. The remaining mouse genome lacks the cognate human KZFPs encoded on other chromosomes not present in the mouse. This model is thus valid to test the hypothesis. The gain of the mouse ZF-CxxC protein CFP1 at a variably methylated LTR12C element in the Tc1 model, but not in human HepG2 cells (that encode both human TEs and cognate KZFPs) convincingly shows that the VM LTR12C element gains binding of the CxxC protein CFP1 when KZFPs are absent and thus supports the main claim of the paper. Supporting their claim, TE-derived sequences targeted by KZFP shared between mouse and human were highly methylated in this model and largely not bound by CFP1 regardless of their CpG content. Thus, their experimental data suggest that TEs that are VM can gain CFP1 binding when their cognate KZFP is missing.

To support that the absence of CxxC domain containing proteins causes changes in methylation, the authors use available mouse DKO cells for the ZF-CxxC proteins TET2 and 3. In this model, the absence of ZF-CxxC proteins leads to increased methylation at VM-IAPs, but not non-VM IAPs thus supporting a causative role of CxxC proteins in maintaining a hypomethylated state at metastable alleles in mice.

The hypothesis that TEs prone to VM rely on escaping the KZP system is further substantiated by profiling of CFP1 binding and H4K4me3 in livers of mice lacking one copy of KAP1, a universal co-repressor used by KZFPs to repress TEs. In livers of these mice, which are known to affect metastability at the agouti coat color gene, ~70 loci gain H3K4me3. These loci susceptible to KAP1 levels are largely young TEs bound by CFP1. Analysis of upregulated TEs in these mice identifies them as particularly CpG-rich supporting a model in which reduced KAP1-mediated silencing would allow CpG-rich TEs to be bound by CFP1 to potentially become metastable alleles.

Weaknesses:

The authors propose that specific sequence features of the subclass of IAPLTR1 and 2 elements that often become VM-IAPs lead to lowered KZFP recruitment and thus 'escape' from KZFP-KAP1 mediated silencing. However, the authors do neither describe the specific sequence changes they refer to, which seems relevant, nor call sequence motifs over these sequences, which could uncover whether it is indeed direct competition between KZFPs and CxxC domain containing proteins that leads to the establishment of VM-IAPLTRs.

Similarly, decreased KAP1 recruitment is central to the conclusions of the paper that TEs prone to be VM escape KZFP repression, but KAP1 coverage heatmaps are only shown for IAPLTR1 regions. It thus remains unknown whether KAP1 recruitment is also decreased at internal IAPez sequences flanked by these LTRs that also show diminished mouse-specific KZFP binding or at VM-IAPLTR2s. The lack of the analysis of KAP1 binding seems particularly relevant for the latter IAPLTR2s, as these seem not to lack KZFP binding, at least for the KZFPs assessed (the KZFP ZFP429 (S1.6)). That these class1 IAPLTR2s are nevertheless often VM seems to contradict the model, unless the cognate KZFPs are less efficient at repressing the TE, which Kap1 density plots could reveal. Alternatively, class 1 IAPLTR2s could depend on diminished recruitment of another KZFP or be KZFP-independent. Currently, such analyses are however challenging, as the genomic binding sites of most mouse KZFPs (unlike human, which are better characterized) are unknown. Similarly, internal sequences flanked by class1 IAPLTR2s, unlike class 3 IAPLTRs, are not proximal to the mouse-specific KZFP Gm14419. Here, this lack of diminished KZFP binding ,may however reflect that most VM-IAPLTR2 sequences are solo-LTRs.

While the human data in the Tc1 mouse model provides convincing evidence for the main conclusion of the paper, such evidence is not presented for mouse, which is surprising given the focus of the paper on mouse-specific VM-IAPs. The majority of the mouse data (except for the TET2/3 DKO and KAP1-/+ liver experiment) are correlative. While the correlations characterizing the identified sequence variants in mice support the main hypothesis, they e.g. do not demonstrate that sequence contributes to establish VM-IAPs, that genomic context influences variable methylation or that high CpG density is important to establish variable methylation and also does not show that mouse TET1 is recruited when mouse KZFPs are absent.

This concern also applies to conclusions made on some of the human data. While the human data does demonstrate CFP1 recruitment in the absence of human KZFPs at a LTR12C element, it does not demonstrate that CpG density establishes LTR12C elements as variably methylated. Rather it shows that VM LTR12C elements are also CpG dense, which is a correlation.

The authors suggest that VM-IAPs have unique sequence variants that allow them to have diminished KZFP recruitment and recruit CxxC-domain containing proteins, but only if the genetic context permits it (as sequence variants characterize both VM and non-VM IAPs). Other than being euchromatic it remains open, what exactly this genetic context entails and if it indeed causes VM.

Like others before, the authors observe high CpG content at VM-IAPs and propose that this hypomethylation is caused by the recruitment of CxxC domain containing proteins such as TET1 and CFP1 that can induce active histone marks or DNA demethylation, respectively. It would require further functional studies to understand the mechanism of diminished KZFP recruitment in establishment of VM despite identical sequence to non-VM-IAPs that also have diminished KZFP recruitment.

The conclusions of this work will likely stimulate and impact the further study of KZFP and CxxC proteins in the establishment of metastable alleles. The elegant use of the Tc1 mouse model could be exploited by the field to dissect the mechanism that establishes metastable alleles further. Given the recent discovery of abundant species-specific KZFP binding to species-specific TEs in human and mice, this work adds to the significance of this binding not just for mere TE repression but gene regulation with relevance to biological processes, in this case the establishment of metastable alleles. As metastable alleles in mice control phenotypes such as coat color and obesity, these results are relevant for our understanding of phenotypic variation in genetic identical individuals. It will be exciting to elucidate the mechanism by which KZFP and CxxC proteins confer metastability to some but not other species-specific TEs.

Reviewer #1 (Public Review):

Summary of what the authors were trying to achieve

Background: Myopia (short- or near-sightedness) is an ocular disorder of increasing concern to human individuals and health-care systems; these days one speaks of a "myopia epidemic" in developed countries. Usually it is due to excessive elongation of the optic axis of the eye during the ages of most rapid growth (ca. 5-16 years in humans), causing images of distant objects to be blurred at the retinal photoreceptors. The optical error can be corrected with lenses or corneal surgeries, but this does not reduce the risk of continued progression and vision loss. Despite extensive epidemiological and animal studies in the past several decades, the underlying causal mechanisms remain poorly known, and therapeutic options are limited. Therefore, further discovery of new candidate mechanisms, drug targets and drugs for inhibiting the onset and progression of myopia is urgently needed.

Rationale: The axial length of the eye is regulated mainly by qualities of the visual environment, including light intensity, spectrum, and spatiotemporal characteristics of images on the retina. Thus the retina encodes and integrates visual information over time, and ultimately sends regulatory "grow" or "stop" signals via the choroid - a vascular plexus behind the retina - to the sclera, the fibrous outer coat of the eye. Changes in size (area) of the sclera are responsible for changes in axial length, and thereby, refraction. The choroid is in a critical position, not only to relay "stop" or "go" messages to the sclera, but also potentially to critically modify those signals (or generate signals of its own) and further modulate ocular elongation and refraction. Importantly, very little is known about how the choroid fulfills either of these roles.

Aims of the Study: The authors' purpose was to test, in juvenile chicken models, whether the 'pro-inflammatory' cytokine, interleukin-6 (IL-6) - synthesized and released in the choroid - might play a key role in the developmental regulation of axial elongation and refraction of the eye.

Major strengths and weaknesses of the methods and results

Strengths:

1. The studies are focused on the choroid, which must be important in regulating ocular growth and refraction, but whose role is still not well understood

2. Expert use of front-line tools for quantifying mRNA and protein (microarray, RT-PCR, ELISA)

3. Immunohistochemistry: Good choice of antibody (raised to chicken IL-6), appropriate specificity control (preabsorption with chicken antigen)

4. IL-6 mRNA in choroid was impressively increased during recovery from form-deprivation myopia (FDM) (preliminary results, Fig. 2) - i.e., during strong positive (myopic) defocus - a defocus-dependent effect confirmed by a similar effect of lens-induced myopic defocus (Fig. 5).

5. Good data for the time-course of IL-6 mRNA content in choroid, with some confirmation of protein levels (though at only 2 treatment intervals) (Fig. 3)

6. Choroidal IL-6 mRNA also shown convincingly to increase, going from darkness to light (Fig. 4).

7. It's clever to compare the growth- and myopia-inhibiting effects of positive defocus, with those of other treatments known to do the same - in this case, atropine and nitric oxide (NO). The evidence shows that the effects of these agents on choroidal IL-6 mRNA are similar to the effect of positive defocus, with an NO-donor increasing the amounts of IL-6 mRNA and protein in isolated choroid (Fig. 7), and a NOS-inhibitor decreasing the mRNA levels at an intravitreal dose that inhibits scleral growth (Fig. 6).

8. If my calculations are correct, 0.1% atropine sulfate solution has a molarity of something like

1.3 mM. Since alpha-2A adrenoreceptors are present in the choroid, of mammals at least (e.g., Wikberg-Matsson et al., 1996, Exp Eye Res, 63(1):57-66), it might be interesting to explore the possibility that atropine is stimulating IL-6 production in the choroid by acting as agonist via these receptors (cf. Carr et al., 2018, IOVS, 59m2778-2791). The isolated choroid, with IL-6 mRNA and protein synthesis as read-outs, should be an exceptional (and novel) model for testing this and other possible signalling pathways in the choroid.

Weaknesses:

1. Immunolocalization of IL-6: The images (Fig. 1) are not good enough to identify cellular localization of immunoreactive structures; identification of RPE is questionable (no DAPI+ nuclei in labeled 'RPE'); nucleated erythrocytes should be visible in vessel lumina.

2. Many important details of methods have been left out. Spectral peaks of LED light-sources need to be given, lines 409-412; that's just one of many examples.

3. Intensity (illuminance) of "red" and "blue" lights seems unnecessarily low (58 and 111 lux, respectively, far below the "medium" and "high" intensities of white lights that were used; Fig. 4).

4. Also, given that red and blue lights have been found to have opposite effects on FDM in chicks (e.g., Wang et al., 2018, IOVS, 59(11):4413-4424), the similarity of IL-6 responses to red and blue in the present study strikes me as a point against a role for IL-6 in regulation of eye growth.

5. I admire the thoroughness of confirming that some of the treatments did in fact have the predicted effects on ocular enlargement, by performing assays for scleral proteoglycan synthesis. This might not be essential to this work, although it is well done, and the scleral data won't detract from the value of the paper if retained. But the induction of opposite effects on eye (scleral) growth by such manipulations is well established, and much simpler (cheaper and faster) refraction and/or caliper measurements would have served the same purpose.

6. I don't buy the argument that the source of NO is not in the choroid (lines 337-340), based on the failure of L-Arg to change significantly the amount of choroidal IL-6 mRNA (Fig. S1). Several thoughts come to mind here: (a) It is solidly established that the choroid is richly innervated with NO-synthesizing nerve fibres, and that its content of NOS is very high [e.g.: "NOS activity is widely distributed in the eye, (choroid > retina > CP > TM) ..."; Geyer et al., 1997, Graefes Arch, 235(12):786-93; also (among others): Wu et al., 2007, Brain Res., 1186m155-63; Hashitani et al., 1998, J Physiol, 510(1):209-223; Fischer & Stell, 1999, cited in the present MS.]. So, there clearly are sources of NO within the choroid, in chicks as in mammals. (b) It's extremely unlikely that "NO, released from the retina ... diffuses to the choroid to stimulate IL-6 synthesis", because NO is highly reactive and has a short half-life, restricting its diffusion. But yes, NO generated by iNOS in the RPE certainly could reach choroidal targets; is there anything in the literature to indicate that iNOS mRNA and protein are increased in the RPE, under conditions or treatments that inhibit axial elongation? (c) The critical experiment to test this idea - treating the isolated choroid with a NOS-inhibitor, to block synthesis of NO by cells in the choroid - was not performed here.

That would be a complicated and difficult exercise, however, requiring the invention of a way to stimulate NOS activity to a new base-level, and then being able to detect effects due to the inhibition of NO-synthesis. It would be good to discuss the issues raised in this point, but acceptable to suggest this as another of the questions that would be suitable to address by further experimentation beyond the scope of this paper.

7. Since it's overwhelmingly likely that NO is synthesized and released locally in the choroid, alternative explanations must be considered for why the NO-donor, PAPA-NONOate, caused increases in IL-6, while L-Arg didn't. Might it have been the case, for example, that the NOS- containing choroidal cells already were fully loaded with L-Arg, under these particular experimental conditions? or that the administered concentration of L-Arg was sub-optimal? or that the proportions of cellular mass to fluid volume in the choroidal samples were highly variable, causing high variance of the individual values? or that the parent compound PAPA- NONOate, however attractive 'his' name, had destinations in mind (molecular targets, actions) in addition to or other than sGC? Any one of these hypotheses might account for the fact that

L-Arg reduced the mean level of IL-6 mRNA by almost 50%, but with p=0.14 despite the sample size n=16.

8. The results of the bulk assays - of whole choroids - are a good beginning, starting to build a map of largely uncharted territory; but they will never be completely satisfactory for constructing signalling pathways or networks for visual regulation of scleral expansion, and will leave one struggling to make sense of it all (cf. lines 345-347). Better immunolabeling, with better image definition and resolution, the addition of single-label images and bright-field images (to locate the RPE securely), and possibly FISH would be helpful for this. If you're rich, have great local resources, and/or are well connected with others who do, scRNA-seq of dissociated choroidal tissue (with RPE and sclera as controls) would have great potential here. If the tissue has been perfused intravascularly or well washed and drained, to get rid of blood cells, there shouldn't be very many cell types to characterize (but, my, wouldn't it be exciting and illuminating if there were!)

9. The relationships between the studies and outcomes reported in this manuscript, and the possible role of choroidal IL-6 and other inflammation-signalling molecules in myopia, is hardly touched upon at all - just a short, very general statement near the end of the Conclusions (lines 368-374).

Whether the authors achieved their aims, and whether the results support their conclusions:

The authors have made a very convincing case that myopic defocus stimulates the synthesis of IL-6 mRNA and protein in the chick choroid. The effects of atropine and NO-related drugs further support the association between this action and the inhibition of excessive axial elongation and myopia.

Likely impact of the work on the field, and the utility of the methods and data to the community: This work - and in particular the use of assays for IL-6 in choroidal explants to assess the actions of candidate signalling molecules in the choroid - should be seen as an important step forward. The methods are up-to-date, but well established and straightforward, and could be easily duplicated by most workers in the field. The chick models for myopia induction and recovery have been used and refined for decades, so they are easy and almost foolproof (used successfully by many undergraduates in my lab). Once the foundations have been laid by studies in chicks, they can be translated rather easily for making similar studies in mammalian models such as guinea pigs and NHPs

Reviewer #2 (Public Review):

This paper reports on several lines of evidence that suggest that up-regulation of IL-6 expression occurs when axial elongation is slowed in response to myopic defocus experienced by the retina, or in response to treatment with atropine. Since imposed myopic defocus and atropine treatment are widely used to control the progression of myopia, these results may have clinical implications. However, evidence that these changes are causally involved in regulation of eye growth is still lacking. The work is likely to be useful in guiding future experimental analysis of the pathways that may be important in the clinical control of myopia progression.

Reviewer #3 (Public Review):

In this paper, the authors performed differential gene expression profiling in the choroid during recovery from form-deprivation myopia or after treatment with +15 D lenses using Affymetrix microarrays. IL-6 was identified as one of the most differential genes. A set of in vivo and in vitro experiments was also conducted, which suggested that there is an interaction between nitric oxide and IL-6 during recovery from induced myopia and during eye's compensation of imposed positive optical defocus.

Reviewer #1 (Public Review): 

Overall, this manuscript presents a careful study of sea star larval nervous system regeneration using new transgenic tools for marking and following cells involved in regeneration. The authors provide a nice, well-written introduction to their study in the Abstract and Introduction sections. I do have one major issue with the wording they are using for describing what can be done with the transgenic tools they have developed. 

They mention in the third paragraph of the Introduction that "Only cell tracking can definitively establish the origin and trajectory of cells during regeneration and resolve the debate as to the role of stem cells versus cellular reprogramming in echinoderms." And then in the final paragraph they state that "We establish a novel cell lineage tracking system to determine the cellular origin of these regenerated neurons." 

The system they develop does mark individual sox2 and sox4 expressing cells but I object to it being called a "cell lineage tracking system" as this is a very specific term used for a set of methods that allow for tracing the fate of individual cells and all of their progeny, traditionally through development or with stem cells. In essence cell lineage tracking/tracing provides the identification of ALL progeny of a single cell. According to a Primer on Lineage Tracing by Kretzschmar and Watt (2012) https://www.cell.com/fulltext/S0092-8674(12)00003-700003-7)<br> "For any lineage tracer, the key features are that it should not change the properties of the marked cell, its progeny, and its neighbors. The label must be passed on to all progeny of the founder cell, should be retained over time, and should never be transferred to unrelated, neighboring cells." 

I strongly believe that the BAC-reporters developed in this manuscript do not fit that definition of a cell lineage tracing/tracking system and new verbiage should be used to describe these tools. These could very simply be referred to as fluorescent BAC-reporters and describe specifically how they are used to mark and follow the fate of cells expressing the Sox2 and Sox4 genes. The only way the language of a cell lineage tracing/tracking system could be used is if they had created a BAC-reporter for a gene that was expressed constitutively throughout a cell lineage as it progresses or if the protein expression (the tracer) was passed along to all progeny of the cell expressing that gene. My understanding is that the gene expression of Sox2 and Sox4 is highly dynamic and thus the label, by definition, is not going to be passed on to all progeny of the founder cell. I do think this is a powerful system, I just object to how the authors have chosen to describe it in the manuscript. Careful rewording can still make the reader aware of the limitations and advantages of this system and will avoid misunderstanding. 

Therefore, all mentions throughout the manuscript of "a lineage tracing system" would need to be removed and replaced with wording that accurately reflects the true nature of these reporters, simply as photoconvertible expression reporters that can show Sox2 or Sox4 expressing cells. This includes text in the Results and Discussion section, e.g. "To our knowledge, this is the first time that any cell lineage tracing studies have been performed in echinoderm regeneration." 

Results: 

The authors nicely present their larval regeneration system and highlight the timeline of when serotonergic neurons regenerate over a period of 21 days. They then demonstrate that embryonic neurogenesis pathways are recapitulated during larval regeneration. Then, they present results from their photoconvertible expression reporters and demonstrate three populations of cells in decapitated larvae. The green-only sox4+ cells are the most interesting population - these are cells that are induced to express sox4 only after decapitation. Comparing embryogenesis and larval development demonstrated that the wound response in larvae involves specifying new sox4+ cells, something that had ended by 4dpf in normally developing larvae. 

The co-injected double BAC recombinant larvae showed colocalization of sox2+ and sox4+ in regenerating larvae. De novo sox2 expression following bisection together with colocalization with sox4 expression nicely shows that these new sox2+ cells contribute to the neural lineage. Considering that the colocalization appeared to be a rare-ish event (only observed in 4 out of 15 larvae), it would be nice if the authors could comment on why this may be. Is it just a truly rare event to catch or could it have anything to do with the reporters themselves? 

Same question about the sox2+ cells that do not express sox4:Cardinal by 3dpb. Can the authors comment specifically on whether they think there are multiple subpopulations of sox2+ cells and why some get specified to the neural fate while others do not? 

The final experiment using cell division inhibitor Aphidicolin was very clever and nicely demonstrates that cells that did not previously express sox2 can be induced in the absence of cell division. It would be helpful if the authors could indicate how many larvae showed this pattern as they did for the previous colocalization experiment. 

Discussion:

In the final paragraph of the Discussion, the authors discuss a dichotomy between the use of stem cells versus de- or trans-differentiation in different model systems of regeneration. They describe the planarian system in the following way: "For example, the freshwater planarian, Schmidtea mediterranea, utilizes a population of heterogeneous, pluripotent somatic stem cells, called neoblasts, to proliferate and differentiate to replace body parts (Sánchez Alvarado, 2006)" and contrast this with Hydra and axolotl, saying "Conversely species such as Hydra and axolotl, refate differentiated cells either through dedifferentiation or transdifferentiation (Gerber et al., 2018)." I think this oversimplifies the current understanding of these systems. For example, a recent paper by Raz et al. 2021 (Cell Stem Cell 28(7): 1307-1322.e5) makes the case that Schmidtea mediterranea is capable of having specialized neoblasts undergo fate-switching and "propose a non-hierarchical lineage model for neoblasts, in which a neoblast can specify one of a diverse set of possible fates in the course of a single division and specialized neoblasts can divide to generate neoblasts that can specify different fates." In essence, this could be considered something more flexible and complicated than what the authors described - just using pluripotent neoblasts to proliferate and differentiate to replace body parts. And although Hydra is known to use trans-differentiation during regeneration, this organism also employs stem cells in the process of regeneration. Please see Siebert et al. 2008 (Developmental Biology 313(1): 13-24) for a discussion of how both mechanisms are employed in this regeneration model. Therefore, I think it is an oversimplification to characterize these regeneration models as either using stem cells OR using de- or trans-differentiation. I think in these systems, there is not a simple dichotomy and more flexibility has been demonstrated in how regeneration is accomplished than the authors describe here and the text would need to be revised accordingly.

Reviewer #2 (Public Review): 

Regeneration is a developmental process that occurs in response to injury. Important questions about regeneration include: 1) to what extent are regeneration gene regulatory networks similar to embryonic gene regulatory networks, 2) what is the source of cells used to build new structures during regeneration, and 3) what aspects of regeneration are deeply conserved and which are taxon specific. Towards addressing these questions, the authors have established the sea star larva as a regeneration model. As invertebrate deuterostomes, echinoderms (such as the sea star) have an informative position on the phylogenetic tree; leveraging these organisms to study regeneration will both help answer outstanding questions in the regenerative biology field and will reveal new insights into the evolutionary history of regeneration in vertebrates. In this study, the authors establish a new method for lineage tracing in the sea star larva to determine the source of cells that participate in regenerating the larval serotonergic nervous system after bisection and regeneration of the anterior end. Previous work by this group has established critical transcription factors in the specification of serotonergic neurons during embryogenesis. Here, the authors demonstrate that during regeneration, these same transcription factors are expressed within one day after bisection in cells that previously did not express these genes. Co-expression analysis revealed that embryonic neurogenesis expression states are repeated during regeneration and that injury induces the formation of proliferative sox4+ neural progenitors. To test the hypothesis that sox4+ progenitors newly arise during regeneration, the authors used BAC recombineering to express the photoconvertible protein Kaede under the sox4 regulatory sequences. The fluorescent protein is stable for at least 7 days, making it possible to use this system to determine the history of the sox4-expressing cells present during regeneration. By converting all existing Kaede to red before bisection, the authors were able to demonstrate that sox4+ neural progenitors are newly specified during regeneration. Importantly, they found that this occurs during a time in normal development when sox4+ progenitor cell specification is complete. The authors next use their BAC Kaede lineage tracing system to determine that the sox4+ cells arise from sox2+ progenitors, similar to what is observed during embryogenesis. Finally, they found that new sox2 expression can be induced in the absence of cell division which strongly suggests that the sox2+ progenitors are derived from the respecification of existing cells, rather than from resident stem cells. Overall, this is an interesting and well-executed study that clearly demonstrates that larval nervous system regeneration involves respecification of cells into the neural lineage and that at least a portion of the embryonic GRN is used to regenerate the larval nervous system. Furthermore, this work firmly establishes sea start larval regeneration as an important model for answering outstanding questions in regenerative biology.

Reviewer #1 (Public Review): 

The relationship between homologous chromosomes sampled in a population can be described by an "ancestral recombination graph" or as a "forest" of correlated coalescent trees describing the relationship at each locus on the chromosome. It has long been clear that this graph contains enormous amounts of information about the history of the population, and should be used in analysis. Hitherto this has been computationally infeasible, but recently developed methods are starting to make it possible, and this paper is one of the first attempts to do so. 

The approach is cutting edge, and appears to work in theory. However, some of the inferences from real data, in this case the 1001 Arabidopsis Genomes seem implausible, in particular the conclusion of rapid recent migration - hundreds of kilometers in tens of generations. At the very least, these estimates need to be put in context of It data on local populations being stable for over 100 years, of outcrossing only happening on the order of every 20 generations, and a general pattern of strong population structure. An alternative explanation that the algorithm is producing biased results needs to be excluded. 

If this can be resolved, the paper should be of major interest to anyone interested in population genetics.

Reviewer #2 (Public Review): 

Strengths: 

The method leverages inferred genealogies to make inferences about location and dispersal rates of ancestors. If the inferred genealogies are sufficiently accurate, this approach should be nearly unrivalled in its power to achieve this aim. Furthermore, the ability to locate ancestors, and trace migrations could be of great importance and has the potential to capture histories more accurately, compared to more simplistic approaches that assume discrete events and no explicit spatial models. 

Weaknesses: 

A potential weakness is that the data contains most likely little information about location of ancestors in deeper times and it is unclear how to identify when estimates become unreliable. Relatedly, there is a potential challenge in interpreting inferred ancestral locations - in particular, when shifts (or lack of them) could be caused by sampling biases or underrepresentation of certain ancestries. Both points are sufficiently caveated in the paper, but could provide challenges when applying this approach in practise. 

Impact and utility: 

The method was applied to A. thaliana, but should readily be applicable to any recombining species and in particular to human data. There, we have extensive data of ancient human groups, which may benefit this approach and could reveal important insights into ancestral migrations of human groups. Quantifying evidence of migrations beyond qualitative measures (e.g., of gene-flow) is important and has the potential to capture more subtle signals, including separation by distance or continued movements through time. 

Overall, I believe that the authors are presenting a good method, which will be difficult to improve upon and can be applied to a wide range of problems. I am convinced about most of their results and conclusions. The overestimation of dispersal rates in simulations is interesting and could be investigated further. The inferred increase in recent dispersal rates of A. thaliana could potentially be (in part) an artefact of excluding rare variants in the data and should be checked by the authors.

Reviewer #3 (Public Review): 

This paper presents a method to estimate the spatial locations of ancestors based on inferred genetic ancestry from recombining species. Given an estimated ancestry (in the form of a sequence of time-resolved trees along the genome), the method can estimate dispersal rates through time as well as the locations of genetic ancestors, based on a Branching Brownian motion model of spatial dispersal. The inference method performs well on data produced by detailed forwards-time simulations, both from the true simulated trees as well as trees inferred by Relate. 

The authors apply the inference method to a large Arabidopsis thaliana data set, first estimating trees using Relate, and then applying their inference methods to the trees. (The estimated trees have been made freely available on Zenodo, which will be a valuable community resource.) They detect very high dispersal rates in the recent past (especially East-West), and show many interesting visualisations of population structure over time. 

This paper is important not just for the method it introduces and the inferences made, but because it showcases what is possible given the newly available estimates of genetic ancestry. Population genetics is today mostly concerned with extant individuals, and the effects of historical processes on their genomes. Now that we have estimated trees we can begin to ask questions directly about genetic ancestors as well as samples. This paper helps to answer one fundamental question (where did the ancestors live?), but there are many more, and the methodology developed here will help shape those questions.

Joint Public Review:

Broad interest:

• This study will be of interest to the large group of neuroscientists interested in delineating the neural and behavioral consequences of heterogeneous neurological diseases

• This manuscript will be of interest to researchers studying complex brain-behavior links

The study tackles the crucial problem of phenotypic complexity head-on:

• This is an important work exploring the complex links between the brain's white matter structure and the symptomatology of concussions in children and adolescents

• Concussion and related brain diseases have great heterogeneity at multiple levels, including variable symptom profiles, different progression patterns, and distinct neurobiological mechanisms. This work acknowledges heterogeneity and the proposed methodology aims to leverage it, rather than averaging it out

Rigor:

• The methods are rigorous and original

• The methods used for analysis of dMRI are the state of the art in the field: excellent modeling techniques that overcome some of the known challenges of dMRI, and judicious use of PCA to summarize the multiple dMRI metrics that were derived into concise and interpretable components

• Data processing and analysis are of very high quality

• The main strength of the work is the expertise of the authors in advanced neuroimaging and statistical analysis methodologies. The brain's white matter structure and its connectivity pattern are quantified using a reliable diffusion imaging and procession pipeline. Multiple diffusion metrics are computed and then summarized using PCA. PLSc is used to discover multi-track multi-symptom relationships. Permutation testing and bootstrapping are used to identify significant links and significant weights. It is always a pleasure to read a manuscript that includes a "reliable" analysis pipeline

• Regarding the statistical methodology -- here as well, the methods are sophisticated and seem to be well-executed. The use of a discovery sample and application to a separate replication sample is an excellent approach.

• Another strength is the inclusion of a replication sample

The approach has broad utility:

• As the methods are agnostic to the nature of concussion, the proposed methodology may benefit studies dealing with other brain diseases and disorders as well

• The methodology introduced here is potentially quite valuable and its application to this dataset and to other datasets where there is substantial heterogeneity could improve the inferences made from measurements of large samples of individuals with these disorders.

Reviewer #3 (Public Review): 

Meiotic drivers act when the driver allele is in a heterozygous state, which is most likely when outcrossing is frequent. To explain the existence of drivers in the highly selfing fission yeast, the authors investigate the effect of selfing on spread of a driver allele. They show that outcrossing in fission yeast varies between different natural isolates, and show in a competition experiment that a meiotic driver increases in the population more rapidly under outcrossing than under selfing conditions. Additionally, they show that fitness costs associated with a driver will reduce the speed of invasion and the initial frequency required for invasion to be possible. 

Strengths:

The research shows experimentally the change of allele frequency for drivers and how this spread differs between heterothallic (obligatory outcrossing) and homothallic (haploid selfing capable) strains. The experiments are further supported by a simple model in which the invasion trajectories are well predicted. The addition of competitions where a driver is linked to a deleterious allele (GFP) shows the importance of the cost of drivers for their evolutionary dynamics which again fits with the predictions of the model. 

The authors further show that the amount of haploid selfing, sporulation efficiency and probably mating type switching varies between natural isolates. These observations suggest that mating in nature is probably more messy than would be expected based on Leupold derived strains. They describe interesting observations in natural strains, such as shmoo-length variation, uncleaved (filamentous) cells that mate and potential mating interference. 

These experiments are very promising and show that meiotic drive in the fission yeast system can be used as a tool to study meiotic drivers, being able to manipulate driver linked cost (and probably also benefits when linking the driver to an prototrophic marker), the ability to self- or outcross and possibly associate these markers to mating type alleles simulating "sex" specific drive. 

Weaknesses:

Even though the experiments find some important parameters for meiotic-driver spread in fission yeast, the results are not sufficient to explain the apparent "success of meiotic drivers in this species". The links that the authors suggest between mating type switching efficiency, the amount of outcrossing, the speed of invasion of the driver and the cost associated with the driver cannot explain the success of drivers. Furthermore, the causality of the different factors is not explained. 

That outcrossing increases the speed of invasion is true (see also Durand et al 1997 PMID:9093861), but the argument that 'reduced levels of mating type switching could lead to less inbreeding' is not supported. There are two problems with this statement. First, it is not clear to me if this is theoretically true. If switching occurs infrequently but consistently, the chances of a cell to be positioned to another cell of the opposite mating type either from self or opposite type will probably not be that different. Only in a narrow range of cell density will this probably play a role, however, this should be properly modelled in a structural environment or tested experimentally. A comparison between heterothallic and homothallic strains is - contrary to what the authors argue in line 138 - not appropriate for this test, as the first cannot reproduce by selfing. Using strains that have intermediate amounts of mating type switching (e.g. using h90 Sp strains mutant in the switching pathway; Maki et al. PMID:29852001) could give more insight in this. Reduced switching will lead to reduced spore production, because fewer of the cells will be located next to a cell of the opposite mating type (as shown in Nieuwenhuis et al. 2018 PMID:29691402 and by the authors in Fig. 1-S3), but this does not have to affect outcrossing efficiency. This also becomes apparent from the data presented in Figs 1D and 1E, which do not seem correlated. Second, the authors have not measured mating-type switching, but used the amount of mother daughter matings as a proxy for mating type switching. This method introduces a bias towards the correlation switching and selfing, because the latter is used as a proxy for the first. Fluorescent proteins under control of a mating-type specific promotor is an established method (e.g. Jakočiūnas et al 2013 PMCID:PMC6420890, Vještica et al 2021 PMID:33406066), which will give direct observations of the mating type. The observation that the shmoo length is associated with outcrossing is very interesting, and - without changing switching frequency - appears to affect outcrossing. 

Finally, the authors argue that meiotic drivers are evolving rapidly, can invade fast and that this can occur even when selfing is prevalent. The model seems to contract this. Let's start with the claim that novel drivers can invade in a population. Novel alleles arise at a frequency of 1/N (N = population size, bottom left corner Fig 3A, not at 5% as used in the analyses) and as drive is as strong as the inverse of the population size the fitness difference is initially extremely low giving plenty of time for drift (when driver is neutral) or selection (when driver is deleterious) to remove the novel allele. In order for drivers to increase to levels that will give 'rapid wtf gene evolution' (line 112) a prolonged level of mostly drift is probably necessary. 

It is difficult to make statements about the speed of wtf evolution in the fission yeast system, without having a better description of the variation of the paralogs and their ages in fission yeast. The speed of wtf evolution is not clear, as shown in earlier findings from this group that shows very old wtf loci; Eickbush et al. 2019. Comparing wtf evolution relative to neutrally evolving loci might give more insight in wtf evolution speed. Especially when drive is costly (as suggested by the authors, though not shown or quantified) the time to substantial frequencies is large. It could also be possible that drive itself is beneficial (e.g. resources from the killed spores made available to the killers or through released local competitive pressure), which will lead to increased fitness though combined drive and increased viability, even at low frequencies. 

# Minor comments 

The loss of mCherry alleles due to reversion of ura+ occurs more rapidly than that in GFP. It is likely that this variable change in reversion affects the observed change in frequency. This should be corrected for in the raw data. 

Inbreeding is a term generally used in population genetics, where it refers the the amount of mating between related individuals. Even though it is fundamentally correct, a more appropriate term would be haploid selfing or intra-clonal mating, as mating in these strains and experiments is actually between clones. Inbreeding in this context is confusing to people who are not familiar with facultatively sexual species. 

The effect of inbreeding on driver alleles has been studied theoretically before, showing qualitatively similar results (e.g. see Durand et al 1997 PMID:9093861; Martinossi-Allibert et al. 2021 PMID:33764512, Ament-Velásquez). 

Reference to driver systems in other fungal species (Neurospora and Podospora) that are highly selfing is completely missing (Svedberg et al. 2018, 2021; Vogan et al 2019, 2020; Martinossi-Allibert et al. 2021) 

There seems to be quite some variation between the different replicate experiments (Fig 1E vs Fig 2-S3 for example). 

Line 76: This paragraph is a bit misleading and internally contradicting. The data from Farlow et al. does not take into consideration the recent hybridisation of diverged populations as shown in Tusso et al. 2018 and thus overestimates the time between outcrossing events. The estimate that 20-60 outcrossing events (underestimate due to homogenization and potential meitotic drive) occurred in the last 500 years suggests a higher number than 1 per 800,000. Citing this number is obsolete. 

line 730: The inbreeding coefficient in Sun et al. 2017 (probability of IBD which is between 0 and 1) is different from the one used in Hartl & Clark 2007 between -1 and 1. 

The speculations on the 4913bp insertion and its effect on mating type switching is not substantiated. Variation around the mating type is rampant (see for example Beach 1986 and Nieuwenhuis et al. 2018) and the authors even show that is likely is not the case that this element affects switching in FY29033. The insertion is an interesting observation, but just that.

Reviewer #1 (Public Review): 

This manuscript reports theoretical and experimental analyses of a meiotic drive element in the yeast Schizosaccharomyces pombe, to understand whether the outcrossing rate is high enough in this species, long thought to undergo mostly same-clone mating, to explain the spread of multiple meiotic drive elements. The topic is of general interest, the experiments and analyses are clever and sound, and provide interesting answers. The experiments indeed show that the outcrossing rate in the laboratory varies among natural isolates and density conditions, and can be substantial ; the theoretical model shows that the estimated outcrossing rates do allow meiotic drive to spread. However, the outcrossing rates measured in the laboratory may be really different from those in nature and population genomic data are available that could allow estimate actual outcrossing rates in natural populations. Indeed, outcrossing rates depend mostly on where and when in nature dispersal and clonal multiplication occur, while laboratory experiments typically use high densities of clonemates on plates. Overall this study brings support to the idea that homothallic fungi probably do not undergo mostly same-clone mating in nature, in contrast to the most accepted view in the fungal literature, but in agreement with evolutionary considerations ; the study and the findings would thus benefit from being placed in the right evolutionary context (doi: 10.1111/j.1420-9101.2012.02495.x ; 10.1111/j.1469-185X.2010.00153.x ; 10.1128/EC.00440-07 ; 10.1038/hdy.2014.37 ; 10.1111/nph.17039). The terms selfing and outcrossing as used in the manuscript does not correspond to the diploid selfing and outcrossing that occur in plants and animals, and the term can thus be misleading.

Reviewer #2 (Public Review): 

The authors combine cytological, genetic, mathematical, and experimental evolution techniques to connect variation in mating behavior with variation in the population dynamics of meiotic drive in the yeast Schizosaccharomyces pombe. First, the authors use cytological and genetic methods to document variation across strains of pombe in their (i) propensity to inbreed, (ii) efficiency of mating, (iii) rate of mate type switching, and (iv) variability of ascus morphology. These results will be of major standalone interest to the yeast community, and will likely find experimental use in many settings. Then the authors use population genetic modeling to study the theoretical implications of this variation in mating behavior for the spread of meiotic drivers (which have recently been shown to be pervasive in pombe). Finally, the authors use cytological techniques to track the spread of introduced drivers in experimental populations of pombe, and show that the drivers follow frequency trajectories that agree well with predictions from the theoretical analysis. These results will be of major interest to geneticists working on meiotic drive, as well as workers in the currently burgeoning field of synthetic gene drives for population control. 

The analysis is carefully done, and I am confident in the results as presented (with one minor exception detailed below). My only major suggestion for improvement concerns the scope of the population genetic modeling. As it stands, this modeling is primarily used to generate predicted frequency trajectories of meiotic drivers against which the trajectories observed in the evolution experiments can be compared. The fact that the experimental and theoretical trajectories match well is impressive, and very promising for the future of pombe as an experimental system in meiotic drive research. However, substantively, as the authors recognize, this agreement tells us mainly that the population genetic model that they use to generate the predicted trajectories takes into account all relevant parameters and is well calibrated. Thus, from the population genetic modeling and evolution experiments, we get only an indirect picture of how variation in mating behavior has actually impacted the natural spread of drivers in this species. I believe that the population genetic modeling, with minor modifications, could in fact be used to make more direct predictions about the natural history of drive in pombe. For example, should strains with less inbreeding harbor more fixed drivers? And in strains with more inbreeding, should drivers---because they have very long fixation times---be more likely to be observed as polymorphisms? Such questions are, I believe, well within reach of the authors' population genetic modeling. 

A minor concern: To track the spread of a driver introduced in their experimental populations, the authors linked the driving allele to one fluorescent marker (GFP/mCherry) and the non-driving allele to the other (mCherry/GFP), and compared the spread of the one marker relative to the other. To use their model to generate expected frequency trajectories for these experiments, the authors needed to measure, in controlled settings, the intrinsic fitness costs of GFP vs mCherry; they estimate that GFP is relatively costly in sexual (but not vegetative) reproduction. However, their estimates of the relative fitness cost of GFP are based on frequency trajectories across just 6 generations, and assume additive dominance, so that the fitness cost to a GFP homozygote is twice that to a heterozygote. It is unclear how statistically noisy the estimation procedure is given the small number of generations used, and whether it is justified to assume additive dominance (which is especially relevant since the dominance of fitness costs is known to be a critical factor in determining the frequency dynamics of meiotic drive).

Reviewer #1 (Public Review): 

The authors present a prototype of a nanofabricated microfluidic device that allows plunge-freezing of sample solutions for high-resolution cryo-EM analysis, specifically single-particle cryo-EM. The design of the device includes cavities (nanochannels) of defined thickness, made of ~10nm thick silicon nitride membranes (SiN). The authors demonstrate the loading of a device consisting of 22 connected cavities with three different samples - apoferritin, 20S proteasome and tobacco mosaic virus - requiring only a few picoliters of sample. Subsequent data collection from the frozen samples produced reconstructions with resolutions between 3 and 5.4 Å. 

The new device solves one of the most difficult problems in cryo-EM sample preparation: controlling the sample thickness. This is a major advance as it is a prerequisite for full reproducibility and automation of single-particle cryo-EM, which is an important goal of the field. The device also addresses sample denaturation at the air-water interface, another problem currently limiting single-particle cryo-EM. However, how replacement of the air-water interface with a SiN-water interface affects denaturation and potential preferred orientation of particles remains to be shown. The SiN membranes also add background to the cryo-EM images that may interfere with the high-resolution signal of the particles. Indeed, the additional background may have limited the resolution obtained from the three datasets. The reconstructions also displayed substantially higher B-factors than previously obtained from similar samples. It is possible that this is related to the enclosure of the sample in a cavity: The cavity prevents the escape of gases from the sample that evolve as a result of radiolysis under the electron beam, potentially leading to increased beam-induced motion of the sample. Nevertheless, this work serves as a convincing proof of principle for a promising new sample preparation technique.

Reviewer #2 (Public Review): 

This is an excellent paper that describes a conceptual advance in specimen preparation methods for cryo-EM. While technically demanding wrt production, these nanofluidic devices may simplify reproducible sample prepapartion from a user perspective. The authors' cryoChip design represents an alternative to conventional well-established methods and adds to a growing number of novel specimen preparation devices. The paper is well written, concise, and comprehensively covers all relevant validations and control experiments. I enjoyed reading it. The discussion was refreshingly objective and critical about the pros and cons of this novel approach. I agree with the authors that the present MEMS devices and their future derivatives are only the beginning of exciting novel applications such as on-chip mixing, high throughput screening, time resolution and other lab-on-chip experiments.

Reviewer #3 (Public Review): 

This new nanofluidic sample cell for cryo-EM is a revolutionary idea and implementation. It addresses one of the most important and most painful bottlenecks in the cryo-EM workflow. The unreliable and wasteful conventional grid preparation method is the most impactful remaining bottleneck of today's cryo-EM. A better alternative method is desperately needed. <br> This manuscript describes such a promising alternative method. However, the described method has its own problems and difficulties. All of these are excellently presented and discussed in this outstandingly well written manuscript. But the manuscript fails to convince this reviewer that the new method is (already) better than the alternative methods. 

As the authors correctly report, quantify and discuss, the new method faces several hurdles: 

• The SiNx membranes on top and bottom of the frozen sample provide additional noise background to the images. The SiNx surfaces lead to massive adsorption of the samples, leading to overcrowding, but at least the sample doesn't seem to denature as much as when adsorbing to an air-water interface. However, the sample will not be in bulk water and for other samples that are not as strongly symmetric as these, preferred orientation is still very likely. CTF estimation on recorded images is difficult due to the signal from the SiNx layers.

• The vitrification of the SiNx grids is difficult and often leads to crystalline ice in the images. But as the authors state, vitrification methods might require adaptation to this new grid type.

• The signal to noise ratio of the recorded images is very low. The calculated ResLog B-factor of 217 Å2 for apoferritin from a Titan Krios with Gif-K2 is actually *not* in range expected for such structures, but rather terrible. The JEOL data with ResLog B-factors of 215 and 490 Å2 are not better. These should be compared to the ResLog B-factors from the same instruments for the same samples, when conventional grids are loaded. But in any case, the authors correctly discuss this issue and correctly estimate that the lower SNR from the SiNx will restrict the method to particles larger than 200kDa.

• The utilized pixel size for these datasets was surprisingly large. If 2.99Å target resolution is achieved, then I'd have expected smaller pixels in the images, such as 0.6Å/px, not 0.8127Å/px.

The most impressive feature about this new cryoChip for me is the tiny amount of sample required to prepare grids. Blotting requires microliters of sample. CryoWriting (your reference 16) requires nanoliters of sample. And your new cryoChip now seems to work with picoliters, which is another reduction of 1000 fold. That is an extremely impressive feature that may enable unique applications in special settings, as for example when imaging (in the future) the entire non-diluted cytosol of a single cell. Also, the possible extensions to prepare multiple different samples on the same nanofluidic chip are very interesting. 

In summary, this is an outstandingly well written manuscript of a revolutionary method that is still in its infancy. At its present state, this technology may be useful only in very few exceptional cases, but fails to convince (yet) as general method. However, further development of this young method holds great promises to still strongly improve this novel technology.

Reviewer #1 (Public Review): 

In this study, the authors use CyTOF-based analysis to characterise spike-specific T cell responses following mRNA vaccination. They seek to understand both the breadth of responses to 'wildtype'-like and variant spikes, as well as the differences between T cell responses from convalescent and previously uninfected subjects. Consistent with other studies, they find that spike-specific T cell responses are similar across different variants, both in frequency and phenotype. In contrast, however, they identify several phenotypic differences in the T cell response elicited by infection, vaccination, or vaccination following infection. 

Despite a somewhat limited sample size, they clearly identify changes in memory phenotype and chemokine receptor expression that may affect T cell trafficking to mucosal tissues across infection and vaccination. While inclusion of additional chemokine receptors (such as CXCR3) in the CyTOF panel would have aided in characterising these cells, this data highlights how infection and vaccination may elicit distinct T cell responses. Future studies will be required to better assess the functional impacts of these phenotypic differences on T cell recall and contribution to protective immunity.

Reviewer #2 (Public Review): 

The authors address an important question, whether it people who have had Covid19 and are then vaccinated with one mRNA Spike vaccines made better immune responses than those who had not previously been infected and have two shots of the vaccine. They also compare responses to different virus variants and find extensive cross reactions and no differences between the groups - an important result.Their main finding is a difference in the quality of the CD4+ T cells in the 'Covid-vaccinees' compared to the 'naive double vaccines'. They suggest that T cells in the former may home better to the respiratory tract and persist longer. 

The major strengths are: 

• The methodology used, based on Cytof multiparameter analysis of antigen responding CD4 and CD8 T cells.

• Demonstration that the second vaccine dose in the naive group 'improves' the T cell response.

• Demonstration that a second vaccination in the Covid19 group does not improve the T cells.

Weaknesses:

Fully (and commendably) acknowledged in the manuscript: 

• The study groups are small

• The antigen specific T cells are stimulated in vitro so may be distorted, nevertheless there were still differences

Not acknowledged but possibly outside the scope of this study:

• The reader will wonder how this affects the antibody response which ultimately is the main protector from reinfection and also how the T cell responses might impact on disease severity after post vaccination (re)-inrfection

Reviewer #1 (Public Review): 

The paper by Saraswathy et al follows up on an astute observation in a previous study from the same group, which showed that mib1 morphants are associated with an early phenotype reflective of impaired convergence extension. This study allows them to link this observation to a role for Mib1 in internalization of Ryk, which previous studies have identified as a target of Mib1 dependent ubiquitination and for its role in Planar Cell Polarity. The study suggests that RING domains required for ubiquitination of Mib1 target genes are required for this function as ectopic expression of mutant forms of Mib1 that lack either all three RING domains or just the last RING domain are capable of exaggerating CE deficits in morphants or specific mib1 mutants, which retain some maternal mib1 function. 

In this context, it remains unclear why the ta52 allele, which has a point mutation that interferes with mib1 function in the context of its other interactions, fails to have this CE phenotype or why ectopic expression of this mutant form of mib1 does not seem to cause similar exaggeration of the early CE phenotype. 

The authors go on to suggest the CE phenotype is not related to a role for mib1 in Notch signaling as ectopic expression of an activated form of Notch (ICD) does reduce the CE phenotype. On the other hand, ectopic expression of RhoA, which functions downstream of Ryk in PCP, is able to reduce the CE phenotype. 

The authors demonstrate that impaired mib1 function is associated with increased accumulation of Ryk-GFP on the cell surface and reduced intracellular Ryk-GFP, which the authors describe are part of an endocytic compartment based on its colocalization with Rab5. To determine if the CE phenotype produced by reduced Mib1-dependent endocytosis, arises as a result of too much Ryk on the cell surface or of as a result of reduction in the endosomal compartment, they determine the effects of ectopically expressing Ryk-GFP. Ectopic Ryk-GFP reduces CE associated defects and results in both increased surface expression and an increase in the amount of internalized Ryk-GFP. This leads the authors to conclude that it is not the increase of Ryk at the cell but rather, the reduction of internalized Ryk that is responsible for CE defect in embryos with impaired Mib1 function. 

The authors go on to demonstrate that effects of Mib1 on internalization of Ryk are specific and not seen with other PCP components like Vangl2, Fz2 or Fz7. <br> Finally, as Mib1 function is hypothesized to be relevant only in the context of Ryk endocytosis for its CE defects, the authors show that loss of Mib1 function does not increase the severityof these defects in Ryk null mutants. 

The paper describes a straightforward set of logical experiments to explore the link between Mib1, Ryk and its role in CE. The primary conclusions of the paper are supported by data presented. Nevertheless, many questions remain unanswered. It remains unclear to me why the ta52b allele has such a minimal effect, the authors could elaborate on this. It also remains unclear how Ryk endocytosis contributes to effective PCP or how reduced Ryk endocytosis alters the organization or function of other PCP components.

Reviewer #2 (Public Review): 

In this study, the authors identify a role for mindbomb (mib) in the trafficking of the Ryk receptor, a Receptor Like Tyrosine Kinase with roles in planar cell polarity (PCP). The authors use a combination of tools and mutants in the zebrafish to propose that the role in gastrulation is notch-independent. A strength is the generation of new genetic alleles. The authors show a genetic interaction between Mib and Ryk. An intriguing finding from the double mutant (mib;MZryk), is that the impact of Mib is mediated through Ryk. This study highlights the need to consider complex signaling networks when evaluating phenotypes, especially convergent extension. 

A concern is that most of the analyses are with morpholino knockdown. It is understood that the use of the MO allows for higher sample sizes, but the authors would need to also investigate the functional rescue in the genetic mutants. Moreover, the CE defects shown in Figures 2-4 were analyzed by morphological characteristics. The conclusions can be strengthened by measurement of axial structures with molecular markers. 

Ryk has been proposed to be a substrate for Mib and in the manuscript, there is a clear demonstration that Mib overexpression is sufficient to stimulate Ryk-GFP internalization. It is noteworthy that Mib over-expression does not stimulate Vangl internalization. However, the difference of the number of Ryk-GFP endosomes/cell in the mib mutant compared to would-type is less clear. Also, the MO alone is not shown. Using an early endosomal marker with the Ryk-GFP would be a more effective way to evaluate endocytosis. 

In the manuscript, the authors generate a zebrafish ryk mutant. In reference to the functional analysis of the ryknce4g product, the authors would need to perform a Western blot of the injected embryos to determine how efficiently the RNA is made into protein and how stable the protein product is, compared to levels of the expressed wt RNA. This will be important when making conclusions about a lack of rescue and lack of over-expression defects.

Reviewer #3 (Public Review): 

The manuscript by Muraleedharan Saraswathy et al. describes a novel role of the E3 ubiquitin ligase Mindbomb1 (Mib1), a known key regulator of Notch signaling, in regulating convergent extension (CE) movements of the zebrafish gastrula, which is dependent on planar cell polarity (PCP) signaling. The authors demonstrate that mib1 null mutant embryos exhibit CE defects and modulate PCP, independently of Notch signaling. The authors also show that the ability of Mib1 to modulate PCP is totally dependent on the receptor tyrosine kinase Ryk via endocytosis. 

The strength of the manuscript is that Mib1 regulates PCP-dependent CE in a Notch-independent manner. Also, the genetic epistasis analysis demonstrates that Mib1 loss of function has no effect on CE in maternal zygotic ryk null mutants, suggesting that Mib1 is an essential regulator of Ryk endocytosis during zebrafish CE. These demonstrations are convincing. However, the weakness is the conclusion that Mib1 is an 'essential' regulator of PCP is not fully supported by the data. Rather, Mib1 could be a context-dependent modulator of PCP. To fulfill the scope of eLife requires the clarification of the weakness. It would be worth considering publication if this is clarified.

Reviewer #1 (Public Review): 

Sanchez et al investigate how the complex morphogenetic movements that drive epithelial tube formation are patterned to occur with the correct spatiotemporal dynamics by upstream transcription factor expression, a fundamental question in developmental biology. In prior work, the authors defined the cell behaviors that drive tube formation in the Drosophila salivary gland, demonstrating that localized apical constriction induces epithelial bending to form a pit and circumferential cell intercalations narrow the primordium. In this study, they show that as more peripheral cells flow into the deepening pit they switch behaviors to constrict apically, promoting continued morphogenesis of the structure. These behaviors, as well as correlated patterns of myosin pathway activation, require transcription factors Fkh and Hkb suggesting the expression pattern of these TFs may drive the dynamic changes in cell behavior. Using endogenously tagged protein reporters of Fkh and Hkb, they show the two TFs display dynamic expression patterns that initiate roughly where apical constriction will predominate and spread outward to cells that will later constrict into the pit. Fog appears to be a key downstream target of Fkh and Hkb, and interfering with the radial pattern Fog expression disrupts tube formation. Strengths of the study include the high quality, quantitative morphometric analysis in both time and space, the use of endogenously tagged reporters of Fkh and Hkb together with time-lapse analysis, and the multiscale nature of the study encompassing and connecting upstream patterning events to intermediate regulators of cell shape to downstream cell and tissue-level behaviors. A minor weakness of the study in its current form is a lack of cell tracking data that connect cell identity with the stated changes in behavior, something that should be straightforward to address. A limitation of the study is whether the timing of events described here are consistent their model - the dynamics of TF expression patterns precede the corresponding Fog/myosin patterns and morphogenetic changes by ~30-60 minutes. An analysis of Fog transcriptional dynamics could fill this gap.

Reviewer #2 (Public Review): 

The authors analyze the cellular dynamics responsible for the formation of a tubular structure, taking as a model the formation of salivary gland in Drosophila embryo, which is initiated by the asymmetric invagination of cells from a circular placode. They observed a regionalized cellular behavior, dependent on Hkb and Fkh dynamic expression in the placode. Both these transcription factors are required to ensure the correct expression of Fog, which drives localized apical constriction and ensures correct morphogenesis of the salivary gland. 

Strengths: 

This is a detailed analysis of cellular dynamics during salivary gland morphogenesis. This study highlights the regionalized behavior of cells from the presumptive gland (or placode) with a region close to the invagination pit where Hbk and Fkh drive Fog expression, leading to medio-apical myosin accumulation and apical constriction and a more distant region where cells mostly intercalate, a process driven by a junctional Myosin polarity. Although mainly descriptive, these data are precise and convincing. The conclusions fit with the observations. 

Weaknesses: 

Although this work is interesting, it raises a lot of unanswered questions. How is the timing of apical constriction in the placode controlled? What is responsible for the delay in apical constriction observed in Hkb mutant? How does apical constriction propagate? How is the switch between apical contraction and intercalating domains regulated?

Reviewer #3 (Public Review): 

In the manuscript by Sanchez-Corrales, Blanchard, and Röper, the authors examine how the Drosophila salivary glands form from a primordium where the specifying transcription factors are expressed asymmetrically. Previous recent studies have shown the roles of apical constriction and cell intercalation to salivary gland formation (Sanchez-Corrales et al., 2018; Chung et al., 2017). In this study, the authors characterize the pattern of these cell behaviors across the primordium and correlate these with the expression of transcription factors, Hkb and Fkh. The authors found that cells in the dorsal-posterior of the primordium apically constrict and invaginate and that the position of this behavior is controlled by Hkb. The authors conclude that the expression of a GPCR ligand, Fog, is patterned by the transcription factors Hkb and Fkh, which leads to primordium regionalization. 

The finding of this transcription factor patterning and switches in cell behaviors at fixed positions in the primordium is likely to be more generally important for the development of complex tubular organs. The authors' conclusions are mostly well supported by the data, which is rigorously quantified. One limitation is that the conclusion that the pattern of Hkb expression regulates this cell behavior is based on analyzing a loss-of-function mutant in hkb, rather than altering its expression pattern.

Reviewer #1 (Public Review): 

The central process of meiotic cell division is the repair of induced double-strand breaks by meiotic recombination, which mediates the pairing of homologous chromosomes and their exchange of genetic material by crossing-over. In this study, Rousova et al aimed to determine the molecular basis of the essential function of Mer2 in regulating the double-strand break response in yeast meiosis. In this, they studied Mer2's interactions with Spp1, Hop1 and Mre11, in order to determine how Mer2 brings together nucleosomal DNA, the meiotic chromosome axis and the double-strand break resection machinery. Through a robust biochemical approach, they determined that Mer2 has a core tetrameric structure that recruits two Spp1 molecules in a high-affinity 4:2 complex. This complex interacts with nucleosomes in a manner that depends on the induced dimerisation of Spp1 and is strengthened by an unanticipated direct interaction between Mer2 and nucleosomes. They further show that Mer2's core/C-terminal end binds directly to Hop1 in a manner that is inhibited by the closure interaction of Hop1's HORMA domain. Finally, they identify point mutations of conserved amino-acids within Mer2's N-terminus that block sporulation and impair double-strand break formation in vivo, thereby indicating their importance in this aspect of its function. These mutations weakened the Mer2-Mre11 interaction by yeast two-hybrid, and to a lesser extent through pull-down of recombinant proteins, suggesting that a weakening of the interaction between Mer2's N-terminus may at least in part explain the in vivo phenotype. Together, this paper provides important molecular insight into Mer2's interactions with three distinct components of double-strand break sites. These findings will provide essential foundations for ultimately uncovering the full three-dimensional structure of Mer2's ternary assembly at double-strand breaks and the precise molecular mechanism whereby it regulates the double-strand break machinery.

Reviewer #2 (Public Review): 

The study of eukaryotic meiosis has for many years been hampered by the inability to purify and reconstitute important meiotic DNA-binding and DNA cleavage/recombination factors, limiting the field's ability to perform detailed interaction analysis and structure-function dissection of the large set of proteins known to be important for meiotic recombination. Recent years have seen major advances in biochemical reconstitution of meiotic chromosome-associated proteins, and this manuscript is a prime example of what can be learned through this kind of analysis. The authors demonstrate direct interaction between Mer2 and: (1) Spp1 (a previously known interaction); (2) nucleosomes; (3) the axis protein Hop1; and (4) Mre11, a member of the conserved MRX/MRN complex required for DNA break formation. Despite limited purity for many of their Mer2 constructs, the authors nicely dissect these interactions and map them to different parts of the Mer2. A clear understanding of Mer2 clearly will require some structural analysis, especially of its tetrameric coiled-coil core region and its interactions with nucleosomes and other factors. But in the absence of 3D structure (which will be very difficult), the current study takes us a long way in understanding the roles of Mer2 in meiotic recombination. 

While most of the conclusions are well-supported by the data, a few observations would benefit from either stronger in vitro data or supporting genetic data. In particular, the conclusion that Mer2 specifically interacts with "open/unlocked" Hop1 and not its closed form is very interesting, but the pulldown gel in Figure 4C is not 100% convincing due to the presence of faint bands in pulldown lanes (perhaps contaminants?). More importantly, the finding that Mer2 interact directly with Mre11 is exciting, but needs some additional support - potentially in the form of chromosome spreads showing Mre11 or its partners are not recruited in the identified 3A/4A mutants of Mer2.

Reviewer #3 (Public Review): 

The manuscript addresses the mechanistic roles of Mer2 in meiotic DNA double-stranded break (DSB) formation in budding yeast. Mer2 is a central component of the meiotic DNA break machinery. Accordingly, Mer2 has numerous known interactions that are thought to (i) promote assembly of the DSB machinery, (ii) promote anchoring of the DSB machinery to chromosome cores and (iii) enable the recruitment of DSB sites to the DSB machinery. The manuscript makes an impressive effort to comprehensively characterize protein interactions of Mer2 that are relevant for the listed functions in DNA break formation. The experiments are well executed and informative. The results are mainly derived from in vitro experiments which uncover potentially important new functions and mechanisms for Mer2. In particular, the newly characterized interaction of Mer2 with nucleosomes and Hop1 will shape our models of meiotic recombination initiation. These discoveries will likely guide and necessitate future functional analysis.

Reviewer #1 (Public Review): 

Huang et al. investigated the role of SOX4 in endometrial stromal cell decidualization. The studies support the conclusion that SOX4 regulates endometrial stromal cell decidualization via enhancing PGR stability. Further, SOX4 deficiency is associated with embryo implantation failure in women with endometriosis that are undergoing IVF treatment. The data are novel and provide new information on how stromal cell decidualization occurs in women. With some exceptions, the quality of the studies is acceptable, and the conclusions are mostly well supported. 

Careful review found that some aspects of the experiments need to be clarified and extended. A technical concern is that the studies utilize immortalized cells and overexpression of SOX4 to conduct the ChIP-seq analyses rather than native stromal cells. Further, the studies of endometrial stromal cells from endometriosis patients are incomplete.

Reviewer #2 (Public Review): 

Proper development a receptive endometrium is essential to support embryo implantation and successful pregnancy establishment. Huang et.al. uncover the critical role of SOX4 in stromal cell decidualization by RNA-Seq and ChIP-Seq, and have further found that that SOX4 affects the protein stability of progesterone receptor (PGR), which is a master regulator of the decidualization process. In addition, the authors have also identified a direct regulation of SOX4 on the expression of FOSL2, which belong to AP1 complex that are involved in inflammation. The regulation of SOX4 on FOLS2 implicates an alternative role of SOX4 regulating decidualization beyond the PGR signal, which is interesting and opening new avenues for future investigations. Finally, the authors provide compelling evidence that the aberrantly decreased endometrial SOX4 expression is associated with patients with endometriosis and recurrent implantation failure in IVF clinics that showing implantation failure, supporting the clinical relevance of the study and pave way for therapeutic strategies for pregnancy improvement. 

The study is overall nicely performed, the data are convincing and overall support the conclusion. Some additional analyses and discussions may further improve the manuscript, as listed separately.

Reviewer #3 (Public Review): 

This study presents a detailed molecular description of an interesting molecular pathway for the understanding of decidualization in health and disease using an experimental design that cover in vitro and in vivo findings including a wide variety of experimental techniques. Conclusions are initially supported by the results obtained. However, the authors should apply more efforts providing more details about experimental design, methodology, and consistency in the structure and flow of the results to ensure their understanding, transparency, and reproducibility. It would be especially valuable to analyse PGR isoforms A and B since PGR-B played a predominant role in hESC decidualization and both isoforms influenced each other's transcriptional activity.

Reviewer #1 (Public Review): 

The authors demonstrate a translational mathematical model dependent on three key parameters for describing efficacy of checkpoint inhibitors in human cancer.The model parameters may serve as early and robust biomarkers of the efficacy of checkpoint inhibitor therapy on an individualized per‐patient basis.

Reviewer #2 (Public Review): 

The study by Butner et al. leverages a previously derived mathematical model (Butner et al. Sci Adv 2020) to predict immunotherapy response using published clinical data from immunotherapy-treated cancer cohorts. The model was fitted to a calibration cohort (meta-analysis, n=189) and then applied to a smaller validation cohort (n=62, Welsh et al., JITC, 2020). The estimated model parameters were tested for their ability to classify responders and non-responders. Using the tumour volume estimated from CT scans as input for the model, the immunotherapy response was predicted with an accuracy of 81.4% (n=62) within 2 months from treatment onset. 

Modelling of the anti-tumour response under immunotherapy is a relevant approach to understand the dynamics behind this process. The results from this study suggest that the model parameters for the tumour-cell killing rate and the ratio of cancer cells to cytotoxic cells are different on average between patients with objective responses and stable/progressive disease. The main advantage of this approach is that the estimations are derived solely using the CT scans to infer tumour volume. However, given that therapy response is characterized by a large tumour and immune heterogeneity, clonal selection over time, and importantly immune escape mechanisms, which were not considered in the model, a larger validation cohort is needed to confirm that the estimated parameters are robust predictors. Their predictive value also needs to be compared to current biomarkers of response. 

Interestingly, PDL1 positive cell percentage and CD8 T cell count were estimated based on the model parameters and compared between the different response groups. The average levels of estimated and observed T cell counts and %PLD1+ cells were comparable between the groups. However, to demonstrate correlation, the estimated and observed values need to be compared on patient level. This has the potential to be the focus and significance of the study, as it could be relevant in the absence of biopsy data.

Reviewer #3 (Public Review): 

This work proposed a non-linear mathematical model with a particular ordinary differential equation to capture the dynamics of tumor size over time in response to the immune microenvironment with treatment of checkpoint inhibitors. The parameters in the model are initially trained by a time-course dataset from six clinical trials consisting size changes over time of six types of tumors from 189 patients in response to the treatment of PD1 or PDL1, which were validated by an additional dataset from a study of 64 patients with non-small cell lung cancer. The authors further investigated the biological relevance of each parameter and found two of them μ and Λ were capable in classification of patients who are response or not to the treatment. However, the training procedure as well as the validation/testing of the model is not carefully evaluated, which could result in overfitting of the parameters to some datasets. 

Strengths: 

Instead of training a classification model to directly fitting tumor characters to the drug treatment in bulk level, this study built a non-linear mechanistic model to capture the dynamics of tumor size over time in response to tumor microenvironment and indirectly using its key parameters to classify the drug effects. This approach, integrated more intrinsic information at cell-cell interaction level, is potentially allowing build a more reliable predictive model across different cancers and treatments. 

Weaknesses: 

1) The prediction power of model is high depended on the robustness of performance in different tumors at different stage under different treatment, however, this study did not provide data on the effects of tumor heterogeneity. 

2) The parameter of proliferation constants (α) defined in the study is coupled and vary with dataset structure of each clinical trial, which should be evaluated independently by patients without the treatments or controlled data from in vitro experiments. 

3) It is unknown how the parameters of the model were trained or validated in batches and whether parameters were overfitting to the datasets.

Reviewer #1 (Public Review): 

In their manuscript "Repair of Noise-Induced Damage to Stereocilia F-actin Cores is Facilitated by XIRP2," Wagner et al. make a significant contribution to understanding how noise-induced damage to stereocilia is repaired. Through immunostaining of XIRP2 in isolated mouse inner hair cells (IHCs) exposed to noise-induced damage, as well as IHCs isolated from human patients, the authors clearly demonstrate the recruitment of XIRP2 to sites of damage within the stereocilia F-actin core (referred to as "gaps"). These gaps are ultimately repaired and filled in with actin monomers. They furthermore show through immunostaining that Xirp2 knockout IHCs lack the recruitment of γ-actin to gaps. Notably, the authors identify the enrichment of the short isoform of XIRP2 at gaps, but not the long isoform through antibodies specific to each isoform. The authors identified a predicted LIM domain in the C-terminal sequence of the short isoform of XIRP2. Truncation of the short isoform of XIRP2 (Xirp2-ΔCterm) resulted in a loss of γ-actin recruitment to gaps, phenocopying Xirp2 knockout mice. Finally, the authors clearly show that Xirp2 knockout mice are more susceptible to various types of hearing loss through hearing tests, including noise-induced hearing loss. In sum, their experiments support a mechanism by which the short isoform of XIRP2 is localized to gaps in stereocilia and facilitates repair by recruiting γ-actin to fill them. Repair of these gaps significantly contributes to stereocilia maintenance and prevention of permanent hearing loss in mice. Overall, the experimental work presented is convincing, and strongly supports a requirement for the XIRP2 short isoform in repairing stereocilia gaps. However, a substantive concern is that while the authors extensively suggest that XIRP2 likely employs a mechanism similar to other LIM domain-containing proteins for repairing mechanical damage to actin bundles, the evidence provided to support this claim is modest.

Reviewer #2 (Public Review): 

The 'hair' cells of the inner ear and their sensory apparatuses, called stereocilia, are non-renewable but have a limited capacity for self-repair in mammals. Following exposure to noise, a temporary reduction in hearing sensitivity can be followed by recovery. Repair of inter-stereocilia 'tip links' was proposed to contribute to this phenomenon, for example. Noise-induced actin gaps in the shaft of the stereocilia have been noted as well, but have been little studied to date. Here, Wagner et al. pick up on a story left off over a decade ago, in which Belyantseva et al. (2009) noted that damage-induced gaps in stereocilia may undergo actin remodeling, suggesting a new mode of hair cell repair. In this new manuscript, Wagner et al. sought to provide bona fide evidence of actin gap repair and to delve into the mechanism by which these gaps can be resolved. 

Wagner et al. show that noise-induced stereocilia actin gaps are repaired within 1 week following noise exposure, in itself an important new contribution addressing a long-standing question in the field. They also leverage two mutant mouse strains as genetic susceptibility models that show gaps without noise-exposure, broadening the importance of gaps and repair. They show that the actin binding protein XIRP2 is enriched in actin gaps in mouse auditory and vestibular hair cells, as well as human vestibular hair cells, and is able to bind actin monomers. Remarkably, XIRP2 knock-out mutants have a reduced ability to repair gaps, to accumulate monomeric actin at gap sites, or to recover their hearing post-noise damage as compared to control mice. Having identified XIRP2 short isoform as the protein form enriched at gaps, the authors engineer a clever XIRP2 mutant mouse line that specifically targets the C-terminus and LIM (putative actin-binding) domain of the short isoform, presumably leaving other XIRP2 functions intact. This model has the potential to specifically ablate XIRP2's gap filling behavior, and thus to assess directly how gap repair impacts hearing upon noise challenges. 

The authors provide painstaking controls to ensure that the observed reduction in actin gaps post-noise over time is due to gap repair, not hair cell death or stereocilia loss. In general, the authors achieved their goals of verifying the hypothesis that stereocilia actin gaps can be repaired after damage, and provided strong support that XIRP2 is required for actin gap repair. 

However several areas can be improved to further strengthen the manuscript. Among others, the gamma-actin immunolabeling raises some questions, as signal is absent at stereocilia shafts and the antibody used does not seem to be characterized. The XIRP2 DelCter mutant is used to show lack of XIRP2 and actin enrichment at gaps, but not to verify the prediction of an excess number of gaps and defective ABR threshold recovery in this model. The manuscript would also greatly benefit from modifications and additions in data reporting. 

Overall, this study provides a substantial and long-awaited contribution to our limited understanding of how mammalian hair cells maintain their precise cellular architecture over a lifetime of wear.

Reviewer #3 (Public Review): 

Wagner et al. address a form of damage to the bundle of F-actin filaments that comprises the core of stereocilia, which are mechanosensitive protrusions on the surface of auditory and vestibular sensory cells. The authors detect breaks in the F-actin core, primarily occurring following noise exposure in auditory stereocilia and for unknown reasons in vestibular stereocilia. Following noise damage, the number of gaps in auditory stereocilia increases within an hour and then decreases back to baseline over two weeks. Newly synthesized GFP-actin integrates into the stereocilia core, which is otherwise stably maintained with little to no actin turnover. 

Similar breaks in stereocilia were previously known to contain monomeric actin, espin, and cofilin. The authors have added the crosslinker XIRP2 to the list, showing compelling immunofluorescent localization data placing XIRP2 either in or at either edge of the break. XIRP2 knockout mice have more stereocilia breaks in both auditory and vestibular stereocilia. In auditory stereocilia the number of breaks steadily increases after noise damage over the normal recovery period, suggesting repair requires XIRP2. The mechanism may involve recruiting XIRP2 associated with monomeric gamma-actin, which is reduced in breaks in XIRP2 knockout vestibular stereocilia. XIRP2 is proposed to bind damaged F-actin in breaks via its C-terminal LIM domains. Correspondingly, XIRP2 lacking these LIM domains does not localize to breaks in vestibular stereocilia and the breaks also have lower levels of gamma-actin. 

This paper provides significant new insights into the generation and resolution of breaks in the F-actin core, establishing the general time course of repair following noise damage and showing that repair proceeds by an infill mechanism rather than wholesale replacement of stereocilia actin. The molecular process by which these patches arise is still unclear, but the data implicate XIRP2. 

The authors present a very interesting hypothesis regarding how XIRP2 functions, which is that XIRP2 is recruited to strained actin filaments in damaged regions of the stereocilia core via its C-terminal LIM domains, where it recruits monomeric actin used to repair broken filaments. Showing that XIRP2 LIM domains bound strained actin filaments and/or that strained filaments existed in the stereocilia breaks would make this hypothesis more compelling, as would demonstrating that XIRP2 recruited G-actin to breaks in noise-damaged auditory stereocilia as well as vestibular stereocilia.

Reviewer #1 (Public Review): 

The eukaryotic mitochondrial acyl carrier protein (mACP) has been shown to have two functions; as the acyl-chain carrier for FASII lipoic acid biosynthesis and as a chaperone for the heterodimeric cysteine desulphurase complex (IsD11-Nfs1) involved in the synthesis of Fe-S complexes. Previous studies have shown that the evolutionarily divergent protist, Plasmodium falciparum, lacks a mitochondrial FASII pathway but retains a putative mitochondrially located ACP. In this study, the P. falciparum mACP is shown to be essential for Fe-S complex formation, the assembly of the mitochondrial respiratory chain Complex III and the viability of red blood cell parasite stages. Using a conditional TetR knock-down system, pull-down experiments and homology modelling the authors demonstrate the mACP binds to the LYR protein Isd11 via a novel interface and stabilizes Nfs1, which in turn is required for expression of the Rieske protein and Complex III function. The conclusions are well supported by the data which are of very high quality. This study is important in identifying new mitochondrial processes that are essential for Plasmodium infectivity. More broadly, the study highlights the important and evolutionarily conserved role that mACP has in assembly of Fe-S complexes in eukaryotic cells and the extent to which this function can be decoupled from FASII fatty acid biosynthesis.

Reviewer #2 (Public Review): 

A summary of what the authors aimed to achieve:

Seeing that Plasmodium's FAS II pathway is hosted in the apicoplast, the role of a putative mitochondrial ACP is an enigma. The curious nature of this puzzle is enhanced by evidence that mACP is essential in red blood stages where few of the mitochondrial metabolic pathways are essential. The authors set to examine if mACP may have alternative functions to FASII, such as respiratory chain complex assembly, Fe-S biosynthesis and translation, as shown for mACP in other eukaryotes in addition to a role if FAS II. 

Furthermore, the Plasmodium mACP lack the 4-phosphopantetheine (Ppant) prosthetic group, whereas the acylation of mACP of other organisms is necessary for these "alternative" functions, because it mediates interactions with the so called LYR-proteins that are involved in those functions. Thus, the authors explored weather an interaction with LYR proteins occurs in Plasmodium, and how it is mediated in the absence of Ppant. 

An account of the major strengths and weaknesses of the methods and results:

Major strengths:

Since mATC wasn't studied in Plasmodium, it was necessary to confirm its mitochondrial localisation and assess its essential role in blood stages, for which the authors provided completing evidence. 

In light of the lack of FASII in Plasmodium mitochondria and known role for mACP in other mitochondrial function via interaction with LYR proteins, the authors identify a Plasmodium LYR protein to be homolog of the Fe-S biosyntheis component Isd11 and demonstrate direct interaction between PfmACP and Isd11 via reciprocal co-IP of tagged proteins, and through a heterologous system. Unbiased IP and MS further identify Nsf1, provides additional support to interaction with the Fe-S synthesis pathway. 

Through sequence analysis and comparison to the described interaction between mACP and Isd11 in other systems, the authors formulate an elegant hypothesis about their co-evolution in Plasmodium. This is backed with nice studied of the amino acid required for their interaction. 

Finally, the authors provide evidence for mACP essentiality in blood staged and deliver extensive characterisation of the defect of mETC function in the mitochondrial of parasites depleted from mACP. 

Major weaknesses:

Throughout the result and discussion the authors conclude that mACP is essential for Fe-S cluster biogenesis (importantly, lines 480-486, and figure 6, are extrapolating). However, while the interaction with Fe-S cluster biosynthesis pathway component is established, a role of mACP in Fe-S cluster biosynthesis or its control is implied from indirect evidence. It is possible that the depletion of complex III subunits and defect in mETC functions are an outcome of other mitochondrial defects. One example would be a defect in mitochondrial translation that leads to complex disassembly with an outcome on the abundance of nuclear encoded complex components. 

Nsf1 and Rieske instability is used as support for defect in assembly of Fe-S cluster biosynthesis pathway and indirect support for defect in Fe-S cluster biogenesis. However, the data is not presented with independent repetitions and statistical analysis nor with quantification of the EF1 control. Moreover, it is not specified if the EF1 proteins used for control is mitochondrial. An unrelated mitochondrial protein that is not down-regulated is essential to support the conclusion about specific instability of NSf1 and Rieske. 

The characterisation of the mACP phenotype is driven by the elegant hypothesis that it performs the same alternative roles that other mACPs perform in addition to FASII in other organisms. However, the work ignores other possibilities - what is the effect of depletion on other mitochondrial functions (e.g. biogenesis pathways such as protein import, division and translation) and is the effect on mETC primary or secondary. Likewise, what is the effect on other cellular functions, is the mitochondrial defect primary? 

An appraisal of whether the authors achieved their aims, and whether the results support their conclusions:

The authors establish that mACP is mitochondrial, and that it binds the LYR protein they had identified. They further established the mACP is essential for blood stages and that its depletion results in defects in mitochondrial mETC function. Finally, they provide a nice characterisation of the potential changes in interaction between mACP and a LYR protein that compensate for the lack of Ppant. 

A direct role is Fe-S cluster pathway assembly and Fe-S cluster biosynthesis is not directly established, and other mitochondrial functions are not examined. Finally, it is also not clear weather mitochondrial functions are the primary defect, since other cellular functions are not tested. 

A discussion of the likely impact of the work on the field, and the utility of the methods and data to the community:

The work will have impact on the understanding of cellular evolution as it highlights similarities and differences to the functions and mechanism of partner interactions of mACT between a model organism of the apicomplexans and previous observations from human and yeast. 

The work also unravels a new essential function in the disease-causing stage of Plasmodium. This has potential to produce impact via informing drug discovery efforts with a putative new target, however will be improved by further characterisation of mACP functions as mentioned in the weaknesses.

Reviewer #3 (Public Review): 

The authors set out to explore the function of mitochondrial acyl carrier protein (mACP) in the malaria parasite Plasmodium falciparum given that no acyl chains are 'carried' in the parasite mitochondrion because it does not make acyl chains therein. 

Targeting of mACP to the parasite mitochondrion is nicely confirmed, as is essentiality via conditional knockdown. Some effort was undertaken to home in on leucine 51 as a likely point of cleavage for the removal of the mitochondrial targeting leader (Figure 1C & Figure 1 supplement 2). Is it possible to make a targeted search through the peptide hits and look for a peptide commencing with leucine 51 as a sort of poor man's N-terminome? 

Binding of mACP to Isd11 is clearly demonstrated, as is further linking to Nsf1 to create a likely iron sulfur complex forming machine. I'm no structural biologist but it strikes me that a single protruding hydrophobic residue (F113) docked into a hydrophobic pocket on Isd11, plus a little cooperation from mACP V117, would make for a very weak interaction. Is that the sole binding interface? The mutagenesis (mACP F113A) abrogating pull down by nickel chromatography when expressed heterologously in bacteria is compelling. Are there data to show this pull down fails in the presence of detergent? Are there comparable examples of weak hydrophobic interactions generating such good binding? 

Line 900 - Figure 2 supplements 4 & 5. There appear to be many spectral counts in the table of mass spec hits for mAPC/Isd11 complex retrieved from bacteria by nickel chromatography, but only one peptide (being the largest possible) is displayed in supplement 5. Is there a reason that smaller, sub-peptides were not observed? 

On line 548 the authors speculate that a small molecule inhibitor of the protein/protein interaction between mACP and Isd11 might be a pan-apicomplexan drug. To better substantiate this speculation, it would be nice to include an alignment of the chromerid and apicomplexan mACP proteins to illustrate the apparent switch from a 4-phosphopantetheine prosthetic group attached via serine to a phenylalanine and the postulated hydrophobic binding interaction. 

Why was 1µm proguanil used for the 7 day growth assay (Fig 5A) and 5 day assay (Fig 5B) but 5µm for the MitoTracker imaging?

Reviewer #1 (Public Review): 

This manuscript investigated the localisation of infant cortical maps related to touch and noxious stimulation. Using a sample of 32 infants and cot-side measures of evoked brain activity assessed using fNIRS, they localise activity evoked by a non-noxious touch stimulus and a clinically required skin breaking blood test procedure (heel lance). The authors present results that suggest that the infants touch maps more closely overlap the somatotopic heel region than the nociceptive map, and that the nociceptive map is more poorly localised and extends inappropriately into the hand region. However, the conclusions of this paper are mostly unsupported by the results. The following four issues should be noted when reading the current manuscript. 

1) Differences in regions of activations constituting touch maps and nociceptive maps are never actually statistically tested: 

The core finding of this paper is that the infants' nociceptive maps and touch maps are not aligned, they are different: "The key difference between the two patterns of activation is that the foot touch response is limited to the areas of the S1/M1 associated with the foot, whereas the lance response extends towards other more ventral regions of S1". However, this claim of differences between touch and nociceptive maps is never actually statistically tested, and therefore this core finding is currently not backed by the analysis and results. 

The way the analysis is conducted is as follows: (i) a region of activation for the noxious stimulus is found by comparing evoked activity to pre-stimulus baseline and statistically significant regions identified, (ii) a region of activation for the touch stimulus is found by comparing evoked activity to pre-stimulus baseline and statistically significant regions identified, (iii) these noxious and touch activation maps are overlaid and regions of overlap are identified as well as regions of non-overlap, and it is these regions of non-overlap that are used to conclude that the maps differ. However, these analyses are invalid as the only statistical comparisons made are relative to baseline. The activation evoked by the noxious stimulus is never directly statistically compared to the activation evoked by the touch stimulus i.e. the regions of non-overlap differ in their statistical relationship with baseline, not with their statistical relationship to each other.

The nature of this analysis flaw is expanded upon in the following eLife paper "Science Forum: Ten common statistical mistakes to watch out for when writing or reviewing a manuscript" (https://elifesciences.org/articles/48175), problem number 2 "Interpreting comparisons between two effects without directly comparing them". This eLife paper also advises pointing to the following paper: https://doi.org/10.1038/nn.2886. And the issue is also detailed here: https://doi.org/10.1198/000313006X152649. As outlined in the references provided, the way to demonstrate a difference in the two maps in a valid manner is to directly compare them and identify statistically significant differences in regions of activation. 

This error is major flaw in the study, as the inappropriate analysis underlies the core novelty of the paper. All other statistically significant findings in the paper have been previously identified: somatotopic organisation of infant S1 has been reported by Dall'Orso et al., 2018; much larger haemodynamic responses to noxious stimulus compared to touch stimulus have been reported by Verriotis et al., 2016; and greater spread (FWHM) of the noxious evoked response is related to the larger response amplitude and "smearing" of signal due to low resolution of haemodynamic responses in general i.e. it would be odd if the FWHM was not greater for the larger evoked amplitude. 

2) Nociceptive maps are equally as widespread as touch maps: 

Describing the nociceptive maps as widespread seems inappropriate, unless the touch maps are also described as widespread, which they are not. The authors explicitly demonstrate that both stimulus types (noxious lance and non-noxious touch) evoke activations over areas that are matched in extent i.e. areas of activation do not differ, Fig 3 part b and Figure 3 figure supplement 1 part b. Given that the authors explicitly demonstrate that there is no statistically significant difference in activation extent, both noxious and touch maps should be considered equally widespread. However, this point that nociceptive maps are not uniquely widespread, that this is explicitly demonstrated to be equally the case for non-noxious touch maps, is not made, but is an important point. 

In contrast, the key point is the apparent difference in location of the activated areas i.e. a point about the activity being "mislocalised" rather than being "widespread". The authors show that both stimulus types evoke activations that are centred on the same location (i.e. peak locations do not differ), but the touch stimulus evoked activity extends more medially (toward the location of the foot region), while the noxious stimulus evoked activity extends more laterally (away from the location of the foot region). 

3) The analyses that produce areas of activation in source space (image space) are circular: 

The assessment of changes in [HbO] and [Hb], relative to baseline, in image space is a circular analysis, as it is restricted (informed) by a strongly related analysis in sensor space - "significant peak reconstructed changes in d[HbO] and d[Hb] were identified with a two-tailed t-test (a = 0.01) comparing the peak time point within a 5-second window around the peak latency derived from the channel-wise analysis against the baseline". That is, analysis of the data in sensor space (channel-wise analysis) identified the temporal region-of-interest by comparing evoked activity to pre-stimulus baseline, and this information was used to restrict analysis in source space (image space) to identify regions-of-interest by comparing evoked activity to pre-stimulus baseline. This sequence of analyses is double dipping and therefore circular. 

This issues is outlined in the eLife paper "Science Forum: Ten common statistical mistakes to watch out for when writing or reviewing a manuscript" (https://elifesciences.org/articles/48175), problem number 6 "Circular analysis". This eLife paper also advises pointing to the following paper: https://doi.org/10.1038/nn.2303. 

4) Limitations of the fNRIS technique and of the link between properties of nociceptive maps (widespread/mislocalisation) and poorly directed behaviour are not adequately discussed: 

Tissue damaging painful events elicit cardiovascular responses and infants have underdeveloped cerebral autoregulation. A major potential difference between the touch and noxious stimuli is that the noxious stimuli can elicit blood pressure changes that get reflected in a haemodynamic response measure. This is true for all haemodynamic imaging modalities such as fMRI, fUS, as well as fNIRS. Thus haemodynamic responses are not a perfect reflection of neurodynamic responses, especially in infants. In this manuscript, the authors appear to be aware of this potential issue as in the methods section they use PCA denoising to remove global signals that could be related to blood pressure effects and reference relevant literature: "Optical density changes recorded from all channels (likely related to stimulus dependent systemic physiological changes) were removed using Principal Component Analysis ((Kozberg and Hillman, 2016; Tachtsidis and Scholkmann, 2016); 1 component removed)". 

However, as in all physiological denoising methods, limitations need to be clearly outlined. It is highly unlikely that removing a single PCA component fully eradicates any risk of haemodynamic responses that might not be reflective of neural activity. And as the authors state, the noxious evoked haemodynamic signal appears to be quite widespread including the control channel. While the widespread responses is not statistically significant (see comment 2 above), the fact that the authors are claiming widespread responses extending into control channels seems inconsistent with their claim in the discussion that "The change in the d[HbO] and d[Hb] following sensory stimulation is a measure of neural activity", the NIRS signal "is presumably due to greater depolarisation and spike activity within the activated areas", and their lack of discussion around the limitations of using haemodynamic measures such as fNIRS. In the discussion, fNIRS is contrasted with other modalities such as EEG and fMRI, and fNIRS is claimed to be "ideally suited to a study of this kind". While one can agree that fNIRS has several advantages over techniques such as EEG and MEG, one major issue with fNIRS and (fMRI) is that it is indirect and relies on underdeveloped cerebral vasculature and underdeveloped neurovascular coupling. 

Given that the core claim of this manuscript is that the noxious evoked activity is poorly localised (and maybe widespread), one major risk of using haemodynamic measures for pain studies in infants is that they can easily reflect cardiovascular effects in addition to neural effects. And while removal of a single PCA component was performed, it is not convincing that this simplistic clean-up step removes all risk of non-neural cardiovascular effects.

Reviewer #2 (Public Review): 

Jones and colleagues investigated the topographical similarity of primary somatosensory cortex responses to painful and non-painful touch stimuli in newborn human infants.  Their hypothesis was that, as in non-human animal models, responses to non-painful stimuli would be more mature and organized than responses to painful stimuli, which would spread to parts of the somatosensory map other than the affected region.  They assessed responses to touch stimuli (repeated, light hammer taps) on the hand and foot as well as responses to a painful heel stick performed for a necessary blood draw.  Statistically significant, non-overlapping responses to the hand and foot touches were observed in an organization similar to that seen in adults.  The painful heel stick had a similar peak to the non-painful foot stimulation, but the extent of this peak was larger, and the response to the painful stimulus spread into non-overlapping parts of somatosensory cortex, including one channel that had responded to hand stimulation.  They conclude that pain responses are more widespread and disorganized relative to innocuous touch sensations early in post-natal development. 

Strengths:

The basic pieces of the methods used here are well chosen to answer this fundamental question about brain development. As the paper points out, fNIRS is an ideal imaging modality here.  It provides the necessary spatial specificity and resolution (relative to EEG), while allowing for imaging to occur in flexible environments (relative to fMRI), including the medically necessary, painful procedure used here.  The array of optodes was also well designed, providing good density over key regions of interest, as well as a control channel.  The authors report changes in both oxygenated and deoxygenate hemoglobin, which helps to support the claim that changes represent canonical hemodynamic responses. I also appreciate the variety of dimensions in which the responses to touch and pain stimuli were compared, including comparisons of the amplitude and spread the peak response using a full width half max calculation, in addition to the number of channels showing significant responses relative to baseline. 

Weaknesses:

Two elements of the experimental design and implementation make it difficult to tell if the differences in responses to painful and non-painful stimulation are actually a product of different properties of early cortical processing of nociceptive vs. innocuous sensation.  First, the touch stimuli were repeated over approximately 10 trials per participant, whereas there was only a single trial of the nociceptive stimulus.  In infants, as in adults, repeated stimuli lead to habituation, which is associated with suppression of the neural response.  Thus, averaging over 10 trials of the touch stimuli may result in an estimated response that is substantially dampened compared to the response that would be observed to a single, initial trial.  This may explain some of the differences in peak amplitude and extent for painful vs. non-painful stimulation.  

Second, when infants moved in the vicinity of a touch trial, that trial was excluded; in contrast, when infants moved after the heel stick, those data were retained.  The latter choice was essentially a necessity, as nearly all of the heel stick participants moved.  The authors justify the choice by saying that the movements were idiosyncratic and therefore "any associated cortical response would be removed during the averaging process" (line 247).  Of course, the responses to such motion would not be removed but rather averaged into the mean group response.  It is understandable to argue that this would result in unavoidable noise that could be overcome by strong signal, but under this justification and for parity between conditions, the authors should also retain motion-contaminated trials in the touch conditions.  I do also worry that the conditions of the experiment (infants were held closely by their mothers) could result in an underestimate of the motor responses produced by infants in the pain condition, who may have attempted to move their arms or legs more than observed but were held back by their parent, leading to additional pressure and sensation across additional, non-stimulated body parts in that condition.  Moreover, the authors do not discuss any differences in gasping or crying between the two conditions, which can also influence cerebral blood oxygenation.  (Perhaps this could account for the significant pain response in the control channel, which is largely ignored in interpretation of the results.) 

One final concern regards the channel-wise statistical analysis.  The methods state that participants' responses were averaged into a single time course, each time point of which was compared to a baseline distribution (rather than comparing a distribution of responses to a single baseline estimate, or pairing individual participants' response and baseline values).  It would seem, however, that the variability in somatosensory responses across participants, and not just variability in baseline, ought to be retained and used to assess statistical significance.  This is particularly a concern given the motion confound.  For example, strong responses in four infants making large arm or head movements would differentially affect analyses in which 1) those responses only pull up a single average value for each time point, versus 2) those responses also increase the variability of a distribution. Retaining variability in responses to trials would also facilitate directly comparing responses to foot touch and pain in channels claimed to be the locus of the disorganized pain response, which would strengthen the claim that early pain responses are misaligned with the topographical map of responses to touch. 

Conclusion: 

Given these concerns about the implementation and analyses, it is not clear whether the authors' conclusions are supported by their data.  However, adjustments or supplementary analyses could bolster the case for their interpretation that humans, like other animals, initially show exaggerated and disorganized cortical responses to pain in early development.

Reviewer #1 (Public Review): 

In this manuscript, the authors isolate a cell culture adapted HCV with 2 point mutation in E2. This is associated with increased infection, rate of entry, and less reliance on the entry factor SR-B1. Tradeoffs are that the virion is less stable in culture and is much more sensitive to antibody neutralization. Biophysical analysis suggests that the mutations stabilize HVR1 with properties that resemble a deletion of HVR1. Indeed deletion of HVR1 has enhanced properties of the double point mutant: higher infection, but less thermal stability and resistance to neutralization. This produces a model wherein HVR1 is a disordered peptide tail that blocks antibody neutralization and inhibits CD81 interaction. Binding of E2 to SR-B1would alter HVR1 conformation, exposing the CD81 binding site and leading to productive entry. 

I found the experiments to be well done and interpreted. The model explains a number of observations regarding receptor usage and how HCV evades antibody control via HVR1, whose disordered nature enables mutation to continually evade antibody responses. 

I thought the manuscript was complete with only minor corrections necessary.

Reviewer #2 (Public Review): 

The molecular mechanism of how HCV enters a host cell remain undefined. Envelope glycoprotein 1 and 2 (E1 and E2) are responsible for facilitating entry. E2 demonstrates conformational flexibility with three hypervariable regions (HVR) and several disordered regions. Using a series of diverse virology and computational methods, the authors claim that HVR1 acts as a "safety catch" that regulates entry. The employment of MD simulations to look at conformational mobility is compelling.

Reviewer #3 (Public Review): 

The complex mechanism of HCV entry has long been an area of focus for both fundamental virologists as well as groups interested in generating vaccines. The virus employs multiple surface receptors to mediate cell entry, which occurs via clathrin-mediated endocytosis and is linked to a complex cascade of both host signalling and the induction of additional responses. 

The study utilises cell culture adaptation to decipher how HCV might improve its ability to enter and infect cells. This is done twice, once in the absence of antibodies, once with them present. Resultant mutations in the absence of antibody selection lead to lesser dependence on SR-B1, increased affinity for CD81, and enhanced susceptibility to neutralising antibodies. In addition, the thermal stability of virions is reduced. This phenotype is recapitulated by genetic deletion of the hypervariable region 1 (HVR1). The enhanced ability for these mutants to enter cells is supported by mathematical models and MD simulations support that the disordered HVR1 region becomes stabilised as a result of these changes. 

The authors therefore propose that a loss of entropy in the E2 protein is responsible for this "hyper-reactive" phenotype. In nature, this would be achieved through SR-B1 interactions, hence the loss of dependence upon this receptor in vitro. 

Strengths:

This paper provides a new concept for how the multi-step entry process in HCV takes place, based upon the two known physical interactions with cellular receptors. The enhanced affinity for CD81 appears directly related to the loss of entropy in the system, and according to the mathematical model, ensuing events post-CD81 binding might also be accelerated. The authors propose that by understanding the nature of the switch that the development of vaccines might be enhanced. The combination of MD, virus culture, biophysical methods and mathematical models provides a strong argument in support of the hypothesis. The work has been carried out to a very high standard and presented very nicely. 

Weaknesses:

Whilst the mathematical models support accelerated entry at multiple stages, steps following the interaction with CD81 are not explored beyond determination of ensuing viral titres.

It is also unclear whether there might be a region of E2 in which the changes in overall structure can be rationalised in terms of mechanism. For example, do regions interacting with CD81 and/or E1 show any alterations that might explain altered behaviour? It is difficult to make out from the RMSF plots whether this might be the case. 

Given that HCV enters cells via clathrin-mediated endocytosis, it would be of interest to determine whether the enhanced entry phenotype was also related to pH. HCV is unusual compared to e.g. influenza in that the loss of infectivity following acid pH treatment can be restored by re-buffering virions to neutral conditions - could this be related to the entropic catch and associated structural changes? 

Whilst this may be difficult to address, would antibody binding to HVR1 also be expected to reduce entropy? Could non-neutralising, yet HVR-binding antibodies therefore lead to enhanced entry and SR-B1 independence? 

Lastly, one presumes that the antibody selected long term culture failed to select any notable mutations? Was this assessed?

Reviewer #1 (Public Review): 

The authors present a comprehensive, well documented, and easy to use toolbox for processing and analysing gene expression data for comparisons with neuroimaging data. 

The tool is well designed and well documented. In the paper, it is used to show how different choices of processing can affect the outcome of 3 different types of gene expression analyses. The fact that they can do such analyses, as well as replicate published analyses and examine the outcome of their processing choices nicely illustrates how flexible this toolbox is.

Reviewer #2 (Public Review): 

In this manuscript, Markello and colleagues exhaustively characterize the impact and relative importance of the many data-processing decisions that go into constructing whole-brain transcriptomic maps from microarray data in the Allen Human Brain Atlas. The authors motivate the need for and have developed an open-source toolbox, abagen, for standardizing workflows in imaging transcriptomics. The authors propose a taxonomy of analyses commonly performed on these data in the literature; they then use abagen to compute the distributions of statistical outcomes for three prototypical analyses across 750,000 combinatorial choices of end-to-end data-processing pipelines. Informed by these findings, the authors then place into context several specific pipelines reported in recent and influential studies. 

The paper is well-written and the authors are successful in illustrating and attempting to address the need for standardized and systematic research in the burgeoning field of imaging transcriptomics. The abagen toolbox is an important contribution and is to my knowledge the current state-of-the-art. The code is clean, flexible, and very well-documented. The chief weakness of this paper is the lack of clear guidance on best practices. Readers should, however, be sympathetic to the fact that there is currently a lack of ground-truth data against which to benchmark different data-processing pipelines. 

Even after reading the paper thoroughly, it's still not completely clear to me whether the analyses in this study are performed for cortex only, or at the whole-brain level (or bi- or uni-laterally for that matter). I'm assuming this study is cortex-only as you say in the methods that "the brain atlas used in the current manuscript represents only cortical parcels." But abagen supports joint cortical+subcortical atlases too. It'd be helpful to readers to make this explicit. Along similar lines, do you expect any of the main findings of this study to change when deriving whole-brain maps? 

Would it make sense to use PET maps or another type of neuroimaging data as a (pseudo-)benchmark in a future study? What about a cross-validation strategy where data are selectively withheld during processing and then predicted after the fact? This may only be possible for a subset of genes and/or pipelines, but it could nonetheless be informative. 

In the discussion, you claim that "the optimal set of processing parameters will very likely vary based on research question." I'd like to see this elaborated on a bit further, at least for the most important parameters. For example, when would it make more sense to use one form of gene normalization over the other? What are the implicit assumptions underlying each choice? 

Is there anything to be said about the order of operations? There seem to be several steps in Table 1 which could conceivably be interchanged. If nothing else, this procedural ambiguity is yet another good reason to standardize workflows. 

I particularly liked the analysis in Figure 2A and thought it made a nice contribution to the paper.

Reviewer #3 (Public Review): 

The work “Standardizing workflows in imaging transcriptomics with the Abagen toolbox” is a major meta analysis pipeline workflow for comparing and integrating parameter choices in imaging transcriptomics using the Allen Human Brain Atlas (AHBA). The release of the AHBA has strongly increased the interest in determining transcriptomic associations in brain imaging studies, yet there is much variability in the analysis, methods used, and subsequent interpretation. 

This work is illustrative of an important trend in informatics analysis allowing strong metadata control by users so as to access, and implement optimal choices of parameters and to study there distribution. The work implemented as an open source Python toolkit is likely to be of importance to analysts working in these areas. 

It would be helpful to clarify and specifically define the term pipeline as a specific set of parameter, normalization, and other choices that are selected. Whereas this term is in common use in the field, in the present work the meaning is specific to a set of selectable options. Of course the any number of such variable selections could be implemented in the Abagen toolbox, it will help for clarity to more clearly define this term up front. 

My major consideration in this work concerns are two issues. The first is how to characterize and summarize the results of pipeline output produced by Abagen. The manuscript illustrates the workflows and various means of summarizing results but does not offer guidance into preferred interpretation of relative value of the results. Whereas we may argue that the primary purpose of Abagen is to run the various pipelines, allowing downstream interpretation to the user, it would be helpful to understand how the Abagen toolbox organizes, summarizes, and sets this output options up for interpretation. This appears to be only weakly addressed in the present manuscript. 

The second point I of importance I believe is more description of the available functionality in the toolbox, perhaps as more of a specific use case analysis. The authors provide substantial documentation on installing and working with Abagen, and but some more direct indication of how the toolkit would be used would be valuable. 

The scale of this work is impressive and the work may be widely used by the neuroimaging community. 

I am enthusiastic about this work but would like to see more description of how the Abagen toolbox might be used better converge on more strongly interpretable results. I certainly understand the issue of ground truth remains open but it would seem that the toolkit might be able to summarize and/or statistically priorize pipeline results for users so as to afford better interpretation. 

A second point concerns somewhat more description of what is actually available functionally in the toolkit, at least as a summary.

Reviewer #1 (Public Review): 

Pöge at al present a cryo-tomographic study of the rod outer segment (ROS). These are specialised cilia of rod photoreceptor cells, which are essential for sensing light cues and initiating the vision process. They are elongated structures filled with stacks of membranous discs which contain rhodopsin, the major light-sensing protein, and other proteins, some of which maintain the disc architecture. 

In this paper, they take advantage of the structural preservation of cryo-FIB milling, and of the unprecedented molecular details available in the resulting cryo-tomograms, to investigate the elements that maintain the elaborated disc architecture. 

The authors provide a general description of the rod ultrastructure, providing precise description of the membrane arrangement and geometry. They then focus their attention to two regions: 

1) Disc connections. Here the authors are able to identify connecting densities and to propose, based on their length and density distribution, that the connections at the disc rims and those on the 'flat' internal surface are of different protein composition. They discuss which protein complexes are likely to contribute to these two types of connectors, although the experimental evidence for this is weak and theirs remains a model to be tested. 

2) Proteins generating/maintaining curvatures at the disc rims. Here subtomogram averaging provides intermediate resolution structures that are entirely novel, and clearly show a regular array of proteins holding the high membrane curvature. As before, the authors tentatively propose the identity of the proteins forming these structures, and while their assignment seems entirely reasonable there is no direct experimental evidence to support it. 

Overall, the quality of the reconstructions presented allows the achievement of a level of insight into rod membrane architecture that will be of interest to scientists working in the fields of in situ structural biology, and of ROS and more generally cell ultrastructure.

Reviewer #2 (Public Review): 

1) The novelty of the current observation of two types of links is overstated, for example, in the abstract: "Our data reveal the existence of two molecular connectors/spacers which likely contribute to the nanometer scale precise stacking of the ROS disks" (Line 25). In fact, both of these links have been shown before (Usukura and Yamada, 1981; Roof and Heuser, 1982; Corless and Schneider, 1987; Corless et al., 1987; Kajimura et al., 2000). These previous studies deserve to be recognized. Of special note is the paper by Usukura and Yamada whose images of the disc rim connectors are by no means less convincing than shown in the current manuscript. On the other hand, the novelty and impact of the data related to peripherin appears to be understated, particularly in the abstract. 

2) Notably, ROM-1 has not been found in peripherin oligomers larger than octamers (e.g. Loewen and Molday, 2000 and subsequent studies by Naash and colleagues). This should be discussed in the context of the current model. 

3) The following statement should be reconsidered given the established role of cysteine-150 in peripherin oligomerization: "We hypothesize that the necessary cysteine residues are located in the head domain of the tetramers (Figure 5B), ..." It has been firmly established that only one cysteine (C150) located in the intradiscal loop is not engaged in intramolecular interactions and is essential for peripherin oligomerization. \

4) Line 340: "A model involving V-shaped tetramers for membrane curvature formation was proposed recently (Milstein et al., 2020), but it comprises two rows of tetramers which are linked in a head-to-head manner. Our analysis instead resolves three rows organized side-by side in situ (Figure 5A)." I am confused by this statement: doesn't your model also show long rows connected head-to-head? The real difference is that Milstein and colleagues proposed four tetramers per rim whereas the current data reveal three. 

5) Line 347: "Our data indicate that the luminal domains of tetramers hold the disk rim scaffold together (Figure 3C), which is supported by the fact that most pathological mutations of PRPH2 affect its luminal domain (Boon et al., 2008; Goldberg et al., 2001). It is possible that these mutations impair the formation of tetramers, rows of tetramers, and their disulfide bond-stabilized oligomerization. These alterations could impede or completely prevent disk morphogenesis which, in turn, would disrupt the structural integrity of ROS, compromise the viability of the retina and ultimately lead to blindness." This is not an original idea, as many studies showed that disruptions in peripherin oligomerization lead to anatomical defects in disc formation and subsequent photoreceptor cell death. 

6) In regards to the distance between disc rims and plasma membrane, the authors cite the data obtained with frogs (10 nm) but not a more relevant, previously reported measurement in mice (Gilliam et al, 2012). The value of 18 nm reported in that study is much closer to the currently reported value. 

7) The authors are (correctly) being very careful in assigning the molecular identity of disc interior connectors to PDE6. However, they are more confident in assigning the disc rim connectors to GARP2, which is reflected in the labeling of these links in figure 5. Their arguments are valid, but these links are not attached to peripherin (a protein considered to be the membrane binding partner for GARPs), which is not immediately consistent with this hypothesis. Perhaps it would be fair to re-label the corresponding links in figure 5 as "disc rim connectors". 

8) On a similar note, the disc rim connectors seem to be located where ABCA4 is presumed to be localized within the rim, which may not be just a coincidence. The authors already have tomograms obtained from ABCA4 knockout animals. Is it possible to analyze whether these links are preserved in these tomograms?

Reviewer #3 (Public Review): 

This manuscript provides an updated and more detailed view on mouse rod outer segment (ROS). Using cryoFIB milling and cryoET the authors imaged intact ROS. ROS comprises of highly ordered, densely packed membrane disks that anchor rhodopsin. ROS ordered ultrastructure is crucial for vision and the authors aimed in this study to determine the structural basis for achieving this order. 

The authors revealed potential molecular scaffolds both in the lumen of the membrane stacks and on the surface of the stack. On the outer surface, two distinct molecular connectors were observed, likely contributing to the precise stacking of the membrane disks. One class of connectors specific to the disk periphery/rim and one distributed on the rest of the membrane. Within the lumen and specific to the curved edge a continuous supermolecular assembly of three rows was observed. Based on previous literature, the authors speculate on the identity of the scaffold proteins they observed. 

The imaging methods and image processing applied in this study are state-of-the-art. The methods are described in detail so can be used as a resource for other groups applying similar techniques. The provided ROS ultrastructure with the averaging of the macromolecular assemblies of the disk rim are of the highest quality and will be highly valuable for the field of phototrunsduction. 

The weakness of this study is in the molecular identification of the components forming the macromolecular assemblies observed. The resolution obtained by subtomogram averaging doesn't allow molecular identification and there weren't any direct experiments attempting to identify the components. Thus the molecular basis in this study is highly speculative.

Reviewer #1 (Public Review):

In this manuscript, Akaki et al. describe a new mechanism by which the activity of Regnase-1, an endonuclease that degrades mRNAs encoding inflammatory mediators, can be regulated. By determining the interactome of Regnase-1 in IL-1b or TLR-ligand stimulated cells, they found that Regnase-1 binds to bTRCP (as previously described) as well as to 14-3-3 proteins, which is novel. The authors further identify the phosphorylation sites on Regnase-1 that are required for the Regnase-1:14-3-3 interaction, and show that the interaction is mediated by the activity of IRAK1/2. By generating knock-in mice carrying a phosphodeficient mutant of Regnase-1, the authors demonstrate that the interaction with 14-3-3 blocks the ability of Regnase-1 to degrade its target mRNA IL-6, as it can no longer bind to the target mRNA. Finally the authors show that binding to 14-3-3 prevents nucleocytoplasmic shuttling of Regnase-1 and therefore target mRNA recognition.

General comment:

This is an important study that describes a new mechanism by which Regnase-1 is inhibited upon immune activation, which mediates efficient synthesis of inflammatory mediators whose mRNAs are normally degraded by Regnase-1. The interaction with 14-3-3 presented here was not known before, and the authors describe the interaction and its consequences in great detail. In general, the study is well conducted and the results are both clear and convincing. The analysis of phosphodeficident Regnase-1 knock-in mice is a major strength of the study. However, there are some smaller points that the authors could address to further strengthen the manuscript, e.g. the mutually exclusive binding of Regnase-1 to bTRCP or 14-3-3, and the possibility that IRAK1/2 may directly phosphorylate Regnase-1. In addition, they should more directly measure the effect of phosphodeficient Regnase-1 on IL-6 mRNA decay, and generalize their observation that 14-3-3 binding prevents Regnase-1 mRNA binding and decay.

Specific comments:

The data suggest that IRAK1/2 may directly phosphorylate Regnase-1 (Fig.2G-I), although the authors do not address this question either experimentally or in the discussion. Do the authors have evidence that Regnase-1 is a direct target of IRAK1/2? Minimally, the authors should discuss this point and assess whether the identified phosphorylation sites conform to consensus IRAK target motifs.

The evidence for mutually exclusive binding of Regnase-1 to bTRCP or 14-3-3 is rather indirect, through the analysis of Regnase-1 phosphorylation status and phosphomutants (Fig.3). This point could be strengthened by competition assays, in which expressing increasing amounts of one protein should weaken the interaction with the other.

The authors show that IL-6 mRNA levels are similar in Regnase-1 WT and S513A mutant cells (Fig.4D-F). This notion is difficult to reconcile with the author's finding that 14-3-3 binding to Regnase-1 prevents it from binding to and inducing the degradation of IL-6 mRNA. This discrepancy should be discussed critically. Moreover, previous studies have shown that changes in mRNA stability are not necessarily reflected by changes in mRNA steady state levels, since they can be buffered by altered transcription (see e.g. Singh et al. 2019; PMID: 31116665). Therefore, mRNA degradation rates should be measured directly by actinomycin D chase or similar experiments.

The authors claim "that 14-3-3 inhibits Regnase-1-mRNA binding, thereby abrogating Regnase-1-mediated mRNA degradation". However, this is only shown for IL-6 mRNA (Fig.5G). Since this is a key result of the study, the authors should also test other targets of Regnase-1 so as to generalize their finding.

Reviewer #2 (Public Review):

The authors used immunoprecipitation followed by mass spectrometry to identify proteins interacting with Regnase-1 before and after stimulation with IL-1β. IL-1β treatment induced a previously unknown interaction between Regnase-1 and 14-3-3 proteins. 14-3-3 bound predominantly to phosphorylated Regnase-1 and specific phosphorylation sites were identified. 14-3-3 binding to Regnase-1 was mutually exclusive with βTRCP, binding of which is known to induce ubiquitination and degradation of Regnase-1. 14-3-3 binding prevented Regnase-1 degradation, but also inactivated it by blocking mRNA binding. 14-3-3 binding also prevented translocation of Regnase-1 from the cytoplasm to the nucleus. This study has identified a second mechanism by which Regnase function can be blocked to increase expression of inflammation-related mRNAs.

Overall, the authors' conclusions are supported by the data. The results of this study significantly advance the understanding of the regulation of Regnase-1 activity in inflammatory gene expression. The data are likely to be of interest to those investigating the intracellular signaling pathways that control gene expression in response to inflammation. The authors identified important sites for Regnase-1 regulation and created several mutant Regnase-1 constructs that will be of use to the research community. In addition, the transcriptomic and proteomic datasets generated in this study are likely to be of further benefit.

Reviewer #3 (Public Review):

Here, Akaki and colleagues set out to identify how Regnase1 is regulated upon cells being stimulated with IL-1Beta or TLR ligand stimulation. To do this they stimulated cells and then carried out a proteomic analysis to identify proteins that specifically interact with Regnase1 in stimulated cells. They identified Rengase1 interacting with the Beta-transducin-repeat containing complex (TRCP), a previously published interaction, which leads to Regnase1 ubiquitination and degradation. Interestingly, they also identify 14-3-3 proteins. Based on other data, they conclude that TRCP and 14-3-3 interact with Regnase1 in a mutually exclusive manner. They go on to show that the interaction between 14-3-3 and Regnase1 is mediated in IL-1B/TLR-stimulated cells by IRAK1/2 through an uncharacterized C-terminal domain. Two phosphorylation sites (S494 and S513) regulate 14-3-3 interaction with Regnase1, while different sites are required for Regnase1 interaction with TRCP and proteosomal mediated degradation. Finally, they conclude based on their data that 14-3-3 binding to Regnase1 stabilizes Regnase1 but prevents nuclear-cytoplasmic shuttling of Regnase and also Regnase1-mRNA association.

The manuscript is interesting and presents another layer with respect to how Regnase-1 activity is regulated during the immune response. However, several points should be addressed in this reviewer's opinion that would help strengthen the manuscript.

Reviewer #1 (Public Review):

Lee et al. report on population-level cell type changes in primary motor cortex (M1) as water-deprived, head-fixed mice undergo associative learning. Mice were exposed to a condition stimulus (auditory tone), followed by a 1.5s delay and then delivery of an unconditioned stimulus (water reward), while the group simultaneously carried out in vivo two-photon calcium imaging of pyramidal neurons (PNs), somatostatin-, parvalbumin-, and vasoactive intestinal peptide-expressing inhibitory neurons (SOM-Ins, PV-Ins, and VIP-Ins, respectively) on the first and seventh day of the task. This investigation indicates that there are cell-type dependent responses encoding the cue and reward stimulus in M1. The group asserts that all of these four major cell types show distinct modifications after associative learning and that cue- and reward- related signals are coded for by major cell types in M1. In particular, it is suggested that PV-INs modulate local microcircuit activity related to the CS association in M1 and that VIP-INs act as a context-dependent switch following the reward delivery. This study provides evidence that M1 may have a broader, more diverse range of functions than previously appreciated

Strengths:

The paper takes a broad, comprehensive look at non-motor related responses in M1 during associative learning, dissecting responses across all the major cell types. This work provides a better appreciation of the heterogeneity of responses observed across the cortex and how those changes may emerge through experience. Overall, the experiments, data analysis, and results reported are highly rigorous.

Weaknesses:

The major weakness of the manuscript is the lack of analysis of motor-related activity that presumably exists in their data set. While the authors intentionally focus on non-motor responses and select a behavior that does not necessarily involve motor learning, the significance and interpretation of their findings seems incomplete without examining the motor activity and its relationship to cue and reward activity. This could better disambiguated whether the plasticity in cue and reward activity in M1 reflect local circuit changes as the authors implies or changes occurring in upstream areas that are then inherited by specific cell types in M1.

Reviewer #2 (Public Review):

Using advanced live brain imaging techniques, the authors studied the activities of neurons in the primary motor cortex of mice during a classical conditional task, in which a tone is paired with water reward. They found that distinct types of neurons respond differently to the auditory cue or the reward, and the responses evolve differentially as learning proceeds. This work reveals an interesting role of the motor cortex beyond its well-recognized function in motor control, and suggests distinct functions of pyramidal neurons as well as various interneurons in reinforcement learning.

The investigation of the M1's role in classical condition is intriguing and the finding is interesting. The behavioral paradigm is straightforward and the systematic examination of different neuronal types is careful.

However, a few technical concerns remain to be addressed or clarified and it will also be very helpful to add in some discussion to put their findings in a framework.

Reviewer #3 (Public Review):

This study investigated how reward-associated signals are represented in layer 2/3 neurons of the primary motor cortex. Water-restricted mice were trained to respond to a conditioned auditory stimulus in order to receive a water reward. Behavior analysis showed that mice quickly learned the association between the sound stimulus and the reward as indicated by increased anticipatory lick rate. Using this behavioral paradigm, neuronal activity was monitored throughout the training (7 days). Two-photon calcium imaging was performed separately from four different types of neurons; pyramidal neurons, PV-, VIP-, and SOM-positive interneurons. Tuning of individual neurons to the tone and reward stimuli were analyzed by using Spearman correlation between the trial-averaged fluorescence and the timing of stimulus delivery. Results showed that PV-positive interneuron responses became more reliable to the cue stimulus, whereas VIP-positive interneuron responses became more reliable to the reward stimulus. Some SOM-INs that were not responsive to the tone before training became responsive at day 7 of training. Activity of SOM-INs became more reliable to the reward after learning. The main findings are quite novel and may provide a new insight into the specific roles of interneurons. More representative imaging data and control experiments will make the story even more complete and convincing.

1) Imaging calcium responses from individual types of interneurons are important and challenging. Tracing activity changes from same population of neurons is especially important because it will show how learning shapes the pattern of changes in each neuron. Despite such powerful approaches, activity changes from each neuron were not shown. Calcium transients measured at day 1 were re-sorted at day 7, so it is not clear whether the same neurons responsive to cue or reward stimulus are still responsive to the same stimuli and, if so, how their onset timing is changed. Knowing whether the cue- or reward-sensitive population is the same population or not may lead to a different conclusion, so plotting calcium signals over days without resorting would be important.

In addition, representative calcium images from interneurons were not shown (like Fig. 1A). It seemed that about 80-90 cells of PV-INs, VIP-INs, SOM-INs were observed (Fig. 2D). Showing some representative images from individual cell types would be helpful for readers to better understand the results.

2) Identifying active cells that are above the chance level was good to define a subset of neurons responsive to a period of cue- or reward-stimulus. Quantifying the tuning of each cell's average response during the tone and reward response periods using non-parametric Spearman correlation was also a powerful way to display a subset of neurons with high or low trial-by-trial reliability. Results suggest that there are changes of less reliable neurons to more reliable ones in the case of PV-INs and VIP-INs (Fig. 4 and 5). However, whether these changes are specifically associated with learning is not clear. Running control experiments without water restriction or with random reward presentation independent of cues would be a good comparison. These experiments will help to rule out the possibility of naturally happening learning-independent changes from day 1 to day 7.