Reviewer #1 (Public Review): 

In this manuscript, Mauduit et al described the comprehensive analysis of a panel of cell state-specific cis-regulatory elements (CREs) by massively parallel reporter assay (MPRA) using libraries with different resolution, including H3K27ac ChIP-seq, ATAC-seq and tiling based design. The authors first identified a set of differentially enriched CREs as candidate enhancers by comparing H3K27ac ChIP-seq, ATAC-seq, and gene expression in melanoma cells representing melanocytic (MEL) and mesenchymal-like (MES) states. They next evaluated enhancer activity of selected regions by MPRA in multiple MEL, MES and intermediate state melanoma cells, and uncovered that more than 60% of the selected regions harboring enhancer activity are within ATAC-seq peaks. They further used the deep learning model DeepMEL2 to discover candidate transcription factors (TFs) responsible for cell state-specific enhancer activity, and identified SOX-MITF and AP1 as the major TFs required for enhancer function in MEL and MES cells, respectively. 

Overall, this is an important and well-executed study describing multiple new findings related to the association between enhancer activity as measured by MPRA, chromatin features, and underlying TF binding signals. There are several technical advances in the refinement of MPRA assays that will be valuable to other investigators to adapt the experimental design and data analysis strategies. The results in this study not only confirmed several established concepts in enhancer biology (e.g. ATAC-seq peak is more predictive of enhancer activity than H3K27ac), but also uncovered several new insights into mechanisms controlling cell state-specific enhancer function. The integrative analysis of multiple data types and the inference of the regulatory logics therein are well performed, and the data are of high quality. There are several remaining questions, including how MES and MEL-specific regions were selected, the nomenclature, the correlation with TF binding signals, and discussion/clarification points, that if addressed, will further enhance the strength of conclusions and overall impact of this manuscript.

Reviewer #2 (Public Review): 

Mauduit et al aimed to gain mechanistic insights into the enhancer regulatory circuits underlying different cellular states of melanoma cells. To this aim, they utilized high-throughput enhancer reporter assays in multiple primary melanoma cell lines to identify cell state specific enhancers, and by transcription factor (TF) binding sites discovery, they further explored the regulator mechanism behind the activity and specificity of cell state specific enhancers. Their results indicated that differentially expressed TFs, specifically, AP-1 in the mesenchymal-like cancer cell and SOX and MITF in the melanocytic cell, were controlling cell state specific enhancers. These massive and systematic efforts could be of great interest to understanding pathological mechanisms underlying melanoma progression and provide a rich resource of the design, experimental setup, and interpretation of enhancer reporter assay to the general study of enhancer biology.

Reviewer #3 (Public Review): 

Mauduit et al. apply leading edge analysis with multiple levels of MPRA analysis of melanoma-state specific enhancer regions guided by both ChIP- and ATAC-seq data and melded with AI-type predictions of transcription factor function. Overall, they find that chromatin regions marked by H3K27Ac (putatively enhancers) and open chromatin regions (less predictive) function as enhancers in MPRAs using large (1.2 to 2.9 kb sequences), middle-sized (501 bp), and small tiled 190 bp analyses. Broadly speaking, MEL-state specific enhancers report stronger activity in more MEL-like cell lines, with correspondingly similar findings for MES-state enhancers and MES-like cells, albeit with lower MES activity in melanoma lines with less-defined states. Particularly interesting is the relationship of increased SOX motifs in synthetic contexts to increased activity in MEL lines, and the central role of AP-1 sites in MES-type enhancers. 

Overall, the data are high quality, the conclusions are well-supported, and the study contributes to a better understanding of how to use and interpret approaches studying specific enhancers, or parts of enhancers, or even specific TF binding sites. 

Strengths:

1) The use of multiple sizes of MPRA test sequences including the use of overlapping, tiled short sequences is particularly interesting and valuable. In addition, the testing of multiple MRPA backbones adds to generalizability of the findings, as does the analysis across different cell lines. Harkening back even to gel-shift assays/EMSAs, there is always that tension of focused analysis on a defined, short sequence vs the more technically hard-to-produce/expensive (but potentially more "chromatin-like") larger sequence, and this study will provide a nice guide for how we think about these competing aspects. The example of the proposed ZEB-1 repressor interaction with AP-1 activator activity (page 16, line 427) is particularly interesting, and again highlights the importance of context for multiple TF binding sites. 

2) The study also adds additional context to the use of TF prediction algorithms with the incorporation of the DeepMEL2 tool previously developed by the authors, which indicates such tools can be useful in highlighting regions of interest as candidate gene regulatory elements. 

3) The type of data presented is necessarily very dense, and overall, the data are well-displayed to incorporate the multiple layers of -omics data from gene structure, ChIP-seq, ATAC-Seq, MPRA design, etc., with some minor readability issues. 

Weaknesses:

1) The major weakness of the study is one shared by all approaches using isolated stretches of putative enhancer DNA in a plasmid or reporter vector. Even if one posits some level of chromatinization of the vectors (as noted in the Discussion, page 20), the reporter assay relies on a somewhat artificial context. 

2) While the incorporation of the DeepMEL2 analysis is interesting, it seems hard to fully know the generalizability of such a deep learning tool in the sense that it was trained/developed on melanoma data (melanoma cells). Perhaps this is a strength in its specificity/ability to work in predicting melanoma-specific regulatory networks (or look at how are MES vs MEL states are regulated), but developing or applying an in silico tool with good predictive utility for enhancer function in different cell state contexts is not addressed.

Joint Public Review: 

Using an impressive combination of endothelial-specific knockouts, the investigators provide strong evidence for the following signaling pathway in pulmonary arteries: Activation of Pannexin1 to the release of ATP-activation to P2Y receptors activation of PKC to activation of TRPV4 channels, anchored by caveolin1. 

The study by Daneva et al examines the link between Cav-1, Panx1, P2Y2R and PKC in modulating TRPV4 channel activity. The authors hypothesize that activation of this signaling pathway, and specifically TRPV4, controls the reactivity of pulmonary arteries and contributes to fine-tuning pulmonary arterial pressure. To examine this hypothesis, the authors deployed an impressive number of techniques that include several endothelial-specific knockouts of key members of the signaling pathway, optical and electrical patch clamping, MRI and in situ proximity ligation assay (PLA). The data seem of high quality and for the most part, supportive of the conclusions of the study. The results may have broad implications for regulating pulmonary artery regulation and potential identification of novel targets to treat pulmonary artery dysfunction. 

1) The physiological role of the proposed pathway is unclear. PAP is normally low (8 - 20 mm Hg). Are the authors proposing that this pathway is always engaged to maintain low PAP? If so, then how is Pannexin 1 being tonically activated? This would also imply that there exists a tonic constrictor pathway which Pannexin1-TRPV4 opposes. Does this exist? Or the proposed Panx1-V4 pathway only engaged in the face of pulmonary hypertension. It is hard to envision a dilatory pathway when the system is already at low pressure, i.e., relaxed. 

2) The use of the term, "small, resistance-sized pulmonary arteries" is curious. Pulmonary arteries have low resistance and pressure. What is the basis of using this term? 

3) The major concern is related to conceptual significance. The reviewer appreciates that the work presented here connects Cav-1, Panx1, P2Y2R, PKC and TRPV4 into a signaling axis regulating pulmonary artery reactivity. However, this group has already published similar papers implicating a role for this axis in pulmonary arteries (and a similar axis in systemic arteries), and comparable conclusions have been reached by examining members of the pathway independently. Therefore, it is unclear what new conceptual information is gained, other than the link between all the proteins in the complex. Perhaps the authors could highlight more the major gaps in knowledge and novel aspects of their work. 

4) The major strengths of the study include the use of EC-specific conditional knockouts of Panx1, TRPV4 and P2Y2R that allowed them to focus on the role played by these protein in the endothelium; the state-of-the-art measurement of TRPV4 Ca2+ sparklets and TRPV4 currents; the use of pressure myography to close-the-loop between the ex vivo studies of TRPV4 sparklets and their in vivo measurement of Right ventricular systolic pressure (RVSP as a surrogate for PAP); their measurement of Right heart mass and function to exclude major effects on heart function as a cause of the observed increase in RVSP; and the use of transfected HEK293 cells to examine the role played by caveolin-1 in the signaling pathway. 

5) The only weaknesses are: the lack of experimental evidence in this study that eNOS and NO are the downstream effectors of the reduced pulmonary vascular reactivity and PAP (although this has been demonstrated in other studies) and the lack of evidence for what is responsible for the activation of EC Panx1 and the release of EC ATP that is key to the process. 

6) The results shown support the authors hypothesis and provide new drug and molecular targets to modulate pulmonary vascular resistance, particularly in disease states where endothelial function is compromised.

Reviewer #1 (Public Review): 

This work provides insight into the effects of tetraplegia on the cortical representation of the body in S1. By using fMRI and an attempted finger movement task, the researchers were able to show preserved fine-grained digit maps - even in patients without sensory and motor hand function as well as no spared spinal tissue bridges. The authors also explored whether certain clinical and behavioral determinates may contribute to preserving S1 somatotopy after spinal cord injury. 

Overall I found the manuscript to be well-written, the study to be interesting, and the analysis reasonable. I do, however, think the manuscript would benefit by considering and addressing two main suggestions. 

1) Provide additional context / rationale for some of the methods. Specific examples below: 

a) The rationale behind using the RSA analysis seemed to be predicated on the notion that the signals elicited via a phase-encoded design can only yield information about each voxel's preferred digit and little-to-no information about the degree of digit overlap (see lines 163-166 and 571-575). While this is the case for conventional analyses of these signals, there are more recently developed approaches that are now capable of estimating the degree of somatotopic overlap from phase-encoded data (see: Da Rocha Amaral et al., 2020; Puckett et al., 2020). Although I personally would be interested in seeing one of these types of analyses run on this data, I do not think it is necessary given the RSA data / analysis. Rather, I merely think it is important to add some context so that the reader is not misled into believing that there is no way to estimate this type of information from phase-encoded signals. <br> - Da Rocha Amaral S, Sanchez Panchuelo RM, Francis S (2020) A Data-Driven Multi-scale Technique for fMRI Mapping of the Human Somatosensory Cortex. Brain Topogr 33 (1):22-36. doi:10.1007/s10548-019-00728-6 <br> - Puckett AM, Bollmann S, Junday K, Barth M, Cunnington R (2020) Bayesian population receptive field modeling in human somatosensory cortex. Neuroimage 208:116465. doi:10.1016/j.neuroimage.2019.116465 

b. The rationale for using minimally thresholded (Z>2) data for the Dice overlap analysis as opposed to the threshold used in data visualization (q<0.05) was unclear. Providing the minimally thresholded maps (in Supplementary) would also aid interpretation of the Dice overlap results. 

2) Provide a more thorough discussion - particularly with respect to the possible role of top-down processes (e.g., attention). 

a) The authors discuss a few potential signal sources that may contribute to the maintenance of (and ability to measure) the somatotopic maps; however, the overall interpretation seems a bit "motor efferent heavy". That is, it seems the authors favor an explanation that the activity patterns measured in S1 were elicited by efference copies from the motor system and that occasional corollary discharges or attempted motor movements play a role in their maintenance over time. The authors consider other explanations, noting - for example - the potential role of attention in preserving the somatotopic representations given that attention has been shown to be able to activate S1 hand representations. The mention of this was, however, rather brief - and I believe the issue deserves a bit more of a balanced consideration. 

When the authors consider the possible role of attention in maintaining the somatotopic representations (lines 329-333), they mention that observing others' fingers being touched or attending to others' finger movements may contribute. But there is no mention of attending to one's own fingers (which has been shown to elicit activity as cited). I realize that the patients lack sensorimotor function (and hence may find it difficult to "attend" to their fingers); however, they have all had prior experience with their fingers and therefore might still be able to attend to them (or at least the idea of their digits) such that activity is elicited. For example, it is not clear to me that it would be any more difficult for the patients to be asked to attend to their digits compared to being asked to attempt to move their digits. I would even suggest that attempting to move a digit (regardless of whether you can or not) requires that one attends to the digit before attempting to initiate the movement as well as throughout the attempted motor movement. Because of this, it seems possible that attention-related processes could be playing a role in or even driving the signals measured during the attempted movement task - as well as those involved in the ongoing maintenance of the maps after injury. I don't think this possibility can be dismissed given the data in hand, but perhaps the issue could be addressed by a bit more thorough of a discussion on the process of "attempting to move" a digit (even one that does not move) - and the various top-down processes that might be involved.

Reviewer #2 (Public Review): 

The authors investigate SCI patients and characterize the topographic representation of the hand in sensorimotor cortex when asked to move their hand (which controls could do but patients could not). The authors compare some parameters of topographic map organization and conclude that they do not differ between patients and controls, whereas they find changes in the typicality of the maps that decrease with years since disease onset in patients. Whereas these initial analyses are interesting, they are not clearly related to a mechanistic model of the disorder and the underlying pathophysiology that is expected in the patients. Furthermore, additional analyses on more fine-grained map changes are needed to support the authors' claims. Finally, the major result of changed typicality in the patients is in my view not valid. 

- Concept 1. At present, there is no clear hypotheses about the (expected or hypothesized) mechanistic changes of the sensorimotor maps in the patients. The authors refer to "altered" maps and repeatedly say that "results are mixed" (3 times in the introduction). They do not in detail report which results actually have been reported before, which is a major problem, because those prior results should have motivated the analyses the authors conducted. For instance, two of the cited studies found that in SCI patients, only ONE FINGER shifted towards the malfunctioning area (i.e., the small finger) whereas all other fingers were the same. However, the authors do NOT perform single finger analyses but always average their results ACROSS fingers. This is even true in spite of some patients indeed showing MISSING FINGERS as is clearly evident in the figure, and in spite of the clearly reduced distance of the thumb in the patients as is also visible in another figure. Nothing of this is seen in the results, because the ANOVA and analyses never have the factor of "finger". Instead, the authors always average the analyses across finger. The conclusion that the maps do not differ is therefore not justified at present. This severely reduces any conclusions that an be drawn from the data at present. 

- Concept 2: This also relates to the fact that the most prominent and consistent finding of prior studies was to show changes in map AMPLITUDE in the maps of patients. It is not clear to me how amplitude was measured here, because the text says "average BOLD activity". What should be reported are standard measures of signal amplitude both across the map area and for individual fingers. 

- Concept 3: The authors present a hypothesis on the underlying mechanisms of SCI that does not seem to reflect prior data. The argument is that changes in map alignment relate to maladaptive changes and pain. However, the literature that the authors cite does not support this claim. In fact, Freund 2011 promotes the importance of map amplitude but not alignment, whereas other studies either show no relation of activation to pain, or they even show that map shift relates to LESS pain, i.e., the reverse argument than what the authors say. My impression is that the model that the authors present is mainly a model that is used for phantom pain but not for SCI. Taking this into consideration, the findings the authors present are not surprising anymore, because in fact none of these studies claimed that the affected area should be absent in SCI patients; these papers only say that the other body parts change in location or amplitude, which is something the authors did not measure. It is important to make this clear in the text. 

- Concept 4: There is yet another more general point on the concept and related hypotheses: Why do the authors assume that immediately after SCI the finger map should disappear? This seems to me the more unlikely hypotheses compared to what the data seem to suggest: preservation and detoriation over time. In my view, there is no biological model that would suggest that a finger map suddenly disappears after input loss. How should this deterioration be mediated? By cellular loss? As already stated above, the finding is therefore much less surprising as the authors argue. 

- Methods & Results. The authors refer to an analyses that they call "typicality" where they say that they assess how "typical" a finger map is. Given this is not a standard measure, I was wondering how the authors decided what a "typical" finger map is. In fact, there are a few papers published on this issue where the average location of each finger in a large number of subjects is detailed. Rather than referring to this literature, the authors use another dataset from another study of themselves that was conduced on n=8 individuals and using 7T MRI (note that in the present study, 3T MRI was used) to define what "typical" is. This approach is not valid. First, this "typical" dataset is not validated for being typical (i.e., it is not compared with standard atlases on hand and finger location), second, it was assessed using a different MRI field strength, third, it was too little subjects to say that this should be a typical dataset, forth, the group differed from the patients in terms of age and gender (i.e., non-matched group), and fifth, the authors even say that the design was different ("was defined similarly", i.e., not the same). This approach is therefore in my view not valid, particularly given the authors measured age- and gender-matched controls that should be used to compare the maps with the patients. This is a critical point because changes in typicality is the main result of the paper. 

- Methods & Results: The authors make a few unproven claims, such as saying "generally, the position, order of finger preference, and extent of the hand maps were qualitatively similar between patients and control". There are no data to support these claims. 

- Methods & Results: The authors argue that the map architecture is topographic as soon as the dissimilarity between two different fingers is above 0. First, what I am really wondering about is why the authors do not provide the exact dissimilarity values in the text but only give the stats for the difference to 0 (t-value, p-value, Bayes factor). Were the dissimilarity values perhaps very low? The values should be reported. Also, when this argument that maps are topographic as long as the value of two different fingers is above 0 should hold, then the authors have to show that the value for mapping the SAME finger is indeed 0. Otherwise, this argument is not convincing. 

- Discussion. The authors argue that spared midsagittal spinal tissue bridges are not necessary because they were not predictive of hand representation typicality. First, the measure of typicality is questionable and should not be used to make general claims about the importance of structural differences. Second, given there were only n=14 patients included, one may question generally whether predictive modelling can be done with these data. This statement should therefore be removed. 

- Discussion. The authors say that hand representation is "preserved" in SCI patients. Perhaps it is better to be precise and to say that they active during movement planning.

Reviewer #3 (Public Review): 

The demonstration that cortex associated with an amputated limb can be activated by other body parts after amputation has been interpreted as evidence that the deafferented cortex "reorganizes" and assumes a new function. However, other studies suggest that the somatotopic organization of somatosensory cortex in amputees is relatively spared, even when probed long after amputation. One possibility is that the stability is due to residual peripheral input. In this study, Kikkert et al. examine the somatotopic organization of somatosensory cortex in patients whose spinal cord injury has led to tetraplegia. They find that the somatotopic organization of the hand representation of somatosensory cortex is relatively spared in these patients. Surprisingly, the amount of spared sensorimotor function is a poor predictor of the stability of the patients' hand somatotopy. Nonethless, the hand representation deteriorates over decades after the injury. These findings contribute to a developing story on how sensory representations are formed and maintained and provide a counterpoint to extreme interpretations of the "reorganization" hypothesis mentioned above. Furthermore, the stability of body maps in somatosensory cortex after spinal cord injury has implications for the development of brain-machine interfaces. 

I have only minor comments: 

1) Given the controversy in the field, the use of the phrase "take over the deprived territory" (line 45) is muddled. Perhaps a more nuanced exposition of this phenomenon is in order? 

2) The statement that "results are mixed" regarding intracortical microstimulation of S1 is dubious. In only one case has the hand representation been mislocalized, out of many cases (several at CalTech, 3 at the University of Pittsburgh, one at Case Western, one at Hopkins/APL, and one at UChicago). Perhaps rephrase to "with one exception?" 

3) The phrase "tetraplegic sinal cord injury" seems awkward. 

4) The stability of the representation is attributed to efference copy from M1. While this is a fine speculation, somatosensory cortex is part of a circuit and is interconnected with many other brain areas, M1 being one. Perhaps the stability is maintained due to the position of somatosensory cortex within this circuit, and not solely by its relationship with M1? There seems to be an overemphasis of this hypothesis at the exclusion of others.

Reviewer #1 (Public Review): 

In this manuscript Hill et al, analyze immune responses to vaccination of adults with the seasonal influenza vaccine. They perform a detailed analysis of the hemagglutinin-specific binding antibody responses against several different strains of influenza, and antigen-specific CD4+ T cells/T follicular cells, and cytokines in the plasma. Their analysis reveals that: (i) tetramer positive, HA-specific T follicular cells induced 7 days post vaccination correlate with the binding Ab response measured 42 days later; (ii) the HA-specific T fh have a diverse TCR repertoire; (iii) Impaired differentiation of HA-specific T fh in the elderly; and (iv) identification of an "inflammatory" gene signature within T fh in the elderly, which is associated with the impaired development of HA-specific Tfh. 

The paper addresses a topic of considerable interest in the fields of human immunology and vaccinology. In general the experiments appear well performed, and support the conclusions. However, the following points should be addressed to enhance the clarity of the paper, and add support to the key conclusions drawn. 

1) Abstract: "(cTfh) cells are the best predictor of high titre antibody responses.." <br> Since the authors have not done any blind prediction using machine learning tools with independent cohort, the sentence should be rephrased thus: "cTfh) cells are were associated with high titre antibody responses." 

2) Figure 1A: Please indicate the age range of the subjects. 

3) Almost all the data in the paper shows binding Ab titers. Yet, typically HAI titers of MN titers are used to assess Ab responses to influenza. Fig 1C shows HAI titers against the H1N1 Cal 09 strain. Can the authors show HAI titers for Cal 09 and the other A and B strains contained in the 2 vaccine cohorts? Do such HAI titers correlate with the tetramer positive cells, similar to the correlations show in Fig 2e. 

4) Fig 2d to i: what % of all bulk activated Tfh at day 7 are tetramer positive? The tetramer positive T cells constitute roughly 0.094% of all CD4 T cells (Fig 2d), of which 1/3rd are CXCR5+, PD1+ (i.e. ~0.03% of CD4 T cells). What fraction of all activated Tfh is this subset of tetramer positive cells? Presumably, there will also be Tfh generated against other viral proteins in the vaccine, and these will constitute a significant fraction of all activated Tfh.

Reviewer #2 (Public Review): 

Hill and colleagues present a comprehensive dataset describing the recall and expansion of HA-specific cTFH cells following influenza immunisation in two cohorts. Using class II tetramers, IgG titres against a large panel of HA antigens, and quantification of plasma cytokines, they find that activated and HA-specific cTFH cells were a strong predictor of the IgG response against the vaccine after 6 weeks. Using RNAseq and TCR clonotype analysis, they find that, in 10/15 individuals, the HA-specific cTFH response at day 7 post-vaccination is recalled from the available CD4 T cell memory pool present prior to vaccination. Post-vaccination HA-specific cTFH cells exhibited a transcriptional profile consistent with lymph node-derived GC TFH, as well as evidence of downregulation of IL-2 signaling pathways relative to pre-vaccine CD4 memory cells. 

The authors then apply these findings to a comparison of vaccine immunogenicity between younger (18-36) and older (>65) adults. As expected, they found lower levels of vaccine-specific IgG responses among the older cohort. Analysis of HA-specific T cell responses indicated that tet+ cTFH fail to properly develop in the older cohort following vaccination. Further analysis suggests that development of HA-specific cTFH in older individuals is not caused by a lack of TCR diversity, but is associated with higher expression of inflammation-associated transcripts in tet+ cTFH. 

Overall this is an impressive study that provides clarity around the recall of HA-specific CD4 T cell memory, and the burst of HA-specific cTFH cells observed 7 days post-vaccination. The association between defective cTFH recall and lower IgG titres post-vaccination in older individuals provides new targets for improving influenza vaccine efficacy in this age group. However, as currently presented, the model of impaired cTFH differentiation in the older cohort and the link to inflammation is somewhat unclear. There are several issues that could be clarified to improve the manuscript in its current form: 

1) It is somewhat unclear the extent to which the reduction in HA-specific cTFH in the older cohort is also related to an overall reduction in T cell expansion - cohort 1 shows a significant reduction in total tet+ CD4 T cells post-vaccination as well as in the cTFH compartment, and while this difference may not reach statistical significance, a similar trend is shown for cohort 2. 

2) Transcriptomic analysis indicates that HA-specific cTFH in the older cohort show impaired downregulation of inflammation, TNF and IL-2-related signaling pathways. The authors therefore conclude that excess inflammation can limit the response to vaccination. In its current presentation, the data does not necessarily support this conclusion. While it is clear that downregulation of TNF and IL-2 signalling pathways occur during cTFH/TFH differentiation, there is no evidence presented to support the idea that (a) vaccination results in increased pro-inflammatory cytokine production in lymphoid organs in older individuals or that (b) these pro-inflammatory cytokines actively promote CXCR5-, rather than cTFH, differentiation of existing memory T cells.

Reviewer #1 (Public Review): 

This paper describes the structure of WIPI2d in association with a region of ATG16L1 that interacts with WIPI. Previous work has demonstrated that WIPI proteins bind to PI3P on the surface of phagophores to initiate autophagy. WIPI also associates with ATG16 orthologs to template assembly of the phagophore expansion machinery. The authors map the interface, and also perform biochemical and cell biological experiments that support the molecular structure. 

Analysis of mutations within the interaction interface reveal the residues required for interaction, confirming the structural data. This information will be very important in efforts to biochemically reconstitute the initial stages of autophagy. This work also reveals the complications of examining mutations in complex interaction surfaces in cells, particularly in cases where multivalent interactions drive the process. The results also explain why there are multiple WIPI proteins in the human genome, reflecting distinct mechanisms of recruitment of ATG16 and ATG2 orthologs to the appropriate membrane.

Reviewer #2 (Public Review): 

This manuscript from Hurley and colleagues uses x-ray crystallography to determine the structure of WIPI-2d in complex with a peptide representing the WIPI-2 binding domain of ATG16L1. WIPI-2 proteins play a crucial role early in autophagosome biogenesis by binding phosphoinositides and then recruiting the ATG16L1 complex to the growing autophagosome membrane. WIPI-2 is a member of the mammalian PROPPIN family of WD-repeat proteins which includes WIPI-1, WIPI-2, WIPI-3 and WIPI-4, but until now, only the WIPI-3 structure had been determined. As expected, the authors reveal that WIPI-2d adopts the classic 7-bladed WD-repeat structure. Further they show that the ATG16L1 helix previously described as a WIPI-2 binding site (Dooley/Tooze 2014), interacts specifically between the 2nd and 3rd blade of the WD-repeat structure through combinations of hydrophobic and charged amino acids on the amphipathic helix. From this result they were able to map the crucial amino acids involved in protein-protein interaction, many of which had previously been identified through elegant biochemistry/mutational analysis experiments (in Dooley..Tooze, 2014), further confirming the validity of the authors' structure. 

They confirm that the crystallographically-determined amino acids forming the peptide binding site are needed for peptide recruitment in an affinity pull-down experiment and then further show that these same amino acids are needed to support recruitment of the full ATG16L1 complex to membranes in an elegant in vitro reconstitution assay of LC3-lipid attachment. Intriguingly, they also observe that recruitment of WIPI2 itself to membranes in the presence of ATG16L1, is reduced when mutations are introduced into the ATG16L1 binding site. It is not clear from their experiments whether this reduction in membrane binding reflects interactions with ATG16L1 or is a general impairment to WIPI2-phosphoinositide interaction. Experiments testing the membrane binding of WIPI2 alone following mutagenesis would help tease out these differences. 

The authors also examine whether mutation of the ATG16L1 binding site they describe impacts autophagy events in cells. To do this, they use siRNA knockdown of WIPI2 and then re-express either wildtype or mutant forms of WIPI2 and follow both WiPI2 puncta formation (suggesting WIPI2 recruitment to new membranes) and LC3 puncta formation (indicative of autophagosome formation), very similar to the original assay design of Dooley/Tooze. The authors observe a modest reduction on LC3 numbers for essentially all mutants tested, thus the direction of the experiment is consistent with their hypothesis. However, the very mild impact here of mutating these presumably key residues is surprising given the all-or-nothing behavior of some mutants in vitro. This assay is essentially a test for dominant-negative behavior and will always be sensitive to the level of knockdown achieved as the expressed protein is competing against the wildtype reserves. The authors do not show what the impact of the knockdown alone is on their system and one possibility is that that the remaining WiPI2 observable by western blot is largely sufficient to support autophagy. The authors also note that the impacts of mutation in cells does not follow the same rank order of impacts in vitro. A more dramatic starting knockdown or a full knockout experiment followed by rescue with wildtype or mutants would likely provide a clearer signal-to-noise in interpreting the significance of these amino acids in supporting ATG16L1 activities in cells.

Reviewer #3 (Public Review): 

In this manuscript, Strong et al. sought to address how WIPI2 recruits the LC3 E3 ligase scaffold component ATG16L1 to the nascent autophagosomal membrane. For this purpose, the authors determined a high-resolution structure of a consensus WIPI2 bound to the WIPI2-interacting region (W2IR) of ATG16L1 using x-ray crystallography. Their model WIPI2 adopted a 7-bladed beta propeller to which the W2IR alpha helix bound in a hydrophobic groove situated between two blades (#2 and #3). Importantly, mutational analysis of W2IR interface residues in WIPI2 confirmed their critical roles in GST pulldown and LC3 lipidation reconstitution experiments in vitro as well as LC3 and WIPI puncta formation assays in starved cells. Rounding off, the authors compared their WIPI2-ATG16L1 W2IR structure to a recently solved structure of WIPI3, which was shown to specifically bind to ATG2A and generated structural models for the two remaining WIPI family members, namely WIPI1 and WIPI4. Overall, Strong and colleagues provide compelling mechanistic insights into WIPI2's role in mediating E3 recruitment and LC3 lipidation. Moreover, the authors' structural data of WIPI2 will help to rationalize functions of the other WIPI proteins.

Reviewer #1 (Public Review): 

A neural network model of a subregion of the hippocampus is trained based on simulated experiences and examined in the context of reactivation during rest. The authors perform several manipulations in order to pinpoint the mechanistic sources of important features of population activity in this brain region, namely sharp wave deflections, ripple oscillations, and replay. This study enhances our understanding of area CA3 and hippocampal replay by pinpointing the anatomical and functional features of CA3 circuitry which contribute to SWR production and replay from experience. 

This study overcomes several significant barriers in the circuit modeling of SWR and replay in the hippocampus. First, the synaptic weights of the network are primarily learned from location inputs drawn from environment trajectories. This stands in contrast to many models of hippocampal activity which rely on synaptic weight matrices established through explicit calculation (e.g. based on the place cell covariance structure) or supervised learning. Second, the synaptic weights are learned from a symmetric STDP rule which has been established experimentally as opposed to an asymmetric STDP rule which has been implemented previously in order to generate sequential place reactivations. In the model presented here, it is shown (through an ablation simulation) that an adaptation mechanism causes sequential place reactivations to emerge and facilitates bidirectional replay. More broadly, previous network models trained online from simulated experiences lack much of the biological detail present here. 

In simulation it was noted that the network spontaneously shifted into high activity modes (generating sharp waves) accompanied by replay and emergent oscillatory activity in the ripple band. The authors were able to tease apart these phenomena in a manner consistent with empirical results. They showed that sharp waves and replay emerged from increased levels of activity in the the pyramidal cell population (containing place cells) spontaneously which then drive ripple oscillations in another subpopulation of basket cells supported by recurrent inhibition. 

Further manipulations of the network are performed in order to relate the learned network structure to functional properties (i.e. SWRs, replay). By shuffling the weight matrix, it is confirmed that the specific pattern of strong synaptic connections is necessary for replay (as opposed to the overall weight statistics). However, this seems unsurprising given the topological nature of the implicit cognitive map and its place cell representation. 

Although some of the results presented here appear in other models, it seems that this study represents an integrative, state-of-the-art account of area CA3 circuitry and the principles employed may be embedded in broader circuit models covering other areas of the hippocampus. Other neuroscientists may be facilitated in accomplishing this by the code repository associated to the manuscript. The authors acknowledge and discuss in detail some significant limitations of their results, most prominently the apparently unending nature of the emergent replay (i.e. replay terminates only on encountering the end of the track). Otherwise, it could be noted that the results are demonstrated exclusively in a linear track environment. It would be interesting to establish whether the model is robust to learning from trajectories in an 2d open box environment for example. Overall, I find the conclusions of this paper to be supported by the results. Though this work provides a comprehensive account of several previous experimental results, it lacks specific experimental predictions thus limiting the significance of this manuscript for experimentalists who may be in a position to test any such novel hypotheses.

Reviewer #2 (Public Review): 

The theoretical study by Ecker et al. uses advanced computational methods to show that the statistics of the connectivity between excitatory principal cells (PCs) determines the content (forward and reverse replay) and the structure (single cell and network activity levels) of replay events in the CA3 region of the hippocampus. The model assumes that a symmetric plasticity rule generates the connectivity between the principal cells in an initial exploratory learning phase. Before the start of this phase, excitatory neurons are coupled randomly and sparsely with 10% connection probability. In a second phase, akin to resting and consummatory behaviours, the model then generates both forward and reverse replay events with biologically plausible single cell and network activity properties. Importantly, this property is not a built-in feature of the model but emerges dynamically. The study uses single-compartment adaptive exponential integrate-and-fire models whose parameters are determined offline (that is, in a single-cell simulation instead of a network simulation) using an evolutionary algorithm (inspyred). In a second step, the synaptic weights, but not the connection probabilities, of the network are determined with another, separate evolutionary alogrithm (BluePyOpt) to yield dynamical properties typically observed during sharp wave/ripple events, such as realistic principal cell firing rates accompanied by significant ripple frequency oscillations. The authors show that the simultaneous presence of forward and reverse replay requires a symmetric plasticity kernel, structured and strong recurrent excitatory connectivity and a spike-triggered adaptation current for the principal cells. The model also supports learning two distinct environments and is robust to scaling the recurrent excitatory weights. It is also shown that ripples together with replay only occurs for sufficiently strong recurrent connections within the inhibitory basket cell (PVBC) population and that with such absent inhibitory connectivity, increasing the strength of the recurrent excitatory weights does not reinstate ripple oscillations, which supports the previously suggested FINO hypothesis for ripple oscillation generation. The manuscript finishes with a thoughtful and extensive discussion about the limitations and shortcomings of the model and its relationship to related studies of hippocampal network oscillations. For example, the authors discuss their main modelling assumption assuming 10% recurrent excitatory connectivity in light of recent experimental results (Guzmán et al. 2016) which suggests that 10% connectivity might actually be too large for CA3. 

The paper uses advanced methodology and the results are well presented. However, certain weaknesses remain, which are listed below. 

1) The main result lacks a mechanistic explanation. My understanding of the main result is that CA3 principal cell firing needs to be tightly controlled by plasticity (so that there are chains of strong bidirectional coupling guiding initial random activity, making it very unlikely that a cell fires outside of its place field) and strong adaptation (so that the participation of a cell in a chain is terminated after a short time, suppressing burst firing) for the model to generate replay events with physiological properties. Is it not possible at all to generate ripples without structured excitatory connectivity in the model? Related to this, if the principal cells fire only once or twice during the whole replay event (as is observed in vivo and also in the present model), how can a spike-triggered adapation current (that is required for replay, as the authors show) exert its influence on the dynamics of the network? 

2) Some of the modelling choices seem to be ad hoc and not well motivated. 

2.1) It is not clear whether the symmetric STDP rule proposed by Mishra et al. is more biologically plausible for CA3 than other rules and/or whether there are other rules at work at all. Currently, it reads as if this is the only STDP rule that is plausible for CA3 recurrent excitatory connections. 

Also, the time constant of the plasticity rule was fixed. It would be good to study the impact of different time constants if this is biologically realistic. The reason for this suggestion is that it might allow for a more mechanistic understanding of the model behavior. Is it possible that there is a certain spatial range for stronger-than-baseline connectivity that is required for there to be replay and if so, does it depend on the time constant? 

2.2) Plasticity between principal cells is switched off after the initial exploration phase and the networks are static thereafter. How realistic is this? It might be possible that spike-triggered synaptic plasticity during a ripple event interferes with the bidirectional coupling established during the initial learning phase, thus rendering replay unstable and/or terminating it before the end of the track is reached. 

3) The authors conclude that the FINO mechanism is at the heart of ripple generation in CA3. I found it confusing that apparently, no ripple oscillations are present in the population rate of the principal cells, but only in the calculated LFP, although the Figure Supplement for Fig. 3 seems to suggest that ripple oscillations are present also in the population rate of the principal cells. Maybe calculating the ripple frequency directly from the population rate of the principal cells, but not from the LFP, would change this conclusion (presented in Fig. 8), because the former still exhibits ripple oscillations, whereas the latter doesn't? 

4) Is it possible to determine which initial conditions cause forward and which reverse replay? Would it be possible to compare this to different behavioural states exhibited by an awake animal traversing a track and then resting? The authors mention in the Introduction that forward replay is associated with memory recall, whereas reverse replay is associated with reward-based learning. How is this reflected in the statistics of these two types of replay events in the resting phase of the model? Is there an expectation that both events should occur equally often or are there any conditions that bias the direction of the replay in the model? This is related to the unstructured input that principal cells receive in the model, so maybe there is an interplay of these two sources of excitation (feedforward via the mossy fibers and recurrent via the learned connectivity) which determines the direction and frequency of replay events. I feel that addressing at least some of these points would allow the authors to gauge the realism of their replay model more finely.

Reviewer #3 (Public Review): 

Computational neural network models are valuable tools for exploring mechanisms of brain dynamics, including generation of different frequencies of oscillation in neural activity and how synaptic plasticity can result in the storage and recall of patterns of information. These aspects are combined here to show that the connectivity patterns that arise when storing activity patterns through synaptic spike-timing-dependent plasticity (STDP) can actually result in the emergence of sharp waves and associated high-frequency ripples (SWR), as seen in the CA3 region of the mammalian hippocampus. This is a result that is of wide interest to neuroscience research. 

Though the model used is rather simple, it contains sufficient detail to be directly constrained by experimental data in key aspects, such as cell spiking responses and the STDP learning rule. The simplicity also allows the operation of the network to be examined and understood. The results clearly show that symmetry in the STDP rule is needed for both forward and backward replay of pattern sequences to emerge, which makes logical sense. Sequence replay depends on spike frequency adaptation in the pyramidal cells. The model confirms the results of other experimental and modelling studies that strong recurrent inhibition between inhibitory interneurons can underpin ripple oscillations. 

Appropriate analysis methods are used to identify sequence replay and the power of different oscillation frequencies in the neural activity. 

The authors are careful to discuss the limitations of their model and the mismatches with experimental data. Such mismatches do not invalidate their main conclusions, but do highlight that there are potentially other neural mechanisms that are absent from the model that contribute to shaping the spatio-temporal characteristics of sequence replay and SWRs.

Reviewer #1 (Public Review): 

Widmer and Keller investigate the role of NMDAR-dependent plasticity in mouse V1 in the development of proper motor-sensory mismatch detection. To control motor-visual experience, the authors reared mice in the dark and then controlled their only motor-visual experience in a VR corridor. Using in vivo 2p imaging of L2/3 excitatory neurons, the authors find that unilateral inactivation of V1 NMDARs led to a decrease in mismatch signals when optic flow is halted; a decrease in visually evoked responses; and abnormal integration of sensory and motor signals at locomotion onset. These findings required NMDAR inactivation during the mouse's first motor-visual experiences. NMDAR-inactivated mice also showed an inability to adjust their behavior in a motor-visual task. Finally, the authors show that inhibiting CAMKII function in excitatory neurons partially recapitulates the effect of V1-wide NMDAR inactivation, while inhibiting CaMKII function in SST neurons has some opposite effects. The authors conclude by suggesting that plasticity of excitatory synapses onto SST neurons is important for learning motor-visual expectations. 

The experiments are well thought out, the paper is well presented, and this is an important contribution to our understanding of where in the brain plasticity happens as animals learn how their actions influence the world. There are a couple of small caveats in interpreting these results, many of which the authors go through in detail in their Discussion section. The authors focus their neuronal readout on the activity of excitatory neurons in L2/3. While there is precedent for focusing on these cells, the perturbations that the authors induce in their experiments are not limited to L2/3 cells. Therefore it remains to be tested at exactly which synaptic sites plasticity matters. Second, the two plasticity manipulations don't lead to exactly the same phenomenology in L2/3 excitatory neurons, leading to some residual questions about exactly how plasticity in V1 contributes to motor-sensory learning. Third, the magnitude of mismatch signals in their various control mice varies quite a bit, meaning that what looks like a big effect in one experimental group might actually be an artifact of the control condition.

Reviewer #2 (Public Review): 

Authors and others previously showed that coupling of motor output and visual sensory feedback is necessary to establish proper visuomotor integration in visual cortex. Here, authors investigated the roles of NMDA receptor and related CaMK2 activity in primary visual cortex to establish visuomotor integration during the first visual experience, and visuomotor skill learning later in life. They found that conditional knock out of NMDA receptor within V1 from juvenile period reared in dark from birth (but not in adult knock out) show diminished V1 mismatch onset response, suggesting a key role for visuomotor integration. The same manipulations also impaired visuomotor skill learning later in life. To gain insight into the cellular/circuit mechanism, authors also employed photoactivable inhibitor of CaM2K in cell-type specific manner, and found changes in the correlation of V1 activity and visual flow when manipulating CaMK2 activity in SST interneurons consistent with the known role of this cell type in regulating visuomotor integration. Overall, these findings support a key role for NMDA receptor and associated CaMK2 signaling in V1 neurons (especially SST interneurons) in shaping visuomotor integration during the first visual experience. These findings are significant to the field as it provides the first molecular mechanism supporting the role of visuomotor integration in V1. Use of photoactivatable peptide to inhibit CaMK2 in cell type and temporally selective manner is especially elegant. 

While the questions and findings are overall interesting, there are some issues with 1) the interpretation of the data related to plasticity, and 2) insufficient integration of data analysis and interpretation between Grin KO (Fig1) and CaMK2 manipulations (Fig5). 

Regarding the data interpretation, authors interpreted that it is the "plasticity" which is necessary for visuomotor integration through the study of conditional manipulations of NMDA receptor or CaMK2 activity in mouse V1. However, data do not convincingly support this claim. Given that not only the response to mismatch onset response, but also grating onset response was reduced by juvenile V1 NMDAR knock-out, it is totally possible that diminished mismatch response may not directly require plasticity mechanism for visuomotor integration (bottom up/top-down integration) per se, but can be secondary to diminished visual response due to plasticity deficits of bottom-up visual input connectivity only. This possibility was not considered well in the current manuscript. Similar concerns on interpretations apply to CaMK2 manipulations which also led to changes in both mismatch onset response and grating onset response. 

There is also a challenge in conceptually putting together the findings from Grin and CaMK2 manipulations. While manipulations of Grin and CaMK2 led to consistent changes at the level of activity correlation with visual flow, these manipulations led to very different V1 responses especially for grating onset response (Fig1F vs Fig5DG). It would be helpful if authors can attempt to better explain why this discrepancy does not contradict to the authors' model. It would be informative to consider the possibility that manipulation in single cell type (Fig5) vs multiple cell types (Fig1 uses non-cell type selective promotor for cre expression) leads to a different outcome. This possibility can be experimentally supported and/or discussed to enhance the integration of the study. 

Other weaknesses, which are minor, include 1) lack of quantification of Grin KO, 2) lack of control experimental groups for better interpretation, 3) lack of consideration of dark rearing to delay the maturation program, and 4) insufficient descriptions of animal number for each figure. Some of these issues can be addressed with more description or experimentally. Otherwise, unaddressed issues need to be discussed thoroughly.

Reviewer #1 (Public Review): 

1) Counter to what one would expect, the effects of introducing the aquaporins seem to be more on Vmax than on Km for CO2, and it is unclear why that would be. Does this not rule out the proposed mechanism? 

2) A more serious issue is that no details were given about how the transformant lines were selected. This has the potential to negate the claims of the paper. The Materials and Methods are so terse as to be unhelpful on this important point. Out of many, only three "positives" and one azygous transformant line were investigated. Given that past work of this type from other labs has fallen into the trap of "selecting for the desired results", it would be crucial to include both descriptions of how these were chosen as well as more detailed statistical comparisons with randomly-chosen transformants, non-transformed lines to show that the phenotypic effects can be reasonably attributed to the expression of aquaporins.

Reviewer #2 (Public Review): 

Strengths: 

Past work looking at mesophyll conductance has approached the problem mainly using the same set of tools (transgenic transformation combined with a fluorescence or isotope-based estimate of mesophyll conductance) in C3 plants. This study adds confirmatory work with yeast to study the functional role of the transformed PIP in facilitating CO2 diffusion in a C4 species. They also examined the C4 transgenics using oxygen isotope approaches that have been recently developed to tackle the difficulty of measuring mesophyll conductance in a C4 system. This isotope approach is validated by comparing it to an independent method based on gas exchange alone. 

Another strength in this work is that the lines selected for further analysis were selected based on the strength of the transgene expression and not a "self-fulfilling" phenotype such as improved photochemical efficiency. This is important since in many other studies investigating improved photosynthetic capacity, large numbers of transgenic plants are screened using approaches that would preferentially select for improved photosynthesis that could have arisen through a pleiotropic effect such as where the insert occurred. By selecting based on transgene expression, then phenotyping photosynthesis, the group more effectively ties the transgene to the phenotype. Their point could be improved if they better highlighted the relationship between the protein abundance of the transgene and the strength of a phenotype. 

Weaknesses 

While there is compelling support for the role of aquaporins in mesophyll conductance in C3 plants, there are also many studies where no difference is seen. The paper could be improved by discussing this work as well and attempting to rectify it with their study's results. 

It is not clear why the transgenics have higher photosynthetic rates under elevated CO2. It would be expected that if mesophyll conductance was the only thing altered, that the benefit to the photosynthesis would be most obvious in the initial portion of the A-Ci curve and then decrease as CO2 becomes saturating. This could be discussed more in depth if there is a ready explanation.

Reviewer #3 (Public Review): 

The potential of using aquaporins to engineer and improved mesophyll conductance and, consequently, an improved photosynthetic efficiency has long been mooted. In C3 plants there are data to support this approach (although nt without some controversy). In this paper the authors set out to test the hypothess that engineering aquaporin expression in a plant with C4 photosynthesis might also function to improve mesophyll conductance. 

Using Setaria as the platform for engineering, they first performed a very careful analysis of genes encoding a type of aquaporinm the plasma membrane intrinsiv proteins (PIPs). This involved functional analaysis of CO2 and water permeability in yeast engineered to express the various PIP genes from setaria. This provided convincing evidence that at least one of the PIPs functioned to enhance CO2 flux. This gene was then used to generate transgenic setaria plants in which expression of the selected PIPs gene was targetted towards the leaf mespohyll. This was achieved, with subsequent analaysis of leaf physiology supporting the conclusion that the expresison of the PIPs gene did lead to an enhancement of carbon assimilation rate and that this was at least partially related to an increased measurement of mesophyll conductance. 

The results extend the work performed on C3 plants by showing that in C4 plants mesophyll conductance can be enginnered by increased aquaporin expression, with the increased rate of carbon assimilation and, thus, the potential for improving crop yields in future work.

Reviewer #1 (Public Review): 

This is a very interesting manuscript that attempts to provide evidence of a case of evolutionary interaction (i..e. natural selection) between two human pharmacogenes: ADCY9 and CETP, suggesting also interaction with sex and a case for pleiotropy. This is likely one of the few examples (or maybe the only one) of evolutionary interaction between two genes in the field of human evolutionary genetics. The authors provide a set of genomics-based evidences to support their case of natural selection that include: (i) a replication of population genetics results in another Peruvian cohort, (ii) evidence of epistatic effect from RNAseq data on public datasets and on ad-hoc experiments, (iii) genotype/phenotypes associations in the UK Biobank. While the use of data from different sources (in opposition to a trans-omic approach in a Peruvian population) may be a weakness of the paper, it is important to recognize that performing a large set of experiments in the Peruvian cohort using the required sample sizes may be logistically prohibitive. Therefore, the author's approach is acceptable. 

Specifically, it is interesting that some of the phenotypes found in the UK Biobank is related with adaptation to high altitude (FVC) 

The only result that is difficult to explain of the difference in the level of long-range LD observed between males and females from Peru, both for the discovery and the replication datasets. The authors should elaborate more quantitatively on the plausibility of this finding.

Reviewer #2 (Public Review): 

This work attempts to identify possible functional links between two pharmacogenetic relevant loci, ADCY9 and CETP, by using signals of positive selection as a starting point. Starting with the 1000 Genomes (1kG) dataset, the authors identify complementary signals of positive selection in the 1kG Peruvian cohort (PEL) using iHS and PBS analyses, specifically in a LD block in ADCY9. They use these results then to support investigating possible coevolution between ADCY9 and CETP in the form of long-range linkage disequilibrium (LRLD), a clever way to identify possible cosegregation between variants that should otherwise not be present. This analysis is particularly apt since two regions have already been identified, one of which is now suggested to have experienced a rapid increase in allele frequency. The authors not only find evidence of LRLD between SNPs in ADCY9 and CETP, but they also identify these results occur in a sex-specific manner. The authors then begin investigating possible functional connections between these two loci, specifically in the form of gene expression analyses. Using both human cell ADCY9 knockdown lines and GEUVADIS/GTEx data, the authors identify that ADCY9 impacts CETP expression, both broadly and through a specific interaction between rs1967309 and rs158477. And lastly, the authors further investigate potential interactions between ADCY9 and CETP via epistatic association analyses using UK BioBank and GTEx data. Encouragingly, among a handful of relevant phenotypes, they find marginally significant, sex-specific interaction effects with CAD and Lp(a), and in both cases the direction of effects match those seen in their LRLD analyses -- eg rs1967309-AA + rs158477-GG containing a protective effect in males. 

Overall, the authors make a strong case that there is coevolution occurring between ADCY9 and CETP. That they are able to also continuously replicate some of their findings using an independent dataset, LIMAA, also strengthens their results. The authors do acknowledge that their sample sizes may be limiting their power at times, but they point out that finding multiple, concordant marginally significant results may represent an unlikely outcome. However, currently these are presented as just observations and not as a formal, integrated test. Therefore we cannot adjudicate whether these concordant, marginally significant results are occurring more than we would expect by chance. 

It is worth noting that by beginning with two loci, the authors are able to make use of 'pairwise' approaches such as LRLD and epistatic analyses. Normally, these types of tests come with large multiple testing burdens due to the rapid increase in test combinations. In fact, a priori one would potentially not expect such analyses to perform well with the limited sample sizes. However, by beginning with a hypothesis that focused on two loci, the authors are able to overcome this normal statistical challenge. 

It is also worth noting that these results are only possible due to the inclusion of diverse datasets. The initial selection signals would not have been identified if the datasets only contained individuals of European ancestry. Additionally, even with the limited sample sizes, the authors are still able to identify statistically significant results. Therefore this work is another example of what can be gained when incorporating more diverse human genomics datasets. 

In terms of identifying a functional or molecular link between ADCY9 and CETP, the authors have begun the work of finding some possible connections between these two loci. However, this goal was not completely met with the current work. There is a clear effect of ADCY9 on CETP gene expression, though whether this is ultimately a direct effect or indirect is unclear. And while the epistatic analyses using phenotypes such as clinical outcomes and biomarkers are encouraging, and concordant in terms of direction of effects, they still do not elucidate a mechanism by which the variants of interest in ADCY9 and CETP are functionally interacting. The finding that sex plays a role in this interaction, and may be important to the mechanistic link as well, is an important result though.

Reviewer #3 (Public Review): 

The authors have analysed genomic data from populations of South Americans to assess the genetic and functional link between ADCY9 and CETP. These two genes, which are both on chromosome 16 separated by over 50Mbp, show weak but significant long range linkage disequilibrium in some subpopulations (ie from Peru). The genetic link between these genes (and SNPs), despite being weak, is suggestive of positive selection for haplotypes that appear with higher frequency in the population compared to populations from Africa, Asia and Europe. What is surprising is the sex-specific linkage, with most if not all of the association signal being driven by male samples. The explanation behind this remains open, however, with multiple explanations from population dynamics and drift, to potential functional benefits selecting this association. 

Strengths:

The work carefully assesses this linkage through robust statistical frameworks. Despite weak effects and low sample sizes, there is a replicable signal in two other populations. The data is also easily accessible and I appreciate the author's documentation of their work. 

Weaknesses:

The effects are still weak, and may still be explained by multiple factors that aren't directly addressed in the work. Most of the caveats of the study have been pointed out by the authors in their discussion, yet some of these detract from their claims and findings. If this is a sexually dimorphic trait, or a sex-specific effect, most of the functional analyses are shown without this distinction.

Reviewer #1 (Public Review):

Garcia-Sifuentes and Maney provide a descriptive analysis, using a set of articles previously screened by Woitowich et al. (2020), on how sex differences are reported and evaluated in biological articles. Beyond the inclusion of both males and females in research, it is crucial that authors employ appropriate design and analysis strategies to be able to identify sex-specific biological effects and realize the potential of studying sex as a biological variable. On the one hand, Garcia-Sifuentes and Maney reassuringly show that researchers are often including both sexes in their studies. On the other, concerningly, appropriate statistical tests are not often being used to correctly test for sex-specific differences. In addition, their results suggest that authors often pool data from both sexes together without appropriate statistical tests, or even do it in the presence of evidence for differences.

The article is well written and organized through four questions: whether sex differences were reported by authors of the included studies, 2) whether the studies used a factorial design and whether an appropriate statistical comparison was used to conclude a sex-specific effect, 3) whether data from males and females were pooled, and 4) terminology used (gender vs. sex) to describe non-human animals. Though the analysis is not systematic in nature, it uses a convenience sample from Woitowich et al. (2020) who surveyed 34 journals over 9 disciplines. The subset of included in this work is limited within some of these journals and disciplines as the authors note, and as such future surveys may be useful to quantify within-discipline practices more rigorously.

I note only one potential weakness:

Per line 374, it is noted that there was a single coder for all extractions. Given the complexity of some classifications, this may be of some concern. As the authors noted in places, language used to describe statistical tests, results, and interpretations can vary considerably, and having a second set of eyes review each paper would have been good to reduce the potential for some systematic misclassification. It is also noted that "A subset of the articles was independently coded by YGS and any discrepancies discussed between the authors until agreement was reached." More information would be helpful about this process.

As a point of interpretation, the authors selected the article as the unit of analysis which is reasonable. Some articles however may have many experiments within, and therefore, the results may be different (though unlikely qualitatively) if considered at the level of the experiment. Within articles, is there reason to think there is much variability on the questions addressed herein?

Reviewer #2 (Public Review):

The authors performed an additional meta-analysis on articles collected for another study considering the use of sex as a biological variable. In this study, they examined a subset of articles that did include male and female subjects to see how sex differences were reported on and tested.

One criticism I have of the phrase "sex as a biological variable" is that it does not immediately fit into a traditional statistical framework. Does it mean that sex should be treated as a possible confounder or mediator in the scope of causal inference? One or both of those? This gets to the crux of what a "sex difference" is: is it a between-groups difference in an independent variable of interest (e.g. treatment), the outcome variable, or a moderating effect on the relationship between a treatment and outcome? Should sex be controlled for, stratified by, or used in an interaction term? These distinctions are important, as the different questions the authors pose of the sample of articles correspond to different kinds of sex differences. It is a strength of this study that all of these different types of sex differences are considered, though the ways in which these fit into different parts of an analysis plan is not thoroughly discussed.

The authors were very thorough in their coding of the articles for how different aspects of the reported analyses were presented. For example, there were nine possible results for how sex as a moderating variable was addressed. The river plots were very useful to map out some of those differences. However, this thoroughness means that they sometimes missed the forest for the trees, as they did not report overall percentages of articles that "did it right" (e.g. the original authors appropriately reported on and correctly interpreted the results of an interaction). There was a focus on whether the results in the articles were positive or negative and what the fields the articles came from, but the small sample sizes for these made it difficult at times to see a big picture. The discussion of errors in analysis and reporting by the original authors was very through, though.

The broader picture the authors paint with their findings is one where it seems like the analysis and reporting of sex differences has substantial room for improvement in the surveyed fields. I agree with this study's conclusion that more statistical training is needed for scientists in both clinical and basic science research. Unfortunately, this is not a new problem as evidenced by the wealth of prior literature the authors cite on that issue.

Reviewer #3 (Public Review):

Garcia-Sifuentes and Maney have conducted a comprehensive descriptive audit on statistically treatment, reporting and interpretation of the effects of SABV on the studied outcomes in articles published across nine scholarly disciplines. In this manuscript, the authors present the proportion of studies examining and reporting sex differences and the proportion of studies that have correctly treated, reported and interpreted SABV (i.e. modeling the main effects and the interaction of examined treatment(s) and sex) as well as less than ideal or inaccurate methods employed by researchers to examine the effects of sex. The authors also present the prevalence of inconsistencies in reporting (e.g. reporting a sex difference when a difference was not examined for / does not exist). Furthermore, the authors reported data on the use of the term "gender" in place of "sex". Overall, the manuscript provides a valuable, data-driven summary on statistical treatment of SABV in recently published articles across the reviewed nine disciplines, while providing data to suggest the fact that the practices followed in a majority of researchers to examine for the effects of SABV are less than ideal, if not inaccurate.<br> The manuscript is well-written in general and most conclusions are mostly supported by the data. The use of river plots to summarize the data is commendable. However, a few minor limitations of the manuscript are noteworthy.

1. The manuscript presents overall summary statistics of the responses for each questions the authors asked about the reviewed articles and also discipline-wise summaries. However, it is not very clear as to how the authors classified the articles into these disciplines (especially given that some articles may have covered the scope of more than one discipline). This limits the ability to make inferences about the treatment of SABV in the considered disciplines.

2. The authors also acknowledge that the coding was based on their interpretation of the data presentation and wording. As such, there is a possibility of introducing bias due to the subjectivity of at least some decisions that had to be made. It would have been ideal if all articles were coded by at least two independent coders and a third coder voted on decisions on which a disagreement was observed. However, only one author (author 2) has made all the coding decisions and the validity of the decisions have not been checked. As such, the validity of the reported proportions and percentages remain questionable.

Reviewer #3 (Public Review):

The balance between excitation and inhibition in the cortex is an interesting topic, and it has already been a focus of study for a while.<br> The current manuscript focuses on the 1/f slope of the EEG spectra as the neural substrate of the change in the balance between excitation and inhibition.<br> While the approach they use to analyze their data is interesting,  unfortunately, for the reasons I'll outline below the study's conclusions are not supported by the data, and the findings do not add any new insight conceptually or mechanistically to our understanding of attention, excitation or inhibition.<br> While the study aims to "test the conjecture that 1/f-like EEG activity captures changes in the E:I balance of underlying neural populations.", ultimately the central conclusions of the work is just conjecture in that they are inference formed without sufficient evidence. 

Anaesthesia study: EEG spectral exponents as a non-invasive approximation of E:I balance<br> The authors observe the 1/f slope was different over pre-selected central electrodes sites between 4 participants undergoing ketamine and propofol anaesthesia.  The rather small sample size is a cause for concern, as are the authors' rationale for looking at the central electrodes -they claim these electrodes receive contributions from many cortical and subcortical sources, but that can be said of any other electrodes at the scalp.<br> But I believe the most critical weakness here is the authors' claim that during anaesthesia is that propofol is "known" to result in a "net" increase of inhibition, while ketamine an increase in net excitation.  We still know very little about what neurophysiologically is happening under anaesthesia and the concept of "net" inhibition and excitation is rather a gross simplification of what happens to the central nervous system under these two agents. Just as an example, propofol has been found to have some excitatory influence on brain function, with dosage of the anaesthetic also playing role: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2717965/. On the other hand, ketamine has been observed to inhibit interneurons and cortical stimulus-locked responses, but cause excitation in the auditory cortex : https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP279705.  

Suffice to say the interaction between anaesthetic agents and the brain is rather complex. Decades of research has shown that the EEG spectra changes during anaesthesia. To rather arbitrarily say one agent has a net inhibitory impact while another excitatory impact, then link those to qualitative changes in the EEG spectra of 4 participants, and further link that back to E:I ratio is committing the scientific fallacy of Begging the Claim.

Cross-modal study: EEG spectral exponents track modality-specific, attention-induced changes in E:I<br> Here the authors observe a difference in 1/f slope depending on if the participants (n=24) were paying attention to the auditory or visual stream.   My central issue here is again with the authors' assumptions: cross-modal attention reflects attention-induced E/I.  While attention to a single sensory modality can result in decreased activity in cortical regions that process information from an unattended sensory modality, there is no basis here to say that the task-irrelevant region is actually inhibited.<br> The authors here do observe differences in 1/f slope as a function of attentional location, and these differences do account for some of the variances in behavior in the task.

But unfortunately other than a purely descriptive exercise, there is not any sort of mechanistic insight is revealed here with regards to attentional allocation, excitation, and inhibition.

Reviewer #1 (Public Review):

"Modality-specific tracking of attention and sensory statistics in the human electrophysiological spectral exponent," Waschke et al. This paper follows upon a recent paper by a subset of the same authors that laid out the signal processing-bases for decomposing the EEG signal into periodic (i.e., "oscillatory") and aperiodic components (Donoghue et al., 2020). Here, the focus is on establishing physiological and functional interpretations of one of these aperiodic components: the exponent term of the 1/f(to the x power) fit to the power spectrum (a.k.a., its 'slope'). This is very important work that will have strong and lasting impact on how people design and interpret the results from EEG experiments, and is also likely to trigger many reanalyses of previously published data sets. However, the manuscript could do a better job of explain WHY this is so. In this reviewer's opinion, more linkage with elements of Donoghue et al. (2020). would help considerably.

First, a brief summary of what this manuscript does, and why it is important. The first section reanalyzes data sets in human subjects undergoing ketamine or propofol anaesthesia, known to influence the E:I balance in the neural circuits that give rise to the EEG. This is an important step in establishing the physiological validity of the fundamental proposition that flattening of the 1/f component reflects an increase in the E:I balance whereas steepening reflects a decrease. This is because these effects of these two anaesthetic agents has been well established in several invasive studies. The second section demonstrates the functional properties of 1/f slope, in that tracks shifts of attention between visual and auditory stimuli in an electrode-specific manner (i.e., posterior for visual, central for auditory), and it also captures aperiodic stucture in these stimuli. It's not too strong to say that, after this paper, EEG-related research will never be the same again. The reason for this, however, isn't stated as clearly as it could be.

With regard to exposition, the manuscript could be improved in terms of building on Donoghue et al. (2020). To simplify, a main take-away from Donoghue et al. (2020) is that many past interpretations of EEG signals have mistakenly attributed to task- (or state-) related changes to changes in one or more oscillatory components of the signal. Perhaps most egregiously, what can appear as a change in power in the alpha band can often be shown to be better explained as no change in alpha but instead a change in either the slope or the offset of the 1/f component of the power spectrum. (E.g., the bump at 10 Hz will increase or decrease if the slope of the 1/f component changes, even though the 'true' oscillator centered at 10 Hz hasn't changed.) In this paper, the authors demonstrate that many conditions, physiological state and cognitive challenge, influence 1/f slope in ways that are systematic and that occur independent of changes that may or may not be occuring simultaneously in oscillatory alpha. Broadly, the authors should consider two modifications: first, point out for each key experimental finding how attributing everything to changes in oscillatory alpha (or sometimes other frequencies) would lead to flawed inference; second, don't stop at demonstrating that the slope effects hold when alpha dynamics are partialed out, but also report the converse -- in what ways is oscillatory alpha sensitive to aspects of physiology and/or behavior that 1/f slope is not? Even if there aren't any such cases (which seems unlikely) it would be informative for this to be tested and reported.

Some more specific suggestions follow in the Recommendations for the authors.

Reviewer #2 (Public Review):

The paper investigates two separate studies looking at the spectral exponent of the EEG 1/f-like spectrum: one a study of the effect of anesthesia type (propofol vs. ketamine), using publicly available data, and the other a traditional study of auditory and visual processing relying on selective attention to one modality vs. the other. The authors make a strong case that the value of the spectral exponent depends on the relevant condition, in both studies, but the case for the spectral exponent's dependence on the Excitation:Inhibition balance is much weaker.

The paper presents the two separate studies as tightly linked, but by the end of the paper it appears they may be quite separate.

The anesthesia study is brief and compelling. With respect to the effect of anesthesia type on spectral exponent, the results are very strong, and, given the results of Gao et al. (2017) and the stated properties of propofol vs. ketamine, the connection to E:I balance follows naturally.

The auditory and spectral 1/f tracking study suffers from some weaknesses.

Most importantly, the design is elegant and the results presented are very compelling. 1) Modality-specific attention selectively reduces the EEG spectral exponent (for relevant electrodes reflecting cortical processing of that modality); 2) Changing the value of the spectral exponent in the stimulus results in a similar change in the value of the spectral exponent of the response, but only for the selectively attended modality (and only for relevant electrodes); and 3) the amount of modality-specific spectral-exponent tracking predicts behavior. The interactions and main effects found all support the importance of the spectral exponent as a physiologically and behaviorally important index.

The main problem is a weakness in analysis regarding whether the mechanistic origin of the above effects may be due to temporal tracking of the stimulus waveform (visual contrast/acoustic envelope) by the response waveform. [In the speech literature this would be referred to as "speech tracking", or, sometimes, as speech entrainment (in the weak sense of "entrainment").] As pointed out by the authors, this is not a steady state response because the instantaneous fluctuation rate of the stimulus is constantly changing, and so cannot be analyzed as such (it is also distinct from the evoked responses analyzed). But it is a good match for other analysis methods, for instance Ed Lalor's VESPA and AESPA methods, and their reverse-correlation descendants. Specifically, Lalor et al., 2009 analyzed EEG responses to a non-sinusoidal envelope modulation of a broadband noise carrier and found strong evidence for robust temporal locking. The success of such linear methods there (AESPA for auditory; VESPA for visual) implies that a change in the stimulus spectrum exponent would produce a similar change in the response spectrum exponent, having nothing to do with E:I balance.

The evoked response analysis clearly aims to go in this direction, but since it does not reflect ongoing response properties, it cannot alone speak to this.

Because this plausible mechanism for the spectral-exponent-tracking has not been explored, it is much harder to associate the observed spectral-exponent-tracking as originating from E:I balance. The study does not then hold together well with the anesthesia study, and weakens the links to E:I balance rather than strengthening it.

Reviewer #1 (Public Review):

This manuscript describes the role of PMd cck neurons in the invigoration of escape behavior (ie retreat from aversive stimuli located in a circumscribed area of the environment in which testing was conducted). Further, PMd cck neurons are shown to exert their effect on escape via the dorsal PAG. Finally, in an intriguing twist, aversive images are shown to increase the functional coupling between hypothalamus and PAG in the human brain.

The manuscript is broadly interdisciplinary, spanning multiple subfields of neuroscience research from slice physiology to human brain imaging.

To understand the novelty of the results obtained in the rodent studies, it is important to note that these data are a replication and elaboration of work published recently in Neuron by the primary authors of this manuscript. The current manuscript does not cite the Neuron paper.

The most novel aspect of the rodent experiments presented in this manuscript is the demonstration of a role for cck PMd neurons in invigorating behavioral withdrawal from cues associated with the kind of artificial stimuli commonly used in laboratory settings (ie a grid floor associated with shock). Unfortunately, these results are made somewhat difficult to interpret by a lack of counterbalancing - all subjects receive an assay of escape from a predator prior to the shock floor assay. Certainly, research on stress and sensitization tells us that prior experience with aversive stimuli can influence the response to aversive stimuli encountered in the future. Because the role of this pMD circuitry in predatory escape has already been demonstrated, this counterbalancing issues does somewhat diminish the impact of the most novel rodent data presented here.

The manuscript concludes with an fMRI experiment in which the BOLD response to aversive images is reported to covary across the hypothalamus and PAG. It is intriguing that unpleasant pictures influence BOLD in regions that might be expected to contain circuits homologous to those demonstrated in rodents. It is important to note that viewing images is passive for the subjects of this experiment, and the data include no behavioral analogue of the escape responses that are the focus of the rest of the manuscript.

Reviewer #2 (Public Review):

The manuscript by Wang et al. addresses neuronal mechanisms underlying conserved escape behaviors. The study targets the midbrain periaqueductal grey, specifically the dorsolateral aspect (dlPAG), since previous research demonstrated that activation of dlPAG leads to escape behaviors in rodents and panic-related symptoms in humans. The hypothalamic dorsal premammillary nucleus (PMd) monosynaptically projects to the dlPAG and thus could play a role in escape behavior. The authors test whether cholecystokinin (CCK)-expressing PMd cells could be involved in escape behaviors from innate and conditioned threats using mainly two behavioral paradigms in mice: exposure to a live rat and electrical foot shocks.<br> Different approaches are used to test the main hypothesis. Using fiber photometry and microendoscopy calcium imaging in freely moving mice, the study finds that PMd CCK+ neurons were more active when mice are close to threats and during escape behaviors. Furthermore, PMD CCK+ activation patterns predicted escape behavior in a general linearized model.<br> Chemogenetic inhibition of CCK+ PMd cells decreased escape speed from threats in both behavioral paradigms, while optogenetic activation of those cells lead to an increase in speed. Observation of c-fos expression after optogenetic activation revealed activation within two target areas of the PMd, the dlPAG and anteromedial ventral thalamus (AMv), in which cellular activity measured by fiber photometry also increased during escape behaviors. Interestingly, inhibition of PMd-to-dlPAG pathway, but not PMd-to-AMv, caused a decrease in escape velocity. Lastly, the authors investigated the response of several human participants to threatening images in an fMRI scan. These results suggest that similar to mice, an activation proportional to the threat intensity within a functional connection between hypothalamus and PAG pathway may occur in humans.<br> The authors conclude that a pathway from the PMd to the dlPAG, characterized by expression of CCK, control escape vigor and responsiveness to threat in mice, and that a similar pathway could be present in humans.

Overall, the comprehensive data from multiple approaches support a role of the identified pathway in escape behavior. However, an insufficient description of the used methods and experimental details makes it difficult to assess the validity and conclusivity of some findings. In addition, the strong interpretation emphasis on the functional specificity of the CCK+ PMd-dlPAG pathway appears not fully supported by the data.<br> -<br> 1) The rationale for selection of CCK+ cells of the PMd is missing in the current manuscript. Despite methodological considerations, a clear description of these cells' role and characteristics from the existing literature is needed.

2) The narrowness of the conclusions of the article is unnecessary. Although CCK+ PMd cells could play a role in regulating escape vigor, some of the presented results rather support the notion of a more general role of these cells in mediating defensive states. For example, the photometry data shows correlation of activity with other active defensive behavior. To address this point, a better analysis of the relation between neuronal activity and the general locomotor behavior of the animals is lacking. In addition, the already presented relation with the measured behaviors is not taken into account when interpreting the results (e.g. Fig 7 E). This description would be relevant to more comprehensively attributing functional roles for CCK + PMd cells.

3) The imprecision of the methods description, especially the behavioral analysis is contributing to the previous point. In particular, the escape criterion itself seems to include a vague classification based on movement away from the threat- this should be more concretely defined (e.g. using angle of escape direction). In any case, the different behavioral context dimensions between the two paradigms would probably affect the escape criterion itself and thus have to be taken into account when interpreting the results.

4) In line, more detailed descriptions of the animal's behavior are needed to support assessment of the results regarding the event-related fiber photometry results. Measures like frequency of escape, duration of freezing bouts and angle, duration and total speed of the escape bouts, and a better description of measures like Δ escape speed could be relevant for interpreting the results. In addition, there is no explanation of how the possible overlapping of behaviors in the broad time frame used in the experiments was regarded.

5) Part of the experimental results provide suboptimal evidence for the provided interpretation. That is, the lack of clear quantification and statistical analysis of the microendoscopy calcium imaging data on PMd-CCK+ cells makes it hard to reconcile this data with the photometry data. Furthermore, evidence through c-Fos staining after optogenetically stimulation of PMd-cck+ cells is insufficient evidence for the interpretation of broad, but functionally specific, recruitment of defensive networks. While the data on optogenetic inhibition of the PMd-CCK+ projection to the dlPAG seems to confirm the main hypothesis, both an intra-animal control and demonstration of statistical significance in the analysis are desirable to fully support that role.

6) The provided fMRI data only provides circumstantial evidence to support a functionally specific hypothalamus to PAG pathway especially due to the technical characteristics and limitations of the experimental setup and behavioral paradigm.

Reviewer #3 (Public Review):

This manuscript by Wang et al extends the Adhikari lab's earlier findings of the hypothalamic dorsal premammillary nucleus' role in defensive behavior. Using cell-type specific calcium imaging, the authors show that the activity of CCK-expressing PMd neurons precedes and predicts escape from both learned and unlearned threats. Optogenetic/chemogenetic inhibition revealed that the PMd-dlPAG pathway contributes to escape vigor. Additionally, optogenetic activation of CCK PMd neurons induces Fos in numerous brain regions implicated in fear and escape behaviors. Last, an analogous hypothalamic-PAG pathway in humans is shown to be activated by aversive images in humans.

Although these findings are potentially impactful, additional clarification and data are needed to strengthen and streamline the manuscript, as outlined below.

1. The results of the authors' recent publications (Wang et al Neuron 2021, Reis et al J. Neuro 2021) should be integrated into the manuscript. For example, the rationale for selectively manipulating CCK+ PMd neurons is not stated. Likewise, histological validation that the Cre-dependent GCaMP expression is restricted to CCK+ neurons should be shown or referenced. The authors should also provide discussion as to how the current results integrate with their other recent findings.

2. The authors used male and female mice in their experiments but there are no analyses of potential sex differences in threat responses or escape vigor. Were there any significant sex differences in the measurements presented in Figure 1? A supplementary figure showing data for male and female mice would be helpful. Also, for Figure 1, please display the individual data points so that the reader can appreciate the variability in the behavioral responses. How many approaches and escapes are observed in each test? What is the average duration of a freezing bout?

3. In Fig. 2, there appears to be sustained activity of CCK+ neurons after the onset of threat approach, and ramping activity preceding stretch-attend. In-depth analysis of these responses may be beyond the scope of this study, but the findings should be discussed since the representation of approach-related behaviors indicates the PMd is involved in more general representation of threat proximity, rather than simply escape vigor.

4. The authors state that PMd CCK neuronal activity regulates escape vigor. Although the authors show a correlation of the calcium signal amplitude and escape distance in Fig. 2I, a correlation with escape velocity would be a more convincing measure of vigor.

5. The changes in prediction error from control to threat contexts in Figs. 4B and 4D are compelling, but the prediction error in the threat context seems high. Can the authors provide a basis for what constitutes a 'good' error score?

6. Off-target effects are a potential concern at the dose of CNO used (5 mg/kg). For example, the increased approach speed with CNO in the YFP control group (Fig. 5D) may be a result of the high CNO dose. How was the dose of CNO selected?

7. Given the visible trends in the data, the number of animals used in Fig. 6B is insufficient to make conclusions about the behavioral effect of optogenetic excitation of PMd CCK neurons. Either more animals should be added, or the analysis should be limited to the Fos staining.

Reviewer #1 (Public Review):

Here the authors use a variety of sophisticated approaches to assess the contribution of synaptic parameters to dendritic integration across neuronal maturation. They provide high-quality data identifying cellular parameters that underlie differences in AMPAR-mediated synaptic currents measured between adolescent and adult cerebellar stellate cells, and conclude that differences are attributed to an increase in the complexity of the dendritic arbor. This conclusion relies primarily on the ability of a previously described model for adult stellate cells to recapitulate the age-dependent changes in EPSCs by a change in dendritic branching with no change in synapse density. These rigorous results have implications for understanding how changing structure during neuronal development affects integration of AMPR-mediated synaptic responses.

The data showing that younger SCs have smaller dendritic arbors but similar synapse density is well-documented and provides compelling evidence that these structural changes affect dendritic integration. But the main conclusion also relies on the assumption that the biophysical model built for adult SCs applies to adolescent SCs, and there are additional relevant variables related to synaptic function that have not been fully assessed. Thus, the main conclusions would be strengthened and broadened by additional experimental validation.

Reviewer #2 (Public Review):

This manuscript investigates the cellular mechanisms underlying the maturation of synaptic integration in molecular layer interneurons in the cerebellar cortex. The authors use an impressive combination of techniques to address this question: patch-clamp recordings, 2-photon and electron microscopy, and compartmental modelling. The study builds conceptually and technically on previous work by these authors (Abrahamsson et al. 2012) and extends the principles described in that paper to investigate how developmental changes in dendritic morphology, synapse distribution and strength combine to determine the impact of synaptic inputs at the soma.

1. Models are constructed to confirm the interpretation of experimental results, mostly repeating the simulations from Abrahamsson et al. (2012) using 3D reconstructed morphologies. The results are as expected from cable theory, given the (passive) model assumptions. While this confirmation is welcome and important, it is disappointing to see the opportunity missed to explore the implications of the experimental findings in greater detail. For instance, with the observed distributions of synapses, are there more segregated subunits available for computation in adult vs immature neurons? How do SCs respond at different developmental stages with in vivo-like patterns of input, rather than isolated activation of synapses? Answering these sorts of questions would provide quantitative support for the conclusion that computational properties evolve with development.

2. From a technical perspective, the modeling appears to be well-executed, though more methodological detail is required for it to be reproducible. The AMPA receptor model and reversal potential are unspecified, as is the procedure for fitting the kinetics to data. Were simulations performed at resting potential, and if yes, what was the value? How was the quality of the morphological reconstructions assessed? Accurate measurement of dendritic diameters is crucial to the simulations in this study, so providing additional morphometrics would be helpful for assessing the results. Will the models and morphologies be deposited in ModelDB or similar?

3. The Discussion should justify the assumption of AMPA-only synapses in the model (by citing available experimental data) as well as the limitations of this assumption in the case of different spatiotemporal patterns of parallel fiber activation.

4. What is the likely influence of gap junction coupling between SCs on the results presented here, and on synaptic integration in SCs more generally - and how does it change during development? This should also be discussed.

5. All experiments and all simulations in the manuscript were done in voltage clamp (the Methods section should give further details, including the series resistance). What is the significance of the key results of the manuscript on synapse distribution and branching pattern of postsynaptic dendrites in immature and adult SCs for the typical mode of synaptic integration in vivo, i.e. in current clamp? What is their significance for neuronal output, considering that SCs are spontaneously active?

Reviewer #3 (Public Review):

1) Although the authors were thorough in their efforts to find the mechanism underlying the differences in the young and adult SC synaptic event time course, the authors should consider the possibility of inherently different glutamate receptors, either by alterations in the subunit composition or by an additional modulatory subunit. The literature actually suggests that this might be the case, as several publications described altered AMPA receptor properties (not just density) during development in stellate cells (Bureau, Mulle 2004; Sun, Liu 2007; Liu, Cull-Candy 2002). The authors need to address these possibilities, as modulatory subunits are known to alter receptor kinetics and conductance as well.

The authors correctly identify the relationship between local dendritic resistance and the reduction of driving force, but they assume the same relationship for young SCs as well in their model. This assumption is not supported by recordings, and as there are several publications about the disparity of input impedance for young versus adult cells (Schmidt-Hieber, Bischoffberger 2007).

2) The authors use extracellular stimulation of parallel fibers. The authors note that due to the orientation of the PF, and the slicing angle, they can restrict the spatial extent of the stimuli. However, this method does not guarantee that the stimulated fibers will all connect to the same dendritic branch. Whether two stimulated synapses connect to the same dendrite or not can heavily influence summation. This is especially a great concern for these cells as the Scholl analysis showed that young and adult SC cells have different amount of distal dendrites. Therefore, if the stimulated axons connect to several different neighboring dendrites instead of the one or two in case of young SC cells, then the model calculations and the conclusions about the summation rules may be erroneous.

Joint Public Review:

The authors established rank in cohorts of four animals using the tube test and calculated weighted David's scores (DS) to classify animals as subordinate, dominant, or intermediate. The validity of the DS was determined using a warm corner test. After ranking, animals were exposed to social or non-social stress and evaluated for both social interaction and active coping. Finally, the authors asked whether a change in dominance would affect stress susceptibility. The major findings were that dominant males, but not females, exhibited resilience phenotypes, and that subordinates of both sexes exhibit a reduction in social interaction after non-social stress. A major strength of the research is in its approach to sex as a biological variable - males and females were studied in parallel, allowing for confident evaluation of the role sex plays in experimental outcomes. The discussion is well written and thoughtful. Although the directionality of the results is not particularly surprising (broadly speaking, resilience in dominant animals and susceptibility in subordinates), this is a thorough characterization of the relationship between social status and stress effects. That said, at least some investigation into known endocrine factors that relate to social hierarchies (e.g. testosterone) would have been appropriate.

Reviewer #1 (Public Review):

Nordén et al. investigated the mechanisms that underlie the brilliant colours of iridescent feathers that more than double the range of colours. They successfully do this first by identifying the three key modifications of melanosomes in brilliant iridescent structures, followed by extensive optical modelling of nanostructures (up to 4500 different combinations) which is validated by spectral data from 120 plumage regions across 80 diverse bird species.

The conclusions of this paper are well supported and I have only few minor suggestions left for the authors. The strength of this paper is not in the novelty of the hypothesis (the importance of thin layers for iridescence is well known), however, as the researchers point out they are the first to place this in a broader context identifying out some general principles that underlie irididescent colour production. The methodology used is state-of-the art and the dataset used is relatively big. The major methodological weakness, if there is any, would be that the validation dataset is a bit low (80 species) which might explain why some patterns observed in the optical simulations were not recovered in the real feathers. However, I think the authors provide a sound explanation of why simulations and feathers not always agree, so I'm not sure if adding more reflectance measurements is necessary. Additionally, in the introduction the question "Why have bird species with brilliant iridescence evolved not one but four different melanosome types?" is, for me, not entirely solved and is out of the scope of this manuscript and I would have like more functional and mechanism-related questions such as "at what point are melanosomes too thick to make iridescent color?". Regardless, I do believe that this manuscript is a good starting point to solve other evolutionary questions. In addition to providing a census on how iridescent colouration is produced, and what the effects are on colour diversity, it provides an overview of where these different mechanisms are present. In this way, future work can use this dataset to test additional evolutionary and genetic questions currently unsolved.

Reviewer #2 (Public Review):

In this manuscript, the authors aimed to identify the mechanisms that separate 'brilliant' from 'weakly' iridescent feathers in birds. Specifically, by leveraging theoretical predictions and the diversity of melanosome shapes in birds, the authors were able to conclude that thin melanin layers were the primary mechanism driving the production of 'brilliant' iridescence. FDTD models of bird melanosomes further support their conclusions that thin layers can greatly expand the range of possible colors that can be produced.

The volume and comprehensiveness of the data in this manuscript will be greatly beneficial to others studying structural color, and there is good alignment between theoretical predictions, morphological observations, and model results. Further, despite some discrepancies between the optical modeling results and the results from spectrophotometry (a common issue in any FDTD modeling comparison) the authors did a thorough job of reconciling those differences with the available data. Finally, the authors do a good job of walking the reader through each set of predictions and comparing those predictions to their results. This adds significant value as a framework for others investigating iridescence across taxa, or researchers exploring specific clades of birds in detail.

Reviewer #1 (Public Review):

Salehinejad et al. run a battery of tests to investigate the effects of sleep deprivation on cortical excitability using TMS, LTP/LTD-like plasticity using tDCS, EEG-derived measures and behavioral task-performance. The study confirms evidence for sleep deprivation resulting in an increase in cortical excitability, diminishing LTP-like plasticity changes, increase in EEG theta band-power and worse task-performance. Additionally, a protocol usual resulting in LTD-like plasticity results in LTP-like changes in the sleep deprivation condition.

1) My main comment is regarding the motivation for executing this specific study setup, which did not become clear to me. It's a robust experimental design, with general approach quite similar to the (in the current manuscript heavily cited) Kuhn et al. 2016 study (which investigates cortical excitability, EEG markers, and changes in LTP mechanisms), with additional inclusion of LTD-plasticity measures. The authors list comprehensiveness as motivation, but the power of a comprehensive study like this would lie in being able to make comparisons across measures to identify new interrelations or interesting subgroups of participants differentially affected by sleep deprivations. These comparisons are presented in l. 322 and otherwise at the end of the supplementary material and the study does not seem to be designed with these as the main motivation in mind. Can the authors could comment on this & clarify their motivation? Maybe the authors can highlight in what way their study constitutes a methodological improvement and incorporates new aspects regarding hypothesis development as compared to e.g. Kuhn et al. 2016; currently, the authors highlight mainly the addition of LTD-plasticity protocols. Similarly, no motivation/context/hypotheses are given for saliva testing. There are a lot of different results, but e.g. the cortical excitability results are not discussed in depth, e.g. there is no effect on IO curve, but on other measures of excitability, the conclusion of that paragraph is only "our results demonstrate that corticocortical and corticospinal excitability are upscaled after sleep deprivation." There are some conflicting results regarding cortical excitability measures in the literature, possibly this could be discussed, so the reader can evaluate in what way the current study constitutes an improvement, for instance methodologically, over previous studies.

2) EEG-measures. In general, I find the presented evidence regarding a link between synaptic strength and human theta-power is weak. In humans, rhythmic theta activity can be found mostly in the form of midfrontal theta. Here, the largest changes seem to be in posterior electrodes (judging according to in Fig 4 bottom row), which will not capture rhythmic midfrontal theta in humans. Can the authors explain the scaling of the Fig. 4 top vs. bottom row, there seems to be a mismatch? No legend is given for the bottom row. The activity captured here is probably related to changes in _nonrhythmic_ 1/f-type activity (which displays large changes relating to arousal: e.g. https://elifesciences.org/articles/55092). It would be of benefit to see a power spectrum for the EEG-measures to see the specific type of power changes across all frequencies & to verify that these are actually oscillatory peaks in individual subjects. As far as I understood, the referenced study Vyazovskiy et al., 2008 contains no information regarding theta as a marker for synaptic potentiation. The evidence that synaptic strength is captured by the specifically used measures needs to be strengthened or statements like "measured synaptic strength via the resting-EEG theta/alpha pattern" need to be more carefully stated.

3) In general, the authors generally do a good job pointing out multiple comparison corrected tests. In some cases, e.g. for their correlational analyses across measures, significant results are reported, but without a clearer discussion on what other tests were computed and how correction was applied, the evidence strength of these are hard to evaluate. Please check for all presented correlations.

There are also inconsistencies like:<br> > " The average levels of cortisol and melatonin were lower after sleep deprivation vs sufficient sleep (cortisol: 3.51{plus minus}2.20 vs 4.85{plus minus}3.23, p=0.05; melatonin 10.50{plus minus}10.66 vs 16.07{plus minus}14.94, p=0.16)"

The p-values are not significant here?

Reviewer #2 (Public Review):

This study represents the currently most comprehensive characterization of indices of synaptic plasticity and cognition in humans in the context of sleep deprivation. It provides further support for an interplay between the time course of synaptic strength/cortical excitability (homeostatic plasticity) and the inducibility of associative synaptic LTP- LTD-like plasticity. The study is of great interest, the translation of findings is of potential clinical relevance, the methods appear to be solid and the results are mostly convincing. I believe that the writing of the manuscript should be improved (e.g. quality of referencing), clearer framework and hypothesis, reduction of redundancies, and more precise discussion. However, all of these points can be addressed since the overall concept, design, conduct and findings are convincing and of great interest to the field of sleep research, but also more broader to the neurosciences, to clinicians and the public.

Reviewer #1 (Public Review): 

This is an interesting and well-written paper which presents the results of a model for cortical plasticity and resulting increase in neuronal responses to unexpected stimuli. Overall, this is an elegant study that provides a number of interesting, experimentally testable, hypotheses and develops a prediction for a mechanism for novelty response generation. The results are clearly presented, and provide for an insightful contribution in linking circuit models with function. 

Below, we comment on the relation between the model findings and published experimental results. <br> The authors use the term "novelty" response, it is unclear to me whether this is a true novelty response because a number of stimulus parameters differs from those identified in the literature. The network is not sensitive to temporal structure, which suggests that it does not completely replicate certain aspects of neuronal adaptation in cortex. MEG studies in humans and electrophysiology in rats suggest that cortex should exhibit a differential response to different sequences of stimuli drawn from the same distribution. This model may be adjusted to produce temporal dependencies. 

In Homann 2017, the stimuli were scenes comprised of many Gabor patches. It would be interesting to consider how a broader pattern of stimulation would affect the results. 

One of the primary findings of Natan et al., 2015 was the differential effect of optogenetic disinhibition of SOM vs PV interneurons on the SSA response. It is possible that the differential learning rules for PVs and SOMs could produce variations in the novelty responses. 

An emerging theme from this paper and from other models such as Yarden and Nelken, 2017 and Park and Geffen, 2020, may be that the tonotopic organization of similarly tuned neurons helps facilitate adaptation. Tuned assemblies were a key feature of the models in these papers. Here, the tonotopic organization arises from the STDP rule here, with plasticity supporting tonotopic organization.

Reviewer #2 (Public Review): 

This study by Schulz et al flushes out in fine detail an interesting consequences of inhibitory synaptic plasticity in plastic neuronal networks, showing that its ability to balance precisely only previously experienced stimuli makes this type pf plasticity an excellent candidate mechanism to allow novelty detection. Strong transient responses will be evoked only by those stimuli which have not previously activated (and thus trained) the stimulus specific set of inhibitory stimuli. An open question remains with regard to the time scales and speed of learning at these inhibitory sites, something that will be answerable by the experimental audience of this paper (but could be investigated in a bit more detail in the model as well).

Reviewer #3 (Public Review): 

This computational paper addresses the mechanisms of sensory adaptation and novelty detection in the auditory cortex. A spiking RNN model of 5000 (4000 excitatory/1000 inhibitory) units is developed and adapted to sequences of inputs (ABC...) followed by a novel stimulus N. As with experiments the model captures the adaptation to the repeated stimuli, as well as a strong response to a novel stimulus. In contrast to many models of sensory adaptation that rely on short-term synaptic plasticity, here adaptation arises from STDP at the Inh->Ex connection. Specifically, during ABCABC ... the inhibitory connection onto active Ex units is enhanced through associative plasticity mechanisms, but not onto the inactive Ex units, thus the adaptation does not apply to the novel stimuli. While the approach seems fairly novel it is also speculative and seems to run contrary to the existing experimental data. 

1) The reason many models of adaptation focus on short-term synaptic plasticity (STP) as opposed to STDP is that the later generally is not thought to operate on the adaptation time scale of seconds. Specifically, STDP is generally considered to be a form of long-term associative plasticity, and thus to rely on mechanisms such as the insertion of new receptors-a processes that is unlikely to operate on a time scale of a second or so. Adaptation is robustly observed at 400 ms (e.g., Natan et al 2017), a time scale that is generally considered to be incompatible with STDP. E.g. in the D'Amour paper the authors cite, STDP is induced over a 5 minute pairing protocol, and can still increase over the course of 5-10 minutes post pairing (e.g., Fig 1H). I'm not aware of any evidence suggesting that iSTDP could be induced and expressed on the subsecond to a few seconds time scales. So this seems to be a fundamental issue that needs to addressed or at least discussed. 

Related to the point above the model also contains subtractive normalization implemented with a time step of 20 ms. Again if this normalization is critical to the model this assumption would pose a serious challenge to the model because there is little or no experimental data suggesting that normalization can operate at that time scale. 

2) A further issue relates to the temporal structure of adaptation. The authors show that adaption is independent of the sequence of the stimuli (ABCABC vs BACCBA) (it would be best to refer to this as sequential structure not temporal structure, which would often include the duration of stimuli and interval between stimuli). It is well established that the longer the interstimulus interval the less the adaptation. The model may or may not capture this effect depending on the assumptions regarding the spontaneous activity during the ISI as a result of the non-associative (pre-only) iLTD. However, given that STDP generally grows after induction it seems like the model is not likely to capture the standard observation that adapation should be less if the stimuli are presented with an ISI of 800 ms versus 400 ms. In figure 5, for example what happens if stimuli are presented for 20 seconds consecutively versus for 10 second then a silence of 2 seconds before another 10 second stimulation? No mention of the time course of spontaneous recovery from adaptation is made. 

3) It also does not seem like the model will capture recently reported effects such as the observation that optogenetic inactivation of inhibitory neurons during pulse n can actually increase adaptation to tone n+1 (Seay et al, 2020), indeed I believe the current model would make the opposite prediction 

4) In its current state I don't think (I may be mistaken) the model accounts for a related and very general property of auditory cortex: lateral inhibition (e.g., Brosch and Schreiner, 1997; Phillips, Schreiner, Hasenstaub, 2017).

Reviewer #1 (Public Review): 

This manuscript applies the framework of information theory to study a subset of cellular receptors (called lectins) that bind to glycan molecules, with a specific focus on the kinds of glycans that are typical of fungal pathogens. The authors use the concentration of various types of ligands as the input to the signaling channel, and measure the "response" of individual cells using a GFP reporter whose expression is driven by a promoter that responds to NFκB. While this work is overall technically solid, I would suggest that readers keep several issues in mind while evaluating these results. 

1) One of the largest potential limitations of the study is the reliance of the authors on exogenous expression of the relevant receptors in U937 cells. Using a cell-line system like this has several advantages, most notably the fact that the authors can engineer different reporters and different combinations of receptors easily into the same cells. This would be much more difficult with, say, primary cells extracted from a mouse or a human. While the ability to introduce different proteins into the cells is a benefit, the problem is that it is not clear how physiologically relevant the results are. To their credit, the authors perform several controls that suggest that differences in transfection efficiency are not the source of the differences in channel capacity between, say, dectin-1 and dectin-2. As the authors themselves clearly demonstrate, however, the differences in the properties of these signaling system are not based on receptor expression levels, but rather on some other property of the receptor. Now, it could be that the dectin-2 receptor is somehow just more "noisy" in terms of its activity compared to, say, dectin-1. This seems a somewhat less likely explanation, however, and so it is likely that downstream details of the signaling systems differ in some way between dectin-2 and the more "information efficient" receptors studied by the authors. 

The channel capacity of a cell signaling network depends critically on the distributions of the downstream signaling molecules in question: see the original paper by Cheong et al. (2011, Science 334 (6054), 354-8) and subsequent papers (notably Selimkhanov et al. (2014) Science 346 (6215), 1370-3 and Suderman et al. (2018) Interface Focus 8 (6), 20180039). The U937 cells considered here clearly don't serve the physiological function of detecting the glycans considered by the authors; despite the fact that this is an artificial cell line, the fact the authors have to exogenously express the relevant receptors indicates that these cells are not necessarily a good model for the types of cells in the body that actually have evolved to sense these glycan molecules. Signaling molecules readily exhibit cell-type-specific expression levels that influence cellular responses to external stimuli (Rowland et al. (2017) Nat Commun 8, 16009). So it is unclear that the distributions of downstream signaling molecules in U937 cells mirror those that would be observed in the immune cell types relevant to this response. As such, the physiological relevance of the differences between dectin-2 channel capacities and those exhibited by the other receptors are currently unclear. 

2) Another issue that readers might want to keep in mind is that the details of the channel capacity calculation are a bit unclear as the manuscript is currently written. The authors indicate that their channel capacity calculations follow the approach of Cheong et al. (2011) Science 334 (6054), 354-8. However, the extent to which they follow that previous approach is not obvious. For instance, the calculations presented in the 2011 work use a combined bootstrapping/linear extrapolation approach to estimate the mutual information at infinite population size in order to deal with known inaccuracies in the calculation that arise from finite-size effects. The Cheong approach also deals with the question of how many bins to use in order to estimate the joint probability distribution across signal and response. They do this by comparing the mutual information they calculate for the real data with that calculated for random data to ensure that they are not calculating spuriously high mutual information based on having too many bins. While the Cheong et al. paper does a great job explaining why these steps need to be undertaken, a subsequent paper by Suderman et al. (2017, PNAS 114 (22), 5755-60) explains the approach in even greater detail in the supporting information. Those authors also implemented several improvements to the general approach, including a bootstrap method for more accurately estimating the error in the mutual information and channel capacity estimates. 

The problem here is that, while the authors claim to follow the approach of Cheong et al., it seems that they have re-implemented the calculation, and they do not provide sufficient detail to evaluate the extent to which they are performing the same exact calculation. Since estimates of mutual information are technically challenging, specific details of the steps in their approach would be helpful in order to understand how closely their results can be compared with the results of previous authors. For instance, Cheong et al. estimate the "channel capacity" by trying a set of likely unimodal and bimodal distributions for the input to the channel, and choosing the maximal value as the channel capacity. This is clearly a very approximate approach, since the channel capacity is defined as the supremum over an (uncountably infinite) set of input probability distributions. In any case, the authors of the current manuscript use a different approach to this maximization problem. Although it is a bit unclear how their approach works, it seems that they treat the probability of each input bin as an independent parameter (under the constraint that the probabilities sum to one) and then use an optimization algorithm implemented in Python to maximize the mutual information. In principle, this could be a better approach, since the set of input distributions considered is potentially much larger. The details of the optimization algorithm matter, however, and those are currently unclear as the paper is written. 

3) Another issue to be careful about when interpreting these findings is the fact that the authors use logarithmic bins when calculating the channel capacity estimates. This is equivalent to saying that the "output" of the cell signaling channel is not the amount of protein produced under the control of the NFκB promoter, but rather the log of the protein level. Essentially, the authors are considering a case where the relevant output of the system is not the amount of protein itself, but the fold change in the amount of protein. That might be a reasonable assumption, especially if the protein being produced is a transcription factor whose own promoters have evolved to detect fold changes. For many proteins, however, the cell is likely responsive to linear changes in protein concentration, not fold changes. And so choosing the log of the protein level as the output may not make sense in terms of understanding how much information is actually contained in this particular output variable. Regardless, choosing logarithmic bins is not purely a matter of convenience or arbitrary choice, but rather corresponds to a very strong statement about what the relevant output of the channel is.

Reviewer #2 (Public Review): 

My expertise is more on the theoretical than the experimental aspects of this paper, so those will be the focus of these comments. 

Signal transduction is an important area of study for mathematical biologists and biophysicists. This setting is a natural one for information-theoretic methods, and such methods are attracting increasing research interest. Experimental results that attempt to directly quantify the Shannon capacity of signal transduction are particularly interesting. This paper represents an important contribution to this emerging field. 

My main comments are about the rigorousness and correctness of the theoretical results. More details about these results would improve the paper and help the reader understand the results. 

The calculation of channel capacity, given in the methods, is quite a standard calculation and appears to be correct. However, I was confused by the use of the "weighting value" w_i, which is not specified in the manuscript. The input distribution appears to be a product of the weight w_i and the input probability value p_i, and these appear always to occur together as a product w_i p_i. (In joint probabilities w_i p(i,j), the input probability can be extracted using Bayes' rule, leaving w_i p_i p(j|i).) This leads met wonder two things. First, what role does w_i play (is it even necessary)? Second, of particular interest here is the capacity-achieving input distribution p_i, but w_i obscures it; is the physical input distribution p_i equal to the capacity-achieving distribution? If not, what is the meaning of capacity? 

A more minor but important point: the inputs and outputs of the communication channel are never explicitly defined, which makes the meaning of the results unclear. When evaluating the capacity of an information channel, the inputs X and outputs Y should be carefully defined, so that the mutual information I(X;Y) is meaningful; the mutual information is then maximized to obtain capacity. Although it can be inferred that the input X is the ligand concentration, and the output Y is the expression of GFP, it would be helpful if this were stated explicitly.

Reviewer #1 (Public Review): 

This paper by R. Venz et al. describes careful work involving the use of auxin-mediated degradation of the DAF-2 insulin/IGF-1 receptor in the nematode Caenorhabditis elegans. Since its initial discovery as a gene that could dramatically alter lifespan in this organism, daf-2 has been very extensively studied. Nevertheless, the authors make excellent and thorough use of their novel reagents to successfully add important new findings to our understanding of this broadly conserved aging pathway. 

One of the most impactful findings described here is that initiating auxin-mediated degradation of DAF-2 at a time late in life, when ~50% of the worms have already died, can still lead to dramatic increases in lifespan, and does not appear to lead to any other "side" effects of lowered DAF-2 activity. As the authors point out, this is informative in terms of what might be possible in intervening in human aging. The authors are also similarly able to restrict the effects only to those involving lifespan by limiting auxin-mediated degradation of DAF-2 only to specific tissues. 

Interestingly, in addition to the extensive results focused on the lifespan-related phenotypes mediated by DAF-2, the authors are able to draw a useful new conclusion about regulation of dauer diapause by DAF-2, namely that "AID-degradation of DAF-2 during a narrow time period in the DAF-2 degron mid-L1 stage is sufficient to induce dauer formation, suggesting that the decision to enter dauer relies upon an all-or-nothing threshold of DAF-2 protein levels." 

Finally, it is worth noting that successful auxin-induced degradation of transmembrane proteins has not been previously published in C. elegans, and the authors' use of this technique here broadens future application of degron-linked degradation for this class of proteins in this model. 

Overall, these findings are of broad interest to the field. The work is thorough, and the manuscript is extremely clearly written overall, with only a few small exceptions as noted in the Specific Comments. The main and supplementary figures are also very clear, as are their captions, and these add greatly to the overall understanding of the authors' conclusions. I do not have any additional experiments to suggest that would obviously be within the scope of the current study.

Reviewer #2 (Public Review): 

One of the remarkable examples of genetic input into lifespan and healthy aging is insulin/IGF-1 signaling (IIS). Reduced IIS promotes both lifespan and healthspan across species. This phenomenon was first discovered in C. elegans and later translated into other animals including mammals. In C. elegans, there is a single insulin/IGF receptor called DAF-2. Reduced daf-2 activity increases lifespan, but this finding is complicated by other phenotypes present when daf-2 activity is reduced during development. Performing RNAi post-developmentally solves some of these problems, but RNAi is apparently ineffective later in adult life. In this study, Venz et al. create and test a new tool to conditionally degrade DAF-2 protein. Specifically, they add an AID degron to the endogenous daf-2 locus, apparently creating the first degron within a transmembrane protein in C. elegans. When this daf-2 allele is combined with a transgene expressing the TIR1 ubiquitin ligase and in the presence of auxin, DAF-2 will be degraded. The authors use Western blots to show that about 40% of DAF-2 protein is degraded under these conditions. Phenotypic analysis shows that this degradation is sufficient to cause various daf-2 phenotypes during development and young adulthood that are reminiscent of phenotypes observed in daf-2 mutants. Therefore, the authors have established an important new tool to study IIS in worms. Next, the authors use the DAF-2 degron system to determine tissue requirements for DAF-2 and to show that depletion of DAF-2 late in life can increase lifespan without any of the earlier developmental effects. 

This study provides a valuable tool and new information regarding daf-2 function in development and aging. Overall, the study is clear and well-controlled, however there are some points that require clarification. These points may or may not require more experiments and could potentially be addressed by rewording and adding more information and explanation. 

1) An important claim made in the abstract and several times in the paper is that depleting DAF-2 late in life increases lifespan while avoiding "dauer-associated" phenotypes associated with earlier depletion of DAF-2. The increase in lifespan is shown in figure 5, but "dauer-associated" phenotypes are not addressed in worms with this treatment. The authors could add experiments directly examining these phenotypes in worms treated with auxin in mid/late adulthood. Alternatively, if the phenotypes in question could not occur at this late stage, an explanation could be added to clarify this point. 

2) Another claim that is emphasized is regarding the time when DAF-2 activity is required to promote non-dauer development. There are three aspects related to this claim that require more explanation. 

First, statements are made on Line 264 and line 464 that reducing daf-2 during the mid-L1 stage causes dauer formation irrespective of conditions later in development. However, I did not see data addressing the environment at L2d or later stages, other than the observation made earlier that dauer formation is induced at all temperatures tested. 

Second, the statement on line 465 that the abundance of DAF-2 appears to be the key or sole factor in the dauer formation decision should be revised to reflect the fact that there are several interconnected dauer formation pathways and mutations in genes in any of these pathways are sufficient to cause dauer formation. There is no indication that daf-2 is more important than daf-7 or daf-12, for example. 

Third, experiments shown in supplemental figure 3 address the precise timing when daf-2 activity is needed to promote non-dauer development. There are three different criteria used to describe the stages in question. The text describes animals as mid-L1. Supplemental figures 1A-B show hours since the onset of feeding. And finally supplemental figure 3C shows the number of germ cells per animal. However, these three criteria are never put in context relative to each other, so it is difficult to understand how they all relate to each other. Under the conditions in the experiments, when does the L1 molt occur in terms of hours post-feeding? How many germ cells are present at different stages of the L1? Is the correlation between germ cell number and developmental stage based on your own observations or previously published work? 

3) A more minor, but very interesting claim made on line 180 concerns the effect of the environment on DAF-2 levels. Building on the results of Kimura et al., the authors show two different experiments, one in which starvation causes a dramatic reduction in DAF-2 levels (Fig 1F-1G) and a second experiment in which no effect on DAF-2 levels was seen (Fig 1H). They interpret the difference between these experiments to be the different bacterial strains that worms were fed prior to starvation. However, there appears to be many differences between these two experiments. The OP50 experiments were presumably carried out on standard NGM plates, since no additional information was provided. In contrast, the HT1115 experiments mention L4440, which is the empty vector control used for RNAi. This presumably means that worms were exposed to RNAi conditions, including IPTG and ampicillin, as well as the double-stranded RNAi coming from the empty vector. Furthermore, FUDR was mentioned for this experiment but not the OP50 experiment. Therefore, in principle any one of these conditions, or a combination, could have affected DAF-2 levels. If the authors wish to make conclusions about the bacterial strain they should perform an experiment where the bacterial strain is the only difference in experimental conditions. Alternatively, they can adjust their wording to encompass all of the differences between the experiments. In any case, the experimental details for these experiments should be added to the methods section. 

4) As alluded to above, there are several places where the information in figure legends and/or methods is incomplete. Some additional examples are given directly to authors, but in general the authors are encouraged to revisit the legends and methods section and make sure the experiments are clearly described.

Reviewer #3 (Public Review): 

The authors created an auxin-inducible system to degrade the DAF-2/insulin receptor protein in specific tissues and at different ages of the worm C. elegans. Because DAF-2 has pleiotropic functions, the authors used their system to identify the specific tissues from where DAF-2 acts to regulate certain physiological processes. For example, they showed that neuronal DAF-2 regulates reproductive physiology and longevity, whereas intestinal DAF-2 regulates both longevity and oxidative stress responses. On the other hand, body wall muscle-expressed DAF-2 had little or no effect on longevity or oxidative stress, although it remains unclear how effective the auxin-mediated knockdown of DAF-2 is in body wall muscles. By using this system, they also extended worm lifespan when they reduced DAF-2 levels in post-reproductive and very old adult animals. Importantly, they show that DAF-2 functions during most of post-reproductive adulthood to regulate longevity, highlighting the potential of the insulin receptor as a therapeutic target in improving the life quality of older adults in our population. 

This auxin-inducible system allows for a finer dissection of the spatial and temporal requirements for DAF-2 in multiple processes, including the decision window for entering dauer arrest, a developmental program that the animal undergoes during harsh environments. For example, Golden and Riddle (Dev Biol 1984, vol. 102, pp 368-378) previously showed that the dauer entry decision window is around the mid- to late period of the first larval (L1) stage through temperature upshift and downshift experiments under dauer-inducing conditions. In this manuscript, the authors refined this window to mid-L1-specifically, a certain threshold of DAF-2 levels at mid-L1 is required to inhibit dauer entry. 

Another example is the necessity for DAF-2 in multiple tissues in regulating longevity. Apfeld and Kenyon (Cell 1998, vol. 95, pp. 199-210) previously showed through genetic mosaic analysis that daf-2 is necessary in the EMS and AB cell lineages to regulate worm longevity. Here, the authors show that DAF-2 is required in intestinal cells, which arise from the EMS cell lineage, and in neuronal cells, which arise from the AB cell lineage, in regulating lifespan. Interestingly, while the daf-2 EMS mosaic mutants (Apfeld and Kenyon, 1998) and the intestinal-depleted DAF-2 animals (this manuscript) exhibited a similar increase in lifespan, the daf-2 AB mosaic mutants (Apfeld and Kenyon, 1998) and the neuronal-depleted DAF-2 animals (this manuscript) did not. The daf-2 AB mosaic mutants lived twice as long as control (Apfeld and Kenyon, 1998), whereas the neuronal-depleted DAF-2 animals (this manuscript) only lived about 44% longer than control (Supplementary Table 1). This could suggest the requirement for DAF-2 in non-neuronal tissues that arise from the AB cell lineage, in regulating lifespan. 

The authors also show that the lack of germline shrinkage at 15C in DAF-2-depleted animals (DAF-2::degron; auxin-treated; Figure 3C) is still accompanied by a smaller brood size at the same temperature (Figure 3D), which might suggest that DAF-2 affects multiple aspects of the animal's reproductive physiology. This again highlights how this inducible system can uncouple the DAF-2 requirements in multiple processes. Thus, this manuscript shows how this tool, which will likely be in high demand among the C. elegans community, can significantly impact the study of DAF-2/insulin receptor signaling and its pleiotropy.

Reviewer #1 (Public Review): 

This is an interesting and thoroughly performed study that addresses the crucial problem of how the brain optimally encode time magnitudes to reproduce durations with the bias property of timing. The authors developed task where initially the gerbils measured the duration of a black screen stimulus and then reproduced this interval by walking on a treadmill to obtain a food reward. Seven durations in the range of seconds were randomly sampled and used in the task. The animals' behavior showed the regression effect and the scalar property of timing. Next, the authors recorded from neurons in medial prefrontal cortex (mPFC) during task performance. They found that neural responses could be grouped into different categories based on their time varying profiles of activation, including ramp-type neurons with or without slope changes as a function of the times duration. Notably, the activity patterns were different at the single cell and population levels between the measurement and reproduction epochs of the task. In contrast, the state-space trajectories showed similar properties between epochs, although temporal scaling was mainly present during reproduction. The time decoding showed that only combinations of ramp-to-threshold and linear increasing neurons could reproduce the regression effect. However, is not evident how this mixing is accomplished.

Reviewer #2 (Public Review): 

Henke and colleagues study the activity of neurons in the prefrontal cortex of Mongolian gerbils while they perform a time interval reproduction task. They provide behavioral evidence that the model animals exhibit features typical of interval timing, including temporal scaling and the regression effect. They first analyzed the activity of individual neurons finding response profiles that have been observed. Critically, the observed profiles include neurons with constant-slope ramping activity (time accumulators) and slope-adjusting neurons. They also show that individual neurons exhibit activity that scales with interval duration. Moreover, they performed population-level analyses and they found neural dynamics that closely resemble previous observations. They went one step forward and classified the response patterns of individual neurons, and were able to identify a specific response profile that contributed the most to the most to the regression effect of interval timing. 

Well performed phenomenological description of the observed neuronal responses using sound analytical approaches, combined with an appropriate behavioral task that allows to separate time encoding from timed motor execution. 

Importantly, the paper provides confirmation of previous observations made using different tasks, model organisms, brain regions, and interval duration ranges. It builds up on previous work and proposes a mechanism that potentially explains the regression effect. 

Population-level analysis rely on pseudo-population data, which is a suboptimal situation when we want to study neural population coding and population dynamics. Although this is something to consider, by no means this invalidate the results. 

However, the framing of the findings in the broad context of magnitude estimation is not fully supported. The manuscript would largely benefit from a more focused framing, narrowing it to time estimation, rather than magnitude estimation in general.

Reviewer #3 (Public Review): 

Henke and colleagues describe a novel behavioral protocol where Gerbilles are trained to estimate and reproduce time intervals while running on a treadmill in a virtual reality environment. By using this method combined with tetrode-based multiunitary neuronal recordings, authors explored the prefrontal cortex neural activity at the single cell and population levels. The results are consistent with previous examples from the primate and rodent literature, and the main observation indicates that interval magnitudes are represented in the population but not in individual neurons. The main strength of this work is the application of several interesting and well conducted analyses on the neural data. The behavioral model is also very interesting and, in my opinion, has great potential. However, the way the introduction and discussion are presented, makes it difficult to extract the novelty of the results. Furthermore, there are several major concerns that should be addressed to increase the clarity of the study. Specially, some important aspects in the behavioral protocol and performance must be clarified and justified before accepting the interpretations.

Reviewer #1 (Public Review): 

Extracellular vesicles (EVs) are critical for intercellular communication by delivering signal molecules, miRNAs, and also nutrients to recipient cells. The mechanisms on selective sorting of cargos into EVs are still largely unknown. The authors previously revealed that the ribonucleoprotein (RBP) YBX1 is required for sorting of miR-233 to exosomes. In this manuscript, the authors further investigated the detailed mechanism. By conducting extensive imaging, fractionation, and biochemical-based experiments, they convincingly demonstrate that YBX1 undergoes liquid liquid phase separation both in vitro and in vivo through its C-terminal IDR. The YBX1 droplets specifically partition miR-233 and mediate its packing into exosomes. Interestingly, the authors demonstrate a close connection between YBX1 condensates and P-bodies during cargo sorting into EVs. Overall, this is an elegant, well-conducted and-controlled study.

Reviewer #2 (Public Review): 

Using a variety of cell biological and biochemical techniques, Liu et al. demonstrate that the RNA-binding protein YBX1 forms biomolecular condensates in cells and in vitro. Condensation is required for loading of YBX1 and miR-233 into exosomes, while other miRNAs are not recruited. This is a great example how liquid-liquid phase separation of an RNA-binding protein can selectively recruit a specific RNA, both in vitro and in cells, and how condensation is essential for the targeting and 'function' of this RNP complex. Furthermore, the authors demonstrate that certain P-body factors are co-recruited into MVBs, revealing a connection between these membraneless and membrane-bound organelles. 

Overall, the conclusions of the manuscript are well supported by the experiments, and the experiments very cleanly performed and well controlled. This work is a great starting point in dissecting which proteins and RNAs are loaded into exosomes, how condensation of one protein contributes to the recruitment of itself but also other RNA and protein factors, and puts forward several follow-up research questions (e.g. RNA selectivity, the connection to P-bodies).

Reviewer #3 (Public Review): 

How cells signal each other is one of the key questions in biology and for the understanding of tissue's disease states like cancer. Whereas much work focused on classic receptor kinase pathways, comparatively little is known about other means of signaling. 

Interestingly, cells can secrete - among many other molecules - RNA and although it has been under debate whether secreted RNA is indeed found inside secreted membrane bound vesicles it has been suggested that secretion of RNA could be a powerful way of signaling neighboring cells their own disease state. 

Previous work from the Schekman lab demonstrated that a small class of RNA (miRNA) becomes enriched in biochemical fractions of exosomal markers. Although miRNA have been found in exosomal preparations before, Shurtleff et al. 2016 developed an advanced purification protocol which excluded the presence of non-membrane bound "contaminants", among them RNA binding protein complexes. They as well demonstrated the presence of miRNA inside vesicles and identified in them the protein YBX1 - which was found by mass spec to be enriched in exosomal preps - to be responsible for exosomal enrichment of certain miRNA. YBX-1 was identified by RNA affinity chromatography and its overexpression is linked to cancer. Previously, it was thought that Argonautes could be involved in miRNA targeting, but that could be ruled out now. 

In this manuscript by Liu et al., the authors now study how YBX-1 can specifically sort miRNA into exosomes and come to the conclusion that condensation of YBX-1 by phase separation could be the mechanism to enrich not only YBX-1 but as well the specific RNA type miR-223 in exosomes. 

They find that YBX-1 forms liquid-like condensates in cells which are sensitive to treatment with 1,6 hexanediol, find that these condensates can fuse and turn over quickly to 70% of fluorescence after FRAP and show that the C-terminal IDR is responsible for condensate formation. When analyzing the IDR further, they find that YBX-1 IDR mutants, in which either all tyrosines or all arginines and lysines were replaced, showed lower enrichment of miR-223 compared to wt YBX-1 (which is in the range of 100-1000 fold depending on the cell type; see Fig 5A). 

The respective YBX-1 IDR mutant Y to S mutant formed smaller condensates in vitro but still recruited miR-223, whereas the RK to G mutant did not form condensates, indicating that RNA binding and condensate formation are mediated by different domains. miR-223 recruitment is mediated by the cold shock domain (CSD), a domain described in LIN28 to recruit LIN7 miRNA. A conserved phenylalanine residue in the CSD (F85 in YBX-1) has previously been shown to be required for RNA binding. 

Overall, this manuscript is very well written, the experiments are well described and the quality of the data is very good. Although several of the experiments seem at first sight not to be novel, the quality of this manuscript is that these experiments are combined to a new idea, which is how cells could make use of condensate biology to extract certain molecules from P bodies and secrete them to neighboring cells. 

The YBX-1 mutant analysis is pretty convincing, demonstrating that the aromatic or basic charges are required to sort YBX-1 into exosomes. Previous work has shown that aromatic and basic residues play a role in condensate formation of RNA binding proteins with PLD's (Wang et al., 2018). Importantly, the F85A mutant is not affected in condensate formation in vitro, showing that miRNA is not involved in condensate formation but rather becomes enriched in exosomes by association with the CSD domain.

Reviewer #1 (Public Review): 

In this manuscript, de Saint Jan examines the diversity of responses of periglomerular (PG) neurons in response to cell-type-specific stimulation of basal forebrain (BF) inputs. These experiments build upon previous work done by the lab examining the circuit organization of BF inputs to the olfactory bulb. For type II PG neurons that are responsive to BF stimulation, the author finds that responses can be grouped into three broad classes containing: excitation, inhibition, and the excitation-inhibition. The manuscript primarily focuses on the third cell type which has the unique characteristic of having low activity at baseline, is inhibited by BF stimulation, but then increases their firing rate following the first optical stimulation. Through a rigorous set of experiments, the author shows that this bidirectional response is due to the interaction of GABA and ACh release. He then goes on to show that the increase in firing in type 2.3 PG neurons is due to activation of M1 AChRs and closure of M-type potassium channels. 

The characterization of the range of responses of PG 2.1-2.3 neurons is interesting and novel. On the whole, this is a well-designed and systematic study that may be appropriate for the journal. However, since the data comes only from slice recordings it is unclear what the in vivo relevance, such as in odor discrimination, maybe. Addressing this point would greatly increase the impact of this work, but it may be beyond the scope of the current manuscript.

Reviewer #2 (Public Review): 

The author used AAVs to conditionally express ChR2 in subsets of HDB neurons, using Dlx5/6-Cre mice and ChAT-Cre mice to restrict expressions to GABAergic and cholinergic neurons, respectively. In Dlx5/6 mice, optogenetic stimulations of labelled fibers in the OB result in observable modulation of spiking patterns in some PG cells. These effects were heterogeneous: (1) brief excitation, (2) mixture of phasic inhibition and slower excitation, or (3) inhibition on its own. The author teased out the basis of this heterogeneity, by revealing that the three types of optogenetic effects correspond to excitatory GABAA-mediaed effect on "type 2.1" PG cells, GABAA-mediated and M1-mediated effects on "type 2.3" PG cells, and finally GABAA-mediated effect on "type 2.X" cells. Further, the cholinergic effect was separable from the GABAergic effect in ChAT-Cre mice. 

Both HDB and the PG cells comprise heterogenous groups of cells, so a mechanistic understanding of cholinergic modulation of the OB function necessarily requires a cell-type specific investigation of this type and precision. The current study therefore advances our understanding in this crucial aspect. The specific receptor contributions with pharmacology is convincing. It is useful to know that the run-down prevents long-term monitoring of M1-mediated effect. 

The definition of PG cell types as presented is somewhat ambiguous and not done consistently, especially with respect to the author's previous publication, but this is something that further analyses of the existing data will be able to address.

Reviewer #3 (Public Review): 

This manuscript dissects the synaptic impact of basal forebrain (BF) afferents onto periglomerular (PG) cells in the olfactory bulb (OB) of mice using optogenetic stimulation of BF axons and in vitro slice electrophysiological recordings of postsynaptic responses. It is known that the BF projects widely throughout the brain, including in the OB, where it influences cellular physiology and behavior. The BF consists of multiple populations of cholinergic, glutamatergic and GABAergic neurons, and it is unclear how each of these populations contributes to BF's various effects on behavior. Here, the author carries out a series of experiments to dissect the synaptic actions of GABAergic and cholinergic axons on the discharge of PG cells, which function as inhibitory interneurons to OB output neurons. In a series of carefully conducted and clearly described experiments, the author uncovers heterogeneity in how PG cells respond to optogenetic stimulation of BF afferents: some are inhibited by GABA, others are excited by GABA, and others still are excited by type 1 muscarinic acetylcholine receptors (M1 receptors). The study focuses most of the latter, which has not been described previously and reliably evokes a prolonged increase in the firing in approximately 12% of PG cells via closure of M-type potassium channels. Despite the small number of PG cells responding to BF cholinergic axon stimulation in this fashion, a significant increase in spontaneous inhibitory synaptic currents (IPSCs) can be detected in tufted cells, potentially underlying an effect of BF afferents on the gain of OB output. 

One of the strengths of this study is that the experiments are conducted carefully and are well described, making it easy for experimental results to be understood and interpreted. Each finding is supported by a sufficient number of observations, large enough to reveal the rich diversity of cell types and their subtypes within the OB and differences in their response to BF axon stimulation. An exciting discovery is that a small population of poorly studied PG cells displays a classical M1 receptor and M-channel-mediated increase in tonic firing to cholinergic afferent stimulation. Another interesting finding, though not entirely novel nor extensively investigated here is that GABAergic inputs can drive another subpopulation of PG cells to fire an action potential, possibly because their chloride reversal potential is elevated above spike threshold. Overall, the analyses are thorough and statistics sound, yielding a solid functional and anatomical foundation for future analyses investigating the impact of BF axons on OB function. 

A weakness of this study is that it is largely observational and amounts to a functional anatomical study of a specialized brain circuit in vitro, offering little insights as to when or how the synapses under investigation contribute to brain function in vivo. For example, the authors show that optogenetic stimulation of BF cholinergic axons increases the firing of a small fraction of PG cells for several seconds, which in turn increases the frequency of spontaneous IPSCs in principal neurons. It is unclear when these BF axons are active in vivo: if tonically active at >0.5 Hz, the effect revealed here may be occluded at baseline. It is also unclear how the increase in IPSC frequency in tufted cells influences their discharge or that or mitral cells to alter the encoding of odorants. The same is true for GABAergic inputs to the OB, which are only peripherally studied here, and mainly exert an inhibitory effect on PG cells. The conclusions of this study therefore remain limited to the existence at least 2 populations of afferents from the BF - one cholinergic and another GABAergic - that synaptically influence largely non-overlapping subpopulations of PG cells to mediate phasic inhibition through GABA release or phasic excitation through acetylcholine release. Overall, the manuscript offers few additional insights into how these specialized BF signaling pathways arise (i.e. pre or postsynaptic specialization), or how differential modulation of subgroups in interneurons affects OB output and the sense of smell in the mammalian brain.

Reviewer #1 (Public Review): 

For this manuscript, I focused on the metabolite analysis. The data is presented as supporting a common response based on shared selective histories if I'm understanding properly. However, primary metabolite data is hard to interpret in the same fashion as genetic data. This arises because of the high degree of pleiotropy wherein it is very hard to find a mutant or variant that doesn't alter primary metabolism. As such, it is possible that there is a common response less because of shared history and more because there is constraining selection that shapes what is the optimal primary metabolite response to cold in photosynthetic organisms. For example, in Arabidopsis, it has been found that accessions tend to have a highly similar primary metabolism but when they are crossed, the progeny have a vastly wider array of primary metabolism phenotypes, suggesting that the similarity in accessions is not shared genetics but constraining selection that forces compensatory variants. I don't think this detracts from the utility of including the primary metabolism but it would help to have more clarity in the strengths and weaknesses in using metabolite data to track theories and arguments that are largely genetic based.

Reviewer #2 (Public Review): 

Cochlearia, and other species that have rapidly evolved new ecological niches, represent excellent systems to study adaptation to past, present, future and changing environments. Furthermore, reticulate evolution within such groups offers a natural experiment to test hypotheses about the roles of hybridization, introgression, etc. on evolutionary dynamics, including pre-adaptation. However, there are also several significant challenges to using such systems, most crucially separating adaptation as the causal mechanism from the wide array of non-adaptive processes that could also cause the observed patterns. Overall, Wolf and colleagues do a nice job describing this complex taxonomic system and provide multiple lines of inquiry into how observed patterns may align with various adaptive scenarios. Despite the strong descriptive framework, I had trouble understanding exactly how causality could be assigned. Thus, the interpretation and discussion of the results felt speculative.

Reviewer #3 (Public Review): 

There has been intense interest in how plants have responded during periods of rapid climate change in the past. Understanding these responses can increase our understanding of how plants might respond to rapidly accelerating anthropogenic climate shifts and help set conservation priorities. Many paleoecological studies have provided insight on how plants have migrated and persisted in suitable climate refugia (i.e. pockets of suitable habitat that exist even if regional climate is unfavorable for the persistence of a species) throughout glacial cycles, however there has been considerably less work that details the evolutionary dynamics of plants during these periods. This piece provides timely and valuable analyses illustrating the potential influence of pronounced climate change on the evolutionary dynamics of the genus Cochlearia. 

The authors' use of cytogenetic analyses, organellar phylogenies, and demographic modeling allows for insights into the geographic patterns of diversity, speciation rates, and postglacial expansion scenarios of Cochlearia. Drawing unique conclusions from these different lines of evidence provides new understandings into the putative role of Pleistocene glacial cycles in driving evolutionary processes such as speciation. The study also aims to provide insight into the origins of the stated putative cold tolerance exhibited by Cochlearia by using a metabolomics approach; however, the framing and use of a single related outgroup (sister genus Ionopsidium) obfuscate the link between the results and stated conclusions. 

Specifically, regarding the approach that resulted in figure 4 which encompassed the metabolomics and related analyses, the initial climate groupings into 'climate ecotypes' would benefit from clarification and consideration of assignment methods. Typically, using the term ecotype invokes the idea of distinct forms of a species with phenotypic differences adapted to local conditions rather than groupings to those under broad climate regimes. While grouping populations according to climate origin can be useful, it is not clear how the final 9 WorldClim bioclimatic variables were selected (e.g. it is not apparent how importance of or correlations between climate variables, etc. were considered). Consequently, knowing this information would help understand the patterns in figure 4b, which seems to indicate that geographically distant populations experience very similar climate conditions (understanding that similarities can exist but variable selection can greatly influence these patterns). The other concern is in regards to the framing and interpretation of these results. For instance, in the results (lines 329-330) and discussion (lines 419-423), the impression is given that experimental results here match those found in plants belonging to a different genus (i.e. Arabidopsis). However, rather than attributing this to more generally conserved mechanisms in response to considerable cold stress, the authors relate this to the unique history of Cochlearia (and its relationship to the drought adapted sister genus). The authors also note that surprisingly there was no demarcation of cold responses between the climate-defined groupings. Detailing why this is surprising given some of the other conclusion statements would be helpful. Some targeted revision to strengthen this link would be useful to bolster the inference of about the origins of cold tolerance in Cochlearia, rather than making it seem like this result could be expected in other taxa. 

Lastly, another area that would benefit from some clarification and tightening is revisiting the connection between the results and stated conclusions. For instance, some of the statements from the introduction and conclusions indicate the reader might expect explicit niche exploration analyses and more detailed genomic approaches. It is not abundantly clear for a general audience how these results definitively demonstrate how genetic diversity was rescued in reticulate and polyploid gene pools or species barriers were torn down. These are very specific, strong claims that do not appear to be explicitly discussed outside of the introduction/discussion or directly related to the results presented in this manuscript. This is no way diminishes the considerable effort of the authors to conduct the informative array of presented analyses, but more closely aligning the conclusions within the scope of presented results (or providing direct links on how the results provide these insights) would help increase the effectiveness of this manuscript.

Reviewer #1 (Public Review): 

The authors used a high-resolution barcoding approach to probe the complex infection dynamics of systemic ExPEC infection. They identified organ-specific population dynamics and discovered that clearance failures are pervasive, as the results of dramatic expansions of very few bacterial cells from the inoculum. These expansions may not be easily identified by routine CFU-based metrics, but could be readily detected and quantified by their methods. In general, this study is well performed and the manuscript is well written. Additional details on how the parameters (e.g. Nb, Nr, GD, RD, FRD) were calculated should be added in order to help the readers to understand the models.

Reviewer #2 (Public Review): 

This is a very interesting study analyzing the pathogen population trajectories during systemic infections. Utilizing the extra-intestinal pathogenic Escherichia coli CFT073, the authors analyzed the trajectories and mechanisms used by this bacterium to persist in particular niches. The authors found that although most of the inoculum was eradicated after injection, some bacteria were able to persist (in some tissues/organs while not in others), mainly because the ability of some individual clones to expand. The authors also analyzed the consequences of this clonal expansion in the ability of the bacterium to disseminate to other locations. While some of the conclusions obtained in this study are not completely novel, no previous studies have analyzed these processes with the detail and depth shown here. And therefore, this work represents an impressive tour de force analyzing these important processes in vivo. Since the quantity of work that this manuscript already has is impressive, more than asking for additional experiments I would like to ask a few general questions, with the hope that they will better connect the results obtained here with what is observed in clinical infections: 

1) The authors have shown in their study that changes in different parameters (i.e. inoculum) impact the trajectories followed by the pathogens. One can imagine that in a normal scenario, the number of bacterial cells that will arrive to the blood will be much lower than the inoculum used in these experiments. Similarly, I do not know if additional host factors will also influence which people will be infected or not. I would like the authors to explain more clearly why the model used is relevant to understand what is happening during normal infections. 

2) The authors clearly show that E. coli CFT073 colonizes the liver much better than other organs. Is there any evidence that liver infections are over-represented in patients suffering with E. coli bacteremia? 

3) In the last part of the paper, the authors identified several genes involved in pathogen survival after infection (early stages). I'm not sure this part is related with the previous part of the manuscript, because my feeling is that the observed clonal expansion was a stochastic process, not driven by mutations in specific genes. I would suggest that some of the clones that expanded were sequenced, to clearly show if this process of clonal expansion was random (or conversely, was driven in some cases by mutations in specific genes). 

4) Finally, there is an interesting paper from Fitzgerald's group (PMID: 31807698), analyzing both intra and inter-host dissemination that should be referenced here.

Reviewer #3 (Public Review): 

Changes in the population of bacteria that cause systemic infections after bloodstream infection are poorly understood. New genetic analytic tools, such as those developed and employed by the investigators here, afford opportunities to dissect bacterial populations at sites of infection, and understand the kinetics of pathogenesis. The authors of this study infected mice intravenously with a library of extra-intestinal E coli strains that were genetically identical except for unique DNA tags that did not affect their growth. By harvesting bacteria from different sites and times of infection, and performing deep sequencing to define the distribution of specific tagged-strains, investigators provide a highly detailed snapshot of populations, with discriminatory power superior to prior studies. Strengths of the project include logical and straightforward study design, and the novelty of the molecular and analytic tools employed. The paper is cogently written, which is essential given the attention to detail necessary to follow the work and results clearly. Weaknesses are that this is a preliminary, proof of concept type of project, and reproducibility of findings, mechanistic studies, and deeper scientific investigation of temporal-spatial patterns will require future study. 

It is also unclear if results are specific to the strain of extra-intestinal E coli or strain of mice used here. However, this study and the methods used provide a necessary foundation for such future studies. The results are presented in descriptive, straightforward fashion, and interpretations and conclusions are supported by the data. The study will have impact on future studies to better understand pathogenesis, and in providing methods, tools and experimental templates for future studies. As such, the paper is a solid advance for the field.

Joint Public Review: 

The authors combine their existing software (ISETBio) for simulating the responses of the human cone mosaic to arbitrary spatiochromatic stimuli with a Bayesian model of image reconstruction, in order to explore hypotheses concerning the efficiency and the perceptual consequences of early visual processing. The simulated cone signals include the effect of human optics, the spectral sensitivities of the cones, a hexagonal cone mosaic, the relative numbers of L, M and S cones, and the Poisson noise of photopigment isomerization. They do not include any noise of biological origin or any synaptic processing beyond the photoreceptors. The image-reconstruction model computes Bayes-optimal estimates of ground-truth natural images given a root mean-squared-error cost function and an estimated prior for the spatio-chromatic structure of natural images. The assumption is that the average reconstruction error for natural images is a good proxy for the average information contained in the cone responses, under natural conditions like those that drove the evolution of the human retina. 

The authors' simulations address several interesting questions. The first is whether there is a principled explanation for why the human cone mosaic contains mostly L and M cones and relatively few S cones, and why the smaller number of S cones is fairly consistent across individuals, whereas the relative number of L and M cones is highly variable. They vary the relative numbers of cones in their simulations and find that ~10% of S cones is optimal for image reconstruction and that the quality of image reconstruction is largely invariant to the relative numbers of L and M cones. It is not clear whether accommodation of the lens was allowed to vary in order to determine whether the overall minimum RMS reconstruction error occurs when accommodation is centered on a wavelength near the peak of the L and M cone spectral sensitivities (which is where accommodation tends to be centered in humans). 

The second question concerns color appearance in color-deficient individuals. The authors created reconstructions of natural images for protanopia (no M cones), deuteranopia (no L cones), tritanopia (no S cones), and deuteranomaly (highly overlapping L and M cone spectral sensitivities). They found reconstructions qualitatively similar to other attempts to estimate color appearance in these individuals. While this is an important result, it is not clear whether there is any way to determine which estimates of color appearance are more accurate. Also, it is not clear what approaches would lead to qualitative failures. 

The third question concerns the effect of cone sampling on image appearance as a function of retinal eccentricity and when the optics of the eye are bypassed so that high-contrast, high-frequency grating (striped) patterns can be formed on the retina. Simulations of reconstructed natural images as a function of retinal eccentricity show complex effects because of the simultaneous changes in optics and the cone mosaic with eccentricity. For example, color can be more desaturated at intermediate eccentricities than in the fovea and far periphery. These are interesting predictions that should be testable in perception experiments. Simulations of reconstructed images of gratings when the optics are bypassed are qualitatively consistent with previous reports of grating appearance in the fovea and periphery. 

The last question concerns the effect of the reconstruction computation on predicted detection performance for sinewave grating stimuli (the contrast sensitivity function (CSF)). The authors' find that for diffraction limited optics, the CSFs of the ideal observer for yellow (L+M) grating and red-green (L-M) gratings of equal cone contrast are approximately the same. However, for the image-reconstruction-based observer that applies a matched template to the reconstructed image, the CSFs are much different for L+M and L-M stimuli, and they differ in a way that is qualitatively (but not quantitatively) consistent with human CSFs. This is a striking difference, but it is not obvious whether this result would hold for the more realistic optics assumed in most of the other simulations. As the authors show, the real optics matter because they strongly affect the reconstruction. 

The proposed reconstruction-error cost function is appealing because it can be applied to a wide range of different questions and can generate predictions for both performance and subjective appearance. One potential weakness of reconstruction error as a cost function is that not all the information captured by the error measure may be relevant for the tasks that humans (and other primates) needed to perform in order to survive and reproduce during the time period where the properties of the retina evolved. It would be useful (in the future) to compare reconstruction-based and task-based cost functions that both take into account the statistics of natural images. Other recommendations include a more realistic model of eye movements during human vision, and the inclusion of neural processing beyond the cone photoreceptors.

Reviewer #1 (Public Review): 

The authors show that Mtb infection of resting bone marrow derived macrophages stimulates host fatty acid synthesis including that of certain polyunsaturated fatty acids (PUFAS). The work additionally demonstrates that this has wider relevance in that other bacterial danger signals similarly induce these changes. The effect of interferon-gamma which is known to activate macrophages and effect lipid droplet formation and fatty acid homeostasis in infected macrophages is not investigated. The authors focus on the upstream increase in PUFAS and show that inhibition of their downstream conversion to arachidonic acid and eicosanoids does not affect Mtb growth in macrophages or in mice which is perhaps a bit surprising but expected effects on eicosanoid on macrophage necrosis/ autophagy/ apoptosis are not investigated. Moreover, whether interferon-gamma stimulation which is known to affect eicosanoid production in macrophages would affect this outcome is not explored although the mouse data may suggest that Fads2 inhibition in the context of immune activation will not affect this outcome. In vitro cultured Mtb acquires the saturated, mono-unsaturated as well as PUFAS by the Mce1 transporter with competition assays with various fatty acids suggesting different binding sites depending on fatty acid conformation but acquire these fatty acids by additional mce1-independent transporters in the host macrophage. Interestingly, it is found that during host cell infection, Mtb preferentially acquires host arachidonic acid although the effect of interferon-gamma stimulation on host fatty acid acquisition are not explored.

Reviewer #2 (Public Review): 

In the manuscript by Laval et al, the authors profile fatty acids after Mtb infection in a BMDM system. This is a timely and important piece of work, since metabolic reprogramming of macrophages has emerged as directive in responses to infection and activation but can also be modulated by Mtb to evade host defence. Much of the published work thus far has focused on glucose metabolism and although it is long-known that Mtb alters lipid levels in cells (causing a foamy macrophage-like phenotype), little work has profiled the various lipid species and the mechanisms regulating their production and fate. The authors find, consistent with other work in activated macrophages, that Saturated Fatty Acid (SFAs) and Mono-unsaturated Fatty Acids (MUFAs) are increased after infection by mycobacterial species, but interestingly here, also characterize an increase in certain Polyunsaturated Fatty Acids, but not one of the products of PUFA metabolism, Arachidonic Acid (AA). This is surprising since AA has a major role in host/Mtb macrophage responses particularly the production of various eicosanoids and when an earlier enzyme in the PUFA metabolism pathway is targeted, FADS2, this has a major effect on the host macrophage response, down-regulating inflammatory gene expression and eicosanoid production. Using elegant technologies, the authors go on to reveal an usual dichotomy whereby Mtb itself imports AA alongside some other PUFA & FA species. This preference for AA explains why AA itself is not up-regulated after Mtb infection, but also why targeting the FADS2 enzyme which blunts AA levels, does not affect mycobacterial survival in in-vitro or in-vivo assays. This is particularly interesting since the interaction of different host signals (TLR stimulation, IFN production) in the modulation of specific lipid species in macrophages is receiving a lot of attention lately (Knight et al, 2018, Hsieh et al 2020), yet the importance of these for host immunity versus metabolism of infecting pathogens can be difficult to discern. The authors develop a strain of Mtb in which the import of AA & other FAs is decreased. However, it is still unclear if the bioavailability of AA itself is controlling this phenotype and what effect blocking its import by Mtb has on the host macrophage response. This should be examined further alongside a more complete characterization of the role of FADS2 in both bacterial survival, macrophage metabolism and inflammatory activation.

Reviewer #3 (Public Review): 

Laval et al examine the metabolism of polyunsaturated fatty acids, in particular arachidonic acid (AA), during macrophage infection with M. tuberculosis. First they demonstrate that infection results in increased levels of linoleic acid and other fatty acid precursors of AA in macrophages. Using an inhibitor of FADS2, an enzyme that mediates one step in the pathway by which linoleic acid is converted to arachidonic acid, they demonstrate that synthesis of AA is required for production of immunomodulatory eicosanoids. In addition, treatment of macrophages with the FADS2 inhibitor results in decreased expression of numerous cytokines and effectors of innate immunity. However, treatment of infected macrophages or mice with the inhibitor has no impact on bacterial growth. In parallel experiments, the authors use a click chemistry approach to label specific fatty acids and demonstrate that M. tuberculosis takes up AA and other PUFAs using the known FA transporter Mce1. Finally, they demonstrate that M. tuberculosis takes up these PUFAs during macrophage infection from the host, a phenomenon that is also partially Mce1 dependent. The experiments are for the most part technically sound and well controlled, and their specific findings are plausible. However, the findings represent only an incremental advance over what is already known about lipid metabolism during M. tuberculosis infection.

Reviewer #1 (Public Review):

In this manuscript the authors have addressed the structural reasons for the ability of fatty acyl-AMP ligases (FAAL) to exclude condensation of activated fatty acids with coenzyme-A and facilitate the reaction with other 4-phosphopantetheine linked acceptors. This issue is of significant interest with regard to understanding how certain fatty acids are channeled to specific metabolic fates. The structural issue is the apparent discrimination of the CoA moiety (adenosine 3',5'-bisphosphate) versus a holo-ACP tethered to the 4-phosphopantethein head group. The authors identified a number of probable issues between FAAL and FACL enzymes.

1. The authors have shown that many of the FAAL enzymes lack the positively charged residues that have been shown previously to function in recognition of the CoA moiety (Figure 1a).

2. They have highlighted a number of residues within the putative binding site for the 4'-pantatheine moieties in the FAAL enzymes that likely preclude the binding of this portion of the substrate. They have subsequently mutated these residues in FAAL enzymes from three different organisms and have shown in certain instances that the mutated enzymes are now able to functionally activate CoA (Figure 2c). The authors, however, should have attempted to explain why some of the FAALs behaved differently than others. For example, why does the F284A/M233A mutant of MsFAAL32 function so differently than the corresponding mutant of RsFAAL? Also, the residue numbering is confusing to this reviewer. Thus, in Figures 1b and F1c the specific methionine and phenylalanine residues that are highlighted are labeled as M231 and F279, respectively. Yet in Figure 2 the methionine mutated is listed as M227 and the phenylalanine is listed as F275. Why is there a residue difference of 4?

3. The authors have provided further support for the inability of the apparent canonical site in the FAAL enzymes to be functional by mutating the residues within the active sites of certain FACL enzymes to the bulkier ones found in the FAAL enzymes. Many of the constructs resulted in the loss of function and their ability to activate CoA. However, the loss of function was not uniform across the three FACL enzymes chosen, and the authors have done an insufficient job of explaining the differences. For example, the A276F/A232M mutant of AfFACL is devoid of CoA activity but the corresponding mutants of MtFACL13 and EcFACL are fully functional.

4. The authors have identified a putative alternative binding site for the 4'-pantatheine moiety using various computational searches that can apparently distinguish between CoA and holo-ACP (Figure 3). Mutations of residue within this newly identified pocket (Figure 4c) significantly diminishes the condensation with the ACP from E. coli.

Reviewer #2 (Public Review):

This manuscript succeeds in experimentally establishing the rationale for a acyl-carrier protein (ACP) substrate specificity of the fatty acyl-AMP ligases. While these enzymes structurally resemble the CoA-dependent fatty acyl CoA-ligases (FACLs), the authors demonstrate that the FAALs use a novel binding site to accommodate the ACP substrate. The biochemical studies are solid and clear cut but the evolutionary analysis could be bolstered with additional bioinformatics analysis. With said analysis, this manuscript would contribute significantly to our current knowledge of distinct classes of enzymes that divert fatty acids to virulent lipids in mycobacteria.

Reviewer #3 (Public Review):

This study that attempted to understand how an important family of enzymes (FAALs) involved in fatty acid activation and transfer can recognise one form or a substrate in preference to another. They were trying to identify how certain enzymes recognise the 4'-PP-SH arm of CoASH is not recognised but the same arm attached to a small acyl carrier protein (ACP) is preferred.<br> The author used detailed analysis of crystal structures in the PDB of many examples of ligand-bound ANLs, sequence analysis and molecular modelling to direct site directed mutagenesis of the FAALs and reveal important elements in the enzyme involved in substrate recognition and discriminaton. This is a major strength of the paper. e.g. SFig. 7. There is also a nice evolution discussion about the origin of the ANL family.

Once they have identified potential residues involved in CoSH or ACP-SH (holo) binding they make a number of mutants of various enzymes.<br> These included EcFAAL, MsFAAL32, RsFAAL, MxFAAL, AfFACL etc. They appear to use a radioactive substrate assay (using hot FA) and measure incorporation with various gels which are scanned and counted. This is where I began to get lost and to me it is a major weakness. They compare WT with mutants including site directed mutants (made deletions of deltaFS1 etc) .<br> I found Fig. 2 confusing.<br> The gels are also confusing.<br> There appears to be a problem with the apo- to holo-ACP conversion - why didn't Bs Sfp work to 100%?<br> There is one mass spec analysis - why wasn't more used?

The reaction is not clearly drawn to begin with - I work in this area.<br> Be clear in the steps.<br> Draw out the FA + MgATP to give acyl-adenylate + PPi. Then, add CoASH or ACP-SH and show formation of acyl-CoA + AMP or acyl-ACP + AMP.<br> There are many other assays to measure activity e.g. PPi coupled or measure AMP formation to back up data from the radioactivity.<br> Figure 4 confuses me a lot. As does Supp Fig 6.

It appears the authors are looking for a yes/no answer - active with CoASH or ACP-SH.<br> A Table would help to summarise.

Reviewer #1 (Public Review):

The paper by Messelink analyzes the cell growth of the bacterium Corynebacterium glutamicum by tracking the length of individual bacteria in a microfluidic device and analyzing the data with a new inference method. Analysis of cell growth has been established as a useful tool to get insight into fundamental biological processes including the coordination of biosynthetic pathways, DNA replication and cell division, and ageing. The complication of such analysis is that single-cell growth data is noisy and cell not only show a range of growth rates, but also a range of lengths at birth and at division. One goal is thus to characterize the cell size (here: length L) as a function of the time since birth (t) and the length at birth (lb), from which growth starts, L(t,lb). One approach is to consider bins of birth length and then determine the average growth curve for each bin, L(t | lb), i.e. conditioned on birth length. Here the authors chose a new, complementary approach and consider instead the length as a function of birth length, conditional on a given cell age, L (lb | t). Since the relation turns out to be linear, a linear fit provides an accurate dependence on birth length, the less well sampled of the two parameters and thus a good description of L(t, lb), which is then analyzed further.

This is a great method of analysis and the authors use it to identify several phases of growth. Initially, the length growth rate of C. glutamicum increases and then is constant, indicating linear length growth. In the final phase, cells start to divide, in particular, the faster growing ones, making the interpretation of the data more difficult.

The linear growth is then interpreted in terms of a model based on cell wall synthesis at the pole, which is supported by a simple calculations as well as behavior of some mutants.

While the results of the paper are nice and the inference method used here is very promising, the framing of the discussion is in some places less convincing. The authors argue that their results go against a commonly accepted paradigm of exponential growth, which appears greatly exaggerated. There is a long history of debate about linear versus exponential growth, so even though some prominent recent results have supported exponential growth, I think the idea of a dominant paradigm here is a straw man and ignores the history of the subject. There is no reason not to expect a variety of growth patterns in different microbes, given their diversity in many other features. Rather than viewing the growth pattern described here as one that contradicts a consensus, I would see it as an indication of such diversity.

The part that remains somewhat unsatisfying about this work is the treatment of the late phase of growth, in which analysis is difficult due to the division of cells that removes fast growing cells from the sampled population. The authors solve this issue by analyzing growth only up to the point where the first cell divides. However, they appear to suggest that linear growth continues beyond that, for which some more support would be desirable.

Reviewer #2 (Public Review):

The work tracks single-cell growth trajectories of the bacterieum C. glutamicum and uses quantitative analysis methods to probe growth rate within the cell-cycle. The growth is found to be asymptotically linear, with the onset of linear growth not corresponding to the onset of septum formation. However, the paper supports a model where polar cell-growth is rate-limiting. The authors find a reasonable agreement between the hypothesized RAG model and the elongation rate vs time plots obtained from the experiments.

Reviewer #2 (Public Review):

Caramellino et al., investigated whether rat sensitivity to multipoint correlations show a similar rank order as observed in humans. They show that rat sensitivity to multipoint correlations exhibits a rank order similar to what was shown in Hermundstad et al., 2014. Interestingly, they also show that such rank order is robust within-group and within-subject. The authors further claim that this similarity indicates that rat sensitivity to multipoint correlations follows efficient coding of natural scenes.

The main conclusion of the paper is mostly supported by the data. However, the presentation of results may benefit from some points of clarification:

1) In Hermundstad et al., 2014, the degree of variation in images themselves (Figure 1E) as well as perceptual sensitivity comparison (Figure 2D and 3A) was done among: 2nd-order horizontal/vertical edges, 2nd-order diagonal edges,-shaped 3rd-order correlations, and 4th-order correlations. However, the comparisons here are: first-order, all 2nd-order correlations including ALL horizontal/vertical/diagonal edges, L-shape 3rd-order correlations, and 4th-order correlations. It is unclear how these two rank-order results are parallel given that Hermundstad et al., 2014 did not include 1st-order at all.

2) In Hermundstad et al., 2014, the paper emphasized that the difference of perceptual sensitivity between horizontal/vertical edges and diagonal edges is not merely an "oblique effect": Horizontal and vertical pairwise correlation share an edge, while pixels involved in diagonal pairwise correlation only share a corner. One wonders whether rats show any sensitivity difference between horizontal/vertical edges and diagonal edges. The manuscript in its current form misses this important comparison. Without showing this, the rat sensitivity does not fully reproduce the trend previously observed in humans.

3) Combining 1) and 2), it is unclear why the ranking in rat sensitivity is evidence for efficient coding. In Hermundstad et al., 2014, efficient coding was established by comparing the image-based precision matrix with the human perceptual isodiscrimination contours. There is no such comparison here.

4) If the authors would like to hypothesize that the rat sensitivity shows efficient coding simply because its ranking is similar to humans, more needs to be done to shore up the quantitative comparison between the two.

Reviewer #1 (Public Review):

This research follows up on prior work showing that human visual pattern discriminability is closely related to the statistical features of natural scenes. The present work developed a behavioral choice paradigm to test whether rats could discriminate between patterns, and then measured their sensitivity to different spatial correlation structures. This allowed them to test whether rats possess the same sensitivity to spatial correlation patterns as had been observed in human psychophysical experiments. The experiments found that the ordering of the sensitivity to spatial correlation patterns matched that measured in humans and follows the frequency of different structures in natural scenes, in accordance with an efficient coding hypothesis.

This work has a strong theoretical grounding. The behavioral experiments are well executed and the results show convincingly that rat behavior follows the same pattern as measured in humans. One strength of this data is that is shows that the order of presenting of these patterns during training did not matter for the eventual relative sensitivity measured in the rats.

This research opens up the opportunity to test whether the correlational sensitivities can be altered by changing visual environments, and if so, what neural substrates might be plastic in these cases.

Reviewer #1 (Public Review):

In this study Scheeringa and colleagues make use of an attention manipulation to examine changes in BOLD connectivity between and regions and layers of the human visual cortex. The modulation of this connectivity (over subjects) with attention is then correlated with EEG changes in alpha, beta and gamma bands. The main findings are that decreases in beta power were associated with increasing coupling between deep layers in separate regions and from deep to superficial layers within regions. Within-region deep to superficial modulation of connectivity was also linked to changes in alpha band power.

Strengths:

This is an ambitious study which pushes the limits of what is possible with non-invasive imaging. It combines laminar- resolution fMRI with concurrent EEG recordings. There is a huge potential here and extensions of this method promise a bright future for high spatial and temporal resolution of brain imaging data using concurrent EEG and fMRI.

The BOLD laminar connectivity analysis is impressive and compelling and the electrophysiological data is remarkably clear given the challenging recording conditions.

The article raises interesting questions about the relationship between the electrophysiological and BOLD signals. Surprisingly, whilst the beta and alpha band signals covary negatively with BOLD generally, connectivity metrics in these bands seems to be linked with BOLD coupling changes of opposite sign.

Weaknesses:

Although the BOLD data is highly spatially specific, there is just one electrophysiological time-series per subject. This is no doubt a bi-product of the extensive noise cancellation that is necessary to record within the scanner. The caveat therefore is that the covarying BOLD and electrophysiological changes may derive from different regions.

The analysis methods are slightly non-standard, perhaps for good reason. The main thing that stands out is the use of correlation coefficients, rather than regression coefficients, at the first level of analysis. This could potentially conflate changes in signal with changes in noise or unexplained variance.

Reviewer #2 (Public Review):

Major concerns

- Introduction. The introduction provides an overoptimistic view on the current possibilities with respect to the investigation of layer-specific activation or connectivity in the living human brain. Cortical layers cannot yet be segmented, the fMRI measures only provide an indirect signal that is heavily influenced by partial voluming between cortical depths, and EEG and MEG approaches often only measure two compartments due to low spatial resolution. The introduction, however, gives the impression that layer-specific neuronal connectivity can precisely be measured in the living human brain, which is not the case. The authors should take considerably more care with respect to how they introduce the methodology with clear references to the limitations. Also, statements such as "laminar fMRI allows us to study connectivity.." should be removed. In the same vein, I would suggest to replace laminar fMRI and laminar connectivity with cortical depth-dependent fMRI and connectivity to account for the above mentioned aspects.

- Concept. Whereas the authors provide a model in the introduction that specifies how different frequency bands could relate to cortical depth-dependent connectivity, they do not develop a working hypotheses based on their experimental design. One conceptual step is therefore missing in the introduction, which has to combine present knowledge on the relationship between different frequency bands and present knowledge on how attention influences frequency-specific activation in the visual system to then make statements about which analyses can be performed to test which aspect of the model.

- Concept & Methods: With respect to both the concept and the analyses, what is missing is taking into considerations the brain areas that were investigated. Wheres in the abstract the authors only mention "within brain region connectivity" and "between brain region connectivity" also in the Methods section there is no clear relation to the anatomical areas that were investigated, being V1, V2 and V3. The authors rather classify the areas as "high level" and "low level" where V2 is sometimes classified as high-level and sometimes as low-level. The data are therefore not investigated with reference to the anatomy of the visual system. In my view, it would be beneficial if all analyses could be performed with respect specifically to V1-V2 connecitivity and V2-V3 connectivity as well as V1-V3 connectivity so that the specific anatomical interrelations are taken into account. Also, the authors should develop a conceptual framework of how layer-specific attention-driven connectivity changes should influence the visual cortex, and why.

- Methods. Given the missing conceptual overview over how attention-induced changes in EEG frequency bands should influence laminar connectivity in the visual system, also the methods lack a clear analyses strategy. The authors computed one correlation between power level of different frequency bands and connectivity between different brain areas without providing an explanation of which question this analysis addresses. The offered results therefore seem random to me, without a clear relationship to an investigated hypothesis.

- Methods. The authors mention that they only analyzed the strongest two connecting vertices within a layer, which was done to improve SNR. In my view, for a connectivity analyses, this is not valid, as it can bias the effect towards superficial connectivity where the SNR and thus correlation is always higher.

- Methods. The authors report 21 correlations in cortical depth, where their resolution allows to only sample perhaps 2-3 data points. The correlation analyses are therefore oversampled, which influences the statistical results. I would suggest to first run a component analyses across cortical depth, and to then correlate independent components to one another to investigate independent data points.

- Methods. The authors refer to their previously published paper with respect to the methods, and do not give any speficiations on the image sequence, image resolution, and image processing in this paper. In my view, all basic methodological steps that are critical to understand the paper should be described here.

- Results. The figure captions are too short and do not explain the presented data in an appropriate way. In Figure 1, details on the calculated contrasts, number of participants investigated, sampling and analyses methods should be given that allows interpreting the data. Also, it would be beneficial to explain the attention paradigm in a bit more detail in the figure caption so that panel A can be interpreted. In Figure 3, more details should be given on what data are shown, particularly for panel C where the only information given is "attention effect on laminar connectivity" with no further axes labels.

- Results. I do not fully understand the results as shown in Figure 3. As those form the major part of the manuscript, this needs revision. As said before, I think that the figure and results section would benefit from region-specific data analyses and presentation, but also clear axes labels are needed to allow interpretation of the data. Also, when I interpret the data correctly, correlations are done for altogether 21 different cortical depth, which would not be valid because of artificially inflating the number of correlations, as pointed out above.

Reviewer #3 (Public Review):

Scheeringa et al. examine laminar connectivity using fMRI and its relationship to neuronal oscillations. The results presented here are a re-analysis of a previously recorded dataset. The novelty here is the analytic technique used to relate laminar connectivity to rhythms. The main result, which is convincingly demonstrated, is that task modulations of deep-to-deep layer connectivity between areas is related to beta power. This is interesting because other studies (including studies by the authors) have found a negative relationship between fMRI BOLD and beta power within an area. In addition, studies in animals have found something similar: that while beta can decrease in a given area that is involved in a task, it can simultaneously increase in synchronization between areas. The present observation provides a link between these animal studies and non-invasive studies in humans. In addition, this work provides an important bridge between laminar connectivity and oscillations, previously only accessible with invasive recordings. Therefore, in the present study Scheeringa et al. push the field of neuroimaging into an exciting, more mechanistic direction, which promises insights into how cognition modulates laminar circuits.

However, a weakness of the technique as currently presented is that patterns of connectivity are only related to oscillations across subjects. It would be more powerful to examine whether the current network state (estimated by trial-by-trial power estimates) relates to laminar connectivity within subject. This would indeed speak to the nature of neuronal communication, which takes place on a moment-to-moment time scale, and which is not reflected in the current analysis. This may explain why laminar patterns of fMRI connectivity were not found to correlate with gamma-band oscillatory activity. In addition, the negative effects of attention on fMRI connectivity itself are somewhat puzzling. This may related to the limitations of the task design which do not perfectly separate attention vs. arousal/expectation, as the authors readily discuss.

Reviewer #1 (Public Review): 

Hülle cells are formed by fungal species of the genus Aspergillus such as Aspergillus nidulans. These cells are unusual because they surround cleistothecia, the fruiting bodies of ascomycetes. Until now, the current hypothesis is that they mainly serve to nurse cleistothecia during development. It was a matter of debate whether this is the real function of these cells or whether they have additional functions. In the current manuscript the authors provide compelling evidence that these cells have a strong ecological impact as they contain specific secondary metabolites (SMs) helping the fungus to "withstand" the attack by fungivorous animals like springtails and furthermore, these SMs inhibit sexual reproduction of other fungi. This way the results of this manuscript have a major impact on our view on the development and ecology of fungi and at the same time also contribute to unravel the function of SMs. 

Overall, the study is novel, highly informative with regard to fungal development and the ecology of fungi, and very well presented. The biology of Hülle cells is fascinating and there is only little information about this type of cells. In this paper the major function of Hülle cells was discovered. The introduction provides an excellent overview about the topic, methods are described in sufficient detail, results are clearly presented, most Figures are of high quality and in the discussion, the results are comprehensively interpreted.

Reviewer #2 (Public Review): 

The submitted manuscript sought to show that xanthones produced by Aspergillus nidulans are associated with Hulle cells that nurse cleistothecia. The authors present compelling evidence that xanthones are produced in Hulle cells during sexual development of A. nidulans. This raises important ecological questions about what important ecological interactions these compounds may have with cleistothecia. The authors investigate potential roles for these compounds in protecting the fungus from fungivory and in suppressing the growth of competing fungi. While these experiments do point to possible ecological functionalities of these compounds, more work is needed to clarify if these experiments represent interactions that could occur in nature and how such interactions might ultimately impact the fitness of A. nidulans. The current ecological experiments do, however, point toward interesting possible functions of these compounds, offering rare direction to investigate the ecological function of a fungal secondary metabolite.

Reviewer #3 (Public Review): 

There are two major strengths this reviewer can find in this manuscript. First, the authors suggest a novel role of Hülle cells in providing benefits in the niche competition against other fungi and in protection of reproductive structures from animal predators by accumulating certain secondary metabolites. Despite the initial identification of Hülle cells in the late 19th century, previous studies on Hülle cells only suggested that these cells actively support and protect the developing fruiting bodies. It has remained largely speculative what advantage the energy-consuming production of Hülle cells confers or how this cell type contributes to the success of aspergilli. In this manuscript, the authors have elucidated an important protective role of Hülle cells ensuring the long-term survival of A. nidulans in the semi-wildlife situation. This discovery could be perceived as a major breakthrough in expanding the knowledge of fungal biology, which can be included in textbooks for undergraduates to fungal experts. Moreover, these findings can provide insights into various fields; for example, these chemicals can be extracted from fungi or chemically synthesized and utilized as an insect repellent or fungicide. Second, the way how they draw conclusions is also the main strength of the manuscript. The authors have extensive expertise in metabolomic research and have revealed which proteins encoded by the mdp/xpt genes are exclusively located in Hülle cells and sexual hyphae by utilizing the GFP fusion technique and quantifying levels of each protein. Furthermore, they performed extensive metabolomic experiments on the mdp/xpt cluster metabolites with wild-type and all single null mutants of the cluster and unveiled the metabolite accumulation pattern of individual strains. Based on the results from these metabolomic data, the authors designed fungal and animal experiments to observe a potential role Hülle cells during interaction. This extensive metabolomic work is thorough and precise and the data support their conclusions well.

Reviewer #1 (Public Review): 

This manuscript describes extensive phenotypic analyses of Flo11, a multifunctional adhesin in S. cerevisiae. Flo11 mediates a variety of adhesin-related activities including flocculation and adhesion to surfaces, and these activities can lead to biofilm and pellicle (floating biofilm) formation, as well as ability to mediate agar invasion. Flo11 activities are activated after shear force, which facilitates surface amyloid formation. Flo11 is highly diverse, with different yeast strains expressing alleles with sequence and repeat number differences. These differences contribute to differences in adhesion behaviors. The alleles have conserved regions including a secretion signal, a globular N-terminal region, and a C-terminal GPI anchor that mediates cell surface attachment. In between these regions are ~1000-1500 amino acids with variable sequence and repeat numbers and a variable number of potential amyloid-core sequences. The manuscript includes extensive sequence analysis of alleles from 4 strains, and analysis of the consequences of expression of different regions of the protein from various alleles. This major study includes detailed AFM analyses showing that a Flo11 allele from a wine-fermenting strain expression leads to presence of cell surface patches of Flo11 adhesin 10-100 nm diameter 10-20 nm elevation from the cell surface. SiN AFM probing determines both adhesive strength and stiffness of the adhesion patches. The regions containing the amyloid-core sequences are essential for most of the Flo11 activities, because the activities are inhibited by deletion of amyloid-forming core regions or by treatment of intact cells with a general amyloid inhibitor or a sequence-specific anti-amyloid peptide. 

The work is extensive and well-documented. Qualitative data on the different constructs shows how activity correlates to specific sequences in the protein. However, the analysis is compromised by the lack of data on the levels of surface expression of the different alleles and constructs. mRNA levels are reported, but for fungal adhesins these values often do not correlate with surface expression levels. The problem is especially confounding for the C-terminal deletion construct, because the deleted region includes the GPI addition signal: its deletion leads to secretion of free adhesin into the medium and decreased surface anchorage (Douglas et al. 1996 Eukaryot. Cell6:2214-2221). 

The quantitative biophysical conclusions would be strengthened if the data from all tested cells were aggregated and presented, not just the data from the individual cell shown in each figure. 

If these critiques are addressed, the manuscript would greatly benefit from a summary figure and paragraph describing the activities attributed to each region of the sequence in L69 and other alleles. It would also benefit from reference to recent work demonstrating that cell-cell adhesion can be mediated by formation of amyloid-like bonds between cells.

Reviewer #2 (Public Review): 

The authors argue that the Flo1 protein of yeast can mediate cell-cell aggregation through amyloid formation. They identified a wine-making yeast strain that is particularly strong in this phenotype and this correlates with the presence of aggregation prone repeat regions in the protein. Their frequency seems to be important for the effect. 

The paper makes an extensive case, although I have reservations on experimental specifics. They show AFM imaging of the cell surface (nanodomain formation) and extensive genome editing to alter the suspected regions, which largely (but not perfectly produces effects in line with the hypothesis). Also, they use an inhibitory peptide (not so well validated) and a widely used amyloid dye to disturb these amyloid formation. Taken together these data point toward a role for amyloid formation, similar to what was previously shown by Lipke for candida. However, I remain with some concerns regarding data, controls, interpretation that preclude firmly backing the story. 

Major Concerns:

There is no direct evidence of amyloid formation in situ. 

The hydrophobic cluster analysis in figure 1 is not very clear - it is almost impossible to make anything out in those graphs 

The term beta-aggregation potential is used throughout and not defined. I presume this refers to the TANGO score? You should call it TANGO score then. <br> Where was the arbitrary cutoff of 70 for TANGO aggregation propensity determined? In itself it is only a prediction, not a demonstration of amyloid potential. To do that you need to make the peptide and show it makes amyloid (and that this can be suppressed by thioS and your breaker peptide, see below) 

line 124: how many cells were analysed, is this pattern present in the population? analysing a single cell would seem insufficient to me - this comment applies to several analyses below. <br> line 124: "this apparent roughness can be account for by proteins" - there is no data to support this, right? could be anything at this stage, it is just patch. I my mind you would need AFM-IR to ensure that you are actually observing a protein structure. This would immediately show you if its in the amyloid state as well. 

line 134: possibly due to my ignorance: how do you conclude from those forces that one is the hydrophobic interaction and the other is protein unfolding? comes as a complete deus ex machina to me. 

line 151 (also mentioned later): the 'anti-amyloid peptide' is taken with a lot of faith: how do we know it works by disrupting amyloid? At least show it works on the peptide level (Tht curves) <br> (again the AFM-IR measurement would solve this in one single effort) <br> This experiment at least needs a control to show the cell is otherwise intact and other structures are still present. <br> Designing potent amyloid disrupting peptide is not trivial - see various papers on the topic by David Eisenberg. I find it hard to believe these subtle mutations achieve all that they are claimed to achieve. 

line 157: Maybe I am confused, but why would thioflavin S destabilise amyloid? it is widely used as an amyloid-specific dye, and I would expect its binding energy to stabilise the amyloid state. Later, around line 310, the compound is called a drug, but really, it is just a rotor dye with an affinity for amyloid. Do other amyloid dyes show similar effect (oligothiophenes, congo red, curcumin)? 

line 250: are we really comparing single cell instances of each construct? 

Suggestions: 

If the hypothesis of the authors is true, it should be possible to replace the amyloid domains with synthetic ones (STVIIE eg, Serrano and co-workers), or from another protein. Once you have sufficient of these, you should see the cell-interactions etc. 

Also, introducing structure breaking residues in the repeats, like proline should stop the effect and would give strong support for the amyloid nature of the interactions over general hydrophobic patches.

Reviewer #3 (Public Review): 

The authors observed nanodomains in FLO11 expressed S. cerevisiae strains when imaging the cell wall of the cell with a bare AFM tip. The adhesion to the tip as well as the stiffness of these domains were characterized using AFM. However, there was no direct proof/confirmation that these nanodomains were actually composed of Flo11 proteins. 

Previous research showed that Flo11p trans interacts and this trans interaction between cells could be (hypothesis) preceded by a cis interaction between the Flo proteins via beta-aggregation-prone amyloid-forming sequences. It seems that the obtained results could explain this trans-interaction (i.e. the "nanodomains"). However, in this model the main interaction is based on the Flo11 A-domain interaction that is responsible for the trans-interaction. This could not be confirmed in the present work where the cell-cell interaction results were only based on (qualitative) microscopic observation of cell aggregates. Additional, it has been previously shown that homotypic Flo11 A-domain interaction is pH dependent since only at low pH these domains hydrophobically interact. It seems that the current experiments were performed at a too high pH.

Reviewer #1 (Public Review): 

T cell-independent type 2 (TI-2) antigen represents highly repetitive multivalent molecule such as bacterial capsular polysaccharides and viral capsids. Although B cell receptor play role in the center of antibody response to TI-2 antigen, necessary signaling molecules for isotype switched antibody production are unknown. In the present work, Fukao et al. first described that IL-1 and IFN-α promote TI-2-induced class switch recombination (CSR) to IgG3 primarily. They identified that Protein Kinase Cδ (PKCδ) and downstream transcriptional regulator, BATF, play critical role in IgG production stimulated by TI-2 antigen as well as commensal bacteria. Present study must be important work in the B cell biology field providing valuable data to understand immune response occurred against TI-2 antigens. Overall conclusions seems appropriately led. Considering that PKCδ is known as a key signaling molecule suppressing autoimmune diseases, results may suggest possible connection between TI-2 response and autoimmune diseases. Although following points could be a future issue, weakness of this study are, 1) Why and how PKCδ is required differently downstream of TI-2 BCR crosslinking but not in TD antigen exposure has not been addressed. 2) How IL-1α/β, IFNα or TLR ligands synergistically regulates Aicda gene expression together with BATF is still obscure.

Reviewer #2 (Public Review): 

The ligation of the B cell antigen receptor (BCR) by multivalent and repetitively arranged haptens can elicit the production of class-switched Ig isotypes such as IgG without the necessity of T cell help. How exactly these so-called T cell-independent type 2 (TI-2) antigens bypass the need of co-stimulatory signals through CD40 or other co-receptors in the course of T cell-dependent responses is unclear. One hypothesis states that based on their multivalency, TI-2 antigens cluster the BCR more potently compared to T cell-dependent antigens, and hence, the different signaling capabilities of the two types of B cell antigens might be quantitative rather than qualitative. The work by Fukao and colleagues challenges that view by providing compelling evidence for a more unique and specific role of protein kinase C-delta (PKC-delta) in TI-2 B cell responses. This conclusion is based on carefully controlled sets of in vitro and in vivo experiments showing that loss of PKC-delta expression in B cells basically abrogates class-switch recombination to and production of IgG3 antibodies associated with compromised protection against bacteremia. The key function of PKC-delta in TI-2 responses appears to be the induction of the transcription factor BATF that is critical for the expression of the Aicda gene, encoding the class-switching factor Activation-induced Cytidine Deaminase (AID). 

Clear strengths of this work are the well-performed studies to monitor B cell functions at all levels by using appropriate mouse models e.g. measuring antibody production in response to hapten immunization or bacterial challenge, or monitoring BCR signaling events and gene expression profiles. The combination of these different read-out systems is sound and renders the data convincing. What is less clear or could be improved is the working model of how BCR ligation by TI-2 antigens translates into PKC-delta activation (Figure 7) while that by T cell-dependent antigens does not or is at least less efficient in terms of kinetic and extent of PKC-delta activation (Figure 2A and others). And what exactly then is the role of secondary stimulation? The presented data and the discussion are a bit weak on these aspects as well as on the obvious role of marginal zone B cells, which represent the main source of IgG3 and possess a particularly low activation threshold. To this reviewer, these aspects are not just additional details that would be helpful to assess. These points address the original and basic question raised above, namely, is it the quantity or the quality of BCR signal output (or BCR signal strength) that accounts for differences in biological B cell responses? 

The concept provided in this manuscript on the role of PKC-delta as a decisive BCR signal effector that "distinguishes" TI-2 from T cell-dependent B cell responses is intriguing and will stimulate further discussions on this important aspect of humoral immunity.

Reviewer #1 (Public Review): 

In this manuscript, Piccolo and collaborators provide a detailed overview of the Common Workflow Language (CWL) for beginner bioinformaticians, and perhaps even older ones that may not be that up-to-date with the latest developments in reproducible research. They also provide a webpage, ToolJig, to create CWL documents without needing to install any software nor learn the specifications of the format in much detail. Written in the form of a tutorial, its major strengths are that explanations are very clear, and are accompanied by illustrative figures and examples in their Github repository. I do not see any major weaknesses that need to be fixed. As science is currently undergoing a major reproducibility crisis, I think that it is crucial that detailed and accessible pieces such as this one are published to teach scientists to create fully reproducible code. If CWL is adopted widely then I believe that it may help alleviate these issues.

Reviewer #2 (Public Review): 

The paper provides working examples of various bioinformatic pipelines written in CWL. The aim is to provide a general introduction by example to the CWL intended for researchers with limited programming skills. The authors also created the web application ToolJig intended to facilitate the generation of CWL in an interactive way. 

Strengths:

- Working examples of useful bioinformatic pipelines written in CWL can provide a quick template for people to start on their own CWL projects. <br> - Enfasis of Docker and containers as the building blocks of pipelines that are portable and reproducible. <br> - ToolJig application as an interactive aid for unexperienced users for building their CWL documents. 

Weaknesses:

Confusion regarding the usage of containers in regards to the location of the workflow manager ( production vs publication workflows). A container with all required analysis software, CWL document and a workflow manager seems well suited for distribution with a publication for reproducible calculations. However for production needs, a modular design with various containers and a workflow manager outside of the containers seem a better choice. It's hard to distinguish these two usages in the manuscript.

Reviewer #1 (Public Review): 

Histone modifications play diverse roles in regulating DNA based transactions. Understanding the biochemical mechanisms underlying such roles has been greatly enabled by the ability to reconstitute chromatin templates with defined modifications. Poly-ADP-ribosylation (poly-ADPr) is a major modification that plays critical roles in DNA damage. However a few features have slowed careful mechanistic dissection of the role of poly-ADP-ribosylation. These features include that (i) the modification is found on multiple proteins such as histones and the PARP enzymes that deposit the modification and (ii) there is heterogeneity in terms of the poly-ADP-ribosylation chain length and the amino-acid location of the modification. Additionally, unlike with other histone modifications such as acetylation, methylation and ubiquitylation, generating homogenously modified poly-ADPr chromatin in vitro has been technically challenging. Here the authors generate nucleosomes with homogenously and site-specifically modified poly-ADPr marks through a series of readily separable steps. Using these modified nucleosomes the authors test the impact of the poly-ADPr modifications on the activity of the chromatin remodeler ALC1. Previous work has shown that ALC1 is activated by binding poly-ADPr chains that have 3+ units. However, it has not been clear which modified substrates are being recognized by ALC1. Using their site-specifically modified reagents the authors show clearly that poly-ADP-ribosylation of H2A (and H3) has a much larger effect on stimulating nucleosome remodeling than auto-ADP-ribosylated PARP. 

Overall, the work in this resource article is carefully carried out with several thoughtful controls and provides a highly enabling new method for the chromatin field. What further raises the impact of this work is the use of this new method to make a significant new mechanistic finding about ALC1regulation. I think the work in this resource article will be of much interest to the chromatin community and those studying DNA damage.

Reviewer #2 (Public Review): 

This interesting manuscript from Liszczak and coworkers highlights the strength of a semisynthetic chemical approach to obtain both monomeric and well-defined polymeric ADP-ribosylated histones H2B and H3. The authors have applied these protein reagents toward dissecting the role of ADP-ribose in facilitating nucleosome remodeling by the macrodomain-containing remodeler ALC1 in the presence and absence of the PARP1 polymerase. The authors demonstrated a clear preference of ALC1 for longer chains of ADP-ribose and also that ADP-ribosylated PARP1 can stimulate ALC1 activity on nucleosome substrates. Finally, they demonstrated that ADP-ribosylated nucleosomes are reasonable substrates for remodeling in nuclear lysates, and that ALC1-knockout cells do not efficiently remodel ADP-ribosylated nucleosomes suggesting that it is the major remodeler of ADP-ribosylated mononucleosomes. Thus, the manuscript is an elegant demonstration of the utility of histone semisynthesis in testing biochemical mechanisms, such as chromatin remodeling, modulated by histone post-translational modifications. The minor weakness of the current method is that all experiments were done with mononucleosomes alone, which likely do not reflect the distribution of ADP-ribosylation in chromatin. However, some key observations, such as the fact that ADP-ribsoylated PARP1 supports nucleosome remodeling by ALC1 and that ALC1 preferentially remodels mononucleosomes modified with longer ADP-ribose chains are noteworthy and expand our understanding of the roles for histone ADP ribosylation in critical processes such as DNA damage in cells.

Reviewer #3 (Public Review): 

In this manuscript, the authors have developed a PARP1-dependent semi-synthetic strategy to generate mono and poly ADP ribosylated nucleosomes at specific histone serine residues. The authors find thatPARP1/2 interacting protein HPF1 restricts PARP1 activity to favor mono ADP ribosylation and that PARP1 can generate poly ADP ribosylated proteins after HPF1 is removed from 'primed' mono ADP ribosylated substrates. This chemoenzymatic strategy, a close derivative of previously published approach for enzymatic ADP ribosylation, enables generation of hetereogenous PAR chains on peptide substrates, which can be purified to homogeneity by chromatography. The modified peptide is then ligated, using canonical native protein ligation strategies, to the protein of interest. As such, the methodology is well suited for N-terminal modifications of proteins lacking Cys. While this limits broad application of the method, it is well suited for the preparation of modified histones and nucleosomes used in this study. 

Using ADP ribose (ADPr) modified histones, the authors have reconstituted a numbed of nucleosomes with specific lengths and defined sites of ADP ribosylation. These substrates were used to investigate how the chromatin remodeler ALC1 responds to defined lengths and positions of ADP ribosylation, either on nucleosomes or on PARP1 itself. The key finding of the study is that the remodeling activity of ALC1 is dramatically enhanced by poly ADP ribosylation (poly = 4,5), but not mono ADP ribosylation on its nucleosome substrate. This enhancement is greater than that provided by auto ribosylated PARP1 binding to ALC1, which has been previously shown to relieve autoinhibition in ALC1. Furthermore, the stimulatory effect is specific to ALC1 and not to other remodelers tested. Authors further show that the stimulatory effect observed with purified components is also present, albeit to an attenuated extent, in nuclear extracts. Lack of remodeling of modified nucleosomes by extracts of ALC1 knockout cells further implicates this enzyme in remodeling of ADPr modified nucleosomes (of note, this point requires additional analysis - see additional suggestions section). 

While the authors show engagement of the ALC1 macrodomain with polyADP ribosylated histone peptides, the link between macrodomain engagement and enhanced remodeling by the presence of polyADP ribose on the nucleosome substrate is not clearly established. Use of macrodomain mutants that abrogate binding of ADPr, or remodeler deletion constructs would allow for better assessment of whether macrodomain-ADPr engagement is necessary and sufficient for the observed stimulation of the remodeling activity. This would help exclude alternative models by which ADPr of histones can promote remodeling (eg, by potentially impacting chromatin structure or DNA accessibility that might be uniquely suited for ALC1 rather than other remodelers), and might shed light on how ADPr of H3 and H2B yield similar effects, despite different topological constraints. 

Finally, manuscript would benefit from additional contextualization of these findings within the existing knowledge in the field of DNA damage response (DDR). Better understanding of mechanisms that control mono vs polyADP ribosylation (and control chain length for polymerization), nature of HPF1-PARP complex and whether it is constitutive, as well as any prior knowledge that may link HPF1 and ALC1 activity, would help readers further evaluate significance of these findings.

Reviewer #1 (Public Review): 

This is an excellent paper that refines our understanding of how FMRP functions in cortical neurons to shape their synaptic development. The authors use viral channelrhodopsin to selectively activate callosal synapses and find that they are weaker onto L2/3 and L5 pyramidal neurons in the mouse somatosensory cortex. Using sparse deletion, the effect is found to be cell autonomous. Under these conditions it occurs during late development (P23-30) when these synapses are normally being refined and strengthened. The effect is believed to be due to a retention of so-called "silent" synapses which express NMDA receptors but not AMPA receptors, since the input via NMDA receptors is normal and the analysis of evoked quantal events using strontium reveals a reduction in frequency but not a reduction in amplitude. However the existence of silent synapses is not demonstrated directly and the change in frequency observed is small relative to the change in overall callosal synaptic drive. The authors address a prior finding that loss of Fmr1 weakens synapses from L4 neurons onto L2/3 neurons. They find that this does not occur cell autonomously or with postnatal deletion, but does occur in the germline knockout. The authors also find that sensory deprivation via whisker trimming can normalize the callosal input when applied to the whiskers providing the majority of input to the neurons lacking Fmr1, but not when applied to the whiskers providing the majority of input to the contralateral neurons providing the callosal inputs to the KO neurons. This would seem to implicate Fmr1 in an activity-dependent competition between local and long-range synapses. Finally, the authors have reanalyzed prior bilateral EEG recordings in germ-line KO animals and find reduced interhemispheric synchrony, consistent with reduced callosal function. Although these changes could also reflect other effects in the KO brain. 

The experiments are convincing and are well illustrated and analyzed. Taken together they suggest a key role for the regulation of translation by Fmr1 in the activity-dependent refinement of callosal synapses, and likely by extension, many long range connections in the brain. This role could contribute to intellectual disability following loss of Fmr1 function and the sensitivity of this activity-dependent refinement process may explain the frequent observation of callosal abnormalities in other forms of autism spectrum disorders and intellectual disability.

Reviewer #2 (Public Review): 

Previous studies showed reduced structural and functional connectivity of two hemispheres in the autism spectrum disorder (ASD), but little is known about its cellular mechanism. This paper tried to fill this knowledge gap using a mouse model of ASD. By combining optogenetics, slice electrophysiology, and Fmr1 gene knockout (KO) approaches (a leading monogenic cause of ASD), the paper demonstrated that callosal inputs to L2/3 pyramidal neurons in whisker somatosensory cortex is reduced in Fmr1 KO mice/neurons compared to wild-type mice/neurons. This reduction was due in part to the selective reduction in AMPA receptor-containing synapses and was restored by sensory input deprivation. The local circuit connection was unaffected in a sparse Fmr1 KO mouse. The paper also recapitulated previously reported the reduced coherence between two hemispheres in vivo. The data is a welcome new addition to the previous studies concerned with circuit abnormality in Fmr1 KO mice and supports the paper's main claim. 

The strength of the paper is the simultaneous comparison of the callosal inputs to L2/3 neurons with or without the Fmr1 gene in the same brain slice. This directly demonstrated the role of the Fmr1 gene on the formation of the callosal synapse, a key piece of information that could guide future basic, translational, and/or therapeutic studies. 

The weakness of the paper is the sparse Fmr1 KO model. An apparent delay in onset of Fmr1 KO effect and unclearness in the extent of Fmr1 KO achieved in one hemisphere makes comparison with global Fmr1 KO model (in this study or previous studies) and assessment of data interpretation made in the paper difficult.

Reviewer #3 (Public Review): 

This is an expertly-carried out study that rigorously examines the effects of global and post-synaptic FMR1 deletion on excitatory synaptic function in primary sensory cortex. These data add to a growing literature about the effects of FMR1 on different subnetworks in the brain, including that from the authors' own labs. Results indicate that L5 neurons show hyperconnectivity within the layer in the global KO, but the postsynaptic effects of FMR1 deletion are only transiently manifested at the L4 to L2/3 connections, and local connections between L2/3 neurons are not affected. Interestingly, callosal inputs show selective impairment into the 4th postnatal week, and mosaic knock-out of FMR1 indicates that this effect is postsynaptic in origin. Although the degree of specificity by which this gene can influence different types of excitatory synapses is provocative (i.e. short-range inputs appear mostly normal in superficial layers but long-range callosal inputs are depressed), the mechanism for this selectivity remains obscure. Although the authors suggest that this reduction in synaptic strength can drive reduced coherence in the interhemispheric EEG, important controls for the analysis of EEG coherence and the circuit relationship between callosal inputs and the EEG are missing, such as the effect of FMR1 KO on inhibitory neurons, as well as lack of correspondence in age- and brain region between the synaptic and EEG studies.

Reviewer #1 (Public Review):

Sleep staging is the key first step in the analysis of human sleep architecture which is fast becoming an important medical tool. Currently scoring is overwhelmingly manual but recently machine learning techniques has led to algorithms with human level performance. They are as yet not widely used in the clinical setting. The current manuscript aims to remove some of the stumbling blocks for the use of automated methods by providing an easy to use, well-validated algorithm that relies on polysomnographic features familiar in sleep literature so as to facilitate interpretation of results.

The paper clearly shows that the algorithm performs within the range of the current state of the art for automated sleep scoring and is unbiased with regard to race, sex and age. It provides demanding cross-validation by using a completely external dataset for performance evaluation. The algorithm seems easy to use and is supplied with a well-documented code that should facilitate its application as the authors intended.

Establishing whether automated methods tend to make similar "errors" as human scorers or whether they display different behaviour is essential if a transition is to be confidently made from one technique to the other. This is not systematically explored in the manuscript. A key advantage of the automated approach is the provision of a confidence measure for each epoch. However how this relates to human inter-scorer variability and accuracy is not fully explored and so the potential of this measure is not clear.

Reviewer #2 (Public Review):

1. The authors describe their algorithm as a tool that (i) was validated "across heterogeneous populations around the world"; (ii) has an "accuracy matching or exceeding human accuracy"; (iii) "is easy to use". I take issue with these three statements.<br> First, the authors did not test the performance of their algorithm in clinical populations with sleep disorders, despite the fact that individuals with sleep disorders represent (logically) the vast majority of sleep recordings. Crucially, such a comparison was made in the best (to my knowledge) published automated sleep staging algorithm (Stephansen et al. Nature Communications 2018, doi: 10.1038/s41467-018-07229-3). The omission of this work is very surprising. Quantifying the impact of sleep disorders on a sleep scoring algorithm is critical for its deployment in sleep clinics.<br> Second, the authors wrote that their algorithm is "matching or exceeding" human accuracy but seem to present uncorrected one-to-one comparisons to support their claim. The fact that an algorithm is better than some humans do not mean it exceeds human performance.<br> Third, although I agree that the tool seems easy to use even for individuals with limited programming skills, it still requires some. I don't think someone who is used to software with graphical interfaces and who has never used (or heard of!) python would describe the tool as easy to use. This poses an important implementation challenge.

2. An important limitation of this algorithm is that it captures only one part of the visual examination of sleep data. Indeed, especially in clinical settings, the data is not only examined to establish the hypnogram but to also identify markers of common sleep disorders (e.g. sleep apnea, leg movements, etc). Although this algorithm could significantly speed up sleep scoring, it does not allow to detect these other important markers. Currently, and in link with the previous comment, the algorithm could not replace the visual inspection of the data for clinical diagnoses.

3. The data were curated with some recordings or portions of recordings being excluded (see p. 7). While I understand that this curation is important for the training set, I think it should not be applied to the test set. Indeed, it goes contrary to the logic of automating sleep staging. For example, cutting the beginning and end of the recording according to sleep start and end (p. 7) supposes that the start and end of sleep are already known (i.e. it has already been scored).

4. Two types of EEG derivations were used (C4-M1 or C4-Fpz). Was the performance impacted by this variable? Is it fair to assume that the choice of features (spectral features or summary statistics of time series data) could explain the absence of differences but that introducing new features (i.e. phase-sensitive features) could increase the influence of the choice of the derivation?

5. Given that markers of sleep stages are very different in EOG, EMG and EEG time series, could the authors explain the logic behind applying the same pre-processing and extracting the same features on these three very different types of data? Could this explain why the majority of the features in the top-20 features were EEG features?

6. Sleep scoring guidelines incorporate not only what can be observed on a given epoch of data but also what is observed in the previous epoch(s). For example, an epoch can be scored as N2 even if there is no marker of N2 but there was (i) a marker of N2 in a previous epoch, (ii) no reason to change the score since. To reproduce this, the authors employed a symmetrical smoothing approach (a combination of a triangular-weighted rolling average and asymmetrical rolling average). Why did the authors choose to incorporate data from following epochs, which is not implemented in established guidelines? How was the duration of the smoothing window chosen? Indeed, 5 minutes appear as rather long could explain the poor performance of the algorithm for fast changing portions of the data (i.e. N1 or transitions). Importantly, these transitions can be very relevant in clinical settings and to establish a diagnosis.

Reviewer #3 (Public Review):

This study presents a new sleep scoring tool that is based on a classification algorithm using machine-learning approaches in which a set of features is extracted from the EEG signal. The algorithm was trained and validated on a very large number of nocturnal sleep datasets including participants with various ethnicities, age and health status. Results show that the algorithm offers a high level of sensitivity, specificity and accuracy matching or sometimes even exceeding that of typical interscorer agreement. The conclusions are supported by the data. Importantly, a measure of the algorithm's confidence is provided for each scored epoch in order to guide users during their review of the output. The software is described as easy to use, computationally low-demanding, open source and free.<br> This paper addresses an important need for the field of sleep research. There is indeed a lack of accurate, flexible and open source sleep scoring tools. I would like to commend the authors for their efforts in providing such a tool for the community and for their adherence to the open science framework as the data and codes related to the current manuscript are made available. I predict that this automated tool will be of use for a large number of researchers in the field.<br> However, there are plenty of automated sleep scoring tools already available in the field (most of them are not open source and rather expensive, as noted by the authors). The current work does not provide a clear view on whether the new algorithm presented in this research performs better than algorithms already available in the field. No formal comparisons between algorithms is provided and the matter is not discussed in the paper.

There are some overstatements in the manuscript. For example, the algorithm was trained and validated on nocturnal sleep data. Sleep characteristics (eg duration and distribution of sleep stages etc.) are different, for example, during diurnal sleep (nap) and the algorithm might not perform as well on nap data. As such, the tool might not be as "universal" as stated in the title. Additionally, as human scores are used as the ground-truth for the validation step, it might be misleading to state that "this tool offers high sleep-staging accuracy matching or exceeding human accuracy". The algorithm exceeded the accuracy of some human scorers and matched the scores of the best scorer.

No reflection on further improvement is offered in the paper. The algorithm performs worse on N1 stage, older individuals and patients presenting sleep disorders (sleep fragmentation) and it is unclear how this could be improved in future research. In the same vein, the current work does not present performance accuracy separately for healthy individuals and patients when it is expected that accuracy would be poorer in the patient group.

There is series of methodological choices that is not justified. For example, nights were cropped to 15 minutes before and after sleep to remove irrelevant extra periods of wakefulness or artefacts on both ends of the recording. This represents an issue for the computation of important sleep measures such as sleep efficiency and latency as the onset/offset of sleep might be missed. It is also unclear how the features were selected and a description of said features is currently missing. The custom sleep stage weights procedure is unclear. The length of the time window for the smoothing procedure seems arbitrary. Last, it is currently unclear when / how the EEG and EMG data were analyzed.

Reviewer #1 (Public Review):

This manuscript presents a systematic review and regression analysis to analyze the association between upper and lower respiratory tract shedding (URT and LRT) of SARS-CoV-2 and disease severity. In addition, the authors study the impact of the days from symptoms onset on shedding in the two compartments. Overall, the presented results provide an interesting synthesis of the literature on these issues.

Reviewer #2 (Public Review):

The study appears robust and comprehensive, and relevant quality checks for systematic review have been applied. The results are valuable and contribute to the scientific knowledge in this field.

Interesting findings include:<br> - Adult patients with severe disease had on average a somewhat higher upper respiratory tract viral load at 1 day from symptom onset than patients with non-severe disease. After this stratification for severity, respiratory viral loads did not differ significantly for age and sex. Rates of viral clearing were similar. Children and adults with non-severe disease had similar upper respiratory tract viral loads and viral clearance rates.

- High and persistent lower respiratory tract shedding of SARS-CoV-2 was associated with severe but not non-severe illness. The difference in lower respiratory viral load for severe and non-severe cases was more pronounced than for upper respiratory tract viral loads. In contrast to the upper respiratory tract, viral clearance from the lower respiratory tract was more rapid in non-severe than in severe cases. Again, age and sex did not differ significantly after stratification for severity.

- The authors then aimed to assess whether the observed difference in shedding in the first days after start of symptoms could be used to predict which people would develop more severe COVID-19. Typically, deterioration into severe disease only happens around 10 days from symptom onset. The authors conclude that upper respiratory tract viral shedding is so heterogeneous that its predictive capacity of disease severity is inaccurate. In contrast, lower respiratory tract shedding does have a predictive accuracy of up to 81% for disease severity.

Potential impact: Lower respiratory tract viral load could thus potentially be used as an early warning for developing severe COVID-19. However, lower respiratory tract samples are not routinely taken, the standard nasopharyngeal swab is an upper respiratory sample. Some discussion on the practical applicability of this suggestion could enhance the paper's impact.

Reviewer #1 (Public Review):

In 2007 the authors published a paper demonstrating that GDP-fucose, the donor substrate for fucosyltransferases (FUTs), is essential for Trypanosoma brucei growth in procyclic and bloodstream forms (Turnock et al., 2007). At that time, no FUTs or fucosylated proteins were known to exist in T. brucei. Here, using bioinformatic analysis, the authors identified a single FUT homologue in the T. brucei genome, TbFUT1, which belongs to the GT11 CAZy family of a1,2-FUTs. They expressed TbFUT1 in E. coli and demonstrated that it is a FUT that transfers fucose to acceptor substrates, preferentially with a terminal Galb1,3GlcNAc. Glycosidase digestions, mass spectral and NMR analyses demonstrated addition of fucose to the 2'-OH of the galactose in alpha-linkage. They generated conditional null (with a tetracycline-inducible promoter) alleles of TbFUT1 in both bloodstream (BSF) and procyclic (PCF) forms of the parasite. Deletion of TbFUT1 resulted in a significant growth defect in both forms when grown in vitro. Surprisingly, TbFUT1 localized to the single mitochondrion in T. brucei, and deletion of TbFUT1 disrupted mitochondrial activity. A homolog exists in Leishmania major (LmjFUT1), which has recently been reported to be essential for L. major growth (Guo et al. 2021). The authors showed that TbFUT1 rescued the growth defect in LmjFUT1 mutants, and also localized to the mitochondrion in L. major.

This study is the first to identify a FUT in T. brucei, and the first to identify a FUT localized to mitochondria in any species. FUTs are typically localized in the Golgi, ER, or as more recently described, in the cytoplasm, but this is the first demonstration of a FUT in mitochondria. Only one glycosyltransferase has been reported in mitochondria, O-GlcNAc transferase (OGT). OGT transfers GlcNAc directly to protein, so no additional glycosylation machinery is necessary. The fact that TbFUT1 prefers a Galb1,3GlcNAc acceptor substrate strongly suggests additional glycosyltransferases will also be localized to the mitochondria in T. brucei and L. major. Thus, these are highly significant and novel studies. They are well done and rigorously performed. A few minor suggestions for clarity and presentation of the data are needed.

Reviewer #2 (Public Review):

The authors used a nice combination of biochemical methodologies, including NMR and mass spectrometry, to work out the substrate specificity of this activity; they concluded that it mainly transfers alpha1-2Fuc to Galbeta1-3GlcNAc but not Galbeta1-4GlcNAc dissacharides. Interestingly, Galbeta1-3GlcNAc residues are highly abundant on surface glycoconjugates expressed by the procyclic (midgut) stage and also as terminal sugars on N-glycans from flagellar pocket glycoproteins expressed in the blood stages, but their mitochondrial presence has never been found in any organism. It remains to be determined what is the natural substrate of this enzyme and whether fucosylation indeed occurs in the mitochondrion of kinetoplastid organisms.

I only have few comments and no further experiments are requested.

1. Did the authors try to determine the exact location of TbFUT1 within the parasite mitochondrion, for instance using TEM immunogold? This could help understanding access to GDP-Fuc (although I believe the enzyme is facing the cytosol) and also the possible location of the natural acceptor molecule(s).

2. In relation with the previous point, given the challenges in trying to localise fucosylated glycans using fucose-specific lectins, I wonder if there is a precedent for detecting terminal beta-Gal residues on the trypanosome mitochondrion using lectins.

3. I found interesting that N-terminal tagging of TbTUF1 sends the protein to the Golgi apparatus. This seems like a great coincidence for a protein that normally would be predicted to be Golgi-resident, so I wonder if there is any identifiable Golgi targeting sequence within TbTUF1. Also, there was any attempt to localise the protein after deletion of the mitochondrial signal (no tagging)?

Reviewer #3 (Public Review):

A great strength of the manuscript is the meticulous, very convincing biochemical analysis of the recombinant trypanosomal fucosyltransferase. No surprise since it has been done by leaders in the field. Also the immunofluoresecence localization is convincing, within the limits of the technique. The authors went through the great effort of producing a monospecific antiserum against the fucosyltransferase which allows them to localize the natively expressed untagged protein. The conditional knock out cell lines for both cell cycle stages unambiguously demonstrate that the enzyme is essential for cell growth.<br> The manuscript is well written, has clearly composed figures and the results are presented in a logical way. I agree with the authors general conclusions. My major criticism is that I think both the localization of the enzyme as well as its physiological properties could have been analyzed in some more detail using the great tools the authors produced in their study (cell lines, antiserum). Having said that I am aware that characterizing the exact function of the protein might be beyond the scope of this study.

Reviewer #2 (Public Review):

In this study, the authors attempted to generate the classification of the human DRG neurons, and compare it to the organization of the mouse DRGs in an effort to better understand the species differences, and to help advance the translational aspect of pain research.

This is an interesting and timely study. It is also perform by some of the labs that either pioneered the use of human tissue for translational pain research (Davidson), or the identification of markers of sensory neurons (Ryba).

Overall, the authors achieved their goal and present a detailed description of the clusters of human sensory neurons. The results support their conclusions.<br> The database produced by this work will be of great help to the community of sensory physiologists.

One caution that readers must bear in mind is that access to human neurons is difficult, and the sample sizes remain unfortunately low. For instance, one mouse DRH snRNAseq study referred to in this work is based on 141,000 nuclei. The human data, comes from around 1,800 nuclei. Therefore, the work provides a great clue to the clusters of sensory neurons in human DRGs, but this clustering may be refined down the road when more data becomes available.

Reviewer #1 (Public Review):

In this work, Minh Nguyen and colleagues performed single nuclear RNAseq of human dorsal root ganglion neurons and classified human DRG neurons into 15 clusters. By comparing the transcriptomic features between human and mouse DRG neurons, they found that there are similar as well as human specific DRG neuron cell-types. Based on this analysis, they assigned the human DRG neuron clusters into different functional groups. In addition, the authors provided multiplexed ISH data to support their sequencing and classification results. Overall, this study is straightforward and the quantify of data is good. Given that the field is just starting to understand the human specific molecular signature of primary somatosensory neurons using single cell/nuclear RNAseq, this study is valuable and provides one of the first gene expression databases of individual human DRG neurons. Below are our critiques and suggestions for this manuscript.

Comments:

1. The authors enriched neuronal nuclei by NeuN antibody selection. However, no direct evidence was provided to show this method can unbiasedly select all types of DRG neurons. If it is not the case, this NeuN selection/enrichment might introduce the bias to some types of neurons and lose some other types.

2. The authors separated the non-neuronal from the neuronal nuclei in analysis, based on marker genes PRP1, MBP and APOE. From figure supplement 1, however, we still see the expression of APOE in quite a lot of selected neuronal nuclei. One possibility is that these transcriptomes are from more than one nucleus as the authors hinted in the figure legend. It would be the best if the authors could give an estimation about the percentage of multiple nuclei from their sequencing result. Since this is one of the first set of papers performing human DRG neuron single RNAseq, adding discussion about the strength and potential technical caveats of this method would be beneficial for the field.

3. The authors should provide a figure or figure panels to show the sequencing quality and depth. For example, the reads number, mapped rates, detected genes numbers, etc. Again, this would facilitate evaluation/comparison of different human DRG neuron sequencing methods and data.

4. The sequencing results show that Nppb is expressed in H10 and H11. Please add fluorescent in situ to validate the point.

5. The authors compared the human and mouse transcriptome data. For a fair comparison and for the readers' information, the author may need to provide some basic information about the two sequencing results, such as the cell number, sequence depth, sex, etc. A table like this would be nice.

6. In Fig. 4, they authors proposed that "transcriptomically related classes of human DRG neurons are spatially clustered in the ganglion" by showing staining of two sections. They provided some quantification in the supplementary figure. Overall, the data in supporting this claim is not strong, and it is not clear whether this is a human specific phenomenon or generally true in other species. Similar trends were noticed for the mouse small diameter vs large-diameter neurons as well, but no careful quantification and careful comparison were made. Depending on how strongly the authors want to claim the "spatial clustering" model, the authors may need additional quantification/modeling and perform similar analysis with mouse DRG neurons.

Reviewer #3 (Public Review):

The manuscript by Nguyen et al describes the assignment of neuronal cell types to human L4 or L5 DRG based on single nucleus sequencing and bioinformatic analyses of freshly isolated samples (1837 neuronal nuclei and an average of 2839 genes per nucleus). A bioinformatic comparison to mouse lumbar DRG single nucleus sequencing results by Renthal et al is also described. Additionally, the manuscript describes results of Hi-Plex RNAscope performed on L4 or L5 DRG from the same donors using 9 different probes. Lastly, the authors make an observation for a potential organization of NEFH and SCN10A expressing neurons in the DRG in situ that has not been reported in mice. The study uses DRG from one male and five female donors age 34-55. The importance of obtaining single nucleus or single cell molecular profiles from humans cannot be overstated. Properly directed therapeutic translation requires knowing this information. The authors have immediate access to freshly removed human tissue which is critical for obtaining high quality samples. Preceding this submission is the submission of a human single DRG gene expression study performed in situ in BioRxiv and the publication of a single nucleus study of macaque DRG. This study was only cursorily compared to those studies. A bioinformatic comparison to the macaque study would have a lot of value and a more in depth comparison to the other human study would also add value. The authors have created a searchable database free and easily accessible to researchers.

Reviewer #1 (Public Review): 

The authors have studied the visual-ocular response in subjects who were exposed to a static magnetic field of 3 tesla. This magnetic field strength is only encountered in the environment of a high field magnetic resonance scanner. The magnetic field induces a response in the vestibular system, and which is highly dependent on head orientation with respect to the magnetic field. The response is hypothesised to mimic that of a caloric vestibular stimulation event. The authors have constructed a very well-designed study to determine the bias in fully dark conditions following a specific visual cued tracking point. The eye movements were tracked with an infra-red camera to ensure completely dark conditions. Data is presented and statistical analysis given which supports the author's questions and hypothesis. 

In a historical context this article is a further development of the field of interest in the effect of magnetic fields on the human senses. Originally the questions were ones of bio-mechanism involved and safety of MRI. Now those questions are largely satisfied, the field moves to using the observed effects for neuroscience and even clinical applications (in their broadest terms). The authors show a distinct bias in visual tracking which could potentially be used as a model for stroke patients with spatial neglect. There are also implications for those studying default networks or visual tracking using fMRI. The effect measured here is still quite subtle at 3 T, but has implications for higher field fMRI at 7 T or even higher.

Reviewer #2 (Public Review): 

Overall, the study is well written, appropriately analyzed, and the methods are clear and well described. I appreciate the simple and straightforward differential in the study design, the authors attempt to maximize the stimulus within the confines of the head coil, and the presentation of the data. The magnetic field used here (3 Tesla) is commonly used in research and increasingly used in clinical MRI scans. The only major comment is that the relationship between the VOR and visual tasks was not analyzed-which seems to me to be a key relationship. A process of adaptation occurs in the MRI machine that is reflected in the after effect where nystagmus (and sense of rotation) reverses direction if the participant spends greater than a few minutes in the MRI. These effects occur on a time scale that would be lost by averaging over a 10 minute period. Nevertheless, the authors achieved their objective of determining whether there is a spatial bias induced by the MRI magnetic field. The temporal effects would not detract from the principle discovery that the MRI machine induces a spatial bias. The results will have implications for fMRI studies and may lead to new ways of treating spatial neglect. 

Studies have also found that the perception of vertigo is shorter than the nystagmus time-constant. What is the strength of the relationship between VOR slow-phase eye velocity and saccade bias? As commented on in the supplement material, typical MVS VOR shows adaptation within a minute or two. Does the VOR and saccade bias follow a similar time-constant?

Reviewer #3 (Public Review): 

The authors report a study in which they show that the involuntary eye movements induced by immersion within the dense magnetic field of an MRI scanner lateralises the spatial deployment of visual attention. This is an important finding because it shows, for the first time, that this source of involuntary movement is of functional significance. The implication is that an individual who is placed within a scanner will likely show attentional biases and vestibular-based neural activations that may need to be accommodated by relevant control conditions/participant groups. In short, the very procedure of imaging the metabolic processes that underpin visual cognition has now been shown to induce visual artifacts that need separate consideration. More generally, the study reaffirms the pervasive effect of vestibular stimulation on lateralised visuo-spatial behaviour. 

The experimental methodology appears robust, and the data interpretation raises no real concern. 

My main query is the extent to which the findings are generalisable. The visual search task (and straight ahead task) were conducted in complete darkness which is, of course, an uncommon viewing condition. One would expect the degree of bias to be strongest under such conditions, and the question arises as to how much the bias is weakened when other stimuli, especially those in the 'neglected' field or which are strongly salient, become visible and compete for attention? On a related note, the degree of bias may also be weakened in scanners that generate a smaller (e.g. 1.5T) magnetic field. 

My other query concerns the author's suggestion that MRI exposure could be used to help treat unilateral perceptual disorders such as hemi-spatial neglect. One questions the effectiveness of this approach given (1) the cost and availability of MRI, not least relative to the galvanic and caloric vestibular stimulators that are available and which enable a more diverse range of stimulation protocols, and (2) the patient burden associated with having to lie inside a high-strength scanner, likely exacerbated in the case of neglect by the presence of allied motor impairment and/or contraindicated by the presence of metal inside the body.

Reviewer #1 (Public Review):

The manuscript provides very high quality single-cell physiology combined with population physiology to reveal distinctives roles for two anatomically dfferent LN populations in the cockroach antennal lobe. The conclusion that non-spiking LNs with graded responses show glomerular-restricted responses to odorants and spiking LNs show similar responses across glomeruli generally supported with strong and clean data, although the possibility of selective interglomerular inhibition has not been ruled out. On balance, the single-cell biophysics and physiology provides foundational information useful for well-grounded mechanistic understanding of how information is processed in insect antennal lobes, and how each LN class contributes to odor perception and behavior.

Reviewer #2 (Public Review):

The manuscript "Task-specific roles of local interneurons for inter- and intraglomerular signaling in the insect antennal lobe" evaluates the spatial distribution of calcium signals evoked by odors in two major classes of olfactory local neurons (LNs) in the cockroach P. Americana, which are defined by their physiological and morphological properties. Spiking type I LNs have a patchy innervation pattern of a subset of glomeruli, whereas non-spiking type II LNs innervate almost all glomeruli (Type II). The authors' overall conclusion is that odors evoke calcium signals globally and relatively uniformly across glomeruli in type I spiking LNs, and LN neurites in each glomerulus are broadly tuned to odor. In contrast, the authors conclude that they observe odor-specific patterns of calcium signals in type II nonspiking LNs, and LN neurites in different glomeruli display distinct local odor tuning. Blockade of action potentials in type I LNs eliminates global calcium signaling and decorrelates glomerular tuning curves, converting their response profile to be more similar to that of type II LNs. From these conclusions, the authors infer a primary role of type I LNs in interglomerular signaling and type III LNs in intraglomerular signaling.

The question investigated by this study - to understand the computational significance of different types of LNs in olfactory circuits - is an important and significant problem. The design of the study is straightforward, but methodological and conceptual gaps raise some concerns about the authors' interpretation of their results. These can be broadly grouped into three main areas.

1) The comparison of the spatial (glomerular) pattern of odor-evoked calcium signals in type I versus type II LNs may not necessarily be a true apples-to-apples comparison. Odor-evoked calcium signals are an order of magnitude larger in type I versus type II cells, which will lead to a higher apparent correlation in type I cells. In type IIb cells, and type I cells with sodium channel blockade, odor-evoked calcium signals are much smaller, and the method of quantification of odor tuning (normalized area under the curve) is noisy. Compare, for instance, ROI 4 & 15 (Figure 4) or ROI 16 & 23 (Figure 5) which are pairs of ROIs that their quantification concludes have dramatically different odor tuning, but which visual inspection shows to be less convincing. The fact that glomerular tuning looks more correlated in type IIa cells, which have larger, more reliable responses compared to type IIb cells, also supports this concern.

2) An additional methodological issue that compounds the first concern is that calcium signals are imaged with wide-field imaging, and signals from each ROI likely reflect out of plane signals. Out of plane artifacts will be larger for larger calcium signals, which may also make it impossible to resolve any glomerular-specific signals in the type I LNs.

3) Apart from the above methodological concerns, the authors' interpretation of these data as supporting inter- versus intra-glomerular signaling are not well supported. The odors used in the study are general odors that presumably excite feedforward input to many glomeruli. Since the glomerular source of excitation is not determined, it's not possible to assign the signals in type II LNs as arising locally - selective interglomerular signal propagation is entirely possible. Likewise, the study design does not allow the authors to rule out the possibility that significant intraglomerular inhibition may be mediated by type I LNs.

Reviewer #3 (Public Review):

To elucidate the role of the two types of LNs, the authors combined whole-cell patch clamp recordings with calcium imaging via single cell dye injection. This method enables to monitor calcium dynamics of the different axons and branches of single LNs in identified glomeruli of the antennal lobe, while the membrane potential can be recorded at the same time. The authors recorded in total from 23 spiking (type I LN) and 18 non-spiking (type II LN) neurons to a set of 9 odors and analyzed the firing pattern as well as calcium signals during odor stimulation for individual glomeruli. The recordings reveal on one side that odor-evoked calcium responses of type I LNs are odor-specific, but homogeneous across glomeruli and therefore highly correlated regarding the tuning curves. In contrast, odor-evoked responses of type II LNs show less correlated tuning patterns and rather specific odor-evoked calcium signals for each glomerulus. Moreover the authors demonstrate that both LN types exhibit distinct glomerular branching patterns, with type I innervating many, but not all glomeruli, while type II LNs branch in all glomeruli.

From these results and further experiments using pharmacological manipulation, the authors conclude that type I LNs rather play a role regarding interglomerular inhibition in form of lateral inhibition between different glomeruli, while type II LNs are involved in intraglomerular signaling by developing microcircuits in individual glomeruli.

In my opinion the methodological approach is quite challenging and all subsequent analyses have been carried out thoroughly. The obtained data are highly relevant, but provide rather an indirect proof regarding the distinct roles of the two LN types investigated. Nevertheless, the conclusions are convincing and the study generally represents a valuable and important contribution to our understanding of the neuronal mechanisms underlying odor processing in the insect antennal lobe. I think the authors should emphasize their take-home messages and resulting conclusions even stronger. They do a good job in explaining their results in their discussion, but need to improve and highlight the outcome and meaning of their individual experiments in their results section.

Reviewer #1 (Public Review): 

This paper sought to dissect the relative impact of history, selection, and chance, on the evolution of antibiotic resistance in the clinically relevant species Acinetobacter baumannii. The authors conducted adaptive evolutions of A. baumannii isolates that had been previously adapted to diverse environments, thus establishing distinct histories. The authors show that the impact of history becomes increasingly diminished as selection strength increases, and several specific observations were made about resistance to beta lactams and their collateral effects of ciprofloxacin resistance. Overall the question being asked is important and the observations made are quite interesting. However, the analysis lacks sufficient depth to draw specific conclusions, and many confounding effects (such as the lack of propagation in a drug-free environment) are not taken into account. 

Minor comments: 

The authors seem to cite themselves an inappropriate amount of times for key findings, and many highly established evolutionary studies on this very topic were not included. For example in line 79 - mutation rate is a well documented parameter that has been estimated long before their work in 2019. Likewise, there have been a large number of studies that leverage population data that were not included.

Reviewer #2 (Public Review): 

The experimental design in this manuscript is exquisite. Its is simple in rationale yet also very clever and the work is performed to an excellent standard. The authors clearly address the extent to which history, chance and selection lead to the evolution of AMR, and it is all the more stronger that this is done in a real MDR clinical pathogen (A. baumannii) rather than lab E. coli. 

The work shows that history can influence AMR evolution, but that clearly natural selection is a dominant driver. This provides clear unambiguous data on the importance of antibiotic exposure on the evolution of AMR and will interest evolutionary biologists, microbiologists and clinicians.

Reviewer #3 (Public Review): 

The manuscript by Santos-Lopez and colleagues investigates the roles of history, chance, and selection on the evolution of antibiotic resistance in the pathogen A. Baumannii. In previous work, they showed that the genotypic and phenotypic evolution of (fluoroquinolone) resistance differed between well-mixed and spatially extended (biofilm) environments; this work uses laboratory evolution experiments to investigate further evolution in response to new (beta lactam) drugs. Their experimental design is based on a simple but elegant assay for distinguishing the impact of previous adaptation ("history"), random deviations across replicate populations ("chance"), and selective pressure from the newly applied drug ("selection"). They found that while prior history of selection (including prior growth environment) often impacts evolution of resistance to a new drug, increasing concentrations of that drug generally reduced historical contingencies-that is, the prior selecting conditions became less influential on the new adaptation trajectories (quantified by MIC-based direct and collateral resistances). They also performed extensive population sequencing of the evolved populations and similarly quantified the effects of history, chance, and selection using aggregate measures of genome similarity based on Manhattan distance metrics. Notably, they found that strains originally selected in structured environments exhibited genetic reversion and a corresponding loss of resistance to the initial drug. 

Overall, this study addresses an interesting and important problem. It is well designed, with careful attention to both the phenotypic and genotypic analysis of evolved strains, and the results contribute new insight into the trade-offs associated with antibiotic resistance in an ESKAPE pathogen. I enjoyed reading this work. My comments below are suggestions to improve the paper and can be addressed by additional clarification and/or discussion of the limitations of the approach. 

Minor: 

- Define / cite ESKAPE pathogens for readers not familiar <br> - Why choose the Manhattan metric? It is not unreasonable, but I am wondering 1) if there is a deeper theoretical justification and 2) whether other metrics could be expected to give similar qualitative results.

Reviewer #1 (Public Review): 

The authors present a new computational method, named MetaGPA, for performing enrichment analysis on cohorts of metagenomes. The method works similarly to GWAS, in that case and control groups are defined. They apply their computational method to the DNA extracted from multiple environmental metagenomes split into cohorts based on the presence or absence of cytosine modifications to the DNA. They uncover an enzyme that they show converts 5hmdC to 5cmdC. Much work has gone into enrichment analysis using individual genomes, but here they perform it on metagenomes. While the MetaGPA method shows promise, it was not fully described or characterized, and attempts to reuse the code failed.

Reviewer #2 (Public Review): 

This manuscript describes a methodology in which bacteriophages are isolated from natural sources such as sewage and their DNA is extracted. The DNA is divided into two parts, the control DNA is treated using an enzyme cocktail to break it at every unmodified cytosine while the modified cytosines are protected in the "case" sample prior to its treatment that cleaves at the unmodified cytosines. The expectation is that organisms that modify most of their cytosines will survive the cleavage step and this unbroken DNA will be enriched in the subsequent amplification and NextGen sequencing steps. The sequencing reads are then compared between the two samples to find genes that are enriched in the case samples and the Pfam database is used to identify potential DNA base-modifying enzymes. Such a search revealed many amino acid sequence motifs associated with base-modifying enzymes and the presence of nearby thymidylate synthase gene was used to identify enzymes with carbamoyltransferase domains. One such enzyme is cloned, purified and biochemically characterized. The authors demonstrate that it transfers the carbamoyl moiety to the oxygen in 5-hydroxymethylcytosine in DNA, RNA, dCMP. The authors suggest that this methodology could be generalized to find other base modifying enzymes. While it is impressive that the investigators are able to find a new base-modifying enzyme in the absence of any prior sequence information or direct selection for the activity, I have several concerns about the methodology and its potential as a general search tool for base modifying enzymes.

Reviewer #1 (Public Review):

The paper is a tour-de-force across multiple techniques and model systems from classical forward screening in C. elegans over ChIP to targeted CRISPR mutagenesis. The data is of a very high quality and supports most of the authors' claims strongly and convincingly. Finally, the manuscript is well written and, in spite of complex experiments and genetics, interesting and easy to comprehend.

- CAMTA, as the name CaM-binding transcription activator implies, have been studied previously and across many different organisms including plants, mice and humans. It was thus presumed and in part shown that CAMTAs regulate transcription depending on CaM levels.<br> - The authors confirm that the gene cmd-1 (encoding CaM) is directly regulated by Camt-1 by using a combination of cell-specific RNAseq and ChIP. This allows them to identify three binding sites upstream of the cmb-1 gene that bind to Camt-1.<br> - Moreover, the authors show that overexpression of CaM in the nervous system fully rescues the observed behavioral phenotypes.<br> - Importantly, the authors make another discovery. They show that CaM can directly repress its own transcription by binding to specific residues of Camt-1. Thereby, the authors argue, Camt-1 is used to precisely and bidirectionally regulate CaM levels dependent on the cell, animal's state etc.

The reported data are interesting and, in particular, the aspect that CAMTAs likely act as activators AND repressors is a novel aspect previously not appreciated. In spite of all these strengths, a potential weakness is that it remains open whether this mechanism is primarily a house-keeping mechanism or is indeed, as the authors speculate, regulated by internal and external factors that might, through CAMTA, make cells more or less responsive to Ca2+-CaM signaling.

Reviewer #2 (Public Review):

Vuong-Brender, Flynn, and de Bono report a detailed analysis of the function of a highly conserved calcium-calmodulin-dependent transcriptional regulator in the function of the C. elegans sensory nervous system. The C. elegans homolog of this factor - CAMT-1 - emerged from a genetic screen for mutants defective in a sensory-driven aggregation behavior. The authors find that multiple chemosensory modalities are disrupted by loss of CAMT-1, and this factor has distributed functions in the nervous system, including in interneurons that receive inputs from sensory neurons. A major finding of this study is that many of the effects of CAMT-1 mutation can be linked to a critical role for CAMT-1 in regulating expression of calmodulin itself. This finding is supported by multiple lines of experimentation, including a demonstration that the effects of losing CAMT-1 can be compensated by restoring expression of calmodulin. The authors further show that what is true for CAMT-1 and calmodulin in C. elegans also applies to Drosophila, indicating that CAMT-1 is a regulator of calmodulin expression whose function has been conserved throughout evolution. This manuscript has many strengths. Key hypotheses are tested using quantitative and technically independent experimental methods. The case that CAMT-1 is a regulator of calmodulin expression is built carefully and, for the most part, the logic of the argument is made clearly and supported by compelling data. Another strength of the manuscript is its candid exposition of data that do not fit neatly into the most simple and accessible model. It is refreshing to see authors who freely admit that they haven't neatly wrapped up every question in a field. The loose ends in this study do not impact the authors' main conclusions. However, some observations seem to consume more bandwidth than warranted, and the authors should consider reorganizing the manuscript so that the loose ends do not distract from the main thread of the narrative. The paper does have a few minor weaknesses that could be addressed. These are listed below.

Specific comments:

1. The initial description of the isolation of camt-1 mutants seemed a bit disorganized. A description of the gene and gene product preceded descriptions of the mutants. Also, some mutants were mentioned in the text but not presented in the corresponding figure. The authors should consider minor changes to better communicate how the mutations were cloned.

2. In Fig. 2 npr-1 baselines vary a great deal between panels A, B, and C. It is not clear why npr-1 behavior is this variable, and the authors do not mention this obvious feature of their data. Data presented in Fig. 2 indicate that heat-shock-induced expression of camt-1 restores a defect in basal locomotion, but it is unclear whether it restores O2-sensitivity - the effect of oxygen on speed of transgenics seems the same +/- heatshock (compare black traces in panels 2B and 2C).

3. Unlike other datasets, the responses of wild-type AFDs to CO2 do not look particularly convincing (panel 3C). There is clearly an effect of camt-1 mutation on AFD calcium, but the AFD responses seem qualitatively different from the responses of BAGs to CO2 or URXs to O2. The authors might consider moving these data to a supplementary figure and tempering their description of wild-type AFDs as CO2-sensors.

4. The authors candidly present data that do not conform to a simple model for how camt-1 affects behavior. Loss of camt-1 increases calcium in sensory neurons that activate the speed-controlling interneuron RMG. However, RMG calcium is reduced in camt-1 mutants. This inversion in the effect of camt-1 mutation might be caused by a homeostatic mechanism, as the authors propose. It might be possible to test this hypothesis by testing whether reducing excitatory input into RMGs elevates resting calcium in camt-1 mutants, for example via mutations that affect sensory transduction.

5. In Fig. 4H RMG data are presented as fractional ratio change - all other imaging data are presented as absolute ratios of YFP and CFP fluorescence. It is not clear why these data are treated differently. It is also no clear that these data are consistent with data shown in Fig. 3F. Which dataset represents the effect of camt-1 mutation on RMG calcium? More measurements might be warranted.

6. Nice experiments show that regulation of calmodulin in Drosophila requires a CAMT-1 homolog. The bar graphs showing unity for values normalized to themselves are a bit odd - perhaps there's a more compact way to plot these data.

7. ChIPseq analysis of CAMT-1 is also quite nice. Is there a sequence motif for CAMT-1 binding that emerges from this study? If so, how does this motif compare to motifs from studies of CAMT-1 homologs in other species?

8. Figure 7 shows that CMD-1 inhibits cmd-1 expression via interaction with CAMT-1. These data are interesting, but it is not clear how this effect can be related to prior data showing that forced expression of CMD-1 can compensate for loss of CAMT-1. The authors behavioral and physiological studies suggest that in vivo CAMT-1 promotes CMD-1 expression. In Figure 7, they suggest that CAMT-1 inhibits expression of CMD-1, but there is no clear link to behavior or physiology for this repressor-function of CAMT-1. The manuscript might be more clear without these data, and the absence of these data would not affect the overall impact of the study.

Reviewer #3 (Public Review):

Vuong-Brender et al present a thorough study investigating how CaM-binding transcription activators (CAMTAs) in C. elegans and Drosophila are required for numerous behaviors and proper neuronal function. The study is strong in how it uses a variety of approaches to study a major underlying mechanism for CAMTA. First, they use reporters, mutant analysis, and heat-shock rescue to show how cart-1 is expressed widely in neurons and functions in adults in several behaviors. They used transcriptional profiling to show that cart-1 is required to upregulate CaM in subsets of neurons in worm. They next use ChIP-seq to zero in on where worm CAMT-1 binds regulatory regions upstream of the CaM gene cmd-1 to promote its expression. They find that overexpression of CaM compensates for behavioral and neuronal response deficits in a cart-1 mutants. Lastly, they propose that when CaM highly expressed, it may down regulate its own expression by binding CART-1.

1. Overall, I feel that the study is excellent and most conclusions are justified by evidence. However, I do not think the title is supported by the data. It currently is listed as: CAMTA TUNES NEURAL EXCITABILITY AND BEHAVIOR BY MODULATING CALMODULIN EXPRESSION. The authors show evidence that camt-1 is required for the normal function of neurons and behavior by promoting expression of CaM. Their only evidence that camt-1 downregulates CaM is a more artificial situation where CaM is overexpressed. I don't think they provide any evidence that camt-1 is used to "tune" behavior or neuron activity up and down in a wild-type strain. Tuning implies that the molecule modulates a physiological system bidirectionally in a natural situation. I suggest using a more accurate title that better fits the experimental evidence.

2. They show ample evidence that cart-1 appears to promote the expression of cmd-1 in most cases. This includes showing that overexpression of cmd-1 suppresses the behavioral and imaging phenotypes of cart-1. But they didn't perform the more straight forward epistasis test with the cart-1;cmd-1 double mutant in worm or fly , presumably because there is no viable loss-of-function allele in the coding area of the cmd-1 gene. It would help the readers understand why this simpler experiment was not performed if they explain this in the paper. A good place would be near line 220, where they generate hypomorphic promoter alleles using CRISPR. If they have tried to make their own loss-of-function alleles by mutating the coding area of cmd-1, but it resulted in presumed lethality, this might be mentioned here too.

3. I am most worried about the potential caveats with the calcium imaging experiments. As the authors note, it is challenging to infer absolute levels of calcium using the ratiometric sensor cameleon across different individuals and genotypes. However, the authors do not note that the YFP/CFP FRET signal from cameleon might be perturbed because it uses calmodulin to bind calcium. At the end of their study (line 244), they provide evidence that calmodulin may bind to CART-1 to suppress its own expression when calmodulin is highly expressed. This is worrisome because cameleon is probably expressed highly in some or most of these strains. The authors may want to re-examine neuronal activity for a subset of experiments with a method that is independent of a calmodulin-based sensor (if possible).

4. The title of "Fig 3 - Figure supplement 1" is confusing because it suggests that they measured the levels of YC2.60 cameleon, when in fact they measured a separate GFP reporter, albeit using the same promoter. So they could clarify the figure title.

Reviewer #1 (Public Review):

Experiments reported in this manuscript aimed to determine if long ascending propriospinal neurons (LAPN) located in the lumbar prominence facilitated locomotor recovery after a moderate severity spinal cord contusion in rats using measurements of gait and swimming ability. Prior studies from this group implicated LAPN in fine tuning coordination of forelimbs and hindlimb during locomotion. The authors used viral constructs to infect cell bodies of LAPN in lumbar spinal cord and their synapses in cervical spinal cord to express botulinum toxin when doxycycline was administered in drinking water. This technique appeared to infect both inhibitory and excitatory LAPN. Strengths of the manuscript include a very thorough gait analysis, evidence through histologic analysis that the desired viral infections occurred, and an experiment in which it was shown that effects of doxycycline on gait resolved when this medication was removed then reappeared when it was added back. Weaknesses include lack of controls that demonstrate lack of any effect of doxycycline on gait parameters in animals injected with control viruses to control for any unexpected effect on gait of the medication or infection of neurons with the virus.

The authors were able to evaluate effects of synaptic silencing of LAPN after SCI. The data showed, unexpectedly, that silencing improved ipsilateral hindlimb gait parameters, left-right hindlimb coordination and forelimb hindlimb coordination.

How the spinal cord is rewired after SCI remains poorly understood. The data presented in this manuscript will advance understanding of the role of LAPN in this process. An important question posed by the authors results is what the respective role(s) of the excitatory and inhibitory LAPN are in the biology they uncovered. Also of interest is whether this biology results from changes in local inputs to LAPN through the sensory nervous systems which could be interpreted as maladaptive given the authors results.

Reviewer #2 (Public Review):

Shepard et al. investigated the role of long ascending propriospinal neurons (LAPNs) in recovery of locomotor function after a moderate spinal contusion injury. Their previous observations had determined that LAPNs have a context dependent function in coordinating fore- and hindlimbs patterns during locomotion. Surprisingly, silencing LAPNs after contusion SCI improved several parameter of locomotion.

Strengths

The novelty and unexpected results are certainly the mains strengths of this report. Given all previous evidence, mostly originating from their own work, they logically hypothesized that disrupting this usually spared pathway after contusion injuries would have a detrimental effect on functional recovery. Interestingly, interlimb coordination was improved after silencing of LAPNs with limit or no change to forelimb-hindlimb coordination patterns.<br> They provide an extensive study of overground kinematics and swimming test. The detailed kinematics parameters provided the basis for their novel results. The addition of the swimming test provided conflicting results compared to the overground walking tests. This aspect was quite interesting and informative as swimming in rodents is a bipedal motor skill devoid of weight bearing. Hence, control of overground locomotion and swimming may be differentially dependent on LAPN activity as clearly illustrated here.

Weaknesses

Although the results are novel, the complete reliance on kinematic analysis limited the reach and conclusions of this manuscript. Their previous published work using these methods also used similar techniques for kinematics analysis. Two fundamental points of weakness here in relation to analysis of hindlimb movements after SCI is the lack of markers defining the knee but importantly the distal phalange of the toes. They defined their hindlimb segments as hip/knee and knee/ankle, when the knee joint was clearly not labelled. There are obvious issues with skin markers especially on the knee joint of rodents; however, given that virtually all of their evidence is based on kinematics, inclusion of such marker would have strengthen their analysis. Knee joint issues apart, the exclusion of a marker on the distal phalange of the toe is an equally critical variable in SCI locomotion. They described they placed markers on the “iliac crest, hip, ankle and toe”, but more precise location would have been helpful, e.g., head of the femur, lateral malleolus, etc. Especially after moderate contusion, toe dragging in different phases of stepping is a hallmark of deficits in this condition. This was partially illustrated by analysis of dorsal versus plantar stepping. However, toe dragging does not always result in dorsal stepping. From the figures and previous publications from the authors, the toe marker seems to have been placed between the metatarsal and the proximal phalange, which does not describe the location of the toes in relation to ground contact. The 2-dimensional analysis also limits any analysis of potential rotations.

There was a complete lack of mechanisms, which is really disappointing. No anatomy or physiology studies were presented to substantiate or investigate possible mechanisms. Figure 2 is the only one showing anatomy, but it is limited to confirming lesion/spared tissue and location of viral vectors. Several of these issues were later mentioned in the Discussion, but this makes this manuscript purely descriptive.

Reviewer #3 (Public Review):

The manuscript by Shepard et al determines the effects of silencing long ascending propriospinal neurons after spinal cord injury. The authors use an elegant reversable silencing strategy that involves a dual virus system to specifically label long ascending propriospinal neurons (LAPNs). This follows a recent study by the group which demonstrated that silencing these neurons in intact rats, resulted in disrupted left-right coordination between the hindlimbs (and between the forelimbs), in addition to alterations in coordination of diagonal forelimb-hindlimb pairs, all seen only in certain contexts (Procratsky et al 2020). Here, the same silencing strategy is used but to determine the functional role these neurons play during locomotor recovery post-incomplete SCI. The reversibility of the silencing was fully leveraged in order to examine the functional consequences of LAPN loss both pre- and post-SCI (multiple times) in the same subject. It was expected that the silencing of the LAPNs spared after incomplete SCI would further disrupt locomotion but the opposite occurred, hindlimb coordination improved. Both the improvement in left-right alternation of the hindlimbs and the lack of major detrimental effect on forelimb-hindlimb coordination are surprising. Having spared fiber tracts leading to worse functional outcomes has interesting implications for strategies to restore function after SCI.

Overall, this is a well-written manuscript with data that are clearly presented and support the conclusions. However, some aspects of the analysis should be clarified. Possible functional explanations for the findings are discussed and it may be possible to narrow down some of these possibilities with additional histological data.

Joint Public Review: 

The authors investigated presynaptic homeostatic plasticity (PHP) at the glutamatergic larval Drosophila neuromuscular junction (NMJ). In this facet of synaptic plasticity, the presynapse increases neurotransmitter release to compensate for diminished postsynaptic sensitivity. To study functional pathways and to identify new molecular components of PHP, the authors carried out a large, electrophysiology-based genetic screen focusing on E3 ubiquitin ligases - key elements of the major cellular protein degradation pathway. This genetic screen, which forms the backbone of the paper, generated an extensive data set encompassing 180 genotypes. This is an impressive achievement. The authors find that the E3 ligase Thin suppresses glutamate release by downregulating Dysbindin, a transmitter release-promoting presynaptic protein. Based on the experimental data, a model is put forward according to which PHP arises by relieving Dysbindin of Thin-dependent ubiquitination and degradation. 

This is a strong paper that adds a highly interesting feature to our understanding of the molecular mechanisms that control synaptic strength. On the other hand, two key aspects of the major conclusions remain equivocal, requiring rectification: (i) A precise interpretation of the morphological phenotypes and the corresponding molecular processes controlled by Thin is not yet fully possible. (ii) Some aspects of the Thin-Dysbindin interplay remain unclear.

Reviewer #1 (Public Review): 

Ter-Ovanesyan, Dmitry et al. reported a novel strategy for quantifying EV yield and purity through detecting three widely expressed transmembrane proteins in EVs (i.e., the tetraspanins CD9, CD63, and CD81) and one most common contamination protein (i.e., albumin) using Simoa. By applying this method, the authors directly compared the yield and purity of commonly used EV isolation methods, i.e., ultracentrifugation, precipitation and SEC columns. According to the report, SEC columns outperform other methods, and the 10 mL Sepharose CL-6B column is the best choice for EV isolation from plasma or CSF, which showed better performance than the most widely used 10 mL Sepharose CL-2B columns. <br> This work is interesting as it provided a rapid framework for comparing and improving EV isolation methods from biofluids. However, some conclusions are not entirely clear and need further clarification. 

1) As both using Simoa to quantify EVs and the different isolations methods will lead to different EV yield and purity are both published previously. The most novelty part of this manuscript is combining these two aspects, i.e., using Sioma to determine the yield and purity of EV samples. However, to address this part, this method should be directly compared with some commonly used techniques such as NTA to test the consistency and different performance between those methods. Especially, the authors indicated in the discussion part that this method is possibly better than commonly used methods. 

2) Interestingly, the authors claimed that lipoproteins are the hardest to be distinguished from EVs due to their similar size but still using albumin as the contamination marker. Due to the smaller size, albumins can be easily excluded in the NTA analysis. The success of using albumin to determine the purity of the EV sample does not support the conclusion that Simoa is better than the NTA. 

3) Some sub-conclusions are not solid enough. For example, whether different centrifuge speeds and time will affect the claim that SEC outperformance ultracentrifuge in regard to EV yield. As it is well known that longer and quicker spinning will increase the yield of EVs. Moreover, whether 10 mL Sepharose CL-6B column still the best choice for EV isolation from plasma or CSF when extending contamination markers to lipoprotein markers. As the size of lipoproteins is much larger, the purity of 10 mL Sepharose CL-6B column may be more affected than 10 mL Sepharose CL-2B columns.

Reviewer #2 (Public Review): 

The authors used a series of sensitive Simoa assays to evaluate EVs and contamination from albumin across common isolation techniques. The experiments were performed rigorously using the same pool of samples from two biofluids, plasma and CSF, for all comparisons. Both biofluids were examined across each isolation method and both the recovery of EVs and contaminants were reported. The number of isolation methods tested was good, though the rationale for choosing each method was missing. The inclusion of albumin as a contaminant was useful information for assessing each isolation method. The additional measurement of lipoprotein contaminants would have strengthened the available data. SEC, once chosen, was very thoroughly examined along with changes such as the volume of the column and the pore sizes.

Reviewer #3 (Public Review): 

There is a lack of consensus about the best way to isolate EVs from biofluids, particularly for human biofluids such as cerebrospinal fluid (CSF) where samples sizes are limited due to the invasive nature of biofluid collection. The authors report that size exclusion chromatography (SEC) is superior for isolation of EVs from 0.5 mL of plasma or CSF versus using ultracentrifugation or commercial precipitation kits. They report the outcomes of isolating EVs using various SEC parameters (Sepharose size, column length, fractions collected) and using single molecule array (Simoa) assays and three commonly used tetraspanins (CD9, CD63, and CD81) to quantify EVs. They report the EV yields and purity with respect to albumin by the various SEC parameters for both plasma and CSF, and the authors conclusions are justified by their data. This is the first demonstration of the use of Simoa with three commonly used tetraspanins to measure EVs from small volumes of CSF, of great relevance to human CSF biomarker studies. Their findings also support that SEC isolation of EVs combined with Simoa using common tetraspanins and additional selective markers could be applied to compare EV isolation methods from other fluids such as cell culture media.

Reviewer #1 (Public Review): 

The goal of this study was to generate and analyze phenotypes associated with alleles of ssl2 that alter transcription start site (TSS) selection. Specifically, the authors created a screening method to detect either upstream TSS shifts or downstream distinctly and then use this approach to create new alleles of ssl2 that might reveal important residues/regions of ssl2 involved in promoter scanning. To that end, new alleles were not only characterized using reporter-based methods but were broadened to genome-wide analysis through a transcriptional start site sequencing approach. Ssl2 alleles were further characterized in combination with Pol II alleles that also impact TSS selection to build a framework of additive and epistatic interactions. Overall, this study was conducted in a very thorough manner and is likely to make an impact on the field by provided insight as to how the key enzymatic players work to initiate transcription. The primary weakness of the paper is that of presentation as the authors attention to detail at times was very dense and thus reduced readability of the study.

Reviewer #2 (Public Review): 

This work is based on the use of multiple reporter systems for TSS usage based on yeast growth (Fig 1A). Those reporters enable to use cell growth as a proxy for changes in TSS usage. 

The authors combine the use those systems and with primer extension assays to screen a big number of known and novel mutants. Using complementary screen strategies, they screen for novel mutants of sll2 associated with changes in TSS usage. The authors analyze the distribution of the identified mutations and focus on 2 groups of alleles conferring upstream or downstream TSS shifts. Then, they confirm for selected mutants that the identified phenotypes are indeed associated to genome-wide changes in the distribution of TSS. Next, they investigate the genetic interaction between the new sll2 alleles and known mutants able to change TSS that the authors have previously characterized. Finally, the authors show that the changes in TSS are also associated with subtle differences in PIC position as measured by ChIP-Exo. This is a strong work based on meticulous use of yeast genetics and confirmed by direct measure of the molecular phenotypes by TSS-Seq and ChIP-Exo. By its own nature, the presented work is complex and difficult to follow. Despite that, the authors include multiple schemes to facilitate its interpretation.

Reviewer #3 (Public Review): 

Biochemical and structural work has produced a model in which TFIIH translocase pushes downstream DNA back towards the active site to create a transcription bubble. In most eukaryotes, the transcription start site (TSS) appears within this initial bubble at a relatively fixed distance from the TATA box. However, at Saccharomyces cerevisiae promoters, multiple TSSs are spread in a window over a larger range. To explain this, a "scanning" mechanism has been proposed. Earlier work from the Kaplan lab and others provides support for directional scanning, where individual TSSs each have a certain probability of functioning. The Kaplan lab has shown that mutations in the RNA pol II active site can shift TSS choice upstream or downstream, with polymerases having faster elongation rates tending to choose upstream TSSs, while slower polymerases start further downstream. Here, Kaplan and colleagues extend their analysis to TFIIH mutants, providing genetic evidence that TFIIH function (and presumably translocation rate and/or processivity) is also a determinant of TSS choice. The one major weakness of the paper is that the TFIIH mutants are not tested biochemically to determine their actual defects (i.e. translocation rate, processivity, etc.), but I think that's a major undertaking that would be outside the scope of this paper. Based on genetic interactions between mutant alleles of TFIIH and other PIC components, a plausible and reasonably persuasive model is proposed for how the initiation/polymerization rate determines TSS choice within a window set by TFIIH translocase processivity.

Reviewer #1 (Public Review): 

In this manuscript "Spatial alanine metabolism determines local growth dynamics of Escherichia coli colonies" by Dias-Pascual, et. al., the authors present a detailed and comprehensive spatial and temporal characterization of the molecular mechanisms involved in E. coli colony biofilm formation. With their studies the authors propose a nutrient cross-feeding model explaining the phenotypic heterogeneity observed within the biofilm. They propose that in nutrient rich regions of the biofilm where carbon and nitrogen sources are not limiting, but oxygen is limited, cells secrete alanine, this amino acid is then used by the surface (and oxygen) exposed regions of the biofilms. Mutants impaired in cross-feeding through impairment of alanine secretion and degradation, have a higher number of dead cells in the surface exposed aerobic region of the biofilm. 

The manuscript is technically very solid. The authors use a combination of temporal-spatial transcriptomics and metabolomics in addition to novel methods to measure growth and cell viability within the biofilm. These methods can be applied to study biofilms in other bacteria. 

Overall, this study highlights the importance of nutrient cross-feeding within biofilm colonies, which according to the authors is likely to be important for biofilms in other systems, including multi-species biofilms, because cross-feeding interactions have already been described in multi-species bacterial interactions. Therefore, the current study, and the innovative methods applied here, will motivate novel studies to address the importance of nutrient cross-feeding interactions in other single, as well as, multi-species biofilms.

Reviewer #2 (Public Review): 

The manuscript by Diaz-Pascual et al shows that alanine metabolism varies spatially across a colony, and is related to both growth dynamics and the presence of oxygen. The authors use a combination of transcriptomics, metabolomics, and quantitative measurements of growth at the single-cell and colony levels to link these properties, all of which have potentially broad impact for thinking about the spatial structure of communities of all bacterial species. 

The data supporting the links between oxygen and alanine secretion/metabolism is convincing, and the authors provide genetic evidence that links alanine usage to both carbon and nitrogen metabolism.

Reviewer #3 (Public Review): 

In the work, spatial alanine metabolism determines local growth dynamics of Escherichia coli colonies, Diaz-Pascual et all used a transcriptomic-metabolomic approach on E. coli biofilms to show that alanine production occurs in different regions of the biofilm than alanine consumption and that this alanine cross-feeding mechanism is important for biofilm growth. 

The mayor strength of this work is the optimal combination of transcriptomics and metabolomics over time during bacterial growth. In my opinion, it is a technological hallmark in the study of biofilms. In my opinion, the weakness lies on certain concepts that are presented and are difficult to understand. 

A number of studies show that glucose and oxygen gradients give rise to distinct metabolic phenotypes within biofilms. A glucose-rich environment promotes acetate-producing fermentative metabolism; a glucose-deprived environment, however, promotes acetate consumption. This is consistent with classical experiments in liquid cultures; cells grow exponentially using substrate-level phosphorylation and when the concentration of glucose decreases use other non-preferred catabolizable substrates (acetate or alanine) via TCA cycle. As E. coli cells cross-feed acetate within biofilm, a similar pattern should occur with alanine metabolism, which is the subject of this study.

Reviewer #1 (Public Review):

The manuscript uses a new screen called MassTitr, a method that is based on fluorescence-activated cell sorting (FACS) of a library of peptide-displaying bacteria and subsequent deconvolution of signals by deep sequencing. This method allows the authors to display long (36-mer) peptides derived from human proteome to screen for peptides that can bind the EVH1 domain of ENAH protein. About 100 peptides were identified and further analysis identified sequence features that contribute to the binding of EVH1 domain, including an additional proline after the FP4 motif and double FP4 motif.

The strengths of the manuscript are: (1) the MassTitr method is very nice and has advantageous features, which allow the screening of longer peptides to explore sequences surrounding the FP4 motif (2) By screening the entire human proteome-derived peptides (416,611 36-mer peptides with 7-residue overlaps), many EVH1 binders have been identified, including previously unknown ones that do not contain an FP4 motif. (3) A nice structural explanation for why EVH1 of ENAH prefers FP4 motif with an additional proline is provided. (4) Mutational studies have nicely confirmed that the double FP4 motif contributes to stronger binding. (5) The thermodynamic analysis (entropy versus enthalpy contribution) of binding is interesting and makes sense.

The weaknesses of the manuscript include: (1) Among the 108 binders selected, the authors further limited the 108 binders to 33 after bioinformatics analysis among which 14 are previously known to interact or co-localize with an Ena/VASP protein. However, whether the new ones really bind to ENAH is not tested. (2) The structure analysis for EVH1 of ENAH is nice, but inference to specificity was not tested by mutagenesis. (3) Although the MassTitr method works nicely, it is not easily apparent how much new insights the study has provided from the MassTitr screening. Perhaps the additional proline is one new insight. The double FP4 motif seems to be already known in the literature.

Reviewer #2 (Public Review):

Hwang et al. reports an exciting study that screened a library of 36-residue peptides that span the entire human protein coding space to identify binding partners of a key interaction domain that regulates cytoskeletal remodeling. Novel binding partners were identified and regions of the ligands beyond the canonical binding motif were shown to play critical roles in determining specificity and affinity.

The conclusions of the paper are mostly supported by the data, but clarifications of experimental details are needed as well as qualifications of some claims.

Strengths:<br> A powerful high-throughput screening of a peptide library was applied to search for binding partners of the EVAH EVH1 domain. Known binding partners provided controls, and new binding partners were identified. The longer length of peptides (36 amino acid residues) provided the opportunity to detect multivalent interactions. Indeed, many of the identified hits contained multiple EVH1 binding motifs.

Complementary techniques were applied to assess the relevance of hits and to select a subset for experimental validation.

Biophysical studies were employed to carefully quantify the most interesting interactions that were identified.

Weaknesses:

The claim that the reported studies show "that proteins with two EVH1-binding SLiMs can wrap around a single domain" is overstated. No structural data is presented to show that dual ligands "wrap around a single domain". It is a logical possibility that is consistent with the data presented, and should be described as such.

To explain why lengthening the linker between two binding motifs in a dual ligand weakens binding by only 2-fold, the authors suggest that the linker may interact favorably with the non canonical site. A likely alternative explanation is the effect of linker length on the effective concentration, Ceff, which reflects the concentration of a second binding motif after the first motif is bound.

Based on the materials and methods, the authors appear to have employed an "ENAH tetramer construct" (human EVH1 fused to ENAH mouse coiled coil) for all of the reported experimental interaction studies. However, the ITC methods describe sample preparation with "ENAH EVH1 domain". The monomer versus tetramer has important implications for interpretation of the ITC data, since a tetramer introduces potential inter-domain binding by dual ligands. Clarification is needed to be able to evaluate fully the implications of the data.

Reviewer #3 (Public Review):

The manuscript by Hwang et al. establishes a new technique for proteome-wide screening of interacting SLiMs via bacterial display (MassTitr). Results from this screening analysis reveals non-obvious influence that context can play in specificity of interaction for polyproline-containing motifs. Specifically, MassTitr revealed an unusual FP8 motif that is not precedented for binding to the ENAH EVH1 domain, and the presence of repeating FP4 motifs from the screen provided evidence that a second non-canonical FP4 binding site may exist, again a novel finding that was enabled by the MassTitr analysis.

Strengths: The MassTitr methodology appears to be straightforward and reliable, and thus should be of broad interest to protein and peptide chemists and potentially implementable in other laboratories. The identification of non-obvious context contributions identified for ENAH EVH1 domain provides an important example of how recognition likely includes more than the simple minimal motifs that are widely studied. Since such SLiM recognition proteins are widely distributed and important for a number of critical biological effects, this research is also of fundamental biological significance as well.

Weaknesses: While the specific examples provided do highlight the significance of the MassTitr screening, these would be burnished with additional examples such as those described in the accompanying report.

In several cases, comparisons are made between KD values that are relatively similar (e.g., 3-fold or less). Additional analysis is required to demonstrate that these differences fall outside the range of experimental error. In addition, the Methods appear to be missing description of the BLI assay even though it is referred to in the text of the Results.

The description of the selection and screening procedure in the Results is vague and lacks critical details.

Reviewer #1 (Public Review): 

In this project, the Takamori and the Haucke lab joined forces to investigate the mechanisms of endocytosis at central synapses. Particularly, the authors aimed to resolve the role of clathrin and AP-2 for endocytosis. To this end, they systematically investigated the kinetics of endocytosis with pH sensitive fluorophores (pHluorins) coupled to six major synaptic vesicle proteins: Synaptotagmin 1 (Syt1), the synaptic vesicle glycoprotein 2A, VAMP/Synaptobrevin 2 (Syb2), Synaptophysin (Syp), the vesicular glutamate transporter 1 (VGLUT1), and the vesicular GABA transporter (VGAT). The function of clathrin and AP-2 were perturbed by knock-down experiments using lentivirus based expression of shRNA validated by immunohistochemistry and rescue experiments. In addition, pharmacological tools were tested (pitstop 2). To control for overexpression of the pHluorin-linked synaptic proteins, antibodies directed against luminal domain of VGAT coupled to the pH-sensitive fluorophore cypHer 5E were used. 

The data show that clathrin is dispensable for endocytosis of all six tested proteins when tested at near-physiological temperatures (33-37 degree celsius) but clathrin can have effects on endocytosis kinetics in experiments performed at room temperature. The data support recent evidence for clathrin-independent endocytosis and resolve discrepancies with previous studies performed at room temperature. The most interesting findings of the study is that AP-2 is required for retrieval of some of the tested proteins (VGAT and VGLUT1) and this function is independent of clathrin. The study therefore reveals a clathrin-independent function of AP-2 in the endocytic sorting of a subset of synaptic vesicle proteins at central synapses. The study is of high technical quality and carefully designed. The manuscript is carefully written and the conclusions are supported by the data. The results provide exciting insights into the mechanisms of endocytosis.

Reviewer #2 (Public Review): 

Previous work has shown that clathrin and AP-2 operate to form synaptic vesicles from an endosomal intermediate that follows clathrin-independent endocytosis. The current study corroborates this finding and its importance at physiological rather than room temperature. However, this study goes one step further, to suggest a role for AP-2 but not clathrin in endocytosis of particular SV proteins, the neurotransmitter transporters. This raises the possibility of a novel mechanism by which AP-2, generally thought to depend on clathrin, can promote endocytosis independent of clathrin.

Reviewer #3 (Public Review): 

Here the authors have further tested the role of clathrin and clathrin adaptors in synaptic vesicle endocytosis. What is known is that endocytosis of membrane is fast and appears to be clathrin independent at synapses, at least during low frequency stimulation. Oddly, the clathrin adaptor complex AP-2 plays some role at synapses, potentially cargo recruitment to endocytic sites, or assembly of a clathrin coat to execute endocytosis. What was not known was whether synaptic vesicle proteins are retrieved by a clathrin-independent mechanism, and whether the roles for AP-2 are separable: specifically, does AP-2 recruit cargo for ultrafast endocytosis? 

The authors claim: 

1. Knockdown of clathrin heavy chain or application of the clathrin inhibitor Pitstop2 do not change re-acidification time constants of six different synaptic vesicle proteins at physiological temperature (200 AP at 40 Hz). The clathrin-independent mechanism is apparently universal for synaptic vesicle proteins and not exhausted by intense stimulation. 

2. Knockdown of the AP-2 mu subunit slows down the re-acidification of the vesicular transporters (VGLUT1 and VGAT), synaptophysin, and synaptobrevin, but has no effect on synaptotagmin or SV2 (200 AP at 40 Hz). There is a differential requirement by synaptic vesicle proteins for AP-2 function during endocytosis. 

3. The requirement for AP-2 mu is not simply due to fatigue caused by intense stimulation, but is required under less intense stimulations to retrieve proteins from the readily releasable pool of synaptic vesicles (50 AP at 20 Hz). 

4. The requirement for AP-2u is not caused by overexpression of tagged proteins, demonstrated by using fluorescently tagged antibodies and an acid-dependent dye. 

5. The transporters are stranded on the surface in the AP-2 mu knockdown as determined by biotin labeling of membrane proteins, and by immunofluorescence. 

6. Mutations of VGLUT1 and VGAT AP-2-binding sites slow recycling and acidification. 

There are two major conclusions from these data: 

1- Recycling of synaptic vesicle proteins is fast and clathrin-independent at physiological temperatures. These results complement previous morphological and electrophysiological studies tracking ultrafast membrane endocytosis at the synapse, and fundamentally change our understanding of protein recycling at the synapse. 

2- Despite the absence of a role for clathrin, the clathrin adaptor AP-2 is required for retrieval of some proteins. The role for AP-2 is not to assemble clathrin or some other membrane-bending protein to execute membrane invagination, because endocytosis of synaptotagmin and SV2 is independent of AP-2. Rather, AP-2 is likely to be recruiting proteins to the endocytic site. 

These conclusions will be of great interest to both the neuroscience community and broader cell biology community. The data are largely convincing, but because they are at odds with the accepted conclusions about both clathrin and AP-2, the results will be disputed.

Reviewer #1 (Public Review): 

The Ena/VASP family of actin regulators contain three highly homologous proteins, ENAH, EVL, and VASP. They have partially overlapping functions but also has unique functions. They use and EVH1 domain to bind many partner proteins to regulate the actin cytoskeleton. How binding specificity is determined is not well known. Here, the authors characterized one peptide binder, derived from a protein called PCARE. This peptide binder is a hit from proteome-wide screen that the authors report in a separate manuscript. Surprisingly, the PCARE peptided binds ENAH with 0.19 uM Kd and is very selective for ENAH over VASP and EVL. Using X-ray crystal structure and mutagenesis, the authors provide a very nice explanation for how the binding specificity is achieved, which is not trivial. Furthermore, using the understanding of the binding mode and specificity, the authors developed a peptide, PCARE-Dual, with a Kd of 50 nM for ENAH, which could be a useful research tool for studying the functions of ENAH. 

Overall, this is very interesting work with many strengths. The selectivity of PCARE for ENAH over VASP and EVL is very interesting. The FP4 motif binds EVH1 in the opposite orientation is interesting. The 14-residue C-terminal to the FP4 motif make additional contacts to EVH1, explaining its high affinity. The binding-induced conformation change in ENAH nicely explained the specificity of PCARE for ENAH over VASP and EVL. Using modeling method dTERMen to come out with a mutant of EVL that binds to PCARE B with 0.35 uM Kd is very interesting. Getting a peptide PCARE-Dual with 50 nM affinity to ENAH and selective for ENAH is impressive. 

The reviewer only found a few weaknesses/questions, which are relatively minor and can be addressed by better explanation or minimum experiments: 

Even though PCARE peptide is strong binder for ENAH, whether the endogenous PCARE protein binds to ENAH or not was not demonstrated. Even though this is not directly relevant for the main conclusion of the manuscript, the discussion of the manuscript does hint on the physiological function of this interaction. Thus, the protein-protein interaction should be validated in cells. 

"Mito-PCARE B serves to sink ENAH away from its normal sites of action" may not be correct, as there are still a lot of mito-PCARE B and ENAH near the plasma membrane based on Figure 2F. 

Why fusing the 36-mer PCARE to EVH1? What happened with the peptide not fused to EVH1? 

The authors used EVL V65P Y62C and showed that this mutant binds PCARE B 10-fold better than WT. How about the EVL Y62C mutant? Do you really need the V65P mutation? 

The authors envision that PCARE and PCARE-Dual could be promising leads for developing therapies to treat ENAH-dependent diseases, but this could be difficult to achieve in practice as it is likely going to be very difficult to shorten such long peptides while maintaining the potency and specificity.

Reviewer #2 (Public Review): 

Hwang et al. reports an interesting study that elucidates the structural mechanism by which only one of the three Ena/VASP paralogs, ENAH, selectively and tightly binds via its EVH1 domain to a newly discovered cellular target, PCARE. Peptides with enhanced affinities were generated by adding a second motif to contact a previously identified secondary binding site on the opposite face of the EVH1 domain. The conclusions of the paper are supported by the data. 

Strengths: 

Important findings include the identification of PCARE (photoreceptor cilium actin regulator) as a high affinity ENAH-specific cellular binding partner. PCARE is a vision-related protein involved in photoreceptor cell maintenance. How PCARE achieves its high affinity is revealed in the high-resolution crystal structure of ENAH EVH1 fused at its C-terminus to the PCARE peptide. This structure shows that the peptide binds in reversed N-to-C orientation and displays contacts beyond the canonical binding surface. How PCARE selectively binds tightly to EVAH but not to the close paralogs VASP and EVL was a surprise, also revealed by this structure. A loop fully conserved in all three proteins adopts a different conformation in EVAH due to a surrounding network of non-conserved residues. Application of the computational tool dTERMen was able to identify this network and guided a series of mutations in the EVL EVH1 to definitively demonstrate these specificity determinants imparted high affinity binding to the EVL paralog. 

The PCARE peptide affinity was further enhanced by fusing to its C-terminus a linker and a second motif designed to engage with a previously characterized secondary binding site on the opposite side of the EVH1 domain of VASP EVH1. This secondary site is conserved in EVAH, and indeed the dual-motif PCARE-DUAL peptide bound with nanomolar affinity to EVAH EVH1 domain. 

Weaknesses: 

The structural conclusions are based on the crystal structure of a fusion protein, where the C-terminus of EVAH EVH1 is fused to the PCARE peptide. The main concern is that tethering the peptide to the C-terminus could induce the peptide to bind in the opposite orientation from the canonical binding interaction. 

The function of PCARE in the cell, and the relevance of its specific and tight interaction with EVAH is not explored. Interestingly, there is other recent literature that connects PCARE to actin dynamics. 

The design of PCARE-DUAL and other dual-motif ligands utilized in this study was based on work by another group that characterized the secondary binding site (Acevedo et al., 2017, "A Noncanonical Binding Site in the EVH1 Domain of Vasodilator-Stimulated Phosphoprotein Regulates Its Interactions with the Proline Rich Region of Zyxin", Biochemistry 56(35):4626-4636, doi: 10.1021/acs.biochem.7b00618). This prior work was not credited as the foundation on which their affinity enhancement strategy was built. 

An additional key finding of the Acevedo 2017 paper was that ligand binding to the novel secondary site is inhibited by the phosphorylation-mimicking mutation of Y39 to E, a site known to be phosphorylated by Abl. Since ENAH has Y at the equivalent site and is a reported target of Abl, the bivalent advantage the authors have designed into their dual-motif ligands might be disrupted by phosphorylation. This has implications for their proposed use of dual-motif ligands for "dissecting specific Ena/VASP functions in processes including cancer cell invasion" and "for developing therapies to treat ENAH-dependent diseases."

Reviewer #3 (Public Review): 

In the accompanying manuscript, the MassTitr method for profiling short linear motifs on a proteome-wide scale is described and identifies PCARE as a high affinity binding partner for EVAH EVH1 domain. The unique observation is that contextural sequence surrounding the polyproline motif in PCARE underlies the specificity for its interaction affinity relative to other binding partners, in that a reverse-peptide binding orientation is favored. While this result may, in isolation, be perhaps significant only to a subset of protein chemists, when combined with the description of MassTitr development, it provides an additional example of how the methodology can uncover non-canonical binding interactions. 

Strengths:

The work provides another powerful example of how the MassTitr method can identify non-obvious binding partners with unique interaction features. The biochemical studies that were employed to elucidate the mechanism for this specificity strongly support the conclusions. 

Weaknesses:

These results in isolation may impact only a limited number of protein chemists working on this or related systems; however, when coupled with the general description of the MassTitr method, the impact is much broader. 

One of the EVL mutants that was generated and characterized was Y62C. This variant has the capacity to form inter-chain disulfide bonds since the newly incorporated Cys is unpaired, but this possibility is not discussed or examined.

Reviewer #1 (Public Review):

The methods appear sound. The introduction of vaccines for COVID-19 and the emergence of variants in South Africa and how they may impact PLWH is well discussed making the findings presented a good reference backdrop for future assessment. Good literature review is also presented. Specific suggestions for improving the manuscript have been identified and conveyed to the authors.

Reviewer #2 (Public Review):

Karima, Gazy, Cele, Zungu, Krause et al. described the impact of HIV status on the immune cell dynamics in response to SARS-CoV-2 infection. To do so, during the peak of the KwaZulu-Natal pandemic, in July 2020, they enrolled a robust observational longitudinal cohort of 124 participants all positive for SARS-CoV-2. Of the participants, a group of 55 people (44%) were HIV-infected individuals. No difference is COVID-19 high risk comorbidities of clinical manifestations were observed in people living with HIV (PLWH) versus HIV-uninfected individuals exception made for joint ache which was more present in HIV-uninfected individuals. In this study, the authors leverage and combine extensive clinical information, virologic data and immune cells quantification by flow cytometry to show changes in T cells such as post-SARS-CoV-2 infection expansion of CD8 T cells and reduced expression CXCR3 on T cells in specific post-SARS-CoV-2 infection time points. The authors also conclude that the HIV status attenuates the expansion of antibody secreting cells. The correlative analyses in this study show that low CXCR3 expression on CD8 and CD4 T cells correlates with Covid-19 disease severity, especially in PLWH. The authors did not observe differences in SARS-CoV-2 shedding time frame in the two groups excluding that HIV serostatus plays a role in the emergency of SARS-CoV-2 variants. However, the authors clarify that their PLWH group consisted of mostly ART suppressed participants whose CD4 counts were reasonably high. The study presents the following strengths and limitations

Strengths:

A. A robust longitudinal observational cohort of 124 study participants, 55 of whom were people living with HIV. This cohort was enrolled in KwaZulu-Natal,South Africa during the peak of the pandemic. The participants were followed for up to 5 follow up visits and around 50% of the participants have completed the study.

B. A broad characterization of blood circulating cell subsets by flow cytometry able to identify and characterize T cells, B cells and innate cells.

Weaknesses:

The study design does not include

A. a robust group of HIV-infected individuals with low CD4 counts, as also stated by the authors

B. a group of HIV-uninfected individuals and PLWH with severe COVID-19. As stated in the manuscript the majority of our participants did not progress beyond outcome 4 of the WHO ordinal scale. This is also reflected in the age average of the participants. Limiting the number of participants characterized by severe COVID-19 limits the study to an observational correlative study

C. a control group enrolled at the same time of the study of HIV-uninfected and infected individuals.

D. results that elucidate the mechanisms and functions of immune cells subsets in the contest of COVID-19.

Reviewer #3 (Public Review):

Karim et al have assembled a large cohort of PLWH with acute COVID-19 and well-matched controls. The main finding is that, despite similar clinical and viral (e.g., shedding) outcomes, the immune response to COVID-19 in PLWH differs from the immune response to COVID-19 in HIV uninfected individuals. More specifically, they find that viral loads are comparable between the groups at the time of diagnosis, and that the time to viral clearance (by PCR) is also similar between the two groups. They find that PLWH have higher proportions and also higher absolute number of CD8 cells in the 2-3 weeks after initial infection.

The authors do a wonderful job of clinically characterizing the research participants. I was most impressed by the attention to detail with respect to timing of viral diagnosis as it related to symptom onset and specimen collection. I was also impressed by the number of longitudinal samples included in this study.

Reviewer #1 (Public Review): 

The data in this manuscript describe a previously unseen conformation of CaSR in complex with a nanobody that has been shown to inhibit calcium activation in functional experiments. Structural studies show that the nanobody physically prevents the dimerization of the extracellular domains. Although the structure of the nanobody-bound, putative inactive conformation is at relatively low resolution, the global conformational changes can be resolved. Such a nanobody may prove to be very useful as a antagonist of CaSR activation. 

Additional information and controls are required to support some of the main claims. The mutants are tested using an intracellular flux assay, but there are no controls to show that the mutant proteins have expressed and trafficked to the membrane. For mutants that inhibit function, cell surface expression controls are essential to establish that the effect of the mutant is specific. 

Additional details are required to evaluate the cryo-EM data, such as an FSC curve and the distribution of viewing angles of the particles used in the reconstruction. To evaluate the maps compared to the model, more extensive maps are required, especially in the regions where calcium ions and ligands are modelled. (Although the maps for the ligands are shown the surrounding regions are not). In Figure 1-Fgure Supplement 2, this appears to be a surface rendering rather than a map as indicated in the title of the figure. This rendering is more featureful than the provided maps, especially for the inactive conformation. In addition, there is no analysis of the quality of the model (bonds, angles, C-beta deviations, etc). 

This is presented as a more valid structure of the inactive state, compared to the "intermediate" state previously reported by Ling et al., 2021, Cell Res. 31: 383-394. This prior model suggested a conformational ensemble of inactivated states with varying degrees of separation of the extracellular domains. However, it is possible the nanobody in the present study simply stabilizes one state from this conformational ensemble and that there are other inactive states as well, such as those reported by Ling et al.

Reviewer #2 (Public Review): 

This manuscript reports the structure of the CaSR in both inactive and agonist+PAM bound states. Although a recent paper by Ling et al. (Cell Res 2021) first reported the structure of the CaSR in its inactive and active states, the present study adds novel and important information. The authors get good resolution of the CaSR extracellular domain (ECD), up to 3 A, allowing a better view of the Ca and amino acid binding sites. This paper is the first to report one of the important Ca site within the cleft of the VFT domain. Indeed, although several structures of the isolated ECD, and the structure of the full length CaSR were already reported, none really clarified the mode of action of Calcium ions leading to the activation of the receptor. However, the evidence that the density observed correspond to a Ca ion, rather than to another ion, remains to be confirmed. 

The second important information is the description of what is very likely the fully inactive state of the receptor, with both VFTs in the open state, and with a relative orientation leading to a large separation of the lobes 2 of the VFTs, as observed in the apo or antagonist bound states of most mGlu VFT dimers. Of special interest is the report of a nanobody that stabilize the inactive state of the CaSR, that was used to obtain the "inactive" structure. However, the possible influence of the nanobody in specifying this specific state is not being considered. 

The authors also identified a key element involved in the allosteric transition between the VFT and the 7TM domain, involving a contact between the C terminal part of the CRD and the ECL2 loop. This proposal based on the solved structures is validated by functional analysis of various CaSR mutants. 

A intriguing observation is that the 7TM conformation of both subunits within the dimer are almost identical in the "inactive" and ago+PAM state, indicating that the ago+PAM state is unlikely corresponding to the active conformation of the receptor. As such, the later state should not be named the "active" state, but rather the agonist+PAM state to be scientifically accurate.

Reviewer #3 (Public Review): 

In this study, the authors aim at deciphering the structural and molecular mechanisms of the human calcium sensing receptor function. For this purpose, they solved the cryoEM structures of inactive and active states of the receptor and provide further functional/pharmacological evidence to support the activation mechanism proposed from the structural data. Overall the data and approach are robust. It would be useful for the readers to better present/discuss the comparison of their structures with the recent structural data published recently for the same receptor (Ling et al 2021).

Reviewer #1 (Public Review):

By linking several databases, the authors tried to measure the impact of trade on all amphibian species. Thereby origins of species traded, volumes, the purpose for trade have been assessed while noting that several loopholes exist in making overall robust assessments e.g., dynamics in trade and taxonomy. However, indicating that gaps and shortcomings are based on the way current databases available have been set up, the authors did an enormous job in applying the extensive digital methods to achieve the best possible reflection of the current amphibian trade. Through the use of several software programmes to measure/analyse current databases, also figures could be built to visualize e.g. trends. This is one of the major strengths of this paper. With the various specifically named methodological queries as well as the use of categorized keywords, which were essential for the inclusion and linking of the various databases, unambiguous results could be generated in the best possible sense, on the basis of which correct and convincing recommendations were made. To make it explicitly clear again, due to the lack of data on the global anthropogenic use of amphibian species, this seemingly complex applied methodological approach was necessary to shed light on the dark. In return, the effort would have been many times less, there would have been more comprehensive and informative databases that transparently and up-to-date illuminate the population status of species, their threats and the impact of trade. The importance of this work is essential to understand how much the various interest groups are lagging behind in order to communicate responsibly and transparently the use of resources, in this case the most threatened group of vertebrates, amphibians; thus, it is difficult to understand as the reader learns here how easy it is to trade species that have already been classified as threatened.

If I would have to mention weaknesses of this paper there are none that I would address explicitly. Apart from the comprehensive Suppl. Mat., I would just not overload the actual manuscript with figures and make sure that these are self-explanatory. As already mentioned, the methodological part in particular is very extensive and complex, but this is essential for this type of study.

Reviewer #2 (Public Review):

While it is wildly assumed that the trade in wildlife is well documented and data are thorough and widely available, this is not the case. The authors scour online sources (databases, websites, marketplaces), in multiple languages, to assess the true extent of wildlife trade related to those values reported and find large discrepancies. Wildlife trade has both direct and indirect effects on wild populations of amphibian species and therefore having more accurate values is essential for measuring potential effects. They call for change in how data are collected and reported so that those data can properly influence policy and conservation measures.

Reviewer #1 (Public Review): 

The authors demonstrate deficits in perceptual tests related to fine-time perception in non-speech and speech sounds in a group of patients with stroke aphasia compared to a control group without a lesion. A subgroup of patients with deficits in spectrotemporal processing at a fine timescale have lesions mapped to the posterior STS, MTG and adjacent white matter. The area associated with deficits in spectrotemporal analysis with a fine timescale is then used as a seed for probabilistic fibre tractography based on diffusion MR. These results show connectivity of the functionally defined seed region with a number of areas including the cerebellum. 

The work is carefully done and I think interesting in demonstrating the cerebellar connections of the functionally defined region associated with deficits in fine temporal analysis that might be a basis for event representation at this temporal level.

Reviewer #2 (Public Review): 

Based on consideration of supportive evidence in the literature, the authors propose that a cerebellar-temporal lobe functional network plays a key role in auditory temporal processing. The precise parsing of temporal information is critical to understanding dynamic auditory processing and thus is an interesting area of study. Better understanding of how the cerebellum and temporal lobe may interact to achieve such parsing of the dynamic signal in a generative/predictive internal model is of clear interest to a broad readership. This idea is put to the test by first having individuals with lesions in the posterior portion of left temporal lobe perform speech perception and timing tasks and comparing performance with 12 healthy controls to establish the role of this region in tasks reliant on intact fine temporal processing. Typically, a lesion model will be helpful when a dissociation between structure and function can be demonstrated, and preferably this would be a double dissociation. Here, while lesions to auditory regions of the left temporal lobe are associated with impoverished performance on speech and temporal order tasks relative to a healthy control group, performance on comparably difficult auditory tasks that do not require good temporal discrimination is not tested to determine if there is such a dissociation. Given the extensive discussion of hypothesized different time sensitivities of right and left auditory cortices in the Introduction, patients with right homologous lesions might also have served as an interesting control and could have supported a double dissociation. In a second step to their study, a seed region was generated based on comparison of the lesion loci for the half of the patients who performed most poorly on the behavioral tasks to the other half, and this was used to explore anatomical tract connectivity of the seed region to the rest of the brain in the neuroimaging data from the healthy controls, with a focus on connections with the cerebellum. This approach to establishing that "temporo-cerebellar connectivity underlies timing constraints in audition" is unfortunately just not that convincing. The data are interesting, but taken alone they simply do not support such a conclusion. In the data, there is no clear functional link established or even hinted at between the temporal lobe and the cerebellum.

Reviewer #3 (Public Review): 

Stockert et al. investigate the cortico-cerebellar network underpinning rapid temporal auditory analysis. This study uses a well-defined group of stroke participants with mostly circumscribed lesions to the left posterior superior temporal lobe to motivate probabilistic tractography from cortical regions associated with verbal and non-verbal rapid auditory temporal analysis. Lesion-symptom mapping identifies a specific region of the posterior superior temporal sulcus and underlying white matter as statistically associated with impairment in rapid auditory temporal analysis. Tractography results demonstrate that these regions have high structural connectivity to wider regions of the left hemisphere cortical language network and ipsilateral and contralateral connectivity to postero-lateral cerebellum and dentate nucleus. It is interpreted that this cortico-cerebellar network is crucial to developing representations of fine auditory temporal structure. 

The conclusions of the paper are an interpretation which is based on integrating previous neuropsychology with the current tractography results and based on well-defined models in the motor domain. Such conclusions are not unreasonable but there is no direct (associative) evidence linking this network to the cognitive function of interest. 

Strengths: 

The paper integrates neuropsychology and neuroimaging methodologies to build a coherent picture which is more than the sum of its parts. The stroke group has well-defined and selected lesions which enable testing of the hypotheses put forward by the authors. The behavioural measures are sensitive and suitable to identify impairments in the behaviours of interest. There has been a detailed analysis of the behavioural speech perception data in the stroke group which largely, although perhaps not entirely, conforms to the asymmetric temporal sampling hypothesis. The lesion-symptom mapping approach is suitable for the nature of the population (small group with similar lesion distributions) and has allowed neuropsychologically guided tractography in the neurotypical population. This has clearly illustrated the complexity of the structural connectivity of the posterior superior temporal sulcus and underlying regions. 

Weaknesses: 

The selective nature of the stroke population - relatively small, chronic lesions - has resulted in only mild impairments for a small number of participants (6/12 participants). At the group level there is no difference between the stroke and neurotypical population on speech perception measures - group statistics do not reach one tailed significance. This reduces the certainty with which the regions identified are associated with the behaviour or interest. However, the results do conform to previous neuropsychology and lesion studies and it is likely that this lack of effect is due to low statistical power. 

All the stroke participants have a similar lesion distribution, and this makes lesion-symptom mapping challenging. For example, lesion data do not give an indication of the functional integrity of perilesional regions which can be reduced, even at the chronic stage, therefore the superior temporal sulcus may not be functioning effectively, even in the proportion of the group without lesions to this area. Lesion symptom mapping is more robust with a wider distribution of lesions and the inclusion of participants with lesions remote from the area of interest. Having said that, the behavioural measures appear sensitive enough to identify mild impairments and the authors, for good reason, wished to reduce the extension of lesion into primary auditory regions. As above, given the limited sample and homogeneous lesion, the lesion symptom mapping approach is reasonable. 

The authors suggest that the behavioural results conform to the asymmetric temporal sampling hypothesis in that only place of articulation discrimination impairments in the stroke group can be (just about) detected, whereas there were no significant stroke-neurotypical differences in other phonetic contrasts. It is not clear that the VOT differences associated with plosive voicing changes and the cues associated with place changes happen over fundamentally different time-scales and, therefore, it is important to further justify the interpretation of the data. In the future it will be helpful to have this level of analysis applied to individuals with lesions to the wider speech perception network to draw conclusions about the specificity of the impairment to these regions - for example, impairments in phoneme discrimination have been associated with frontal lobe lesions. 

The tractography results reveal a complex pattern of structural connectivity, including other regions associated with speech perception. The authors have a theoretical motivation to focus on the importance of the temporo-cerebellar pathway but there is no correlation evidence to link auditory temporal analysis to the integrity of this pathway in the neurotypical population. The non-verbal measures appear to be sufficiently sensitive for this type of analysis. This lack of association with behaviour makes it hard to draw conclusions about the functional role of this network. 

Despite some limitations, this paper highlights the potential importance of a previously under-studied network supporting auditory perception. The paper is, therefore, impactful in motivating future research into the cerebellar contribution of auditory processing.

Reviewer #1 (Public Review):

Strengths

This study is a technical and analytical tour de force. The evolution experiments with barcoded lineages involved an immense amount of work and clever design, and the scale of the data challenged the authors to develop new statistical summaries. The figures are clear and results easy to interpret, even outside the evolution-experiment bubble. While the essential findings are not especially surprising, the robustness enabled by this level of replication is appreciated.

Weaknesses

I'm not exactly sure what I learned. I'm biased to like this work and while I'm confident that if I studied these findings more I would learn more, it wasn't obvious. For example - I want to know more about the effects of ploidy on pleiotropy, and while there are some differences e.g in Figure 4A, I don't know what these PCs actually are saying. If particular phenotypes associate with PC's, it'd be helpful to "load" them on these axes. Also, do some treatments lead to faster or more complete diminishing returns than others, and does this influence pleiotropy? In total I think this manuscript can be improved by being presented / read by others, which is the job of peer review but here I think it's also to broaden its implications.

Reviewer #2 (Public Review):

The pleiotropic adaptive effects of mutations have been extensively characterized, with numerous examples of both specialist and generalist adaptations. However, it is less clear whether and how these pleiotropic patterns change over the course of evolution for a given evolving species. The authors attempt to answer this question by characterizing the pleiotropic fitness effects of ~150 adapted S. cerevisiae populations at 200-generation intervals over 1000 generations of evolution. The authors provide strong quantitative evidence that patterns of pleiotropy are highly dependent on the evolution environment, and changes substantially over relatively short evolutionary timescales (1000 generations). They further show that even with a fixed genotype and selective environment, replicate populations can substantially vary in their pleiotropic profiles (again, in an environment-dependent manner), and this variation tends to increase over evolutionary time. The authors thus provide substantial evidence that the evolution of "generalist" or "specialist" types is not deterministic but rather strongly dependent on both the specific evolving system and evolutionary stochasticity.

The authors use DNA barcodes to uniquely tag each of their evolving populations, then use bulk fitness assays to measure the fitness of each population at 5 different timepoints across five different environments. These methods are well established, and appear to have been implemented correctly. The resulting data clearly answers the proposed question, and supports the major claims of the paper.

Most previous studies of the pleiotropic effects of adaptation characterize a relatively small number of independently evolved lines, limiting our ability to draw quantitative conclusions (but see Kinser et al 2020 eLife). Furthermore, I am unaware of any large-scale study characterizing how the pleiotropic effects of mutations change over evolutionary time. While the specific experiments and data are not of use to the broader community, the major findings substantially advance our understanding of the evolutionary process and open up new avenues of both theoretical and empirical research.

Reviewer #3 (Public Review):

Bakerlee, Phillips et al. investigated the dynamics of pleiotropic effects of laboratory adaptation in yeast. They do this by experimentally evolving 172 yeast populations in three different laboratory conditions. After which, they examine the dynamics of (environmental) pleiotropy by tracking the relative fitness change of each population in a focal and alternative environment (a total of five measurement environments).

Even though various experimental evolution studies examined the same question (some of which testing it on the same environmental conditions and the same model organism), this manuscript stands out with a significant improvement in its experimental design, i.e., testing the pleiotropic effects of local adaptation on a large number of replicate populations across a series of evolutionary time points. As a result, they could observe that the pleiotropic effects show a considerable variation within treatment groups, and this variability is more pronounced in away than in home environments.

The question of how adaptation to a stable environment affects the evolutionary fitness of populations in a different environment represents a crucial research topic. This work displays the presence of considerable variability in the evolution of pleiotropy, stressing the importance of chance and contingency. Overall, this is a nice study leveraging the recently developed microbial barcoding techniques, but there is still room for significant improvement to strengthen and clarify the results and conclusions.

Reviewer #1 (Public Review): 

The manuscript is clearly written and was a pleasure to read. Rigorous example data from several of the mouse lines were presented, nicely illustrating their utility. The manuscript could be improved by an open discussion of potential problems/limitations with using the transgenic mice described here.

Reviewer #2 (Public Review): 

This paper is well-written and discusses an important and efficient approach to generating optogenetic mice. The major strength of the paper is the combination of optogenetic activators and sensors that are expressed in complex tissue networks. The ease of breeding to generate these mice is an important step that will allow testing of the interactions between cells in complex networks. Generation of lines that enable ratiometric measurements of Ca2+ are also important development that should allow better assessment of intracellular Ca2+ levels. 

A weakness of the paper is that it is essentially a methods paper that describes the characteristics of the mice and gives proof-of-principal examples of various combinations of activators and sensors in cardiac, smooth muscle and endothelial cells. The paper lacks applications that test novel hypotheses and demonstrate new findings. That being said, it is likely that the mice described and available to all investigators, will provide myriad new experimental opportunities. It will be up to the creativity of investigators at large to devise these applications. 

While the breeding tactics to generate experimentally useful chimeras are clear, it would also be helpful to compare relative expression/signal strengths for Cre recombinase driver mice vs. the monoallelic mice promoted in the present study. Comments about breeding success/difficulties would also be useful for those planning experiments using combinations of the mice from CHROMus.

Reviewer #3 (Public Review): 

The authors describe the generation of a battery of monoallelic homozygous transgenic mice expressing optogenetic effectors and sensors in specific cells primarily in the cardiovascular system. The paper potentially provides a relatively simple framework for expressing optogenetic reporters and effectors in any desired cellular system and/or disease model. An important advantage of this approach is that a simple breeding strategy can be used to produce animals expressing both sensor and effector in the desired cell-type. The paper provides convincing direct evidence of the utility of the resulting effector/sensor pairs in a variety of cell types. The development and demonstration of utility of animals expressing "ratiometric" cytoplasmic Ca2+ probes is particularly impressive. These probes are demonstrated to have advantages for challenging in vivo imaging where movement artifacts can confound interpretation. These mice are freely available and will be a powerful and valuable resource to the community.

Reviewer #1 (Public Review): 

In this paper, Bali et al. use an assay in which alkaline phosphatase-tagged cell surface proteins are used to stain live embryos overexpressing other cell surface proteins to identify Sticks and stones (Sns) as a possible ligand for the Lar receptor protein tyrosine phosphatase. Direct binding between the two proteins was confirmed in vitro, although it requires extensive multimerization of Lar, and is conserved between PTPRF/D and Nephrin homologues. The authors use GAL4 insertions to compare the expression patterns of Lar and sns in the nervous system, and observe genetic interactions in transheterozygotes as well as similar knockdown phenotypes. They conclude that Lar and Sns both act in motor neurons to promote NMJ growth, but act in different neuronal populations to control mushroom body morphogenesis and R7 photoreceptor axon targeting. 

Although it would be of great interest to confirm a receptor-ligand interaction between Lar and Sns, this paper falls short of that goal. Transheterozygous phenotypes are suggestive but do not prove that the two proteins act as a receptor-ligand pair. The physical interaction is quite weak and the evidence that Sns acts in the cells most likely to provide a Lar ligand is not compelling. 

First, Sns has a close homologue, Hibris. The authors do not examine whether Lar also binds to Hbs, or specifically interacts only with Sns. 

Second, only a single allele of sns is used to generate tran-sheterozygotes. It is possible that this chromosome carries another interacting mutation or even an allele of Lar. 

Third, the authors show that loss of Sns has a similar phenotype to loss of Lar in NMJ growth and mushroom body development, but do not demonstrate this for R7 axon targeting. 

Fourth, it is possible that the sns phenotypes are due to interactions with its known binding partners, Roughest and Kirre, and the transheterozygous interaction with Lar simply reflects the involvement of both Sns and Lar in the same process. The authors should more directly test the ligand hypothesis.

Reviewer #2 (Public Review): 

This manuscript from Bali, Lee and Zinn addresses a long-standing mystery in the receptor protein tyrosine phosphatase (RPTPs) field regarding the conserved extracellular protein binding partners of the LAR subfamily of RPTPs. Although mammalian LAR-family proteins bind and interact with several ligand molecules, and LAR-family receptors associate with heparan sulfate across species, the question of what protein ligands might represent the conserved partners of this well-conserved set of RPTPs has lingered. The authors present an exciting discovery that the Sns/Nephrin family of transmembrane proteins bind to LAR-family RPTPs in situ and in vitro. A powerful in situ tissue-binding assay that can overcome low-affinity interactions was developed by this group for screening and analysis of new ligands. Compelling Drosophila binding data implicating Sns as a candidate LAR ligand led the authors to analysis of mammalian Sns (Nephrin) for in situ binding in the same assay. By creating MIMIC reporters of the endogenous transcription units, the authors show that Sns and LAR are both expressed in larval motor neurons, where they are both required to promote NMJ morphogenesis. Consistent with the binding and very similar NMJ phenotypes of Sns and LAR, transheterozygous mutants show a very clear 'synthetic' phenotype strongly suggesting the type of strong functional interdependence characteristic of genes in a common signaling pathway. In contrast to the cis requirement in motor neurons, the authors discover that other regions of the nervous system display largely complementary expression of the reporters, raising the possibility that Sns and LAR act in trans in the larval mushroom body, and the visual system. The authors go on to confirm the same type of strong genetic interaction between Sns and LAR in these other contexts, leading to a working model that the two proteins function as a ligand-receptor pair in these other circuits at different stages of larval and pupal development. The data are very high in quality and the manuscript is clearly written. The study represents an exciting advance in this field from a leading lab in RPTP and receptor-ligand function in the fly nervous system.

Reviewer #1 (Public Review): 

In this manuscript by Gilbert et al., the authors found that MLC1 is required for postnatal maturation of perivascular astrocyte coverage. Through various detailed experiments, they further found that Mlc1 KO mice showed a number of defects, including the reduced VSMC contractility, neurovascular coupling and parenchymal CSF flow. The data is well presented, however there are several points that need to be addressed to strengthen the manuscript. 

1) Since many PvAP proteins showed the normal expression after P60 in Mlc1 KO mice, it is also possible that many of the phenotype that the authors presented, such as the reduced VSMC contractility, neurovascular coupling and parenchymal CSF flow, can be recovered after P60. This would be important as well to understand the prime pathological cause of megalencephalic leukoencephalopathy induced by MLC1 deletion. 

2) Does CBF get differed only after neuronal stimulation in Mlc1 KO mice? It is unclear whether the basal CBF/neurovascular coupling level is disrupted as well in Mlc KO brains and how this defect is related to the reduced vasoconstriction in these mice. 

3) The reduced cohesiveness of PvAPs and the associated neuronal fibers to the vessel in Mlc1 KO brains should be validated with additional experimental approach. 

4) The defective polarity of astrocytes should be better described by using other markers other than GFAP. The distribution of Aquaporin4, Cx43 or several glutamate transporters in the specific compartment of astrocytes can be examined. 

5) The authors provide interesting observations such that the formation of perivascular astrocyte coverages is required for the dissociation of the contacts between neuronal components and the vessel during development. The authors need to discuss more about potential regulation and implication of this phenomenon. 

6) It is interesting that DOTA-Gd tracer shows different traces in Mlc1 KO brains. However, it is unclear how MLC1 deletion affects glymphatic system. Does the tracer normally enter to the perivascular spaces in Mlc KO brains? Does the tracer leak out more from the perivascular spaces in Mlc1 KO mice? Is the general clearance or drainages of the tracer impaired in Mlc1 KO mice? Would these defects be originated by the reduced perivascular astrocyte coverage or the reduced vasoconstriction itself?

Reviewer #2 (Public Review): 

This very interesting manuscript by Gilbert and colleagues uncovers that the astrocyte specific membrane protein MLC1, the mutation of which causes a rare disease called megalencephalic leukoencephalopathy with subcortical Cysts (MLC), plays a fundamental role in the postnatal development of the gliovascular unit and the organization of the perivascular astrocyte processes, in particular. To reach this conclusion, the authors used an elegant multiscale approach including in vivo MRI, in vivo functional ultrasound, ex vivo analysis of vascular constriction, anatomical approaches at the light and electron microscopic level, and molecular characterization of the gliovascular unit from isolated microvessels. The manuscript is very well-written although it uses too many (unnecessary) abbreviations, which prevents a fluid reading of the manuscript, results are well illustrated and convincing and the discussion is reasonable. 

I have a major concern regarding the results reported in Figure 4D, which seem somewhat contradictory to those shown in Figure 6A-F. Indeed, the authors report in Figure 4D that there is less Neurofilament-M protein around isolated microvessels in MLC1 KO mice, whereas Figure 6A-F shows that these animals have more neuronal processes in contact with the vessels than in wiltypes. How can the authors explain this?

Reviewer #3 (Public Review): 

Astrocytes unique to the central nervous system such as the brain is known to play a role in maintaining BBB integrity by surrounding brain vessels tightly vis their end-feet structure. However, how astrocytes form end-feet along brain vessels and how they regulate gliovascular function are largely unknown. The present manuscript by Gilbert et al. describes how perivascular astrocytic processes are established during postnatal development and pinpoint a player involved. Moreover, it revealed dysmorphic astrocytes could affect neurovascular coupling and CSF transport. They find that MLC1 deletion resulted in disorganized perivascular astrocytic processes and defective contractility of vascular smooth muscle cells, thereby leading to impaired neurovascular coupling and intraparenchymal fluid clearance.

Reviewer #1 (Public Review):

Identifying private peptides for generating personalised cancer vaccines is a promising approach to launch robust anti-tumor response; however, the challenges remain in developing an effective process to achieve that. In this manuscript, the authors present an interesting and powerful pipeline (PeptiCRAD) to achieve this goal by examining CT26 model. Overall, this manuscript is well written and presented. Despite that this work presents interesting findings and pipeline, I have the following concerns. I do feel that this manuscript will improve if these concerns can be addressed.

1. It will be critical to confirm TILs and T cells in draining lymph nodes indeed recognise the peptide used in Figure 7-8 by ELISPOT of IFNg.

2. It would be interesting to see if this pipeline can be used to identify human peptides in human melanomas.

Reviewer #2 (Public Review):

Summary:

In Feola et al., the authors applied an immunopeptidomic-based pipeline to discover tumor antigens that can be incorporated into an adenovirus/vaccine platform to potentially cure Balb/c mice with established CT26 tumors. The authors performed immunopeptidomic experiments to isolate Kd/Dd-associated peptides, selected the most promising peptide candidates based on differential mRNA expression (mTEC and normal colon as control) and sequence similarities with pathogens (NEX software). Then, they tested a list of 26 peptide candidates in vitro (ELISPOT), selected the 6 most immunogenic peptides and tested their therapeutic potential in vivo using their PeptiCRAd vaccine platform in Balb/c mice with established CT26 tumors. Finally, they measured tumor growth and analyzed tumor infiltrating T cells in vaccinated mice bearing tumor cells to confirm the feasibility of applying their pipeline for the generation of therapeutic oncolytic cancer vaccine.

Strengths of the methods and results:

The pipeline appears to be straightforward. A clear strength of the study is that the authors applied immunopeptidomics to measure MHCI-peptides that are genuinely presented on the cell surface. The method provides physical measurements of MHCI-peptides using mass spectrometry. This is a direct approach that differ from the application of MHC-peptide binding prediction algorithms, which are cheap/highly accessible and useful but often lead to relatively large numbers of false positive peptides that are non-immunogenic and therapeutically irrelevant. Another strength is that the authors tested their peptide candidates in vivo (in mice) using their PeptiCRAd vaccine platform.

Weaknesses of the methods and results:

The study provides a complete pipeline, from antigen discovery to pre-clinical validation in mice using a vaccine platform. This could be viewed as a strength, but there are important limitations in each component of the pipeline, making the overall pipeline relatively suboptimal. The rational for the selection of the peptide candidates could be better explained and additional controls could be used throughout the study. To what extent the selected tumor antigens are tumor-specific is unclear. The therapeutical potential of HEX-identified peptides compared to previously reported TSA (mutated or aberrantly expressed) is not evaluated in this study. Most importantly, the anti-tumor response measured in vivo appear to be MHCIpeptide-independent since the uncoated adenovirus (VALO-mD901) improved tumor growth control in the infection site, raising doubts about the overall approach. No conceptual advancement is provided by the study.

Is the aim achieved, and results support their conclusions?

There is a lot of work in this study. The authors aimed to show the feasibility of applying their pipeline to identify tumor antigens, and then test the therapeutic potential of those peptides in mice; this is achieved. However, it would be important to not oversell the pipeline since many tumor antigens that have been identified are non-immunogenic and have little impact in vivo. The current pipeline is a good start but still need significant improvements to rationalize the peptide selection process and to clearly understand why those peptides are, or are not, immunogenetic in vivo. This is important as the authors envision the development of a robust platform that can be successful in the clinic.

Impact and significance of the work:

The work shows that immunopeptidomics is useful to identify tumor peptide antigens. The peptide selection is also based on viral molecular mimicry (using the HEX software), which is relatively new in the field and certainly worth exploring by others. I envision that this type of pipeline can be useful in the context of a biotech company that aims to rapidly identify and test potential peptide targets.

Reviewer #3 (Public Review):

The goal of the study was to establish a first-in-the-world personalized oncolytic viral vaccine pipeline that covers the entire process, from the identification of candidate vaccine targets from the primary tumor to the in vivo efficacy testing.

By using their pipeline, the authors identified peptides that could be successfully loaded onto the oncolytic virus. In addition, the generated oncolytic vaccine was effective not only in the treated tumor but also in the untreated distant tumor. These observations indeed demonstrate that their pipeline platform can successfully generate oncolytic viral vaccines that can induce a strong systemic anti-tumor immunity.

However, this study's major weakness is that it has no experimental data that support their conclusion that this pipeline can be translationally applied to the clinical setting as "PERSONALIZED" vaccine development.

Personalized medicine, by definition, is a treatment that is tailored specifically to a particular patient. One of the challenges in developing effective personalized medicine for cancer patients is how to utilize the materials taken from the patients to generate the effective treatment in a timely manner so that it can be therapeutically utilized before the patients reach the untreatable stage of the disease.

While the pipeline demonstrated in the study was effective, the authors had access to the primary tumor material, an immortalized transplantable tumor cell line grown in vitro, to identify the peptide ahead of time, then to treat cancer developed by injecting the same cell line into the mice. Although this perfectly validates the pipeline as a capable system to induce anti-tumor immunity, it fails to demonstrate the challenges of applying this as personalized medicine in the future, as they concluded.

Reviewer #1 (Public Review): 

General overview and merit of academic rigor:

Xu et. al put forth an innovative experimental pipeline to examine the connections of the raphe nuclei. This manuscript details elegantly designed viral tract-tracing methods coupled with fMOST intact imaging and sophisticated analyses. All figures are of good quality. The studies presented in the current manuscript will be a valuable contribution to the field, therefore an enthusiastic recommendation for publication is endorsed presently. However, there is a cluster of revisions and clarifications warranted before publication. 

Major concerns: 

1. The manuscript's English needs to be proofread extensively for readability and clarity. 

2. The term MR (median raphe) is used in the atlas of Paxinos and Franklin. But, the entire study follows the Allen Reference Atlas nomenclature, in which the same raphe nucleus is called the "Superior center nucleus" (CS). To keep consistency, I suggest using "CS" instead of "MR". Alternatively, in the Introduction, please make a clear statement that the MR is equivalent to CS in the Allen Reference Atlas. 

3. In the Introduction, it is unclear the rationale behind the decision to selectively study the DR and MR here (why other raphe nuclei are not included?). 

4. In the Results, I did not find any figure panel or images to show the anatomical location of the MR. Figure 1 shows only one injection site in DR. It is necessary to also show at least one representative injection site in the MR. 

5. This study is designed to map the input/output of two major populations of neurons (Glu+ and GABAergic) in the DR and MR using two cre-driver lines (Vglut2-cre and Gad2-cre). Please clarify how these two cre lines were characterized and whether those cre expressions are consistent with endogenous gene expressions. What are their distribution patterns in the DR and MR? Are they intermingled or relatively segregated? How are their distributions in comparison with that of serotonergic neurons? 

6. Overall Discussion is not well organized. I suggest to start with a clear statement about the novel discoveries of this study in comparison with existing literature, and then compare the overall input/output patterns of Glu+ and GABAergic populations in the DR and MR. The current discussion focuses on a few major targets (i.e., CEA, LH), but missed a big picture. Additionally, it is necessary and important to carefully compare their connectivity patterns with that of serotonergic neurons in these two raphe nuclei. 

Minor concerns:

1. The Impact statement reads, "We reconstructed the input-output circuits of glutamatergic and GABAergic neurons in the dorsal raphe nucleus and median raphe nucleus and proposed a more refined model of the habenula-raphe circuit." When a comparison like this is put down, a specific reference to what your method is more refined than is required. This is well explained in lines 242 and 243, "Based on the conventional model of the habenula-raphe circuit (Hikosaka, 2010; Hu et al., 2020), we proposed a more refined model of the habenula-raphe circuit (Figure 5C)." Make a similar claim earlier in the Impact statement. 

2. For Figure 2A, it would be easier on the reader if inputs for each region (DR and MR) and each plane of the section were placed on the same image akin to the inputs presented on coronal maps in Figures 2B and 5A and the inputs/outputs for each region (MR and DR) in the sagittal summary diagrams in Figure 7. 

3. It is unclear what the nonsignificant grey open circles represent in Figures 3A-D; 4D and E. 

4. In Figure 4A, the imaging portion would be clearer if it read "optical sectioning." 

5. In Figure 7A and B, the position of ACA on the flatmap looks odd to me (it is a little bit too caudal).

Reviewer #2 (Public Review): 

This work from Xu et. al. "Whole-brain connectivity atlas of glutamatergic and GABAergic neurons in mouse dorsal and median raphe nucleus" provided a comprehensive brain-wide analysis for input and output patterns to/from specific DR/MR neuronal populations in adult mouse brain. With exceptional strength in experimental approaches for high quality whole brain imaging that this group is famous for, their data and analysis are thorough and convincing for the general conclusion of the manuscript for describing both convergent and divergent patterns of DR/MR connectivity. While the current study is based on structural but not functional correlation analysis, the results are validated with prior knowledge of the field. It will provide a more complete picture to facilitate future investigation of DR/MR connectivity and physiological functions. 

The work would provide a significant and useful knowledge for the field, while also promoting the generation and application of advanced brain-wide profiling resource to advance board neuroscience research topics. However, there are still a few technical and analytical concerns that need to be addressed or discussed to refine the conclusions. 

Major concerns: 

1. For targeted injection-based analysis, it is critical to carefully analyze and discuss on-target vs off-target rates of labeled cells in DR/MR to validate the datasets. Whole mount data would best fit for such accurate analysis not possible before. 

2. It is also important to know what percentage of the cells get labeled over individual samples, and how many samples and in total what coverage/saturation over the entire anatomical structure has been achieved to justify a complete/comprehensive analysis. 

3. Further on last point, the labeling rates need to be small enough to warrant a more meaningful analysis in Figure 6. From another aspect, is there any anatomical correlation of the target sites in DR/MR for the distinct input/output clusters? This can probably be best addressed with single neuron resolution analysis that this group is good at. For the current study it is a vital part to include this detailed information for better resource to the field (e.g. to guide or map to future spatial transcriptomic analysis to study molecular-cellular correlations).

Reviewer #3 (Public Review): 

Xu et al utilize retrograde and anterograde viral tracing in Cre-transgenic mouse lines to map the inputs and outputs of glutamatergic and GABAergic neuronal populations in the dorsal (DR) and median raphe (MR) nucleus. The experiments generate a large anatomical dataset which the authors analyse with correlation analysis, revealing subtle differences in connectivity patterns between the targeted cell types and nuclei. The study furthermore focuses on the lateral habenula (LH) to raphe nucleus circuit, identifying large amounts of inputs from the LH to both glutamatergic and GABAergic DR and MR populations, but scarce projections from these cells back to the LH, with some cell-type specific differences. In particular, MR glutamatergic neurons send the strongest projections to LH among the targeted populations, supporting previous studies which identified this pathway as playing a role in aversive behaviors. 

Overall, this study nicely complements previously published work on whole-brain connectivity of the DR and MR which have chiefly focused on the main neuromodulatory neurons found in these nuclei, ie. serotonin and dopamine neurons. Some of the experiments in the study are not completely novel, such as input tracing to GAD2-expressing neurons in DR (Weissbourd et al, 2014). However, comprehensive side-by-side comparison analysis between glutamatergic and GABAergic connectivity of both DR and MR nuclei has not been performed before, and will provide a welcome resource to circuit neuroscientist looking to elucidate functional circuits of the raphe nuclei. A further strength of the study is the high-resolution 3D imaging, revealing three distinct projection pathways from MR glutamatergic neurons to LH. 

Two main concerns regarding the study are:

1) The authors do not sufficiently justify the use of Vglut2 as a marker for glutamatergic neurons in DR and MR. The majority of previous studies, especially of the DR, use another glutamatergic marker which is more specifically expressed in the raphe nuclei, namely Vglut3. Vglut3 is much more anatomically restricted to the DR and MR (but has also been shown to partially overlap with serotonergic expression). In contrast, Vglut2 is very broadly expressed throughout the brain and in regions adjacent to DR and MR. For this reason, and from the data in the main manuscript as well as raw microscopy images provided in the accompanying website, it is unclear how specific the starter neuron targeting really is. The authors should show more detailed starter neuron analysis for both the broadly expressed Vglut2 and Gad2 in the DR and MR, showing the histology of the helper virus BFP and RV-ΔG-EnvA-GFP, their anatomical locations, and some quantification of proportion of starter cells within DR/MR (Fig 1B-C shows it only for Vglut2, but in insufficient detail). Furthermore, a rationale for using Vglut2 instead of Vglut3 would be appreciated, especially given that the vast majority of functional studies of the DR have used Vglut3. 

The authors also miss the chance to characterize the topography of Vglut2 and Gad2 starter cell expression within the DR and MR and emphasize the interesting differences between these two populations, which may be relevant to the differences in input and output connectivity. 

2) The quantification throughout the manuscript refers to the relative proportion of inputs or outputs for each cell population and nucleus. The manuscript would be strengthened by also including total cell counts for starter cells in each group, as well as total numbers of input neurons. For example, is the Vglut2 population in DR much larger than the Gad2 population, and do DR Vglut2 neurons receive more inputs in total than DR Gad2 neurons? Including raw numbers would provide concrete information to contextualize connectivity patters between cell types and nuclei to the readers.

Reviewer #1 (Public Review): 

In this study, Sharaf et al., characterize the role of the NmMetQ, a substrate binding protein (SBP) in Neisseria meningitidis, a gram-negative bacteria. Typically, SBPs are localized in the periplasmic space and play the primary role of sequestering rare substrates like amino acids, in this case methionine. Once bound, they bind to a receptor transporter, usually from the ATP binding cassette (ABC) transporter family, which hydrolyzes intracellular ATP to enable uptake of the substrate across the inner membrane. However, the SBP NmMetQ has raised some questions, since the protein has been found on the outer membrane of the cell and is a contender for a surface antigen for the development of meningococcal vaccines. To consolidate this finding, the researchers study NmMetQ to determine whether it still participates in modulating methionine transport. Through the use of the signal peptide server SignalP5.0, they hypothesize that NmMetQ is actually a lipoprotein, which they confirm by expression in E. coli and by mass spectrometry analysis. Then, to examine whether NmMetQ couples to the activity of the ABC transporter NmMetNI, they purify the transporter and measure ATPase activity in detergent micelles with and without NmMetQ. They find that ATP hydrolysis is strongly coupled to the presence of NmMetQ bound to methionine, indicating a preserved functional interaction between these proteins. Next, they investigate the binding of different methionine analogues to NmMetQ by "Fluorine chemical shift Anisotropy and eXchange for Screening (FAXS)", a rigorous screening method that shows a preference for the L-methionine form. Finally they determine cryoEM structures of the NmMetNI transporter at 3.6 Å resolution, and the NmMetQ complexed structure at > 6 Å. While the structural details are low for the complex, they are able to confirm binding as well as the global conformation. The studies clearly demonstrate that NmMetQ can act in the canonical mechanism as a sequestering protein that couples to activity of the ABC transporter. However, the discovery that these proteins are lipoproteins, and the analysis that other Proteobacterial families may possess similar lipoproteins, opens up a new area of biology, paving the way to a better understanding of why these SBPs are also surface localized. Altogether, the research that is presented is clear and rigorous, and substantially increases our understanding of SBP-transporter partnerships. This is a thoroughly interesting discovery.

Reviewer #2 (Public Review): 

To address the roles of NmMetQ protein, the authors used multiple biochemical and biophysical techniques to characterize the structure and function of NmMetQ without and with its cognate ABC transporter NmMetNI. However, considering the similar substrate binding protein EcMetQ from E. coli has been experimentally verified to be a lipoprotein, the major conclusion of this manuscript is not particularly novel. Besides, the authors should address some points to further strengthen their conclusion. 

Major points: 

1) The LC-MS results suggest that the recombinantly expressed and purified lipo-NmMetQ protein has lipid modifications, mainly deduced from the molecular masses. Did the authors perform other experiments to further support the presence of lipid modifications? 

2) I noticed that the NmMetQC20A protein was also purified with DDM detergent, could the mutant protein be purified without detergent? And could the WT protein be purified without detergent? This experiment could be an additional evidence to support the absence of lipid modifications on the mutant protein and presence of lipid modifications on the WT protein. 

3) It is very interesting to see that the lipid moiety of lipo-NmMetQ is required for maximal NmMetNI stimulation, especially compared to the secreted NmMetQ. This result suggests that the lipid moiety could participate in the NmMetNI stimulation directly. But the lipid moiety could not be resolved in the lipo-NmMetQ:NmMetNI complex structure, probably due to the limited resolution at 6.4 Å. This point is quite novel, unfortunately, this manuscript provided little insight on this. 

4) The inward-facing NmMetNI structure was resolved in the presence of lipo-NmMetQ and AMPPNP, but only the apo NmMetNI structure was captured. This is unexpected, and the authors should comment on this. 

5) The bioinformatic prediction of the distribution of lipid-modified MetQ proteins in different classes of Proteobacteria is very weak. As the authors mentioned in the Discussion part, future efforts should be made to experimentally determine which SBPs have lipid modifications. I would suggest that the authors should move the Fig. 5 to supplementary information and move the corresponding text to Discussion. 

Minor points: 

1) The authors should provide the details for cryo-EM sample preparations in the Methods and Materials, such as how the protein complex was assembled, the protein complex concentrations, AMP-PNP/ATP concentrations, methionine concentrations, et al. 

2) Please state the ATP concentration used in ATPase experiments in Methods and Materials.

Reviewer #3 (Public Review): 

Although the lipidated MetQ protein is isolated from E.coli and not the native Neisseria, this reviewer is satisfied that the MetQ proteins lipidation state is valid when produced in E.coli. 

The ATPase assay of MetNI coupled to lipidated MetQ (MET) binding is a nice way of testing binding and correlating this with the F-NMR FAXS data to examine substrate specificity and diversity is clever. 

The authors present 2 Cryo-EM structures of NmMetNI in the inward-facing and outward facing (with MetQ bound) forms. The inward facing adds to our understanding of how interactions between the NBDs of ABC transporters without a C2 autoinhibitory domain differ from their homologous structures with autoinhibitory domains. 

Enthusiasm for this study is slightly dampened by the lack of resolution of the ternary complex that doesn't give insights on how lipidated MetQ binds to MetNI whereas the nonlipidated form does not stimulate activity. This would have been a nice structural outcome. The lower resolution structure of the Lipidated MetQ bound to MetNI would have been more impactful if we could see how/why the lipid on MetQ is required for stimulation of ATPase activity or if it told us whether NmMETQ is located in the inner leaflet of the OM or outer leaflet of the IM - presumable it is the latter. But this highlights an important question: what regulates the transport of lipidated-MetQ to the OM -is it the presence of substrate Met and its interaction with the integral membrane protein MetNI that retains it? 

Given all this work, the title could be changed as it should reflect that the characterization of the ABC methionine transporter from Neisseria reveals that Lipidated MetQ is required for interaction. 

Overall this is an important contribution to our understanding of the function of lipidated MetQ as a SBP and component of an ABC importer in Neisseria. Especially as MetQ is being considered as a vaccine candidate to prevent Neisseria infections and groups are working towards this goal.

Reviewer #1 (Public Review): 

The manuscript by Takahashi et al describes the interaction between MLL fusion proteins with HBO1 and its role in leukemogenesis. Myeloid progenitor transformation assays using various MLL fusion proteins reveal that MLL fusion proteins requires the TRX2 domain of MLL for effective leukemic transformation. IP-MS identifies HBO1 as a bona fide binding partner of the MLL TRX2 domain. ChIP-seq experiments show genome-wide colocalization of HBO1 complex with MLL-ENL and the WT MLL in MLL-fusion leukemia cells and MLL WT cells, respectively. ChIP-qPCR in MLL-deficient cells suggest that recruitment of HBO1 to MLL target genes (such as MYC and CDKN2C) depends on MLL. Truncation analysis of the ELL part of the MLL-ELL fusion reveal that MLL-ELL transformation activity requires OHD domain-mediated recruitment of AF4 and EAF1. Furthermore, co-IP and ChIP experiments with various fragments show that AF4 and EAF1 form two distinct SL1/MED26-containing complexes and likely the AEP/SL1/MED26 complex is competent for transactivation. Series of transformation assays suggest that MLL-ELL transforms hematopoietic progenitors via association with AEP, but not other ELL-associated proteins. Finally, the authors also show that NUP98-HBO1 fusion transforms myeloid progenitors through interaction with MLL. Overall, this is a quite comprehensive study demonstrating that various MLL fusions and NUP98 fusions transform hematopoietic progenitors via HBO1-MLL interaction, which suggests that targeting their interaction might be s new therapeutic approach.

Reviewer #2 (Public Review): 

In this manuscript, the authors identified an interesting interaction of MLL (a methyltransferase) with an HBO1-JADE complex (an acetyltransferase) and investigated the functional impact in leukemogenesis by fusion proteins containing MLL or HBO1. The data is clear and the connection between MLL and HBO1 is unexpected. The manuscript is also well organized and relatively easy to follow. 

Comments: 

1) The functional relevance of the interaction between MLL and HBO1 is still correlative. It would be important to know whether there are any results directly about the impact of the loss of the HBO1 complex on the function of MLL. 

2) It is important to show the source and specificity of the antibodies that were used for ChIP of the HBO1 complex. 

3) It might be interesting to check whether other JADE proteins and also BRD1 (another partner of HBO1) are involved. 

4) The acronym TRX2 may be confusing as some might think that it is thioredoxin.

Reviewer #3 (Public Review): 

This paper starts with a series of bone marrow transformation assays comparing MLL fusions and domain-deletion mutants thereof to define the minimal elements for robust leukemic transformation and surveying growth and attendant common fusion targets HoxA9, Meis1 in colony replanting assays. Here they discover that a region of the MLL-N portion just upstream of the well-studied CXXC domain, termed in their previous work the "TRX2 domain" is important for the transformation capacity for several different MLL-fusions (and more minimal chimeras of key modules). A small region of the MLL-N protein encompassing the TRX2 domain and the CXXC module are subjected to complex purification, it is clear from comparison to number of controls that the TRX2 domain is an important mediator of association, perhaps indirect, with the HBO complex. Drop out experiments confirm that HBO1 knockout is lethal to MLL-rearranged leukemia, nicely confirming recent work (Ay et al., MacPherson et al.). 

ChIP-seq experiments in an ALL with MLL-ENL fusion, and then more extensively in a kidney cancer cell line indicate overlap with some of the HBO complex subunits and MLL, however this does not establish recruitment at these sites. ChIP-qPCR at a few MLL-fusion target genes with MLL depletion supports the recruitment hypothesis somewhat although mixed and modest effect sizes indicate that alternate pathways for HBO1 recruitment are involved, and could also be explained as reduced deposition of marks known to recruit HBO1, rather than direct recruitment. Sadly, the real potential strength of this work goes unrealized, as the recruitment of HBO1 mechanism remains tantalizingly out of reach. More experiments in this space could conclusively establish the molecular mechanism of a seemingly biomedically important recruitment paradigm, and thereby have much more impact. 

At this point the paper shifts to a seemingly distinct line of inquiry, which is not closely related to the HBO1-TRX2 story to the first three figures. The new direction examines the ELL fusion partner in some detail using similar fusion protein chimeras, but a portion of Figure 4, is largely confirmatory of previously established findings about the critical regions of ELL for transformation and its AF4/EAF1 partners, adding only that portions of the MLL fusion protein are dispensable, provided that they are replaced with the PWWP of LEDGF. It is a little bit of a Frankenstein's monster experiment, and does not add much new to the field. Further experiments define potentially two distinct complexes that have already been characterized being recruited by ELL, although there is overlap here again with their previous studies, and the results are a little hard to interpret. 

The authors create structure-guided separation of function mutants in the ELL domain that binds both AEP and SL1, permitting them to specifically disrupt EAF1 interactions but not AF4. Further experiments solidify this interpretation, and find that this mutant shows no deficits in hematopoietic progenitor transformation or primary leukemia lethality, although there appears to be some effect upon reimplantation. 

The last figure in the paper tackles the seemingly unrelated Nup98-HBO1 fusion, a rare patient mutation-they demonstrate a requirement for MLL for viability of hematopoietic progenitors transformed by this fusion, connecting back to the TRX2 interaction, and show that menin inhibitors slow growth. 

Strengths: 

The identification of the TRX2 region of the MLL-N protein as the major point of contact (perhaps not direct), to the HBO1 complex adds mechanistic depth to the really important recent discovery (confirmed in this work) the MLL-fusion leukemias rely on HBO1 function. This lab has published a number of technically similar types of papers defining minimal regions of MLL and distinct interacting partners by chimeric fusions, with bone marrow transformation assays, mouse model engrafting studies, IP's, ChIP etc. In my view they are very much under cited, likely because they are similarly so challenging to read. 

The mixture of Co-IP biochemistry, bone marrow transformation assays, and ChIP, to define interactions, minimal requirements for transformation, and their chromatin consequences for a host of different MLL-fusions and HBO1-fusions has the potential to define the key interfaces underlying recruitment. 

Weaknesses: 

The mechanistic inquiry stops short of really defining the critical MLL-HBO1 complex interface. Defining the point of contact on the HBO1 side (even which subunit) and determining whether it is direct, or bridged by some, as yet unidentified factor, as well as conclusively demonstrating that this is the mechanism of HBO1 recruitment remain the major shortcomings. 

And the follow-on figures apart from the last one, appear disconnected from this portion of the story and distract from it. 

The complex nomenclature and density/organization/logic of the presentation of experiments makes this paper difficult to read. Absence of sufficient grounding in the broader literature much beyond their own lab's work further compounds the problem. 

There is a lot of overlap, particularly in parts of figure 1 and figure 4 with previously published results. So perhaps re-organizing the display of data, and the organization of presentation, putting confirmatory work in the supplementary figures, would improve accessibility. 

Impact: 

In its present state, this is an incremental, but important advance for the field. With more mechanistic depth, particularly on the HBO1-MLL interface would substantially increase general interest.

Reviewer #1 (Public Review): 

This paper addresses the role of the DNA damage and DNA replication stress proteins Mrc1 and Rad53 kinase in DNA replication fork progression. The authors confirm previous research that Rad53 can block the initiation of DNA replication by phosphorylating either Dbf4 or Sld3. They also show that Mrc1 stimulates the movement of the CMG helicase on a n activated dsDNA substrate and that Rad53 phosphorylation of Mrc1 blocks this helicase stimulation and thus DNA replication fork progression. They also show that Rad53 can target Mcm10, but the consequences of this phosphorylation are only revealed in the absence of Mrc1. The authors confirm the effect of the Mrc1 phosphorylation mutations in vivo. 

This paper reports new functions for the Rad53 checkpoint kinase and for Mcm10 and Mrc1 at DNA replication forks. The results have important implications for understanding the response of the DNA replication machinery to DNA damage and the separation of this activity from global checkpoint signaling.

Reviewer #2 (Public Review): 

In this manuscript, McClure and Diffley take advantage of their powerful in vitro reconstitution system of budding yeast DNA replication to explore the role of Rad53, a key DNA damage signaling kinase, in the slowdown of DNA replication progression. They identify two potential Rad53 targets: Mrc1 and MCM10 and show that Mrc1 phosphorylation is both essential and sufficient for slowing down DNA replication by slowing down the unwinding rate of the replicative helicase. 

Overall, this manuscript makes some very useful contributions to our understanding of the control of DNA replication progression: it identifies Mrc1 as a key and direct Rad53 target in the replisome and it also provides the basis of Mrc1-dependent inhibition of replisome progression via slowing down the unwinding rate of the replicative helicase. While the results are broadly supportive of their conclusions, the work would be strengthened if the authors could provide additional evidence that the sites on Mrc1 they identify as being critical for slowing down replisome progression in vitro, are also phosphorylated by Rad53 in yeast cells.

Reviewer #3 (Public Review): 

Background and Significance:

DNA replication of in the presence of DNA damage or nucleotide depletion results in activation of the DNA checkpoint signaling pathways that trigger numerous responses to minimize genome damage. The signaling pathway results in the Mec1/ATR kinase-dependent phospho-activation of the Rad53 effector kinase that in turn targets multiple replication and DNA repair factors to maintain the integrity of the replication process. Previous in vivo studies have determined that Rad53 responds to replication stress by inhibiting the activation of additional, unfired replication origins, and by stabilizing replication forks to prevent their collapse into dsDNA breaks or other destabilizing structures. Previous studies have also indicated that fork stabilization rather than origin inhibition is the critical essential function of Rad53 in maintaining genome stability and cellular viability. However, exactly how Rad53 activity stabilizes forks remains incompletely understood. Some in vivo studies have suggested that Rad53 slows the progression of the replicating fork(s), perhaps to provide time for repair of template damage or restoration of dNTP levels before attempted replication. However, the effect of Rad53 on fork rate has been somewhat controversial, and critical targets of Rad53 at the fork required for fork stabilization and fork rate control have remained obscure despite much interest in identifying these factors, which are considered to be central to genome stability. 

Detailed Review:

McClure and Diffley's current work reports on the function of Rad53 in regulating origin firing and particularly the elongation phase of replication. The authors use an in vitro replication system using step-wise addition of purified proteins to examine the effects of Rad53 activity in the absence of DNA damage or dNTP depletion, thereby eliminating these as confounding factors. 

The authors begin by loading MCMs onto a plasmid template, followed by DDK activity, which has been pre-incubated with active Rad53 kinase or mutant Rad53 as control. Consistent with expectations from previous in vivo studies, phosphorylation of DDK inhibits replication initiation in this system. Similar treatment of Sld3 by Rad53 similarly inhibits replication initiation as expected from previous in vivo work. The experiments indicate that Rad53-dependent phosphorylation of either Dbf4 or Sld3 is sufficient to inhibit replication initiation. 

Next, the authors examine the effect of exposing elongation factors to Rad53 following the execution of the initiation steps. This results in a reduction of elongation rate from ~700bp/min to ~400bp/min. The authors narrow the focus to Mrc1, Tof1, and Csm3, which together comprise a so-called replication fork protection complex, by specifically pre-incubating these proteins with Rad53. They find that pre-incubation of Mrc1 alone is sufficient to slow elongation whereas pre-treatment of Csm3 and Tof1 had little or no effect. They further showed that Rad53-mediated inhibition of elongation requires only Mrc1 phosphorylation, as add-back of Rad53-untreated Mrc1 to replication reaction with other elongation factors pre-treated with Rad53 restored normal elongation rate. 

In possible conflict with these results, the Remus lab has recently reported that Rad53-mediated phosphorylation of Mcm10 slows elongation in reconstituted replication reactions lacking Mrc1, Csm3 and Tof1 (M/C/T). The authors here confirm this result but show that the effect is only observed in the absence of M/C/T, and add back of untreated M/C/T rescues elongation rate to wild-type level. These results suggest that Mcm10 phosphorylation may regulate fork rate in concert with phosphorylated Mrc1, but not unphosphorylated Mrc1. 

The authors further investigated the mechanism of Mrc1 fork slowing in the presence of different replicative DNA polymerases, and find that slowing is independent of the DNA polymerase involved. To determine whether Mrc1 affects DNA unwinding, they developed an assay for unwinding based on DNA cleavage by a restriction endonuclease, which can only cut dsDNA. Thus, DNA unwinding can be measured by loss of cleavage at increasingly distal sites from the replication origin. In this assay, presence of Mrc1 stimulates the unwinding rate approximately two-fold, whereas phosphorylated Mrc1 does not, suggesting that Rad53 phosphorylation of Mrc1 inhibits its stimulatory effect of the helicase. It's notable that the rate of unwinding in this assay is roughly five-fold slower than progression of the synthesizing replication fork in this system. This, of course, suggests that additional elongation factor(s) contribute to the full rate of replication in this system. 

To dig deeper into mechanism of Mrc1 inhibition, the authors examine mutant alleles lacking numerous potential phosphorylation sites, starting with the 17AQ mutant, which lacks Mec1-target sites, and is defective in signal transduction to Rad53. Interestingly, this mutant has no effect on Rad53 inhibition. A C-terminal truncation of Mrc1 eliminated inhibition by Rad53; however, this allele also fails to stimulate replication like wild-type Mrc1, complicating interpretation. The authors take multiple approaches to map Rad53 phosphotargets in Mrc1, including incubation of Rad53 with Mrc1 protein fragments, with Mrc1 peptide arrays, and MS. Most of the Rad53-dependent sites are located to the C-terminus and the authors create multisite mutants (to alanine) to determine functional effects of eliminating phosphorylation. Although they failed to identify a perfect separation of function mutant that retained full stimulation of replication while eliminating inhibition by Rad53, the 14A and 19A mutants exhibit intermediate phenotypes for both stimulation and inhibition, consistent with these residues serving a regulatory function. 

They alternatively mutated a subset of the 14A residues to phosphomimics, creating an 8D allele. With or without phosphorylation by Rad53, the Mrc1-8D supports replication equivalently to phosphorylated wild-type Rad53 as predicted for this phosphomimicking allele. One concern, of course, is whether the residue changes have inactivated normal Mrc1 stimulation of replication independently of effective phosphomimics. The finding that 14A affects Mrc1 function in the absence of Rad53 contributes to this concern as mutation to A is expected to results in less potential pleiotropic effects than changes to D. 

The authors go on to test the effects of the 8D allele in vivo on response to DNA damaging (MMS) and nucleotide depleting (HU) agents. Intriguingly, they find that combination of the Mrc1-8D allele with rad53∆, partially suppresses the extreme drug sensitivity of rad53∆. This is an interesting finding that supports the idea that phosphorylation of Mrc1 by Rad53 contributes to survival in the presence of these damaging agents. Arguably, this may also support the idea that the 8D allele is not simply broken (comparison to mrc1∆ might help make that point though rad53∆ mrc1∆ may be lethal, which also supports this conclusion). Indeed, the 8D allele supports Rad53 activation in vivo, indicating that it retains some function. Overall, this is a compelling result despite the modest effect and lingering caveats about this allele. 

Strengths: 

- Answers, at least partially, the long-standing question in the field: How does Mrc1 control replication forks normally and under replication stress? This has been challenging to address by in vivo experiments for multiple reasons including multiple pathways, targets, and use of drugs to elicit Rad53 activation, which has independent/additional effects on forks. 

- Given the complex role of Mrc1 in coordinating multiple activities at the replication fork, the highly defined biochemical system with purified proteins is appropriate to the task. Experiments are well-controlled and replication is robust. Many complicating effects are eliminated. Most results are clear-cut and convincing. 

- Identification of Rad53 phosphosites in Mrc1 and likely role of these sites in regulating the rate of fork progression is a substantial accomplishment and these results provide a defined mechanism to incorporate into existing models of Rad53 and replication control. 

Weaknesses: 

- The in vivo suppression phenotype is relatively minor and suggests that other factors or pathways play a significant role in vivo. For example, the Diffley lab has previously shown that deletion of EXO1 almost completely suppresses the drug sensitivity of rad53∆ cells. This suggests that the Rad53-Mrc1 axis of regulation described here remains insufficient to prevent the accumulation of Exo1-sensitive lesions or conditions. 

- Despite the development of a quantitative method for DNA unwinding and the high quality of the data, there is no quantitative analysis of the data by statistical method. At least there needs to be clear evidence of reproducibility.

Reviewer #1 (Public Review): 

Heinze and colleagues present InsectBrainDatabase (IBdb), a resource that collects, displays and shares mostly neuroanatomical data from several insect species. Users are able to search and visualise neuronal morphologies in several ways, brain regions and species' information. On the whole, the site is very well built, with a clear intent on providing a good user experience, both for experienced insect researchers and naïve users. The authors have designed the site in a way that it could be used as a general data hub, pre-publication, with a clear versioning system, and this will be appealing for some researchers. The manuscript describes the site accurately, focusing on how a user would interact with it. 

Although the authors intend for IBdb to become a major resource for their community, it is not made clear how they will increase data deposition beyond simplifying this process. With a distributed curation approach, and the expectation that researchers will be submitting their own data, it is also not clear how they will ensure consistency, completeness and accuracy of data curation, an essential aspect to broaden the usage of their platform, guarantee transparency to users, while, at the same time, enriching the amount of searchable data. 

1) The authors use the term 'cell type' often in the manuscript, and a few times in the user guide. The concept of 'cell type' is an essential one for neuroscience researchers. However, I was unable to find any reference to it on the site, particularly in the information pages for neurons. This led to ambiguity about what data was displayed, what constitutes a cell type and how I would search for it. For example, a pair of neurons that look of the same type are am-AMMC-1 (NIN-0000159) and am-AMMC-1' (NIN-0000171), although they have distinct, yet related, full names. A similar case for am-AMMC-2' (NIN-0000172) and am-AMMC-2 (NIN-0000160) although it looks like this could be in fact the same cell - the neuron image caption of the former notes that it is a mirror image of the actual cell though this information is not given elsewhere. Lastly, the authors mention that they do not require neuron morphologies to be registered to a particular template. This point only makes it more essential for neuron types to be transparently labelled. 

2) I found the curation of data was often incomplete or inconsistent. This might be a consequence of the distributed strategy for curation together with the significant direct input from users. A few illustrative examples: <br> a. Completeness: the curation of neuroanatomical information for neurons is often missing, although there is enough information to do this for 'soma location', 'fiber bundle', 'morphology description'. <br> b. Consistency: (1) I found 3 different forms to designate AMMC, used in 'keyword'. Furthermore, I found functional terms used for 'keyword' ('mechanosensory'), not captured by a 'Modality' descriptor. (2) There is significant inconsistency also in the species's pages. Some have descriptions taken from the web, others cite academic literature while others are missing this section entirely. Often, no linked publications are available. 

3) The authors present the platform as a data hub for not only neuroanatomical but also functional data. There is of course potential, but currently there is very little functional data on the site. Thus I find the authors' claim on the abstract "by intimately linking data on structure and function" unproven at this point.

Reviewer #2 (Public Review): 

Heinze et al. created a public online database for morphology and function of neurons in many insect species (IBdb). This database is a platform for shared and searchable data repository. It can visualize the morphology of multiple identified neurons in the context of neuropils with options to control various visualization features (i.e. color, transparency, etc.). The uniqueness of IBdb is to index brains and neuron data of many insect species, allowing comparative approaches. The structure and functions of the IBdb are uncomplicated and intuitive. 

All these features were described clearly, and the paper can also serve as an instruction of the database. My comments and concerns are therefore mainly about usability and future development. 

1) This comparative database does not include some species, most critically Drosophila melanogaster. This exclusion is a drawback, as searching homologous neurons of the Drosophila neurons in other insects, or vice versa, would be inspiring and promote further comparative approach. As the same neuropil nomenclature was used in the largest and probably most elaborate database with similar functions for the Drosophila brain, VirtualFlyBrain, and IBdb, it would be helpful to implement cross-species neuron search based on arbor areas (as mentioned in Line 508). 

2) More comprehensive 'preset' depository of published data would make this database more attractive, as users naturally tend to first go to the largest and most comprehensive one. VFB also made a big success in this respect by actively indexing massive data taken in different labs. 

3) There is a concern on sustainability, as administration/management (e.g. species, curation, approval) continuously need expertise. It would be powerful to come up with a mechanism to encourage participation of more active users. 

4) This database doesn't seem to require registration of neurons to a standard brain of the species (Line 502). It is unclear how one can make visualization as in Fig. 4 without registration. It would be helpful to detail what one can/cannot do depending on the data type.

Reviewer #1 (Public Review): 

Twitchin kinase is a sarcomeric protein found in the muscles of the nematode, C. elegans, whose activity is thought to be mechanically regulated by stretch of the muscle. The authors construct transgenic animals containing a twitchin variant with built in FRET sensors that should report the structural state and, by inference, the activity state of the protein in the muscle as the worm moves in a dish. A control variant is constructed containing the same two FRET pairs, but located in a position of the molecule that should not be sensitive to stretch. They show periodic changes in the FRET signal that correlates with the state of muscle contraction/relaxation. 

A major strength of the study is the technical novelty of combining brightfield imaging of the undulating worms with FRET measurements in real time. Care has been taken to essentially align many events as the worm moves, correcting for motion in order to align the two imaging modalities and to be able to present the data in the form of kymographs. The construction of the stretch-sensitive and control twitchin constructs are based on crystal structures and are logical. The manuscript is well written and possible artifacts are discussed. The results are convincing, but not overstated. The problem is that the study is dealing with a complex living system and signal noise arises from several sources. While a good control was designed that, ideally, should not show any change in FRET signal, there are changes that are in the same "direction" as those seen with the experimental construct. The changes seen with the control however are smaller. Another concern, which is well discussed, is that the constructs for most of the experiments were overexpressed in the worms in order to gain a larger signal and it is not clear now the "extra" molecules are binding within the sarcomere and whether they would be subjected to the same stresses as the fraction that is bound in the spot usually occupied by native twitchin. This is addressed using CRISPR/Cas9 to create a knock-in worm containing the full length FRET-twitchin construct. This worm shows the expected changes, but with a lower signal to noise ratio due to the weaker signal.

Reviewer #2 (Public Review): 

Porto et al. developed a method to image the force-induced unfolding of twitchin kinase (TwcK) by FRET at the whole organism (C. elegans) level and correlated to its motion. The twitchin kinase contains autoinhibitory domains (NL+CRD) that can be mechanically unwrapped from the kinase, leading to its activation. The authors designed experiments to see whether this mechanically-coupled activation happens or not during the muscle contraction of live C. elegans. To do so, they engineered FRET donors and acceptors (mCFP and mCit) to flanking positions around TwcK. They did so without affecting the auto-inhibitory functions of TwcK. As a control, they flanked a nearby Ig domain with the same FRET pairs. The hypothesis is that under force, the unravelling and stretching of the NL domain from TwcK will lead to FRET decrease. While in the control case, Ig domains can resist mechanical unfolding, and FRET would not change. The authors were able to image the whole organism with NIR-brightfield, mCFP and mCit. Channels simultaneously with reasonable spatial and temporal resolution. Image processing algorithms were developed to correlate the C. elegans motion (curvature along the body of C.elegans) to the fluorescence at corresponding body locations. They found propagating "waves" of curvature from the head to tail of the worm in kymographs of body curvature (as expected) and fluorescence in mCFP, mCit, and FRET channels, which is somewhat unexpected. They found the magnitude of FRET change for the mCit-TwcK-mCFP to be statistically greater than the control mCFP-Ig-mCit, although both exhibited a significant correlation with the worm's body curvature. During the contraction and relaxation cycles of the muscles, the FRET change in the mCit-TwcK-mCFP construct is also greater than that of the control group. The authors contributed the observed FRET changes in mCFP-Ig-mCit to various factors, including imaging and motion artifacts. And therefore, the difference observed between the experiment and control is due to the force-induced unfolding of the NL domain from TwcK. The authors suggested that the force-induced unfolding of autoinhibitory peptide activates the kinase, which may have physiological relevance to be discovered in future studies. 

Strengths: 

1) Traditionally, in vitro characterizations of molecules were performed to infer their biological function, which requires knowledge of their complex physiological environment that may not be available. Direct, in vivo, experimental measurement of mechanically-coupled activation of signalling proteins in this study is crucial to understand their physiologically-relevant function and allows correlation to the function at the organism level. This is a beautiful demonstration relating molecular mechanical events in the context of the system it works in. I also recognize and respect the challenges undertaken by the authors in this research. 

2) The overall idea and design of the experiment are elegant. The authors created a transgenic worm that allowed the FRET to be monitored in vivo over the entire body of the worm while simultaneously tracking its motion. The new methodology developed in this study can be potentially adopted to study other mechanically-activated systems in vivo. 

Some aspects of the analysis and interpretation of the fluorescence result require clarification: 

1) The propagating patterns in the kymographs of mCFP and mCit (Fig 5BCFG) are puzzling. The authors contributed it to inherent locomotion artifacts, noise and internal sarcomere rearrangement during motion. While some of these may be true, could these be image processing artifacts? The authors stated in the method section that the fluorescence intensity at a particular body segment is obtained by drawing a perpendicular line to the midline. The pixels that it intersects will provide the fluorescence intensity. This approach does not seem to account for the fluorophore density change due to tissue compaction and expansion, resulting in overcounting intensity in the inner circle and undercounting at the outer circle - similar to the observed intensity patterns Fig 5BCFG. Perhaps it would have been helpful if the intensities were normalized by the arclength at the different radii from the center of the curvature. 

2) As related to the previous comment, but more generally, image analysis is a critical and sensitive step towards the interpretation of the fluorescence results. The authors would need to elaborate if and how errors in the image processing might contribute to the emergence of correlation between FRET and curvature. For instance, the CFP and mCit expression levels vary significantly along the body of the worm (Fig 5) and should be time-invariant. If an error in image processing picks up nearby spatial variations as the worm moves, the detected fluorescence will become time-variant and correlate with the worm's motion. It is unclear whether this could this happen with the current algorithm. This is a crucial assessment as it is crucial to ensure the observed small FRET changes (+/- 0.015) are due to molecular stretching and not artifacts of image processing. 

3) The shape and meaning of FRET change in the contraction-relaxation cycles (Fig 7) would require further interpretation. The data shows that the extrema and phase of the FRET signal correlate to curvature, and thereby, sarcomere stretching. It is unclear whether it is valid to assume the stretching or relaxing of sarcomeres apply tension directly over each twitchin. Is the binding-unbinding transition of NL to TwcK two-state? If so, would this lead to two-state behaviour in the observed FRET? 

4) The reason behind the small observed FRET change (+/-0.015) requires further clarification. Is it because (1) all FRET sensors changed slightly, or (2) a small fraction of FRET sensors changed from high to low FRET. 

5) The manuscript provides strong evidence of FRET correlating to curvature during the muscle contraction cycle. However, the causality is less clear. It is unclear whether the contraction force causes the FRET change, or can curvature without any active contraction cause FRET change. For instance, it is unclear whether, if the worm were dead or myosin activity inhibited, the bending of the worm would cause FRET change.

Reviewer #3 (Public Review): 

Porto and colleagues explore the possibility that cytoskeletal protein kinases like twitchin and its vertebrate analogues may play a role in mechanosensing. The cytoskeletal integration of proteins like twitchin, titin or obscurin/unc-89, their established roles in maintaining elastic connections between myofilaments and evidence from single-molecule studies on force-induced conformational changes all suggest that such a role is plausible. However, direct proof in functional sarcomeres that physiological mechanical forces can lead to functionally significant conformational changes in cytoskeletal kinases have so far been missing. 

The authors fill this important gap by using a cell biophysical approach in the transparent nematode C. elegans, employing transgenic and genome-edited animals expressing FRET mechanosensor constructs to explore whether mechanical stress associated with nematode locomotion could induce conformational changes in the twitchin kinase region. The development for this aim of an imaging and image analysis platform that combines information on locomotion with fluorescence output is an important advance in itself. 

This is a technically cutting-edge, interdisciplinary work, employing novel imaging approaches that will pave the way for many other studies on different proteins in the nematode and possible other model organisms. The findings are highly intriguing and could represent ground-breaking work into cytoskeletal signalling mechanisms. 

The data suggest that the twitchin kinase FRET mechanosensor responds to changes in body curvature occurring during locomotion and therefore reports on conformational changes in the kinase region that would be compatible with a mechanically gated activation mechanism. <br> The possibility of intra- versus inter-molecular FRET is discussed and addressed but may require further validation. 

The paper is well illustrated and very clearly written and explained.

Reviewer #1 (Public Review):

This is a carefully performed but largely descriptive analysis of 27 cases of T-cell lymphoma, 25 of whom had AITL. The authors attempted to connect mutations identified in the lymphoma to mutations arising in clonal hematopoiesis (CH) identified in paired bone marrow samples. They also identified differences in mutation patterns between early mutations (primarily C>T, an aging signature) and late mutations (primarily C>A), which they argue is evidence of a smoking signature. They do identify an enrichment of lung cancer among the patients in their cohort compared to rates for other cancers, such as CLL. Finally, they determine that TET2 mutations confer a poorer prognosis among their AITL patients, and show that those with 2 or more TET2 mutations in their CH clones have an increased risk of concurrent hematologic neoplasm (HN).

Strengths:

- the analysis of mutation signatures is novel, and the association with smoking, if confirmed, would be an interesting insight.

Weaknesses:

- many of the findings reported here have been reported elsewhere, including association between AITL and concurrent myeloid neoplasms earlier this year in Blood Advances (Lewis et al, PMID 32442302), the frequency of mutations in DNMT3A/TET2 in AITL (Couronnee et al, NEJM 2012, Odejide et al, Blood 2014), and the prognostic impact of TET2 mutations in T cell lymphoma (Lemonnier, Blood 2012).

- To define the likelihood of developing a concurrent hematologic neoplasm among those with multiple TET2 mutations, the authors incorporated the findings from the Lewis paper into their analysis. It appears that they took this straight from the paper, rather than obtaining the primary data (which would have been more robust).

- the C>A mutation signature can also been an oxidative artifact of sequencing library preparation. It is not clear from the methods or data supplement that the authors considered this possibility.

- the authors do not appear to have reported smoking history for these patients. If available, pairing smoking data with presence or absence of the C>A mutational signature would be helpful.

- care must be taken to distinguish these patients, who already have one hematologic neoplasm, from patients with CH who do not have HN. Although two TET2 mutations in this cohort increases the risk of another concurrent HN, one cannot extrapolate the same risk to a non-HN CH population.

Reviewer #2 (Public Review):

In this study, Cheng and colleagues use a comparison of mutational landscape in matched AITL tumor samples and Bone marrow sample/peripheral blood to describe the link between clonal hematopoiesis (CH) originating from precursor marrow cells (HSC) and AITL as well as other hematologic malignancies. Using sequential or synchronous samples, they show that myeloid or B-cell neoplasms and AITL can arise from CH. They report enrichment in APOBEC/AID activity-associated substitutions and tobacco smoking associated-substitutions in the variants observed in AITL, the later being associated with a higher risk of lung cancers in their cohort of AITL patients with comparison of epidemiological data. They finally found that the presence of 2 or ore TET2 mutations with >15% VAFs associate with a higher risk of second hematologic malignancy.

This study is well designed and overall original, based for the first time on a systematic comparison of matched bone marrow/peripheral blood and primary AITL tumors in a series of 25 AITL (angioimmunoblastic T-cell lymphoma) patients, using a large targeted gene panel. Through this original design, the results presented further expand recent reports of the link between clonal hematopoiesis (CH) and AITL, one the most frequent T-cell lymphoma worldwide and also brings important novel findings with potential clinical and biological relevance with respect to the association with APOBEC/AID signature and significance of multiple TET2 mutations with a high allele burden. Overall, the conclusions are fully justified by the obtained data. The main weakness of the study relies on the fact that the association with APOBEC/AID signature, and also with lung cancers is found in the context of a rather limited number of AITL patients (n=25) and, though statistically significant, would ideally require validation in independent publicly available series.

Reviewer #3 (Public Review):

This study examined patterns of CH-related mutations in the tumors and matched bone marrow samples from patients with AITL/TFH-PTCL. AITL is an uncommon T-cell lymphoma with a high prevalence of epigenetic mutations (TET2, DNMT3A, IDH2) also typical of age-related CH, and it has association with other hematologic malignancies. The authors suggest that AITL may be derived from a CH stem cell progenitor, which could explain some of the prior associations. This hypothesis has previously emerged because co-occurring mutations were found in B-cells from patients with AITL, and a population-based study discovered a particularly high incidence of T-cell lymphoma among patients with MPNs. The paper confirms the prior observation from a paper by Lewis et al., Blood Adv, 2020, which showed co-occurrence of TET2 and DNMT3A mutations in 22 paired samples of AITL(or TFH-PTCL)/marrow. The present paper additionally provides information about the differential types of single-nucleotide variants among "CH-derived" and "lymphoma specific/late-acquired" mutations, and suggests possible association with tobacco use as a modifying factor (by showing that C>A or "COSMIC Signature 4" mutations are prevalent in AITL-specific mutations but not CH mutations). The authors also report that multiple (>1) TET2 mutations at high VAF (>15%) in CH are associated with subsequent hematologic neoplasms.<br> The strength of the paper is reliance on diagnostic samples (as opposed to post-treatment or post-transplantation samples which are known to harbor higher rates of clonal hematopoiesis), and more in-depth evaluation of a few patients who developed other hematologic malignancies before or after AITL.

The major weakness of the paper is the speculative nature of assignment which mutations may be in which tissue (lymphoma, "HSC", other concurrent myeloid malignancy) based solely VAF %. In only one patient authors substantiate this weak assumption by analysis of purified blood granulocytes. Therefore, the discussion should be prefaced by disclosing this limitation and the "evidence" from data should be really framed as hypotheses which need extensive validation. In particular, it is not clear if CH mutations were in lymphoma (AITL) cells. Some other sources of bias (e.g. immortal-time bias when studying "secondary hematologic cancer-free survival") should also be discussed, and some methodologic aspects need clarification.

However, these data add to accumulating circumstantial evidence that supports the model of CH as a background for development of AITL and associated hematologic malignancies through sequential acquisition of additional genomic hits that may predispose to T-cell (RHOA), B-cell (EZH2), or myeloid (ASXL1, SRSF3) malignancies.

Reviewer #1 (Public Review):

Activated neutrophils are known to contribute to inflammation and oxidative stress by generating reactive oxygen species (ROS) via NADPH oxidase and superoxide dismutase (SOD) during infections. The ROS can not only damage pathogens but also host blood plasma proteins and lipids. Severe COVID-19 is characterised by a high neutrophil to lymphocyte ratio and an underlying state of oxidative stress in blood has also been hypothesized. In the study presented here for the first time provide evidence that hydrogen peroxide generated by the neutrophil specific enzyme myeloperoxidase is not only accumulated in plasma but also perpetrates structural damage to the strong and weak lipid binding sites on albumin, a key antioxidant in blood plasma, using SLFAs and EPR spectroscopy. This study is timely and relevant in deciding treatment of severe COVID-19 patients with human serum albumin and possibly supplementation of FDA approved antioxidants like glutathione.

Reviewer #2 (Public Review): 

In this manuscript the authors performed a prospective study in 25 COVID-19 hospitalized patients and try to associate oxidative stress and hypoalbuminemia with COVID-19 mortality. Although it is an interesting study, the small sample size does not allow safe conclusions. I doubt which may be the clinical impact of this study on developing a better treatment of patients with COVID-19.

Reviewer #1 (Public Review):

This work by Avetisyan et al. is focused on uncovering the developmental processes underlying cell type differentiation events during later stages of larval proprioceptive chordotonal organs (ChO) in Drosophila. Making use of an enhancer recently identified by this research group, and building on a previous RNAi screen, the authors find that two transcription factors known to regulate sensory organ development in other contexts - D-Pax2 and Prospero (pros) - function antagonistically to regulate the expression the expression of another transcription factor, Delilah (dei), in two "cousin" cell types in the LCh5 cluster of ChOs. Studies include standard loss- and gain-of-function experiments, gel shift assays, and in vivo reporter analysis to show that Pax2 promotes dei expression in PIIa-derived cap cells, whereas pros suppresses this activation in the PIIb-derived scolopale cell through experimentally-defined binding sites. CRISPR-mediated deletion of the enhancer confirms that the enhancer is sufficient to drive expression of dei in ChOs, with its deletion leading to defects in larval locomotion.

Overall, the study is carefully performed and the data generally support the model that Pax2 and Pros function antagonistically to control dei expression during ChO development. Quantifications of the expression levels of site-directed mutants, and of the larval locomotion were also valuable for assessing the specific contributions of particularly cis-regulatory sequences. Appropriate discussion of previous findings in the fly eye and mechanosensory system demonstrating a similar genetic relationship between Pax2 and Pros is included. The authors also recognize that the precise mechanism by which this antagonism is achieved remains unanswered, despite their identification of overlapping binding sites within the minimal dei ChO enhancer. Together, this study helps to clarify downstream roles for Pros and Pax2 in chordotonal organ formation, and further establishes that the antagonistic relationship between these factors is broadly required during sensory system development in Drosophila.

Reviewer #2 (Public Review):

In this manuscript, the authors have performed a careful analysis of an enhancer that they have identified within the gene, Dei, in the context of sense organ development in larval Drosophila. The study is interesting and important because the authors demonstrate that a 262bp enhancer of Dei contains binding sites for two transcription factors, sv and pros. They show that sv binds this enhancer to activate Dei expression in some cells of the sense organ, and pros binds it to repress Dei expression in other cells. Thus, this 262 bp region integrates both activation and repression of Dei, ensuring the appropriate expression of this gene and therefore appropriate cell fate decisions within the sense organ lineage. Most interestingly, the authors demonstrate this via a combination of methods, including generating mutants of the enhancer that phenocopies the gene mutant in cell-fate specification as well as behaviour.

Reviewer #3 (Public Review):

In this paper, Avetisyan et al use Drosophila genetics, transgenic reporter assays, in vitro DNA binding assays, genome editing, and larval movement assays to study how the Pros and D-Pax2 transcription factors differentially impact the expression of delilah (dei) and ultimately, the function of the Drosophila abdominal lateral chordotonal organs (lch5). A prior RNAi screen revealed that both Pros and sv (D-Pax2) were required for proper dei expression in the lch5 organ. Building off previous publications, the authors focus on a specific enhancer within an intron of dei that is sufficient to mediate expression in cap and attachment cells (but not scolopale cells) in the lch5 organ. The authors recapitulated the RNAi findings using loss-of-function alleles for sv and Pros, and the authors used gain-of-function assays to further show that these two factors impact dei expression in lch5 organs. A yeast 1-hybrid assay was used to identify D-Pax2 as a direct binding factor to the dei enhancer and EMSAs were used to map D-Pax2 and Pros binding sites, which were found to regulate dei enhancer activity in transgenic reporter assays. Lastly, the authors used genome editing to specifically remove this dei enhancer and found that its loss resulted in abnormal larvae movement.

Overall, the experiments are largely supportive of the main authors conclusions that Pros and D-Pax2 directly regulate dei expression in distinct cell types of the lch5 organ. However, I do have several experimental concerns, which are raised below, that would need to be addressed to further strengthen this conclusion. Moreover, while the finding that D-Pax2 activates and Pros represses dei presents a compelling gene regulatory module, the overall advance is rather modest as these results focus only on a single enhancer in a single gene and do not significantly describe how the differential regulation of dei leads to distinct gene regulatory networks that define scolopale vs attachment/ligament cell fate. In addition, the authors only explore the D-Pax2/Pros/dei relationship within one type of proprioceptor (lch5) that is specified in the Drosophila embryo and functions within the larva, and not within other proprioceptor sensory organs of the embryo (dch3/vchA/B, where I assume it is similar) and not in sensory organs that are also specified by atonal later in the larva (i.e. the leg proprioceptors, Johnston's organs, etc) or in the external sensory organs of the wing (where dei is also expressed). Without such findings, it is unclear if this antagonistic relationship between Prox/dPax2 is broadly used to specify different fates in many different sensory organs.

Reviewer #1 (Public Review): 

This manuscript describes spatial heterogeneity in transcriptional profiles in the syncytial amoeboid protist Physarum using single-cell scRNAseq. Physarum amoebae exists in single nucleated (haploid) and syncytial (multinucleated) life cycle stages, and the syncytial which can grow very large (centimeters) with both networked and front or fan morphologies. This begs the question of how single cells - often larger than some animals - can maintain and differentiate their varied functions and communicate with the environment. Therefore, the overall scientific question and rationale for the work is to better understand spatial gene regulation in large differentiated syncytial organisms that could have implications for multicellular tissue evolution and differentiation. The primary claims of the manuscript include that there exists spatially differentiated transcription in the nuclei of the Physarum syncytia and the authors thus conclude that nuclei are "mobile" processors facilitating specialized functions - essentially that the nuclei integrate spatial cues that result in the syncytium to locally change morphology and behavior. This primarily mechanistic conclusion - while provocative - is not yet adequately supported by the primarily descriptive or correlative experimental data presented in the manuscript.

Reviewer #2 (Public Review): 

Review of "Nuclei are mobile processors enabling specialization in a gigantic single- celled syncytium". Physarum polycephalum is a member of Myxogastria, within the Amoebozoa, which is a sister clade to Opisthokonta, which includes animals and fungi. P. polycephalum grows as a multinucleate syncytial plasmodium, while other members of Amoebozoa grow as single nucleate cells (Dictyostelium). In this study, the authors use a clever method to isolate sections of a diploid plasmodium of P. polycephalum by growing a single plasmodium over a 384 well grid and isolating RNA from each colonized grids by centrifuging the 384 well plate. A total of 4 genetically identical plasmodia were assayed, two of which were grown in a uniform environment and two of which were exposed to oat flakes. The authors show variability in gene expression patterns across a plasmodium, with clustering of gene expression patterns in certain areas of the plasmodium, for example the growth front (fan) versus the vein network. Additionally, each of the four plasmodia also showed a unique expression patterns. Single nuclear RNAseq was also performed to assess expression patterns differences between nuclei within the syncytium and again, clustering of expression profiles was observed. Expression patterns in syncytial plasmodia was also compared to expression profiles obtained from single celled haploid amoeba. As between plasmodia, expression patterns differences were observed between individual amoebae, particularly in cell cycle functions, in addition to the expression of amoebae specific genes. This study is quite elegant and the data supports heterogeneity of gene expression patterns and thus specialization within a syncytial multinucleate plasmodium. 

Some information regarding the genome and predicted genes in the P. polycephalum is needed in the introduction, strengths and deficiencies as a model to understand syncytial function. This would be useful for those readers not familiar with this organism. The expression of how many of the predicted genes in the genome (~48,000) were detected in the clusters? 

The functions of different parts of the plasmodium were often extrapolated based on expression of marker genes whose function has been presumably characterized in other organisms. How these marker genes were identified and justification could be made clearer. 

Functional category analysis of gene clusters would be helpful to know if particular functions are enriched. 

In Videos 1 and 2, the nuclei show very different sizes and morphologies. It would be useful to have additional discussion. 

Within the plasmodium, using single nucleus RNAseq, did the number of reads per nucleus vary in the plasmodium? Could some nuclei be quiescent and transcriptionally inactive (for example, those that are moving) versus others that are anchored, as in Video 1? 

For the supplemental datasets, it would be useful to have detailed information about the expression levels of genes in each cluster, along with the UniProt prediction and gene ID in Physarum. 

The discussion needs more evaluation of methods and results of data presented, for example, what does it mean that each plasmodium (in an identical environmental condition) shows differences in expression patterns, including different areas of the plasmodium?

Reviewer #3 (Public Review): 

Typically, multicellular organisms exhibit cell type specialization, where cells perform distinct roles and have distinct transcriptional states that generate their diversity. Physarum polycephalum is a large, syncytial plasmodium that, despite being only a single cell, still exhibits specialization over different regions or subdomains. Each large cell contains thousands of nuclei. This work by Tobias Gerber, Cristina Loureiro, Nico Schramma, et al. endeavored to find whether that subdomain specialization is supported by spatially localized transcriptional states, and whether heterogeneity in expression between nuclei accounts for those spatially localized transcriptional domains. To do this, they quantified nuclear distribution and mobility using DNA dyes and live imaging and confirmed that the individual networks contain thousands of nuclei, some of which are mobile and others are stationary, as had been anecdotally observed. They spatially profiled gene expression within individual Physarum plasmodia by dividing them in a grid pattern into 384 samples and using scRNAseq library preparation techniques. They found that overarching transcriptional differences in cultures that were dividing, chemotaxing towards a food source, or actively in contact with that food source. By comparing expression within cultures, across space, they found that Physarum progress through the cell cycle and express different cytoskeletal regulators in the network and fan portions, and when in contact with food, transcribe genes involved in nutrient uptake/breakdown specifically in regions in contact with the food source. They then used single-nuclei RNAseq to determine the transcriptional state of individual nuclei within these plasmodia. They found that nuclei were in different cell cycle states, and that some nuclei seemed to be specialized to produce genes involved in fruiting body formation. Finally, to determine whether the observed nuclear heterogeneity was related to earlier expression differences (i.e. developmentally maintained), the authors profiled single-nucleated Physarum amoebae. They found that amoeba have an expression program that is distinct from syncytial Physarum and exhibit transcriptional hetereogeneity that is related to their phase in the cell cycle. 

Overall, the quality of the RNAseq data seems sound, and the conclusions presented seem mostly supported by the data. Additionally, the manuscript is well written and easy to read. 

The spatial profiling of Physarum by physically segregating it by centrifuging it into a 384-well plate is clever. While the approach is probably cannot be generalized to most organisms, it still provides a nice example of creative experimental design that is somewhat lacking in the single-cell genomics field at the moment. Moreover, given that there seem to be no/few published studies with RNA in situ hybridization gene expression patterns in this animal, it probably provides a wealth of information to Physarum researchers. 

Some aspects of the experimental design potentially limit the conclusions that can be drawn from the data. The authors find that plasmodia in distinct states of life (mitotic, non-mitotic, chemotaxing, contacting food) have broad syncytium-wide transcriptional differences. A major caveat of this finding is each separate condition was only profiled once without replicates, which makes it more difficult to tie which of these transcriptional differences are related to the samples' biological differences and which might be a batch effect. Additionally, it's not clear why the authors profiled different timepoints via snRNAseq (1 week with oat flake) and spatial RNAseq (only a few hours with oat flakes) in their experiment to assess feeding behavior. While the work identifies spatial heterogeneity and nuclear heterogeneity, they are not directly compared (how much of the nuclear heterogeneity is explained by spatial heterogeneity?) perhaps because different timepoints were used with the two approaches? 

Additionally, some aspects of the analysis seem to miss opportunities. A significant portion of the presentation of the gene expression results discovered by the authors is focused on the cell cycle, which seems less exciting than perhaps other biological phenomena related to structural specialization within different parts of the organism or related to its feeding and metabolic behaviors might be. Also, while multiple classes of nuclei (stationary and mobile) are identified, it's unclear how those relate to the different transcriptional states identified through the snRNAseq. 

Lastly, some aspects of the presentation detract from the work. Some of the results and discussion focus on 'coordinated intra-syncytial behaviors', and a major one of focus is that a wave of mitosis seems to proceed across the organism. However, to my knowledge, many syncytial systems (e.g. Xenopus or Drosophila embryos) exhibit synchronized mitosis, so I would have expected this to be the default state, rather than an exciting finding. Is this result unexpected? If so, it would be helpful if better contextualized. One aspect that would likely improve this manuscript would be to place it more firmly within the larger context of well-studied syncytial cells that exhibit specialization. For instance, a major example of a well-studied syncytium that exhibits spatial gene expression and nuclear specialization is the Drosophila embryo, which undergoes much of its early patterning while syncytial. Furthermore, muscle cells are typically syncytial, and some exciting recent studies have similarly used snRNAseq to observe heterogeneity and specialization of particular nuclei. Lastly, while not meant to diminish the contributions of this work, it does seem that given the diversity of syncytia that are well studied and exhibit nuclear specification, perhaps the title "Nuclei are mobile processors enabling specialization in a gigantic single-celled syncytium" oversells the results presented in this work.

Reviewer #1 (Public Review): 

In this manuscript, Fong et al. showed that temporary inactivation of the fellow (good) eye through injection of tetrodotoxin (TTX) led to long-lasting recovering from amblyopia beyond the critical period, using both mice and cats as model systems. In contrast, reverse occlusion only had short term effects. This work is built on previous works from the authors, showing that TTX injection did not have any obvious effects on neuronal health (DiCostanzo et al., 2020), and that reverse occlusion in cats induced anatomical recovery from amblyopia (Duffy et al., 2018). In summary, this work is clearly written, with a strong and simple message, and has potentially important clinical implications for the treatment of amblyopia. It could be significantly strengthened by some probing into potential mechanisms, especially whether mechanisms previously shown to be important for critical period plasticity are activated following temporary inactivation of the fellow eye. These may give insight into potential treatment strategies.

Reviewer #2 (Public Review): 

This manuscript by Fong and colleagues explores plasticity of circuits in the visual cortex of normally reared and amblyopic mice and cats. Previous work from this group had reported the exciting finding that transient inactivation of the non-deprived eye by intraocular injection of TTX can trigger substantial recovery of acuity in the deprived eye. Here the authors perform electrophysiological experiments to reveal that: 

1. Temporary inactivation of one eye in normally reared mice increases the visually evoke potential amplitude of the non-inactivated eye for more than 7 days <br> 2. Recovery of amblyopia from long term monocular deprivation is possible even in the adult <br> 3. Recovery is detectable even after 1 day of transient inactivation <br> 4. Recovery persists for a longer period of time than the recovery traditionally observed with reverse occlusion <br> 5. Inactivation has similar effects in mice and cat 

The experiments described in this manuscript are generally carefully performed and results are very clear. The information gained from this study are important in advancing our understanding of adult plasticity and the potential to reverse amblyopia. This reviewer has a few questions/comments about the interpretation of the data that should be addressed to improve the impact and clarity of this study.

Reviewer #1 (Public Review): 

This is collaborative work from a number of groups with complementary expertise in the generation of binders, Biological Chemistry and Structural Biology and super resolution microscopy and Neuroscience. The groups have a long history of effective collaboration and are widely known to consistently produce high quality work. 

The authors here build on previous (published) collaborative work in which they developed and characterised a number of small, non-antibody, protein binders to the important scaffold protein PSD-95, which through its PDZ-domains scaffolds AMPA-receptors via auxiliary proteins in the excitatory post synapse. 

They now in the present work aim to establish these binders as tools to investigate PSD-95 function specifically and synaptic physiology globally by using the binders as fluorescent markers for live and super resolution imaging and for the delivery of sensors for functional imaging. This is an important and worthy goal, as overexpression of fluorescent protein tagged PSD-95 is known to lead to artifacts and PSD-95 is an extremely important molecule to study. 

The authors measure using NMR of in vitro reconstituted binding to purified components and a previously developed dimeric binder to the dual PDZ-domains of PSD-95 whether binding of the proposed binders interferes with binding of PSD-95 to stargazin, an auxiliary protein that mediates AMPA-receptor recruitment to synapses or whether engagement of the binding sites changes attachment of their binder. They find that neither is the case. They then move on to express their binders coupled to fluorophores under control of an expression system that prevents too strong overexpression and see that the binders become recruited to post synapses. They go on to test if the binders change synaptic physiology in measuring EPSCs and find no change. They then move on to test their binders in a number of super resolution modalities by linking them to fluorescent proteins and dyes. They furthermore use the binders to recruit a calcium sensor to the postsynapse, which they successfully use to detect and signalling in single synapses in dendritic spines. 

The authors present very convincing data supporting their claims and provide very thorough characterization of the binders and their interaction with PSD-95 in live cells and during signalling events. The binders they present clearly do not interfere with PSD95 function as characterised and can be used to deliver dyes, probes, fluorescent proteins and sensors to functional postsynapses. This is an important step forward and very useful to the field.

Reviewer #2 (Public Review): 

The precise, simple and non-interfering visualization of neuronal key structures is a major challenge and currently limiting the advancement of our understanding of brain function. This work presents intrabodies as selective and non-interfering tools for the visualization of PSD95 - a major scaffold of the neuronal excitatory postsynapse. A toolbox of fully characterized modular imaging tools for use with various super-resolution microscopy techniques or to achieve synaptic enrichment of activity reporters of choice. 

Here four key issues of current visualization approaches are addressed: The problem of visualizing exclusively the relevant protein target and not the closely related proteins. The problem that conventional probes, like antibodies, may introduce a significant labeling offset or display low binding efficiency. The problem that live imaging approaches still largely rely on excessive genetic manipulation often including the artificial over-expression of a selected isoform or modification variant of the target protein. Finally, and possibly most importantly, the difficulty to visualize the target protein without interfering with its biological functions. 

In contrast to chemical probes and fused fluorescent reporters the provided intrabodies are depending on promoters and regulated expression. As a result the probes are either over-expressed or they display a sub-stoichiometric target binding. As such this work also highlights the need for the development of improved systems to precisely control and adjust the expression level of such probes to avoid probe excess and, as a result of this, low contrast visualization of the target protein. 

Due to the simplicity of application and the modular nature of the provided toolbox, it has the potential to be broadly used, especially in synergy with the emerging imaging techniques. Specifically, the verified target selectivity and the proven absence of functional interference can be expected to motivate the use of the provided PSD95 affinity proteins over the currently existing ones. 

It is worth mentioning that the semi-rational development approach of the PSD-95 intrabodies seems to be applicable to other proteins of choice, enabling the production of new modular imaging tools.

Reviewer #3 (Public Review): 

Rimbault, Breillat, Compans et al., set out to characterize and exploit the use of previously described synthetic binders of PSD-95 as imaging tools for a wide range of techniques. They characterized the binding properties of these probes and tested the impact of these binders on PSD-95 function. The authors also evolved a set of these binders through addition of different fluorescent modules, as well as an expression regulation system to achieve close-to-endogenous labeling of PSD-95. The authors provide an extensive toolbox and an in-depth characterization of the tools used. The set of probes described in this manuscript will be of use to the neuroscience community, but some aspects of the specificity and application of the binders could be clarified.