Reviewer #3 (Public Review):

This study investigated how reward-associated signals are represented in layer 2/3 neurons of the primary motor cortex. Water-restricted mice were trained to respond to a conditioned auditory stimulus in order to receive a water reward. Behavior analysis showed that mice quickly learned the association between the sound stimulus and the reward as indicated by increased anticipatory lick rate. Using this behavioral paradigm, neuronal activity was monitored throughout the training (7 days). Two-photon calcium imaging was performed separately from four different types of neurons; pyramidal neurons, PV-, VIP-, and SOM-positive interneurons. Tuning of individual neurons to the tone and reward stimuli were analyzed by using Spearman correlation between the trial-averaged fluorescence and the timing of stimulus delivery. Results showed that PV-positive interneuron responses became more reliable to the cue stimulus, whereas VIP-positive interneuron responses became more reliable to the reward stimulus. Some SOM-INs that were not responsive to the tone before training became responsive at day 7 of training. Activity of SOM-INs became more reliable to the reward after learning. The main findings are quite novel and may provide a new insight into the specific roles of interneurons. More representative imaging data and control experiments will make the story even more complete and convincing.

1) Imaging calcium responses from individual types of interneurons are important and challenging. Tracing activity changes from same population of neurons is especially important because it will show how learning shapes the pattern of changes in each neuron. Despite such powerful approaches, activity changes from each neuron were not shown. Calcium transients measured at day 1 were re-sorted at day 7, so it is not clear whether the same neurons responsive to cue or reward stimulus are still responsive to the same stimuli and, if so, how their onset timing is changed. Knowing whether the cue- or reward-sensitive population is the same population or not may lead to a different conclusion, so plotting calcium signals over days without resorting would be important.

In addition, representative calcium images from interneurons were not shown (like Fig. 1A). It seemed that about 80-90 cells of PV-INs, VIP-INs, SOM-INs were observed (Fig. 2D). Showing some representative images from individual cell types would be helpful for readers to better understand the results.

2) Identifying active cells that are above the chance level was good to define a subset of neurons responsive to a period of cue- or reward-stimulus. Quantifying the tuning of each cell's average response during the tone and reward response periods using non-parametric Spearman correlation was also a powerful way to display a subset of neurons with high or low trial-by-trial reliability. Results suggest that there are changes of less reliable neurons to more reliable ones in the case of PV-INs and VIP-INs (Fig. 4 and 5). However, whether these changes are specifically associated with learning is not clear. Running control experiments without water restriction or with random reward presentation independent of cues would be a good comparison. These experiments will help to rule out the possibility of naturally happening learning-independent changes from day 1 to day 7.

Reviewer #1 (Public Review):

The manuscript by Jasek et al. uses serial electron microscopy to reconstruct all 852 somatic muscle cells and their desmosomal inter-connectivity and connections to non-muscle cells and extra-cellular matrix structures in a 3-segmented larva of the nereidid Platynereis dumerilii. The study is complementary to a previous report by the same research group on the whole-body neuronal connectome using the same specimen. It is among the first studies to present a thorough and detailed analysis of the full complement of larval muscle cells and their fixations on a whole-body level and classifies them based on their morphological location and local inter-connectivity. The authors' choice to highlight the muscles around the acicula is highly justified given the pivotal role of acicula as endoskeletal anchor points for a highly diverse and complex set of muscles, thus important to understand the muscle movements controlling crawling behaviour. The authors use the morphological positions and connectivity of muscle groups to infer their role in moving acicula during crawling movements. A weakness of the manuscript is that it is quite difficult to follow how the authors inferred these acicular movements mainly due to the difficulty to represent spatial and temporal changes in a two-dimensional way. In addition, these inferred movements could only be directly tested for a small subset of muscles due to technical limitations in imaging the activity of worm muscles during locomotion. A thorough analysis of muscle functions seems however currently impossible due to technical limitations of in vivo calcium imaging of moving animals, and the complexity and speed of the muscular activities occurring during crawling movements. Altogether, the current manuscript forms a comprehensive, detailed and thorough basis for future studies aiming towards an understanding of locomotor activity from both neuronal and muscular perspectives on a whole-body level.

Reviewer #2 (Public Review):

Attachment of muscles to exo- or endoskeleton by desmosomes and hemidesmosomes is necessary for invertebrates to transform the muscle force into body movement. Jasek et al. fully reconstructed these functional connectivity units in a 3-day annelid larva by serial electron microscopy. The final dataset includes desmosomes from 852 muscles to all non-neuromuscular cells, as well as desmosomes that connect these non-neuromuscular cells.

The authors showed that this desmosome connectivity matrix exhibits a highly organized local community structure, and this structure matches the physical organizations of involved cells and desmosomes.

The authors attempt to relate structures to functions. A module with the highest membership diversity that reflects an anatomical organization is the interacicular muscle complex. The connectivity map at this complexity allows the authors to infer this structure's capacity for diverse motor patterns. They verified some of these motor functions by DIC imaging and muscle calcium imaging.

Strengths:

1) This is an impressive trove of data, essentially an annelid anatomical atlas on how muscles connect to the body. It provides the only complete dataset for non-neuronal tissue's connection to the neuromuscular system that underline body movement.

2) Quantitative assessment and description of the demosome network, which defines parameters to highlight its organizational difference from that of the neuromuscular system.

3) Generation of testable hypotheses for possible motor patterns to be produced from the hub of the desmosome connectome. Some, such as the independent and coupled movement of the notopodial and neurpdodial aciculae, were corroborated with behavioral and muscle imaging data.

Weakness:

Behavioral and behavioral calcium imaging is at a fairly early stage, due to understandable technical challenges, such as tracking 3D movements, discerning origins of cytosolic calcium signals from a large population of muscle cells, and the imaging preparation's difficulty to perform nature movement.

Reviewer #3 (Public Review):

This paper is based on digital reconstruction of a serial EM stack of a larva of the annelid Platynereis and presents a complete 3D map of all desmosomes between somatic muscle cells and their attachment partners, including muscle cells, glia, ciliary band cells, epidermal cells and specialized epidermal cells that anchor cuticular chaetae (circumchaetal cells) and aciculae (circumacicular cells). The rationale is that the spatial patterning of desmosomes determines the direction of forces exerted by muscular contraction on the body wall and its appendages will determine movement of these structures, which in turn results in propulsion of the body as part of specific behaviors.

To go a step further, if connecting this desmosome connectome with the (previously published) synaptic connectome, one may gain insight into how a specific spatio-temporal pattern of motor neuron activity will lead, via a resulting pattern of forces caused by muscles, to a specific behavior. In the authors' words: "By combining desmosomal and synaptic connectomes we can infer the impact of motoneuron activation on tissue movements". This is an interesting idea which has the potential to make progress towards understanding in a "holistic" way how a complex neural circuitry controls an equally complex behavior. The analysis of the EM data appears solid; the authors can show convincingly that desmosomes can be resolved in their EM dataset; and the technology used to plot and analyze the data is clearly up to the task. My main concern is with the way in which the desmosome pattern is entered in the analysis, which I think makes it very difficult to extract enough relevant information from the analysis that would reach the stated goal.

1.The context of how different structures of the Platynereis larval body, by changing their position, move the body needs much more introduction than the short paragraph given at the end of the Introduction.<br> -My understanding is that the larval body is segmented, and contraction of the segments can cause a certain type crawling or swimming: does it? Do the longitudinal muscles, for example, insert at segment boundaries, and alternating contraction left-right cause some sort of "wiggling" or peristalsis?<br> -In addition, there are segmental processes (parapodia, neuropodia), and embedded in them are long chitinous hairs (Chaetae, Acicula). Do certain types of the muscles described in the study insert at the base of the parapodia/neuropodia (coming from different angles), such that contraction would move the entire process, including the chaetae/acicula embedded in their tips?<br> -Or is it that only these chaetae/acicula move, by means of muscles inserting at their base (the latter is clearly part of the story)? Or does both happen at the same time: parapodium moves relative to the trunk, and chaeta/acicula moves relative to the parapodium? How would these movements lead to different kind of behaviors?<br> -Diagrams should be provided that shed light on these issues.

2.The main problem I have with the analysis is the way a muscle cell is treated, namely as a "one dimensional" node, rather than a vector.<br> -In the current state of the analysis, the authors have mapped all desmosomes of a given muscle cell to its attached "target" cell. But how is that helpful? The principal way a muscle cell acts is by contracting, thereby pulling the cells it attaches to at its two end closer together. As the authors state (p.4) "...desmosomes..are enriched at the ends of muscle cells indicating that these adhesive structures transmit force upon muscle-cell contraction."<br> -for that reason, the desmosomes at the muscle tips have to be treated as (2) special sets. Aside from these tip desmosomes there are other desmosomes (inbetween muscles, for example), but they (I would presume) have a very different function; maybe to coordinate muscle fiber contraction? Augment the force caused by contraction?<br> -As far as I understand for (all of) the desmosome connectome plots, there is no differentiation made between desmosome subsets located at different positions within the muscle fiber. I therefore don't see how the plots are helpful to shed light on how the multiplicity of muscles represented in the graphs cause specific types of neurons.<br> -As it stands these plots "merely" help to classify muscles, based on their position and what cell type they target: but that (certainly useful) map could have probably also be achieved by light microscopic analysis.

3.Section "Local connectivity and modular structure of the desmosomal connectome" p.4-7" undertakes an analysis of the structure of the desmosome network, comparing it with other networks.<br> -What is the rationale here? How do the conclusions help to understand how the spatial pattern of muscles and their contraction move the body?<br> -Isn't, on the one hand (given that position of the desmosome was apparently not considered), the finding that desmosome networks stand out (from random networks) by their high level of connectivity ("with all cells only connecting to cells in their immediate neighbourhood forming local cliques") completely expected?<br> -On the other hand, does this reflect the reality, given that (many?) muscle cells are quite long, connecting for example the anterior border of a segment with the posterior border.

4.In the section "Acicular movements and the unit muscle contractions that drive them" the authors record movement of the acicula and correlate it with activity (Ca imaging) of specific muscle types. This study gives insightful data, and could be extended to all movements of the larva.<br> -The fact that a certain muscle is active when the acicula moves in a certain direction can be explained (in part) by the "connectivity": as shown in Fig.7L, the muscle inserts at a circumacicular cell on the one side, and to an epithelial (epidermal?) cell and the basal lamina on the other side. But how meaningful is a description at this "cell type level" of resolution? The direction of acicula deflection depends on where (relative to the acicula base) the epithelial cell (or point in the basal lamina) is located. This information is not given in the part of the connectome network shown in Fig.7L, or any of the other graphs.

Reviewer #1 (Public Review): 

This manuscript examines how effort is integrated into economic decisions by recording neural activity from the dorsal anterior cingulate cortex (ACC) in monkeys. The ACC is a relevant area because some theories have suggested it is important for evaluating or selecting among potential actions during decision-making, although evidence supporting this idea has been inconsistent. The study design and analyses follow a long line of research from this group that has mainly focused on responses of OFC neurons during economic choices, so that the results in the present study can be directly compared to OFC which broadens the potential interpretation. 

In the task, subjects made choices between variable amounts of different flavors of juice that required different amounts of effort to obtain. The level of effort was indicated before the direction of movement, effectively separating the effort evaluation from action planning. The main results provide evidence against the notion that ACC contributes to evaluation of potential actions or computation of effort-based choices. Although neural signals correlating with effort evaluation were found, these were nearly all related to post-choice variables (i.e. the information encoded depended on which option had been chosen, meaning that they more likely reflect the outcome of the choice rather than the inputs to it). This is in contrast to OFC encoding in the same task (and same subjects), in which neurons encoded the effort associated with choice options. This contrast supports the role of OFC, but not ACC, in economic choices involving action evaluation, but leaves open the possibility that ACC retrospectively evaluates or tracks past choices that involve different effort costs. 

Overall, the study is well designed and executed, and the comparison with previously published OFC data provides an important contrast. Although the evidence against the notion that ACC is uniquely involved in evaluating actions to make a decision is reasonably strong, the impact of this result is modest given that there have been a number of previous publications refuting this perspective. I have a few specific comments below, but the major consideration in my view is whether the idea that ACC contributes to decisions by evaluating potential actions is still prominent enough for this study to be of high impact. Though I do think it could provide something of a final word on the issue.

Reviewer #2 (Public Review): 

Cai & Padoa-Schioppa recorded from macaque dorsal anterior cingulate cortex (ACCd) while requiring animals to choose between different juice types offered in variable amounts and with different action costs. Authors compared neural activity in ACCd (present study) with previous, directly comparable, findings on this same task when recording in macaque orbitofrontal cortex. The behavioral task is very powerful and the analyses of both the choice behavior and neural data are rigorous. Authors conclude that ACCd is unique in representing more post-decision variables and in its encoding of chosen value and binary outcome in several reference frames (chosen juice, chosen cost, and chosen action), not offer value, like OFC. Indeed, the encoding of choice outcomes in ACCd was skewed toward a cost-based reference frame. Overall, this is important new information about primate ACCd. I have only a few suggestions to enhance clarity. Figures 5 and 7 are maximally informative, but it is not clear that Figure 6 adds much to the reported Results. It is also suggested to abbreviate the comparison with Hosokawa et al. as it presently takes up 3 paragraphs in the Discussion: it is clear the methods and task designs were different enough to not be so easily compared with the present study. An additional suggestion would be to include mention of the comparison with OFC in the abstract and possibly also in the title, since the finding and direct comparison in Figure 7 are some of the most novel and interesting effects of the paper. Other suggestions are minor, and have to do with definition of time windows, variables, and additional papers that authors may cite for a well-rounded Discussion.

Reviewer #3 (Public Review): 

Cai and Padoa-Schioppa present a paper titled 'Neuronal Activity in Dorsal Anterior Cingulate Cortex during Economic Choices under Variable Action Costs'. They used a binary choice task where both offers indicated the reward type, reward amount, and the action cost (but not the specific action.) Variable action costs were then operationalized by placing targets on concentric circles of different radius. Here, and in a previous study that included OFC recordings (Cai and Padoa-Schioppa, 2019), monkeys integrated action costs into their decisions. Single-unit recordings in ACCd revealed that neurons predominantly coded for post-decision variables, such as cost of the chosen target and the juice type of the chosen offer, but not pre-decision variables, such as offer values. Given this finding, the authors compared the percentage of neurons in OFC and ACCd that coded for decision variables. In OFC neurons, the activity was mostly restricted to the offer presentation phase, whereas ACCd neurons showed sustained coding of chosen value and costs that lasted until the appearance of the saccade targets. Overall, this is an interesting study that provides evidence that decision-related signals evolve from coding offer values in the OFC to representing chosen costs in the ACC. This finding could highlight the roles of ACC neurons in learning and decision making. We have only a few questions. 

1) Do any of the variables used in this study correlate with a conflict? When the authors previously studied ACC, they discarded the conflict monitoring hypothesis - a hypothesis that is well established for ACC hemodynamic responses - for ACC single cell activity based on neural data from 'difficult' decisions (Cai and Padoa-Schioppa, 2012). The definition of difficulty they used, then, was descriptive and based on reaction times (RTs). They defined the most difficult trials as those trials with the longest RTs and discovered that those trials had options with similar offer values. This definition of choice difficulty appears to be contrived from evidence accumulation models/tasks, where normatively harder judgments elicit longer RTs. However, there is no normative economic reason that trials with similar offer values are more difficult or should cause conflict. After all, according to theory, choosing between two options with the same value is as easy as flipping a coin. Here, it seems like the authors could have a more fitting definition of conflict. For example, conflict can be operationalized by considering trials when the animal must choose between a high value/high-cost option and a low-value/low-cost option. In that case, the costs and benefits are in conflict. What do the RTs look like? Do the RTs indicate conflict resolution? If so, is this reflected in neuronal responses? 

2) The authors claimed that the ACCd neurons integrated juice identity, juice quantity and action costs later in the trial. As they acknowledge, the evidence for this claim is marginal. The conclusion the authors made in line 211, therefore, could be moderated. Given that the model containing cost-related variables is more complex, it is equally valid and more appropriately to write '... we cannot reject the null hypothesis that action cost was not integrated by chosen value responses later in the trial.

Reviewer #1 (Public Review): 

The authors address an interesting but neglected issue in pigment cell biology, concerning the developmental origin of red erythrophores, especially in relationship to yellow xanthophores, and the genetic basis for their differing pigmentation. Red-yellow colouration in vertebrates usually arises from accumulation of dietary carotenoids, and often has significant behavioural importance, e.g. as an honest signal of individual quality. This and the biochemistry of carotenoid colour variation is nicely covered in the Introduction, providing helpful background to a broad audience. 

The authors document the widespread presence of erythrophore in Danio, highlighting the unusual nature of Zebrafish within the genus as lacking them. They then develop some quantitative and objective measures of the xanthophores and erythrophores based upon Hue and Red:Green autofluorescence ratios, allowing clear distinction of the mature cell-types, and note the often binucleate nature of the erythrophores. 

The authors then use a variety of tools to assess, with differing degrees of certainty, the lineage relationships of the erythrophores; together these provide a consistent and convincing picture of shared lineage between the two cell-types. This is consistent with the observed gradual shift in properties of proximal cells from xanthophore-like to erythrophore. A more direct test of the conversion of early xanthophores to erythrophores comes from the clonal analysis of aox5:nucEosFP cells (Fig. 4). They then use a fin regeneration assay to assess the plasticity of these cells in the mature adult. This is a neat experiment, but I am struggling with the interpretation of Figure 5A: which cells are being used as landmarks to justify the conclusion that the cells shown are clonally-derived form that single cell in the 5 dpa image? It may be that the full series of images could be provided in a supplementary figure and might make this clear, but the current images do not seem convincing to me. The experiment in Fig. 5B is convincing, so conclusion seems sound. 

The authors then use a transcriptomic comparison to identify candidate genes influencing erythrophore v xanthophore differentiation. They study 3 with mutant phenotypes affecting these cell-types, identifying likely roles of 3 erythrophore genes. Whilst most of this analysis is beautifully presented, I am confused by Fig. 7 in which I think panel D and F as described in the legend are inverted. 

As is expected form this lab, the manuscript is generally very carefully and clearly written and includes thorough data presentation and statistical analysis. Conclusions drawn are appropriately nuanced, and justified by data presented. The manuscript provides an important first step in understanding the developmental relationship of erythrophores to xanthophores, and a number of genetic resources for the further exploration of this question.

Reviewer #2 (Public Review): 

In this paper the authors present a number of approaches to address the biology of the erythrophore, a red-pigmented cell present in non-mammalian vertebrates including birds, fishes and amphibians. This cell type is found in many species of the genus Danio but not in the widely-studied zebrafish (Danio rerio), therefore the investigators undertake studies in the relative Danio albolineatus, which has a prominent population of erythrophores in its proximal anal fin, with xanthophores (yellow-pigmented cells) located more distally on the anal fin. Both cell types use carotenoid derivatives as pigments, and this spatial arrangement provides an opportunity to address hypotheses concerning their ontogenetic relationship. The authors initially provide two lines of evidence that erythrophores and xanthophores arise from a common progenitor: fate-mapping via mosaic labeling of cells with transgenic markers as well as a stable line expressing a photoconvertible FP, and analysis of residual clones of pigmented cells in fish targeted by CRISPR to eliminate the scarb1 gene required for carotenoid uptake. Additionally, these experiments suggest that both cell types arise in the fin via differentiation from an unpigmented progenitor, as opposed to migration of committed or differentiatied cells from the body of the fish. The presence of bipotential progenitors is shown to be restricted to cells in the distal portion of the fin, while examination of regrowth of amputated fins reveals that xanthophores can be regenerated from unpigmented precursors as well as from division and transdifferentiation of erythrophores from the proximal fin "stub". In the last half of the paper, the authors identify a set of differentially-expressed genes between the erythrophore-enriched and xanthophore-enriched regions of the anal fin and subject a number of those from the former to CRISPR mutagenesis, with notable hits in the cyp2ae2 and bdh1a loci diminishing red coloration (and as shown by HPLC, levels of the corresponding ketocarotenoid astaxanthin.) The result with cyp2ae2 is especially intriguing considering the association of a related cyp2 P450 gene with red coloration in avians. The authors' conclusions are well-supported by their use of complementary approaches and clear presentation of a wide range of types of data.

Public Review:

Li and colleagues used data from 2000 to 2014 in 54 low and middle-income countries (LMICs) to study the association between exposure to landscape fire smoke PM2.5 and birthweight, including very low birthweight. While there is a relatively robust epidemiological literature that supports an association between non-biomass fire smoke PM2.5 and low birthweight, there are relatively few studies that are specific to biomass smoke PM2.5 and birthweight. The authors of this paper conducted their study to specifically address this data gap. They took advantage of satellite data which provide estimates of PM2.5 levels that are now available for most locations in the world at a high geographic resolution (0.5 x 0.5 km). They enhanced the satellite exposure data using a chemical transport model to distinguish fire-sourced PM2.5 from non-fire PM2.5. The exposure modeling approach is sophisticated as is the statistical analysis of the association between the fire-sourced PM2.5 exposure estimates and birthweight outcomes.

The study has multiple strengths, including the first study of the association between fire-sourced PM2.5 and birthweight to use a sibling-matched case-control design, a large sample size (227,948 births born to 109,137 mothers), the focus on LMICs, the exposure modeling, a careful statistical analytic approach with alternate non-linear regression and sensitivity analyses, and the outcome of very low birthweight that is one of the World Health Organization targets to reduce the global burden of disease. Limitations notwithstanding, this is an impactful study. The results of the authors' analyses provide strong support for the concept that exposure to biomass smoke -- whether from a landscape fire set by farmers, a wildfire, or cooking with solid fuels -- can lead to low birthweight. This concept is especially important for LMICs that have large portions of their populations engaging in slash and burn agriculture and/or cooking with solid fuels. Given that reducing the incidence of low birthweight is a necessary to meet the 2025 United Nations Sustainable Development Goals, it is incumbent that policies to reduce landscape fires and household air pollution from cooking with solid fuels be considered by governments of LMICs. Such policies would also have a climate change mitigation benefit through reduction of greenhouse gases and aerosols.

Future research efforts to actually measure landscape fire smoke PM2.5 in different locations to provide ground-truthing for the chemical transport model exposure estimates used by the authors would be useful as would a study that could obtain gestation duration data.

Public Review (Reviewer #1): 

Sarm is an important neurodegenerative protein. A series of papers over the past year described how Sarm is regulated, identifying an allosteric pocket that binds at least three NAD-related metabolites in order to inhibit or activate Sarm. The current study identifies and characterizes a fourth metabolite that binds in the allosteric pocket. Importantly, this is not a natural NAD-related metabolite, but instead is a metabolite of the neurotoxic pesticide vacor. Hence, this study shows that exogenous toxins can function by direct activation of Sarm. The study convincingly demonstrates that vacor toxicity is Sarm-dependent, identifies the relevant metabolite as vacor mononucleotide (VMN), shows that VMN directly binds and activates Sarm, and solves the structure of Sarm bound to VMN and compares this structure to the binding of the natural activator nicotinamide mononucleotide (NMN). The major strengths of this study are the high quality of the data and the demonstration that an exogenous toxin can activate Sarm. Surprisingly, the manuscript also shows that VMN can inhibit Sarm. The manuscript could be improved by working out how VMN can inhibit as well as activate Sarm.

Reviewer #2 (Public Review): 

This study provides definitive and convincing data that vacor, a neurotoxin, drives neurodegeneration by direct binding and activation of SARM1. The authors show that vacor is metabolized into VMN, which binds the ARM domain of SARM1 with high affinity, and drives activation of the SARM1 NAD hydrolase function to drive neurodegeneration. The study is rigorous, and uses a combination of biochemistry and structural biology with the ARM domain to make a convincing case that even in vivo, this activaiton is direct. The complete suppression of vacor-induced toxicity in DRGs and retina argues further that SARM1 is critical for neurodegeneration in these tissues and SARM1 blockade should be beneficial for patients. Finally, the model they have developed in the visual system will provide a nice assay for in vivo testing of small molecules for inhibition of SARM1.

Reviewer #3 (Public Review): 

In this manuscript, Loreto and colleagues show that vacor, a now banned pesticide, induces neurotoxicity via a Sarm1-dependent mechanism. Exposure to vacor induces both axon and soma degeneration that is blocked in Sarm1-deficient neurons. This is demonstrated in vitro in DRG and SCG neurons and in vivo in the eye with intravitreal injection of vacor. Importantly, the authors examine multiple potential mechanisms by which vacor may activate Sarm1 and elegantly demonstrate that the vacor metabolite VMN can directly bind to and activate Sarm1. The binding of VMN to Sarm1 is then examined at high resolution with crystallography studies with follow up mutational studies that confirm this model of interaction. 

Overall, the experiments conducted are thorough and examined with dose and time dependent studies in primary neurons. The in vivo studies with functional readouts are a very good complement to the in vitro experiments. The mechanisms experiments are also well designed and use complementary approaches to demonstrate that the vacor metabolite NMN is the direct activator of Sarm1. Lastly, the crystallography results provide a detailed perspective of the mechanism of binding of NMN to the ARM domain of Sarm1. 

These results have significant implications. First, they provide a mechanistic insight into the neurotoxicity of vacor and illustrate the critical function on Sarm1 in this context. Second, they identify vacor and NMN as reagents that are potent activators of Sarm1, a finding that other researchers will likely utilize for their studies on Sarm1. Third, the structure studies identify key residues that are involved in Sarm1 activation. This information will be very useful for the screening and identification of Sarm1 inhibitors. Sarm1 inhibition is being actively pursued by pharmaceutical companies as a therapeutic strategy for maintaining neuronal integrity in the context of neuronal injury or disease.

Reviewer #1 (Public Review): 

This manuscript is based on prior work, which demonstrated the inhibition of phage infection by aminoglycosides in Mycobacteria. Following up, the authors demonstrate in this manuscript that the acetylated form of apramycin retains its anti-phage activity while blocking its antibacterial activity. Along these lines, the authors describe that MgCl2 diminished the anti-phage effects of aminoglycosides, presumably due to inhibition of aminoglycoside uptake. Then the authors investigated if apramycin impacts the initial stages of phage infection, and found that DNA injection was not impacted leading to the important question of the apramycin target and if phages would actually encounter apramycin before or after infection, and which effects other aminoglycosides might have on the infection process.

Reviewer #2 (Public Review): 

The primary observations are very interesting. Their impacted is lessened by the prior demonstration of the inhibition of phage infection by aminoglycosides in Mycobacteria (Reference 18). This prior work also suggests a similar stage of phage infection is inhibited. As a consequence, perhaps the most novel and striking aspect of the paper is the demonstration that the acetylated form of apramycin retains its anti-phage activity while blocking its antibacterial activity. Further development of this aspect of the paper, or greater mechanistic insight into how aminoglycosides block phage infection would clearly strengthen the work.

Reviewer #3 (Public Review): 

The authors effectively show that aminoglycosides can protect against phage infection - and that this is by inhibiting lytic infections and not by promoting lysogeny. 

It was notable that not all aminoglycosides had a protective effect. It would be helpful if the authors could comment on the differences between those that had an effect versus those that didn't in their discussion. 

It was observed that MgCl abrogated the anti-phage effects of aminoglycosides, and noted that MgCl can inhibit aminoglycoside uptake. From this, the authors concluded that antibiotic uptake was likely important for the anti-phage effects - and this was nicely (but indirectly) confirmed by showing that the initial stages of phage infection (adsorption and DNA injection) were not impacted by the antibiotics. The observation did, however, raise the question of whether bacteria whose antibiotic resistance is conferred by efflux would be more susceptible to phage infection than those which modify the antibiotic itself? This would be a point worth addressing in the discussion. Is antibiotic efflux more commonly associated with other antibiotic classes? 

Testing the phage resistance of natural aminoglycoside producers effectively showed that these Streptomyces species could produce potent anti-phage compounds. For S. tenebrarius, an apramycin producer, spent medium was observed to have strong protective effects. Are there S. tenebrarius strains available that are defective in their ability to make apramycin? This would be a useful control to include to ensure that these strains aren't making another metabolite/releasing another compound with anti-phage activity. Similarly with the kasugamycin-producing strain - purified kasugamycin didn't have a profound effect on phage defense, based on the data presented in Fig. 1c or 2a, yet the addition of culture supernatant from the producing strain was as effective as apramycin in protecting against phage predation. It would be useful to be able to separate the aminoglycoside-specific effects, from the effect of other compounds produced by these strains (it would also be useful to know if they make anti-phage adjuvants that enhance the effects of the aminoglycosides). 

The demonstration of distinct anti-bacterial versus anti-phage properties for apramycin is fascinating. Given that many of the initial experiments presented in the manuscript were done using host strains expressing antibiotic-modifying resistance determinants, might this suggest that all results obtained to this point were with modified antibiotics? (despite unmodified variants being added to the cultures) This could suggest a trade-off between anti-phage effects versus anti-bacterial effects (for microbes that are not resistant - or that do not encode antibiotic-modifying resistance determinants). Which effects are more potent? This would be an important point to discuss/address (in some cases, the concentration of antibiotic needed to protect against phage infection seems quite high).

Reviewer #1 (Public Review):

This article describes single-molecule measurements of two conformational changes that take place during transcription initiation: DNA unwinding and RNAP clamp movements. The work is technically sound and the incredibly clean single-molecule traces we have grown accustomed to from this group are beautiful.

The claim is that the data exclude a model where the RNAP clamp needs to acquire more open conformations in order for DNA unwinding to take place. The measurements appear to support this claim, but fall short in completely excluding the model due to some missing experiments.

The two main data sets presented show that within the time required for DNA unwinding, no RNAP conformational changes consistent with the open-most clamp structure are observed. However, as this is performed on a single, unnatural promoter sequence without presenting negative controls for specificity, no broad conclusions can be made. This severely limits the impact that this work would have on the field of bacterial transcription.

There is also a danger in referring to "open" and "closed" states in general as it appears that a range of clamp openings are possible. Therefore, what someone labels as "open" from structural work may or may not be what is considered "open" in these relative fluorescent experiments.

Reviewer #2 (Public Review):

The structure of cellular multi-subunit RNA polymerases is highly conserved, with a notable feature being their "crab claw" shape. Since the structure of RNA polymerase was first determined there has been a great deal of interest in how the opening and closing of that claw might affect the steps in transcription initiation, elongation and termination. Transcription initiation poses particular problems, as RNA polymerase must unwind the double-stranded DNA to create a region of single-stranded DNA that can be used as a template for RNA synthesis. Different models have proposed that RNA polymerase opens up to bind double-stranded DNA that is then unwound, or that it is unwound first and then a single strand is loaded into the narrower opening in a "closed claw". There is experimental evidence in favour of each of these models. In this paper single molecule techniques are used to monitor both the unwinding of the double stranded DNA and the opening of RNA polymerase. The results suggest that RNA polymerase does not need to adopt an open conformation in order to unwind and load DNA, and hence that unwinding is likely to occur outside the main cleft of the enzyme. The work also detects a post-loading step in which the RNAP closes more tightly on the DNA, which is proposed to act to "lock" the polymerase onto the DNA. The results presented in the paper are an important addition to our understanding of this key step in gene expression.

Strengths

The work has been done with RNA polymerase from Escherichia coli, which is perhaps one of the best-studied RNA polymerases, and uses the most prominent sigma factor binding to a canonical promoter. The system has therefore been chosen to be off wide interest and applicability, and also sets the groundwork for similar studies in other systems. The development of ideas is logical and multiple lines of evidence are used to test the validity of the conclusions. For example, on finding a difference in kinetics between the unwinding of the upstream and downstream ends of the bubble the assumption that this represents directional unwinding is tested by increasing the strength of the intervening base pairing, and the expected changes in rate are seen. Similarly, the requirement (or otherwise) for clamp opening is demonstrated both by single molecule FRET experiments and by testing the effect of an inhibitor of clamp opening. The experiments are presented clearly in both the text and the figures, and the results are convincing. Overall the correlation between the rates of strand opening and clamp locking is compelling, and the conclusion that the clamp need not open to form the open complex, and closes more tightly after bubble opening seems to me to be convincing.

Weaknesses

I have noted above that the experimental system was carefully and appropriately chosen. And the authors note that the tools developed here can subsequently be applied to other systems, and list some of those that they feel may be of interest. This acknowledges one potential weakness of the study, which is that it studies essentially a single promoter, and what is true of this system may not be true of all. Given that the main conclusion is that there is not an absolute requirement for clamp opening in order for an open complex to form I feel a single system is sufficient to make this claim. Further limitations of the experimental systems also leave some potential room for argument: the time resolution of the experiments means that very rapid opening would not be detected, and as DNA opening and RNA polymerase opening were measured in separate experiments the conclusions rely in part on correlation between results gained in slightly different conditions. But I do not think that these caveats undermine the authors' conclusions. Finally, the smFRET experiments monitor the relative position of two residues on either side of the clamp. These report on clamp opening (as stated in the paper) but do not reflect movement of other parts of RNA polymerase that might facilitate entry of double-stranded DNA into the cleft. Movement of the beta lobe is of particular interest in this context, given prior structural and modelling work: Unarta et al PNAS 2021 118 e2024324118, Chen et al Mol Cell 2020 78 275 and references therein. If, as suggested in these other papers, movement of the beta lobe might allow double-stranded DNA to be accommodated in the RNA polymerase cleft without substantial clamp opening the logical link between the data presented here and an "unwind then load" model might be weakened.

Reviewer #1 (Public Review): 

The paper describes the development of a mechano-chemical model for plant root development that incorporates mechanical aspects of cell growth and division, chemical aspects of (amongst other auxin processes) polar auxin transport and auxin patterning, and the feedbacks between these processes. As such it presents a significant advance relative to other root models that have focussed predominantly on either the mechanical or auxin patterning aspects of root development, as evidenced by the potential of the model to reproduce a series of hormonal and mechanical perturbation experiments. Additionally, the efficient flexible manner in which mechanics are incorporated make the model potentially suitable for studying e.g. conditions in which aberrant division planes lead to additional cell file formation, or for studying tissue shape regeneration after root tip excision. 

Still, the claim that this study reveals a set of minimal principles for self-organized root tip patterning in which interplays between mechanics, growth and auxin patterning are essential is less strongly substantiated. By superimposing an auxin source in the middle vasculature cells and an auxin sink in the topmost outer cells the authors effectively impose the polar auxin transport directions of the tissue outside the simulated domain, which likely causes the simulated domain to align with this rather than displaying truly self-organized patterning. Also, in the model mechanics has been made to impact PIN polarity, but it has not been demonstrated if in absence of PIN polarity dependence on mechanics PIN patterning would be different, i.e. if the mechanical feedback on PIN patterning is necessary or rather that the source and sink prepattern dominate. Similarly, the feedback of auxin on mechanics is I believe limited to cellular auxin levels determining cellular growth rates and does not appear to control cellular growth anisotropy (i.e. predominant longitudinal growth of cells), which arose from the initial symmetry breaking of differential growth rates. As such the actual coupling between mechanics and auxin patterning and the extent of self-organization is less than suggested. 

Some matters are rather unclear in the manuscript in its current form. It is for example unclear how exactly does auxin translate into cellular growth rate and how does this result in stable, coordinated growth across cell files. In a previous study it was shown that since auxin levels differ significantly across cell files (e.g. much higher in vasculature than in neighboring cell files), problems in coordinated cell growth may occur (https://pubmed.ncbi.nlm.nih.gov/25358093/). It is unclear how such problems are avoided here. Similarly, in the discussion authors mention possible uses of the model such as studying tropisms. However the latter requires incorporating an elongation zone in which cells undergo rapid and extreme cell elongation. It seems that the current model only incorporates slow cytoplasmic cell growth and division occurring in the meristem, and it is unclear whether the used model formalism would be capable of stably simulating these far more anisotropic growth processes in a numerically stable and efficient manner. 

Summarizing the authors have generated a highly valuable combined mechano-chemical modeling framework for root tip development that can be used to various applications, but that was somewhat oversold.

Reviewer #2 (Public Review): 

The work by Marconi et al. echoes previous work conducted at the shoot apical meristem where PIN transport has been modeled in realistic templates, and with mechanics, for iterative morphogenesis (phyllotaxis). The two most appealing parts of this study to me are: 

(i) The initial symmetry breaking event where differential growth can promote anisotropic root growth (a trichome-like model for the radicle) <br> (ii) The link between growth and reflux loop, including the tests in mutants and decapitation/regeneration. In particular "auxin influx from the LRC and subsequent 'bipolar' PINs localization in the cortex tissues may be important elements of the sustained auxin-dependent root growth" is key. 

Somehow, the underlined molecular hypothesis (phosphatase model) is a bit secondary - this is rather an algorithmic way to generate emerging properties; other molecular mechanisms might provide the same outcome. Overall, this represents a major advance in the field. However, some points would require attention: 1/ PIN polarity rather depends on actin filaments, thus the relation between cytoplasmic microtubules, cortical microtubules, and mechanical stress is a bit confusing, 2/ the paper describes two systems (the radicle from the embryo and the growing primary root) with no clear link between both, 3/ Although I understand that this paper only describes a model (based on existing wet data), the wording is sometimes misleading and suggests that digital data are also real data.

Reviewer #3 (Public Review): 

Marconi, M et al. developed a new mechano-biochemical computational framework to study plant morphogenesis. A positional information is self-organized by a diffusing substance that regulates acquiring cell polarity and modulates cell growth by changing the cell's biomechanical properties. The model for the root meristem functioning in Arabidopsis thaliana is composed of a minimal set of experimentally derived principles for self-organization of organ patterning. This study is an excellent methodological achievement that also brought some new biological results. Although all the results are there, the manuscript requires major revision to present the framework better to avoid miscommunication. 

Framework: 

Good: The framework looks very promising to study tissue morphogenesis (not necessarily in plants). So it has the sense to make it available for the scientific community. Now the code can be downloaded from the google disc using a password; I encourage the authors to add a permanent link and the tutorial to the framework in the final version of the paper. 

Not good: The manuscript structure does not allow us to see all the advantages of the framework; the information about it is spread throughout the text. The main text misses essential details, while the materials and methods section contains a lot of discussions. E.g., CTM are mentioned several times in the results section without explaining how you modeled them. Another example is on the line 188 where the reference about "put together growth biomechanics" leads us the Figure 1 without any details about that. I would suggest adding a first section of the Results and the main figure about the framework, its basics, advantages, and limitations. Also, to add a bit more about PBD technology. 

Root model: 

Good: 

• The application of the model delivered some new and important results for plant biologists. E.g., the authors confirmed the hypothesis that the start of anisotropic root growth (elongation) results from the differential expansion of neighboring tissues.

• They also showed self-organization of a complex auxin distribution (PIN polarization) map, which reproduces tiny properties like the switch in PIN polarity from rootward to shootward in the cortex.

• The authors also showed that auxin reflux in the meristem is not that important for auxin maximum maintenance under normal conditions but gives an advantage in shoot-independent growth.

• The fact that the authors were able to "grow" a "whole root" from the embryonic-like structure under a limited number of predefined rules is inspiring and promising.

Can be better: 

1) I am not sure that I understand the visualization of PINs on the figures. The authors do not distinguish between PIN levels and PIN polarities. It is clear that they managed to model PIN polarities correctly, but I am not sure about PIN levels. If PIN levels are shown by red rectangular, then columella does not have any PINs (which is incorrect). Is it so? 

2) It sounds great that you simulated an oscillating behavior in the root growth, but actually, you did not. Simulating periodic application of auxin treatment to simulate oscillation is a trivial solution, that only sounds great, but disappoints when you look into detail. Instead, I would suggest simulation of restoration of root growth after it is inhibited by auxin application. 

3) Looking at the resulting solution, it looks like that auxin maximum in QC emerges because one QC cell adopts auxin from two vascular cells. Or maybe because columella does not have PINs. If it is so, it has to be stated; otherwise, there is an impression that auxin maximum self-organizes due to flux pattern only. =

4) It also looks like that QC does not grow during calculation, if it so, it has to be stated, because then it is one of the factors that "eventually reproducing the non-trivial shape of the root" (line 214). The rules to get "the non-trivial shape" should be explicitly stated. E.g., how did you get that only columella stem cells have divided and not other columella cells that are bigger in size? 

5) I noticed the formation of left-to-write asymmetry in auxin distribution in the root meristem that was self-organized. It will be great if you elaborate on that more. Especially considering that you simulated pin2 mutant where this asymmetry got lost. 

6) I was also quite excited looking on the bending root (Figure 2 - supplement 8) correlated with this left-to-write auxin distribution asymmetry. It will be great if you elaborate on the root bending more in the manuscript. 

Not good: 

1) There is a mess with anatomical terminology usage in the manuscript. The authors name the basal part of the embryo as the root or radicle, which is incorrect. There is no root or hypocotyl at the heart stage and even at the torpedo or bending-cotyledon stages. Consider starting growing the root from the mature embryo stage and not from the heart one. Otherwise, you study the formation of "apical-basal polarity" in embryo development and not root development. 

2) Another problem with the terminology is that the definitions are sometimes given later than they were used, and there is a lot of introduction in the results and material and methods section, which disturb the comprehension of the text. E.g., QC is introduced in the last results chapter (line 324). 

3) The only experimental data given in the manuscript is an analysis of hypocotyl and root growth in the seedling soon after germination. Using this data to confirm the results on "growing the root" from the heart-stage embryo is confusing. The results of the experimental data analysis are also quite obvious and did not require an experiment :). Having Eva Benkova among the co-authors, the manuscript should be supplemented with better experimental data that confirm or demonstrate the modeling results' appropriateness. The authors refer to confirmations from the published data in the rest of the text instead of explicitly comparing computational and experimental results. The manuscript will certainly win from such a direct comparison. 

Actually, it would not be bad if the authors did not use the experimental data at all, because all the facts discussed are well known; the authors can give just references. It just looks strange why only one (and not very relevant) experiment is shown from a vast collection of Prof. Eva Benkova.

Reviewer #1 (Public Review): 

This work demonstrates that functional KATP channels exist in most neuronal cell types in the mouse somatosensory cortex. While the transcriptomic profiling of electrophysiologically characterized neurons is only indicative of the existence of the Kir6.2/SUR1 KATP channel, the acute slice pharmacological/electrophysiological experiments convincingly supports this notion. The uncertainty of single-cell RT-PCR is likely due to a small amount of starting material inherent to the sample collection method. As the authors discuss, low copy numbers of target transcripts may also have contributed to the negative/uncertain results. Next, the authors demonstrate that lactate is taken up by neurons and elevates the discharge rate via an increased ATP production due to the oxidative metabolism downstream of lactate, which is in line with earlier studies including Ivanov et al. (2011, doi: 10.3389/fnene.2011.00002). The authors showed this by introducing 15 mM lactate, and discuss a possibility that extracellular lactate can be elevated by a systemic increase of lactate. However, such an increase is likely more modest in the brain (Carrard et al., 2018, doi: 10.1038/mp.2016.179). So, the lactate-enhanced firing might occur in extreme conditions such as during anoxia or ischemia; however, intracellular ATP would most probably decrease and hence KATP channels would open in this case. A discussion on extracellular lactate levels in physiological conditions would be helpful. Overall, this is a rigorous study that confirms the existence of functional KATP and dominant oxidative metabolism in most types of juvenile somatosensory cortical neurons.

Reviewer #2 (Public Review): 

The authors present an impressive array of experiments testing the effect of lactate on a number of neocortical cell types. They uncover a mechanism by which lactate might enhance neuronal firing although direct physiological relevance needs further support for CSF lactate concentrations. Most of the experiments are sound and interesting and the remaining experiments have limitations inherent to the methodology and presented accordingly in the discussion. The results are convincing, however a number of specific points need to be addressed. 

Specific points:

• Page 6 line 21 onwards. The authors state consistent expression of Kir6.2 and SUR1 in various cortical cell types. Data presented in Fig1 challenge this statement showing that Kir6.2 and/or SUR1 was expressed in the minority of cells tested regardless of cell type. For example, out of the 10 intrinsically bursting cells shown in the Ward cluster plot on Fig1A-B, only two was positive for Kir6.2 according to Fig1D. Surprisingly, Fig1F shows that 10% of intrinsically bursting cells express Kir6.2 which is clearly not the case (it is 20%). Amplification was used for the detection of mRNAs by the authors, thus it is unlikely that detection threshold plays a role in having Kir6.2 or SUR1 negative cells. Along the same vein, amplification makes it difficult to understand what the authors mean by "low copy number at single cell level". Specifically, the sentence (p6l22-25) is self-conflicting suggesting reliable detection of KATP subunits yet downplaying the significance of moderate single cell detection rates. I think a moderate statement with percentages of expression would adequately describe the findings with an emphasis on potential variability between individual cells regardless of cell type. Throughout the text, the authors should avoid the use of uniform expression of KATP channels in neurons.

• Page 6 line 30. The authors conclude co-expression of Kir6.2 and SUR1 subunits. Fig1D shows that out of approximately n=71 Kir6.2 positive cells and n=28 SUR1 positive cells only n=16 expresses Kir6.2 and SUR1 together and the evidence presented shows that n=83 cells do not co-express Kir6.2 and SUR1. Again, the conclusion in the manuscript seems biased towards the minority of cases and does not reflect the overall dataset. Accordingly, the suggestion that neurons and beta cells use the same KATP channel is not supported (p6l32).

• KATP channel presence in neurons. With respect to the points above, it would be helpful to see in the results section and possibly on Fig2 whether there is an electrophysiological indication of pharmacologically unresponsive cells. This would help in assessing the relative sensitivity of the two approaches. Fig.2G is helpful here, however signal to noise is hard to assess in the current version in individual experiments. Please state if single cell PCR was performed on any pharmacologically examined cells. Fig3B recapitulates the results of Fig1 that only a small fraction of RS cells express Kir6.2 and SUR1. In spite having a clever pharmacological design, due to limitations inherent to spatially nonspecific drug application methods, one cannot exclude that the results measured on individual cells could also reflect network interactions with astrocytes and/or neurons and should be discussed.

• Lactate concentration in blood vs CSF. As the authors point out, there is a discrepancy in glucose concentration between the blood and CSF, yet they use lactate concentrations measured in the blood (and not in the CSF) during exercise in their experiments. The physiological relevance of these experiments is unclear unless there is evidence that lactate concentration in the CSF is indeed in the range found effective here.

• MCT1 and MCT2 expression and widespread lactate effects. Here, the authors admit that relatively low single cell detection rates were observed for MCT1 (19%) and MCT2 (28%). It seems consistent (and a bit worrisome) throughout the manuscript that expression of mRNAs additionally tested functionally have a limited range of PCR detection yet (again) ubiquitous presence was found when tested pharmacologically.

Reviewer #1 (Public Review): 

This study sets out to critically test two competing models of Start activation, namely an increase in the Start activator Cln3 versus a decrease in the Start repressor Whi5 as cells progress through G1 phase. Evidence has been previously published in support of both models by different groups. The authors use a dual tagged strain (CLN3-HA6, WHI5-HA3) in which both Cln3 and Whi5 can be detected on the same blot with the same antibody to enable parallel quantitative measurement of each protein in the same samples. To follow each protein as a function of cell size and cell cycle position, early G1 phase fractions of small cells are obtained by elutriation followed by inoculation into either fresh rich or poor nutrient medium. Many replicates of these elutriation experiments in different contexts show that Cln3 levels rise substantially as cells approach Start in both rich and poor nutrient conditions (about 10X and 2.5X respectively for absolute amount per cell, 7X and 2X for calculated concentrations), whereas Whi5 levels remain roughly constant throughout G1 phase in both nutrient conditions. The decrease in Cln3 occurs within minutes of a shift into poor medium, consistent with a role for Cln3 in coupling growth to division. To further test the Whi5 dilution model, the authors re-examine the genetic effects of altered WHI5 gene dosage and show that doubling the WHI5 dosage has no effect on size in wild type cells. Even very high level overexpression of WHI5 from the TEF1 or GAL1 promoters has only a modest effect on size, measured on both asynchronous and synchronized cultures. The authors then test various ways to interfere with growth in G1 phase and find that inactivation of the secretory factor SEC7 by an auxin inducible degron completely halts growth and the accumulation of Cln3, but without any apparent effect on bulk protein synthesis. The dependence of size and Cln3 accumulation on different candidate upstream kinases is then evaluated. The authors find that the redundant Yck1 and Yck2 kinases are partially required for cell growth in G1 phase and completely necessary for Cln3 accumulation in both rich and poor nutrient conditions. Cln3 accumulation thus fails to occur in inhibitor treated ypk1-as ypk2 cells. Consistently, Ypk1/2 activity as measured by Thr662 phosphorylation is rapidly lost upon shift of cells from rich to poor nutrient conditions. The loss of Ypk1/2 does not appear to affect bulk protein synthesis, suggesting that Ypk1/2 may couple growth to Cln3 accumulation. Finally, given the known role if Ypk1/2 in sphingolipid biosynthesis, the authors test for effects of sphingolipid synthesis inhibitors. The inhibitor myriocin reduces growth rate/budding onset and delays Start but only modestly attenuates Cln3 accumulation. However, genetic inhibition of ceramide production in a lac1 lag1 strain dramatically reduces Cln3 levels. Addition of exogenous phytosphingosine similarly reduces Cln3 levels, particularly in the lac1 lag1 strain. From this result, the authors conclude that feedback inhibition of Ypk1/2 by sphingosine, either due to a block in conversion to ceramides or addition of excess exogenous sphingosine, blocks the accumulation of Cln3. Overall, this study helps to resolve the on-going controversy regarding the mechanism of Start activation in budding yeast and adds new insight into Cln3 regulation by Yck1/2.

Reviewer #2 (Public Review): 

This paper investigates cell size-dependent regulation of G1/S cell cycle transition in budding yeast, with a focus on the relationship between the activator Cln3 and the inhibitor Whi5. A prominent 2015 paper proposed that cell growth dilutes the inhibitor Whi5 while Cln3 levels remain constant. This 'inhibitor dilution' model has been challenged by several recent papers. In the present paper, Sommer et al. perform a series of quantitative western blots of whole cell extracts from synchronized cell cultures. They show that Cln3 concentration increases 10-fold before bud emergence (i.e. G1/S) but Whi5 concentration is largely constant, at least in rich media. Similar results were obtained in poor carbon media with a smaller increase in Cln3. These data argue against the inhibitor-dilution model and indicate that Cln3 levels are tuned by carbon availability and cell growth rate. Interestingly, Cln3 increases are not dependent on actin-based growth or bud emergence, but rather depend on membrane trafficking and TORC-SGK signaling. A series of experiments altering ceramide synthesis identify a link with Cln3 synthesis, although it remains unclear how directly this ceramide-Cln3 connection occurs. 

The combination of results in this paper represent a significant contribution to the field. Major strengths include the careful quantitation of Whi5/Cln3 levels, and the clear effects on Cln3 from membrane trafficking events. I also appreciated the balanced tone of the text, which describes the strengths and weaknesses of each experiment and interpretation. I have a series of comments/concerns that could be addressed to strengthen the paper, as described below. 

1) I understand why cells were pre-grown in poor carbon media for these experiments, but it seems important to know how Cln3 and Whi5 levels change for cells pre-grown in rich media. Otherwise, each paper reporting different results for Cln3/Whi5 could be dismissed as using a unique set of growth conditions. Along these lines, it would be ideal for the authors to test Cln3/Whi5 levels in their western blot assay using the same strain background and media as the Schmoller paper. It would be very interesting if the inhibitor-dilution model were observed under these conditions, whereas alternative mechanisms like Cln3 accumulation were observed under other conditions. 

2) The authors over-express WHI5 to test the inhibitor-dilution. Their results dovetail with a recent study from the Murray lab (Barber et al., PNAS) suggesting that cells are not very sensitive to Whi5 levels. However, one can envision mechanisms (e.g. PTMs) that inhibit Whi5 molecules when expressed beyond their physiological concentration. Instead, it would be interesting to know what happens in WHI5/whi5 heterozygous diploid mutants that cut Whi5 levels in half. Perhaps this experiment exists in the literature, but it would be an ideal setting for the authors to perturb the inhibitor-activator ratio, and test Cln3/Whi5 protein levels along with cell size in synchronized cultures. 

3) I found the result in Figure 5E very correlative and hard to interpret. For example, Ypk1 phosphorylation is lost at 2.5 min, but Cln3 levels seem unaffected at this timepoint and the next (?). I would suggest softening the (already soft) tone of explaining these results. In general, the connection between ceramide synthesis and Cln3 levels remains quite unclear to me. 

4) The text would need to describe a potential role for protein localization in this pathway. All the results come from cell extracts, whereas local protein concentration in the nucleus could be changing and impact the pathway.

Reviewer #3 (Public Review): 

In this manuscript Sommer et al. investigate how cell growth in G1 induces cell cycle entry in budding yeast. This mechanism is fundamental to cell size homeostasis in eukaryotes and has been investigated for decades. Several models have been proposed, including a Cln3 accumulation model and a Whi5 dilution model, but conclusive evidence has been lacking mainly because Cln3 is of low abundance and difficult to detect in cells. Using a 6xHA tagged version of Cln3 and centrifugal elutriation to isolate early G1 cells, the authors are able to detect Cln3 on western blots and show that its levels increase during G1 while Whi5 levels remains nearly constant. Importantly, Cln3 levels rise more abruptly (10-fold in rich, 2.4-fold in poor carbon medium) than cell volume (1.7-fold in rich, 1.2-fold in poor carbon medium) leading to an increase in Cln3 concentration (7-fold in rich, 2.2 in poor medium) that peaks at Start. Preventing the advertised Whi5 dilution by doubling WHI5 gene dosage did not change cell size distribution (Fig. 3). These observations clearly favour the Cln3 accumulation over the Whi5 dilution model, which confirms other reports (Litsios et al., 2019; Barber et al., 2020). Interestingly, Cln3 levels drop abruptly within minutes when cells are shifted from rich to poor carbon sources (as seen previously by Perviz et al., 1998 and Hall et al., 1998), while Whi5 levels increase slightly (Fig. 2). This creates a conundrum because cells grown in poor carbon medium initiate Start at a smaller cell size despite a stark reduction of the Cln3/Whi5 ratio. The authors also looked at what might regulate cell growth and Cln3 accumulation during G1. They find that cell growth in G1 is independent of Cdk1, Pho85 and of actin or tubulin (Fig.4A and Fig.5_Sup1), but dependent on ER-Golgi trafficking (Sec7), ceramide biosynthesis (Lac1, Lag1) and the TORC2-regulated Ypk1,2 kinases. In all cases, reduced growth in G1 is accompanied by lower Cln3 levels, indicating that sustained growth might be necessary for strong Cln3 accumulation. These data bring new players to the Cln3 accumulation model. 

The experiments are in most cases well designed and performed, and the main conclusions supported by the data. Using population-based methods and western-blotting, the authors succeed at demonstrating that Cln3 levels rise during G1, in contrast to early assumptions, but which is something that has been documented previously by the same (Zapata et al. 2014) and other labs (Thornton et al., 2013). They convincingly demonstrate that Whi5 dilution during G1, as proposed by Schmoller et al. (2015), cannot be responsible for setting the critical cell size at budding. Similar conclusions were drawn by others using single-cell imaging (Litsios et al., 2019) or ectopic WHI5 expression (Barber et al., 2020). Regrettably, most studies on the coupling of cell entry to cell growth focus on Cln3 and Whi5 levels, not activity embodied by Whi5 phosphorylation's status. The key question for how cells growing on poor carbon medium reach Start at a smaller cell size despite much reduced Cln3 levels remains unanswered by the authors, and a paper (Talarek et al., 2017) proposing down-regulation of the Whi5 phosphatase PP2A-Cdc55 by the Rim15-Igo1,2 pathway to trigger Start when Cln3 levels are low in poor medium is not cited. The identification of Sec7, Ypk1,2 and Lac1,Lag1 as regulators of growth in G1 is interesting, but the link to Cln3 accumulation is not explained by a coherent model.

Reviewer #1 (Public Review): 

Using Tet-off system, Kir2.1 was expressed (or not) during the key time of callosal development from E15 to P15. Restoring activity either by adding Dox during a critical period from P6 to P15 or using DREADDs from P10-14 could rescue the callosal projection to the cortex, whereas later restoration of activity (with Dox) was not successful. Did this successful rescue lead to normal activity? Calcium imaging in animals with Kir2.1 had low levels of any kind of activity, both highly correlated and low correlation, but P6-13 dox treatment partially restored only low-correlation activity and not high correlation activity at P13. The effects of DREADDs on activity was not similarly measured though it was effective for at least partially restoring the callosal projection. 

Overall this study builds on earlier findings regarding the importance of neuronal activity in the formation of a normal callosal projection, using in utero electroporation which is particularly well suited for this subject. It makes the case very compellingly that near-normal callosal connectivity can be produced if activity is permitted during a critical period window from P6 or P10 to P15, though the exact timing of this window is imprecise because the elimination of Kir expression was not systematically quantified. For transmembrane proteins like channels it can often take many days for protein expression to completely abate. 

I found the quantification of the callosal projection to be rather minimal and the normalization approach not entirely transparent. For example does activity from P10-15 restore the full normal PATTERN of callosal connectivity or merely the density of input overall? Also in the discussion it would be nice to more clearly establish whether activity is thought to be maintaining a projection already formed by P10 or permitting the emergence of such a pattern. 

The calcium imaging is a valuable validation of the Kir expression approach, but it the study here appears to overinterpret what may simply be an intermediate level of activity restoration rather than a specific restoration of L events, as it seems that L events would be the most likely to occur under conditions of reduced overall activity. One possibility is that the absence of H events at P13 in the calcium is due to residual Kir expression creating a drag on high level network activation rather than any more complicated change in patterned spontaneous activity/connectivity. The conclusions from this study regarding the permissive role of activity during a critical window and the lack of a requirement for highly correlated activity are valuable, even if somewhat imprecise on both counts. The authors should probably refrain from use of the term patterned activity given that this was measured but not systematically compared to unpatterned spontaneous activity.

Reviewer #2 (Public Review): 

Tezuka et al. use in vivo manipulations of spontaneous activity to identify the activity-dependent mechanisms of callosal projection development. Previous research of the authors' and other labs had shown that overexpressing the potassium channel Kir2.1, which reduces activity levels in the developing cortical network, blocks the formation of callosal connections almost entirely. 

The current manuscript corroborates and extends these previous discoveries by: 

1) Demonstrating that the effect of Kir overexpression can be rescued by pharmacogenetic network activation using DREADDs. 

2) Revealing the requirement of network activity for the development of callosal projections during a particular developmental time window and by 

3) Directly relating perturbed callosal development to the actual changes in activity patterns caused by the experimental manipulations. 

Thus, this paper is important for our understanding of the role of neuronal activity in the development of long-range connections in the brain. In addition it provides strong evidence for a role of specific activity patterns in this process. 

In general, the approach is very straightforward and the results clearly interpreted. Nevertheless, there are a few points to consider. 

1) It is not clear in which cortical area(s) the in vivo 2-photon recordings were performed and in how far cortical areas that actually receive/send callosal projections were included or not in the analysis. 

2) It is not discussed what the duration of the CNO effect is. Do daily injections rescue activity patterns for 24 hours or a significant proportion of this period?

Reviewer #3 (Public Review): 

The manuscript by Tezuka adds to an emerging story about the role of activity in the formation of callosal connections across the brain. Here, the authors show that they can use a TET system to switch off the activity of an exogenous potassium channel, in order to probe when activity might be necessary or sufficient for the formation of callosal connections. The authors find that artificial restoration of activity with DREADS is sufficient to rescue the formation of callosal connections, and that there is a critical period (somewhere between P5-P15) where activity must occur in order for the connections to form within the cortex. Finally, the authors show that when the potassium channel is removed during the critical period, the cortex exhibits activity, but few highly synchronous events. These results indicate that it is activity in general and not specifically highly synchronous activity that is necessary for the final innervation of the callosal cortex. <br>  <br> In general, the study is well done, and the writeup is polished, well summarized. The figures are solid. There are only a few criticisms/suggestions. 

Major issue: Have the authors demonstrated a requirement for "patterned spontaneous activity"? <br>  <br> The authors claim variously in the abstract ("a distinct pattern of spontaneous activity") and in the results (pg 6, "our observations indicate that patterned spontaneous activity") and discussion (pg 6, "we demonstrated that patterned spontaneous activity") that it is "patterned" spontaneous activity that is key for the formation of callosal connections. However, when I was reading the paper, I came to the opposite conclusion: that any sufficiently high spontaneous activity is sufficient for the formation of these connections. <br>  <br> The authors showed that relieving the KIR expression from P5-15 allows the connections to form; however, in Figure 4, the authors show that the nature of the activity produced in the cortex (in terms of mixtures of H and L events) is very different. Nevertheless, the connections can form. Further, the authors showed that increasing activity when KIR is expressed using DREADS restores the connections. The pattern of activity produced by this DREADS + KIR expression is likely to be very different from the pattern of activity of a typically-developing animal. In total, I thought that the authors demonstrated, quite nicely, that it is just the presence of sufficient activity that is key to the innervation of the contralateral cortex. (It's not cell autonomous, as the authors showed before; there seems to be a "sufficient activity" requirement). 

Therefore, I think the authors should remove references to the requirement of patterned activity and instead say something about sufficiently high activity (or some characterization that the authors choose). I think they've shown quite nicely that a specific pattern of the spontaneous activity is not important.

Reviewer #3 (Public Review):<br> Approaching the search of novel viruses while in an endogenized stage, rather than as free virions, this study reveals a large diversity of complete and fragmented virophages genomes - termed EMALEs-scattered throughout the genomes of four strains of Cafeteria (a marine protist) expanding the known diversity of the Lavidaviridae family from a fresh slant. Given that the activation of the integrated virophage mavirus during infection by the giant virus, CroV, has been shown to have a protective effect on the Cafeteria population, this study provides a tantalizing window into the traces of virophage-giant virus¬-protist interactions in the marine environment. Given the enormous diversity of virophages and giant viruses that have been found in metagenomes with no known hosts, this study is a step towards deciphering the biology of these viruses. Intriguingly, the authors show that endogenized virophages themselves are predisposed to being targets of NGARO transposons, pointing to another potential player in an already complex (virophage-giant virus-protist) biological system.

Strengths

The article is well-written and presented, so it is easy to follow the bioinformatic process and evaluate its technical soundness. Namely, how the search for endogenized elements was conducted, then how the EMALEs/NGAROs were classified by GC/nucleotide identity/whole genome synteny and phylogenetically.<br> The authors mapped each complete genomic element its genetic content and genomic context in great detail. This allows clear comparisons to be made between the EMALE types.<br> All genomic sequences and analyses are available, and the depth of the analyses available in supplementary is encyclopedic in scope, making this work highly transparent and also readily exploitable by others in the field.<br> The conclusions are well-supported by the data and the authors keep their discussion close to their findings.

Weaknesses

There are no identifiable weaknesses in the technical analysis. There were a few points in the descriptions where a few points of clarification would be welcome.

Reviewer #1 (Public Review):

This study is well-written and well-presented. The conclusions are clear and robustly supported by the data. The figures provide useful visualizations for the major findings. Virophage are an important and underappreciated component of global viral diversity, and they likely play important roles in eukaryotic genome evolution; this work is therefore quite timely. Relatively few studies focus on virophage or giant viruses compared to other viral lineages, so studies like this are highly valuable.

Strengths of this work include the high quality of the reference genomes, which were constructed using both short-read and long-read sequencing, as well as the diverse locations and isolation times of the host genomes.

I found no major weaknesses in this study. One minor issue is that the details of how EMALEs were delineated and initially detected seem a bit unclear to me. Based on my reading I am curious if some divergent or degraded EMALEs could have been missed. This may be important for assessing the consequences of possible retrotransposition-mediated EMALE inactivation.

Reviewer #2 (Public Review):

The manuscript reports the in-depth analysis of 4 Cafeteria Burkhardea strains genome sequences revealing several integration events of mavirus-related virophages (EMALE) which can either correspond to ancestral integration events (present in several strains at the same genomic location) or unique to a given strain as a result of recent integration. Given the protective effect that Mavirus virophage have on the host against CroV infection, they are likely functional mobile genetic elements that can reactivate upon giant virus infections thus providing adaptative anti giant viruses defense to the host. Some integrated virophage genomes are incomplete and thus not functional anymore.

The EMALEs can be clustered into 8 different groups each endowed with different sets of gene promoters and are thus proposed to be virophages infecting different giant viruses, all infecting Cafeteria Burkhardea. Half of them are AT-rich and the others present medium AT-richness. The isolated mavirus prototype falls into type 4 of AT-rich EMALE. Type 2 could be the result of recombination between EMALE 1 and a polinton-like virus. The truncated type 8 lacks morphogenesis genes.

Interestingly, the study also reveals nested parasitism with integration of tyrosine recombinase retrotransposons of the Ngaro super family that can occur both in the host genome and in the virophage, with an example of a virophage integrating inside a Ngaro retrotransposon. There are four types of Ngaro retrotransposons based on their nucleotide sequences that share the same coding potential but differed in terms of integration preferences. When integrating into the host genome, the ORF1 encoding a GAG-like protein appears to be missing and this ORF could be determinant for integration site specificity.

Reviewer #1 (Public Review):

Mulholland et al show that there is a very close relationship between the development of excitatory and inhibitory networks in the developing cortex. This paper makes an important contribution to our understanding of the structure of inhibition during an early stage in cortical development. The work has been carefully performed and analysed. The changes I suggest are principally to improve clarity in some places.

Lines 48-55: express these possibilities more didactically as individual items in a list, rather than grouping the first two together.

Fig 2: Show a raw image of the imaging window, with a scale bar.

Line 81: clarify the type of imaging used (both wide-field and 2-photon are mentioned in the Introduction).

Line 89: This is presumably in a different set of animals: make this clearer. n value?

Fig 3b: "Example mean trace..."

Fig 3e: Is this for one animal or averaged over all? Why not show negative correlations as well?

Fig 3f: This appears to be the same image as S1c?

Line 119-120: Suggest you explain more clearly that this is a preliminary step towards the later simultaneous E-I imaging, and why it's still useful given that it's somewhat indirect compared to the later simultaneous imaging.

Line 124: Why tenth of maximum? Can you confirm that the main results are not highly sensitive to this choice?

Fig 4: It would be clearer to give the graph parts of a and d their own panel labels. Is it worth explicitly flagging that none of the E-I distributions are significantly different? It would be useful to explicitly define pink and blue in the caption. Does panel e have units? In panel a I'm confused about what is being referred to as "circles" and what as "lines". Panel f caption: "principal components"

Line 143: suggest "similarity in the statistical structure", to make it clearer what you mean by "structure".

Line 159: state the n value here.

Fig 5: It would be interesting to speculate about the significance of the correlation fractures. In the rhs of panel b the red seed points are almost impossible to see, since they occur on top of a red patch (perhaps include a black or white border to these circles?). a caption: "-2-2 z-score" is confusing; perhaps say "-2 to +2 z-score". g caption: Is this correlation similarity for E, I or both?

Reviewer #2 (Public Review):

The paper by Mullholland and colleagues investigates the spatial structure of spontaneous events in inhibitory neurons in ferret primary visual cortex before eye opening. This work is a logical and important extension of previous studies by the senior author which investigated the same issue from the perspective of excitatory neurons. Two major findings result from this work: First, spontaneous events are tightly coupled in inhibitory and excitatory networks. Neurons of both kinds participate in the same event, and the strength of inhibitory and excitatory activation in each event is well matched, presumably to maintain a somewhat constant overall activation level. Second, the spontaneous activity reveals spatial structure in excitatory and inhibitory networks already at this relatively early developmental stage, with very similar spatial scales for both kinds of networks. Both findings are well supported by careful experimental work and matching quantitative analyses. At the same time, the conclusion that spatial scales for inhibitory and excitatory networks match could be further strengthened. A particular concern in this respect are the potentially quite different firing rates of inhibitory and excitatory neurons. In combination with the non-linearities of calcium imaging, they introduce potential confounds when comparing measurements between inhibitory and excitatory networks. The two calcium sensors used for some of the work similarly might introduce systematic differences between the 2 types of cells. Finally, the paper could be strengthened by a more thorough analysis confirming that differences in calcium indicator expression levels are not limiting the detectable spatial scale of spontaneous events.

Reviewer #3 (Public Review):

The manuscript provides key information about the state of cortical networks at around the time that feed-forward input can begin to provide external drive. The data show convincingly that early activity in inhibitory networks exhibits a modular structure that overlaps with the same modular activity of immediately neighboring excitatory neurons. This early architecture contrasts strongly with the non-specific pooling that has been shown in rodent visual cortex. The paper also contains a small related gem: evidence that GABAergic signaling has a net inhibitory effect on cortical activity at P21 in the ferret, showing that the transition from GABAergic signaling from depolarizing to hyperpolarizing has already occurred during this period. The paper borrows analysis techniques from a prior publication that examined excitatory activity (Smith et al. 2018), so the methods and techniques are already proven. The paper is highly polished.

Reviewer #1 (Public Review): 

Although the results are impressive, the paper has several weaknesses, particularly in terms of analysis, as listed below. 

As far as I can tell, the input to the model are raw diffusion data plus a couple of maps extracted from T2 and MT data. While this is ok for the kind of models used here, it means that the networks trained will not generalise to other diffusion protocols (e.g with different bvecs). This greatly reduces to usefulness of this model and hinders transfer to e.g. human data. Why not use summary measures from the data as an input. There are a number of rotationally invariant summary measures that one can extract. I suspect that the first layers of the network may be performing operations such as averaging that are akin to calculating summary measures, so the authors should consider doing that prior to feeding the network. 

The noise sensitivity analysis is misleading. The authors add noise to each channel and examine the output, they do this to find which input is important. They find that T2/MT are more important for the prediction of the AF data, But majority of the channels are diffusion data, where there is a lot of redundant information across channels. So it is not surprising that these channels are more robust to noise. In general, the authors make the point that they not only predict histology but can also interpret their model, but I am not sure what to make of either the t-SNE plots or the rose plots. I am not sure that these plots are helping with understanding the model and the contribution of the different modalities to the predictions. 

Is deep learning really required here? The authors are using a super deep network, mostly doing combinations of modalities. is the mapping really highly nonlinear? How does it compare with a linear or close to linear mapping (e.e. regression of output onto input and quadratic combinations of input)? How many neurons are actually doing any work and how many are silent (this can happen a lot with ReLU nonlinearities)? In general, not much is done to convince the reader that such a complex model is needed and whether a much simpler regression approach can do the job. 

Relatedly, the comparison between the MRH approach and some standard measures such as FA, MD, and MTR is unfair. Their network is trained to match the histology data, but the standard measures are not. How does the MRH approach compare to e.g. simply combining FA/MD/MTR to map to histology? This to me would be a more relevant comparison. 

Some details of the modelling is missing (apologies if it's there but I missed it): <br> - Not clear if there are 64 layers or 64 residual blocks. Also, is the convolution only doing something across channels? i.e. do we get the same performance by simply averaging the 3x3 voxels? <br> - The result in the shii]verer mouse is most impressive. Were the shiverer mice data included in the traning? If not, this should be mentioned/highlighted as it is very cool.

Reviewer #2 (Public Review): 

The overall idea behind this paper is exciting - use multimodal MRI, co-registered to histology, to train a predictor of a histology image given the MRI input. They demonstrate the potential of this technique using a series of datasets from the Allen Brain Institute, showing decent face validity and further being able to probe which MRI contrasts most contribute to which cellular predictions. The resulting maps of predicted myelin basic protein and predicted Nissl staining appear convincing, lending credibility to the overall effort. Furthermore, the MRI acquisitions and image processing and deep learning steps all seem very well executed. 

The most important limitation in this study is, to me, that the MRI and histology samples come from different animals. It is not immediately clear that the prediction obtained by comparing different brain areas, which is essential what this prediction will produce, will be accurate in the case of different types of subtle or coarse pathology. As such, this manucript would need ideally a full training set of MRI and histology from the same mice, again ideally consisting of both healthy brains as well as models with pathology. In the absence of such a rich training set at least a strong validation set with histology and MRI in the same animals is needed in order to prove the validity of the proposed method. This does exist, as shown for example for the shiverer mice in Fig 3. Yet the comparison is purely qualitative, with no attempt made to directly quantitate how accurate the predicted histology maps are compared to the real thing, which is a missed opportunity. 

There is definitely value to the proposed approach, in particular around understanding which MRI modalities are most likely to contribute to certain cellular staining techniques. The trained networks for different cellular contrasts look intriguing. Yet without samples from the same animal directly and quantitatively compared the results cannot be clearly evaluated.

Reviewer #1 (Public Review): 

In this work, the authors present a model for double nerve transfers in the forelimb of the rat. The authors provide a detailed description of how the model is developed and they characterize neuromuscular regeneration through nerve crush, neurotomy, behavioral analysis, and retrograde labeling. The peripheral innervation of muscle with a double nerve transfer is compared to that with a single nerve transfer. 

Major strengths: 

- Strong motivation for necessity of this model given <br> - Experimental design and surgical techniques are clearly described. The authors include methodologies, materials used, figures, and supplementary videos to support the discussion of how the experimental model is developed. <br> - Large number of animals are used for both the double nerve transfer and single nerve transfer, and results appear to be consistent within these populations. 

Weaknesses: 

- The work assumes specialized knowledge of peripheral nerve anatomy and some surgical techniques. The article may be less accessible to someone without a background in these areas who seeks to learn more about nerve transfer models. 

The authors do a rigorous job of describing the techniques used to develop the double nerve transfer model. The experimental design and surgical methods provide detailed accounts of how the model is realized, including descriptions of the techniques as well as highlighting materials that are necessary for the procedures. This is particularly valuable for a reader who desires to replicate this model. The efficacy of the nerve transfer is examined in multiple ways and compared to a single nerve transfer model. These results, which are statistically verified, demonstrate that the double nerve transfer more effectively reinnervates muscles and, in some measures, that there is no difference between animals with double nerve transfers and healthy comparisons. This provides confidence and excitement for how this model may be used in the future for studies involving therapeutics.

Reviewer #2 (Public Review): 

The paper is well-written with sufficient sample sizes. The figures are generally clear and easy to understand. 

The clinical utility of multiple nerve transfers should be better delineated in the introduction. Current limitations, difficulties, and strategies to mitigate such limitations should also be overviewed to provide better context to the reader as to the importance of this work. 

What were the relative proportions of innervation between the two nerves in the dual-innervation model based on the retrograde labeling? 

Section 3.4.1: Please present the raw data and mentioned scatterplot? What was the correlation coefficient for the linear regression? 

Section 3.2. The distribution of scores from each group should be presented in a graphical or tabular format. What was the course of recovery? Were the evaluations performed anytime before 12 weeks? 

Section 3.3: SNT vs DNT should be evaluated in a comparative fashion. 

Section 3.4 How is 'adequate' muscle fibrillation determined? 

Was any electrophysiology performed to assess the quality of reinnervation and nerve conduction velocity? Compound muscle action potentials or motor unit counts would help identify the proportion of the muscle that was reinnervated. 

Can the authors please comment on the way in which the DNT was adopted by the animals as opposed to the SNT? Was there any noticeable difference in the functional recovery of the muscle or retraining process? The neuroplastic adaptation would be an interesting characterization. 

In the discussion, the authors suggest that "hindlimb models do not adequately represent the physiology of upper extremity nerve transfers and targeted muscle reinnervation procedures." based on outcomes for lower vs upper transfer. A number of additional factors, including usage of the limbs, weight bearing, sensorimotor circuitry etc. play a role and should be accounted for. 

In the DNT model, was the distance between entry points of the coaptation held constant between animals or optimized? The increased muscle mass observed in the DNT group is likely a result of a better axon : myocyte ratio and spatial distribution. This could be studied and optimized to improve the outcomes and utility of DNT. 

The discussion should more thoroughly explore the limitations of this model and experimental constraints. 

What remains to be optimized prior to clinical translation? What types of scientific questions can this help answer? In which types of clinical cases would DNT not be appropriate?

Reviewer #3 (Public Review): 

The authors aimed to establish a rodent upper limb model to test double vs. single nerve transfers, and provided base results for histological retrograde labeling, topographic findings, functional (behavioural) analysis and outcomes for reinnervated vs. control muscle mass. The manuscript is well balanced, and contains a detailed description to reproduce the experimental model. 

The authors demonstrated equal functional outcomes for both types of transfers and have in this reviewers opinion succeeded in establishing a novel model for future (experimental) tests before (clinical) application.

Reviewer #1 (Public Review):

Watanabe presents a set of EEG-TMS experiments to show that brain stimulation in specific frontal regions affect both perception and brain states during bistable perception. The patterns of results appear interesting and potentially significant. The work uses relatively idiosyncratic methodologies in terms of data analysis and modelling, which makes the work harder to relate to extant literature. This situation requires authors to "go the extra mile" in explaining their approach and ensuring that readers can easily understand the findings in the light of what they're likely to already know - and here, I find steps could be taken.

Specific comments;

- The author has an idiosyncratic definition of DLPFC; especially pDLPFC seems to coincide with iPCS retinotopic regions as found by Mackey et al, and with rIFJ from earlier work such as that by Sterzer and Kleinschmidt. When you look up DLPFC on wikipedia, this shows a region more 'superior' than both regions designated DLPFC in the present work, closer to aDLPFC than pDLPFC. Perhaps the author could try to more explicitly connect his nomenclature to the literature?

- The possible relation of the individual brain state dynamics with the ongoing sequence of bistable perception apart from the TMS manipulation is not treated. This may feel self-evident to the author perhaps because this is the topic of previous studies, but it's confusing to a novice reader. To me, linking the sequence of bistable perceptual states to the sequence of brain states as found using the author's methodology is a fundamental step to allow interpretation of all of the subsequent results, because it speaks to the meaning and significance of the existence of these brain states. Without this step, I find it difficult to interpret figures 2a and 2b (which, I have to say, do indeed look like enticing patterns in the data). So, specifically, does the author replicate the brain-state vs behavior correlations that he reported in his earlier (2014) work on this topic? And, because this publication reported mainly across-observer correlations, what about relations between brain states and their transitions and perceptual events on a within-subject basis?

- The exact analysis procedure that leads up to the brain state designation is not very transparent. What, for example, is the brain state that is "Frontal"? I would appreciate to see state-transition triggered time-frequency plots to be able to understand what exactly in the EEG the procedure picks up. The same holds for the TMS-triggered changes; is there any pattern in terms of TMS-induced time-frequency changes?

- It would be a valuable addition if the author could clarify what he means with a brain state; this term means different things in different fields. The concept of brain state is now primarily based on high frequency EEG signatures, but there are likely many other possible measurements that could produce estimated brain state. How would the findings change if other measures were used as a basis for the same methods?

- Figure 1l. From methods and explanations it's not really clear how this figure is produced. If it is created from single-subject surface locations that were explicitly targeted, and these locations are then transformed into an average-subject surface, that would be correct. But weren't these locations targeted based on MNI coordinates? In this case one would expect more of a spread in specific locations because of the across-subject variability in surface folding. So, could the author please explain in more detail how this figure is generated?

- Page 8, I appreciate the logic that "barrier heights are associated with the dwelling time in the brain states and inversely correlated with the transition frequency between them", but this needs to be fleshed out more. What are the numerical simulations here? These aren't described in the text and as a reader, I'm left having to believe the accuracy of the 'numerical simulations' without being given the opportunity to understand them. This explanation would be a nice opportunity to go into detail about how the author does (and, consequently, the audience should) understand and interpret both the brain states, and their transitions.

Reviewer #2 (Public Review):

The author tested the hypothesis that the causal influence of the PFC on bistable perception is dynamic and depends on the (fluctuating) state of the cortical networks. Using offline and online EEG measurements and a sophisticated analysis procedure, the author characterized their dynamic brain states when observers perceived a bistable rotating sphere defined by Structure-from-Motion with alternating perceived direction of rotation. TMS applied to different regions of the frontal, parietal, and visual areas had different effects on observers perceptual dynamics, depending on the dynamic state of the cortical networks. It's quite impressive to see the large effect size from TMS to the aDLPFC, and the opposite direction of the effects observed from aDLPFC and pDLPFC/FEF stimulation makes it more convincing that the PFC has specific and robust roles in bistable perception.

Although the effect on bistable perception from state-dependent TMS of DLPFC is robust and very interesting, the functional mechanism of how different regions of DLPFC contribute to the perceptual dynamics remains unclear. I find it surprising that the author did not address the potential role of attention in mediating DLPFC's contribution to observers' perceptual dynamics. Given that attention does play a role in the dynamics of many forms of bistable perception, it is important to distinguish between an intrinsic contribution of DLPFC to bistable perception vs. an effect mediated by changes in attentional state. It is also useful for the author to discuss how and why certain brain states are linked to certain perceptual states.

There are many forms of bistable perception, and their dynamics are controlled or influenced by shared as well as independent mechanisms (e.g., Cao T et al, Frontiers in Psychology 2018). It would be useful to discuss the generalizability and limitations of the current results in relation to different types of bistable stimuli. The methodological approach developed by the author will be quite useful in researching the neural mechanisms of other types of bistable perception.

Reviewer #3 (Public Review):

This is an ambitious study by a competent single author who has previously published highly innovative work on this topic. The study incorporates real-time closed-loop EEG-TMS and computational modeling to causally test the role of PFC in perceptual switching of bistable perception triggered by ambiguous visual input. While the work is technically impressive and involves a substantial amount of work spread over multiple experiments (especially notable in the context of a single-author manuscript), I have some major concerns as described below.

1) The author's previous work on energy landscape in the context of bistable perception was conducted using fMRI. This current study employs EEG, and records time series from 7 ROIs (Fig. 1a-b). Some of these ROIs are very close together, less than a few centimeters (e.g., a-p SPL; a-p DLPFC; LOC-V5). The conventional thinking is that scalp EEG does not have the spatial resolution to separate signals from such closely spaced areas. While the author employs a Laplacian montage, validation data suggesting that the resulting signal had high SNR and could differentiate between neighboring regions is missing.

2) The EEG analysis rests on gamma band (30-80 Hz) power. This should be explained in the main text. It is technically risky to record gamma band activity using scalp EEG, due to muscle, eye, and, most concerningly, microsaccade-related artifacts (see work by Yuval-Greenberg). Since the task employs a structure-from-motion stimulus, the effect of microsaccades is especially worrying. No control data was presented to suggest that these artifacts do not contribute to the analyses.

3) P. 10 There is a concern here that the hypothesis testing is circular. The models were fit by using EEG data (and behavior?) to calculate the energy landscape, so is it trivially expected then that the dwell times seen behaviorally correlate with the energy barrier estimated by the model?

4) It's not clear to me why pDLPFC's result was interpreted as "functional diversity".

5) The pDLPFC region here would be more accurately referred to as inferior frontal gyrus (IFG) or ventral frontal cortex (VFC), or inferior frontal cortex (IFC). It is not part of the classic DLPFC.

Reviewer #1 (Public Review):

The authors initially showed that M. brevicollis were killed by Pseudomonas aeruginosa and avoided ingesting them, as opposed to other bacteria, like E. coli. The next sought to ask why this was occurring and found that M. brevicollis STING was upregulated in the presence of Pseudomonas aeruginosa but not bacteria that do not kill M. brevicollis. While previous studies have shown that STING orthologs exist in choanoflagellates, the authors next sought to determine how M. brevicollis STING functions, especially with regard to the presence of pathogenic bacteria. The authors achieved this goal through the generation of STING-null M. brevicollis and showed that these mutants were less susceptible to Pseudomonas-induced killing. They next found that M. brevicollis STING induces autophagy, which the authors conclude is the mechanism for Pseudomonas-induced, STING-mediated killing of M. brevicollis.

The strengths of this manuscript are that the authors convincingly show evolutionary conservation of STING function in a choanoflagellate, the closest relative to animals. Furthermore, through their research they developed novel methods for genetic editing in M. brevicollis that is useful to the scientific community who use similar organisms, or for the development of these techniques in other organisms where they have yet to be used. A weakness of the study is that some gaps remain with the full story. From the start, why does Pseudomonas aeruginosa, but not other bacteria, induce STING and the generation of cyclic dinucleotides? What is the source of these STING-inducing cyclic dinucleotides? And in the end, is the STING-mediated induction of autophagy (Atg8 lipidation) the cause of M. brevicollis death? Overall, this is a nice study by Woznica et al. that is of interest to the broad scientific community.

Reviewer #2 (Public Review):

In this novel and interesting manuscript by Woznica, et al. the authors show that STING signaling, known in mammals as a key regulator of the interferon response, is found in the evolutionarily closest precursor to mammals, choanoflagellates. The authors show that there appears to be a choanoflagellate cell death pathway that is a specific response to Pseudomonas aeruginosa, although it is not clear how this is related to pathogen protection. The authors clearly and convincingly show that the STING-mediated response is dependent on the endogenously produced 2'3' cGAMP, presumably synthesized by the cGAS ortholog encoded by M. brevicollis. In contrast, microbial cyclic dinucleotides were not inducers of this system. By developing a system for genetic manipulation of M. brevicollis, they were able to evaluate the role of mutations, perform complementation analysis, and localize fluorescent proteins after transfections. Most convincingly, they showed a connection between the evolutionarily ancient autophagy pathway, STING and 2'3' cGAMP, showing a strong connection between this system and mammalian cell autophagy. Whether the autophagic response is the key link to pathogen protection, or host cell death to bacterial products is the critical link to interfering with spread and host disease is not clear. Also of unknown relevance is the observation that there is phagocytosis resistance toward P. aeruginosa, as it was not shown to be directly linked to the STING regulation, and it wasn't directly demonstrated that phagocytosis avoidance was a choanoflagellate activity or a bacterial activity. The fact that there also appears to be an LPS-responsive pathway leading to death indicates that the organism may have multiple pattern recognition receptors yet to be identified.

Reviewer #3 (Public Review):

In general, this is a well-performed study. The experiments are well-controlled and interpreted. There were a number of strengths, including (i) the development of a Choanaoflagellate model system to study innate immunity, (ii) the development of genetics for M. brevicollis, and (iii) the demonstration of a functional STING system in one of the closest relatives to animals.

The weaknesses are that as of now (i) these are really two unrelated stories that are each incomplete. It is unknown what P. aeruginosa factor is affecting M. brevicollis growth and whether this factor requires STING. Likewise, it is unknown what cGAS/STING is responding to in M. brevicollis. Is it a virus? (ii) M. brevecollis is 2 of >20 Choaoflagellates with a predicted intact cGAS-STING conservation. As mentioned by the authors, there is evidence of horizontal transfer of cGAS/STING, which may have occurred after the evolutionary fork to animals. Is it true that this represents an evolutionary precursor of animal innate immunity?

Reviewer #1 (Public Review):

Review:

The manuscript "Fast and accurate annotation of acoustic signals with deep neural networks" by Elsa Steinfath, Adrian Palacios, Julian Rottschäfer, Deniz Yuezak, and Jan Clemens describes a new piece of software that, building on previous work, trains a deep network classifier to segment audio signals. The main advances are speed (real time), works on standard hardware, and a user-friendly interface, which allows users with little machine learning experience to train and use a deep neural network classifier.

I. Results

A. How good is it?

1. How fast is it?<br> a. How long to train?<br> Train time depends on the amount of data, but the ranges quotes (10 minutes to 5 hours) are quite reasonable. It works on reasonable hardware (I tested on a laptop with a GPU).

b. How long to classify?<br> Latency to classification is between 7-15ms, which is a little long for triggered optogenetics, but not bad, and certainly reasonable for acoustic feedback.

2. How accurate is it?<br> a. In absolute terms<br> Accuracy is improved relative to Fly Song Segmenter, particularly in recall (Arthur et al., 2013; Coen et al., 2014).<br> Pulse song:<br> DeepSS precision: 97%, recall: 96%<br> FlySongSegmenter: precision: 99%, recall 87%.<br> Sine song:<br> DeepSS precision: 92%, recall: 98%<br> FlySongSegmenter: precision: 91%, recall: 91%.

b. One main concern I have is that all the signals described, with the exception of pulse song, are relatively simple tonally. Bengalese finch song is much less noisy than zebra finch song. Mouse vocalizations are quite tonal. How would this method work on acoustic signals with noise components, like zebra finches or some non-human primate signals? Some signals can have variable spectrotemporal structure based on the distortion due to increased intensity of the signal (see, for example, Fitch, Neubauer, & Hertzel, 2002).

W.Tecumseh Fitch, Jürgen Neubauer & Hanspeter Herzel (2002) "Calls out of chaos: the adaptive significance of nonlinear phenomena in mammalian vocal production" Animal Behaviour, 63: 407-418. doi:10.1006/anbe.2001.1912

B. How easy to use?

0. "our method can be optimized for new species without requiring expert knowledge and with little manual annotation work." There isn't a lot of explanation, either in the paper or in the associated documentation, of how to select network parameters for a new vocalization type. However, it does appear that small amounts of annotation are sufficient to train a reasonable classifier.

1. How much pre-processing of signals is necessary?<br> All the claims of the paper are based on pre-processed audio data, although they state, in the Methods section that preprocessing is not necessary. It's not clear how important this pre-processing is for achieving the kinds of accuracy observed. Certainly I would expect the speed to drop if high frequency signals like mouse vocalizations aren't downsampled. However, I tried it on raw, un-preprocessed mouse vocalizations, without downsampling and using very few training examples, and it worked quite well, only missing low signal-to-noise vocalizations.

C. How different from other things out there?

It would strengthen the paper to include some numbers on other mouse and birdsong methods, rather than simple and vague assertions "These performance values compare favorably to that of methods specialized to annotate USVs (Coffey et al., 2019; Tachibana et al., 2020; Van Segbroeck et al., 2017)." "Thus, DeepSS performs as well as or better than specialized deep learning-based methods for annotating bird song (Cohen et al., 2020; Koumura and Okanoya, 2016).

D. Miscellaneous comments

1. Interestingly, the song types don't appear to be mutually exclusive. One can have pulse song in the middle of sine song. That might be useful to be able to toggle...I can imagine cases where it would be nice to be able to label things that overlap, but in general if something is sine song, it can't be pulse song. And my assumption certainly was that song types would be mutually exclusive. Adding some explanation of that to the text/user's manual would be useful.

2. How information is combined across channels is alluded to several times but not described well in the body of the manuscript, though it is mentioned in the methods in vague terms:<br> "several channel convolutions, 𝑘𝛾(1, 𝛾), combine information across channels."

II. Usability

A. Getting it installed<br> Installing on Windows 10, was a bit involved if you were not already using python: Anaconda, python, tensorflow, CUDA libraries, create an account to download cuDNN, and update NVIDIA drivers.

Reviewer #2 (Public Review):

Steinfath et al. developed a new toolbox to segment animal communication within acoustic signals. Specifically, they use temporal convolutional neural networks with dilated convolutions -- this is an excellent algorithmic choice: not least since WaveNets have demonstrated the power of those building blocks for acoustic signal processing in 2016, and they are now running in every modern phone.

Steinfath et al. evaluated the performance of their toolbox for courtship songs in flies, ultrasonic vocalizations in mice and songs of Bengalese birds. While the performance is generally convincing, the specific examples do not seem to be (very) challenging pattern recognition problems and comparisons to other tools are lacking. In particular, for the examples regarding mice and birds, no other methods are considered (or quantitatively discussed), not even simple baselines such as SVMs applied to spectrograms. Another concern is that currently the toolbox only works for single animals, but often (especially) for courtship songs at least a couple is involved.

Overall, this is an appealing tool with good performance. The toolbox seems well written, boasts a GUI for annotation and proof-reading and a clean documentation. Thus, it might be broadly used within the auditory community.

Reviewer #3 (Public Review):

The authors have developed a software package called Deep Song Segmenter (DeepSS) that uses the deep learning framework known as temporal convolutional networks to segment and annotate communication signals. The package is intended to be a general and flexible framework for annotating communication signals in many experimental settings. The paper reports performance on recordings from drosophila, mice and songbirds. The classification and detection performance across these diverse data sets is quite good, generally in the mid to high 90 percent. This suggests that the framework could be useful for a wide range of researchers studying communication signals.

Strengths:

In addition to overall classification, the authors do a good job of answering the most important additional questions about their package's performance. Namely, they show that: (i) classification/detection does not degrade dramatically at lower signal to noise ratios; (ii) segmentation is temporally precise, generally in the submillisecond range for temporally well-defined events; (iii) classification is fast after training and can be run on a standard PC, facilitating use in real time; and (iv) good classification performance can be achieved with a relatively modest number of hand annotations (generally in the 100s).

Weaknesses:

There are two main weaknesses of this paper. First, although the authors claim that the method compares favorably to other machine learning methods, they only provide head-to-head performance comparisons for drosophila songs. There are no direct comparisons against other methods for mouse ultrasonic or songbird vocalizations, nor is there a readable summary of performance numbers from the literature. This makes it difficult for the reader to assess the authors' claim that DeepSS is a significant advance over the state of the art.

Second, the authors provide little discussion about optimizing network parameters for a given data set. If the software is to be useful for the non-expert, a broader discussion of considerations for setting parameters would be useful. How should one choose the stack number, or the size or number of kernels? Moreover, early in the paper DeepSS is touted as a method that learns directly from the raw audio data, in contrast to methods that rely on Fourier or Wavelet transforms. Yet in the methods it is revealed that a short-term Fourier front-end was used for both the mouse and songbird data.

Reviewer #1 (Public Review): 

Between-group competition is thought to be an important selective force shaping animal social behavior and social structure. However, because levels of between-group competition often covary with other aspects of the environment (e.g., food/resource density), it is difficult to evaluate how conflict between groups affects fitness outcomes per se, especially over longer time scales. To address this question, Braga Goncalves and Radford experimentally manipulate intergroup conflict in daffodil cichlids to compare fitness-related outcomes between groups chronically exposed to intruders versus those that were not. They find evidence that repeated exposure to intergroup conflict affects rates of clutch production, parental care behavior, egg size and composition, and, crucially, the overall rate of surviving offspring. 

The manuscript is clear and compelling, with a simple experimental design that is exploited to investigate a range of outcomes. The results indicate that repeated exposure to intergroup conflict has multimodal effects on fitness, not only for breeding adults, but also for the next generation (offspring of fish regularly exposed to intruders are smaller, slower to respond to a sudden stimulus, and less likely to survive their first month of life). Interestingly, these consequences are observable even in the absence of differences in resource availability, and without physical contact between intruders and resident fish. These findings highlight the importance of social stress caused by intergroup conflict, and suggest that species that encounter frequent conflict have experienced strong selection pressures to sense and respond to intrusion. Although the generalizability of these findings remains to be tested, they provide convincing empirical evidence in favor of the importance of between-group competition in social evolution.

Reviewer #2 (Public Review): 

Overall, this study has the potential to be illuminating to research in both intergroup conflict and social evolution. By staging repeated conflicts in a controlled environment, the researchers (to their knowledge, for the first time) measured a greater array of relevant responses to chronic intergroup conflict. They also discuss how they reveal indirect effects on offspring-i.e., not just immediate death or injury as a result of conflict. However, there may be some significant statistical issues to be addressed before the results are fully accepted. 

Pros: The authors made clear how this study was relevant, important, and timely. Intergroup conflict is a fast-growing field of research because we are beginning to understand that it might affect fitness, and therefore evolution, more than previously appreciated. By directly, and experimentally, studying impacts on fitness-not just in the short term (e.g., egg production) but in the longer term (e.g., young survival to one month)-the authors suggest important negative effects of intergroup conflict on fitness. It is also useful to note that these types of fitness effects have generally only been studied in mammals. Showing similar results in fish lends broader relevance to the importance of intergroup conflict to social evolution. 

Areas for Improvement: While I believe the overall experiment seemed quite strong, the statistical approach does not seem in-line with current recommendations. Most importantly, the authors appear to have used stepwise model reduction-including, importantly, removing non-significant fixed effects-to test the significance of predictors. Several simulation studies have shown that this increases the likelihood of false positive results (e.g., Mundry & Nunn 2009 Am. Nat., Forstmeier & Schielzeth 2001 Behav. Ecol. Sociobiol.). The above concern related to the fact that the authors reported some "trending" results (e.g., 0.05 < P < 0.10) as being relevant, but not others. A revised statistical approach may clear up confusion with this. I give recommendations to the authors regarding these issues in the "recommendations for authors" section. 

The authors also appeared to "Gaussianise" response variable distributions. As a reader, I could not understand whether this took the place of fitting models using an appropriate error distributions for the response variable (e.g., Poisson), or whether this was a necessary step just for the distributions of the model residuals. It is important to give more details on this, to know if did or did not affect the conclusions of the study. 

Finally, but perhaps most importantly, the lack of a clear set of hypotheses relevant to the specific variables measured here made it hard to understand the Results section. In the Results, the relevance of a given predictor was only described after the statistical significance of that predictor was revealed. This gives the appearance that the authors measured a wide range of factors and only describe the relevance of the ones revealed as significant.

Reviewer #3 (Public Review): 

In this paper, Goncalves and Radford report the results of two treatment/control experiments in which they simulated territorial incursions between groups of daffodil cichlids in the lab. In the treatment condition, fish from a neighbouring tank were temporarily placed behind a transparent panel in the tank of the focal group. The results of the experiments are quantified in great detail: the authors describe findings relating to the effect of the experimental treatment on (to name but a few measures): egg count, interclutch interval, offspring size and survival to one month, egg protein, and parental care. With so many results (I counted 43 p-values in the Results section) and several LMMs, there is a lot to consider the results. But the key finding is that the absolute number of offspring surviving to one month decreased over time in the treatment group. In my view, this is an important finding that will provide an experimental basis for solidifying the commonly made assumption that the well-established short-term costs of intergroup conflict is wild vertebrates will have fitness consequences in the medium/long term. 

In my opinion the striking thing this about these results is that the intrusions were not physical encounters: it is not that the incurred groups were physically attacked or that eggs were eaten. Similarly, it wasn't that the 'invading' individuals were directly competing for food with the target group (or if they were eating some residual food then it would be no more than in the control condition). Although the strength of the paper is in the neat experimental design, this is unfortunately obscured due to insufficient explanation of the experiment prior to the results. But this shortcoming is superficial - the authors could move up some more details from the methods. Apart from some minor changes suggested to the authors in another section I have no major concerns about this paper. I expect that these results will be of interest to a broad audience in behavioural ecology - they make an important contribution to our understanding of intergroup conflict, a phenomenon of clear importance to the evolution of animal behaviour which has been relatively neglected in the literature compared to studies of intragroup cooperation.

Reviewer #1 (Public Review): 

Wang and colleagues investigate epistatic interaction between a group of seven positions in the influenza H3N2 neuraminidase protein by constructing all possible combinations of amino acids that were observed at these seven sites in the past 50 years. This exhaustive characterization in done in 6 different genetic backgrounds spanning the documented evolution of the virus since 1968. The study is impressive in its exhaustiveness and the systematic approach. The paper is well written and clear. The main findings of the paper are:

• Epistatic interactions between this set of seven residues in proximity do not depend much of the genetic background.

• Later genetic backgrounds are much more permissive to changes in these 7 positions.

• Changes in local charge result in reduced fitness.

Measuring all possible combinations of the residues in 6 different backgrounds is a very powerful and elegant way to explore how the evolutionary constraints on the protein changed over time and this paper stands out in presenting clean and well controlled data that supports the main claims.

Reviewer #2 (Public Review): 

This is a nice study. The authors use experimental and statistical methods to evaluate the fitness landscape of mutations within a region of the neuraminidase (NA) segment of the A/H3N2 influenza strain. Results from this segment contrast from those obtained from haemagglutinin, and contribute towards a long-term aim in research of predicting the evolution of seasonal influenza strains. This is a question with broad impact as a scientific matter, and with immediate application, touching on the need to regularly update the strains used in influenza vaccination. 

The collection of mutagenesis data via deep mutational scanning creates a useful dataset. Using these data the authors successfully demonstrate that epistatic fitness effects in this region of the NA protein are largely conserved across a range of historical genetic backgrounds.

Reviewer #3 (Public Review): 

The paper contains a substantial amount of novel experimental work, the experiments appear well done, and the analysis of the data makes sense. Raw data and analysis scripts have been made fully available. 

I have two specific comments: 

- While the paper talks extensively about deep mutational scanning, I don't think this is a deep mutational scanning study. In deep mutational scanning, we usually make every possible single-point mutation in a protein. This is not what was done here, as far as I can tell. 

- For the analysis of epistasis vs distance (Fig 4d, e, f), it would be better to look at side-chain distances rather than C_alpha distances. In covariation analyses, it can be seen that C_alpha distances are not a good predictor of pairwise interactions. Similar patterns may be observable here. 

See e.g.: A. J. Hockenberry, C. O. Wilke (2019). Evolutionary couplings detect side-chain interactions. PeerJ 7:e7280.

Reviewer #1 (Public Review): 

The manuscript by Swygert et al. characterizes the local chromatin compaction (less than 1 KB) in quiescent yeast cells and its role in the repression of gene expression. The authors conclude that Q cells have increased heterogeneous interactions between nucleosomes beyond N+1, resulting in a disordered 10-30 nm fiber. These interactions require the basic patch of histone H4 and are blocked by H4 acetylation. Q induced local compaction occurs independent of condensin and represses transcription at a subset of genes that partially overlap genes repressed by condensin. These conclusions are supported by well-executed and generally well-performed experiments. These insights into the changes in chromatin structure and their impact on transcription silencing in quiescent cells will provide a useful conceptual and technical paradigm for studies of quiescence in other organisms.

Reviewer #2 (Public Review): 

In their paper, 'Local chromatin fiber folding represses transcription and loop extrusion in quiescent cells,' Swygert et al combine high-resolution Hi-C analyses with a slew of imaging, chemical perturbation, and computational experiments to investigate how chromatin folding in quiescent S. cerevisiae cells. This is an impressive paper - the authors perform multiple orthogonal analyses to demonstrate that quiescent yeast nuclei do, indeed, harbor a distinct chromatin structure compared to log-phase yeast nuclei, one in which the chromatin fiber is (on average) folded into a more compact structure during quiescence. The authors then employ a mixture of chemical and genetic perturbations to provide compelling evidence that 1.) this folded fiber structure represses transcription (though fiber folding is not absolutely necessary for transcriptional repression), and 2.) that a basic patch in the tail of core histone H4 is necessary for maintaining fiber folding in quiescence, in part through interactions with a lysine deacetylase. While  these findings are likely to be of interest to the chromatin field at large, this paper does have some weaknesses relating to data interpretation and experimental interpretation that should be addressed by the authors.

Reviewer #1 (Public Review): 

The recordings done by the authors are impressive and rare, and I appreciate the efforts of the authors to bridge very different types of signals that are generally recorded in different paradigms. However, the analysis at many places is quite nuanced and high-level, making it difficult to directly compare these findings with previous results. I think several additional analyses are needed to properly place these findings with previous results. 

1) Effects of attention in V4 generally start earlier (~100 ms). It is unclear why no effect is observed during earlier time periods in these data. To make better comparison with previous studies (such as Nandy et al., 2017), the authors should show the average PSTHs in supragranular, granular and infragranular layers during both target-out versus target-in conditions. Interestingly, Nandy and colleagues found largest changes in firing rates in the granular layer. To better understand the ERP outside the cortex, the authors should also show the average LFPs in the three layers, for target-in and target-out conditions. It is surprising that MI analysis reveals no significant information about the target in granular layer - given that some attentional effects are seen in upstream areas such as V1 and V2. 

2) Eye position analysis: my understanding is that the animals could make a saccade as soon as the arrays were displayed. Given that the main effect of attention is observed after ~150-200 ms, the potential effect of saccade preparation could be important. There could also be small eye movements before the saccade. Given that the RFs were quite fovial for one monkey and not too far from the fixation window, and the effect of attention appears to be quite late, detailed analysis of eye position and microsaccades is needed to rule out the possibility of differences in eye movements between target in and target-out conditions influencing the results. A timeline and some analysis of eye movement patterns would be appropriate. The authors should also clearly mention the mean and SD of the saccade onset. 

3) Attention studies typically keep the stimulus in the RF the same to tease out the effect of attention from stimulus selectivity. Ideally, the comparison should be between the two green (or red) in RF conditions as shown in Figure 4A. However, these results are shown only after pooling across all color selective columns. This comparison should be shown from Figure 2 itself (i.e., Figure 2C should have green in the RF and red target outside). 

4) Information has been well characterized in a large number of previous studies (generally yielding values between a few bits/s, see for example, Reich et. al, 2001, JNP). Here, the absolute value of mutual information seems rather low. This may be due to the way the information is computed. A discussion about these reasons would be useful for scientists interested in information-theoretic measures. 

5) Dependence on feature preference: The effect of spatial and feature attention is well studied. A multiplicative gain model of spatial attention would predict a larger increase in firing rates )and perhaps other signals such as CSD) for preferred versus non-preferred signals. Feature similarity gain model would predict the red preferring columns to increase their activity and green preferring columns to reduce their activity when the animal is attending to the feature red, irrespective of which stimulus is in the receptive field. Here, the task is a pop-out task which likely has both a spatial and feature attention component. The authors should discuss their findings in these contexts. Further, the authors should discuss whether their findings could just be a reflection of the magnitude of the change (which could be larger for preferred versus non-preferred stimulus). The information-theoretic measure should ideally not depend on the absolute magnitude, but these quantities often get biased in non-trivial ways based on the magnitude. Does information transmission depend on the magnitudes of firing rates/CSDs? 

6) For columns that were not feature selective, is there an effect of attention? Does the magnitude of N2pc change depend on color selectivity? I think that should be the case based on Figure 4H and 4I, but a plot and/or some quantification would be useful.

Reviewer #2 (Public Review): 

Scalp ERPs are widely used in human neuroscience research to understand basic mechanisms of neural and cognitive function and to understand the nature of neurological and psychiatric research. However, this research is hampered by a surprising lack of research in animal models exploring the neural mechanisms that produce specific ERP components. 

Previous research by this research group identified a potential monkey homologue of the N2pc component, a neural correlate of the focusing of attention onto visual objects embedded in arrays of distractors. The present study took a giant leap forward by recording extracellular potentials from densely spaced arrays of electrodes (.1 mm spacing) on probes that extended perpendicular to the cortical surface. These electrode arrays made it possible to simultaneously record voltages throughout the different layers of a cortical column and convert these voltages into current source density (CSD, which isolates local synaptic current flow and minimize volume-conducted activity from other brain regions). In addition, simultaneously recorded voltage from an electrode just above the cortical surface was used as a proxy for scalp potentials. Scalp ERP recordings were also obtained from separate monkeys to measure the actual scalp ERPs and verify that an N2pc-like ERP was elicited by the task (a simple visual search task in which the monkey made an eye movement to the location of a color popout item). 

Very clear CSD was observed in V4 in both supragranular and infragranular layers that was stronger when attention was directed to the contralateral visual field than when attention was directed to the ipsilateral visual field, which is the hallmark of the N2pc component. Little or no such activity was observed in the granular layer (the primary recipient of feedforward projections). In addition, the effects were observed primarily when the column was selective for the target's color. An information theory analysis showed that these intracortical current flows contained significant information about the voltage measured on the cortical surface and the location of the target object. 

All of these results were clear and convincing. Moreover, the laminar and columnar analyses provide interesting new evidence about attention-related neural activity independent of any considerations about ERPs. The most challenging aspect of the study is to provide a solid link from the intracortical activity to the voltage on the cortical surface, and then to the monkey scalp ERPs, and finally to human ERPs. Toward that end, the present study relied entirely on correlational evidence, rather than experimental manipulations. That's quite appropriate for a first step, but it must be considered an important limitation on the conclusions that can be drawn. It would be wonderful if future research took the next step of providing experimental evidence. 

There are also some troubling aspects of the existing evidence. The scalp ERP effect in this study and the prior work from this groups is a positive voltage over the contralateral hemisphere, whereas in humans the voltage is negative. This may well reflect the orientation of the relevant cortical surface in monkeys versus humans. However, the voltage on the cortical surface in the present study was negative contralateral to the target, not positive. Unless this opposite voltage on the cortical surface relative to the scalp reflects something about the reference site for the cortical surface electrode, then this makes it difficult to link the intracortical effects and cortical surface effects to the scalp ERP effects. Also, the CSD was negative in the upper layers and positive in the lower layers, again suggesting that the voltage should be negative contralateral to the target on the surface. Ironically, this polarity is what would be expected from the human brain, where a contralateral negativity is observed. The oddity seems to be the contralateral positivity in the monkey scalp data. Also, the cortical surface voltage exhibits a polarity reversal at approximately 180 ms, which is not seen in the intracortical CSD. 

One possible explanation for the discrepancy is that the scalp voltage likely comes from multiple brain areas besides V4. If, for example, areas on the ventral surface of the occipital and temporal lobes produce stronger scalp voltages than V4 under the present conditions, the opposite orientation of these areas relative to the cortical surface would be expected to produce a positive voltage at the scalp electrodes. 

The manuscript notes that multiple areas probably contribute to the scalp ERPs and argues that the pattern of intracortical CSD results obtained in V4 will likely generalize to those areas. That seems quite plausible. Moreover, the results are interesting independent of their link to scalp ERPs. Thus, the present results are important even if the scalp polarity issue cannot be definitively resolved at this time. 

There are also some significant concerns about the filters. The high-pass cutoff was high enough that it could have produced artifactual opposite polarity deflections in the data. If causal filters were applied (e.g., in hardware during the recordings), these artifactual deflections would have been after rather than before the initial deflection, possibly explaining the polarity reversal at 180 ms. If noncausal filters were applied in software, this would be a larger problem and could produce artifacts at both the beginning and end of the waveform. Moreover, the filters were different for the CSD data and the extracortical voltages, which is somewhat problematic for the information theoretic comparisons of these two data sources (but is likely to reduce rather than inflate the effects).

Reviewer #3 (Public Review): 

In this study, Westerberg et al., investigate the cortical origins of the N2pc, an ERP for selective attention. By using a combination of indefinite inverse models of cranial EEG and translaminar electrophysiology, the authors demonstrate that dipoles in V4 are the source of the N2pc. 

The study is well conducted and the manuscript is well written. 

I have a few comments about the CSD, RF alignment profiles, and LFP based analyses: 

(A) The method section states correctly that "current sinks following visual stimulation first appear in the granular input layer of the cortex, then ascend and descend to extra granular compartments". However in the example CSDs shown in Fig 2, Fig 3, Fig S3 there is no visible current sink in the infra-granular layers. Instead, the identified infra-granular layers show a prolonged current source (e.g. Fig S5B,C), which is unexpected. 

(B) The example RF profile shown in Fig S5A, although aligned, looks a little strange in that the RFs taper off rapidly in the infra-granular layer. Is this the best representative example? It will be important to see other examples of RF alignment. 

(C) The study used LFP power in the gamma range to compute the response ratio between red and green stimuli. LFPs measured across the cortical depth are highly correlated, and so would gamma power estimated from the LFPs. Given this, how meaningful is the laminar analysis shown in Fig 4B? How confidently can it be established that the LFP derived gamma power estimates have laminar specificity?

Reviewer #1 (Public Review): 

Ring-shaped proteins known as AAA+ ATPases consume chemical energy in form of ATP to catalyze essential cellular reactions ranging from the copying and repair of DNA to signal transmission at neuronal synapses. In the present study, computational and biochemical approaches are used to model how the six subunits of a particular AAA+ ATPase known as Rpt1-6 coordinate their enzymatic activity with each other to exert unidirectional pulling forces on target polypeptide chains that promote protein unfolding. Although the technical aspects of the work can be difficult to follow at times, the conclusions seem largely supported by the observations. The findings indicate that the order in which ATPase active sites fire around the is generally sequential, much like a rotary engine, but that the system can tolerate "misfires" - instances in which a subunit fails to hydrolyze ATP - by skipping the problematic subunit. The work should appeal to the broad AAA+ community and researchers trying to understand the biophysical principles by which complex biological machines operate.

Reviewer #2 (Public Review): 

I found the manuscript difficult to follow and assess in part because of the vast scope of the experimental approach. At the same time, the conclusions could be worked out more clearly and put into the context of the current understanding of proteasome action.

Reviewer #3 (Public Review): 

This paper presents a simple mathematical model that explains how conformational dynamics of the 26S proteosome enable it to perform mechanical work. The model/approach also have broad implications for other members of the AAA+ family of motors, adding to the significance of the results. Despite a number of simplifying assumptions (e.g. that all 6 subunits have identical behavior), the model does remarkably well at reproducing a variety of experimental data. In particular, the model recapitulates the dominant states/transitions seen experimentally without building in serial transitions. Other transitions are possible, they just have lower probability. Importantly, this heterogeneity of possible pathways allows the motors to continue functioning even when deleterious mutations are introduced, in agreement with experiments. Deleterious mutations are modeled by removing key interactions in the model, rather than explicitly modeling the physical chemistry of different amino acids. This example highlights how the model can be used to explore the motors conceptually, but is not suited to predicting the impact of specific mutations (or other perturbations). The authors also do a reasonable job of acknowledging the limitations of their model.

Reviewer #1 (Public Review): 

In this manuscript by Rankin et al., the authors proposes a model of reciprocal mesoderm-endoderm interactions involving Tbx5 activation of retinoic acid (RA) production in the posterior second heart field (pSHF) that activates endoderm expression of patterning ligands such as Shh, which feeds back to activate the pSHF to coordinate cardiopulmonary development. This is a nice model that bridges previous work from the same authors, which had shown that Tbx5 can alter Shh expression in a non-cell autonomous manner (Steimle et al, 2018), along with prior experiments showing that RA induces Shh expression (Rankin et al, 2016). As such, the novelty here lies in the mechanistic portion of the paper that describes how Tbx5 induces Aldh1a2, a gene responsible for RA production in the pSHF, along with the interactions of Tbx5 with putative enhancers of Aldh1a2. The use of the xenopus model system allows the authors to perform elegant epistasis experiments using morpholinos and Crispr/CAS9 excisions in the whole embryos, which nicely illustrates the role of Tbx5 in inducing Ald1a2 expression and the role of Tbx5 in downstream pathways described. The experiments are well-presented and follows a clear logic, and they are mostly supportive of the experimental models presented at the end of the manuscript. However, a potential weakness of this manuscript is the reliance on pharmacologic methods of modifying RA pathways rather than using a genetic/RNA targeting approach, which would confer more specificity related to the functional importance of the Tbx5 transcriptional target described (in this case, Aldh1a2). Furthermore, more precise colocalization of Tbx5 and Aldh1a2 within the developing cardiopulmonary tissues is important, with some clarification as to why there appears to be broad Aldh1a2 expression independent of Tbx5.

Reviewer #2 (Public Review): 

In this manuscript Rankin et al. combined mouse and frog genetic models to study the gene regulatory network orchestrated by the transcription factor Tbx5. The authors demonstrated that Tbx5 regulates expression of Aldh1a2 which catalyzes the production of retinoid acid (RA) in the lateral plate mesenchyme, thereby activating RA signaling which then signals to the foregut endoderm and induces Shh expression there. In turn Shh activates Hedgehog signaling in the mesenchyme where it works with Tbx5 to promote expression of Wnt2/2b. Wnt2/2b then initiates lung specification from the anterior foregut endoderm. Biochemistry assays were used to assess howTbx5 and RA regulate the transcription of Aldh1a2 and Shh, respectively. Two evolutionarily conserved enhancers were identified for the regulation of the transcription of Aldh1a2 and Shh. The authors in the end suggested that their work provides knowledge basis for better understanding the pathogenesis of the human birth defects DiGeorge Syndrome and Holt-Oram Syndrome. The findings help to fill the knowledge gap, connecting several observations made by previous studies. Moving forward, Tbx5 and other Tbx genes (e.g. Tbx4) continue to be expressed in the developing lungs. Whether a similar regulatory axis is present to modulate lung epithelial and mesenchymal development remains to be explored.

Reviewer #3 (Public Review): 

Previously the Zorn lab has published that retinoic acid-hedgehog signaling is a key step in lung specification. (Rankin et al, Cell Rep 2016, 66-78.) Previously, molecular networks have been proposed for the early heart/lung differentiation. Examples include: (Xie, L.et al (2012). Dev. Cell 23 (2), 280-291; Steimle, J. D., et al. (2018). Proc. Natl. Acad. Sci. USA 115 (45), E10615-E10624. Peng T, et al, (2013) Nature 500(7464):589-92). Although many pieces of these signaling and transcription factor activities have been described, this manuscript demonstrates additional information. The strengths are the use of an in vivo system to tease out the transcriptional elements regulated by Tbx5 and RA. The authors perform sufficient experiments to support their claims although some of these readouts are qualitative rather than quantitative. The authors include relevant controls where possible. The authors were also rigorous by providing a time window for when Tbx5 control of Raldh2 occurs. One weakness is that the manuscript is difficult to follow for individuals who are not familiar with past published data in these networks. Another weakness is that some of the data is drawn from whole mount images and bulk sequencing which could lead to overstatements. A third weakness is that the manuscript does not have a clear focus. Its main concept is filling in the gaps for some of the gene transcription networks that have been described previously. An additional weakness is that almost all of the gene manipulations are global either by morpholino or chemical treatment (inhibition and activation). Finally, it is unclear what the outcomes of the signaling disruptions are in the embryos. We see a snapshot of gene expression but not how this affects organ development in the long run.

Reviewer #1 (Public Review): 

In this work authors focused on RIM neurons which are part of the locomotion circuit of C. elegans but whose role was enigmatic. They put particular focus on the fact that RIM has multiple ways to communicate to other neurons, through glutamate, tyramine gap junctions as well as neuropeptides. 

It is praised that they made precise cell-specific manipulation of each of these transmission pathways and clarified the different roles of the transmission by combining with detailed quantification of behavior. 

There are several interesting observations: lack of AIM glutamate and/or tyramine both caused increased reversal initiation rate. This is similar to the effect of RIM ablation previously reported. Lack of glutamate and/or tyramine also shortened reversal duration. However, of all reversals, lack of tyramine caused decreased rate of omega, while increased pure reversals occurred instead. It is a good contrast to the previously reported role of tyramine in suppressing head swing during reversal. 

Most interestingly, hyperpolarization of RIM by the HisCl channel caused reduced rate of reversal initiation and extended forward run duration. This was opposite to the effect of removal of RIM, and suggested that RIM hyperpolarization has some active effect. This was proposed to be caused through gap junctions. Authors tried to inhibit gap junctions by cell-specific expression of unc-1(n494), which by itself caused increased rate of reversal initiation and shortened forward runs. This effect was antagonistic to effect of unc-1(n494). 

A previous paper from the same lab (Gordus et al. 2015) showed that RIM neurons are activated (or inactivated) along with AVA and AIB in response to odor stimulus and this was assumed to be mediated by gap junctions between these neurons and correspond to the reversal behavior. On the other hand, an opposite activation pattern was sometimes observed for RIM (for example Piggott et al. 2011). The current manuscript solves the confusion to some extent. 

It is important to note that the current manuscript focuses on spontaneous activity changes during "local search" and "global search" (especially the former) in unstimulated animals, because the role of RIM is likely different between evoked and spontaneous behaviors. Under these settings, it looks like glutamate and tyramine mainly acts for prolonging reversal behavior, while gap junction looks to have a major role in prolonging forward motion. Overall, RIM seems to have a role for "behavioral inertia" as proposed by the authors. 

This is a very informative work and the idea of positive and negative regulation causing "behavioral inertia" is novel and interesting.

Reviewer #2 (Public Review): 

The interneuron RIM affects many behaviours. Attempts to understand or manipulate its function have sometimes led to conflicting and difficult to interpret results. Sordillo and Bargmann investigate the role of the RIM interneuron in locomotion by manipulating it in multiple ways: chemogenetic hyperpolarization, optogenetic depolarization, and genetic disruption of glutamate, tyramine, syanptobrevin-dependent, or electrical synaptic signaling. They reach two primary conclusions: 1. RIM depolarization extends reversals via synaptic (glutamatergic) and secretory (tyraminergic) signaling; 2. RIM hyperpolarization promotes forward locomotion via gap junctions. 

These conclusions of the study are well supported by the experiments. Overall, it is of interest to the subfield because it resolves some conflicting interpretations of the role of RIM in regulating locomotion. Of broader interest, it illustrates an interesting case of a multifunctional neuron involved in stabilizing behavioral states, the complexity of the interplay between electrical and chemical signaling, insight into network/circuit functions of electrical synapses, and the caution needed in interpreting results from various kinds of neural manipulation.

Reviewer #3 (Public Review): 

Sordillo and Bargmann report a detailed study of mechanisms by which RIM interneurons control foraging by C. elegans. A comparison of the effects of knocking out the vesicular glutamate transporter in RIMs to the effects of knocking out synthesis of the monoamine transmitter tyramine leads the authors to conclude that glutamate/monoamine cotransmission is required for RIM function. The authors further find that acute perturbation of RIMs by chemogenetics has a surprising effect. Manipulation of RIMs with HisCl - a histamine-gated ion channel that permits rapid and reversible inhibition of target cells - had the opposite effect of RIM ablation or mutations that affect RIM neurotransmission. The effects of HisCl-mediated perturbation require gap junctions, leading the authors to conclude that gap-junction connectivity between RIMs and their targets promotes specific foraging behaviors while neurochemical signaling from RIMs to their targets promotes antagonistic behaviors. 

This study has several strengths. Sophisticated genetic tools are developed to perturb RIMs. Measurements of RIM-dependent foraging behaviors are made using high-resolution video tracking systems, and these rich datasets are clearly presented and rigorously analyzed. The manuscript is clearly written and beautifully illustrated. And the overarching hypothesis that RIM interneurons support distinct behavioral programs when depolarized and hyperpolarized is provocative and significant. The study also has weaknesses, some of which significantly impact the strength of the authors' conclusions. These weaknesses are listed below. 

1) One major conclusion is that RIMs use both glutamate and tyramine as co-transmitters. This conclusion is based on the observed effects of VGLUT knockout in RIMs. It is known, however, that VGLUT facilitates the loading of monoamine neurotransmitters into vesicles, raising the possibility that the observed effects of VGLUT knockdown are via effects on tyraminergic signaling. The authors discuss this point and argue that an observed difference between the effects of tdc-1 mutation and RIM-specific VGLUT mutation indicates separable functions of glutamate and tyramine, i.e. co-transmission. However, most of the data are also consistent with a model in which VGLUT facilitates VMAT function. This point is critical for one of the study's main conclusions and should be resolved. For example, a clear role for a known glutamate receptor in RIM-mediated behavior would support the authors' conclusion.s 

2) The surprising observation that HisCl-silencing of RIMs causes the opposite effect as ablation of RIMs or mutation of the monoamine biosynthetic pathway is the basis for the other major conclusion of the study. The authors conclude that this difference reflects a signaling function for hyperpolarized RIMs that is eliminated by ablation. This difference might also reflect differences between chronic and acute perturbations. Methods exist to chronically silence neurons by expressing hyperpolarizing conductances, and the authors' model suggests that these manipulations would cause effects similar to those caused by acute inhibition via HisCl. 

3) HisCl silencing of RIMs was performed using a tdc-1::HisCl transgene, which supports expression in RIMs and RICs. The authors should be certain that RICs have no role in the effects they see using this transgene. Similarly, perturbation of RIM gap junctions uses a tdc-1-based transgene, which should be paired with a control that allows the authors to rule out any contribution of RICs. 

4) The authors' final model proposes that the constellation of chemical and electrical synapses endows RIMs with a kind of 'inertia.' In the absence of any data that report how perturbation of RIMs affects dynamics of the foraging circuit (AIB/RIB/AVA/AVB) is it difficult to assess this model. 

Minor comments:

The authors use variants of 'RIM glutamate KO' when referring to to the strain carrying RIM-targeted allele of eat-4/vglut. They should be consistent. 

A critical set of control experiments for the eat-4 conditional allele is presented in Figure 2S1 but not mentioned in the manuscript until data from Figure 3 are being discussed. These controls should be clearly described earlier in the results section - they are beautiful experiments and establish confidence in the method. 

Line 192 refers to 'a synaptobrevin-dependent transmitter.' This is correct, but it might be more clear to simply say 'another neurotransmitter.'

Reviewer #1 (Public Review): 

The entrapment of viral DNA and viral capsids in PML cages is efficiently achieved only when the cells are infected with a low MOI (ie one copy of HCMV). This is a well performed work, which describes an interesting pattern of nuclear changes upon viral infection. However, the question that remains is how is PML capable of sensing the number of viral genomes. As PML cages can be formed in the absence of viral infection, exemplified by the fact that IFN signalling and DNA damage can induce their formation, I would expect the authors to deepen this part of the work, which is rather limited to data presented in Figure 4. The functional relevance of disruption of PML cages by viral protein IE1 and its impact on HCMV replication (shown in Figure 6) is a nice demonstration of the strategy that virus evolved to disrupt these structures. It would also be very interesting to understand how are these structures metabolized in the absence of viral infection, especially because the authors highlight the importance of IFN and DNA damage for their formation.

Reviewer #2 (Public Review): 

Utilising a combination of high-resolution light microscopy and electron microscopy imaging, Scherer et al., identify promyelocytic leukaemia nuclear bodies (PML-NBs) to undergo extensive rearrangement during HCMV infection in the absence of the viral PML- antagonist IE1 protein. These data identify PML, the principal scaffolding protein of PML-NBs, to undergo dynamic structural rearrangements throughout the course of HCMV infection in a manner dependent on the activation of IFN-mediated innate immune defences and induction of the cellular DNA damage response (DDR). As such, the authors identify PML to play sequential roles in the spatiotemporal restriction of HCMV at multiple phases of infection dependent on the immunological state of the cell. The manuscript is accessible and well laid out, exceptionally well written, and experiments conducted to a high standard. The authors conclusions are generally supported by the data without over interpretation. Some aspects of the image analysis, including population and statistical testing, and DDR activation during infection require extending to add support to their major conclusions. Nevertheless, the study overall resolves many conflicting issues in the current literature surrounding the antiviral properties of PML during HCMV infection and identifies important future areas of research pertinent to both virology, immunology, and cell biology research communities. Fascinating science!

Reviewer #3 (Public Review): 

In their study Scherer et al. demonstrate that the PML NBs act as a nuclear intrinsic antiviral response against the incoming HCMV genomes and more surprisingly against capsids. Using a set of approaches such as click chemistry to label the viral genomes, immunofluorescence, and electron microscopy their show that PML NBs entrap incoming viral genomes forming giant PML NBs leading to transcriptionally repressed viral genomes. If viral genomes escape the first layer of restriction activity of the PML NBs, to progress into the lytic cycle, they show that the PML NBs are also able to entrap capsids in a second layer of antiviral defense mechanism. Finally, they show that PML cages formation containing viral genomes or nucleocapsids arise via the combined interferon and DNA damage signaling. 

Major strengths and weaknesses:

Strengths <br> - The study nicely demonstrates the major involvement of the PML protein and PML NBs in the control of the incoming viral genomes during infection by the human cytomegalovirus (HCMV). <br> - Results are clear, nicely illustrated and presented in a easily understandable manner. <br> - The methods in use especially click chemistry to visualize the incoming viral genomes and combination of light microscopy with CLEM and FIB-SEM to visualize viral capsids entrapment in the nucleus are really challenging. <br> - The study is of broad interest regarding various pathological situations whether they result from viral infection (HCMV, HSV, VZV, HPV, HBV...) or genetic disorders (ICF, ALT), and in which PML NBs play major roles. 

Weakness <br> - Although at mechanistical and molecular levels the study does not suffer of major weaknesses to reviewer's opinion, at the physiological level it would have been interesting to provide some data on the formation of PML cages in cells supporting HCMV latent infection such as bone marrow CD34+ cells or alternatively THP1 cells. However, the reviewer acknowledges the fact that this could be out of the scoop of this study given the amount of work it could necessitate to provide a complete set of data in cells supporting HCMV latency in physiological conditions. 

Appraisal:

To reviewer's view there is no doubt that the authors achieved their aims to demonstrate the physical interaction between viral genomes and capsids with PML NBs and the role of this epigenetic regulation in the establishment and maintenance of HCMV latency. As such, they data nicely support previous studies either showing entrapment of incoming viral genomes during HSV-1 lytic infections and latency (Everett et al, 2007; Catez et al, 2012; Alandijany et al, 2018; Cohen et al, 2018), and of capsids during VZV infection (Reichelt et al, 2011, 2012). One of the originality of the study by Scherer et al. stands in the fact that it is the first time that such interactions are described for a herpesvirus of a subfamily other than alphaherpesvirinae (HCMV being a betaherpesvirus). Additionally, it is the first time that both, PML cages entrapping viral genomes or capsids are described for the same virus and during the process of infection. 

References <br> Alandijany T, Roberts APE, Conn KL, Loney C, McFarlane S, Orr A & Boutell C (2018) Distinct temporal roles for the promyelocytic leukaemia (PML) protein in the sequential regulation of intracellular host immunity to HSV-1 infection. PLoS Pathog 14: e1006769 <br> Catez F, Picard C, Held K, Gross S, Rousseau A, Theil D, Sawtell N, Labetoulle M & LOMONTE P (2012) HSV-1 Genome Subnuclear Positioning and Associations with Host-Cell PML-NBs and Centromeres Regulate LAT Locus Transcription during Latency in Neurons. PLoS Pathog 8: e1002852 <br> Cohen C, Corpet A, Roubille S, Maroui M-A, Poccardi N, Rousseau A, Kleijwegt C, Binda O, Texier P, Sawtell N, et al (2018) Promyelocytic leukemia (PML) nuclear bodies (NBs) induce latent/quiescent HSV-1 genomes chromatinization through a PML NB/Histone H3.3/H3.3 Chaperone Axis. PLoS Pathog 14: e1007313 <br> Everett RD, Murray J, Orr A & Preston CM (2007) Herpes simplex virus type 1 genomes are associated with ND10 nuclear substructures in quiescently infected human fibroblasts. 81 <br> Reichelt M, Joubert L, Perrino J, Koh AL, Phanwar I & Arvin AM (2012) 3D reconstruction of VZV infected cell nuclei and PML nuclear cages by serial section array scanning electron microscopy and electron tomography. PLoS Pathog 8: e1002740 <br> Reichelt M, Wang L, Sommer M, Perrino J, Nour AM, Sen N, Baiker A, Zerboni L & Arvin AM (2011) Entrapment of viral capsids in nuclear PML cages is an intrinsic antiviral host defense against varicella-zoster virus. PLoS Pathog 7: e1001266 

Likely impact: 

In the field of herpesviruses and other DNA and nuclear replicating viruses the role of PML NBs as part of the intrinsic immunity has become a major subject of investigations for several years. Hence, this work represents a major contribution in the understanding of the role of PML NBs in the antiviral response. 

Additional context:

As investigated by the authors in figure 6 this work could be of interest for scientists working in the field of telomeres biology particularly in the context of cancer cells that maintain telomeres length by the alternative telomere lengthening (ALT) process. Indeed, in those cells telomeres are entrapped in PML NBs forming structures called ALT associated PML NBs (APBs). Any kind of study investigating similar behavior for PML NBs in sequestrating chromatin loci, whatever it is for HCMV, HSV, or other type of viruses are likely to bring new clues and idea concerning the role and the formation of the APBs.

Reviewer #1 (Public Review):

In metazoan systems, YTH-domain proteins recognize N6-methyladenosine (m6A) in mRNA to affect these molecules half-life or their translation efficiency. What remains unknown is whether mammalian YTHDF proteins act redundantly or not. Conversely, in plants EVOLUTIONARILY CONSERVED C-TERMINAL REGION (ECT) 2-4 were reported to work redundantly to promote proliferation of primed stem cells in organ primordial. However, it is largely unknown how these functions are achieved. In this manuscript, the authors rely on the data obtained in a companion manuscript to define ECT2 and ECT3 targeted mRNAs, redundant functions, and specific targets. In addition, the manuscript also describes divergent functions that make plant YTHDF proteins unique compared to mammals.

The authors also used super-resolution microscopy to precisely define ECT2 sub-cellular distribution, demonstrated that ECT2/3/4 do not influence poly(A)-site choice in their targets as previously suggested, and showed that ECT2/3-targets are repressed in the absence of these proteins.

Experimentally this is a solid paper that re-evaluates several aspects of YTHDF proteins disproving some functions wrongly attributed to these proteins. The conclusions achieved in the paper are of general interest to the community studying post-transcriptional gene regulation and especially to RNA biology. The experimental data provided is solid and support the conclusions. This paper is built on a companion paper that provides the foundations for the study.

Reviewer #2 (Public Review):

Many plant mRNAs contain N6-methyladenosine (m6A) towards the 3' end of the transcript. YTH-domain proteins of plants and animals have a binding pocket that accepts the m6A. These YTH proteins fall into two broad classes, YTHDF type and YTHDC type. Arabidopsis has two YTHDC types and one of these is a fusion with the CPSF30 polyadenylation factor and has been implicated in 3' processing of transcripts in the nucleus. In contrast, Arabidopsis, like many plants, has an expanded family of 11 YTHDF type proteins which can be divided into 3 or possibly 4 subgroups based on sequence similarity and evolutionary trees. ECT2, ECT3 and ECT4 fall into one of these subgroups. Clear developmental phenotypes are only seen when both ECT2 and ECT3 are knocked out and these phenotypes are stronger when ECT4 is also absent - suggesting that the four proteins act redundantly (at least in part). Previous publications have suggested that ECT2 can move to/be found in the nucleus and that it may promote alternative polyadenylation. The work reported here shows that ECT2 is cytoplasmic and that the nuclear location previously reported is likely artifactual.

The authors identify a high confidence set of transcripts that are bound by ECT3, and they demonstrate that there is a very high level of overlap between ECT3 and ECT2 targets. Through a series of experiments, they show that binding of these target mRNAs is likely occurring in the same cells (rather than different tissue/cell types), and thus the two proteins are likely to be truly acting redundantly.

A key finding is that in the ECT2, ECT3, ECT4 triple mutant, the high confidence target mRNAs are reduced in abundance - thus suggesting a role in stabilising m6A containing transcripts.

Questions raised

The authors have previously shown that there are subtle differences between ECT2 and ECT3 mutant plants, most notably in the direction of root growth. It will be interesting to determine if this can be explained by the small number of differentially bound target mRNAs.

It remains formally possible that, whilst ECT2 and ECT3 bind the same mRNA targets and behave redundantly under most conditions, under certain treatments or environmental stimuli, they might be differentially post-translationally modified or may interact with different protein partners to bring about different consequences for bound transcripts. A larger question relates to the remaining YTHDF type proteins belonging to the other three subgroups. It will be interesting to determine whether these too act redundantly with ECT2, carry out a similar function in different cell types, or act to bring about different or even antagonistic outcomes.

Perhaps the most exciting finding is that transcripts that are targets of ECT2 and ECT3 are preferentially reduced in abundance in the triple mutant. This suggests a role for these proteins in cytoplasmic stabilisation of their mRNA targets, which would be consistent with observations from low m6A plants. The authors point out that it is also formally possible that this increase in transcript abundance could result from a feedback effect on transcription, and mRNA half-life measurements would be required to exclude this. However, a further possibility is that the stabilising effect could be indirect, many mRNAs are degraded co-translationally; if ECT2 were to act as a translational repressor, then increased translation of target transcripts in the mutant could actually result in their reduced steady-state levels.

Reviewer #1 (Public Review):

In plants, as in many eukaryotic organisms, the m6A modification of RNA molecules plays critical roles in development, stress response and adaptation. This modification alters the RNA molecule properties changing their function, folding, and how they are post-transcriptionally processed and modified. RNA-binding proteins containing YTH domains recognize these modified nucleotides. In this manuscript, the authors used a combination of iCLIP (individual-nucleotide resolution crosslinking and immunoprecipitation) and HyperTRIBE (targets of RNA-binding proteins identified by editing) to identify mRNA targeted by the YTH protein ECT2. In addition, the same data was used to define the sequence motifs bound and around ECT2 crosslink sites.

The authors concluded that RRACH sites are the most enriched motifs at m6A sites in Arabidopsis and that ECT2 binds to these sites. It also reports that the intrinsically disordered region of ECT2 participates in RNA binding through U-rich elements abundant upstream of m6A-sites. Finally, they reported that URUAY motifs, previously reported as targets of ECT2 are not the preferential target sites and even disfavors ECT2 binding.

The conclusion achieved in the paper is of general interest to the community studying post-transcriptional gene regulation as the datasets generated and analyzed here. The experimental data provided is solid and fully support the conclusions. A companion paper provides the missing information regarding specificity by comparing ECT2 and ECT3.

Reviewer #2 (Public Review):

Most eukaryotes can contain N6-methyladenosine (m6A) in a significant proportion of their mRNAs, and in both plants and animals m6A is required for normal developmental programmes. The modification is usually found towards the 3' end of transcripts within the sequence RR(m6A)CH (R=G/A, H=A/C/U). YTH-domain proteins of plants and animals have a binding pocket that accepts the m6A, and binding to methylated transcripts by such proteins is required to bring about many of the developmental programmes associated with the requirement for m6A. However, previous reports have suggested that one of the major Arabidopsis YTH proteins (ECT2) actually binds the sequence URU(m6A)Y. The work reported here uses UV cross linking of a tagged ECT2 as well as A to inosine editing imparted by an ECT2-ADAR fusion, to identify transcripts that are bound by ECT2 and shows that these correlate strongly with sites mRNAs previously identified as containing m6A. Thus supporting ECT2 as a bona fide m6A reader. In addition, the use of UV crosslinking (iCLIP) suggests that ECT2 binds to m6A predominantly in the DR(m6A)CH sequence context in vivo, consistent with major methylation consensus of eukaryotes. The authors suggest that the earlier claim for URUAY as the target site might have been due to the use of formaldehyde for crosslinking (which will promote protein-protein as well as protein-RNA crosslinks). As well as providing a detailed list of ECT2 target transcripts, and showing that this largely overlaps with previously identified m6A containing mRNAs, the authors also undertook a detailed study of other sequence motifs associated with m6A and ECT2 binding. The key conclusions from this are that pyrimidine-rich motifs are enriched around, but not at m6A-sites, which likely reflects a preference for such motifs in the broader recognition of mRNAs by components of the enzyme complex that carries out the methylation. In particular, oligo-U and UNUNU motifs are frequently found upstream of the DR(m6A)CH site and evidence is presented that these can become UV crosslinked to intrinsically disordered region (IDR) N-terminal of ECT2. The relevance and importance of this binding by the IDR remains to be determined, but the suggestion that it helps stabilise binding to m6A containing transcripts is a reasonable hypothesis.

While the authors conclude that URUAY is not the main binding site for ECT2, where it is present close to a DR(m6A)CH site, it could become cross linked to ECT2. However, URUAY was more enriched proximal to previously identified DR(m6A)CH that were not bound by ECT2, which the authors interpret as indicating that another RNA binding protein is competing for such sites.

Questions raised

The current report does a thorough job of identifying transcripts that are bound by ECT2. However, plants have a large family of YTH proteins, indeed, Arabidopsis has 11 of the YTHDF type of which ECT2 is a member, and these fall into four sub-groups. It remains an open question whether all of these family members are binding the same set of transcripts, either competing with each other or acting redundantly. If the different ECT proteins are competing with each other, it will be important to determine whether the binding of different ECTs brings about different outcomes and if so whether they have different affinities for the methylated targets (or whether their binding can be moderated eg by environmental conditions). Answering these questions will likely shed light on the complexity of m6A post-transcriptional regulation.

The suggestion that there is a protein that binds to URUAY sites and competes or displaces ECT2 is intriguing and will require further experiments to verify and hopefully identify. Whether such a protein is recognising just the URUAY motif or is binding this motif in the context of a nearby DR(m6A)CH site is also an open question. However, in the latter case, it might be that such a competitor is one of the other ECT family members.

Reviewer #1 (Public Review):

The authors bring compelling biochemical evidence that MatP and ParC compete to interact with the hinge of MukB. In addition, the authors made an effort to support their hypothesis with the description of the phenotype of a mutant of the hinge, mukBKKK, which fails to interact with both ParC and MatP.

Reviewer #2 (Public Review):

In the manuscript entitled "Competitive binding of MatP and Topoisomerase IV to the MukB hinge modulates chromosome organization-segregation", Fisher and colleagues characterized in vitro the interaction between MukB and MatP or between MukB and TopoIV using different biochemical approaches. First, they identified that a dimer of MatP or a dimer ParC interacts in vitro with the hinge domain of MukB but failed to form a tripartite complex ParC/MatP/MukB. Second, they observed that the interaction of full-length protein MatP with MukB hinge is competed out by ParC suggesting that MatP and ParC share overlapping binding sites on the MukB hinge. Third, using a MukB mutant (MukB-kkk) known to be deficient in ParC binding, the authors did not detect any major topo-IV defective phenotype questioning the requirement of direct interaction between TopoIV and MukBEF for chromosome unlinking by decatenation.

Several major difficulties regarding this manuscript are indicated below:

1) Although the authors have thoroughly and elegantly analyzed the in vitro interactions of the MukB hinge with either MatP or ParCCTD, there is no strong evidence revealing the significance of these interactions in vivo. While the interaction between MukB hinge and ParC has been reported by several groups over the years using two-hybrid screens and pull-down assays, the data presented in this manuscript reveal that the absence of such interactions in vivo do not cause major defects in decatenation/chromosome segregation resulting from an impaired action of TopoIV and do not importantly affect MukBEF activity.

2) A robust readout is required to demonstrate the significance of the MatP-MukBEF interaction in vivo. The read-out used in this study (morphology of MukBEF foci) to reveal the absence of interaction between MatP and MukBEF is not easily quantifiable. Different methods have been used before that revealed the inhibition/displacement of MukBEF by MatP. Such methods (Mäkela and Sherratt, 2020; Lioy et al., 2018; 2020) should be used to estimate the MukBEF positioning/activity in the absence of interaction with MatP.

3) The in vivo interaction of MatP with the MukB hinge revealed by the bacterial two-hybrid assay (Nolivos et al., 2016) seems to be weak. A recent study (Bürmann et al. BioRxiv) has revealed the CryoEM structure of MukBEF complexed with MatP. In the structure obtained, MatP bound to its target matS interacts with MukE and the "joint"region of MukB, not with the hinge. Based on the structure, Bürmann and colleagues propose a very attractive unloading model enhanced at matS sites, explaining how MatP prevents MukBEF activity in the Ter region. In the present manuscript, it is not addressed whether the interaction of MatP with the hinge corresponds to a subsequent stage during unloading or contributes to another aspect of MukBEF activity.

4) DNA entrapment by MukBEF requires ATP hydrolysis (Bürmann et al. BioRxiv). It is therefore not obvious to understand how MukBEFEQ is bound in vivo to matS sites. Also as matS sites compete with the MukB hinge for MatP binding in vitro, it is not clear how MukBEF could interact with MatP bound to matS sites in vivo.

5) It is not clear how matS sites could prevent MatP-MukB in vitro interaction. This is not consistent with the observation that MatP is required bound at matS sites to unload/prevent MukBEF activity.

6) The title is misleading as no clear evidence is reported concerning an effect of the interactions on chromosome organization/segregation.

Reviewer #3 (Public Review):

This paper focuses on protein-protein interactions of the E.coli Structural Maintenance of Chromosomes (SMC) complex MukBEF through the hinge domain of its MukB subunit. Previous work has demonstrated that the MukB hinge interacts with the ParC subunit of topoisomerase IV (ParC2E2), which decatenates sister chromosomes, and the MatP homodimer, which preferentially binds chromosomal matS sites near the replication terminus (ter) and excludes MukBEF from ter. This paper expands on previous studies by using a wide range of in vitro assays (isothermal titration calorimetry, native mass spectrometric, fluorescence correlation spectroscopy, analytical size exclusion chromatography, etc) to characterize MukB-MatP and MukB-ParC interactions qualitatively and quantitatively. One major finding is that MatP and ParC compete for MukB hinge binding, rather than forming a MukB-MatP-ParC ternary complex. Additionally, this study reports that a ParC-binding deficient MukB mutant (MukBKKK) is also deficient in MatP binding, suggesting that ParC and MatP have overlapping binding sites on MukB. Further, MatP-matS binding prevents MatP from binding MukB, suggesting that MukB and matS have overlapping binding sites on MatP. Live-cell fluorescence imaging of WT MukB and the binding-deficient MukBKKK mutant confirm that MukBKKK does not colocalize preferentially with ori or bind ParC, in contrast to WT MukB, although it does not show some of the expected Muk˗ phenotypes such as temperature-sensitive growth.

Strengths of this study:

1) Using a range of experimental techniques to study binding interactions between MukB, ParC/topo IV, and MatP helps increase confidence in the findings. For example, multiple lines of evidence (analytical size exclusion chromatography, native mass spectrometry, and fluorescence correlation spectroscopy) all indicate that there is no or minimal formation of a MukB-MatB-ParC ternary complex and that instead MatP and ParC bind competitively to MukB.

2) The in vitro assays use only partially reconstituted complexes, including a substantially truncated form of MukB containing the hinge and a portion of the coiled-coil domain. The inclusion of in vivo imaging experiments showing that the binding-deficient MukBKKK mutant is impaired in ParC binding and proper localization at ori helps to support the relevance of the in vitro results.

3) Experiments are well-controlled and the use of a number of MukB, ParC, and MatP mutants helps to support the conclusions regarding the interactions between these proteins.

Weaknesses of this study:

1) In some cases, different experimental techniques are used to measure binding interactions between WT and mutant proteins, and no explanation is given for the choice of technique. For example, isothermal titration calorimetry and native mass spectrometry are initially used to measure the MukB hinge binding to MatP, but these techniques are not used to characterize the binding of the MukBKKK hinge mutant to MatP. Comparisons between WT MukB and MukBKKK binding to MatP are provided by other methods (native PAGE, fluorescence correlation spectroscopy, analytical size exclusion chromatography), so the difference in binding between the WT and mutant is clearly demonstrated, but the lack of these corresponding experiments creates some confusion and makes it more challenging to interpret some of the results.

2) Although the experiments and analyses are for the most part rigorous and well-controlled, there are a few minor experiments or analyses with some weaknesses. In the analysis of cell size in the in vivo imaging experiments (Figure 5D), the median cell lengths are reported for WT MukB and different mutants. The authors conclude that the cell size is essentially the same for WT MukB and the binding-deficient MukBKKK mutant, suggesting no major chromosome segregation defects for the mutant. However, the difference in median cell length between WT and MukBKKK is the same as that between WT and ΔmukB cells. Thus it is not clear how the authors draw their conclusion from these data. Further discussion and analysis (perhaps the presentation of the full distribution of cell lengths and/or statistical analysis) might support these claims. Further, the demonstration that the MukBS781F mutant is not growth-sensitive (Supplementary Figure 3C) appears to represent a single experiment, whereas even a second replicate would help increase confidence in the results.

3) Although this study provides a more in-depth characterization of the ParC/topo IV and MatP interactions with the MukB hinge than in previous work, it is not clear that the results lead to substantial changes in our understanding of the role of these protein-protein interactions in coordinating chromosome segregation.

Reviewer #1 (Public Review):

The study describes in detail protein abundance and utilization in C. necator. The authors show highly interesting data and draw conclusions about the strategy of C. necator to be ready for environmental changes. The authors combine experimental data and models to predict environmental adpatation of C. necator and the enzyme utilisation. Especially the data regarding the adapation of the CBB in the new chromosome and plasmid are of special interest.<br> The data are presented in a good fashion and the methods are described sufficient.<br> The hypothesis of enzyme utilization was modeled, tested and analyzed and showed a clear different strategy compared to E. coli. The barcode mutant library is a excellent tool to test enzyme essentiality. Interesting is the high protein abundance of unutilized enzymes.<br> The authors also sufficiently describe the limitations of the study, especially the unanotated proteins and proteins not involved in pathways (sensing, motion etc.)

Reviewer #2 (Public Review):

In this work, authors investigated the versatility of the beta-proteobacterium Cupriavidus necator from the proteome perspective. For this purpose, they cultivated the microorganism in a chemostat using different limiting substrates (fructose, fructose with limited ammonia, formate and succinate) and under different dilution rates. Integration of experimental proteomic data with a resource balance analysis model allowed to understand the relation between enzyme abundances and metabolic fluxes in the central metabolism. Moreover, the use of a transposon mutant library and competition experiments, could add insights regarding the essentiality of the genes studied. This shed light on the (under)utilization of metabolic enzymes, including some interpretations and speculations regarding C. necator's physiological readiness to changes in nutrients within its environmental niche. However, several parts of C. necator metabolism are not yet well analyzed (PHB biosynthesis and photorespiration) and some conclusions are not well reported.

Strengths:

1. The manuscript is well written, easily understandable also for (pure) experimentalists, and adds a novel layer of comprehension in the physiology and metabolism of this biotechnologically relevant microorganism. Therefore, it is likely to raise attention and be well-cited among the metabolic engineering community of this organisms.

2. More generally, the scope of the study is broad enough to potentially attract experts in the wider-field of autotrophic/mixotrophic metabolism, especially regarding the metabolic difference in the transition from heterotrophic to autotrophic growth modes and vice versa.

3. Findings from different experimental techniques (chemostat cultivation, proteomics, modelling, mutant libraries) complement each other and increase the level of understanding. Consistency of the results from these different angles increases the roundness of the study.

Weaknesses:

1. A main conclusion of this paper is that it concludes that the CCB cycle operation in heterotrophic conditions (fructose and succinate) is not useful for the biomass growth. However, Shimizu et al., 2015 claim that the CBB cycle has a benefit for at least PHB production is increased, in the presence of the CCB cycle (as demonstrated by a decrease in PHB production when Rubisco or cbbR are knocked out). In this work the authors do not analyze PHB production, but they do analyze fitness in mutant libraries. They claim not see this benefit in this study, however in their data (Figure 5 F) also small fitness drops are seen for cbbR mutants on fructose, as well as on succinate. So I think the authors have to revisit this conclusion. The type of modelling they use (RBA/FBA) may not explain such re-assimilation as 'a theoretically efficient' route, as this type of modelling assumes ' stochiometric' metabolic efficiency with setting a maximum growth objective, which is not what seems to happen in reality fully.

2. The authors focus a lot on readiness as a rational, but actually cannot really prove readiness as an explanation of the expression of 'unutilized' proteome, in the manuscript they also mention that it maybe a non-optimized, recent evolutionary trait, especially for the Calvin Cycle (especially because of the observed responsiveness to PEP of the cbbR regulator). The authors should discuss and not present as if readiness is the proven hypothesis. It would be interesting (and challenging) if the authors can come up with some further suggestions how to research and potentially proof readiness or ' evolutionary inefficiency'.

3. C. necator is well-known for the production of the storage polymer polyhydroxybutyrate (PHB) under nutrient-limited conditions, such as nitrogen of phosphate starvation. Even though the authors looked at such a nitrogen-limited condition ("ammonia") they do not report on the enzymes involved in this metabolism (phABC), which can be typically very abundant under these conditions. This should be discussed and ideally also analyzed. The formation of storage polymers is hard to incorporate in the flux balance analyze with growth as objective, however in real life C. necator can incorporate over 50% of carbon in PHB rather than biomass, so I suggest the authors discuss this and ideally develop a framework to analyze this, specifically for the ammonia-limited condition

4. The authors extensively discuss the CCB cycle and its proteome abundance. However during autotrophic growth also typically photorespiration/phosphoglycolate salvage pathways are required to deal with the oxygenase side-activity of Rubisco. The authors have not discuss the abundance of the enzymes involved in that key process. Recently, a publication in PNAS on C. necator showed by transcriptomics and knockout that the glycerate pathway on hydrogen and low CO2 is highly abundant (10.1073/pnas.2012288117). Would be good to include these enzymes and the oxygenase side-activity in the modelling, proteome analysis and fitness analysis. An issue with the growth on formate is that the real CO2 concentration in the cells cannot be determined well, but not feeding additional CO2, likely results in substantial oxygenase activity.

Reviewer #3 (Public Review):

This work provides new knowledge regarding the versatile chemolithoautotrophic bacterium C. necator, such as presence of many under-utilized enzymes as well as excess amount of utilized enzymes, suggesting the highly robust properties of this bacterium against environmental perturbations. One weakness is thought to be lack of consideration into polyhydroxyalkanoate synthesis occurred in this bacterium under nitrogen limited condition.

Reviewer #1 (Public Review):

The authors extend their previous work on thymus epithelial cells (TECs) and antigen presenting cells (APCs), which focused on TLR signaling in TECs and monocyte-derived dendritic cells (mDCs), and now focus on medullary (m) TECs and ask whether different subsets of DCs can uniquely serve as APCs for tissue-restricted antigens (TRAs) expressed by mTECs.

The approach used makes use several reporter transgenic mouse models to conditionally express a fluorescent reporter gene in mTECs (Defa6-cre) or in TECs more broadly (Foxn1-cre or Csnb-cre). Their findings show that restricted expression of reporter genes in a subset of mTECs, which typically expressed AIRE-dependent TRAs, are typically presented by a subset of DCs that express XCR1 and CCR7, which, together with mDCs, are able to perform cooperative antigen transfer (CAT) effectively. The authors also examined the ability of different DC subsets to acquire antigens from other DCs, and show that mDCs excel in this ability.

Overall, the work is clearly presented and makes use of several elegant mouse models to further delineate the role of different DC subsets in T cell selection. However, as pointed out by the authors, it remains to be determined whether the differences in the ability of different DC subsets to perform CAT has any fundamental impact in the establishing self-tolerance, either by negative selection or induction of Treg differentiation.

Reviewer #2 (Public Review):

Medullary thymic epithelial cells (mTEC) are a heterogenous population of cells that are best known for their expression of genes that are typically limited to certain tissues. This promiscuous gene expression allows developing T cells to test their antigen receptors against the majority of self-peptides they may encounter when patrolling the peripheral organs to ensure those that may be autoreactive to self-antigens are rendered tolerant before they exit the thymus. The expression of any given tissue restricted antigen by mTEC is rare, and an important role for thymic dendritic cells in taking up and presenting antigen from mTEC has been suggested to broaden the presentation of these rare antigens. Further, it has previously been shown that thymic dendritic cells are also diverse and that the subsets are differentially cross-dressed with self-peptide:MHC complexes from mTECs, In addition, it has been suggested that certain dendritic cell populations are more efficient at cross-presenting secreted versus membrane bound antigen. Here, the authors expand upon this previous work and put forth an interesting hypothesis that there is preferential antigen exchange between specific mTEC and dendritic cell subsets.

To test this hypothesis, the authors employ a number of mouse models in which a fluorescent protein is differentially expressed among mTEC subsets. They demonstrate, as expected, that mTEC expressed fluorescent protein can be taken up by thymic dendritic cells. Perhaps unexpectedly, their data are suggestive of biases in the interactions between different dendritic cell and mTEC subsets that lead to the biased exchange of antigen. The authors go on to suggest differences in the efficiency with which individual dendritic cells from different subsets take up antigen from multiple mTECs using a genetic cell-labeling technique to direct distinct fluorescent protein expression among individual mTECs. Lastly, the authors suggest that exchanges in dendritic cell antigen also occur using dendritic cell-directed fluorescent protein expression and bone marrow chimeras. Important controls are provided, and complementary assays are included that strengthen confidence in the authors observations.

One limitation of the authors' approach is the lack of mTEC subset restricted fluorescent reporters; in addition, several of the conclusions in the manuscript are therefore reliant on linear regression analysis to identify correlations in mTEC subset expression of fluorescent protein and dendritic cell subsets that preferentially take up the fluorescent antigen. The authors acknowledge some caveats to this approach. Regardless, the results are consistent with the idea that dendritic cell subsets are specialized to take up and present antigen from certain mTEC populations and set the stage for further testing of this hypothesis as well as the physiological impact of this mTEC-dendritic cell interaction bias awaits the development of additional tools.

Reviewer #3 (Public Review):

In this manuscript, Voboril et al. address the question of whether specific dendritic cell (DC) types in the thymus acquire antigens from distinct thymic epithelial cell subsets. It is well-documented that both medullary thymic epithelial cells (mTECs) and thymic DCs present self-antigens to thymocytes to induce central tolerance. mTECs express the majority of the proteome, including AIRE-dependent tissue restricted antigens (TRAs), and thymocytes must be tolerized against this diverse set of self-antigens to prevent autoimmunity. While some self-antigens are expressed abundantly in an AIRE-independent manner by mTECs, a given AIRE-dependent TRA is expressed at low levels by only 1-3% of mTECs, raising the question of how thymocytes encounter rare, sparse self-antigens during their residence in the medulla. Part of the answer to this comes from the fact that thymic DCs can acquire self-antigens from mTECs to improve the efficiency of display to thymocytes. As the authors point out, several recent studies have utilized single-cell transcriptional profiling to identify multiple distinct medullary thymic epithelial cell subsets. Furthermore, work from the authors' lab and others has demonstrated heterogeneity within the thymic DC compartment. Thus, the authors set out to address whether DC interactions with mTECs are promiscuous or whether specific DC cell types interact with different mTEC subsets to acquire self-antigens to induce tolerance to different types of self-antigens, such as Aire-dependent versus Aire-independent TRAs or ubiquitous antigens.

Using mouse strains that express a fluorescent protein in TECs under the control of different promoters, the authors use flow cytometry to determine the relative ability of thymic DC subsets to acquire self-antigens, TdTomato in this case, from TEC subsets. Using linear regression modeling to compare the frequency of TEC subsets expressing TdTomato in the different reporter strains to the frequency of DC subsets that acquired TdTomato, the authors conclude that there is specificity to the interactions of different DC subsets with distinct mTECs, resulting in antigen acquisition by the interacting DCs. Specifically, they conclude that pDCs and macrophages acquire antigens from mTEClow cells, cDC1 and activated XCR1+ DCs acquire antigen from mTEChigh cells, activated XCR1+ and activated XCR1- DCs acquire antigen from pre-post-Aire mTECs, cDC2 exclusively acquire antigen from Post-Aire mTECs, and activated XCR1+ DCs acquire antigen from Tuft cells. It is well documented that activated XCR1+ DCs (Ardouin et al. 2016, Oh et al. 2018, Perry et al. 2018) and activated XCR1- DCs (Leventhal 2016) acquire and present Aire-dependent self-antigens from mTECs to induce thymic central tolerance. Thus, the most novel claim is that cDC2 acquire antigens from Post-Aire mTECs. However, given that the model is derived from the linear regression analysis of fluorescent reporter mice in which multiple TEC subsets express each reporter, and there are some known caveats to these analyses, this interesting conclusion is not adequately supported by the data. The authors cleverly use Foxn1-cre ConfettiBrainbow2.1 reporters to conclude that moDCs are particularly adept at acquiring antigen serially from different mTECs. Furthermore, they use mixed bone marrow congenic/reporter mice to demonstrate that moDC are particularly good at acquiring antigens from other DC subsets. These two conclusions are well-supported by the data, although it is notable that while moDC are efficient at these two processes, other DC subsets acquire antigens from DCs, and fluorescent reporter acquisition does not indicate the ability to process and present antigens to developing T cells to promote central tolerance, somewhat reducing the impact of the findings. Altogether, this is a promising study that cleverly uses a variety of mouse models with flow cytometric analysis of recently identified TEC and DC subsets to delve into whether specificity of interactions between TEC and DC subsets could enable distinct DC subsets to contribute differentially to central tolerance induction against distinct types of self-antigens. However, the conclusions could be strengthened by additional analyses.

Reviewer #1 (Public Review):

The authors conducted continuum electrostatics, molecular dynamics and QM/MM calculations to probe the protonation states of two key amino acids, Asp234 and Glu68, in the anion channelrhodopsin (GtACR1). Previous spectroscopic experiments using different mutants of GtACR1 suggested that both residues are likely protonated, leaving open to the question what residues stabilize the protonated Schiff base. The current calculations strongly suggest that Asp234 is, in fact, deprotonated while Glu68 is protonated in the wild type. The protonation pattern changes in some mutants; for example, Glu68 is deprotonated in the D234N mutant, which helps explain why limited spectroscopic perturbation was observed in the mutant. The observed protonation patterns in the WT and mutant proteins also help explain how the Schiff base is stabilized in the relevant kinetic states, and how the photocurrents are affected by various mutations. Overall, this careful study helps highlight the value of detailed atomistic calculations in mechanistic analysis as well as providing explanations for spectroscopic measurements.

Reviewer #2 (Public Review):

This manuscript addresses the protonation state of two carboxyl acids, D234 and E68, in the active site of a channelrhodopsin. The active site is composed of a protonated retinal Schiff base which is positively charged. The authors find that D234, which is next to the Schiff base is deprotonated and E68 is protonated.

The strengths of this work is the broad range of computational methods ranging from membrane embedded molecular dynamics to QM/MM calculations. The simulations are supported by expressing a mutant that was not reported before and measuring its spectrum.

The weaknesses of this manuscript are that some methods are not state of the art and some conclusions are not justified. The method for the calculation of the absorption maxima is not accurate or might not even be physical. These computed absorption maxima are crucial for the conclusion of the paper. Also the gating mechanism is discussed on the basis of the structure of resting state. While the change of some amino acids is mentioned, the retinal is still in all trans and conformational changes of the protein are neglected in the discussion.

This anion channelrhodopsin is considered to be an important to tool in the field of optogenetics, where living cells can be controlled by light. Hence, a molecular level understanding of how this protein works could be highly valuable and could assist further tailoring of this channelrhodopsin for specific applications.

Reviewer #3 (Public Review):

This work probes the role of the acidic residues, E68 and D234, proximal to the Schiff base in the anion channel rhodopson from Guillardia theta (GtACR1). The authors present experimentally measured and QM/MM calculated absorption spectroscopic shifts for a new mutant, E68Q/D234N, as well as pKa estimates for several mutants. Their results further support the notion that Asp234 is deprotonate and Glu68 is protonated in the crystal structure and in the wild type channel, as previously proposed. They also suggest that the absorption spectra are not shifted in the D234N mutant because Glu68 deprotonates and rotates to stabilize the protonated Schiff base. This is potentially valuable both to the rhodopsin field and to the general domain of understanding the sometimes-nuanced influence of Asp to Asn or Glu to Gln mutations. The rotation of deprotonated Glu68 in the D234N mutant puts it in direct path with the ion channel, potentially explaining why photocurrent is abolished in this mutant.

This work would be improved by including E68(H)/D243(-) in their analyses to delineate the potential role of this intermediate in the wild type system. Many of the conclusions drawn could be similarly explained by this protonation state, thus direct comparisons are warranted. Additionally, error analysis is lacking for both measured and calculated quantities and must be added in order to verify several conclusions, including the proposed protonation states. The chosen level of QM should also be justified with higher level benchmarks for this system. Finally, delineating which findings/conclusions are new and which are supporting previous work should be clarified.

Reviewer #1 (Public Review):

Summary:<br> In " Rapid and Sensitive Detection of SARS-CoV-2 Infection Using Quantitative Peptide Enrichment 1 LC-MS/MS Analysis" Hober, A. et al. describe the addition of peptide immunoprecipitation by means of SISCAPA technology to the Sars-Cov2 mass spectrometry-based diagnostics toolbox. The work shows in a straightforward way that this is a huge improvement and of great importance to the field. It shows beyond any doubt that mass spectrometry can become a clinically applied diagnostic instrument to detect (viral) infection.

Overall remark:<br> The main concern is the reported number of 83% sensitivity. This is not because the number is too low, but because the number is misleading. In line with "CLSI EP 12-A2 User Protocol for Evaluation of Qualitative Test Performance guidance" a summary of the sample analysis results are shown in a 2x2 contingency table. Unfortunately, I oppose to this representation of the results at this stage for three reasons: (i) reporting a percentage shouldn't be done on less than 100 samples because of the weight of a few misannotated samples on these numbers, be it in the qPCR or the MS results; (ii) because both assays are imperfect, it is impossible to assess the ground truth for calling patients and thus assess sensitivity and specificity; (iii) the authors still only target a single peptide, which is not conventional in MS-based assays that targets proteins.

Rather than the proposed confusion matrix, which assumes that the ground truth is known to call it e.g. "false negatives", the authors could refer to it as an agreement matrix and not be tempted to calculate threshold values like sensitivity, which have too much of an impact on the clinical readership that is used to seeing this value in a more controlled context. This is in line with the recent Lancet manuscript from Fitzpatrick, M. et al (2021), proposing percent positive agreement (PPA) and percent negative agreement (PNA) instead (Fitzpatrick et al., 2021).

More specifically, as we and others have shown, qPCR Ct values rarely agree in two (consecutive) analyses, even within accredited settings (personal communication NHS). Above Ct30, patients regularly turned negative in our hands (https://doi.org/10.1021/jacsau.1c00048), even with an assay that had proven detectability of 1 plasmid at Ct40. Furthermore, we suspect that freeze-thaw cycles further inflate this uncertainty, two of which the current samples were subjected to. Undetected mRNA would then classify these patient samples as "false positives" if they did yield signal in the LCMS results. By chance, this did not happen in this manuscript, yet this could very well be the reason for the highest signal reported in Figure 3 as a green dot at log2 MRM response of -6 (see minor remarks).

The authors already distinguished the patients in a High Pool of Ct <30, a Low Pool 30{less than or equal to}Ct<33 and the negative samples (Ct>40). It is clear from the gap (no 34<Ct<39) that finding patients between Ct33 and Ct39 is challenging. Indeed, qPCR has its own "diagnostic grey zone" of LOQ negative and LOQ positive that rarely is being referenced. Thus, a "sensitivity" of 95% for patients <Ct30, despite the low number of samples and considering the uncertainties in qPCR (just above or below Ct30) at least limits the comparison to samples that are positive beyond any doubt. But again, we would be thresholding against a trembling metric, in turn making the claim from the authors dangerous that "the estimated LLOQ is 3 amol/μL approximates to Ct {less than or equal to}30". Rather, the Ct30 threshold should be set for a different reason, if one is chosen at all.

What is needed is good thresholding for clinical diagnostics, as is done in qPCR. In the public hospital in Belgium that provided us with patient samples, the positive threshold is set to Ct33 on the first measurement and practitioners use higher Ct values only in the context of physical symptoms of the disease to come to a final conclusion. For MS, we now need to measure >1000 samples in order to decide what log2 MRM response for a given set of peptides corresponds to an LOQ positive from - say - Ct27 to Ct30 and an LOQ Negative from Ct31 to Ct33. In other words, the linearity of the correlation between qPCR and MS illustrates the intrinsic value of MS; the point up until which we can provide clinically relevant information remains to be determined on large patient cohorts. In turn, these large patient cohorts can allow to sort (clinically) validated patients according to signal intensity and set a log2 threshold at which e.g. 2% or 5% negatives are expected, in line with False Discovery calculations for target decoy strategies. At this stage however, it might be most straightforward to conclude with percent positive agreement (PPA) and percent negative agreement (PNA), as is recommended for laminar flow tests validated on <100 samples.

Finally, realizing the importance of this pivotal moment in the implementation of MS in the clinic, I find it somewhat tricky to only focus on one peptide. In fact, the authors perform the qPCR on two genes (three genes being even more common) because of the drop-outs that can occur. I feel like the use of peptide IP with MRM for detecting pathogens has not yet matured enough to rely solely on one peptide. Still, I understand that asking for a second peptide would mean repeating all the measurements, so that is most probably not realistic. Yet, I do consider this to be yet another reason not to report % sensitivity and specificity in the current manuscript and the potential to gain robustness with more peptides should clearly be emphasized at every stage of the manuscript.

In conclusion, because patient batches in the thousands are currently unavailable to MS-oriented diagnostic labs and because of all the reasons mentioned above, we cannot report the numbers of sensitivity and specificity in this manuscript, as they are misleading and do not quantify what they are intended to do.

Fitzpatrick, M. C. et al. (2021) 'Buyer beware: inflated claims of sensitivity for rapid COVID-19 tests', The Lancet. Lancet Publishing Group, pp. 24-25. doi: 10.1016/S0140-6736(20)32635-0.

Major remarks:<br> P3L250: "on-column amount of 60 amol." Because of the enrichment procedure, could the authors specify what initial conditions they spiked into the dilution series prior to enrichment. This would allow recalculation and avoid confusion about the correctness of the 60 amol on column claim (which in our hands is still detectable).

P8L181: "50 μL elution buffer (0.5 % 180 formic acid, 0.03% CHAPS, 1X PBS) and incubated for 5 min at room temperature." This minor sentence is placed under major remarks, because in our understanding the elution buffer needs to be acidic and adding PBS will reduce acidity. If this is a typo, please correct. If this is not, could the authors try and use H2O instead and see if their results improve?

The access to the raw data was denied.

Reviewer #2 (Public Review):

MS-based proteomics is currently discussed as a method for detection of viruses from clinical samples. Several studies have already shown the potential of this method on the example of the detection of SARS-CoV-2 from respiratory specimens. However, one of the major drawbacks still remains the low sensitivity of MS-based virus detection compared to real-time PCR, which is the gold-standard method. In their manuscript Hober and colleagues apply specific antibody-based enrichment of SARS-CoV-2 peptides from upper airway samples to concentrate the analyte prior to analysis by targeted MS (MRM). The authors determined the dynamic range of the method for four different SARS-CoV-2 NCAP peptides using a calibration curve. On the example of the SARS-CoV-2 NCAP peptide AYNVTQAFGR a correlation between the MS result and the cT value is shown. Furthermore, using stable isotope labelled (SIL) peptides as internal reference, a quantitative MS measurement was achieved. The presented approach is able to distinguish real-time PCR SARS-CoV-2 positive samples from negative samples in the used set of 88 samples from asymptomatic patients. Combined with a specificity of 100 % and sensitivities of up to 94.7 % for samples with cT values {less than or equal to} 30 the authors conclude that the method could be an alternative to real-time PCR.

Strengths of the manuscript:

I think the applied technique (SISCAPA) is highly interesting in the context of virus proteomics. This is because virus proteins are often underrepresented in relation to the host proteins, especially during early time points of infection, hampering their detection. Recently, the application of SISCAPA for SARS-CoV-2 diagnostics has been suggested in the discussion of a manuscript from Van Puyvelde and colleagues. The manuscript from Hober and colleagues presents the first study demonstrating that this technique can be applied to enrich, detect and quantify SARS-CoV-2 peptides from upper airway samples. The manuscript is clearly arranged, the data is sound and supports the main conclusions.

Weaknesses of the manuscript:

I think the manuscript in some points underestimates the PCR and vice versa overemphasizes the proteomics approach. For example, I don't agree that real-time PCR generally suffers from technical problems, degraded probes or non-specific amplification. Vice versa I think the LC-MS/MS approach is not inherently absolute specific and does not outperform PCR in terms of specificity. Further, LC-MS/MS does not eliminate the problem of false positives, which could be introduced during sample preparation or by inter-run contaminations. Although in real-time PCR no internal standards analogous to isotopically labelled peptides are used there are internal controls used to assure the quality of the extraction and the PCR reaction itself. The method presented by Hober and colleagues is clearly beneficial for the field of proteomics-based virus detection, but I suggest a more balanced discussion also including also the potential drawbacks of the method.<br> Another point I like to raise is that the authors conclude at the end of the results section that patient samples were collected at an infectious stage. However, an assessment of the infectivity cannot be drawn from the presented data. The analysis of real-time PCR results in the manuscript is based on cT values. But to draw the conclusion, that the analysed samples contained infectious virus particles, the number of viral genome equivalents has to be determined, which in turn can be correlated to infectivity. The detection of viral proteins itself does not proof that samples were collected at an infectious stage and there is currently no correlate of the amount of NCAP protein and infectivity. Since viral proteins are likely more stable than viral RNA, they could even be detectable for a more prolonged time in patient samples.

Reviewer #3 (Public Review):

Major comments

P2, l245, Figure 2: It is not completely clear to me what is represented in panels A and B. Is this the pure SIL peptide of the endogenous peptide in a complex matrix? This may make a large difference for the determination of the LLOQ. Panel B shows a calibration curve and as these are curves for which the signal is detected based on known input amounts of sample, I assume that the input is pure SIL peptide here?

In panel A, what does '3 amol/ul' in the middle chromatogram exactly mean? Is this the endogenous peptide that was calculated to be present at 3 amol/ul based on a known concentration of spiked-in SIL peptide?

P4, l276: The authors need to explain the details of data imputation. It is unclear which data were imputed and how this was done. In Figure 3 the grey data points represent "not detected" or "inconclusively identified" samples by LC-MS, while some of the data points seem to have a higher 'response' values than others. Please explain.

In Figure 3, how is 'response' defined? I don't understand the following sentence (p4, l277): "... for the LC-MS results the lowest response divided by three was used, mimicking....". Which variable does the data point size reflect? There seem to be clear differences in ball sizes. Please explain. For clarity, it would be advisable to keep the y-axes for panels A and B identical. Also, how could RT-PCR values be not obtained, apparently leading to missing Ct values (p5, l278)?

Assuming that all collected samples from individuals in the test group in this study are visualized in Figure 3, the majority was tested positive for SARS-CoV-2. This is very different from the percentages oberserved in regular testing facilities. How was the study group composed? Were these individuals who were already admitted to the hospital? It would be interesting to include more negatively tested individuals to see the distribution of 'MRM response' values in this group, since some of the negatively tested individuals (green data points) show higher than expected MRM response values if no viral protein is present at all. Related to this, I do not understand how a specificity score of 100 % (p5, l292) was obtained while some green data points (negative by RT-PCR) have higher associated MRM response values than some of the blue (positive by RT-PCR) samples. Can the authors explain this?

I find the text from p6, l298 ("However...") onward more suited for the Discussion section, since this is about the interpretation of the results presented here and the use of the described methodology in diagnostics; no results are shown in this part.

Reviewer #1 (Public Review):

This is an interesting manuscript in which Fox-Fisher et al report a novel method for minimally invasive monitoring of the dynamics of the human immune system, based on profiling of cell-type specific cell-free DNA methylation marks. This method implements targeted DNA sequencing of a small panel of cell-type specific CpG markers and has the potential to be useful for the monitoring of a wide range immune-related diseases. The authors have tested the utility of this assay in three settings: influenza vaccination, Eosinophilic Esophagitis and B cell lymphoma. The experiments are well described and well powered, and the results are compelling. Nonetheless, this submission can be improved in several areas, most importantly, i) the presentation of the data in some of the Figures, ii) the statistical analysis of B cell dynamics after vaccination in Figure 3, iii) the discussion of the challenges for diagnostic testing related to the significant inter- and intra-individual variation in immune cfDNA levels observed, and iv) the lack of discussion of other immune monitoring methods that have previously been described in the scientific literature or that are used in diagnostic medicine.

Reviewer #2 (Public Review):

The authors try to dissect the contributions of various types of immune cells in cfDNA. They have mined cell-type-specific DNA methylation biomarkers for 7 common immune cells and validated their assays in 3 clinical scenarios, arguing that the signal in cfDNA could provide better assessment of the immune response in the body than cell counts.

B-cells also have lots of subtypes which could be very different from each other (e.g., plasma B cell vs naive B cell), the authors did not differentiate them while 2 of the clinical applications in the article are closely related to B-cells. Therefore the results need to reinterpreted using more precise measurements.

Reviewer #1 (Public Review): 

There is a critical need for new methodologies to study the physical properties of biomolecular condensates in living cells under normal and pathological conditions. To address this, Schlüßler et al innovatively combined Brillouin microscopy with Optical Diffraction Tomography (ODT) and epi-fluorescence imaging. The current study can have a significant impact on the community. A major strength of the study resides in the application of Brillouin microscopy, which offers a label-free and nondestructive approach to investigate the complex viscoelastic behavior of biological materials. The study initially attempts to benchmark their new methodology using control samples, called cell phantoms. Subsequently, the authors apply their new method to study the physical properties of biological materials including nucleoplasm, cytoplasm, phase-separated organelles, and adipocytes. The results are largely convincing and offer interesting insights into the complex material properties of these subcellular fluids and organelles.

Reviewer #2 (Public Review): 

The multimodal instrument presented here provides an independent measurement of the spatially dependent cellular refractive index, which yields a more quantitative extraction of the longitudinal modulus from Brillouin spectroscopy. To my knowledge, this instrument is unique, and its capability addresses unresolved problems in Brillouin studies. The method was judiciously validated on standard samples. The experiments and analysis were carefully performed, and the statistics seems solid. The manuscript is very well written and clear.

The results highlight some discrepancies between the generally accepted assumptions regarding the cell density and refractive index. One striking example is the finding that the nuclear matter exhibits lower mass density but higher longitudinal modulus. Using the fluorescence channel for specificity, the authors investigated successfully other cellular compartments. 

While the objectives of the study seem to have been achieved, I wonder how large an impact this development will have in the field. At the end of the day, the method yields the longitudinal modulus at GHz frequencies. Cell mechanics is indeed very important, but at much lower frequencies. For example, actin filament lifetime is of the order of minutes. It seems very difficult to infer cell mechanics information relevant to its function, from the GHz range, as the dispersion of the material is unknown.

Reviewer #3 (Public Review): 

In the manuscript by Kim et al, the authors present a combined optical system, termed FOB microscopy which bring together the epi-fluorescence, optical diffraction tomography (ODT), and Brillouin microscopy. The main purpose of FOB is to establish a colocalized measurement of Brillouin shift and the refractive index (RI) to calculate absolute densities of biological sample, especially the biomolecular condensate. 

The major strength of this paper (method development) is the added measurement of ODT which can correct for and thus provide a more precise RI and density of a given sample. If the RI and densities can be reliably measured in in vitro and cell samples, it will be a great tool that can complement other microrheological measurement. The major weakness is the lack of appropriate controls and lack of comparisons to other conventional methods (microrheological), which together lead to questionable outcomes from various measurements shown throughout the manuscript. Another concern is the pervasiveness of the method which involves excessive level of illumination and vibrational excitation. 

If the work could be revised to present careful calibration with samples that are pertinent to biological systems (both in vitro and cellular) and make a comparison to other conventional methods in every possible case, the strength and limitations of the combined microscopy will be clear, making it very helpful for the researchers in the field.

Reviewer #1 (Public Review): 

Meyer and Benoit provide an elegant and timely approach to understanding the brain mechanisms that underlie suppression-induced forgetting. Using a modified version of the "Think/No-Think" paradigm, the authors explored memory suppression across four study phases: pre-test, suppression phase and post-test. Objects were presented with aversive scene backgrounds and participants studied object-scene associations to a criterion level of 60% accuracy. In the MRI scanner, participants were required to recollect ("think') or suppress ("no-think") the aversive scene associated with the presented object. Results indicate that memory suppression was associated with increased activation of the dorsolateral prefrontal cortex and deactivation of the parahippocampal cortex and hippocampus. Additional analyses indicated that the reduction in memory vividness due to memory suppression was associated with deactivation of the parahippocampal cortex. 

The brain mechanisms enabling us to selectively suppress unwanted memories remain unclear, particularly over longer durations. As such, this study addresses an important topic. There are a number of strengths to this study. By testing memory suppression in the scanner after the initial encoding phase, the authors could identify patterns of brain activation and deactivation when memories were successfully recalled versus successfully suppressed. Importantly, one third of the objects were not shown during the fMRI phase, providing a control baseline to assess how memories weaken in general due to the passage of time. Use of representational similarity analysis further permitted the authors to determine whether the neural activity pattern before and after the suppression phase was statistically different. 

Overall, these findings offer new insights into the processes by which the neural activation for suppressed memories is dampened and how such altered neural representations relate to markers of memory quality (e.g., vividness). 

One aspect that was not discussed is how increased activation of the dorsolateral prefrontal cortex relates to individual differences in cognitive control. It may be that some individuals are more adept at inhibiting unwanted memories and how such individual differences relate to mental imagery, emotion regulation and cognitive control would be important to consider.

Reviewer #2 (Public Review): 

In this paper, the authors examine whether suppression of memory retrieval, as implemented with the Think/No-Think paradigm, results in a degradation of a memory trace. They utilize two multi-voxel techniques, pattern classification and representational similarity analysis (RSA), to examine the nature of the representation of memorial information (rather than just examining the degree of activity in a given brain region related to memory). Furthermore, their investigation determines whether the degradation of the memory representation, as indexed by the information derived from the multi-voxel techniques, is related to reported changes the vividness of the memory. This latter approach is laudable in that it attempts to link their neural measure of a memory's robustness to an index of behavior. 

The methods, approach and statistical analayses that the authors take seem quite sound. They use an object-aversive scene pairing for the cue-target pairing in the Think/No-Think task. They obtain vividness ratings of the target after individuals learn the pairings, and then after they have manipulate the target in their mind, either to Think about it, (i.e., retrieve it) or to Not Think about it ( i.e., suppress it). Finally, a task is given in which individuals see a variety of different aversive scenes as compared to morphed versions of the scenes, which allows them to create a classifier to determine the degree to which scene processing is affected by the Think and No-Think manipulations.. 

The first set of findings are confirmatory, indicating that the typical regions that are involved in retrieval suppression becomes activated during the No-Think condition (e..g, lateral prefrontal cortex). They also find that people report memories that were suppressed are less vivid than either those that were retrieved or the control items that were not mentally manipulated. However, in a separate sample of individuals who did a behavior only version of the task, both the control and suppressed items were rated less vivid. Given the demand characteristics (i.e., individuals who were told to Not Think about items, might be inclined to rate them as less vivid to please/go along with the experimenter), these data are not the most convincing. 

The more interesting results come from the multi-voxel approaches. First, it was found the classifier fit for scenes increased with increased vividness rating, suggesting that demand characteristics are not driving the results. With regards to the classifier fits for scenes in general, it is found that the classifier fit is reduced both during active suppression (i.e., during a No-Think trial) and during the subsequent post-test than is observed for baseline trials. Moreover, those participants who showed a greater suppression of scene information during No-Think trials also showed a reduced report of vividness. Much of these results focus on the parahippocampal gyrus, which is known to process scenes. These findings support the idea that retrieval of scene information in general is impacted, and is consistent with prior work that retrieval suppression involves a general (rather than a more specific) inhibition of temporal lobe memory-related regions as to inhibit retrieval. 

Probably the most novel part of the study involves using representational similarity analysis to examine the degree to which a specific scene (rather than scenes more generally) is suppressed and the relationship of such suppression to the post-test vividness rating of the item. Here the results are marginal which makes an interpretation difficult. Had the effect been found it would have suggested an item-specific effect that has not been observed previously. 

Overall, this is a solid study with interesting results that expand our knowledge of the neural mechanisms that support the retrieval and suppression of memories.

Reviewer #3 (Public Review): 

This fMRI study uses classification analysis and representational similarity analysis (RSA) to test the hypothesis that memory suppression sustainably deteriorates the neural representations of previously suppressed memories during their subsequent retrieval. In this study, 33 participants repeatedly suppressed memories of aversive scenes during a Think/No-Think (TNT) paradigm. Before and after the TNT phase, participants also had to recall the scenes and to rate the vividness of their recollection (pre- and post- tests). Classification analysis using the decoding toolbox (Hebart et al., 2015) was computed to distinguish between intact aversive scenes and their morphed versions. The weight pattern obtained was then used to quantify the degree of reactivation during pre-, post- test and suppression phase of the TNT. RSA was performed to obtain the similarity of each scene with itself (same-item) and with the average of the scene with the other scenes (different-item) of the same category (i.e., Think, No-think or Baseline). The difference score between same-item and different-item similarity enabled to assess whether the retrieval of a given scene is associated with similar neural activity pattern before and after suppression (i.e., reinstatement of its unique representation). Results evidenced that the process of memory suppression rendered the memories less vivid and diminished the reactivation of the scenes both globally across the brain and locally in the parahippocampal cortices. The decline in vividness was associated with weaker reinstatement of memory representations in the parahippocampal cortex. These results support the hypothesis that suppression sustainably reduces the neural reactivation of memory traces. 

The conclusions of this paper are well supported by the data. 

One of the strengths of this study is that it does not only examine whether there is less reactivation of memory traces during suppression attempts (as previously shown), but whether this effect maintains during the subsequent reactivation of the previously suppressed memories. This question is tested using innovative multivariate pattern analyses that are not traditionally used in the Think/No-Think literature and that may provide new promising approach for the analysis of such data. Moreover, RSA investigates the neural reinstatement of unique representations, rather than more global category of items (i.e., Think, No-Think and baseline categories). 

Although multivariate methods are promising tools in neuroimaging, the specific advantages of the methods used in this paper, in comparisons to more classical methods (e.g., difference in activation at pre- vs post- tests), are not sufficiently highlighted. 

The approach used in the manuscript do not allow to have a full picture of the neural network involved in the reduction of scene reactivation at the post-test. Authors focused a priori on the parahippocampal cortex for its involvement in memories for complex scenes and in mnemonic vividness. However, other regions such as the hippocampus or the amygdala also play an important role. Concerning the hippocampus, the authors speculate that the hippocampal representations may be largely protected from interference (discussion, page 11), suggesting a different disruption process of these representations in comparison to the parahippocampal cortex. Amygdala is crucial for the type of material used (i.e., aversive scenes from the IAPS) in this study. From a clinical perspective, decreasing the reactivation of the amygdala is important as it should subsequently decrease the negative emotional content of the memory trace. Despite the interest of the hippocampus and amygdala, these regions were not included in the analyses. Focusing on a single region or more globally across the whole grey matter (without distinguishing the involved brain regions) prevents to make a more integrated discussion on the neural mechanisms that may help to sustainably reduce memory reactivation. 

At the end of the experiment, participants had to recall the 48 object-scene pairs to investigate whether suppression during the TNT phase induced forgetting. However, results did not evidence significant suppression-induced forgetting at final recall (Appendix 3). Authors suggest that the additional retrieval tests before and after the TNT may have reduced the forgetting. Additionally, the classifier training task (aversive scenes used), presented before the final recall, may also have introduced interference, and dampened the differences between the conditions (Think, No-think, and baseline). Consequently, the authors cannot rely on this objective measure of forgetting but can only rely on the subjective measure of vividness. This limitation is not sufficiently discussed in the manuscript. 

Although the representational similarity analysis was promising to evidence the reinstatement of unique representations, it did not evidence a significant reinstatement for suppressed memories, and it was not weaker than for baseline memories (Figure 4b). This suggests a lack of power for such neural measure.

Reviewer #1 (Public Review):

'Presynaptic Stochasticity Improves Energy Efficiency and Alleviates the Stability-Plasticity Dilemma' by Schug et al moves energy efficiency questions of stochastic synaptic transmission that were asked at the level of the single synapse and the single cell to the network level. This is important since local advantages in terms of energy cost may have unknown consequences at the larger scale. And stochastic synapses may have an unknown advantage in learning paradigms at the network level.

I have some concerns regarding this work

(1) The considerations are made in one/two particular network architecture with one parameter combination. The generality of the conclusions is not given and there is no reason to believe that the observations made here will hold for other network architectures or even different parameters. In this way, the current manuscript seems to describe the beginning of a project that hasn't really been worked through.

(2) Additionally, the network architectures used here are rather artificial (multilayer perceptron) and come from machine learning. Linking a physical measure in a biological system (the metabolic cost) with task solving in a machine learning setting that does not have a biological pendant seems far-fetched and would not be the first thing in my mind to do to study the information transmission in biological neuronal networks.

(3) A lot of different measures for efficiency of the network are all briefly addressed but not dissected properly. A more fundamental understanding of why and when stochastic synapses in the network might be useful is missing and seems rather unexplored apart from some select manipulations.

To sum up, I think that the question is interesting but the work is yet premature.

Otherwise, the paper is very well written and puts this work very nicely in context with the existing literature.

Reviewer #2 (Public Review):

In this study, every synapse is described by the probability of release (p) and the synaptic strength (m). The learning dynamics for the expected synaptic strength (p*m) follows a classical gradient-based rule while the learning rule for p is such that important synapses (in the Fisher Information sense) increase p while the others decrease p.

Strengths

One of the biggest strength of the proposed learning rule is its simplicity. It is remarkable how a simple learning rule for the probability of release (in addition to the classical gradient rule for the expected synaptic strength) can help in continual learning problems. In some sense, the authors leverage the fact synapses are described by 2 variables in order to keep important memories for a longer time. The simplicity of the learning rule makes it very attractive for practical implementations.

Another strength of the paper is that the learning rule for p has some relationship to the Fisher Information giving thereby a (partial) justification for this rule.

Finally, the energetic considerations are very welcome in a field where those aspects are too often neglected.

Weaknesses

Probably the strongest weakness of the manuscript is the lack of biological evidence for the proposed learning rule.

To my opinion, the second weakness of the paper is that the learning rule for p is more of an heuristic than principally derived. I understand the benefit of its simplicity (as argued above), but the price to pay is that we are not guaranteed that P(g_i^2>g_lim^2) does always depend monotonically on E(g_i^2).

Overall, I find the paper well written and the main claims are well supported by the data/analysis.

Reviewer #2 (Public Review):

This paper by Patel and Matange focuses on understanding the evolutionary response of E. coli cells to the antibiotic trimethoprim (TMP). Mutations in the gene folA, which encodes the dihydrofolate reductase (DHFR) enzyme - an established target of TMP, are known to mediate intrinsic resistance to TMP. This work shows that de-repression of the PhoQ/PhoP pathway via inactivation of MgrB, which is a negative feedback inhibitor of PhoQ, leads to TMP tolerance. Further, this response to TMP is due to the upregulation of DHFR expression. Similarly, inactivation of MgrB and a corresponding increase in the PhoQ/PhoP-regulated gene expression is a prominent mechanism for acquired colistin resistance in clinical isolates of Klebsiella pneumoniae. The authors performed adaptive laboratory evolution of E. coli under TMP selection and identified factors contributing to the transition of TMP-tolerant bacterial cells to TMP-resistant cells. At high TMP concentrations, the cells become resistant via mutations in DHFR. Cells evolved under low (sub-MIC) concentrations of TMP develop mutations inactivating the RpoS sigma factor to offset the cost of PhoQ de-repression. The data presented in the paper are clear and the conclusions are mostly valid. However, the authors need to modify parts of the main text and figure presentations for clarity and a better/more straightforward interpretation of the results by the reader. In general, the results obtained in this study explain well the evolutionary consequences of these mutations and the pathway for the acquisition of the mutations.

1) The authors find that mutations in the mgrB locus precede mutations in folA during E. coli's response to TMP. Why only sequence 5 of the 10 TMPR mutants? Was this subset chosen for sequencing based on any specific criteria? Below are some follow-up comments.

a. Do any of the mutations cause growth defects relative to the wild-type strain?

b. Line 103: What are the mutations in folA promoter region? Only mutations in the coding sequence are listed in table 1 and figure 1A.

c. Line 109: The authors speculate that IS-element insertions in the mgrB promoter region reduce its expression, maybe they can provide a reference here from previous studies that have analyzed such mutations. Also, including details of the length/size of these insertion elements within table 1 would be helpful.

d. Line 111: the phrase "stop-codon readthrough" is misleading. The authors should rephrase to clarify that the single nucleotide deletion leads to a shift in the reading frame leading to an altered protein sequence at the C-terminal end.

2) Based on growth assays including competitions, and measurements of folA gene expression in mgrB-deficient E. coli cells, the authors conclude that tolerance to TMP is caused by PhoP-dependent upregulation of DHFR.

a. The authors should rewrite the text (lines 143-155) to make the experimental design of the competitions more obvious to the reader. Indicating either within the figure legend or main text what ∆mgrB/total means would definitely make analysis of the figure and results easier for the reader The reader needs to go to the materials section to get a full understanding how exactly this experiment was performed.

b. In Figure 1C, the IC50 value for ∆phoP is similar to that of wild type. If PhoP-dependent expression of folA important for TMP tolerance/resistance, shouldn't we expect to see a lower IC50, similar to that of ∆mgrB∆phoP? Intriguingly, the data for wild type in Figure 1C appears to be in conflict with the data in Figure 3B, please clarify.

c. In Figure 1D, it is hard to figure out the exact strains and conditions of each competition. For instance, the ratios 10:1, 100:1 and 1000:1 needs to be clearly labeled, "wild type: mgrB" or "wild type: specific mutant" as applicable, the label on the X-axis is misplaced. Does "WmgrB" refer to ∆mgrB? If yes, change to ∆mgrB. Fitness values need a label or put into a table.

d. Line 172: incorrect figure citation, replace Figure 2B with 2A.

e. Lines 180-181: Only 5 out of the 10 TMPR isolates were sequenced and found to have mutations in the mgrB locus. In the absence of sequencing data confirming such mutations in TMPR 6-10 isolates, the increased levels of DHFR cannot be attributed to loss of mgrB.

f. In Figure 2C, it would be helpful to show the GFP fluorescence data for the single deletions, ΔphoP and ΔrpoS, to further support the claim that TMP tolerance via DHFR upregulation is PhoP dependent. In addition, the X-axis should specify the promoter reporter that was used.

g. Lines 181-183: reference for the previous work on W30G folA is missing.

h. In Figure 2, there is a discrepancy in the level of DHFR observed for both TMPR2 and 3 isolates in panels D and E - the DHFR protein levels are much higher in panel E. Can the authors explain this discrepancy, especially given the W30G mutation in TMPR3 (expected to show reduced levels of DHFR)? Is the same amount of protein loaded in both experiments? If so, why are the levels of protein different (and vastly different for TMPR3)? Better quantification of the western blots depicting the signal for the replicates would be helpful.

3) The data presented here also show that mgrB and folA mutations act in synergy in TMP resistant E. coli.

a. It would be useful to the reader to include a table listing the MIC values in Figure 3. The plate images showing the E-tests are difficult to read and less helpful in interpreting the MICs and can be moved to the supplement.

b. In Figure 3E (and lines 234-238), what was the strain background used for DHFR overexpression? The details are missing from the paper.

4) To follow the adaptive pathway for TMP resistance, the authors sequenced genomes of TMP-resistant isolates.

a. Line 283: How many strains were sequenced at each time point? "3 to 5" is confusing.

b. In Figure 4, the data points/symbols and lines are hard to read in both panels A and B. These graphs can be replotted with open symbols or different colors to help the reader analyze the figure much more easily.

c. Overall, it is still unclear how folA expression is regulated by PhoP regulation. An alternate hypothesis is that loss of MgrB may influence folA gene expression in a PhoP independent manner. Have the authors ruled out this possibility?

Reviewer #1 (Public Review):

This paper looks at the evolutionary trajectory of E. coli populations exposed to the antibiotic trimethoprim. The authors argue that mutations in mgrB are initially selected since these lead to an increase in folA expression. At high drug concentrations, mutations in folA are seleted, but at lower drug concentrations, mutations in rpoS are selected since these compensate for the growth defect of an mgrB mutant. Overall, the authors conclude that the evolution of resistance to trimethoprim first involves the accumulation of mutations outside of folA that modulate gene expression, and that the evolutionary path to resistance depends on the concentration of drug.

The approach taken by the authors is very thorough, and the conclusions are well supported by the data. I think this is an important contribution to the field, and I have only a few specific comments:

– The authors should sequence the mgrB gene and upstream sequence, and the rpoS gene for TMPR6-10. If these strains don't have mutations in mgrB, I think it's important to sequence their genomes to find out why DHFR levels are higher than in wt cells.

– Presumably the higher number of mutations in mgrB rather than folA reflects the mutational space available, i.e. there are more possible mutations that reduce mgrB expression than there are gain-of-function folA mutations. This is worth mentioning, since it has a big impact on the evolutionary path to resistance.

Reviewer #3 (Public Review):

The study by Patel and Matange explores how trimethoprim resistance evolves in E. coli in vitro, with a focus on the role of gene regulatory networks in the evolution of antibiotic resistance. In vitro antibiotic resistance evolution is often studied through a "one-hit" model, wherein antibiotic pressure selects for mutations in a single target gene. Here, the authors explore how different levels of antibiotic selection result in different evolutionary "choices" made by the bacterial population. They also show that mutations affecting the PhoPQ regulatory system decrease antibiotic susceptibility (i.e. increased IC50) but do not cause overt resistance (i.e. increased MIC), and that these mutations are an early adaptation under antibiotic selection.

Major strengths of the study include the clearly-planned and executed experiments, the use of multiple replicates to establish the reproducibility of the observed mutations, and the employment of relevant functional assays to confirm hypotheses generated by the comparative genomics analyses performed. Weaknesses are generally minor, and include an occasional lack of justification and sufficient explanation for some experimental design choices, as well as omission of a few easy-to-perform experiments that would further support the authors' conclusions.

In my view, the authors have achieved their aims and their results largely support the conclusions that are drawn. Overall, this study makes a meaningful contribution to our current understanding of the role of the PhoPQ system in the evolution of antibiotic resistance in E. coli, and with minor modifications the study would be of significant interest to researchers studying in vitro evolution of antibiotic resistance in bacteria.

Reviewer #1 (Public Review):

Overall, this study is well designed with convincing experimental data. The following critiques should be considered:

1. It is important to examine whether the phenotype of METTL18 KO is mediated through change with RPL3 methylation. The functional link between METTL18 and RPL3 methylation on regulating translation elongation need to be examined in details.

2. The obvious discrepancy between the recent NAR an this study lies in the ribosomal profiling results (such as Fig.S5). The cell line specific regulation between HAP1 (previously used in NAR) vs 293T cell used here ( in this study) needs to be explored. For example, would METLL18 KO in HAP1 cells cause polysome profiling difference in this study? Some of negative findings in this study (such as Fig.S3B, Fig.S5A) would need some kind of positive control to make sure that the assay condition would be working.

3. For loss-of-function studies of METLL18, it will be beneficial to have a second sgRNA to KO METLL18 to solidify the conclusion.

4. In addition to loss-of-function studies for METLL18, gain-of-function studies for METLL18 would be helpful for making this study more convincing.

Reviewer #2 (Public Review):

The manuscript by Matsura-Suzuki et al. characterizes the role of METTL18 histidine methyltransferase in protein synthesis. The authors used genetical manipulation, affinity purification and mass spectrometry to indicate METTL18 protein as methyltransferase that specifically modifies ribosomal protein RPL3 during ribosome maturation. Using METTL18 based in vitro methylation system they annotate His245 residue in RPL3 protein as a methylhistidine and confirmed their results using a mass spectrometry on immunopurified RPL3 protein from HEK293 cells. The authors use ribosome profiling techniques, luciferase aggregation assays and mass spectrometry analyses of cellular aggregates to argue for Tyrosine specific effects during protein synthesis that influence protein folding.<br> While METTL18 was recently characterized as a RPL3 specific histidine methyltransferase in HAP1 cells by another group (Malecki et al., 2021). The previously published study also pointed out on the role of this modification in ribosome biogenesis, general translation and GAA codon specific effects, however the study by Matsura-Suzuki et al., argues for Tyrosine specific effects and impact of RPL3 His245 modification on proteostasis maintenance which would clearly distinguish these two studies and importance of the METTL18.

Reviewer #3 (Public Review):

In this article, Matsuura-Suzuki et al provided strong evidence that the mammalian protein METTL18 methylates a histidine residue in the ribosomal protein RPL3 using a combination of Click chemistry, quantitative mass spectrometry, and in vitro methylation assays. They showed that METTL18 was associated with early sucrose gradient fractions prior to the 40S peak on a polysome profile and interpreted that as evidence that RPL3 is modified early in the 60S subunit biogenesis pathway. They performed cryo-EM of ribosomes from a METTL18-knockout strain, and show that the methyl group on the histidine present in published cryo-EM data was missing in their new cryo-EM structure. The missing methyl group gave minor changes in the residue conformation, in keeping with the minor effects observed on translation. They performed ribosome profiling to determine what is being translated efficiently in cells with and without METTL18, and found decreased enrichment of Tyrosine codons in the A site of ribosomes from cells lacking METTL18. They further showed that longer ribosome footprints corresponding to sequences within ribosomes that have already bound to A-site tRNA contained less Tyrosine codons in the A site when lacking METTL18. This suggests methylation normally slows down elongation after tRNA loading but prior to EF-2 dissociation. They hypothesize that this decreased rate affects protein folding and follow up with fluorescence microscopy to show that EGFP aggregated more readily in cells lacking METTL18, suggesting that translation elongation slow down mediated by METTL18 leads to enhanced folding. Finally, they performed SILAC on aggregated proteins to confirm that more tyrosine was incorporated into protein aggregates from cells lacking METTL18.

The article is interesting and uses a large number of different techniques to present evidence that histidine methylation of RPL3 leads to decreased elongation rates at Tyrosine codons, allowing time for effective protein folding. I agree with the interpretation of the results, although I do have minor concerns:

1. The magnitude of each effect observed by ribosome profiling is very small, which is not unusual for ribosome modifications or methylation. Methylation seems to occur on all ribosomes in the cell since the modification is present in several cryo-EM structures. The authors suggest that the modification occurs during biogenesis prior to folding and being inaccessible to METTL18, so it is unlikely to be removed. For that reason, I do not think it is warranted to claim that this is an example of a ribosome code, or translation tuning. Those terms would indicate regulated modifications that come on and off of proteins, but the authors have not presented evidence that the activity is regulated (and don't really need to for this paper to be impactful).

2. In Figure 4-supplement 1, it appears there are slightly more 80S less 60S in the METTL18 knockout with no change in 40S. It might be normal variability in this cell type, but quantitation of the peaks from 2 or more experiments is needed to make the claim that ribosome biogenesis is unaffected by METTL18 deletion. Likewise, the authors need to quantitate the area under the curve for 40S and 60S levels from several replicates and show an average -/+ error for figure 3, supplement 1 because that result is essential to claim that ribosome biogenesis is unaffected.

3. The effect of methylation could be any step after accommodation of tRNA in the A site and before dissociation of EF-2, including peptidyl transfer. More evidence is needed for claiming strongly that methylation slows translocation specifically. This could be followed up in vitro in a new study.

Reviewer #1 (Public Review):

The authors employ CRISPR/Cas9 technology to create 2 knock-in mouse lines harboring separate pathological variations in the HCN1 gene, p.G391D and p.M153I, that have been identified in human epilepsy. The goal was to develop preclinical models that recapitulate key aspects of human disease and provide a novel resource for studying mechanisms and evaluating new therapies for epilepsy.

The authors largely achieved their aims. The key strengths of this work were the expert team assembled to execute the project, the successful use of novel CRISPR/Cas9 technology to generate new mouse lines for the study, and the use of cutting-edge techniques to study the molecular and physiological consequences of the genetic variants in the brain. A "hidden gem" in the work is an exploration of whether lamotrigine directly enhances HCN function and finding it did not. While an important negative result, this was not demonstrated in native tissue, leaving the question open regarding direct effects on the native channel in neurons.

One weakness of the study is the data from the set of experiments exploring impact of overexpression of the variants in neurons. This technique can be highly variable and the data interpretation in this case would benefit from more rigor. There are minor questions about statistical methods for comparing and concluding about the significance of differences between some experimental groups.

An important conceptual gap remains unanswered by the study. Given the phenotypic similarities between patients with sequence variation in Na+ channel and HCN genes, as well as evidence of reduction of other channels or pumps in this case and the strong co-localization of Na+ channels and HCN channels in the PV+ neurons thought critical in the epilepsy of the HCN sequence variants, is it possible that Na+ channels are impacted as a secondary effect of HCN channel dysfunction here?

Overall, the impact of this work to the field will be important. The personalized medicine framework here is a good example of how the field will in the future address diseases caused by distinct sequence variation. The choice to generate mice in which phenotypes analogous human variations were replicated in multiple patients and the phenotype was severe was logical and a good example to others about increasing odds of success in approaches like this....

The work is also important in connecting prior in vitro work, in which changes in channel function in divergent directions predict certain in vivo consequences, to the context of the animal in vivo, where predictions can be tested, and in this case, leading to surprising results. Indeed, these results point in new directions for understanding new aspects of channel "function" such as trafficking and targeting to subcellular domains in neurons.

Reviewer #2 (Public Review):

The authors have created transgenic mouse models incorporating single-nucleotide sequence variations of the HCN1 gene known to cause severe early infantile epileptic encephalopathy in early life in humans. The human syndrome includes severe, drug resistant epilepsy as well as developmental regression and delay.

The study makes some important contributions to our understanding of HCN1 EIEE. First, the two genetically-altered mouse lines harbor an epileptic condition as well as excess mortality similar to that seen in human patients with similar genetic changes. Thus, these mice will be a useful platform for future discovery in understanding the causes of epilepsy. It is not clear whether the mouse equivalent of the severe developmental disability seen in humans was present in mice.

Second, the authors perform some key antiepileptic drug treatment trials in vivo demonstrating that phenytoin and lamotrigine exacerbate seizure frequency whereas valproic acid does not. In this response the mouse models resemble the differential drug responses seen in another human EIEE, Dravet syndrome. Although not a clinical trial, this information will be useful to clinicians encountering children with HCN1-related EIEE.

The authors performed a substantial amount of experimentation to show the consequences of HCN1 mutation on the ion channel's subcellular localization and on neuronal excitability. The results are complicated. It does appear that the subcellular localization of HCN1 channels is disrupted in both pyramidal neurons and interneurons, presumably causing loss of function. Also, current clamp recordings of hippocampal pyramidal neurons suggest some loss of HCN1-mediated contributions to their passive membrane properties, but these effects are not entirely consistent with sole mediation by HCN1 channels. This suggests that there are changes to other ionic conductances as well.

There are some significant limitations to interpretation of this study. First, there is no demonstration of hyperexcitability at a cellular or network level, so we do not know how HCN1 mutation predisposes to seizures. In fact, hippocampal pyramidal neurons were shown to be hypoexcitable, at least to one method of action potential generation. There is a suggestion that parvalbumin-positive interneurons may be affected, but there is no evaluation of their excitability. It is possible that HCN1 mutation is directly causing neuronal hyperexcitability, but this would only be uncovered by studying HCN1 channel effects on pyramidal neuron dendrite excitability (where they are mostly localized); synaptic function; or on interneuron excitability.

There is also no direct demonstration of the effects of channel mutation on HCN1-mediated current (Ih) in native neurons, so we cannot assess how channel biophysics is altered.

Reviewer #3 (Public Review):

This is an innovative and important paper that demonstrates the power of experimental models to advance our understanding of human disease. The authors focus on early-life epilepsy, a devastating and common disorder, and specifically on genetic epilepsies generated via de novo mutations in the hyperpolarization-activated nonspecific cation (HCN) channels subtype 1. They delineate the epileptic phenotype and demonstrate some of the potential mechanisms leading to the generation of spontaneous seizures in genetically engineered mice.

The key strengths of the paper are:

a. Starting off with human mutations known to cause disease, generating knock-in mice, and recapitulating the human phenotype, thus creating a powerful platform to study mechanisms which, in turn, will inform human therapies

b. Identifying plausible mechanisms, including pronounced alterations in the levels and distribution of HCN1 protein, including disrupted targeting to the axon terminals of basket cell interneurons.

c. Probing the functional consequences of the mutation on the properties of the HCN1 channels, discovering a clinically relevant paradoxical seizure induction by anticonvulsant drugs targeting sodium channels (lamotrigine and phenytoin).

d. Identifying a novel pharmacology of the mutated channels, which are unresponsive to classic antagonists. This supports the need to screen mutated channels to identify novel compounds that might overcome the mutation-induced functional deficits of the channels. In other words, the authors make a strong case for employing tailored therapies for specific channel gene variants.

While overall, the work is rigorous and the majority of the conclusions are robust, there are several issues that require addressing:

a. The authors characterize cerebellum-dependent functional deficits in the mutant mice, basing their studies on the high expression levels of HCN1 in cerebellum, citing Notomi & Shigemoto, They do not present phenotypic deficits in function ascribed to hippocampus or cortex. However<br> 1. Notomi and Shigemoto state: "Immunoreactivity for HCN1 showed predominantly cortical distribution, being intense in the neocortex, hippocampus, superior colliculus, and cerebellum,"<br> 2. Importantly, the seizures of the HCN1 mutant mice are unlikely to arise from the cerebellum, and the encephalopathies elements of HCN1-related neonatal epileptic encephalopathies clearly derive from cortex and hippocampus. Therefore, it should be excellent if the authors presented functional tests of hippocampus or cortex dependent behaviors, regardless of the outcome in Fig.2. At a minimum, they should modify the text and downplay the cerebellar emphasis.

b. The authors base their proposed mechanism for the pro-epileptic effects of the mutation on the notion that HCN1 Channels are localized to axons only of PV interneurons. Whereas this fact may be true for the adult, during development, axonal targeting is not unique to basket-type interneurons. It is observed in the developing hippocampal circuit, in medial entorhinal cortex neurons innervating dentate gyrus granule cells, i.e., the perforant path. Have the authors looked at axonal targeting in this region in the mutant mice during appropriate developmental stages? Its absence might modulate the firing of GCs, specifically during development (Bender et al., J Neurosci 2007). At a minimum this point merits discussion, particularly in view of the developmental nature of the epilepsies described.

c. In this context, there are distinct developmental profiles for the 4 HCN subunits, including HCN1, and these profiles might contribute to age-specific defects leading to seizures. This point merits discussion.

d. Whereas the focus of this paper is on the role of genetic mutations in HCN1 in epilepsy, the paper may be enriched by being placed in the context of the overall contributions of HCN1 channels to human epilepsy, including "acquired epilepsy"" via potential epigenetic changes in the expression of normal HCN channels (Bender et al., 2003 and others).

Reviewer #1 (Public Review):

The key question addressed of this MEG study is whether speech is represented singly or multiplexed in the human brain in the linguistic hierarchy. The authors used state-of-the-art analyses (multivariate Temporal Response Functions) and probablilistic information-theoretic measures (entropy, surprisal) to test distinct contextual speech processing models at three hierarchical levels. The authors report evidence for the coexistence of local and global predictive speech processing in the linguistic hierarchy.

The work uses time resolved neuroimaging with state-of-the-art analyses and cognitive (here, linguistic) modeling. The study is very well conducted and draws from very different fields of knowledge in convincing ways. I see one limitation of the current study in that the authors focused on phase-locked responses, and I hope future work could extend to induced activity.

Overall, the flow in the MS could be streamlined. Some smoothing in the introduction would be helpful to extract the main key messages you wish to convey.

For instance, in the abstract:

– Can you explain the two views in a simpler way in the abstract and to a non-linguistic audience? Do you mean to say that classic psycholinguistic models tend to follow a strict hierarchically integration (analysis only) but an alternative model is hierarchically inferential (analysis by synthesis)?

– Indicate early on in abstract or intro where the audience is being led with a concise message on how you address the main question. For instance:

To contrast our working hypotheses A and B, we used a novel information-theoretic modeling approach and associated measures (entropy, surprisal), which make clear predictions on the latency of brain activity in responses to speech at three hierarchal contextual levels (sublexical, word and sentence).

– Why did the authors consider that the evoked response is the proper signal to assess as opposed to oscillatory (or non phase-locked) activity?

– Parallel processing with different levels of context (hence temporal granularities) sounds compatible with temporal multiplexing of speech representation proposed by Giraud & Poeppel (2012) or do the authors consider it a separate issue?

Methods:

– Figure 2: please spell out TRFs and clarify the measured response

– The sample size (N=12) is very low in today standards but the statistical granularity is that of the full MEG recording. Can a power estimate be provided or clear justification of reliability of statistical measures be described.

– The inclusion of a left-handed in speech studies in unusual, please comment on any difference (or lack thereof) for this participant and notably the lateralization tests.

– The authors state that eyes were kept open or close. This is again unusual as we know that eye closure affects not only the degree of concentration/fatigue but directly impact alpha activity (which in turn affects evoked responses (1-40 Hz then 20 Hz) that are being estimated here). Please explain.

– It would be helpful to clarify the final temporal granularity of analysis. The TRFs time courses are said to be resampled to 1kHz (p22) but MEG time courses are said to be resampled at 100 Hz (p18).

– The % of variance explained by acoustic attributes is 15 to 20 folds larger than the that explained by the linguistic models of interest. Can a SNR measure be evaluated on such observations?

Results and Figures:

– The current figures do not give enough credit to the depth of analysis being presented. I understand that this typical for such mTRFs approach but given the level of abstraction being evaluated in the linguistic inputs, it may be helpful to show an exemple of what to expect for low vs. high surprisal for instance from the modeling perspective and over time.

For instance, could Figure 1 already illustrate disctinct predictions of the the local vs. global models?

– Why are visual cortices highlighted in figures?

– Figure 2:

Fig 2A and B: can the authors quantitatively illustrate "5-gram generally leads to a reduction of word surprisal but its magnitude varies substantially between words" by simply showing the mean surprisal and its variance?

Fig 2C: please explain the term "partial response"; please indicate for non M/EEGers what the arrow symbolizes.

– Figure 3:

p8: the authors state controlling for the "acoustic features" but do not clearly describe how in the methods and this control comes as a (positive) surprise but still a bit unexpected at first read. Perhaps include the two acoustic features in Fig2C and provide a short couple sentences on how these could impair or confound mTRF performance.

Have the same analysis been conducted on a control region a priori not implicated in linguistic processing? This would be helpful to comfort the current results.

Fig 3B-C-E: please clearly indicate what single dot or "individual value" represents. Is this average over the full ROI? Was the orientation fixed? Can some measure of variability be provided?

Fig3E: make bigger / more readable (too many colors: significance bars could be black)

– Figure 4: having to go to the next Fig (Fig5) to understand the time windows is inconvenient and difficult to follow. Please, find a work around or combine the two figures. From which ROI are the times series extracted from?

Reviewer #2 (Public Review):

This manuscripts describes an MEG study where N=12 English-speaking participants listened to about 45 minutes of an audiobook story. The key question is what sorts of information guides predictions during this naturalistic comprehension: local information (e.g. phoneme to phoneme transitions) or global information (e.g. sentence context constrains phoneme expectations etc.) These theories were tested by constructing a set of language models that varied the context used to compute phoneme and word probabilities; these probabilities were quantified in terms of surprisal and entropy and those values were fit against source-localized MEG data using standard techniques (mTRFs). Results showed independent contributions of both more local and more local contexts on superior temporal sources.

I really like this manuscript and I think it will make a fine contribution to the literature. A few things to highlight: It is very clearly written. The introduction does a really nice job of integrating current state-of-the-art thinking with classic key psycholinguistic debates; the theoretical stakes are very clear. I also appreciated the relatively cautious aspects of interpretation such as the analysis looking at trade-offs between global and local contexts.

Reviewer #3 (Public Review):

This manuscript presents a neurophysiological investigation of the hierarchical nature of prediction in natural speech comprehension. The authors record MEG data to speech from an audiobook. And they model that MEG using a number of different speech representations in order to explore how context affects the encoding of that speech. In particular, they are interested in testing how the response to phoneme is affected by context at three different levels: sublexical – how the probability of an upcoming phoneme is constrained by previous phonemes; word – how the probability of an upcoming phoneme is affected by its being part of an individual word; sentence – how the probability of an upcoming phoneme is affected by the longer-range context of the speech content. Moreover, the authors are interested in exploring how effects at these different levels might contribute - independently - to explaining the MEG data. In doing so, they argue for parallel contributions to predictive processing from both long-range context and more local context. The authors discuss how this has important implications for how we understand the computational principles underlying natural speech perception, and how it can potentially explain a number of interesting phenomena from the literature.

Overall, I thought this was a very well written and very interesting manuscript. I thought the authors did a really superb job, in general, of describing their questions against the previous literature, and of discussing their results in the context of that literature. I also thought, in general, that the methods and results were well explained. I have a few comments and queries for the authors too, however, most of which are relatively minor.

Main comments:

1) One concerns I had was about the fact that context effects are estimated using 5-gram, models. I appreciate the computational cost involved in modeling more context. But, at the same time, I worry a little that examining the previous 4 phonemes or (especially) words is simply not enough to capture longer-term dependencies that surely exist. The reason I am concerned about this is that the sentence level context you are incorporating here is surely suboptimal. As such, could it be the case that the more local models are performing as well as they are simply because the sentence level context has not been modeled as well as it should be? I appreciate the temporal and spatial patterns appear to differ for the sentence level relative to the other two, so that is good support for the idea that they are genuinely capturing different things. However, I think some discussion of the potential shortcomings of only including 4 tokens of context is worth adding. Particularly when you make strong claims like that on lines 252.

2) I found myself confused about what exactly was being modeled on my first reading of pages 4 through 7. I realized then that all of the models are based on estimating a probability distribution based on phonemes (stated on line 167). I think why I found it so confusing was that the previous section talked about using word forms and phonemes as units of representation (lines 118-119; Fig 2A), and I failed to grasp that, in fact, you were not going to be modeling surprisal or entropy at the word level, but always at the phoneme level (just with different context). Anyway, I just thought I would flag that as other readers might also find themselves thinking in one direction as they read pages 4 and 5, only to find themselves confused further down.

3) I also thought some the formal explanations of surprisal and entropy on lines 610-617 would be valuable if added to the first paragraph on page 6, which, at the moment, is really quite abstract and not as digestible as it could be, particularly for entropy.

4) I like the analysis examining the possibility of tradeoffs between context models. I wonder might such tradeoffs exist as conversational environments vary - if the complexity of the speech varies and/or listening conditions vary - might there be more reliance on local vs global context then. If that seems plausible, then it might be worth adding a caveat that you found no evidence for any tradeoff, but that your experiment was pretty homogenous in terms of speech content.

Reviewer #1 (Public Review):

The authors have shown in a previous paper (ref 20) that relaxation of the plasma membrane after mechanical stretching induces the formation of different membrane structures that transiently accommodate the excess area, among them small membrane evaginations of typically 100 nm diameter on the apical side of the cells. The stretching technique was already described in (20). They focus here on the mechanism of the active re-absorption process of these small buds. Using fluorescence microscopy, correlative fluorescence/SEM microscopy and TEM, they show here that the evaginations are actively re-adsorbed in an actin-dependent manner. They evidence that the I-BAR domain protein, which was previously shown to be a curvature sensor is very important for the flattening of these evaginations by locally triggering actin polymerization in a Rac1/Arp2/3-dependent manner. By using different constructs, they show that the I-BAR and SH3 domains are key, for the recruitment in the curved structures and for the downstream actin polymerization, respectively. They also show that the re-absorption process is independent of Myo2, formins and N-WASP.

Concluding that IRSp53 is a mechanosensor is not so original: it was already proposed for the mechanism of filopodia generation by G. Scita (A. Disanza et al., EMBO J. 32, 2735 (2013)). It is rather a "curvature-sensor" to be more precise, as already shown in vitro as mentioned by the authors. However, the originality of this work is to show that IRSp53 is involved in the resorption of a protrusion rather than on its growth as observed in filopodia for instance. I think this result brings novel insights on the already-rich range of functions in which this protein is involved. It also shows that this process involves branched actin and not linear actin bundles like in filopodia.

To explain how resorption happens, the authors develop a theoretical model based on a composite membrane made of an active gel layer (the cortex) with a frictional coupling to the lipid bilayer. They propose that IRSp53 is recruited to the membrane buds due to its affinity for curved membranes. It locally nucleates branched actin polymerization, with induces a gradient of actin density, and thus a lateral actin flow away from the bud. Due to the frictional coupling between the membrane and the cortex, the evagination eventually vanishes. This is an interesting mechanism, but so far, the authors do not have considered other possibilities, neither ways to test the model, beyond the fact that this mechanism does not require Myo2 in agreement with the experiments.

On the same line, the authors do not discuss why resorption still occurs, although less efficiently, when IRSp53 is silenced or absent.

Reviewer #2 (Public Review):

Quiroga et al investigate what happens with the excess plasma membrane that becomes available when a cell undergoes rapid shrinkage / compression after a phase of stretching. Using fluorescence and electron microscopy they find that apically small bottle-like evaginations form and disappear again after a few minutes. They find that, in analogy to much larger membrane blebs, ezrin and actin get recruited to the evaginations before they disappear. As a possible mechanism, they investigate the involvement of IRSP53 as a putative sensor of membrane curvature. They use IRSP53 deficient fibroblasts, which show a delayed resolution of evaginations, as a platform and rescue with different established deletion constructs and results suggest that Rac1 and is downstream of curvature sensing. This is confirmed by dominant negative Rac1 constructs. Quantitative imaging and APEX based electron microscopy provide evidence that the IRSP53 acts locally at the evaginations, not globally, e.g. at lamellipodia or filopodia.

Finally the authors demonstrate that Arp2/3 mediated polymerisation of actin is required for the effect and suggest a model where in plane polymerisation of actin generates friction with the membrane that flattens out the local curvature.

This is an interesting, well executed study that might be relevant in many different physiological settings. For example, when cells get passively deformed in motile metazoans or when they actively deform upon contraction and migration. Membrane evaginations did get less attention in this context than invaginations and the suggested mechanism is plausible and will trigger future studies that sort out the biophysics and physiological relevance in more detail. At this stage the local action of actin in the context is completely hypothetical.

Reviewer #3 (Public Review):

The study by Quiroga et al. explores an interesting new mechanism used by cells to repair nanoscale outward membrane deformations (i.e. evaginations) that form upon rapid drop in membrane area. Using a combination of fluorescence and electron microscopy, the authors show that formation of these membrane evaginations, which are approximately 150nm in height and 100nm in diameter, cause recruitment of IRSP53 that locally augments Rac and Arp2/3 dependent actin polymerization to flatten and reconnect these membrane folds to the cell cortex.

To quantify the proposed mechanism, the authors use wild-type cells upon knockdown of IRSP53 as well as on IRSP53-/- cells. Reabsorption dynamics of membrane evagination is being measured for several proteins involved in the mechanism, as well as for a set of mutant and deletion constructs of IRSP53. The membrane structure and protein localization at these membrane folds is visualized using SEM, CLEM and APEX. The link from IRSP53 to its actin-regulatory binding partners is being explored using IRSP53 mutants, Rac mutants and chemical perturbations. These experimental data is complemented by a nice mathematical mode that further strengthens the feasibility of the proposed mechanism.

The manuscript is well structured and clearly written. The systematic analysis presents a compelling set of experiments that support the main findings of the manuscript.

Following aspect could benefit from some revisions: The authors show that membrane evagination are still reabsorbed upon knockdown of IRSP53 and in IRSP53-/- cells. If the proposed homeostatic mechanism is mediated solely by IRSP53, as the title implies, this should not be the case. Considering their redundant function, additional I-BAR domain proteins may contribute to this mechanism as well. To address this possibility, the authors would need to monitor the localization of other candidate proteins (e.g. MTSS1, MTSS2, IRTKS) at membrane evagination. Optionally, protein function could be tested via knockdown to delineate the contribution of individual I-BAR domain proteins.

Reviewer #1 (Public Review):

The authors investigate the contribution of a CTCF binding site located 3' of the beta-globin gene cluster to the relative expression of different genes within the cluster, in erythroid cell lines. They determine that deletion of 3'HS1, which harbors the first CTCF binding site downstream of the cluster, results in upregulation of the fetal HBG genes in both the K562 and HUDEP-2 cell lines. Hi-C analysis indicates that, as might be expected, the patterns of nuclear colocalization/interactions between the CTCF binding sites in the region change upon deletion of 3'HS1. Moreover, the authors also examine an inversion of the 3'HS1 binding site at its normal location, and this manipulation results in decreases in HBE/G expression associated with an increased frequency of interaction between 3'HS1 and another CTCF site further downstream of the locus, as determined by Hi-C. Additional disruption of the HBG transcriptional repressor BCL11A results in a further increase in HBG expression, suggesting that the upregulation that occurs upon 3'HS1 deletion results from a distinct mechanism. Deletion of a putative enhancer located downstream of the 3'HS1 site, marked by GATA-1 binding, in the 3'HS1-deletion background results in partial loss of increased HBG expression. Deletion of 48 kb of DNA between this enhancer and 3'HS1 results in a modest (2-fold) increase in HBG and HBE expression. Finally, the authors perform the 3'HS1 deletion in primary human CD34+ erythroid cell cultures, and demonstrate a similar effect on gene expression, as measured by HbF+ cells, albeit not as dramatic as that observed in the HUDEP-2 cell line.

In this study, the authors revisit a decades-old line of research related to deletional HPFH. A number of prior studies had indicated that 3' deletions in the human beta-globin locus resulted in significant upregulation of fetal beta-globin expression. The bulk of this effect has always been ascribed to the loss of the adult beta-globin gene, which then presumably allows the beta-globin LCR to activate the remaining genes within the locus, but there were prior suggestions of a downstream enhancer as well. The current study pinpoints the identity of this downstream enhancer and suggests a contribution for this element, as well as for its relative proximity to the genes after deletion of intervening sequences, in fetal globin activation. The study also clarifies the absence of any effect when 3'HS1 was deleted in the mouse, in that the enhancer is human-specific. In this light, the absence of an accompanying deletion of the HBB and/or HBD genes in combination with the deletion of 3'HS1 is notable; this would have provided a more complete investigation of the relationship with deletional HPFH. The authors potentially provide a wider scope of interest, outside the realm of beta-globin gene expression, by demonstrating specific effects, associated with physiologically relevant phenomena, upon manipulation (deletion and inversion) of a single CTCF binding site. My only complaint about the data is that beta-globin gene expression is not examined comprehensively for each cell line, and also is not presented consistently.

Reviewer #2 (Public Review):

The authors analyzed the functional role of a CTCF binding site in the β-globin gene locus in Hudep-2 and differentiating CD34+ cells. Previous studies have shown that CTCF sites flanking the globin gene locus interact and form chromosomal loops. The authors found that deleting 3'HS1 specifically increased expression of the γ-globin gene and reduced expression of the β-globin gene. This seems independent from the levels of the known γ-globin repressor BCL11a. Through analysis of ATAC seq. and GATA1-ChIP-seq. data, they identified an enhancer that upon deletion reduced activation of globin in the 3'HS1 deficient cells. Deletion of a GATA1 site within this enhancer also reduced globin expression.

This is an interesting manuscript that provides functional insight into regulatory DNA elements located downstream of the β-globin gene cluster. This study is also significant with respect to potentially improving therapeutic fetal globin production. Overall, the experiments include appropriate controls and statistics.

Reviewer #1 (Public Review):

Lenz et al have shown that IP injection of atRA does not affect sEPSC amplitude, sIPSC amplitude and frequency in the denta gyrus of both ventral and dorsal hippocampus. Interestingly, they observed a strong promoting effect of atRA on sEPSC frequency in the denta gyrus of dorsal, but not ventral, hippocampus. Lastly, they did not observe an difference in I/O in vivo, but did observe enhanced in vivo LTP in denta gyrus of mice injected with atRA which is abolished in the synaptopodin KO mice. The effect of atRA on LTP is very interesting as on sEPSC frequency in dorsal denta gyrus.

1) I do not agree with the authors' claim that atRA does not have a major effect on excitatory synaptic transmission. It seems that the sEPSC frequency increase by ~100%. Even if the 4 outlier points are excluded, the rest of the data points still clearly indicate an increase of sEPSC frequency.

What is the possible explanation of increased sEPSC frequency by atRA in dorsal region? Increased excitability of presynaptic neurons? (use TTX to decipher this?) Increased spine density? It seems that the authors did dye fill already... Count spine density? AND/OR increased glutamate release probability? (PPR measurement?) Did the authors perform I/O measurement in slice?

It is imperative that the authors tackle this issue head on.

2) The author need to specify which part of the denta gyrus for their in vivo study, as they discovered difference between ventral and dorsal in sEPSC frequency in slice preparation.

Reviewer #2 (Public Review):

The authors explore the effect of all-trans retinoic acid (atRA) on synaptic function and plasticity in the dentate gyrus of the hippocampus. Previous studies have established that atRA is a critical synaptic regulator that underlies an important form of homeostatic synaptic plasticity. This study builds on recent work by the same group showing that atRA enhances synaptic function in layer 2/3 pyramidal neurons of cortex in both mice and humans, and that atRA's synaptic regulating effects are critically dependent on synaptopodin, a protein that is specifically localized to the spine apparatus. In the present study, the authors used systemic treatment of atRA (via i.p. injection) and then monitored synaptic and intrinsic properties of dentate granule neurons in both dorsal and ventral hippocampus. Under these conditions, they found very little effect of systemic atRA on basal synaptic properties or intrinsic excitability of dentate granule neurons, but observed a striking enhancement of long-term potentiation (LTP) of inputs onto granule cells recorded in vivo in anesthetized mice. In the absence of atRA, LTP was induced but diminished over the next 60 min - systemic atRA rendered this LTP significantly more persistent over this time period. This plasticity modulating effect of atRA is absent in synaptopodin knockout mice, suggesting some mechanistic overlap in the plasticity modulating and synapse enhancing actions of atRA. This work represents a significant advance in demonstrating atRA can modulate enduring forms of synaptic plasticity even in the absence of overt regulation of basal synaptic function.

The experiments presented in this paper have all been well executed and the data presented justify the conclusions made in the paper. The authors also provide a thoughtful discussion of potential limitations of the approaches and outline follow-up experiments that will be informative. In particular, the systemic administration of atRA used in the study has the advantage of examining atRA actions in vivo, but the effective concentration of atRA in different brain regions that follows is unclear. The authors demonstrate that systemic atRA alters gene expression in the hippocampus, an important finding that clearly shows systemic atRA reaches the brain in sufficient concentrations. Yet, whether basal synaptic properties or intrinsic excitability of granule neurons might be altered by different local atRA concentrations remains an open question that should be addressed in future studies. Still, the data clearly demonstrate that atRA can modulate enduring forms of synaptic plasticity in the absence of overt changes in basal synaptic function.

Reviewer #1 (Public Review):

A major strength of the paper is the development of an improved activity assay that allows the authors to reliably measure the IC50 of telacebec with their supercomplex preparation.

A strength of the paper is the development of a 3xFlag-tagged M. smegmatis CIII2CIV2 allowing for one step affinity purification. This purification strategy also allowed the authors to demonstrate that the SOD and LpqE subunits are less tightly associated with the complex and can be lost during size exclusion chromatography. These observations will inform future biochemical work on this complex.

A strength of the paper is the improved cryoEM reconstruction of the M. smegmatis CIII2CIV2 supercomplex, showing clear density for the LqpE subunit and improved density for the SOD subunit.

A strength of the paper is the use of 3D variability analysis (3DVA) to identify a subset of particles that have lost one of the LpqE subunits resulting in the Cyt cc subunit to adopt the "open" conformation suggesting a regulatory role for this subunit. Furthermore, the 3DVA reveals movement of the SOD subunit that brings it into proximity of the cyt cc subunit and may allow for direct electron transfer from superoxide to CIV.

A strength of the paper is the localization of the telacebec biding site to the QP-site of CIII2 and the comparison with the menaquinone bound at the site in the absence of the inhibitor.

Reviewer #2 (Public Review):

Previously, these authors reported the structure of the CIII/CIV super-complex from M. smegmatis (Wiseman et al. 2018). In the current manuscript, the cryo-EM structure of CIII/CIV shows different positions for the attached superoxide dismutase and for the characteristic CIII/CIV cc domain, indicating movement of these units during the catalytic cycle.<br> The authors confirm that the isolated CIII/CIV is enzymatically active and is inhibited by nano-molar Q203, in line with the drug's MIC and previous experiments using membrane fractions.<br> The cryo-EM structure reveals that Q203 binds to the Qp site, which in drug-free CIII/CIV is occupied by a quinone molecule. The structure also shows a network of interactions between Q203 and various residues of CIII/CIV.

The paper is well written and in general the conclusions are supported by the data.

To this reviewer, one technical aspect related to the activity assays is unclear: according to the transparent reporting form, six independent assays were performed (technical replicates). Does that mean that the enzyme was purified once and the assay then carried out six times with that purified sample? Usage of multiple biological replicates (i. e. different batches of isolated CIII/CIV) is important, in particular in case, as the authors indicate, some subunits can partially dissociate from the super-complex.

Reviewer #3 (Public Review):

Telacebec (Q203) is a novel first-in-class antituberculosis drug that targets Mycobacterium tuberculosis respiration and cellular energy production through inhibition of the mycobacterial cytochrome bcc-aa3 (CIII2CIV2) super complex and thus represents the third class of energetic inhibitors against M. tuberculosis. Q203, currently in phase II clinical trials, has proven anti-tuberculosis activity in humans. Q203 also has the potential to be used in the treatment of nontuberculosis mycobacterial infections e.g. Buruli ulcer, a neglected tropical skin disease.

The manuscript by Rubinstein and colleagues reports the atomic-resolution entire structure of the Mycobacterial super CIII2CIV2 complex from Mycobacterium smegmatis inhibited by the anti-TB drug Q203. Other groups have reported the structure of CIII2CIV2 complex at high resolution, but none of this previous work has captured the drug inhibited structure. Yanofsky's new CIII2CIV2 complex captures the binding site of Q203 providing the mechanism (protein-inhibitor contacts) of inhibition of the complex and importantly also sheds light on the mechanism of enzyme catalysis. The structure reveals that Q203 binds with its head group deep within the Qp-binding pocket in a pose similar to UQ thus blocking menaquinol oxidation by the complex in both menaquinol binding modes. Multiple interactions are apparent leading to stabilised inhibitor binding.

The paper provides further structural data on the role of the SOD subunit in the complex and suggests that SOD may indeed transfer electrons from periplasmic superoxide to CIV thus contributing to the PMF and energy generation. The paper is a comprehensive, well written in an easy to follow style and will have broad interest and readership.

Reviewer #1 (Public Review):

In this study, Kuppan, Mitrovich, and Vahey investigated the impact of antibody specificity and virus morphology on complement activation by human respiratory syncytial virus (RSV). By quantifying the deposition of components of the complement system on RSV particles using high-resolution fluorescence microscopy, they found that antibodies that bind towards the apex of the RSV F protein in either the pre- or post-fusion conformation activated complement most efficiently. Additionally, complement deposition was biased towards globular RSV particles, which were frequently enriched in F in the post-fusion conformation compared to filamentous particles on which F exists predominantly in the pre-fusion conformation.

Strengths:<br> 1) While many previous studies have examined the properties of antibodies that impact Fc-mediated effector functions, this study offers a conceptual advance in its demonstration that heterogeneity in virus particle morphology impacts complement activation. This novel finding will motivate further research on this topic both in the context of RSV and other viral infections.

2) The use of site-specific labeling of viral proteins and high-resolution fluorescence microscopy represents a technical advance in monitoring interactions among different components of antiviral immune responses at the level of single virus particles.

3) The paper is well written, data are clearly presented and support key claims of the paper with caveats appropriately acknowledged.

Minor weaknesses:<br> Working models and their implications could be clarified and extended. Specifically:

1) The finding that globular particles enriched in F proteins in the post-fusion conformation (Fig 3F) are dominant targets of complement activation as measured by C3 deposition by not only post-F- but also pre-F-specific antibodies (Fig 4B, left) is interesting. This is despite the fact that, as expected, pre-F antibodies bind less efficiently to globular particles (Fig 4B, right). How do the authors reconcile these observations, given that C3 deposition seems to be IgG-concentration-dependent (Fig 2E)?

2) Based on data in Figure 5-figure supplement 2, the authors argue that "large viruses are poised to evade complement activation when they emerge from cells as highly-curved filaments, but become substantially more susceptible as they age or their morphology is physically disrupted." Could the increase in C3 deposition be alternatively explained by a higher density of F proteins on larger particles instead of / in addition to a larger potential decrease in membrane curvature?

3) In the discussion, the authors acknowledge that the implications based on the findings are speculative. However, more clarity on the basis of these speculative models would be useful. For example, it is not clear how the findings directly inform the presented model of immunodominance hierarchies in infants.

Reviewer #2 (Public Review):

This is an intriguing study that investigates the role of virus particle morphology on the ability of the first few components in the complement pathway to bind and opsonize RSV virions. The authors use primarily fluorescence microscopy with fluorescently tagged F proteins and fluorescently labeled antibodies and complement proteins (C3 and C4). They observed that antibodies against different epitopes exhibited different abilities to induce C3 binding, with a trend reflecting positioning of IgG Fc more distal to the viral membrane resulting in better complement "activation". They also compared the ability of C3 to deposit on virus produced from cells +/- CD55, which inhibits opsonization, and showed knockout led to greater C3 binding, indicating a role for this complement "defense protein" in RSV opsonization. They also examined kinetics of complement protein deposition (probed by C4 binding) to globular vs filamentous particles, observing that deposition occurred more rapidly to non-filaments.

A better understanding of complement activation in response to viruses can lead to a more comprehensive understanding of the immune response to antigen both beneficial and detrimental, when dysfunctional, during infection as well as mechanisms of combating the viral infection. The study provides new mechanistic information for understanding the properties of an enveloped virus that can influence complement activation, at least in an in vitro setting. It remains to be determined whether these effects manifest in the considerably more complex setting of natural infection or even in the presence of a polyclonal antibody mixture.

The studies are elegantly designed and carefully executed with reasonable checks for reproducibility and controls, which is important especially in a relatively complex and heterogeneous experimental system.

Specific points:

1) "Complement activation" involves much more than C3 or C4 binding. Better to use more specific terminology relating to the observable (i.e. fluorescently labeled complement component binding)

2) What is the rationalization for concentrations of antibodies used? What range was tested, and how dependent on antibody concentration were the observed complement deposition trends? How do they relate to physiological concentrations, and how would the presence of a more complex polyclonal response that is typically present (e.g. as the authors noted, the serum prior to antibody depletion already mediates complement activation) affect the complement activation trends? The neat, uniform display of Fc for monoclonals that were tested is likely to be quite garbled in more natural antibody response situations. This should be discussed.

3) Are there artifacts or caveats resulting from immobilization of virus particles on the coverslips?

4) How is the "density of antigen" quantitated? What fraction of F or G is labeled? For fluorescence intensity measurements in general, how did the authors ensure their detection was in a linear sensitivity range for the detectors for the various fluorescent channels? Since quantitation of fluorescence intensities is important in this study, some discussion in methods would be valuable.

5) The authors also show that the particle morphology, whether globular or filamentous, as well as relative size and resulting apparent curvature, correlate with ability of C3 to bind. Some link to the abundance of post-fusion F (post-F) is examined and discussed, but I found the back and forth discussion between morphology, C3 binding, and post-F abundance to be confusing and in need of clarification and streamlining. Is there a mechanistic link between morphology changes and post-F level increases? Are the two linked or coincidental (for example does pre-F interaction with matrix help stabilize that conformation, and if lost lead to spontaneous conversion to post-F?). Please clarify.

6) Since their conclusion is that curvature of the virus surface is a major influence on the ability of complement proteins to bind, I feel that some effort at modeling this effect based upon known structures is warranted. One might also anticipate then that there would be some epitope-dependent effect as a result of changes in curvature that may lead to an exaggeration of the epitope-specific effects for more highly curved particles perhaps than those with lower curvature? Is this true?

7) Line 265: it would be useful to confirm the increase C1 binding as a function of morphology as was done for antibody-angle of binding experiments.

Reviewer #3 (Public Review):

Overall the manuscript is clearly written and the data are displayed well, with helpful diagrams in the figures to illustrate assays and RSV F epitopes. The engineering of the RSV strain to include a fluorescent reporter and tags on F and G that serve as substrates for fluorophore attachment is impressive and is a strength. The RSV literature is well cited and the interpretation of the results is consistent with structure/function data on RSV F and its interaction with antibodies. This reviewer is not an expert on the experiments performed in this manuscript, but they appear to be rigorously performed with appropriate controls. As such, the conclusions are justified by the data. One weakness is the extent to which the results regarding virion morphology are biologically relevant. Non-filamentous forms of the virion are generally obtained only in vitro as a result of virion purification or biochemical treatment. However, these results may be relevant for certain vaccine candidates, including the failed formalin-inactivated RSV vaccine that was evaluated in the late 1960s and caused vaccine-enhanced disease upon natural RSV infection.

Reviewer #1 (Public Review):

In prior work, the authors identified a requirement for an ATR-dependent activation of Chk1 for the maintenance of G2 arrest in larval tracheoblasts. Absent the activity of ATR or pChk1, larval tracheoblasts re-enter the cell cycle early. Here the authors were attempting to build on their prior work by determining the mechanism by which ATR is regulated in larval tracheoblasts. In large measure, the authors are successful in this endeavor, finding that ROS generated by Duox are required to activate ATR. However, what regulates the timing of Duox expression and the precise molecular mechanism by which ROS activates ATR remains unresolved. The closely related ATM kinase has previously been shown to be directly regulated by ROS, through the formation of disulfide bridges. The authors' modeling suggest a different mechanism may be at play.

Reviewer #2 (Public Review):

This paper follows up on a previous eLife paper from the lab in which they showed that multiple Wnts control ATR expression to mediate cell cycle arrest of the trachea stem cells that will repopulate the tracheal system during pupal development. Here they show that activity, but not expression, of ATR is regulated by phosphorylation in response to high reactive oxygens species (ROS) generated by the enzyme Duox, an H2O2 generating-Dual Oxidase. Duox is required for high ROS levels in the trachea cells and suppression of high ROS levels by overexpression of super oxide dismutase causes precocious proliferation. Duox regulation of ATR and cell cycle arrest is independent of DNA-damage or activation of ATRIP, TOPBP1 or Claspin. Exogenous H2O2 can block excessive proliferation and restore ATR phosphorylation deficits resulting from knockdown of Duox. The experiments are well done and rigorous, and the paper is well written. No major concerns are noted.

Reviewer #3 (Public Review):

Kizhedathu et al. propose that Duox generated ROS activate ATR/Chk1 in the tracheoblasts in the Drosophila Tr2 metamere, and that this ROS-mediated ATR/Chk1 activation is DNA damage independent. This is a novel, unusual mechanism of activation of ATR and Chk1 that should be of rather broad interest, though it is not without precedent. They present data supporting the conclusion that ROS-dependent activation of ATR & Chk1 plays an instructive role in arresting tracheoblasts in G2, and that the developmentally controlled loss of ROS during the late L3 larva stage deactivates ATR/Chk1 and thereby activates tracheoblast cell division. The authors have performed extensive experiments to support these claims and the data are, overall, consistent with the model they present. The limitations of this study are that the authors have not elucidated how, molecularly, ROS activates ATR, and that they do not provided data or discussion relevant to how ROS levels might change according to developmental stage.

Reviewer #1 (Public Review):

The authors demonstrate that selective inactivation of carnitine acetyltransferase (Crat) – a key metabolic enzyme – in AgRP neurons attenuates the response of AgRP neurons to peanut butter (PB) chips, the release of dopamine in the nucleus accumbens, and the motivation to work for food when mice are fasted. The strength of this study is the demonstration that metabolic sensing by AgRP neurons is somehow linked to dopamine release in the nucleus accumbens, but a weakness is that it is unclear how the lack of Crat in AgRP neurons affects their responsiveness to PB chips or how AgRP neurons regulate dopamine release. The authors use of contemporary methods to monitor the kinetics of of AgRP neuron activity (fiber photometry) and dopamine release (GRAB-DA) in response to feeding fed or fasted mice with PB chips is commendable. The authors acknowledge that the neural circuits linking changes in AgRP neurons activity to release of dopamine is indirect because AgRP neurons do not directly synapse onto dopamine neurons; thus, their findings provide intriguing correlations without a clear understanding of the circuit(s) involved. The authors clearly demonstrate that the Crat knockout (KO) mice do not respond to PB chips the same was as WT mice; the KO mice respond to the first chip normally but responses to additional chips are blunted. The mechanisms underlying the blunted response are assumed to be due to a failure of metabolic sensing, but the mechanisms involved are not explored.

Reviewer #2 (Public Review):

In their manuscript, Reichenbach et al perform experiments to demonstrate that knocking the metabolic enzyme carnitine acetyltransferase (Crat) out of hunger-promoting AgRP neurons impairs an animal's ability to accurately sense its nutritional state and thus decreases motivation to work for food and attenuates neural response to palatable food rewards. They accomplish this using in vitro and in vivo neural recording techniques, and thoughtful behavioral approaches. Specifically they show 1) impaired responses of AgRP neurons to glucose (in vitro) and sensory cues predicting food (in vivo), 2) impaired striatal dopamine release in response to palatable food presentation, and 3) decreased motivation to work for palatable rewards.

Their work largely substantiates their conclusions. Their experiments are well described, and the phenotypes observed are generally clear. The data partially explain prior behavioral studies performed on these conditional knockout mice. Moreover, their data are consistent with and a valuable addition to previously published data showing how dorsal and ventral striatum differentially respond to nutrient intake with ventral striatum dopamine release increasing in response to sweet taste and dorsal striatum dopamine increasing in response to rewarding post-ingestive effects of nutrients. This study will be of interest to a fairly broad community of feeding, hypothalamus, and dopamine researchers.

A limitation of this study is that it does not adequately address the possibility that decreased AgRP neuron responses to food presentation may be related to altered in vivo baseline activity or attenuated fasting-induced hyperactivity of these neurons. While their slice studies show mostly normal ex vivo electrophysiologic properties of these neurons, in other models ex vivo and in vivo measurements of AgRP neuron activity are not directly correlated. Specifically Kristen O'Connell's group has shown increased baseline AgRP neuron activity in diet-induced obese (DIO) mice that is not further increased by fasting in slice (Baver et al, 2014). By contrast, Michael Krashes's group recently shown that DIO mice have reduced baseline AgRP neuron activity using a fiber photometry approach in vivo (Mazzone et al, 2020). Of note, decreased baseline or fasting-induced AgRP neuron activity would not necessarily diminish the impact of the rest of the results presented. Moreover, it is not necessarily a question that must be answered by this study, but it should be acknowledged as a possibility that is important to test.

Additional minor concerns do not significantly dampen my generally positive opinion of the study. These include: 1) the lack of feeding data associated with AgRP neuron fiber photometry responses, 2) analysis of operant GRAB-DA data by pellet retrieval event rather than by mouse, and 3) incompletely described inclusion criteria for mice in photometry studies.

Reviewer #3 (Public Review):

Reichenbach et al tested the hypothesis whether metabolic sensing in AgRP neurons is required to increase food reward motivation by influencing dopamine release in the striatum. As a model for disrupted metabolic sensing they employed their previous described mouse model that specifically lacks carnitine acetyltransferase (Crat) in AgRP neurons. They confirm using electrophysiology the appropriateness of the model and then conduct a series of elegant experiments measuring short-term dopamine release by fibre optics in response to different feeding manipulations in the nucleus accumbens and striatum. In a final experiment they then also test for longer term responses to Dopamine release utilising 18F-fDOPA in combination with positron electron tomography. Their data reveals that reduced metabolic sensing in the knockout mouse model reduces acute dopamine release in the NAc but not in the dorsal striatum, while in the longer time frame (30 min) the dorsal striatum is also affected. In summary the authors conclude that metabolic sensing in AgRP neurons is necessary to integrate homeostatic regulatory processes in the AgRP neurons with hedonic aspects of dopamine signalling in reward pathways. Taken together, the experiments are well-executed and the results justify the main conclusions of the study.

Reviewer #1 (Public Review):

Schlegel et al. dug into the function of the CRALBP protein in relation to a series of autosomal recessive retinal diseases. They used zebrafish as a model since they hold two duplicated genes (rlbp1a and rlbp1b paralogs) which are orthologs for the human RLBP1. Interestingly, a subfunctionalisation has happened in zebrafish that led each one of the paralogs to be expressed exclusively in either the RPE (rlbp1a) or the Müller glial cells (rlbp1b). This circumstance is very favourable for the study of their function in the visual cycle, independently and together, by knocking them out using goal directed mutagenesis. To do so, Neuhauss lab incorporated to his lab the state-of-art technique CRIPSR-Cas to specifically KO each of the two genes of interest.

Subsequently, to analyse the results, they confirmed the absence of the targeted proteins by immunohistochemistry and afterwards analysed the retinal function by electroretinogram and the presence of different metabolites by HPLC.<br> The results and the conclusions extracted from them are totally sound and the discussion is well-structured touches each single point.

The goal of the study was to shed light on the aetiology of the diseases known to be by associated with mutations in the gene RLBP1 and the authors succeeded by generating a reliable model recapitulating the main features common to these autosomal recessive retinal diseases, including an essential factor as aging is. The generation of a new model opens doors in the respective field by facilitating the study of the related phenotype-related alterations, something important in this case due to the incidence of the blinding diseases caused by RLBP1 mutations.

As a major strength I see the continuity of the reliable, rigorous and meticulous work of Neuhauss lab. It is a pleasure to see regular publications, evolving together with the advances in technology. In this case, incorporating the technique awarded in the last Nobel prize in Chemistry. In addition, the generation of a new model to study RLBP1-KO-related diseases adds a new tool to investigate the molecular mechanisms triggering these diseases.

As a weakness I see the use of a CRALBP1 serum to validate the KOs, instead of using specific antibodies. I understand the difficulty of generating antibodies but perhaps the use of riboprobes could be a more adequate way.

Reviewer #2 (Public Review):

Using CRISPR/Cas9 mutagenesis the authors generated two homozygous knockout fish lines deleting separately rlbp1a in RPE and rlbp1b in MGC and investigated the pathological effects of the protein deficiencies on visual function. It was found that deletion of rlbp1a results in reduced chromophore levels and suppressed cone responses to light. This mutant also accumulated lipid droplets containing 11-cis and all-trans-retinyl esters which are similar to subretinal lesions in patients with RLBP1 mutations. During aging rlbp1a mutant fish develops cone and rod dystrophy and retinal thinning. In contrast, rlbp1b deletion did not result in retina degeneration or dysfunction. The double mutant demonstrated the same phenotype as a single rlbp1a mutant. This means that CRALBP1a is an essential protein which plays an important role in generation of 11-cis-retinal by accelerating the retinol isomerase reaction catalyzed by RPE65 and, probably, assisting to RGR photoisomerase in RPE cells. It may also assist to yet unknown retinyl ester hydrolase which hydrolyzes 11-cis-retinyl esters in retinosomes to produce 11-cis-retinol. The results of this work show that MGC produced CRALBP1b does not participate in visual cycle in zebrafish. Most likely, mammalian CRALBP expressed in MGC also plays only minor role in visual function. It means that Muller cells contribution to chromophore regeneration previously postulated by other authors maybe overestimated. The role of rlbp1b remains to be investigated in future.

The manuscript is written clearly and carefully. Both the data and the conclusion of the manuscript are solid and certainly worthy of publication.

Reviewer #3 (Public Review):

Schlegel et al take advantage of the fact that in teleost fish there has been a duplication of the RLBP1 gene, with rlbp1a expressed in the retinal pigment epithelium (RPE) and rlbp1b expressed in Müller glial cells (MGCs). Selective knockout of these paralogs in zebrafish has enabled the authors to separately examine the roles of the expressed cellular retinaldehyde binding protein, CRALBP, in these two classes of cell. They have thereby been able (in a cone-dominant species) to disentangle the contributions of CRALBP within the two retinoid re-cycling pathways: the canonical RPE cycle and the more recently discovered intra-retinal cycle.

Importantly, the authors show that their zebrafish rlbp1a KO model exhibits features found in certain human blinding diseases caused by mutations in the RLBP1 gene (e.g. Bothnia dystrophy, fundus albipunctatus, etc.). Most notable is the occurrence of sub-retinal lipid droplets (that appear to contain the accumulation of retinyl esters), together with the occurrence of age-related retinal degeneration.

The experiments have been well-designed and for the most part well-executed, and the results and interpretation generally appear very solid. Nevertheless, there are a few areas where clarification or additional data are called for.

1) One area where additional explanation is needed concerns the measurement and interpretation of 11-cis retinaldehyde (11cisRAL) levels. The text refers to "11cisRAL" as the aldehyde that binds to CRALBP. However, it is unclear whether the measurements of "11cisRAL" refer to the total of 11-cis isomer covalently bound as visual pigment (in cones and rods) in addition to the non-covalently bound 11cisRAL. I am not a chemist, and am unable to determine what was extracted. It will be important for the paper to make absolutely clear what was measured, and to interpret the measurements appropriately, according to whether or not covalently bound retinaldehyde was included. If the measurements did include rhodopsin and cone opsins, then I would urge that a term distinct from "11cisRAL" be employed throughout the paper.

2) Another area that would benefit from additional explanation is the considerable degree of variability in the measurements of retinoid content in WT adult animals in Figure 2. Between panels A, C and E, the mean WT levels differ markedly: for 11cisRAL (312, 177 and 280); for 11cisRE (12.7, 86.1 and 7.2); and for atRE (168, 460 and 187). Are these WT values significantly different from each other? Would it be better to use the WT grand means in the tests of significance for the knockouts?

3) Before the authors can state that "We observed no pronounced changes of cone responses in Cralbpb-deficient larval eyes" they need to illustrate rlbp1b KO responses in Figure 4. At present, the double KO plots in panel B2 ("BL", and "BL+reDA") appear to show considerably greater attenuation than for the rlbp1a KO plots in panel A2, strongly suggesting an effect of Cralbpb. So I can't see how the authors can make the above statement unless they present the data.

Subject to clarification of the points above, this paper represents an important contribution to knowledge of the cellular mechanisms of retinoid recycling for cone photoreceptors.

Reviewer #1 (Public Review):

In previous work these authors developed a mathematical model for the pulsatile activity of KNDy neurons responsible for episodic GnRH and LH secretion. Here they use this model and experimental manipulations to test the hypothesis that the KNDy neural network responds differently to optogenetic stimulation on different days of the estrous cycle and that these changes reflect changes in the excitability of this population driven by glutamate signaling. In the first experiment, they demonstrate that optogenetic stimulation increases episodic LH secretion on estrous when endogenous activity is low and inhibits LH pulse frequency on diestrus when endogenous activity is elevated. In the second experiment, they show that pharmacological inhibition of glutamate signaling blocks the stimulatory actions of optogenetic stimulation on estrous and inhibits endogenous LH pulses on diestrus. Morever, the latter effect is partially overcome with optogentic stimulation. The modelling portion of the manuscript provides a simple explanation for these observations and supports the hypothesis that glutamate signaling that increases the network excitability of this population plays an important role in the changes between diestrus and estrus.

Overall, this work has been carefully done and the overall conclusions are consistent with the experimental observations and model. I do have three concerns about the interpretation and discussion of these results. First, although the key characteristic of the proposed model (that there is a limited range of excitability levels compatible with the episodic activity of the KNDy network) provides an elegant explanation for previous work describing apparent paradoxical effects of many neurotransmitters on LH pulses, it also complicates the design of experiments to test it. Thus, if a pharmacological manipulation will either decrease or increase episodic LH secretion depending on the endogenous activity of the network, which may be difficult to determine, then any experimental result is consistent with the model. This caveat appears to be inherent to their model, but it may be prudent to soften the conclusions in light of it. For example, any pharmacological manipulation that affects the level of excitability of the network might have similar effects as the blockade of glutamatergic transmission reported here. Thus, the data are consistent with the proposed role of glutamatergic input to KNDy neurons, but they do not establish that this is the only, or even the most important, input. Second, how do the authors reconcile the positive correlation of NKB and dynorphin signaling with most data indicating that these two neuropeptides usually have opposite effects on episodic LH secretion. Further consideration of these potentially conflicting data and the implications of this positive correlation to the functioning of the KNDy network would strengthen the Discussion. Finally, the discussion on changes in the characteristics of KNDy neurons on different days of the estrous cycle focuses exclusively on the effects of estrogen on this neural network. However, the relevance to these estrogen effects to the changes reported between diestrus and estrus is unclear because the KNDy neural network (based on bursts of calcium and LH pulse frequencies) does not change from diestrus though proestus when estradiol concentrations are increasing from a nadir to peak levels. Since, as pointed out in the Introduction, the low LH pulse frequency on estrus is caused by the proestrous increase in progesterone concentrations some consideration of this steroid should be included in the Discussion.

Reviewer #2 (Public Review):

The paper by Voliotis et al. combines novel experimental evidence with mathematical modelling. The experiments were designed to analyse the effects of activating the kisspeptin population in the arcuate nucleus on the generation of pulsatile LH secretion at different phases of the ovarian cycle, and the involvement in this of glutamate signalling. Experimental data come from transgenic mice engineered to express Cre in kisspeptin-expressing cells that had been injected stereotaxically in the arcuate nucleus (bilaterally) with an adenoviral vector to transduce expression of channelrhodopsin in arcuate Kisspeptin neurones. The authors measured luteinising hormone (LH) in small blood samples taken from the tail tips of conscious mice while these kisspeptin cells were activated optogenetically using an implanted optic fibre. Experiments were performed during icv administration of glutamate antagonists or aCSF.

The experiments are a methodological tour de force, necessitating stress-free sampling and stimulation procedures and precise targeting of adenoviral injections. The design of the experiments is elegant, inspired by a simple mathematical model that generates a concise simulation of the role of kisspeptin neurones in the generation of pulsatile LH secretion.

Kisspeptin neurones co-express two other neuropeptides, dynorphyn and neurokinin, leading them to be known by the acronym KNDy neurones, and these are thought to influence LH secretion by the actions of these peptides on neurones that release GnRH. This paper reproduces previous work by the same authors in showing that optogenetic stimulation of KNDy neurons stimulates pulsatile LH secretion in estrous mice but inhibits it in diestrous mice.

These results are consistent a mathematical model which proposes that that the transition between estrus and diestrus reflects orchestrated changes in the excitability of the KNDy population. On the premise that the excitability of the KNDy cells is primarily controlled by glutamate neurotransmission either between the KNDy cells or from external inputs, the present study studied the consequences of blocking glutamate signaling. Blocking glutamate signalling in diestrous animals inhibited LH pulses, and this inhibition could be mitigated by optogenetic stimulation of the KNDy cells. In estrous mice, blocking glutamate signalling inhibited the pulsatile LH secretion generated by optogenetic stimulation of the KNDy cells.

The model characterises the KNDy cells as a single element that secretes dynorphin and neurokinin independently, in a manner that varies during the ovarian cycle. These two secreted products decay at different rates, and have different (autocrine) feedback effects on the KNdy cells: neurokinin is excitatory and decays rapidly; dynorphin is inhibitory, is secreted at a lower rate but decays more slowly, so accumulates. These characteristics establish a bistable oscillatory system, that oscillates at a frequency determined mainly by the time constants of degradation.

I like the model, it is very simple yet capable of elegantly explaining apparently complex behaviour. However, it seems to me that the assumptions that underlie the model should be explicitly stated and the relevant evidence supporting them clearly stated.

In L209 the authors state "In the equations above neuronal activity stimulates secretion of both neuropeptides, and Dyn represses NKB secretion."

Thus one key assumption is that dynorphin and neurokinin secretion are independent: this seems to require either that they are packaged in separate neurosecretory vesicles that are subject to independent regulation, or that they are predominantly expressed in mainly separate (and functionally distinct) populations of kisspeptin neurones in the arcuate nucleus. I am unsure how tenable either of these assumptions is. I don't mind if they are bold and speculative, but they should be clearly recognised is such. (Key questions appear to be whether dynorphin, NK and kisspeptin are contained in the same or separate vesicles, whether their synthesis is differentially regulated through the estrus cycle, and if their secretion from different compartments is regulated differently).

2. The model collapses two very different roles of glutamate neurotransmission into a single rather vaguely defined variable. The authors recognise that glutamate is likely to be the major excitatory transmitter arising from external inputs that may perhaps be presumed to establish a constant excitatory 'tone.' Understanding 'excitability' as the strength of a tonic input seems reasonable. However, the authors also recognise that glutamate signaling between KNDy neurones might be important in synchronising neural activity in a way important for pulse generation. This seems to be a very different role, a role that is not incorporated in the model.

It seems to me therefore that the experiments don't really test the model predictions because they have a major effect on something not included in the model that would reasonably be expected to have a substantial effect on the way the system behaves.

Nevertheless, this work is an imprtant step forward. It is highly original; its importance in my view is in establishing a clear framework for understanding the function of a complex neuronal system that plays a critically important role in reproduction, and in providing an exemplar of how simple mathematical models can inspire the generation of experimentally testable hypotheses by proposing simple explanations of apparently complex behaviours. The authors have made the raw data from their experiments fully and openly available, as I have checked for myself.

Reviewer #1 (Public Review):

In the present technical report the authors develop a novel and simple strategy to generate and efficiently isolate cre/lox responsive alleles that in the right orientation are highly mutagenic. These allele offer in principle a powerful tool for the tissue specific analysis of gene function giving the ability to inactivate the gene of interest with high temporal and spatial control. The mutagenic targeting cassette contains a marker reporter for the efficient isolation of putative carriers following the integration in injected zebrafish embryos. The cassette in the mutagenic orientation also potentially labels the disrupted gene with a fluorescent tag, with the intent of following the single cells (and their full lineage) in which gene disruption has occurred. This approach is applied to three test loci comparing for which the same group had generated classical null alleles containing small deletions in the coding region.

Reviewer #2 (Public Review):

This manuscript describes a method for generating floxed conditional alleles in the zebrafish. The method employs the authors' previously reported GeneWeld CRISPR/Cas9 short homology-directed targeted integration strategy to introduce a "UFlip" cassette that allows target genes to be "turned off" or "turned on" in a tissue-specific manner with appropriate cre driver lines. The authors provide data to show the efficacy of their new method by targeting hdac1, rbbp4, and rb1. Although a variety of other methods have recently been reported for gene inactivation in the zebrafish (many of which were cited and discussed by the authors of this manuscript), the authors method could provide some notable and significant advances including speed and ease of generation of conditional alleles and flexibility to generate "on" and "off" alleles at the same genomic locations. However, the authors would need to do more to provide additional and more quantitative data validating some of the important features and advantages of their method. A few of the most significant issues that should be addressed include:

Better assessment of the efficiency of cre/dre "flipping" of integrated constructs.

It's important that the authors provide data showing not just that inversion of their integrated constructs can happen, but quantitatively measuring how efficiently this occurs. The authors should provide qPCR or other measurements to assess % inversion of their constructs via injected and transgene-driven cre. The authors should also provide data quantitating the % efficiency of dre/rox inversion used to turn "off" alleles into "on" alleles and vice-versa (see Supp Fig. 5).

Although the authors provide junction qRT-PCR suggesting efficient transcript blocking by "off" insertion alleles (eg Fig 2B,J) it would also be useful to further validate and explore the effects of "off" UFlip transgene insertions on expression of targeted genes by similar qRT-PCR on upstream and downstream exons in control vs. het or homozygous transgene insertion animals to assess whether truncated transcripts are degrading, and whether expression of downstream exons is indeed absent.

Better validation that "null phenotypes" can be generated.

Although the qRT-PCR data the authors provide suggests efficient transcript blocking by "off" insertion alleles, the authors need to strengthen some of their data showing that null phenotypes are being generated by these alleles. In many cases the authors provide only anecdotal images, describe relatively generic phenotypes, or provide quantitative measurements of mutant phenotypes (eg pH3 positive cells) that lack key positive controls such as comparable quantitative measurements on previously generated bona fide "null mutant" alleles of the same genes. All of this is important to demonstrate that this method can generate robust phenotypes that are both qualitatively and quantitatively comparable to null mutants.

Demonstrate tissue-specific "flipping on."

One of the major points of novelty and most exciting features of the authors methods over other recently reported approaches is the potential to carry out tissue-specific gene activation (by cre-flipping on "UFlip-Off" alleles). This would be an exceptionally useful and powerful new tool for fish researchers (and others). Surprisingly, this particularly exciting feature is curiously unexplored in this manuscript, although the authors do generate a number of UFlip-On alleles. The impact and significance of this manuscript would be substantially increased by a well-validated and quantitated demonstration of tissue-specific activation of a "UFlip Off" allele, perhaps demonstrating rescue and lack of rescue of tissue-specific specific mutant phenotypes by activation using different cre drivers.

Reviewer #2 (Public Review): 

To put their study in context, Vance and colleagues state (lines 83-84) that "the sensory feedback that echolocating predators receive from movements of their prey has received little attention." Although there has been considerable investigation of echolocation in odontocetes (toothed cetaceans) as well as in bats-which the authors point out together constitute a remarkable one fourth of all mammals-there have been no successful attempts to compare the kinematic and neural tracking responses of echolocating predators with that of visual predators (such as in well-studied primates). 

The echolocation system of both odontocetes and bats works exactly like sonar: high frequency sonic pulses are emitted, and their reflections are detected to provide information about their surrounding environment, such as obstacles to be avoided or food items to be gained. It has long been recognized that when odontocetes and bats approach targeted prey, their distinct clicks are produced increasingly rapidly, producing, in one burst, what sounds to people like a single buzz. 

The so-called "interclick interval" [ICI] between the individual clicks within this buzz are extraordinarily brief-about 2 to 4 milliseconds. It has been assumed, but never really addressed previously, that odontocetes process this information quickly and make use of all the incredibly rapid clicks to adjust their approach to prey. However, this idea has not been properly studied via scientific investigation. Could it be that the brief IC intervals simply relate to how the clicks are produced, due to the anatomy of the nasal passages or the physiology of the odontocete sound production system? In other words, do these rapid clicks translate to rapid brain processing? This is an important question because primates and other visual predators also use rapid signal processing, but at a rate 10-20 times slower than the odontocete buzzes-about 50 to 200 milliseconds. This neural processing of visual predators (roughly a tenth of a second on average) has been described as "ultrafast." Could odontocete brains work even faster? 

Vance and colleagues collected data from digital logging tags affixed to the backs of two odontocete species: beaked whales that live in the deep open ocean, as well as trained captive porpoises that live in shallow coastal waters. The combination of two species (and habitats), and the inclusion of controlled study from the captive porpoises, is a real strength of this study. A potential weakness of this study could be that the tagged beaked whales were feeding in an uncontrolled setting. Relying on wild animals alone would limit the conclusions, as it is generally thought that predators use a feed-forward control system to anticipate prey movements and strike just as prey respond. Therefore, to bolster their investigation, the scientists conducted experimental trials on the trained porpoises in which the experimenters pulled on targets at varying speeds. This enabled the researchers to confirm that even sudden, unpredictable movements of evasive prey do not affect the results. The porpoises could respond to rapid movements of their intended targets, but the response latencies were not instant and did not correspond to individual clicks; instead, they spanned many clicks. 

The study’s small sample size, particularly of captive animals, with just two captive harbor porpoises along with six wild harbor porpoises and eight beaked whales, could be seen as another limitation. However, it is difficult to work with live cetaceans, and this sort of sample size is not unusual for biologging research. Nonetheless, it would be helpful to know more about the specimens. The data Vance et al. analyzed suggest that control bandwidths scale inversely with body length (e.g., with longer response latencies in larger animals). There was essentially no overlap in response times of the two porpoises, leaving one to wonder if one of the porpoises was notably larger. This sort of information would be useful to include in the paper. Also, potential extension of conclusions on response latencies to a range of other odontocetes, such as large sperm whales, would be useful. 

Nonetheless, the combined data strongly support the conclusion that odontocetes process sound signals about as fast as visual predators-in other words, much more slowly than the abundant acoustic signals arrive. This finding-that there is a confirmed mismatch between odontocetes' sound production (during close-approach buzzes) and their motor control-is important. The combination of similar data from wild beaked whales and captive porpoises in a controlled setting serves to strengthen the conclusions. Not only are these diverse, distantly related odontocetes both relying on the same mechanism, but they are also responding at roughly the same speed as predators that track prey visually. It is not hard to agree with the conclusion that mammals' biosonar response is akin to the visual response. 

Thus the researchers conclude that there is likely an "echokinetic" response similar to the well-studied "optokinetic response" of mammals. Perhaps this stems from common evolution of mammalian brains and neurophysiology before these types of predators diverged and evolved different habitats and prey capture methods, or perhaps it is due to simple limitations of neural processing. Either way, the authors present a solid case that "echo processing and control decisions during buzzes are decoupled from click rate" (line 270), which means that echolocation does not require extremely fast brain speeds.

Reviewer #1 (Public Review):

This manuscript summarizes a heroic effort by the Reiser and Wernet labs, and colleagues, a clear and well-balanced manuscript. I have no doubt it will become an important reference for anyone analyzing color or polarized light circuits in Drosophila.

It is intrinsic to the approach of anatomical connectomic data, that the core of the text is rather descriptive and simply summarizes the quantification and statistical analysis of synaptic connection between individual types of neurons. The most important points that makes this paper so interesting is the following:

– Many types of neurons within the lamina, medulla, lobula and beyond have been studied for their physiological responses or roles in visual behaviors. Thus, there is a strong framework to which the current study contributes and provides a rather extensive circuit structure.

– The structure how individual topics as subheading are organized is clear and repetitive. It allows a reader to clearly repeat the content of each section and also to look for relevant information in the respective section.

– As an elegant way to distinguish between pale and yellow PRs the authors made use of an identified accessory medulla neuron aMe12, which selectively innervates pale but not yellow ommatidial elements.

– The manuscript provides important information on the circuit architecture, commonalities and differences between pale versus yellow inner PRs (R7 and R8) and the DRA-R7/R8 pairs that provide one core element of polarized light sensing.

– There are several previously unknown neurons described here. Based on the connections it is possible to speculate about their function.

Reviewer #2 (Public Review):

This comprehensive manuscript can be regarded as an atlas of synaptic connections of photoreceptors specialized to detect color (R7/R8) and skylight polarization (R7/R8 in the dorsal rim area, DRA) in the fruit fly Drosophila. In contrast to previous attempts, the authors used a full-brain EM volume for the reconstruction of the pathways downstream of these photoreceptors and identified so far unknown photoreceptor targets and pathways into the central brain. Besides describing in detail the different targets of R7/R8 and R7/R8-DRA, the authors revealed highly interesting differences between the two pathways. While in the color vision pathway R8 input is dominant, in the polarisation vision pathway R7 input is dominant. Another interesting difference lies in the innervation of the lobula, which is prominent in the color vision pathway and virtually absent in the polarisation vision pathway. Both pathways innervate the central brain (e.g. the anterior optic tubercles), but their axon terminals remain spatially separated showing that polarization and color vision are processes in parallel.

The study is carefully performed, the methods are described in detail, the results are excellently documented and all conclusions are justified by the data. Furthermore, all data are included in the manuscript and supporting files.

In summary, I can say that this comprehensive catalog establishes a broad foundation for further studies into the mechanistic basis of color and polarization vision and its contributions to perception and behavior.

The manuscript has several strengths:

1) It provides the first connectomic data set for photoreceptors that process skylight information in any insect and will be important for developing refined models of skylight navigation.

2) It establishes the complete morphology of large, muli-columnar cell types that are targetted by the inner photoreceptors R7 and R8 but had eluded previous connectomic reconstruction efforts.

3) The EM data were supplemented by analysis on light microscopic level and compared with previous connectivity data from dense reconstructions generated from smaller-scale medulla volumes.

4) The study discovered important pathways from the inner photoreceptors to the central brain. Completely novel is a pathway formed by only three neurons per brain hemisphere (cell type aMe12) that connects pale R8 photoreceptors with the central brain.

5) The comparative interplay between data sets provides unique advantages and represents an important step in cross-validating and extending the applicability of all related data sets.

6) This atlas provides a valuable basis for all researchers interested in color and/or polarization vision.

Weakness of the study:

1) As true for all anatomical descriptions, the work is hard to read and understanding requires some prior knowledge.

Reviewer #3 (Public Review):

Connectomics analyses of neural circuit structure provide a durable description of neural anatomy that can both inform our understanding of circuit organization and inspire a wide range of subsequent studies. Here Kind, Longden, Nern, Zhao and colleagues provide a thorough reconstruction of the synaptic partners of a subset of photoreceptors in the Drosophila visual system, a widely used system for understanding circuit development and function. Using a reconstruction strategy that begins with individual photoreceptors and endeavors to identify all of their post-synaptic targets across a large brain volume, this work provides a detailed description of the core circuitry involved in processing spectral and polarized light inputs. A core strength of the work lies in its comprehensive characterization - I agree with the authors that they have achieved their goal of identifying all of the major synaptic partners of these cells, as the analysis is very carefully laid out and the data are thoroughly described. A second strength of the work is that, like the best of studies in the connectomics field, it opens up a number of intriguing experimental directions for future work, and presents the data in a form that can be readily accessed by future investigators. I therefore believe that the work will be widely used by a growing field.

Reviewer #3 (Public Review):

The paper by Kumar, Barkai, and Schiller, entitled "Plasticity of olfactory bulb inputs mediated by dendritic NMDA-spikes in piriform cortex" aims to understand LTP induction in intracortical (IC) and lateral olfactory tract (LOT) synapses onto layer 2 pyramidal neurons from the piriform cortex (PCx). Their findings uncover some of the location and pathway-dependent plasticity rules and challenge the notion that LOT inputs (carrying direct odor information from the bulb) become "hardwired" after the critical period. More specifically, Kumar et al., show:

1. LOT inputs onto distal apical (PCx) layer 2 pyramidal neuron dendrites do not trigger LTP when global STDP protocols are being used. However, strong LTP is induced when these inputs are co-activated with other LOT inputs capable of generating local NMDA-spikes delivered at theta frequency.

2. In contrast to LOT inputs, IC inputs directed to more proximal apical dendrites trigger the opposite: locally activated NMDA-spikes fail to trigger LTP, whereas a global STDP protocol – where synaptic inputs are followed by backpropagating action potentials – can effectively trigger LTP.

3. IC inputs in basal dendrites can potentiate both global STDP induction protocols and local NMDA-spikes

Strengths:

Understanding the dendritic mechanisms by which the PCx performs odor discrimination, recognition and memory are fundamental for understanding single-cell and network computations in odor processing. More specifically, this works focuses on understanding the synapses and pathways responsible for these plasticity changes, and the conclusions are clear and well-substantiated by the provided data.

This work challenges the notion that LOT inputs - inputs responsible for carrying information from the olfactory bulb as well as higher brain regions and thought to be important for odor recognition - onto pyramidal neurons become "hardwired" (LOT synapse stability in adulthood) after the olfactory critical period. Importantly, their data clearly demonstrate that LOT inputs onto distal apical dendrites can undergo LTP when these inputs are co-activated with other LOT inputs capable to generate local NMDA-spikes. These data help to reconcile seemingly conflicting previous findings.

Weaknesses:

Major issues:

– Novelty: 1) It is well established that postsynaptic depolarizations given by dendritic spikes (i.e. NMDA-spikes) can trigger LTP in several cell types (i.e. Major et al., 2013; Golding et al, 2012; Remy and Spruston, 2007; Gambino et al., 2014). 2) The finding that LOT inputs onto distal apical layer 2 pyramidal neuron dendrites from PCx do not trigger LTP when global STDP protocols are used corroborates previously published findings [Johenning et al., 2009]. 2) The demonstration that IC inputs trigger LTP via global STDP protocols in proximal distal dendrites also corroborates previous findings [Johenning et al., 2009].

– The most interesting/puzzling finding is how even applying up to 23 NMDA-spikes at 4Hz at more proximal apical locations via activation of IC synapses completely failed to induce LTP of these IC inputs. However, global STDP can effectively trigger plasticity in these synapses. Unfortunately, the authors didn't explore the mechanisms of why distally located LOT synapses can trigger strong LTP via NMDA-spikes, but those more proximally located IC synapses cannot. These mechanisms should be explored, especially since they challenged the local depolarization and calcium influx properties of LTP induction. An experiment to study the role of active conductances that can explain these puzzling results should be designed, including imaging of local calcium concentration using the same tools as those described in the Methods section (although not presented in the results section) and measurements of local voltage changes using dendritic patch recordings (a technique for which the lab is well known).

– A demonstration that NMDA-spikes can occur in vivo in the apical and basal dendrites of PNs from PCx (i.e. during odor discrimination and plasticity task) would greatly strengthen their in vitro findings indicating that LTP can be triggered in LOT-synapses and IC synapses directed to basal dendrites when driven by NMDA-spikes. This is important since LOT synaptic contacts onto distal tuft dendrites of pyramidal neurons are few (~ 200 total contacts, Miyamichi et al, 2011) and sparse (Davison and Ehlers 2011). Hence, for the reported NMDA-spike-dependent plasticity observed in vitro to be the modus operandi for plasticity and memory formation in vivo the need for a significant amount of synchronously activated LOT inputs directed to >20 clustered spines in the apical dendrites of PNs from PCx would be required according to the presented data. Or at least, provide a more extended discussion on this issue.

Reviewer #1 (Public Review):

With this manuscript, Kumar at al. study how plasticity can be achieved at the synapses between the lateral olfactory tract (LOT) and pyramidal neurons in the piriform cortex. These pyramidal neurons receive LOT inputs distally and intracortical input more proximally. The authors use patch clamp electrophysiology and electrical stimulation or optogenetics to show that local dendritic NMDA spikes are required to induce plasticity at these synapses, whereas a protocol (spike-timing dependent plasticity, STDP) that pairs pre- and post-synaptic activity is ineffective. On the other hand, they confirm previous findings that the STDP pairing protocol induces plasticity at intracortical synapses, whereas NMDA spikes are ineffective in this case. Finally, an STDP pairing protocol and NMDA spikes, when applied separately, were both able to induce plasticity at synapses on the basal dendrites although at a lesser degree than in the apical dendrites.

The results are clearly described in the text and figures, and support the conclusions drawn by the authors. This work should therefore clarify some discrepancies in the previous literature about plasticity of excitatory inputs to pyramidal neurons of the piriform cortex, as discussed in the paper. It also fits well with reports from other neurons that show different plasticity requirements for spatially segregated inputs arriving to a specific neuron.

Reviewer #2 (Public Review):

This manuscript uses whole-cell recordings and local synaptic stimulation to characterize the plasticity rules in pyramidal neurons in piriform cortex. From previous work it was typically thought that the ascending sensory inputs to piriform from olfactory bulb were plastic early during development but then became largely fixed during adulthood, with learning depending instead on changes in intracortical synapses. The authors previously showed that distal dendrites show robust NMDA-dependent dendritic spikes, a finding that also overturned previous thinking for these neurons. Here they show that even a small number of distal NMDA spikes causes high levels of plasticity that can more than double the strength of synaptic inputs from olfactory bulb. Plasticity is demonstrated in several ways via different paradigms including selective ChR2 activation of LOT inputs and glutamate uncaging. Intracortical synapses, in contrast, do not change strength with NMDA spikes but are instead enhanced by spike-timing protocols. Ascending and intracortical synapses thus follow distinct rules for engaging plasticity. The data are clearly described and the effects are robust and compelling. Overall the results add to our knowledge of how activity and experience alter synaptic communication in the olfactory circuit. The authors propose that this form of plasticity could shape cortical odor processing in distinct ways than changes in intracortical synapses.

Reviewer #1 (Public Review):

Understanding the genetic basis of infectious disease is a challenge, and even large-scale GWAS for important public health burdens like malaria explain only a small percentage of overall trait variability. An alternative approach is to isolate aspects of disease progression that can be studied using in vitro assays. Here, Ebel et al adopt such an approach to investigate the contribution of common genetic variants (beyond well-established malaria-associated/RBS disease alleles) to variance in Plasmodium falciparium invasion and growth rates. In a panel of 121 donors, they show that RBC phenotypes and parasite invasion/growth rates in non-disease allele carriers largely overlap with the range of phenotypes measured in carriers. Most remarkably, their results suggest that RBC phenotype data on traits like fragility and deformability, combined with genotype data for genetic variants in a small set of 23 RBC proteins, can predict up to 83% of variance in Plasmodium growth rates.

This manuscript has several strengths. First, it highlights the importance of the normal range of phenotypic variation in non-disease allele carriers for explaining variation in malaria infection and growth, and shows that this variation is often not well-correlated with ancestry. Second, it suggests how in vitro studies of malaria infection might be useful in identifying genetic variants that are not captured in GWAS (e.g., because the disease trait modeled in GWAS reflects a complex mixture of many component processes that may be better isolated in experimental models). Third, it suggests that the expectation of higher malaria resistance allele frequencies in African ancestry populations relative to European ancestry populations may be overly simplistic, which serves as an important reminder of the complexity of human biological variation.

However, I also have some concerns about the main predictive model result. Although the parasite invasion/growth phenotypes are arguably simpler than an overall in vivo malaria disease phenotype, the reported 40 - 80% variance explained by the LASSO models strikes me as concerningly optimistic. Notably, the correlation in the growth phenotype for repeated samples from the same individuals (sampled weeks apart) is only rho = 0.34 (and for invasion, it is only 0.05). Given that a trait's repeatability is the upper limit to its heritability, and genetic prediction is based on a trait's heritable component, I do not understand how the trait prediction can be as strong as currently reported. Because the result is so striking, it will be crucial to perform true out-of-sample prediction to evaluate predictive accuracy and generalization error.

Assuming out-of-sample prediction holds up, it is interesting that the genotype data add substantially to predictive accuracy even after directly considering RBC phenotypes themselves. As the authors note, this result suggests that the mechanisms through which the genetic effects act are independent of the measured phenotypes. This prediction should be further evaluated (e.g., by assessing genotype-RBC phenotype correlations).

Finally, although the results suggest no polarization of allele frequencies by European versus African ancestry, this result should be interpreted with caution throughout the manuscript, since it's unlikely that the predictive variants identified by LASSO are in fact causal.

Reviewer #2 (Public Review):

In this comprehensive study, the authors investigate the association of P. falciparum's (pathogen responsible for malaria) ability to invade and grow in red blood cell with red blood cell traits and host genetics. By directly measuring the RBC phenotypes, exome sequencing, and parasite growth rate on these samples, they can better understand how genetic variation directly influences a parasites ability to invade and replicate within RBCs, as well as investigate the role of African ancestry in this relationship. I found the paper well-written with many of the conclusions supported by the evidence presented in the main text. However, I had a few concerns that should be addressed.

1. The authors note that there is one family (mother and five children) are not carriers of known genetic loci. Figure 5-figure supplement 4 shows that they have significantly different distributions than other non-carriers with regards to principal components and parasitic invasion and growth rate. My concern is that many of the tests in the manuscript assume independent observations and related individuals violate this assumption. The children should be removed from all analyses to test for the sensitivity of results to this structure in the data.

2. This is also related to the increase in % variance explained in their lasso models when including genetics. It would be useful to know how much of the outcome variation was from the inclusion of the principal components specifically (capturing the family) versus the variants of interest.

3. It would be helpful to know some more about the variants that were included from exome sequencing. This would include their allele and genotype frequencies, as well as the comparison with reference population frequencies.

4. Are the frequencies of known RBC disease alleles consistent with population estimates? It would be useful to assess the representativeness of the sample.

5. I would appreciate knowing a bit more about the difference between the two strains, one lab adapted and one clinical. Is it known how the lab strain was adapted or how representative it is to circulating strains? If so, may be worth describing in the discussion to explain the differences in results between the strains.

Reviewer #3 (Public Review):

The aim of this study by Ebel et al. was to investigate host genetic and phenotypic factors underlying variation in Plasmodium falciparum fitness in the red blood cell (RBC). They found that variation in common RBC phenotypes such as MCV, deformability, and hydration status contribute substantially to parasite fitness. The RBC phenotypic variations together with genetic variants in genes encoding important RBC proteins, explained 71-83% of variation in parasite growth. Importantly they did not identify large ancestry-specific effects on parasite fitness, highlighting that majority of common human genetic variation is shared among all human populations.

The conclusions of this paper are well supported by their data and analytical approach, but some aspects of the experimental design need to be clarified and expanded.

Strengths:

The main strength of this study is the pragmatic approach used to identify genetic and phenotypic factors underlying variation in parasite fitness in a small sample size: utilising the Lasso approach to guide their variable selection. Their hypothesis-led approach also ensures that their significant results are reliable.

Weaknesses:

The limitation to the targeted variable selection approach, which the authors have acknowledged in their discussion, is that it relies heavily on prior knowledge of genetic variants and RBC phenotypes, which might miss additional host factors that influence parasite fitness.

The authors used one laboratory parasite strain and one field isolate in the study, which might limit their conclusions on parasite fitness to strains that follow a similar invasion and growth profile.

The population studied (admixed African ancestry individuals in a non-malaria endemic setting), might also limit their conclusions on population divergence in genetic variants and RBC phenotypes with substantial effects on parasite fitness.

Reviewer #1 (Public Review):

The genetic code is not a "universal code", but alters in some organisms and organelles. Shulgina and Eddy developped a computational method, termed Codetta, to predict genetic code alteration from the input nucleotide sequences using a probabilistic modeling approach. The basic idea is not novel, yet, the present model differed from previous approaches by taking advantage of the extensive profile HMMs from the Pfam database. Codetta works fairly well in assigning and detecting possible codon reassignment in archaeal and bacterial genome sequences. This bioinformatic prediction should provide a plausible starting point for further experimental validations. Generally, this reviewer finds the research content presented here is well conducted and well written.

Major comments:

In the Introduction, the authors extensively review historical findings of non-stardard genetic codes. In my point of view, however, tRNA modifications play a major roles in reassigning the genetic code. It would be good to revise the introductory part including a role of tRNA modification.

The authors first validate Codetta's performance by predicting the codon decoding of 462 yeast species. Among them, the authors further analyzed S. malanga in which two tRNAs possibly responsible for Ser and Leu are encoded. They just confirm expression and aminoacylation by acid-urea Northern blotting, but they did not demonstrate they are actually Ser-tRNA and Leu-tRNA. To determine the species of amino acid attached to a specific tRNA, there are several methods. After deacylation reaction, total tRNA can be charged with Ser or Leu in the presence of S100 fraction of cell lysate. Then, you can detect the acylated band of the target tRNA by acid-urea northern blotting. Otherwise, MS-based proteomic analysis of cell lysate would be much easier to demonstrate the dual assignment in this yeast species.

In general, MetRS strictly recognizes anticodon sequence of tRNAMet for methionylation. If the authors' hypothesis is true, the anticodon-recognition domains of MetRSs in these species might be different from those of E. coli MetRS. It would be worth investigating and discussing to support their speculation.

Reviewer #2 (Public Review):

There are many known cases of small variations in the genetic code where one or more codons have bee reassigned in a particular group of organisms. This paper gives a method for scanning genome data to identify cases where a codon has been reassigned. It is tested by comparison with known code variants and also shown to identify several previously unknown cases. It promises to be a useful tool in the future.

Reviewer #1 (Public Review):

The authors carried out a post-hoc analyses of a protective gene expression signature previously observed in preclinical trials and clinical trials (RV144 and HVTN505) to identify a possible correlate of reduced risk of infection and whether able to provide a potential mechanism for protection. This monocyte signature they focus on was absent in the DNA/rAd5 human vaccine trial which did not show efficacy and was enriched in the partially effective RV144 human trial where the vaccine was and ALVAC/protein vaccine. Here they indicate that the signature is a correlate of reduced risk of infection.

Identifying signatures of protection is an important issue in the development of a HIV vaccine, and signature analyses might be important to reveal a few markers that might be selected to evaluate vaccine trials. However, this analysis must be able to point to very few genes, as single cell analyses are not an option in a large clinical vaccine trial.

It is unclear whether or how the conclusion of the previous publication by many of the same authors of this paper, including the senior author (Ehrenberg et al., 2019, identification of a gene signature in B cells that is associated with protection from SIV and HIV infection providing a new approach for evaluating future vaccine candidates) is compatible with this new one: signature primarily expressed in myeloid lineage being the one most consistently associated with vaccine efficacy. It is unclear which one of the two is correct or how they are reconciled. Was the single cell analysis done in monocytes only for this paper or simply not reported in the studies of Ehrenberg et al., 2019?

Figure 1: The gene expression score (GES) of this figure does not seem to be for a specific cell type. It is unclear how the GES reported here relates to the final GES of monocytes. What is the utility of this analysis? Can we observe here the same most significant genes that we observe in monocytes? This is important because if bulk analysis gives the same results as looking at monocytes an eventual marker identified in monocytes could be evaluated in luck analysis.

Figure 2: it would be good to know whether the subset of the 63 genes can be restricted to the most significant and their GES can still retain the predictive value.

Figure 3 deals with genes associated with antibody dependent cellular phagocytosis (ADCP). Can one derive a gene or a few genes that are predictive of significant ADCP?

Reviewer #2 (Public Review):

Shangguan, Shida et al report here that a "protective" gene set correlating with decreased risk in sorted B-cells from NHP immunized with Ad26 regimens, is also enriched in microarray of total in cultured PBMCs from human immunized with the moderately protective ALVAC-based vaccine in the RV144 HIV vaccine trial, but not in HVTN505, that afforded no protection against HIV.

Strengths:

The analyses of the efficacy of different HIV vaccine candidates across different species is challenging but is an important goal. Important findings of the current work stem from analysis of sorted B-cells and monocytes from the non-efficacious trial, HVTN505 and from the single cell analyses from peripheral blood in the RV306 immunogenicity human trial, where volunteers were immunized with the same regimen used in RV144 and received an additional vaccine boost. They found that the gene set that correlated with protection in RV144 was associated with monocytes, and not B-cells as previously thought and importantly, that the gene set correlating with protection in RV144, was not enriched in the HVTN505 failed trail. In addition, in the RV306 immunogenicity trial they found that this gene set correlated also with antibody-dependent cellular phagocytosis (ADCP). This finding supports the hypothesis that ADCP could be a mechanism of protection against HIV, a testable hypothesis in future efficacious HIV vaccine trial.

Weaknesses:

The analysis of data from additional failed trails in NHP could strengthen the authors' conclusions

Reviewer #3 (Public Review):

Strong points:

1. This provides a novel mechanism into the RV144-mediated protection of HIV acquisition.

2. The analyses are robust and statistically sound.

3. The flow of the paper/figures is easy to follow.

Weak points:

1. the RV306 trial (Figure 3 A and B) RNA-SEQ analysis vs ADCP could benefit from a little more information:

Are the 118 / 93 genes at Wk2 / Day 3 post-vaccination overlapping a lot ?

What are those genes ? Do they play a known direct role in ADCP or are they upstream regulators? Perhaps a heatmap representation with the ADCP as an annotation track would help unfamiliar readers better understand.

2. I would nuance that ADCP is "A" primary mechanism, not "THE" (title). There could be more potent unidentified mechanisms, so the usage of "THE" in the title is in my opinion premature.

3. While I agree that it is possible that ADCP is a primary mechanism with the previously identified transcriptomic signature given the evidence, we cannot exclude that the signature in fact represents an upstream regulator of ADCP, inducing a myriad of cascades contributing to vaccine-induced protection. If that were the case, ADCP could be higher in individuals with higher protection without it being directly involved in that protection (more of a collateral effect).

Showing an enrichment of ADCP-associated genes from external datasets with the tested gene signature would strengthen at least partly that this is a direct phenomenon.

Otherwise, I would nuance the statement and say that ADCP is a likely/potential mechanism of vaccine-induced protection.

4. Observations in Figure 4 are glanced too quickly in the Results section: this would require a more in-depth description.

5. It is not clear whether the expression level per monocyte for the subset of genes tested in the CITE-seq data is different in patients with higher ADCP vs those with lower ADCP, or is the differential enrichment the result of a different number of cells that express this signature ? Or both?

Reviewer #1 (Public Review): 

This study set out to test whether Eurasian jays take into account the perspective and desire on an observing bird during food caching. The inclusion of multiple cues associated with different mental states is a novel and valuable approach to the field of comparative Theory of Mind and social cognition. The rigorous, elegant design over five experiments aims to pinpoint what mechanisms and cues the jays use when choosing where to cache what food and how much depending on the observer's perspective (caching location out of view or not) and current desire (being pre-fed on a single food type). The reader is guided through complex experimental procedures with clear descriptions of predictions and helpful figures. The manuscript is well written and stays on topic, while also expressing important concerns about the replicability and validity of studies in comparative cognition in general. The results go against many earlier papers by some of the authors, and it is commendable that they set out to replicate their own studies in the first place. This kind of critical assessment of earlier research and presentation of negative results (especially in a replication study of positive results) is certainly welcome after the concerns commonly expressed that this is not done sufficiently often. The results support the conclusion and no inflated interpretations are presented. Other caching corvids can be tested with this method after suitable adjustments, and the general design that includes both perspective and desire of observer has wider uses also in other taxa. The explanation that results did not replicate due to age or experience of the birds is compelling (as is the effect size argument) and would be an interesting avenue for future research if possible. Although it remains an open question why the results differ from those of earlier studies, and what mechanisms the jays employ when caching in the presence of a conspecific, this excellent study will hopefully set the example for many future studies that cover multiple lines of evidence across several experiments to critically examine the validity and replicability of the science of comparative cognition.

Reviewer #2 (Public Review): 

In the present manuscript, Amodio and colleagues investigate whether or not Eurasian jays use cues correlated with the perspective and desires of a potential competitor to shape their caching behavior. The first two studies build on past research to test whether Eurasian jays can integrate cues regarding both the competitor's perspective and desires. The authors fail to find evidence that Eurasian jays integrate these cues to cache undesired food in unobservable locations. Thus, the authors attempt to replicate their own earlier findings showing that Eurasian jays' caching behavior is impacted by these cues when the competitor's perspective and satiety are manipulated in separate tasks. The authors fail to replicate these effects and conclude with a discussion of possible reasons for these failed replications and suggestions for increasing replicability in comparative psychology. 

Strengths: 

The logic of the experiments is very clear. Beginning with the two novel studies, the authors provide convincing justification for why cue integration is an interesting question to investigate in this species. Additionally, the initial null results for Studies 1 and 2 led logically to the replication attempts for the individual effects of competitor perceptual access and desire on caching behavior. 

Notably, several of the authors on this manuscript are also authors on the studies that failed to replicate here. These authors were thus able to discuss the nuances of the past and present experiments in great detail and offer a well-reasoned perspective on why these replication attempts may have failed. This combined with their broader discussion of the importance of replication in comparative psychology make the manuscript even more impactful contribution to the literature. 

Weaknesses: 

The authors find null results across the board on all five studies, yet they do not have a section of the discussion that explores the implications of what it would mean for Eurasian jay cognition. Do the current findings mean that jays cannot, in fact, track the perspectives and desires of conspecifics? Should the current findings prompt a replication attempt of the desire tracking of mates from Ostojic et al., 2014? What other cues might affect this behavior? Etc.). Obviously, the absence of evidence does not mean evidence of absence but the authors should provide a thorough discussion of the possibility that these replication attempts may not simply be "local failures." Given the preponderance of null results (and the fact that these result could be the first null results on this topic published in a high-impact journal), the authors should at least briefly discuss what it would mean for the avian cognition if these replication failures are not just isolated cases but instead indicative of a broader pattern in bird perspective taking capacities. 

In the authors' discussion of why these replications may be inconsistent with the rest of the literature on this topic, they cite the subjects' age and life experience as reasons why they may have failed on these replication attempts. However, these suggestions feel under-explored and therefore tenuous. The authors should provide more context for this claim. For example, are there any empirical or anecdotal data supporting the proposal that older Eurasian jays are less motivated to guard their caches (or are there any related social changes experienced by older Eurasian jays)? Similarly, to what extent does the experience of living in the aviary reflect or diverge from the caching and pilfering experiences of wild Eurasian jays? For example, do birds in the aviary frequently cache and pilfer even though they receive a maintenance diet? (Given the role of experience in scrub jay caching behavior suggested by Emery & Clayton, 2001 additional context here would be informative). 

Impact: 

This work is a well-thought-out series of studies that performs the valuable function of attempting to replicate previously well-cited findings within the field of comparative cognition. Unfortunately, these findings are null results, but given the uniqueness of this captive population of Eurasian jays, the publication of these findings is of critical importance to our understanding of this phenomenon. Additionally, the format of the paper (two new studies, followed by replications) serves as a nice example for how replication attempts can be integrated into novel investigations of comparative cognition.

Reviewer #1 (Public Review): 

This manuscript describes a novel tool for tracking synaptic plasticity at the single synapse resolution with a SEP-tagged GluA1 receptor. The authors rather convincingly demonstrate that this tool does not disturb synaptic physiology or mouse behavior. They also show that this tool can be used to measure the distribution of synaptic weights and its variation during a plasticity protocol in barrel cortex. This tool is useful for more quantitative measurements of synaptic strength in vivo. The main weakness of the method is related to the density of marked synapses which makes the tracking difficult with only 80% reproducibility, probably due to the resolution limits of 2P-microscopy. It could however be improved with in vivo superresolution technique. The other limit of the method is that it is not demonstrated to allow longitudinal studies at the single synapse resolution. The authors do not discuss this issue in detail. It seems feasible with additional markers of the dendrite and spines. But this is not developed in the manuscript. Also, it is not demonstrated to which extent this technique outperforms traditional methods for synaptic weight measurements like spine volume.

Reviewer #2 (Public Review): 

Over the last couple of decades, the development of fluorescent transgenic mouse lines (e.g thy1-GFP) and delivery techniques (e.g., in utero electroporation), as well as the democratization of recording methods in living animals (such as calcium imaging, high-density probes,) have strengthened the link between synaptic plasticity and behavior. Nevertheless, these methods are most of the time limited to hundreds of cells at best (very few in the case of patch-clamp recordings), or failed to achieve a clear synaptic resolution without affecting the tight equilibrium of endogenous proteins. 

In this paper, Graves, Roth, Tan, Zhu, Bygrave, Lopez-Ortega et al present an additional high-resolution optical tool to overcome these limitations. They generated a new knock-in mouse line that fluorescently labels all endogenous AMPA receptors (KI SEP-GluA1). In this mouse line, the extracellular N-terminal domain of the GluA1 subunit of AMPAR is tagged with super ecliptic pHluorin (SEP), a pH sensitive variant of GFP that fluoresces at neutral pH (at cell surface) and is quenched at acidic pH (within the cell). This tool thus avoids the use of antibodies or the over-expression of exogenous tagged receptors. 

They perform a set of convincing experiments showing that synaptic transmission, homeostatic and activity-dependent synaptic plasticity in vitro (Figs2-3), and behavior (Fig. 4), are not affected in KI SEP-GluA1 as compared to wild-type mice. Despite the obvious quality and viability of this mouse line (this is an important tool with no doubt), it is puzzling however that, while the level of GluA2 remains unchanged, the global expression of SEP-GluA1 is twice as low as the expression of GluA1 in wild-type mice (Fig.1). The manuscript would benefit from a clear brain-region specific comparison between the expression pattern of GluA1 and SEP-GluA1. At least, the author should discuss this point, and how this might affect the formation of GluA1/A2 heteromers, the dominant form of AMPAR in pyramidal neurons. 

Then, they provide strong evidence that SEP-GluA1 receptors are mobile in vivo (Fig. 6) and can thus accurately report synaptic plasticity, at least in anesthetized animals (Figs. 7-8). The authors make a point that "this novel SEP-GluA1 knockin mouse is the first tool that enables longitudinal tracking of synaptic plasticity underlying behavior at brain-wide scale with single-synapse resolution". However, given the high density of fluorescent synapses, it remains unclear how effective would be this mouse line in awake mice, and more specifically during behavior, in which movement artifacts could preclude the tracking and registration of the same population of SEP-GluA1 containing synapses over time. 

Finally, they present a new automated analytical tool to detect and register fluorescent synapses (Fig.7). Although the initiative is important and laudable (it is true that imaging approaches are usually plagued by the lack of user-friendly analysis tools), the method would benefit from a comparison with existing methods.

Reviewer #3 (Public Review): 

Understanding the distribution of synaptic strength and plasticity in the brain is paramount for understanding neural circuit function underlying behavior. In the manuscript by Graves et al., the authors developed a novel mouse model for optical detection of synaptic strength and plasticity in live brains. Specifically, they modified the mouse genome by modifying the native GluA1 AMPAR subunit gene (gria1) with a pH-sensitive GFP (pHluorin) -tagged GluA1 at its N-terminus. This sensor is nearly maximally fluorescent when the pH is neutral and quenched in acidic environments and, therefore, preferentially marks AMPARs located on the plasma membrane. Since AMPARs are known to cluster in the postsynaptic density, AMPAR number is the predominant postsynaptic determinant of synaptic strength. Specifically, the trafficking of GluA1 AMPARs is responsible for LTP in CA1 hippocampus. The use of this novel genetic tool raises the possibility of monitoring synaptic strength optically, thus providing a strategy for massively parallel assessment of the distribution of synaptic strength in bulk brain tissue. Even more promising is the use of cranial windows and 2-photon microscopy to assess synaptic strength longitudinally, for example, during learning acquisition. 

The authors performed a comprehensive and rigorous set of control experiments using electrophysiology and behavior experiments. They demonstrated that modified GluA1 acts as native receptors and does not suffer from the shortcomings of overexpression approaches. The authors convincingly demonstrate that the modified receptors generate normal wild-type synaptic physiology and no behavioral alterations. Using glutamate uncaging, they showed that fluorescence changes at synapses were highly correlated with an electrophysiological assessment of synaptic strength and plasticity. Thus the data support claims that synaptic strength and plasticity could be assessed and monitored at unprecedented parallelization. 

Whether SEP-GluA1 can be used to quantify synaptic strength and its changes is uncertain due to an unknown ratio of GluA1/2 versus GluA2/3 receptors, the differential expression of GluA1 in different cell types, and the presence of GluA1-independent plasticities. Another potential shortcoming of the study is the lack of a ground truth demonstration of true synapses in vivo. Given the high-density of synapses, low z-resolution 2P microscopy (> 2 um), and the presence of a significant extrasynaptic pool, a confirmation of their results with superresolution or EM would be essential. Moreover, the lack of cell-type-specific labeling is likely to limit the tool's use for linking behavioral and microcircuit synaptic plasticity. It is possible that control experiments in acute brain slices could circumvent some shortcomings and provide a more quantitative workflow.

Reviewer #1 (Public Review): 

The authors of this paper use contemporary circuit-mapping techniques to identify the excitatory and inhibitory inputs to AVP neurons in the hypothalamus. They then use chemogenetic methods to inhibit potential inputs to AVP neurons while imaging calcium activity of the AVP neurons by fiber photometry to identify the relevant inputs for pre-systemic inhibition of AVP neurons. They conclude that the presystemic regulation AVP neurons by drinking and eating are mediated by non-overlapping neural circuits - drinking inhibits AVP neurons via decreased activity of glutamatergic MnPo/OVLT and SFO neurons and an increased activity of GABAergic MnPO/OVLT neurons. Whereas AVP neuron activation in response to eating comes from unknown neurons in the arcuate nucleus. Overall, this is a comprehensive analysis of the neural circuitry controlling presystemic AVP neuron activity. 

A major disappointment in the study is the failure to identify the neurons in the ARC thT provide excitatory input to AVP neurons in response to eating. The authors suggest that the rapid activation of AVP neurons is likely to glutamatergic (lines 492-497) and go on to suggest the Oxtr-expressing neurons are a good candidate. Thus, it is surprising that they tested Pomc- and Agrp-expressing neurons and not the glutamatergic Oxtr-expressing neurons.

Reviewer #2 (Public Review): 

This study investigates the important question of which afferent circuits are responsible for the rapid, presystemic regulation of AVP neurons by food and water. Using a combination of rabies tracing, CRACM, and chemogenetic silencing in conjunction with photometry, the authors conclude that the MnPO transmits presystemic fluid signals whereas food signals are relayed by unidentified neurons in the ARC. The paper presents a collection of useful data investigating the anatomic connectivity between cell types in the LT and elsewhere and tests a broad range of possible inputs that could mediate these presystemic effects. However there are interpretational problems with the experiments that investigate functional connectivity (which pathways transmit which signals). 

Major points:

1) The conclusion that the ARC is the source of the presystemic signal that activates AVP neurons in response to food is based on an experiment in which the effect of silencing the VMH and DMH individually on a photometry trace is subtracted from the effect of silencing DMH, VMH and ARC simultaneously. This is problematic given the broad and variable spread of such injections, the difficulty of completely and selectively hitting a single nucleus, and the lack of characterization. It also seems likely that there could be synergistic effects of inhibiting multiple adjacent nuclei. From the data in Figure 7F, it appears that most of the effect is driven by a few animals, which also raises the question of sources of variability. 

2) The effect of silencing DMH/VMH/ARC on food intake is not reported. If these mice eat less or more slowly, this would explain the partial reduction in the presystemic activation of AVP neurons. 

3) It is not clearly reported to what extent the chemogenetic silencing of the MnPO/OVLT in the mice used in Fig. 2 and 5 reduces the amount of water consumed and how this relates to the dynamics in each animal/trial. This is confounding in two ways. Given that the silencing does not fully block drinking, this implies that the MnPO/OVLT silencing is incomplete (based on Augustine 2018), and thus the negative photometry result in Fig. 5 is hard to interpret. Conversely, if silencing reduces drinking partially, which seems likely, then this behavioral change could account for the reduced presystemic inhibition of AVP neurons. It is hard to see how the direct effect of MnPO on AVP neural dynamics could be separated from its effects on behavior in this experiment. While there is a pre-ingestive response in the AVP neurons (which does not have this confound), in several experiments in Fig. 2 this is approx. 1% dF/F. 

4) MnPO neurons are heterogeneous in their dynamics, especially the MnPO-GABA neurons, and for this reason ruling out a possible mechanism based on a photometry trace is challenging. For example, compare the interpretation of the photometry recordings of MnPO-Glp1r neurons in (Augustine, 2018) with the results of single cell imaging of the same neurons in (Zimmerman, 2019)).

Reviewer #3 (Public Review): 

This manuscript by Kim and colleagues explores the circuit that communicates food- and water-intake-related presystemic regulations to vasopressinergic endocrine output neurons. Although previous work from several labs had observed food and water intake related anticipatory signaling in cell types in several lamina terminalis (LT) nuclei, the functional significance of these remained unexplored. Here, the authors demonstrate that the neural circuits underlying foo- and water-related presystemic signals are anatomically dissociable at the level of the vasopressin secreting endocrine output neurons. The authors use viral retrograde tracing to identify candidate anatomical regions that could communicate food and water intake related anticipatory signals to VP neurons in the SON and PVN. They show that excitatory neurons in LT nuclei SFO and MnPO/OVLT and inhibitory neurons in the latter make direct synaptic connections onto VP neurons in the SON and PVN. They also perform chemogenetic silencing experiments to elucidate the functional importance of LT and other brain structures for presystemic VP regulation. They show that MnPO/OVLT is important for water drinking but not food intake related presystemic regulation. Furthermore, the authors survey several brain regions that could provide the food intake-related input to VP neurons identifying the arcuate nucleus as the likely source. The experiments in this paper are generally rigorous. Addressing the following points should improve the manuscript: 

• In Figure 2, DREADD was used to suppress the activity of the LT. However, the virus construct is a general promoter, and no data is provided to demonstrate that CNO/DREADD works in this system or cells. In particular, there is no behavioral effect by CNO inhibition of SFO or MnPO/OVLT. To confirm these negative data, slice ephys or similar method should be used to confirm the efficiency of chemogenetic manipulation.

• Although the authors revealed the anatomic sites relevant for different kinds of presystemic regulation of VP neurons, the causal role of specific cell types in these structures that provide this input remains untested/unclear. The manuscript would be significantly more impactful if this was addressed. <br> Specifically, whether excitatory and inhibitory populations in the MnPO/OVLT indeed mediate the pre- and post-ingestive effects on presystemic VP neuronal activity as suggested by GCaMP imaging should be tested.

• The authors rule out two cell populations in ARC as a potential source of food-related presystemic effects on VP neurons. They do suspect a specific cell type in ARC (OXTR+ neurons), which should be tested.

Reviewer #1 (Public Review):

This manuscript was well written and interrogates an exciting and important question about whether thalamic sub-regions serve as essential "hubs" for interconnecting diverse cognitive processes. This lesion dataset, combined with normative imaging analyses, serves as a fairly unique and powerful way to address this question.

Overall, I found the data analysis and processing to be appropriate. I have a few additional questions that remain to be answered to strengthen the conclusions of the authors.

1. The number of cases of thalamic lesions was small (20 participants) and the sites of overlap in this group is at maximum 5 cases. Finding focal thalamic lesions with the appropriate characteristics is likely to be relatively hard, so this smaller sample size is not surprising, but it suggests that the overlap analyses conducted to identify "multi-domain" hub sites will be relatively underpowered. Given these considerations, I was a bit surprised that the authors did not start with a more hypothesis driven approach (i.e., separating the groups into those with damage to hubs vs. non-hubs) rather than using this more exploratory overlap analysis. It is particularly concerning that the primary "multi-domain" overlap site is also the primary site of overlap in general across thalamic lesion cases (Fig. 2A).

2. Many of the comparison lesion sites (Fig. 1A) appear to target white matter rather than grey matter locations. Given that white matter damage may have systematically different consequences as grey matter damage, it may be important to control for these characteristics.

3. The use of cortical lesion locations as generic controls was a bit puzzling to me, as there are hub locations in the cortex as well as in the thalamus. It would be useful to determine whether hub locations in the cortex and thalamus show similar properties, and that an overlap approach such as the one utilized here, is effective at identifying hubs in the cortex given the larger size of this group.

4. While I think the current findings are very intriguing, I think the results would be further strengthened if the authors were able to confirm: (1) that the multi-domain thalamic lesions are not more likely to impact multiple nuclei or borders between nuclei (this could also lead to a multi-domain profile of results) and (2) that the locations of these locations are consistent in their network functions across individuals (perhaps through comparisons with Greene et al., 2020 or more extended analyses of the datasets included in this work) as this would strengthen the connection between the individual lesion cases and the normative sample analyses.

Greene, Deanna J., et al. "Integrative and network-specific connectivity of the basal ganglia and thalamus defined in individuals." Neuron 105.4 (2020): 742-758.

Reviewer #2 (Public Review):

Hwang et al. compared the neuropsychological deficits due to isolated thalamic lesions (n=20), to size-matched lesions outside the thalamus (n=42), which revealed significantly greater likelihood of domain-general deficits due to thalamic lesions. They found the left anterior-medio-dorsal thalamus to be the most likely lesion site to cause widespread cognitive deficits, highlighting its importance as a network hub. Drawing on their large patient database, Hwang et al. were able to verify this finding by also comparing to a larger sample (n=320) of non-size-matched control lesion patients. They utilized the NeuroSynth task fMRI database to show that their lesion-cognitive deficit mapping results were less likely to be due to thalamic lesions damaging multiple functional networks, and more likely due to the thalamus in general and the anterior-medio-dorsal thalamus in particular serving as a network hub. They then analyzed resting state functional connectivity data, which as hypothesized revealed the anterior-medio-dorsal thalamus to have the highest participation coefficient, an indicator of hubness. Finally, they analyzed Allen Brain Atlas data to show that the same segment of the thalamus preferentially expresses CALB1 in comparison to the remainder of the thalamus which expresses more PVALB. This article combines lesion, task fMRI, resting state functional connectivity and gene expression data, all of which strongly point towards the anterio-medio-dorsal thalamus as an integrative hub for cognitive function. This work has implications for making clinical prognoses following thalamic lesions and it strongly suggests that the anterior-medio-dorsal thalamus makes domain general computations to cognition.

Reviewer #1 (Public Review):

The manuscript presented describes the transcriptome of primitive hematopoietoc stem and progenitor cells harvested from untreated control or mice treated with either PGE2, G-CSF, pIpC or indomethacin. These are some of the drugs commonly used to generate experimental models of stressed hematopoiesis. Having observed some patterns of responses and the transcriptomic level, the authors ask whether these may be driven by specific chromatin accessibility patterns in stem and progenitor cells subset. However, ATAC-seq reveals that this is not the case when directly responsive genes are analyzed, and rather differences can be found in the promoter accessibility of genes further downstream.

Strengths. The authors analyze large and challenging datasets, where relatively minor differences in transcription and chromatin accessibility patterns are highlighted.

Weaknesses. The choice of stimuli is somehow arbitrary, and the description of the data presented in the figures is often hard to follow, with some contradictions present and text and figures being ordered differently.

Reviewer #2 (Public Review):

Fast et al here describe responses by hematopoietic stem and progenitor cells to niche signals using scRNseq and ATACseq. The data provide a rich resource to the research community demonstrating a number of distinct cell states, and heterogeneity between cell clusters in their responses to external stimuli. Notable observations are the continuum in cell states among HSCs and LSK cells, and the distinct clusters that are marked by interferon signaling response as opposed to AP-1 family / PGE signaling. This paper is a resource paper that will serve as a starting point for future studies - in depth studies were not undertaken to validate or understand the implications of these findings on disease states or developmental outcomes, although such studies would certainly increase the impact of the work if they were available.

Reviewer #3 (Public Review):

Understanding the molecular determinants controlling hematopoietic stem cell (HSC) biology is critical for myriad clinically-relevant interventions; however, because HSC are rare, this information is limited. Here the authors exploit their considerable facility with HSC isolation and apply single-cell genomics to provide a profile of both normal HSC transcriptional clusters and HSC relevant perturbations (di-methyl-PGE2 vs. the Cox1/2 inhibitor indomethacin, and G-CSF stimulating mobilization, or the TLR3 ligand poly(I:C)) and identify potential underlying regulatory transcription factors based on in silico analyses. They note that they can understand the perturbations as shifts in cells within the unperturbed clusters (with modest gene expression changes in each cluster). There are some aspects of the work that could be changed to improve impact and to clarify the take-home message.

The manuscript leaves the reader with the expectation that the work will biologically dissect the normal and perturbed cluster/populations. This is probably because the authors do not sufficiently clarify the biological impact of the manipulations, the depth of the published record on them, and then convey the expected versus observed transcriptional changes based on that prior published record. In addition, the transcriptional changes in each cluster within the heatmaps relegated to supplementary data probably provide the essential information, but they fail to represent the data across all clusters with all differentially expressed genes to demonstrate common or distinct gene expression changes. This would best be consolidated to a heatmap of differentials instead of the current method of clustering the actual expression metric. To be clear, it would significantly improve the work to show all differentially expressed genes in each HSC cluster across all perturbed clusters in a single heatmap. A viewer other than a genome browser session (which is not easily maintained) would be an essential improvement.

The central claim is that "niche signals regulate continuous transcriptional states in hematopoietic stem cells". As an experimental paradigm, the authors inject mice with different molecules and then purify HSC two hours later to examine changes in gene expression. This experimental paradigm does not represent specific perturbations of niche signaling.

Reviewer #1 (Public Review):

The presented results are convincing, but how diffusely located proteins might accomplish this task is still unknown. The discussion section should be tightened up; many non-pertinent or poorly supported things are discussed, whereas, for example, the anomalous finding that T-DNA insertion alleles are hypomorphic (lines 164-167) are not. This is important given their assertion that the amount of protein influences the phenotype.

Reviewer #2 (Public Review):

Plants have an amazing diversity of pollen morphologies. This study set out to determine how apertures, the gaps in pollen exine wall, are specified during pollen development. The authors focused on the macaron (mcr) mutant that was identified in a previous forward genetic screen to have one aperture that extends around the circumference of the pollen grain, instead of the 3 equidistant apertures in normal Arabidopsis pollen. They identify the MCR gene as ELMOD B, a member of a small gene family with domain similarity to the animal Engulfment and Cell Motility domain (ELMOD) protein , which has been shown to be non-canonical GTPase Activating Proteins (GAPs). Genetic dosage experiments showed that increasing the expression levels of MCR and the closely related ELMOD_A gene in developing pollen also increases the number of apertures. Combined with epistasis analysis showing that MCR is upstream of INP1, INP2, and D6PKL3 (previously published regulators of aperture development), this data provides good evidence that MCR and ELMOD-A are major regulators of the number, positions, and size of pollen apertures.

In the second part of the manuscript, the authors present a phylogenetic analysis of ELMOD proteins in plants. They show that the plant ELMOD genes likely have a common ancestor and that Angiosperms have four distinct ELMOD clades. They hypothesize that the A/B clade is necessary for aperture formation in Angiosperm pollen and that many angiosperms have more than one A/B ELMOD gene in order to provide redundancy for pollen development and/or other important functions. While intriguing, a weakness in this part of the study is that they did not test this hypothesis by checking to see if A/B clade ELMOD genes are expressed during pollen development in other Angiosperm lineages.

In the final experiments, the authors analyzed the predicted GAP domain for amino acids that are highly conserved in Angiosperm ELMODs and that are specific to different clades. They identified a conserved Arginine in the same position of the GAP domain as in animals. This arginine is necessary for GAP function in animals. The authors predicted that this Arginine would also be important for Arabidopsis ELMOD function and mutated this residue to Lys in MCR and ELMOD_A. Neither of these versions could complement the mcr aperture phenotype, confirming their hypothesis. One limitation of this experiment is that it only indicates that the domain might have a similar function to the animal ELMODs but does not directly test whether MCR and ELMOD_A actually have GAP activity.

The most intriguing data comes in the final figures of the paper, where the authors compare the GAP domains in the different ELMOD clades. Sequence comparisons revealed that the A/B clade ELMODs tend to have a glycine at position 129 within the GAP domain, while clade E ELMODs have a cysteine at this position. They predicted that this amino acid position could be important for diversification of ELMOD functions. elmod-b, c, d, and e mutants did not have aperture phenotypes as single mutants nor in combination with mcr, indicating that they probably do not function in aperture development. However, when ELMOD_E was expressed in developing pollen with the MCR regulatory elements the shape of apertures changed to round instead of elongated furrows. A similar dosage study to the one described previously in the manuscript revealed that high levels of MCR protein could counteract the effects of ELMOD_E on aperture shape. When the ELMOD_E protein was mutated to be more like MCR in the GAP domain (Cys129 changed to Gly and Asn129 changed to Asp), aperture number in an mcr background was increased and some of the apertures were furrowed rather than round. A limitation in this study is that the furrowed and round apertures were counted together, thus missing an opportunity to quantify the effects of the mutated ELMOD_E on aperture shape. While the opposite experiment of changing these residues in MCR to ELMOD_E-like residues was not as striking (aperture number was complemented but they did not become round), these data are exciting because they reveal the power of small amino acid changes in one protein to dramatically change aperture number and phenotypes during pollen development.

This manuscript will be of broad interest to scientists interested in cell polarity, patterning, and evolution of diverse morphologies. Diversification of clade A/B ELMOD genes could have played a role in generating the wide range of aperture numbers and shapes seen in Angiosperms. A mystery that remains and that can be addressed in future studies is how a protein that is localized throughout the cytoplasm and in the nucleus is able to regulate polarity during aperture formation.

Reviewer #3 (Public Review):

This largely genetics, morphology, molecular phylogenetic study is thorough, systematic and its presentation clear, allowing the extensive results to be quite easily followed.

The founding mutant is called macaron (mcr) - pollen grains have a ring-shaped aperture formed by two oppositely positioned slits merging to form a sandwiched ring in the equatorial position. The first part of the study established the relationship between MCR and two previously established aperture regulators, INP1 and D6PKL3.

1. Earlier observations showed a ploidy effect on aperture number, suggest a dosage effect. Introducing mcr into mutant plants (MiMe, Mitosis instead of Meiosis, and tes, which produce pollen grains defective in aperture numbers (respectively 50% normal, 50% more apertures in MiMe, and >10 irregularly placed in tes), the authors further established the reducing effect of mcr on aperture number.

2. Two previously studied aperture controlling proteins, INP1 and D6PKL3, were mislocalized in mcr pollen. Instead of the normal location along the three aperture domains in WT grains, these appeared in a ring-shaped location in the mcr mutant. This and the analysis of double mcr-1 inp1/2 (apertureless), mcr-1 d6pkl3 mutants (ring-shaped and partially covered by exine-pollen outer coat), established MCR as acting upstream of D6PKL3 and INP1.

The remainder of the manuscript reports the molecular identification of MCR (with four original alleles mcr1-4; and addition T-DNA alleles mcr5-7) and the two related ELMOD proteins involved in aperture formation [ELMOD_A and _E; A is closest to MCR, in the ELMOD_B clade). MCR has a ELMOD (engulfment and cell motility) domain; in animal cells ELMOD proteins are GTPase activating proteins, activating ADP-ribosylating factors (ARF GTPases). ELMOD proteins in plants have not been studied before. The rather extensive collection of mutations in a single gene, each conferring notable aperture defect, suggests MCR function is not only crucial, but sensibly monitored during the pollen maturation aperture formation window of time; these mutations, not the focus here, will be useful entries to future study of MCR. These are followed by analysis of potential functional interactions between proteins from the various ELMOD clades.

3. The genetic and aperture phenotype analyses of MCR and ELMOD-A (Fig. 4) showing that they are redundant, acting synergistically but MCR being more prominently needed (Fig. 4), localization studies showing that they are not located at the aperture membrane domain, but in the cytoplasm and nucleus (Fig. 5), and that the invariant R residue these noncanonical GAPs was indeed critical for their function (ability to complement the mutants) (Fig. 6) are well done.

These proteins are most likely involved in the secretory process or in polarity (alluded to in the Discussion), but how the location of these proteins relates to a function to select membrane regions as future aperture site (i.e. beyond being needed for INP1 and D6PKL3 but in finer biochemical or cellular details) is left unaddressed. However, these are likely too much to add to an already extensive genetic and morphology-centric study.

4. A dosage effect of MCR and ELMOD_A and aperture number, higher levels of complementing transgene product ] correlates with higher aperture numbers (>3) in rescued mutant plants, was nicely demonstrated (Fig. 7; Supp. Data).

5. Phylogeny analysis-guided amino acid sequence comparisons (Fig. 8 associated) led to insightful predictions, some of which, such as AA121 and 129, the authors were able to test experimentally in a later section. Genetic knockouts of other ELMODs_C,D,E,F (Fig. 9 associated) did not reveal impact on aperture formation. Cross-complementation by MCRp-driven expression of these genes in mcr mutant, and later in WT, authors were able to establish that mis-expression of ELMOD_E (from the MCR promoter) led to a new phenotype of multiple short round apertures in both mcr and WT background, showing a dominant negative effect.

On the other hand, additive expression of ELMOD_E from its own promoter in the WT background did not influence aperture phenotype. This (though not explicitly indicated, I interpreted this as probably together with the non-phenotypic elmod_e property, and low expression level during the tetrad stage) led the authors to conclude that ELMOD E, while it has the capacity to influence, is normally not involved in the aperture formation process [L466]. They went on to show high level of MCR promoter:MCR expression in these MCR promoter:ELMOD_E expressing mcr lines can counteract the impact from ELMOD_e, restoring the apertures (from round) to elongated/furrow.

I find the findings above interesting and the conclusion from the results insightful. However, the details related and also those in the remaining study in Fig. 9 regarding elmod_e,d,f mutants interfering with the more directly related characterization of MCR and ELMOD_E (Fig. 10).

6. The sequence comparison led the authors to identify aa121 and 129 located in the GTP activating region as potentially important, this together with the differential activity between MCR and ELMOD_E, led them to test the functional significance of these residues. [E_clade, 100% Asn121/Cys129; other clades 0%; mcr-2 has a D substituting the conserved G129 ]. (Fig. 10) ELMOD_E(N121D/C129G) acted more like MCR in restoring furrow like apertures; with only of these amino acid residue change, the mutant ELMOD_E induced a mix of multiple round apertures (similar to normal ELMOD_E), but also three furrowed pollen and pollen with a mix of furrowed and round apertures. This is a satisfactory functional follow up from a very nice phylogeny analysis.

7. The Discussion brought out some interesting testable hypotheses for the future, e.g. the role of ELMODs in Pedicularis pollen apertures (line 560).

In summary, this is a strong study, with two minor weaknesses (referred to in points #3 and #5). #5 is easily remedied. #3 is most likely not possible to address. The Discussion on the #3 aspect is largely and necessarily speculative, so not quite reach the intellectual satisfaction one would like to have. But it could be the basis for future investigation.

Reviewer #1 (Public Review): 

In this manuscript, Avraham et al. report their results of profiling different cell types in DRG from mice with different types of injury. In general, injuries occurred in PNS (sciatic nerves or dorsal roots) trigger more drastic effects on nearly all DRG cell types, comparing to those applied to CNS (spinal cord), albeit with some exceptions. Among these responding cell populations is a subset of macrophage expressing a satellite glial cell (SGC) marker, and an another population of SGCs, although their lineage and role remain unknown. Furthermore, fatty acid biosynthesis and PPARgamma signaling pathways are up-regulated after sciatic nerve injury, but down-regulated after dorsal root injury (again these observations are not verified and the underlying mechanisms are elusive). Application of a PPARgamma agonist is able to elevate axon regeneration after dorsal root injury. In general, this manuscript has a large amount of data from bioinformatics analysis but with limited functional verifications. Thus their biological meaning is less clear. <br> In Fig 3, the results imply different signaling involvement in DRG macrophages (cell cycle and DNA replication after SNC and steroid biosynthesis and glycolysis/gluconeogenesis pathways after DRC/SCI). Are these results due to differential resident/infiltrated macrophage in DRG after individual injury types? 

It is also intriguing to note that all injuries down-regulate genes related to antigen processing and presentation (Fig 3). This seems an interesting observation as macrophages often exhibit pro-inflammatory responses to injury. These results should be verified with independent methods. 

In Fig. 4, the authors describe a subset of macrophages expressing glial markers whose numbers become increased after injury. Is it possible that these might be the macrophages with engulfed SGCs? To test this, perhaps the authors could compare the abundance of these type-specific RNAs or use other independent methods (with transgenic mice with GFP-labeled SGCs to see if any GFP signals are in these macrophages). 

The results in Fig. 5 suggest that SGCs represent different cell populations. Again, their biological meaning remains unknown. An obvious possibility is these clusters might reflect their different activation states. It might be useful to apply single cell trajectory analysis to assess their relationship. 

Fig. 7, are the regeneration results after PPARa agonist comparable to those after sciatic nerve injury? Such information might provide insights as to its translational potential.

Reviewer #2 (Public Review): 

Avraham et al. applied single cell RNA seq to characterize the sensory neuron microenvironment in dorsal root ganglia after sciatic nerve crush (SNC), dorsal root crush (DRC) and dorsal column transection spinal cord injury (SCI) 3 days after injury. The data revealed differentially expressed genes and pathways in endothelial cells, Schwann cells, macrophages and satellite glial cells (SGCs), etc. among the different injury models, with SNC and DRC co-clustering, and SCI and uninjured control co-clustering for the most part. While a number of cell types are implicated in the differential responses of the microenvironment to injury, the authors focused on the satellite glial cells (SGCs) in functional validation of PPARa signaling in regeneration after DRC using a PPARa agonist (fenofibrate). 

Strengths: 

1) Many strengths: contrasting injury models, scRNA seq, extensive bioinformatics analyses 

2) Many interesting pathways were found to be differentially expressed after different injuries (e.g. Arg1 in macrophages, the Hippo pathways in Schwann cells, etc). 

3) If immune glia cells prove to be a new subtype (of macrophages or SGCs?), it will be a very interesting finding indeed. 

4) It is interesting that PPARa signaling is upregulated in SNC, unchanged in DRC and reduced after SCI. 

5) The study illustrates the value of using single cell RNA seq to dissect the neuron microenvironment in response to injury and neuron extrinsic influences on axon regeneration. 

Weakness: 

1) The authors have previously shown that PPARa agonist rescues the reduced axon regeneration in fatty acid synthase (Fasn) conditional knockout mice after SNC, so the role of PPARa in SGCs to support regeneration is no longer novel. 

2) This is not a weakness per se, but the two most interesting findings on immune glia cells and PPARa do not appear to be directly related. 

3) Pharmacological test with fenofibrate does not address the cell type specific role of PPARa, so it cannot be firmly established that PPARa in SGCs is most important for regeneration.

Reviewer #1 (Public Review):

This work investigated the mechanism of inhibition of SARS-CoV-2 polymerase by multiple nucleotide analogs using a high-throughput, single-molecule, magnetic tweezers platform. There was particular focus on the remdesivir (RDV) because it is the only FDA approved anti-coronavirus drug on the market at the time of this review. The study shows that remdesivir leads the polymerase to undergo a backtrack in which it moves back as much as 30 nucleotides from the last insertion. The results also show that RDV is not a chain terminator, which is consistent with prior work. In addition to RDV, the authors characterized other nucleotide analogs such as ddhCTP, 3'-dCTP, and Sofosbuvir-TP to propose that the location of the modification in the ribose or in the base dictates the catalytic pathway used for incorporation. The authors also propose that the use of magnetic tweezers is essential towards characterizing and discovering therapeutics that target viral polymerases.

Strengths:<br> A strength of the papers is the utilization of magnetic tweezers to characterize the polymerase at the single molecule level. This provides a unique method to capture less common or difficult to observe phenomena such as backtracking. Most bulk ensemble assays would have difficulty detecting these phenomena.

The characterization of multiple different types of nucleotides analogs to investigate the different mechanisms by which they could inhibit the polymerase is a strength of the paper. The authors elegantly utilize their system to show different pause states and backtracking of the polymerase.

In general, the paper is well written, and the data is clearly presented.

Weakness:

The experiments performed with the magnetic tweezers appear to not have contained the exonuclease domain. This domain would presumably be involved in removing nucleotide analogs that have been inserted and may alter the pause states or backtracking prevalence. For example, does the prevalence of backtracking increase when the exonuclease domain is not present. This is particularly important in regard to the RDV experiments.

A major claim for this study is the utilization of the magnetic tweezers "experimental paradigm" as being essential to the discovery and development of therapeutics to viral polymerases. In addition the authors state this approach is superior to bulk ensemble studies. This reviewer found these conclusions to be an overstatement and unnecessary. The use of magnetic tweezers is not amenable to all laboratories or an easy technique to implement within the therapeutic drug development. In general, the authors also overstate the power and feasibility of the magnetic tweezers in comparison to bulk ensemble studies. All assays have limitations, and the magnetic tweezers is no different in regards to being purified proteins, an in vitro approach, limitations in regards to feasibility for all users, ability to detect the amount of active protein, and multiple other reasons. This is a minor weakness of the paper that can be easily addressed because it detracts from the novelty of the studies.

Reviewer #2 (Public Review):

This study investigates the impact of remdesivir (RDV) and other nucleotide analogs (NAs), 3'-dATP, 3'-dUTP, 3'-dCTP, Sofosbuvir-TP, ddhCTP, and T-1106-TP, on RNA synthesis by the SARS-CoV-2 polymerase using magnetic tweezer. This technique allows to directly quantify termination of viral synthesis, pausing or stalling of the polymerase, thus, defining the effect of these NAs on viral synthesis. The work includes good quality data and nicely stablishes an assay to follow the activity of the SARS-CoV-2 RNA-dependent RNA polymerase. However, the basis of the assay and theory was largely presented before by the authors in Ref 22 and 23 (and other references therein). The main result here is that RDV incorporation does not prevent the complete viral RNA synthesis but causes an increase of pausing and back-tracking. This contrasts with a clear signature of synthesis termination induced by 3'-dATP. The work is complemented with the characterization of other NAs. Despite these results are of merit, I do not see this work to present a sufficient advance of our current knowledge. How these results translate into more physiological conditions at zero force should be addressed. The rationale of testing other NAs apart from the mere systematic characterization of other compounds is unclear. Similarly, I do not see the benefits of adding cell experiments with three compounds and experiments with the nsp14 mutant to address proofreading because they were inconclusive.

Reviewer #3 (Public Review):

This manuscript focuses on understanding the mechanism of action of remdesivir in the inhibition of SARS-Cov2 polymerase, using single molecule methods. The findings are highly original, significant and surprising. The approach is highly robust and supported by a range of orthogonal studies. Overall, these findings should help those engaged directly in drug discovery by providing a critical foundational understanding for the action of remdesivir.

The research described in this manuscript has several findings that significantly impact the broader field polymerase inhibition. First, the authors were able to show using single molecule methods that remdesivir-TP incorporation leads to polymerase backtrack. This is important because the pause is long enough that an ensemble assay could mistake this backtrack for a termination event. Secondly, the researchers found the effective incorporation of remdesivir-TP was determined by its absolute concentration. This suggests remdesivir-TP and similar nucleotide analogs incorporate via the SNA or VSNA pathway and would be more likely to add to the RNA chain when substrate concentration is low (independent of stoichiometry with the competing native nucleotide). Thirdly, the researchers found the effective incorporation rate of obligatory terminators was affected by the stoichiometry of their competing native nucleotide rather than their absolute concentration. This suggests that obligatory terminators are incorporated via the NAB pathway. The pausing that the researchers observed in the polymerase elongation kinetics have recently been demonstrated by two other groups. However, this study improved upon the assay conditions used by other researchers to recapitulate in vivo conditions and remove bias from kinetics measurements.

The authors highlighted the issues with remdesivir, tested other nucleotide analogs, and proposed a better alternative based on their assays (ddhCTP). Interestingly, the ddhCTP didn't actually work in infected cells. However, the authors presented a few theories on why it didn't work and said they plan to follow up to elucidate why it didn't work in cells. I think those results will be very interesting for the larger community working in this area. It's clear that the authors made a substantial enough contribution on the mechanism of inhibition of SARS Cov2 polymerase to merit publication in eLife, independent of the work on the "improved" antiviral candidate.

It would have been useful to clarify for the reader the pharmaceutical import of the putative delayed chain termination (or pausing) relative to actual chemical chain termination. In other words, I'm assuming that in both cases the viral genome is considered to be non-transcribed (in that a chemical agent has been incorporated into the growing strand). This is true for most compounds in this broad class of anti-virals. The issues are usually surrounding the width of the therapeutic index and the degree to which resistant mutants arise.

Overall, this manuscript constitutes a major advance in our understanding of chain termination in polymerases, and provides deep insights into the mechanism of action of remdesivir, which may contribute to further drug discovery efforts targeting this polymerase. Additionally, the authors have highlighted and addressed issues in the methodologies of previous mechanistic studies that led others to erroneous conclusions.

Reviewer #1 (Public Review):

In this work, the authors develop and apply a physical theory for interpreting fluorescence recovery after photobleaching (FRAP) data in the context of phase separation. Physical theories have been developed for interpreting and fitting FRAP data in other contexts, but such an effort has been missing in the context of phase separation.

The authors introduce a new dynamic boundary condition and show how this can be applied practically to extract diffusion coefficients for labeled molecules within condensates. Following this, the authors separate the molecules into bleached and unbleached species, apply mass balance, and rewrite diffusion equations, deriving the chemical potentials from a Flory-Huggins free energy functions, in terms of the temporal and spatial evolution of unbleached molecules. The authors demonstrate that if the internal diffusion coefficient is fixed, the free parameters become the diffusion coefficient outside the condensate, and the partition coefficient.

Their analysis shows that the external diffusion coefficient and the partition coefficient are related to one another. This suggests that their measurement of the internal diffusion coefficient combined with an independent measurement of external diffusion coefficients should enable the extraction of partition coefficients from FRAP data. Their approach avoids the use of empirical fitting functions, and the suggestion is that FRAP data, analyzed using a pre-determined internal diffusion coefficient, based on the dynamic boundary condition, can enable the extraction of transport properties outside condensates and thermodynamic partitioning coefficients.

This work is likely to be of broad interest, providing the numerical apparatus is made more transparent to the average reader and to users of FRAP, of which there are many. The three shortcomings are the lack of a direct connection to the interfacial tension, and the absence of analytical solutions, at least in asymptotic limits, and the description of the so-called cost function, which leads to the finding that the external diffusion coefficient scales essentially linearly with the partition coefficient in the limit of large P.

Overall, the authors seem to have achieved the goals they set for themselves, and given the immense interest in using FRAP measurements, there is a clear need for quantitative approaches to analyzing these data. On a semantic note, it is high time that we dispensed with the insistence that these condensates have to be liquids. Nothing about FRAP measurements or the underlying data stipulates that any of these systems have to be liquids. The prefix of liquid or liquid-liquid is, given all we now know, not really relevant or accurate.

Reviewer #2 (Public Review):

This paper establishes a novel theoretical model for diffusion in a liquid-liquid phase separation (LLPS) system to be used in fluorescence microscopy experiments for determination of diffusion and partition coefficients of the system via FRAP data. The crux of the model presented in this study is that it is the first to be derived using a theoretical framework tailored for a LLPS system. The starting points of the model are bulk diffusion theory combined with solution theory, adapted for diffusion across a droplet interface. As the authors note citing Taylor et. al. 2019, previous analyses relying on various phenomenological fitting schemes result in significant discrepancies between models due to underlying assumptions about the system. As LLPS and complex coacervation are the predominant frameworks that currently underlie the study of biomolecular condensates, a model that enables accurate and consistent determination of the physical properties of such condensates is of great interest to the field. The authors use both experimental and simulated datasets to successfully show that the diffusion constant inside the condensate can be precisely determined from a single FRAP experiment. The biggest strength here is that diffusion within the droplets can be independently determined without regarding any other parameter or situation outside the condensate. Additionally, when given specific data, their model finds a strong relationship between the diffusion constant in the dilute phase and the partition coefficient, such that if one is known, the other can be determined. Furthermore, they show, in principle, a global cost minimum exists and determine both of these parameters simultaneously. The model presented here will be a strong starting point for improved physical characterization of coacervates and protein condensates to be further adapted to complex systems. Despite the success in the primary purpose of the model, we note concerning issues that we recommend the authors address.

There is a recurrent lack of clarity in many sections of text and how the author's claims are supported by the evidence shown. Though we were able to fully understand the study, interpretation was needlessly difficult at times. Outstanding, but not exhaustive, examples are listed below:

1. Overstatement of capabilities of this model that are weakly or not supported by the study. At various points throughout the article, the authors speak of the applicability of their model to nuanced conditions that we believe is either only indirectly supported, or not supported at all by their evidence.

a. On line 62 the authors claim that this model uses non-equilibrium thermodynamics to capture the diffusion across the droplet interface. This implies that the model would be applicable to dynamic processes in which detailed balance is not preserved. While exchange of photobleached and unphotobleached fluorescently labelled components is a dynamic process, the authors explicitly assume that the volume fraction of condensate components within a droplet (Φtot) remains either at equilibrium or quasi-equilibrium when building their model.

b. On line 272 where the authors claim that their model is applicable to non-spherical droplets, referencing Fig. 3 as evidence. However, Fig. 3 and the accompanying text sections starting on lines 163 and 188 describe effects of different environments on an explicitly spherical droplet. In particular, the distance to a coverslip (h) and between neighbouring droplets (d) never drops below the spherical droplet radius (r). We believe this data would constitute evidence of the model's applicability to a non-spherical droplet. Another concern is that the dynamic boundary condition could be dependent on the ratio between the bleaching spot radius and the condensate radius. Thus the authors should discuss the applicability of their theory when Rbleaching spot << Rcondensate and when Rbleaching spot >> Rcondensate.

c. Having claimed that not only Din, but also Dout and P can be determined, in principle, from analyzing a single FRAP experiment, it is unclear why they do not show this capability using the experimental data they have. It is especially obscure because the cost function and how it is calculated is not described at all.

d. In the Discussion section, the authors claimed that their model can be generally applied to study the diffusive properties of biomolecular condensates. However, recent literature (e.g. Biophys. J. 2019, 117, 1285-1300, Nature 2017, 547, 241-245, etc.) reported that the diffusion at the biomolecular condensate interface cannot be treated with local equilibrium due to interfacial resistance. This work did not take these interfacial effects into consideration, and the authors should explain if they expect these phenomenological effects to hamper the application of their theory.

Reviewer #1 (Public Review):

Olivieri, Julia Eve et al., applied their novel statistical approach, the SpliZ (detailed in a separate manuscript but it's very difficult to judge the approach since we do not really have access to it) to high-throughput single-cell RNA-seq datasets collected using the 10x platform to discover novel insights into cell-level heterogeneity of alternative splicing. Previous works in the field of single-cell alternative splicing have relied on single-cell technologies that profile a much lower number of cells. The authors validate their findings using the experimental approaches of single-cell PCR and RNA FISH, and validate that their findings can also be found using Smart-seq2 data, on which the gold standard approaches for single-cell alternative splicing analysis have been developed. They demonstrate conservation analysis of single-cell alternative splicing events across species, examples of genes that are spliced in cellular compartment specific and cell-type specific patterns, cell-type specific alternative splicing changes correlated with psuedotime in spermatogenesis, and, importantly, the discovery of new cell-type subpopulations that are defined by splicing changes but are indistinguishable based on gene expression. They show that the SpliZ score correlates well across replicates on a tissue level and cell-type level, which indicate the robustness of the method.

The conclusions of the paper are reasonably well supported by the data, and the authors have sufficiently proven that their approach allows for the discovery of novel biological phenomena. The authors provide examples in which the key questions that can be addressed with a single-cell splicing technology are investigated. An important question in the field of cellular heterogeneity is whether or not novel cell populations can be detected by clustering based on splicing events that can be not detected based on gene expression. The authors convincingly demonstrate that subpopulations of the blood classical monocyte cell type can be distinguished by a single splicing event captured by their approach that do not separate by gene expression.

Overall this paper reports some novel biological discoveries. The weakness and limitations of the method should be elaborated to guide future usage. When introducing a new technology, it is important for researchers utilizing these findings to be aware of the known limits. Furthermore, evaluation of alternative splicing conservation on a transcriptome-wide scale and reproducibility of splicing change detected on a single cell level are not demonstrated, and could further strengthen the arguments claimed by the authors.

Reviewer #2 (Public Review):

This manuscript from Salzman and colleagues described interesting attempts to study cell type-specific alternative splicing using 10x scRNA-seq data. Given the strong 3' bias, analysis of splicing using such dataset is in general challenging. This work provided evidence that alternative exons with differential splicing between tissue compartments can be identified, and cell types can be revealed by splicing profiles of single cells, to some extent. This work is informative regarding what can be done for alternative splicing using 10X data and filled in a gap in the field in this regard.

Reviewer #1 (Public Review):

The series of experiments to identify and examine changes in chromatin marks in response to flgg22 treatments investigate clearly defined hypotheses. They first present data showing that genes in pathways involved in specialized metabolism are more likely to be associated with both repression (H3K27me3) and activation (H3K18ac) marks than expected by chance in the genome. Using antibodies against H3K27me3 and H3K18ac in pull-down experiments, they show that H3K18ac and H3K27me3 are co-localized at the camalexin biosynthesis genes. They then show that, in response to FLG22, H3K18ac modifications increased and H3K27me3 modifications depleted. In mutant lines that have defective deposition of H3K27me3 expression of camalexin biosynthesis genes in response to FLG22 increased compared to wild type. Together, this progression of experiments provides convincing data of an association between chromatin state and changes in transcription levels. Overall, the model is compelling, though the authors should take care to ensure that the language used is reflective of the association or interplay between chromatin state and transcription factor availability.

What is less clear is the link between chromatin states/transcriptional expression and the abundance of the metabolic pathway products that are required to limit pathogen spread. In this manuscript, Zhao et al, test camalexin content using liquid chromatography-tandem mass spectrometry (LC-MS/MS) following FLG22, finding an initial accumulation, followed by further increases over 6 hours. Plants with reduced H3K27me3 marks started to accumulate camalexin earlier while those with reduced H3K18ac marks accumulated camalexin later. While the altered timings in the mutant lines do support a connection between gene expression dynamics and metabolite accumulation, it does not prove that changes in transcription are wholly responsible for the differences in metabolites. Previously reported Ribo-seq experiments mRNAs from genes, including those for camalexin biosynthesis, suggesting that PTI/RTI-triggered translational regulation has a significant role in changes in expression (Xu et al 2017 Nature 545, 487-490; Yoo et al Molecular Plant 13:1 88-98). This does not need to detract from the main findings of this paper, but the changes in metabolite accumulation should be interpreted with these data in mind and discussed appropriately.

Reviewer #2 (Public Review):

The authors propose that a bivalent chromatin switch exists on genes for the major Arabidopsis phytoalexin and this helps to influence the kinetics of this compounds regulation. This suggests that developmental regulatory designs are also used for defense chemistry. This is an interesting idea and the use of serial ChIP to show that this is not developmentally delineated but in fact occurring on a single promoter at the same time is very interesting. The efforts to extend this to being a general specialized metabolism pathways are less clear given some issues of over-counting pathways when a metabolic pathway is truly cyclical and applied to a hierarchical database design. The phenomon appears limited to fewer pathways then suggested and some statistical analysis are needed to support the claim on this one pathway.

Reviewer #3 (Public Review):

This study proposes the identification of "bivalent chromatin" at specialized metabolic gene clusters. Perturbation of either H3K27me3 or H3K18ac levels using mutants were used to show that there were effects on the expression of these metabolic genes. However, it is not novel that H3K27me3 and H3K18ac colocalize in the Arabidopsis genome. This was shown by Luo C et al in 2012 and is referenced by the authors. In fact, Luo C et al also showed these two modifications are colocalized by Chip-re-ChIP. This current study is presented as if the H3K18ac and H3K27me3 modifications are specific to metabolic genes, but they're not. Instead, it appears H3K18ac is localized to many H3K27me3 genes of which certain specialized metabolic genes are an enriched subset. There are >4000 genes targeted by PRC2/H3K27me3 in Arabidopsis that are enriched for genes that respond to the environment and or developmental cues. It is therefore, not unexpected that specialized metabolites are a subset of this class given they're only expressed in very specific environments. In summary, the results presented in Figure 1, 2 and 4 are essentially already published by Luo C et al, making the results of this study incremental.

Reviewer #1 (Public Review):

The paper by Neofotis and colleagues contains a substantial amount of data that together supports a highly novel and important finding: the pyrenoid of the green alga Chlamydomonas can be induced by hyperoxia, even in the presence of high levels of CO2. More specifically, hyperoxia can be substituted by hydrogen peroxide, indicating that ROS signalling mediates this response. The finding is significant because hyperoxia is a common but understudied event, especially when algae are grown for biotechnological purposes.

The paper leads by exploring the phenotypes of two wild-type strains of Chlamydomonas that show markedly different abilities to grow and tolerate at hyperoxic conditions. Extensive electron microscopy indicates that the resistant strain produces pyrenoids with closed and robust starch sheaths when challenged with oxygen. In contrast the susceptible strain has fragmented starch sheaths surrounding the pyrenoid. Crossing the two strains demonstrated the heritability of this phenotype. Importantly this hyperoxia dependent pyrenoid induction occurs even at the very high levels of 5% CO2, a condition where no pyrenoid would normally be seen.

The TEM images are supplemented using light microscopy, where the presence or absence of the pyrenoid starch sheath can be clearly discerned. Following various control conditions (e.g. no pyrenoid at high CO2 alone), it is discovered that H2O2 in the presence of high CO2 alone can bring about pyrenoid formation. Importantly the effect of H2O2 is validated using another Chlamydomonas strain that expresses fluorescently tagged Rubisco. In addition data is presented that indicates that H2O2 in the hyperoxia resistant strain is localized in small clusters that may represent peroxisomes (and would implicate photorespiratory H2O2 production). The resistant strain was also able to operate at higher O2 levels, having a higher O2 compensation point.

The main weakness of the manuscript is its current poor presentation, which needs to be improved to enhance accessibility of the work. The word limits common in selective journals could have been helpful to better communicate the work's impact and importance. This comment relates mostly to the results section. The discussion, albeit also long, is better written and deeply engages with the literature at a high level.

An aspect that would greatly strengthen the paper overall is to clarify whether the pyrenoids induced at high CO2 by hyperoxia or hydrogen peroxide participate in the CCM. Do these cells now have high affinities for Ci?

Reviewer #2 (Public Review):

A summary of what the authors were trying to achieve:

Under limiting CO2 algae operate CO2-concentrating mechanisms (CCMs) to increase CO2 at the active site of Rubisco to enhance Rubisco's efficiency. To achieve this Rubisco is packaged into a non-membrane bound compartment called the pyrenoid where CO2 is delivered via a series of Ci transporters and carbonic anhydrases. The size and Rubisco content of the pyrenoid has previously been shown to increase when CO2 becomes limiting and the cells are photosynthetically active (i.e. in the light). However, the signal for pyrenoid (and CCM) induction is unclear, with it proposed it could be due to direct sensing of inorganic carbon (i.e. CO2 or HCO3- concentration) or a metabolic signal through reduced photosynthetic capacity (i.e. photorespiratory by-product). In this study, Neofotis et al. explore whether inorganic carbon concentration or a photosynthetic by-product induced by hyperoxia leads to pyrenoid induction. They show that hyperoxia most likely due to increased hydrogen peroxide levels is a prime candidate for pyrenoid induction. With pyrenoid induction even in the presence of high CO2 when cells are hyperoxic or exposed to H2O2.

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

Strengths:

The authors use two independent strains that show differing morphologocial and growth responses to hyperoxia which correlate to different pyrenoid starch sheath induction levels and morphologies; linking for the first time response to hyperoxia with pyrenoid formation. They further analyze progeny of crosses of these two strains indicating that this is potentially linked to a singular genomic locus, although this would require further investigation. They show that both exogenous and endogenous produced H2O2 induces pyrenoid formation independent of inorganic carbon level and that the addition of H2O2 scavengers prevents pyrenoid induction. As far as we are aware this is the first time in Chlamydomonas that a specific metabolic signal, H2O2, induces pyrenoid formation. The results will strongly support efforts to identify the unknown molecular mechanisms that lead to pyrenoid formation. The authors also analyse a large range of conditions relating to carbon source and light, and record the state of pyrenoid induction in these conditions, providing a useful resource for the pyrenoid community.

Weaknesses:

A limited number of crossing progeny are analysed, and in some instances the n values for observational experiments are low. Without quantitative analyses of these data, the phenotypes remain observational and purely correlative with other observations the authors make. The authors mention in several occasions that the genetic variation between the two strains used should modulate differing responses, however they do not give any details on the genetic basis for the differing response - particularly the change in H2O2 production between the two strains used for comparison and the differences in starch synthesis - even though genetic information on both strains are available. The claims that pyrenoid formation is induced by endogenously-produced H2O2 by methyl viologen or metronidazole and is repressed by H2O2 scavengers ascorbic acid and dimethylthiourea requires further experimental proof.

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

The authors provide strong evidence that H2O2 peroxide provides a signal for pyrenoid induction. The data further supports the importance of the pyrenoid starch sheath in efficient pyrenoid function and the CCM. In addition, the data indicates that pyrenoid formation (and CCM induction) is more complex than just H2O2 signalling but is likely a combination of factors including light, carbon source and availability, plus metabolic signalling, likely via H2O2 production.

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

The work provides a foundation for investigating the regulation and control of pyrenoid assembly at a molecular level in Chlamydomonas. For the first time a metabolite (H2O2) was shown to induce pyrenoid formation. This opens a starting point to investigate potential signaling pathways and to identify other components involved in pyrenoid assembly and CCM induction. The findings of this study are likely applicable to pyrenoids across algal lineages, and this study will provide a starting framework for follow-up studies in Chlamydomonas and other pyrenoid-containing species.

Any additional context you think would help readers interpret or understand the significance of the work:

It would be helpful for readers to understand the significance of the work if the authors could put the work into context of ongoing efforts to engineer the pyrenoid into crop plants to increase yield and to highlight the global importance of pyrenoid-mediated algal photosynthesis.

For a better understanding of their proposed model of H2O2 mediated pyrenoid/CCM induction it would be very helpful if the author added a figure of their CCM induction model to the discussion.

Reviewer #3 (Public Review):

The paper presents novel insight into a potential signal, H2O2, and environmental conditions that induce the pyrenoid, which is an essential part of the physical CO2/inorganic carbon (Ci) concentrating mechanism (CCM) of many microalgae One central claim of this study is that pyrenoid formation and morphology with regard to its starch sheath is induced by hyperoxia in two Chlamydomonas wild type strains which otherwise differ in their ability to grow under hyperoxic conditions. Yet, the way some conclusions are presented may need clarification to fully appreciate the impact of the findings presented. The authors have carefully devised growth conditions that allow constant monitoring and control of the O2 and inorganic carbon (Ci; dissolved CO2 and bicarbonate) concentrations in liquid cell cultures. Ci was supplied by addition of NaHCO3, which will provide predominantly bicarbonate as CI source at pH above 8 and by sparging liquid cultures with gas mixes of 5% CO2 with either 21% or 95% O2 for normal, air level or high oxygen conditions respectively. Without hyperoxic stress, cell lines were shown to behave as expected in terms of starch sheath accumulation and formation of pyrenoids in response to light and low Ci. Therefore evidence presented in a series of high quality TEM images for the presence of pyrenoids and their starch sheath morphology specific to hyperoxia is unambiguous. This finding is of great interest with regards to unsolved questions about pyrenoid function and mechanisms underlying responses to Ci and high O2 stress in more general terms.

Furthermore, the authors provide strong evidence in support of their hypothesis that induction and/or formation of the pyrenoid is related to the ability of cells to grow efficiently under hyperoxia conditions (p.3). The study establishes that two wild type isolates show differences in starch sheath morphology. Cell lines with relatively continuous (sealed) starch sheath around pyrenoids (CC1009) accumulated significantly more biomass than cell lines with lesser developed (fragmented) starch sheath (CC2343). The same correlation was found in the meiotic progeny of crosses between CC1009 and CC2343, showing a 2:2 segregation with a more severe growth phenotype in some of the progeny. However, the analysis was not taken any further to investigate the genetic basis and concluded that genetic factors aside from pyrenoid morphology may play a role in tolerance to hyperoxia.

It is not quite clear why the authors included a growth analyses under mixotrophic conditions and solid media and measured of photosystem II efficiency only under these conditions. The results showed faster growth of cells (CC2343) that tend to accumulate fractured pyrenoid starch sheath, however, growth is based on undefined proportions of autotrophy and heterotrophy under these conditions, and changes in metabolism are not well understood or predictable, which - from my point of view - is too confounding for gaining conclusive evidence for hyperoxia tolerance from biomass accumulation. Likewise, the measurements of PSII function which is a product of many factors concertedly impacting photosynthetic electron transport and correlates with growth only conditionally under autotrophy is even less informative under mixotrophy. Differences in respiration rates may have to be considered as well as differential partitioning of metabolites in the two different strains, which is outside the scope of this paper.

Physiological analyses (Rubisco function, photosynthetic O2 evolution and ROS and H2O2 levels), which might have also provided clues as to which processes diverge among progenies, were focussed on the parental strains. A potential correlation between pyrenoid morphology and photosynthetic performance was based to rubisco function (Rubisco content and activation state, rubisco activase levels) and O2 compensation points. It is to note that the O2 compensation point measurements, in contrast to the author's interpretation, are not a very specific measurement of CO2/O2 discrimination by rubisco.

The authors develop the hypothesis that a photosynthesis-derived signal common to low Ci and hyperoxia induces pyrenoid and starch sheath formation, since both conditions are associated with low CO2:O2 ratios in the chloroplast, potentially enhancing H2O2 by photorespiration and O2 acting as alternative electron acceptor in the light reaction. The study firmly establishes that the formation of starch sheath-coated pyrenoids was mediated through elevated H2O2 when applied exogenously or induced endogenously by addition of H2O2 generating chemicals (methylviologen etc.), which is consistent with this hypothesis. However, evidence for elevated H2O2 levels under hypoxia are based on indirect measurement of total ROS, using a general ROS indicator stain, rather than a more specific indicator for H2O2. It was not tested whether other ROS species could induce pyrenoids/starch.<br> For the reader it is not straight-forward to reconcile, and an explanation is not offered, an increase in H2O2 as the pyrenoid inducing signal in wild type CC2343 which showed no differences in ROS levels before and during hypoxia treatment (Figure12), but forms pyrenoids and starch sheaths similar to CC1009 as opposed to hyperoxia when CC2343 had more fractured starch sheaths.

As to the discussion, the implications for H2O2 signalling and the yet to be investigated interaction with other signals are discussed highlighting the differences and commonalities of CCM induction by low CI compared to high Ci and hyperoxia. The eco-physiological implications are presented less comprehensively and focus on functional implications and less on the correlation of species distribution, with presence of pyrenoids and adaptation to a habitat.

Reviewer #1 (Public Review):

This study investigates the brain mechanisms serving attentional modulations of pain. To this end, 39 healthy human participants performed an attention-demanding visual task with simultaneous application of painful thermal stimuli to the arm. In a 2x2 factorial design, visual task difficulty and thermal stimulus intensity were varied. To investigate the contributions of opioidergic and noradrenergic signalling to the attentional modulations, the paradigm was performed three times, i.e. after application of an opioid antagonist, a noradrenergic agonist and a placebo drug. During the paradigm, simultaneous fMRI of the brain and the spinal cord was performed. The results show that the opioid antagonist reduces attentional modulations of pain and changes connectivity between the anterior cingulate cortex, the brainstem, and the spinal cord. Such changes were not observed after application of the noradrenergic agonist. Together, the findings convincingly demonstrate that attentional modulations of pain are associated with opioid-sensitive changes of cortical-brainstem-spinal connectivity.

Strengths:

• The brain mechanisms serving attentional modulations of pain is a timely and relevant topic with potential clinical implications.

• Simultaneous fMRI of the brain and spinal cord is a cutting-edge neuroimaging approach which uniquely allows for cerebral-spinal connectivity analysis.

• Comparing attentional modulations after opioidergic and noradrenergic modulations is innovative and offers a translational perspective.

Weaknesses:

• The comparison of the opioidergic, noradrenergic and placebo manipulations are particularly interesting. These results are mostly based on the presence of effects in the opioidergic as compared to the placebo condition but not in the noradrenergic as compared to the placebo condition. Thus, the negative findings in the noradrenergic condition play an important role. This part might be strengthened by using Bayesian statistics which allow for distinguishing between evidence of absence and absence of evidence. Moreover, directly contrasting the opioidergic and noradrenergic conditions might further strengthen the case.

• A crucial part of the reasoning is the correspondence between behavioral and neuroimaging effects. So far, the correspondence is mostly based on similar patterns of modulations on the group level. Extending this correspondence to the individual level might provide further support. For instance, the authors might relate the individual behavioral effects to the individual neural effects. This would substantially strengthen the relationship between behavioral and neural effects.

• The crucial attentional effects on pain ratings and fMRI responses are interactions between task difficulty and stimulus intensity. These findings are in line with previous findings from the same group. However, a main effect of task difficulty might also be reasonable. This should be discussed with reference to previous studies on attentional effects on pain.

Reviewer #2 (Public Review):

In this impressive study, Oliva and colleagues used a placebo-controlled, double-blind, within-subject, three-armed pharmacological challenge with simultaneous imaging of the entire central nervous system (CNS) of healthy volunteers during attentional modulation of pain. They observed that during a distraction-induced reduction of perceived pain, the first station of nociceptive processing in the CNS - i.e. the spinal cord - exhibited functional magnetic resonance imaging (fMRI) signal changes that went along with the behavioural data. fMRI signal changes in the spinal cord could furthermore be related to connectivity changes with brainstem areas known from animal experiments to influence spinal cord nociceptive processing. Finally, the authors demonstrated how these behavioural and neurobiological effects were disrupted by a blockade of the opioidergic system, but not of the noradrenergic system, conferring a certain pharmacological specificity.

From a technical point of view this is a formidable feat: due to the complexities of data acquisition, there are only a handful of whole-CNS imaging studies and none of them had yet paired this with a pharmacological challenge and connectivity down to the spinal cord. By targeting both the noradrenergic and opioidergic system and obtaining results from the spinal cord, brainstem and brain simultaneously during an analgesic manipulation, the authors provide a so-far missing link between an extensive animal literature that has examined the relevant spinal (and brainstem) processes and human studies that have focussed to a large extent on supra-spinal cortical and sub-cortical processes. Here, especially the connectivity results that provide a critical link from (cortex via) brainstem to the spinal cord are of note.

Nevertheless, there are several - mostly methodological - points that the authors will need to address in order to substantiate the reported results (some of which might simply be addressed by adding currently missing information). These include i) missing statistical tests, ii) a possible bias in the connectivity results, iii) their approach towards identifying the small target areas of interest in the brainstem and spinal cord, iv) the success of double-blinding and v) the quality of the fMRI data.

Reviewer #3 (Public Review):

The strengths of the paper include the comprehensiveness of the experimental design, which involves the spinal-brainstem-whole brain simultaneous fMRI with pharmacological treatment (naltrexone and reboxetine).

The weaknesses of the paper include the insufficient description of experimental methods, statistical analyses, and results. My biggest concerns are mostly about statistical analyses and how they described the statistical analysis results. Most importantly, in their descriptions of results, it was difficult to get details about statistical analyses and tests (e.g., thresholding, masks, etc.) and also quantitative information about the results (e.g., effect size, and test statistics, p-values, etc.).

Reviewer #1 (Public Review):

Stem cells in the Drosophila ovary provide a great model to understand cell behavior and regulation due to its genetic tractability, organized morphological pattern, and ease to perform live imaging. Compared to the adult ovary which have been studied quite extensively, the pupae ovary is a much less explored stage. Here the authors extensively studied the cell lineage in the pupae ovary, which helps understanding the development of early cell fates and the formation of the first set of egg chambers. They first described the different stages of pupal ovary development, followed by several different lineage tracing experiments that conclude a subset of Intermingled Cells (ICs) as Escort Cell (EC)/ Follicle Stem Cell (FSC) common precursors. Then they described Extra-Germarial Crown Cells (EGCs) and basal stalk cells and showed by live imaging that they contribute to the first budding cyst.

Strength:

Several lineage tracing experiments and statistical calculations are performed to conclude that a subset of ICs function as EC/FSC common precursors. The methods are written in great detail to help understand the calculation.

The finding that EGCs and basal stalk cells contribute to the first budding cyst is new and intriguing. This initial developmental process, different from what happens in the adult ovary, might provide insight into how germline and somatic cells are coordinated.

Weakness:

The authors showed in their 2017 NCB paper that FSCs contribute to ECs in adult ovary. Here they showed that there is a common precursor of EC/FSC. Are these two cell types the same? It has been shown in single cell analysis during third instar that the ICs and FSCPs present Con and bond as their specific markers (Slaidina et al. 2020). Both give rise to EC/FSC/FC in lineage tracing experiments. Therefore, the novelty of this finding is weakened.

While the finding that EGCs and basal stalks contribute to the first budding egg chamber is intriguing, the definition of EGCs and basal stalks are quite vague. Do EGCs have a distinct feature that is worth noting as separate cell types, or are they simply early FCs locating posterior of the first germline cyst? Do basal stalks express mature stalk markers, or are they simply accumulating FCs that are not fully differentiated yet? Is the process of EGCs and basal stalk contributing to the first budding egg chamber similar to posterior FCs contributing to the budding egg chamber? Is it because the budding of the first germline cyst takes longer than normal, there are more FCs accumulating at the posterior, thus making this region look like EGCs and basal stalks?

The presentation is too lengthy. A more concisely written paper would help the audience to get the key points that the authors hope to convey.