Reviewer #1 (Public Review):

In this work, the authors set out to use contact tracing and whole-genome sequencing to track the elimination of dog-mediated rabies in Pemba island, Tanzania. A major strength is the use of multiple data types in the analysis. A major limitation is the rudimentary health economics approach to make claims about the cost-effectiveness of different approaches. The work will likely have an impact on influencing the practical policies that can be implemented to target the elimination of dog-mediated rabies in other regions/contexts.

Reviewer #2 (Public Review):

In this paper, the authors illustrate how a One Health approach can strengthen our understanding of the dynamics of the spread and the control of rabies. This is done by analyzing multiple epidemiological and sequence data from both dogs and humans, on the island of Pemba. The joint analyses of these data make it possible to reconstruct the history of rabies introduction and circulation on the island and to quantify the impact of different control measures in particular the cost per death averted.

Data documenting rabies epidemics tend to be rare and of limited quality so the effort to collect these data and analyze them with state-of-the-art statistical techniques should be saluted.

Reviewer #3 (Public Review):

In the proposed paper, the authors use a combination of case data and genetic data to characterise the impact of a dog vaccine campaign on rabies transmission on Pemba island. This represents an impressive set of data to answer key questions linked to rabies control. It is rare to see a combination of detailed genetic and epidemiology data from the same disease system. Overall, I thought it was an impressive paper. My only major concerns were with the phylogenetic analyses.

The phylogenetic analyses were difficult to understand. The authors use a phylogenetic framework to estimate the underlying number of rabid dogs per outbreak (171 in the first outbreak and 140 in the second one), but it was unclear to me where the information was coming from. From the supplementary material, it seems the authors build transmission trees consistent with the phylogenies. However, these are reliant on (a) a serial interval and (b) a dispersal kernel. There is no reference as to what serial interval distribution was used and how it was calculated. Similarly, there is no information on the dispersal kernel, including what data was used to fit it. I suspect that the serial interval for rabies (and probably the dispersal kernel) has a long tail, which would lead to substantial uncertainty in the transmission chains, however, I could not see uncertainty in the outbreak sizes.

Reviewer #1 (Public Review):

The authors developed a new concept: Skeletal age, which is chronological age + years lost due to suffering a low-energy fracture.<br /> There seem to be conceptual problems with this concept: It is not known if the years lost are lost due to the fracture or co-morbidities. In addition, with the possible exception of zoledronate after hip fracture, we have no evidence that this increased risk of mortality can be changed with interventions. Furthermore, it is not clear why the authors think that patients and doctors will better understand the implications of older "skeletal age", on future fracture risk and the need for prevention, for example, the 10-year risk of MOF? Knowing that my bones are older than me, could make a patient feel even more fragile and afraid of being physically active. The treatment will reduce the risk of future fractures, but this study provides no information about the effect on mortality of preventing the subsequent fracture or the risk of mortality associated with recurrent fractures.

Introduction:<br /> The statement that treatment reduces the risk of dying, needs modification as the majority of clinical trials have not demonstrated reduced mortality with treatment.<br /> It is not clear how the skeletal age captures the risk of a future fracture. The other difference between the idea of "skeletal age" and for example "heart age" is that there are treatments available for heart disease that reduce the risk of mortality, as mentioned above this has not been shown consistently in clinical trials in osteoporosis.

Discussion:<br /> The prevalent comorbidities; cardiovascular diseases, cancer, and diabetes, suggest that fracture patients die from their comorbidities and not their fractures.<br /> The discussion should be more balanced as there is a number of clinical trials demonstrating reductions in vertebral and non-vertebral fractures without effect on mortality. There may be specific effects of zoledronate on mortality, but that has not been shown for the vast majority of treatments.<br /> It is not correct that FRAX does not take mortality into account? It does not tell you specifically how high the risk of dying and how high the risk of a fracture is but integrates the two. "Skeletal age" does not provide either information, it just tells you that your skeleton is older than your chronological age - most patients and doctors will not associate that with an increased risk of dying - only of frailty.<br /> The statement that zoledronate reduces the "skeletal age" by 3 years, has not been demonstrated and it is not clear how this can be demonstrated by the analysis reported here. As the reduced mortality has only been shown for the Horizon RFT, this cannot be inferred for other treatments and other fracture types.<br /> The information provided by the "skeletal age" is only that the fracture you already had took x years of your remaining lifetime. With the exception of perhaps zoledronate after hip fracture, we have no indication from clinical trials that the treatment of osteoporosis will change this.

Reviewer #2 (Public Review):

The paper of Tran et al. introduces the concept of 'skeletal age' as a means of conveying the combined risk of fracture and fracture-associated mortality for an individual. Skeletal age is defined as the sum of chronological age and the number of years of life lost associated with a fracture. Using the very comprehensive Danish national registry and employing Cox's proportional hazards model they estimated the hazard of mortality associated with a fracture. Skeletal age was estimated for each age and fracture site stratified by gender. The authors propose to replace the fracture probability with skeletal age for individualized fracture risk assessment.

Strengths of the study lie in the novelty of the concept of 'skeletal age' as an informative metric to internalize the combined risks of fracture and mortality, the very large and well-described Danish National Hospital Discharge Registry, the sophisticated statistical analysis and the clear messages presented in the manuscript. The limitations of the study are acknowledged by the authors.

Reviewer #1 (Public Review):

The authors demonstrate that modest oscillatory changes in the E-I ratio occur throughout the day and are linked to changes in both synaptic excitation and inhibition. These conclusions were based on adequately sampled electrophysiological data of stimulation-driven and non-evoked excitatory and inhibitory currents. For these studies, a fixed stimulation was not used across slice recordings but was limited to intensities where the E-I ratio was stable. Two points may need further clarification in the text. Firstly, authors might comment on whether current magnitudes plateau at these stimulation intensities. Secondly, make clear why the cause of E-I balance changes was not elucidated from convergent, evoked measurements in the same cell, but instead relied on non-evoked measures of spontaneous miniature excitatory postsynaptic currents and miniature inhibitory postsynaptic currents (mEPSCs and mIPSCs) that were recorded separately in different cell populations. mEPSCs and mIPSCs data analysis relied on statistical scrutiny within genotype and could gain additional rigor and benefits to study reproducibility by applying tests ( e.g. two- way repeated measures (RM) ANOVA) that consider the influence of both genotype and time of day. With this approach, the authors could determine in figures 3 and 4 whether control (B6) mice exhibit the predicted increase in mEPSCs and reduction in mIPSCs at ZT0 when compared to its ASD mouse model. In a noteworthy experiment, the authors connect abnormalities in inhibitory oscillations to altered endocannabinoid signaling using measurements of spontaneous (s) IPSCs, where changes in sIPSC charge were noted. The measurements used to make the paper's conclusion lacked consistency and the authors can bridge these differences by testing whether WIN agonist treatment can restore normal daily E/I oscillation in FMR1 KO and BTBR mice using the stimulation-evoked measurements from figure 1. The study used male and female BTBR mice and only male Fmr1 KO mice. Sex- effects in the study were not disclosed, so it is unclear whether daily E/I oscillation changes were similar in male and female BTBR mice or occur at all in female Fmr1 KO mice. Lastly, numerous studies have noted significant changes in the magnitude of the E-I ratio in an autism mouse model and causally linked these changes to alterations in disorder-related behavior or homeostatic regulation of circuit activity. However, in this current study, neither the loss nor reversal of daily E/I ratio oscillation were causally linked to alterations in sleep timing and architecture or any change in behavioral phenotype. On a promising note, the authors did find a slight decrease in NREM delta power in the Fmr1 KO and a larger decrease in the BTBR mice. Future mechanistic studies on this topic may aim to buttress support for E/I oscillations rather than alterations to the overall E/I level in causing autism-related phenotypes by providing supporting examples of biological significance.

Reviewer #2 (Public Review):

In Bridi et al the authors convincingly show alteration of the E/I ratio oscillation in two mouse models (Fmr1 and BTBR) of ASD. They go on to examine two possible mechanisms that may underlie these changes, 1) sleep/wake cycle and timing and 2) eCB signaling, both of which have been shown to change E/I ratio oscillations. They find that eCB signaling is altered in both models while sleep/wake timing and cycle are normal, concluding that dysfunctional eCB signaling is likely contributing to the changes in E/I oscillation. The experiments are extremely well done, and conclusions are mostly supported by the data, however, there are some concerns with the interpretation of their findings which I will detail below.

1) The authors describe the changes in E/I ratio that they observe in the BTBR mouse line as a "phase-shift". However, to actually show a true phase shift they should record at all of the same time points as they did in the Fmr1 model. Based on just two time points the authors have not shown a "Phase-shift" a phase shift would have to show that the other two time points (Z6 and Z18) follow the predicted (-6hr?) shift. These data would also help define the length of the shift.<br /> 2) Are the changes in E/I ratio presynaptic or postsynaptic? The authors seem to suggest that the synaptic changes they observe are a loss or gain in synapses. Mini-analysis alone is not sufficient for this conclusion. Even if the authors have shown in a previous paper that PPR is unchanged in control mice, presynaptic effects could be contributing to the observed changes in the mouse models studied here. As eCB signaling is thought to be primarily presynaptic this lends additional motivation to explore presynaptic contributions to the observed phenotypes.<br /> 3) The authors do not make any comparisons between control and ASD model mice at any of their time points. It would be helpful to have additional comparisons between ASD model and control at each time point tested in Fig 1 to relate back to previously published studies that mostly record in the animals' light phase. In other words, please clarify at which phases the ASD E/I ratio is different from the control.

Reviewer #3 (Public Review):

The authors previously reported a daily oscillation of the excitation/inhibition ratio occurs normally in layer 2/3 cortical neurons in wild-type mice. In this manuscript, they examined the E/I ratio in the primary visual cortex in two different autism mouse models and showed that the daily oscillation was disrupted in both, albeit in different ways. They further demonstrated that complementary changes in excitatory and inhibitory synaptic transmissions were underlying the disrupted E/I ratio, which is also accompanied by alterations in the endocannabinoid signaling but not sleep time in general.

Disruption of the E/I ratio (or balance) has been a major theme of proposed mechanisms underlying sensory and behavioral abnormalities observed in autism spectrum disorder patients and animal models. The demonstration and characterization of the shift/flattening of the daily oscillation of E/I in the two mouse models provide strong evidence for a disruption of the daily dynamic regulation of the E/I ratio instead of an overall change in the absolute level of E/I, at least in layer 2/3 pyramidal neurons in the visual cortex examined here. These results call for a re-visit of previous studies and offer a potential explanation to reconcile conflicting prior reports regarding the valence of E/I ratio changes in different autism models and brain areas, taking the recording time during the day into consideration. It also raises the question of how the dysregulated daily E/I oscillation affects brain functions. On the other hand, the dissociation of sleep and E/I oscillation observed in the autism models may also provide an opportunity to better understand the functional relevance of sleep-dependent E/I oscillation in a normal brain in the future.

Reviewer #1 (Public Review):

Pathological conformation and aggregation of tau protein are involved in several neurodegenerative diseases such as tauopathies and Alzheimer's disease. Identifying drug-candidates capable of interfering with pathological transformation of tau remains a challenge for which sensitive and specific assays are needed. This article describes the development and characterization of tau biosensors based on NanoBit technology (nanoluciferase complementation). It is a well-designed and precise study providing very interesting new tools.

Strengths<br /> 1/ The authors have developed a variety of tau biosensors: some that can be used for basic research to monitor pathological tau transformation and others with properties suitable for drug screening. All biosensors make it possible to evaluate the action of different agents (chemical products, purified proteins, cell or tissue extracts) in a living cell.

2/ The intermolecular biosensors developed from a shortened version of tau (K18), or the full-length tau, and carrying the P301L mutation possess sensibility and specificity allowing their further development to identify drug-candidates interfering with tau self-interaction.

3/ The characterization of the tau probes confirmed the physiological and pathological knowledge concerning the tau protein: - proximity of full-length tau when bound to microtubules, - conformational changes of tau during its phosphorylation and - tau-self interaction induced by pathological seeds.

Weaknesses<br /> 1/ The suitability of tau biosensors for high-throughput screening needs to be further developed as experiments were only performed in 96-well plates. No scaling up in 384- or 1536-well plates was attempted. Moreover, the transfection of the biosensors could be tedious.

2/ The models used are basic (HEK-293T cells associated with tau aggregates, Aβ oligomers or mouse brain lysates). The number of drugs tested is also quite low.

Collectively, the conclusions drawn by the authors are supported by the results. These new biosensors will be easily usable by the scientific community in fundamental research and could also be of interest to pharmaceutical laboratories wishing to carry out screenings of molecules capable of impacting the pathological transformation of tau.

Reviewer #2 (Public Review):

Cecon et al presented a series of tau biosensors using the NanoBiT complementation system to monitor tau intramolecular and intermolecular interactions. Three major findings shown in the paper are discussed below.

(1) The authors added two modifications to the existing NanoBiT complementation-based biosensors including K18(P301L) and TauP301L which have the capabilities of monitoring tau-tau interactions in response to phosphorylation and seeding. It is important to first have a thorough characterization of the biosensors such as the basal comparative signals among the different isoforms/mutations (the data in the paper are mostly normalized) and how these signals correspond to their functional units such as whether they are monomers, oligomers or fibrils as confirmed by other biochemistry assays e.g. ThS staining. The interpretation on the functional effect of these biosensors in response to stimulation such as addition of seeds have to be discussed. For example, K18(P301L) biosensor is responding to both mK18 and aggK18 as well as aggTau but not mTau or oAB. It appears that the biosensor is unable to differentiate monomeric and aggregated species of K18 tau. Also, beta-amyloid oligomers have been shown to seed tau aggregation, but this is not the case shown by the study which warrants some discussion. A more thorough characterization of the luciferase biosensors would be essential before moving into other assays and high-throughput screening as it is important to know exactly what kind of tau species are being targeted.

(2) The authors added colchicine, a MT destabilizing drug, to the luciferase biosensor systems and showed that phosphorylation of WT tau takes place when it is still bound to MTs, as colchicine prevented its phosphorylation and suggested that tau species comprising of K18 and full-length WT tau might represent an interesting new therapeutic target, as K18 tau and tau with P301L mutation renders full-length WT tau responsive to seeding. It is an interesting concept to study how tau aggregation changes with respect to MT destabilization. However, it is worth noting that treatment with chemical compounds may cause many other effects that need to be well controlled/eliminated before reaching a conclusion. The authors showed that treatment with colchicine reduces luciferase signals of the tau biosensors and suggested that the luciferase signals arise from MT bound tau which is interesting. While colchicine is a well-known MT destabilization drug, it is still important to test if colchicine itself is perturbing tau-tau interaction as other studies have shown that colchicine might promote tau aggregation and cause cognitive dysfunction. From a different perspective, one might consider that MT destabilization may result in more tau in the cytosols due to their detachment from MTs and hence resulting in enhanced tau-tau interactions which would be reflected by an increased in biosensor signals. Furthermore, if tau proteins are already interacting when they are on the MTs, a disruption in MTs may not disrupt tau-tau interactions and might lead to enhanced tau-tau interactions. However, this is not the case shown in this study and perhaps a discussion on this interpretation would help to clarify some questions. The luciferase signal for tau on MTs might be due to tau being near one another when they are residing on MTs which acts as a scaffold to hold them together and not exactly due to tau-tau interactions. Hence, upon MT destabilization, the tau proteins lost the scaffolds that hold them together and hence results in a reduction in the luciferase signals. In terms of the therapeutic targeting of K18-WT tau complex, it is important to note that K18 has increased the responsiveness of WT tau to seeding by 2-fold as compared to the 107-fold change upon seeding of K18-K18 tau biosensor. Although significant, it is a very small change as compared to the signal obtained from K18 biosensors.

(3) Finally, the authors conducted a proof-of-concept study to illustrate the potential of the luciferase biosensor to be used in high-throughput screening drug discovery. The authors used tau seeds (Tg brain lysates), and not small molecules, to show the increase in luciferase signals with Z-factors of >0.5, which indicates excellent assay condition. The authors then further showed that known compounds reduced tau aggregation in Tg brain lysates and reduced luciferase signals of the biosensors. High throughput screening capability typically refers to the perturbation of biosensors or tau-tau interactions directly by drug compounds. From the experimental setup, it seems like the authors will be using luciferase biosensor in the presence of Tg brain lysates (together as a system) to screen for drug candidates, instead of using the biosensor directly to screen for compounds that have a direct effect in perturbing the biosensor. In this case, the Z-factor should be calculated for positive-control compounds that are applied to the biosensor+Tg lysates system. The IC50 of the compounds tested in this system should be determined and compared with the known IC50 values of these compounds in the available literature. It appears that the compounds are only exhibiting good inhibition at very high concentrations, suggesting the need to test and eliminate any non-specific effect such as compound aggregation at a very high concentration.

Reviewer #3 (Public Review):

The paper by Cecon et al. presents a novel biosensor approach designed to study aspects of Tau aggregation that employ the luciferase-based NanoLuc Binary Technology (NanoBiT). The last decade has seen a rise in the number and variety of Tau biosensor systems, each with its own strengths and weaknesses to study various aspects of Tau aggregation. So far, these have proven to be extremely useful tools for the detection of proteopathic Tau molecules from different origins, by virtue of their capacity to induce easily detectable aggregation of the "endogenous" reporter Tau proteins in the intracellular environment, enabling for example to interrogate the structural features that render the protein pathogenic; in addition, they have been employed for screening of therapeutic candidates that can inhibit or slow down the aggregation process. As regards the study of the aggregation process itself, such systems encounter important limitations in that the modifications done to the protein likely impact reaction rates (both intramolecular and intermolecular interactions) and the aggregation mechanism itself. Additionally, the majority of them rely on overexpression systems, further altering the dynamics of physiological interactions. This paper implements a recently developed and commercially available technology based on nano-luciferase complementation, which has been used to study transient protein-protein interactions but not yet for Tau, and reports on its utility to study both inter- and intra-molecular interactions of Tau in live-cells and seeding activity of exogenously added Tau.

Strengths<br /> The field of Alzheimer's will benefit greatly from cellular models that enable faithful replication of aggregation mechanisms that occur intracellularly involving Tau. The elucidation of high-resolution molecular structures of Tau fibrils from cryo-electron microscopy and the realisation that fibrils from different tauopathies display characteristic folds point to altered cellular states that drive the intrinsically-disordered protein (IDP) Tau to adopt specific conformations that spur pathological aggregation processes. The aggregate burden is known now to correlate well with disease progression. Tau has otherwise been described as a highly soluble protein, yet under certain circumstances it adopts a misfolded conformation that in the proximity of other monomers can template further misfolding and spur aggregation. Several biosensor systems have been developed that detect proteopathic Tau with high sensitivity, most notably those that consist of cell lines expressing intracellular FRET pairs. These have been invaluable to the field and have served to demonstrate that seeding activity strongly correlates with disease aggressiveness in Alzheimer's patients (see Dujardin et al. Nat Med 2021), among other important contributions. There are however major limitations in using these models to study aggregation mechanisms in a cellular context in that they rely on significant structural modifications to the protein that alter the aggregation energy landscape, among other artefactual concerns (e.g., protein overexpression).<br /> This paper sets out to showcase the applicability of the NanoBiT technology on the strength of the considerably smaller size of the fusion proteins. which comprise one large BiT fragment of 17.6 kDa and a small complementation peptide of only 11 amino acids, compared to for instance the popular Tau RD P301S FRET biosensor line that relies on CFP and YFP (both ~27 kDa) Tau-fused constructs as FRET pairs. This is important for interrogating intracellular inter- and intra-molecular interactions as steric effects impact reaction rates and mechanisms. This, coupled with high sensitivity of the bioluminescence signal and amenability for high throughput, comprise the most important advantages of this approach.

Weaknesses<br /> Perhaps the most significant advantage (conceptually) of the NanoBiT technology in this context is the ability to create intramolecular interaction sensors by fusing the fragments to opposite termini. This is especially useful for the N- and C- termini of Tau which are known to be in proximity in certain conformations. The same can be achieved with fluorescence complementation yet with the caveat of introducing larger molecules. Nevertheless, regardless of the smaller dimensions of the fusion protein, the modifications are likely to still alter protein interaction dynamics - this is relevant to both intra- and inter-molecular sensors. While this may not always be a major concern when working with globular proteins, it should be a key consideration when studying Tau aggregation. The energy landscape of intrinsically disordered proteins is highly sensitive to even small structural changes, as exemplified by conformational changes in Tau that render this otherwise highly-soluble protein aggregation-prone. The interaction between the complementary small and large fragments of NanoBiT is reversible and weak (reported as 190 uM), but may still stabilise non-intrinsic conformations. Demonstrating that interaction and aggregation kinetics are not affected significantly compared to the native protein in vitro would be required to support the physiological relevance of the claims related to inter- and intra-molecular interactions.

An additional concern with the intramolecular sensor is the ability to discriminate whether interactions are indeed intramolecular and not intermolecular, this introduces a confound for instance in the interpretation that a reduction in signal with the WT Tau conformation sensor after treatment with colchicine suggest that microtubules stabilise Tau in a conformation where N- and C- termini of a Tau monomer are in proximity, when this could also well be due to intermolecular interactions, or a combination of both (see the continuous stretch of density of Tau along protofilaments in Kellogg et al. Science 2018). Furthermore, the colocalization data is not of high enough quality to support the claims regarding microtubule interactions, in fact there seems to be stronger colocalization with the intramolecular sensor than with the intermolecular one. Better quality images and co-localization analysis are needed to support these interpretations. The paper thus falls short of providing compelling data to regard this method as a physiologically-relevant approach to study Tau molecular interactions.

Artefactual problems stemming from the aforementioned alterations are likely not as important for their applicability as sensors, as other Tau biosensors have shown the ability to detect proteopathic forms in a way that reflects the severity of pathology in various contexts, regardless of whether the ensuing aggregates faithfully replicate those encountered in pathology. It would then be of interest to assess how the NanoBiT technology fares compared to alternative cell models in regard to sensitivity. The paper provides a response curve with tissue extracted from a mouse model of tauopathy. The extracts are not purified for tau which makes comparison with other data difficult given that the degree of tauopathy is model and mouse dependent. A more extensive evaluation of the sensing capacity would be needed to establish sensitivity in a meaningful way, for instance with Tau forms for which concentration can be more appropriately estimated, e.g., recombinant Tau and IP-purified extracts from mouse and human tissues, or a direct comparison with other methods.

Reviewer #1 (Public Review):

The authors use a newly developed object-space memory task comprising of a "Stable" version and "Overlapping" version where two objects are presented in two locations per trial in a square open field. Each version consists of 5 training trials of 5-min presentations of an object-space configuration, with both object locations staying constant across training trials in the Stable condition, and only one object location staying fixed in the Overlapping condition. Memory is tested in a test trial 24 hours later where the opposite configuration is presented - overlapping configuration presented for the Stable condition and stable configuration presented for the Overlapping condition - with the thesis that memory in this test trial for the Overlapping condition will depend on the accumulated memory of spatial patterns over the training trials, whereas memory for the test trial in the Stable condition can be due to episodic memory of last trial or accumulated memory. Memory is quantified using a Discrimination Index (DI), comparing the amount of time animals spend exploring the two object locations.

Here, animals in other groups are also presented with an interference trial equivalent to the test trial, to test if the memory of the Overlapping condition can be disrupted. The behavioral data show that for RGS14 over-expressing animals, memory in the Overlapping condition is diminished compared to controls with no interference or controls where over-expression is inhibited, whereas memory in the Stable condition is enhanced. This is interpreted as interference in semantic-like memory formation, whereas one-shot episodic memory is improved. The authors speculate that increased cortical plasticity should lead to increased and larger delta waves according to the sleep homeostasis hypothesis, and observe that instead increased cortical plasticity leads to less non-REM sleep and smaller delta waves, with more prefrontal neurons with slower firing rates (presumably more plastic neurons). They further report increased hippocampal-cortical theta coherence during task and REM sleep, increased NonREM oscillatory coupling, and changes in hippocampal ripples in RGS14 over-expressing animals.

While these results are interesting, there are several issues that need to be addressed, and the link between physiology and behavioral results is unclear.

1) The behavioral results rely on the interpretation that the Overlapping condition corresponds to semantic-like memory and the Stable condition corresponds to episodic-like memory. While the dissociation in memory performance due to interference seen in these two conditions is intriguing, the Stable condition can correspond not just to the memory of the previous trial but also accumulated memory of a stable spatial pattern over the 5 testing trials, similar to accumulated memory of a changing spatial pattern in the Overlapping pattern.

Here, it is puzzling that in the behavioral control with no interference (Figure 1D), memory in the Stable and Overlapping condition is unchanged in the test trial, with the DI statistically at 0 in the test trial. In the original description of the Object Space task by the authors in the referenced paper, the measure of memory was a Discrimination Index significantly higher than 0 in both the Stable and Overlapping conditions. This discrepancy needs to be reconciled. Is the DI for the interference trial shown in Fig. S1 significantly different than 0? No statistics or description is provided in the figure legend here.

2) The physiology experiments compare Home cage (HC) conditions to the Object Space task (OS) throughout the manuscript. While some differences are seen in the control and RGS14 over-expressing animals, there is no comparison of the Stable vs. Overlapping condition in the physiology experiments. This precludes making explicit links between physiological observations and behavioral effects.

3) The authors speculate that learning will result in larger and more delta waves as per the synaptic homeostasis hypothesis. It should be noted here that an alternative hypothesis is that there should also be a selective increase in synaptic plasticity for learning and consolidation. The authors do observe that control animals show more frequent and higher-amplitude delta waves, but rather than enhancing this process, RGS14 animals with increased plasticity show the opposite effect. How can this be reconciled and linked with the behavioral data in the Stable and Overlapping condition? Similarly, there is an increase in slower-firing neurons in RGS14 over-expressing animals. Slower-firing neurons have been proposed to be more plastic in the hippocampus based on their participation in learned hippocampal sequences, but appropriate references or data are needed to support the assertion that slower-firing neurons in the prefrontal cortex are more plastic.

4) It is noted that changing cortical plasticity influences hippocampal-cortical coupling and hippocampal ripples, suggesting a cortical influence on hippocampal physiological patterns. It has been previously shown that disrupting prefrontal cortical activity does alter hippocampal ripples and hippocampal theta sequences (Schmidt et al., 2019; Schmidt and Redish, 2021). The current results should be discussed in this context.

Reviewer #2 (Public Review):

In this paper, the authors provide evidence to support the longstanding proposition that a dual-learning system/systems-level consolidation (hippocampus attains memories at a fast pace which are eventually transmitted to the slow-learning neocortex) allows rapid acquisition of new memories while protecting pre-existing memories. The authors leverage many techniques (behavior, pharmacology, electrophysiology, modelling) and report a host of behavioral and electrophysiological changes on induction of increased medial prefrontal cortex (mPFC) plasticity which are interesting and will be of significant interest to the broad readership.

The experimental design and analyses are convincing (barring some instances which are discussed below). The following recommendations will bolster the strength/quality of the manuscript:

1. Certain concerns regarding the interpretation and analysis of the behavioral data remain. The authors need to clarify if increased mPFC plasticity leads to only an increase in one-shot memory or 'also' interference of previous information. It seems that the behavioral results could also be explained by the more parsimonious explanation that one-shot memory is improved. Do the current controls tease apart these two scenarios? Additionally, the authors need to clarify why the 'no trial' and 'anisomycin' controls for the stable task perform at chance levels on exposure to a new object-place association on test day (Fig 1D). Finally, further description of how the discrimination index (exploration time of novel-exploration time of familiar/sum of both) is recommended i.e., in the stable condition, which 'object' is chosen as 'novel' (as both are in the same locations) for computing the index (Fig 1). Do negative DI values imply a neophobia to novel objects (and thus are a form of memory; this is also crucial because the modelling results (Fig 1E) use both neophilia and neophobia while negative discrimination indexes are considered similar to 0 for interpreting the behavioral results, as stated on page 3, lines 84-86?

2. The authors report lower firing rates in RGS14414 animals during the task in Fig 2F. It is indeed remarkable how large the reported differences are. The authors need to rule out any differences in the behavioral state of the animals in the two groups during the task, i.e., rest vs. active exploration/movement dynamics. Are only epochs during the task while the animals interact with the objects used for computing the firing rates (same epochs as Fig 1)? If not, doing so will provide a useful comparison with Fig 1. Additionally, although the authors make the case for slow firing rate neurons being important for plasticity (based on Grosmark and Buzsaki, 2016), it is crucial to note that the firing rate dynamic (slow vs. fast) in that study for the hippocampus is defined based on the whole recorded session (predominated by sleep), indeed the firing rates of the two groups (slow vs. fast/plastic vs. rigid) during the task/maze-running do not differ in that study. Therefore, the results here seem incongruent with the Grosmark and Buzsaki paper. Since this finding is central to the main claim of the authors, it either warrants further investigation or a re-interpretation of their results.

3. A concern remains as to how many of the electrophysiological changes they observe (firing rate differences, LFP differences including coupling, sleep state differences, Figs. 2-4) support their main hypothesis or are a by-product of injection of RGS14414 (for instance, one might argue that an increased 'capability' to learn new information/more plasticity might lead to more NREM sleep for consolidation, etc.). The authors need to carefully interpret all their data in light of their main hypothesis, which will substantially improve the quality/strength of the manuscript.

Reviewer #3 (Public Review):

The authors set out to test the idea that memories involve a fast process (for the acquisition of new information) and a slow process (where these memories are progressively transferred/integrated into more-long term storage). The former process involves the hippocampus and the latter the cerebral cortex. This 'dual-learning' system theoretically allows for new learning without causing interference in the consolidation of older memories. They test this idea by artificially increasing plasticity in the pre-limbic cortex and measuring changes in different learning/memory tasks. They also examined electrophysiological changes in sleep, as sleep is linked to memory formation and synaptic plasticity.

The strengths of the study include a) meticulous analyses of a variety of electrophysiological measurements b) a combination of neurobiological and computational tools c) a largely comprehensive analysis of sleep-based changes. Some weaknesses include questions about the technique for increasing cortical plasticity (is this physiological?) and the absence of some additional experiments that would strengthen the conclusions. However, overall, the findings appear to support the general idea under examination.

This study is likely to be very impactful as it provides some really new information about these important neural processes, as well as data that challenges popular ideas about sleep and synaptic plasticity.

Reviewer #2 (Public Review):

The manuscript by Yildiz et al investigates the early response of BECs to high fatty acid treatment. To achieve this, they employ organoids derived from primary isolated BECs and treat them with a FA mix followed by viability studies and analysis of selected lipid metabolism genes, which are upregulated indicating an adjustment to lipid overload. Both organoids with lipid overload and BECs in mice exposed to a HFD show increased BEC proliferation, indicating BEC activation as seen in DR. Applying bulk RNA-sequencing analysis to sorted BECs from HFD mice identified four E2F transcription factors and target genes as upregulated. Functional analysis of knock-out mice showed a clear requirement for E2F1 in mediating HFD induced BEC proliferation. Given the known function of E2Fs the authors performed cell respiration and transcriptome analysis of organoids challenged with FA treatment and found a shift of BECs towards a glycolytic metabolism.

The study is overall well-constructed, including appropriate analysis. Likewise, the manuscript is written clearly and supported by high-quality figures. My major point is the lack of classification of the progression of DR, since the authors investigate the early stages of DR associated with lipid overload reminiscent of stages preceding late NAFLD fibrosis. How are early stages distinguished from later stages in this study? Molecularly and/or morphologically? While the presented data are very suggestive, a more substantial description would support the findings and resulting claims.

Reviewer #2 (Public Review):

It is believed that the reason why women generally have lower rates of atherosclerotic events than men until menopause is due to the beneficial effects of estrogen on the cardiovascular system. The paper attempts to explain why hormone replacement therapy with estrogen is not effective in preventing atherosclerosis in post-menopausal women. The authors posit that accumulation of iron after menopause inhibits estrogen receptor expression and makes estrogen ineffective. Using mouse model of atherosclerosis and iron overload, they demonstrate that 1)atherosclerosis is increased in overectomized mice 2) estrogen supplement seems to further exacerbate atherosclerosis and this is accompanied by increased total body iron; 3) iron itself causes a decrease in ERa via increased proteasome degradation of Era via E3 ligase MDM2 and 4) iron chelation rescues the protective effects of estrogen in overectomized mice on atherosclerosis progression.

While interesting in terms of hypothesis, I found the manuscript (not the overall themes) but the individual experimental logic difficult to follow with unclear rationale for many of the experiments and timepoints chosen. Moreover the human data supporting these claims are weak in terms of what is shown. The authors only partially achieve their aims as many of the experiments in mice appear incomplete in terms of data shown and transparency. Some important controls are also missing.

This work has important potential to understand the causes of accelerated atherosclerosis in women after menopause and how to better prevent atherosclerosis in women of this age group

Reviewer #2 (Public Review):

In this manuscript, Nguyen et al. make use of recently determined cryo-EM structures of Nav1.7 channels in complex with ProTX-II, a peptide spider toxin that binds to VSD2 and stabilizes the deactivated state of the channel in addition to reducing peak currents. Previous work on making modified spider toxin peptides as potent and selective Nav1.7 inhibitors by Merck, Amgen, and others was conducted in a structure-blind manner. This manuscript demonstrates that it is possible to use structure data and computational tools to identify modified spider toxin peptides that show even better potency and selectivity properties.

The authors did a very nice job presenting their detailed results. This detailed material should be very helpful to researchers wanting to expand on this work toward the development of peptide-based pain drugs that selectively target Nav1.7. Their in-vitro electrophysiological analysis is excellent, showing full selectivity profiles (including difficult to work with channels such as hNav1.8 and hNav1.9) from HEK293 cells and also showing inhibition of the TTX-S current with both mouse and human cultured DRG neurons. The in-vivo work shows very strong analgesia in the hotplate model as well as in a model of oxaliplatin-induced peripheral neuroparthy, showing that PTx2-3127 is a powerful analgesic in rats.

Overall, this is an excellent investigation into the feasibility of using structural information and computational tools to design potent and selective Nav1.7 inhibitors. Such peptide-based inhibitors might be developed in the future as novel pain drugs.

Reviewer #2 (Public Review):

Jelen et al. developed a new taste conditioning paradigm where they pair a tastant (CS) with optogenetic activation of either sensory neurons or dopamine neurons. Activation of different cell types in training led to decreased sugar attraction or decreased salt avoidance. Depending on the activated cell type, the authors could even induce LTM with optogenetic activation. They found that the neural requirement for aversive or appetitive taste learning widely overlaps with the requirement for learning with other modalities (olfaction). They focus also on appetitive taste LTM formation, which requires caloric food intake after training similar to olfactory LTM.

Strengths:

The newly developed operant paradigm has several advantages compared to previous taste learning paradigms. The flies are freely walking and can be monitored throughout training and test. This allowed the authors to describe the temporal dynamics of learning and learned behavior. They could show that a specific type of dopamine neuron enhances salt sipping during training but was not sufficient to induce learning. Furthermore, they could now investigate both, appetitive and aversive learning, which was not possible before in immobilized flies. Optogenetic activation as the US in training allowed the authors to disentangle the need for caloric value in short-term and long-term memory.

Weaknesses:

Artificial activation of neurons seems to be sufficient to induce different memories in the fly. However, as the flies do not receive actual food in the training, those results may not represent the naturally used neural circuits, or only partial circuits underlying the normal taste learning. Also, the new paradigm has operant training, which might change the requirement or recruitment of learning circuits. Thus, the authors find similar neurons involved as in classical conditioning, which is very interesting, but also some differences.

Reviewer #2 (Public Review):

Kankaanpää and colleagues studied how lifestyle factors in adolescence (e.g., smoking, BMI, alcohol and exercise) associate with advanced epigenetic age in early adulthood.

Strengths:

The manuscript is very well written. Although the analyses and results are complex, the authors manage very well to convey the key messages.<br /> The twin dataset is large and longitudinal, making this an excellent resource to assess the research questions.<br /> The analyses are advanced including LCA capitalizing on the strength of these data.<br /> The authors also include a wider range of epigenetic age measures (n=6) as well as a broader range of lifestyle habits. This provides a more comprehensive view that also acknowledges that associations were not uniform across all epigenetic age measures.

Weaknesses:

The accuracy of the epigenetic age predictions was moderate with quite large mean absolute errors (e.g., +7 years for Horvath and -9 years for PhenoAge). Also, no correlations with chronological age are presented. With these large errors it is difficult to tease apart meaningful deviations (between chronological and biological age) from prediction error.

The authors claim that 'the unhealthiest lifestyle class, in which smoking and alcohol use co-occurred, exhibited accelerated biological aging...'. However, this is only partially true. For example, PhenoAge was not accelerated in lifestyle class C5. Similarly, all classes showed some degree of deceleration (not acceleration) with respect to DunedinPACE (Figure 3D). The large degree of heterogeneity across different epigenetic age measures needs to be acknowledged.

The authors claim that 'Practically all variance of AAPheno and DunedinPACE common with adolescent lifestyle was explained by shared genetic factors'. However, Figure 4 suggest that most of the variation (up to 96%) remained unexplained and genetics only explained around 10-15% of total variation. The large amount of unexplained variation should be acknowledged.

Reviewer #2 (Public Review):

Little is known about how the circadian clock regulates the timing of anthesis. This manuscript shows that the circadian clock regulates the diurnal rhythms in floral development of the sunflower. The authors have developed a new method to characterize the timing of floral development under normal conditions as well as constant dark and light conditions. The results from the treatment of darkness during the subjective night and day suggest that the circadian clock regulates the growth of ovary, stamen, and style differently.

All clock papers claim that the circadian clock regulates the fitness of organisms, however, it is hard to evaluate how the circadian clock affects the fitness of organisms. The timing of pollen release and stigma maturity is directly related to plant fitness. That's why the authors suggest that the circadian clock in sunflowers increases plant fitness by regulating the timing of anthesis.

Although the authors manipulated the light and temperature to examine the role of the circadian clock in floral development, the weakness of this manuscript is that there is no molecular evidence to show how the clock regulates floral development.

Reviewer #1 (Public Review):

The authors push a fresh perspective with a sufficiently sophisticated and novel methodology. I have some remaining reservations that concern the actual make-up of the data basis and consistency of results between the two (N=16) samples, the statistical analysis, as well as the "travelling" part.

I previously commented on the fact that findings from both datasets were difficult to discern and more effort should be made to highlight these. Also, a major conclusion "the directionality effect [effect of attention on forward waves] only occurs for visual stimulation" only rested on a qualitative comparison between studies. The authors have improved on this here, e.g., by toning down this conclusion. One thing that is still missing is a graphical representation of the data from Foster et al. (the second dataset analysed here) that would support the statistical results and allow the reader a visual comparison between the sets of findings.

Also, for any naive reader, the concept of travelling waves may be hard to grasp in the way data are currently presented - only based on the results of the 2D-FFT. Can forward and backward-travelling waves be illustrated in a representative example to make this more intuitive?

Finally, the way Bayes Factors from the Bayesian ANOVA are presented, especially with those close to the 'meaningful boundaries' ⅓ and 3, as defined in the 'Statistical analysis' section, requires some unification/revision. For example, here: "We found a positive correlation between contra- and ipsi- lateral backward waves, and occipital (all Pearson's r~=0.4, all BFs 10 ~=3) and -to a smaller extent- frontal areas (all Pearson's r~=0.3, all BFs 10 ~=2).", where the second part should strictly be labelled as inconclusive evidence. In the same vein, there is occasional mention of "negative effects", where it should say that evidence favours the absence of an effect.

Reviewer #2 (Public Review):

The present manuscript takes a new perspective and investigates the functional relevance of traveling alpha waves' direction for visual spatial attention. While the modulation of alpha oscillatory power - and especially the lateralization of alpha power - has been associated with spatial attention in the literature, the present investigation offers a new perspective that helps understand and differentiate the functional roles of alpha oscillations in the ipsi- versus contralateral hemisphere for spatial attention.

The present study uses a straightforward approach and provides an analysis of two EEG datasets, which are convergingly in line with the authors' claim that two patterns of travelling alpha waves need to be differentiated in visual spatial attention. First, backward waves in the ipsilateral hemisphere, and second, forward waves in the contralateral hemisphere, which are only observed during visual stimulation. Importantly, the authors test the relation of these patterns of traveling waves to the overall power of alpha oscillations and to the hemispheric lateralization of alpha power. Furthermore, to test the functional significance, the authors demonstrate that the pattern of forward and backward waves around stimulus onset differentiates between hits and misses in task performance.

Although the results are in line with the conclusions drawn, some questions remain. The authors investigate the relationship between traveling alpha waves and the hemispheric lateralization of alpha power, which is a well-established neural signature of spatial attention. Surprisingly, the lateralization of alpha power shown in Figure 3B appears relatively weak in the present dataset (by visual inspection), which raises the question of whether the investigation of a relation between lateralized alpha power and alpha traveling waves is warranted in the first place.

Furthermore, the authors employ between-subject correlations (with N = 16) to test the relationship between alpha traveling waves and (lateralized) alpha power. However, as inter-individual differences in patterns of travelling waves are not the main focus here, within-subject analyses of the same relations would be able to test the authors' hypotheses much more directly.

It needs to be appreciated that the authors analyze two datasets in the present study. However, the question remains whether the absence of the forward waves effect in paradigms without visual stimulation is a general one and would replicate in other datasets. Moreover, the manuscript would benefit from a discussion of the potential implications of traveling waves for functional connectivity between posterior and anterior regions.

Reviewer #1 (Public Review):

The manuscript by Silva et al. "Evaluation of the highly conserved S2 hairpin hinge as a pan-coronavirus target" seeks to evaluate a new epitope target on the S2 domain of SARS-CoV2 Spike protein and evaluate its potential as a pan-coronavirus target. This is an impressive combination of extensive structural, HDXMS-based dynamics and antibody engineering approaches. What is missing is a detailed correlation of HDXMS with Spike dynamics. The authors have not examined the allosteric effects of 3A3 binding to the Spike trimer, specifically cooperativity in antibody binding. Does binding of one Fab positively or negatively impact the subsequent binding of antibody? In this regard, readers would benefit from HDXMS spectral envelopes in figures, at least for the epitope locus peptides. Further, what is the effect of the intrinsic ensemble behavior of the Spike protein on 3A3 interactions? In a broader sense antibody binding is assisted by intrinsic trimer ensemble behavior, as observed by the lowered binding to the omicron variant- but are there induced binding effects? It would help to better integrate HDXMS with cryo-EM and antibody engineering. It is a novel, less explored epitope target on the S2 domain. Overall, a more definitive mechanistic conclusion for how targeting the S2 hinge can advance future pan-coronavirus strategies is missing.

Major Comments:

1) Given that the authors have demonstrated ensemble switching behavior from 4 ℃ to 37 ℃ (Costello et al. (2021)) why is this not factored in how the HDXMS is carried out? The samples were stored, frozen at -80 ℃, thawed, and equilibrated for 20 min at 20 ℃ with or without antibody present and analyzed by HDXMS. However, the reported t1/2 for trimer tightening at 37 ℃ is t1/2 = 2.5 h (Supplementary Fig. 7). The samples should ideally be analyzed under standardized conditions with the stable conformer. Sample heterogeneity from HDXMS is likely due to any of the following contributing factors:<br /> i) Intrinsic ensemble heterogeneity (Costello et al. (2021)), Kinetics of RBD- up and down conformational switching<br /> ii) Cooperativity of Fab binding.<br /> iii) Partial occupancy of trimer epitopes with bivalent IgG.<br /> iv) Combination of cooperativity effects and partial binding effects

I would predict for any of the above reasons, it is intriguing why are there no bimodal kinetics of deuterium exchange reported. Partial occupancy should be evident from HDXMS paratope analysis.

2) Pan-coronavirus neutralization potential is clearly evident. It is intriguing that the antibodies were isolated after immunization with an authentic MERS S2 domain but showed better selectivity to full-length 6P-engineered Spike. How is cooperativity built into antibody binding, given that the epitope site is occluded to various extents by the S1 domain and access is contingent upon RBD up-down kinetics?

3) I am surprised that there is no allostery described for 3A3 (Supplementary figures 5, 6).

Reviewer #2 (Public Review):

The authors report a conserved spike S2 hinge epitopes and two conformationally selective antibodies that help elucidate spike behavior. This work defines a third class of S2 antibody and provides insights into the potency and limitations of targeting this S2 epitope for future pan-coronavirus strategies.

Reviewer #3 (Public Review):

The study by Silva et al details the discovery and evaluation of a third class of broadly cross-reactive anti-Spike antibody that binds a conserved hinge region in the S2 domain. After immunizing mice with a stabilized S2 protein from MERS and generating scFv phage libraries, the authors were able to identify antibody 3A3, which showed broad cross-reactivity with SARS2 (including Omicron BA.1), SARS1, MERS, and HKU1 spike proteins. Using a combination of a low-resolution cryo-EM structure and HDX mass spectrometry, the authors were able to map amino acids in the antibody paratope and spike epitope, the latter of which is the hinge region of the Spike S2 domain (residues 980-1005) that plays a critical role in pre- to -post-fusion conformational changes. Through well-executed and comprehensive mutagenesis, binding, and functional assays, the authors further validated critical residues that lead to antibody escape, which centered around the 2P residues and diminished viral entry. While 3A3 and an affinity-enhanced engineered version, RAY53, did not show potent in vitro neutralization against the authentic virus, the antibody was shown to recruit Fc effector functions for viral clearance, in vitro.

Overall, the conclusions of this paper are well supported by the data, but the usefulness of such antibodies is likely limited. The work can be strengthened by extending the analysis of 3A3-like antibodies in the context of human immune responses and in vivo effectiveness.

1. Isolation of 3A3 was achieved after the generation of scFv-phage libraries following immunization with a MERS S2-domain immunogen in a mouse model. The fact that 3A3 binds well to 2P-stabilized sequences and binding/neutralization is diminished upon reversion of 2P mutations back to the native spike sequence (Figures 3a, 4c, and 5b), suggest that such antibodies would likely not arise from natural infection. This contrasts the isolation of fusion peptide and stem helix-directed antibodies, which were isolated from both immunized animals and convalescent individuals. To make their results more solid regarding the use of such antibodies in future vaccine strategies, the authors should provide evidence that 3A3-like antibodies can be identified in human donors. For example, they could enrich donor-derived S2-specific antibodies that bind both MERS and SARS2 S2 domains and evaluate the fraction of antibodies that recognize the hinge-epitope using competition binding assays (either ELISA or BLI), which have commonly been used to map epitope-specific sera responses. This could also be achieved with nsEMPEM of polyclonal IgGs bound to S2 proteins.

2. The authors speculate in the discussion that strategies to enhance access to the hinge epitope, which may include ACE2-mimicking antibodies, could promote enhanced viral clearance. In addition to ACE2-mimicking antibodies, several antibodies have been described that bind the RBD and promote S1 shedding (see for instance mAb S2A4 - Piccoli et al, 2020, Cell). Several 2nd generation vaccine platforms utilize RBD-only immunogens that are likely to induce high titers of ACE2-mimicking and cross-reactive S1-shedding antibodies. Thus, adding in vitro neutralization and ADCC experiments to assess synergy between 3A3/RAY53 and such antibodies would booster this speculative claim and be of interest to many in the field developing strategies for pan-coronavirus therapies.

3. The authors provide in vitro evidence in Figure 5c,d for Fc-mediated viral clearance. While in vivo data to show effectiveness in animal models is ideal, additional in vitro data that utilize engineered constructs that modulate effector function (e.g., DLE (+) or LALA (-)) would boost the authors' claims regarding Fc-mediated viral clearance mechanisms by EA3/RAY53.

Reviewer #1 (Public Review):

Nephronophthisis (Nphp) is a multigenic, recessive disorder of the kidney presenting in childhood that is characterized by cysts predominantly at the cortico-medullary junction and progressive fibrosis. An infantile form of the disease presents earlier with more diffuse cystic change. The condition is considered a ciliopathy because most of the genes linked to the condition encode proteins involved in ciliary biogenesis or function. Germline mutations in NPHP2 are associated with a particularly severe, infantile form of the disease. Given that interstitial fibrosis is a more prominent feature of Nphp compared to many other forms of polycystic kidney disease, the authors sought to determine the mutant cell types responsible for the phenotype.

In the current study, the authors generated and characterized mouse lines with Nphp2 selectively inactivated in either renal epithelial cell or stromal cell lineages and found that inactivation in renal epithelial cells was both necessary and sufficient to cause disease. They further showed that markers of interstitial fibrosis and proliferation increase in mutants prior to the onset of histologically evident cystic disease, suggesting that aberrant epithelial-stromal cell signaling is an early and primary feature of the condition (Figures 1-4). The study design was straightforward and appropriate to address the question, and the results support their conclusions.

They next tested whether the cilia-dependent cyst-activating pathway (CDCA) that is "unmasked" by loss of other PKD-related genes is similarly active in Nphp2 mutants by generating Nphp2/Ift88 double mutants. Their studies found that the severity of cystic disease and markers of proliferation and fibrosis was attenuated in double-mutants (Fig 5, 6). These studies were also appropriate for testing the hypothesis and the results were similarly consistent with their interpretation.

In the last set of studies, they tested whether valproic acid (VPA), a drug that has multiple modes of action including acting as a broad inhibitor of HDACs and previously used by the investigators in other forms of polycystic kidney disease, would have similar effects in Nphp2 mutants. The authors tested daily injection from days P10 through P28 in both control and Nphp2 mutant mice with VPA or an appropriate vehicle control and found that VPA was beneficial (Fig 7). The study design was acceptable and the results generally support their conclusions. The one perplexing result is shown in Fig 7B. The Nphp2 mutants, regardless of treatment status, have body weights (BW) that are significantly lower than the controls, with treated mutants even trending lower than their untreated mutant counterparts. This is unexplained and should be addressed. In the mutants with more widespread epithelial cell knock-out of Nphp2 (Ksp-Cre, Fig 1), total body weight decreased as mice became more severely cystic with renal impairment. In the milder form of disease produced with the Pkhd1-Cre (Fig 7), total body weight is inexplicably approx. 2g lower on average despite having much more modestly elevated KBWs and BUNs. Moreover, one might have expected that mutants treated with VPA would have had BWs intermediate between untreated mutants and controls since the severity of the disease was moderately attenuated. These differences raise the question as to whether body weight differences are due to factors independent of disease status, the most likely of which would be that the controls were not littermates. This prompted a careful review of the text for descriptions of the control mice. Throughout the study, the investigators describe selecting animals from the same "cohort", but this term is imprecise. There is little information provided about background strains, whether any of the lines were congenic, or whether any of the studies were done using littermate controls. This must be addressed. It would help if the investigators identified the litter status in their plots. This would clearly show relationships between animals and the number of litters that had animals with these properties. If littermates were not used for each study, the authors must explain both why they didn't do so and how they then selected which animals to use. This information is especially important for interpreting the results of their genetic interaction and drug treatment studies.

Several other considerations. The authors state that the effects of VPA are mediated through the drug's inhibition of HDACs and suggest that future studies could be directed at refining the specific HDAC. While this is certainly possible, the authors should acknowledge that VPAs have been reported to have numerous pharmacologic effects and targets and which of these is mediating the effects in their model is unknown. They would need mechanistic studies to show this, though it doesn't discount their possible efficacy as a therapy for PKD. The authors also state in their abstract that their double knock-out studies "support a significant role of cilia in Nphp2 function in vivo." It is not clear to me how their studies show this nor how they can exclude that ciliary activity is operating in an Nphp2-independent, parallel fashion that modulates some common downstream pathways.

Reviewer #2 (Public Review):

The manuscript by Li et al demonstrates the role of Nphp2/Invs in renal epithelia in preventing NPHP-like phenotypes, such as epithelial/stromal proliferation and stromal fibrosis, in mice. Previously, mutants of the Nphp2 allele in mice, generated by insertional mutagenesis, showed severe cystic kidney disease and fibrosis in neonates.

The authors nicely show that the NPHP-like phenotypes in mutant kidneys arise from abnormal signaling specifically within and from renal epithelial cells. Furthermore, the fibrotic response and abnormal increase of cell proliferation precede cyst formation and could be initiated independently of cyst formation. The authors also show that the removal of cilia reduces the severity of Nphp2 phenotypes. The authors suggest that similar to polycystins, NPHP2 inhibits a cilia-dependent cyst and fibrosis-activating pathway. Finally, the histone deacetylase (HDAC) inhibitor valproic acid (VPA) reduces these phenotypes and preserves kidney function in Nphp2 mutant mice, supporting HDAC inhibitors as potential candidate drugs for treating NPHP.

Overall, understanding the mechanisms driving NPHP phenotypes is important and drugging relevant pathways in treating this disease is an important unmet need in patients. The authors have provided insights into both these aspects in this study. The manuscript is nicely written, and the assays shown are rigorous and insightful.

Reviewer #3 (Public Review):

In this manuscript, Li et. al, investigate whether epithelial or stromal Nphp2 loss, a gene causative of nephronophthisis (NPHP), drives polycystic kidney disease (PKD) and kidney fibrosis in a novel floxed model of Nphp2. The authors found that only epithelial and not stromal Nphp2 loss results in NPHP-like phenotypes in their mouse model. In addition, the authors show that concurrent cilia, via Ift88 loss, and Nphp2 loss within the kidney epithelium as well as HDAC inhibition results in less severe PKD/kidney fibrosis, as has been shown in mouse models of other non-syndromic forms of PKD, such as autosomal dominant PKD caused by mutations to Pkd1 or Pkd2.

The authors aimed to understand (1) whether the published NPHP phenotype (kidney cysts and fibrosis), known from the global Nphp2 knockout mouse, is driven by the function of NPHP2 in the kidney epithelium or stromal cells; (2) if kidney fibrosis in NPHP is linked to kidney damage caused by cysts, or independent and preceding of the PKD phenotype; (3) whether cilia are required, causative, or prohibitive of NPHP cystogenesis; and (4) if a broad spectrum HDAC inhibitor is a potential therapeutic approach for NPHP.

With the provided results, the authors established that epithelial Nphp2 loss is likely a predominant driver of PKD in their model; however, they cannot exclude that stromal NPHP2 does not play a role in cysts growth post-initiation because the authors failed to directly compare their cell type-specific models to a global cre knockout (e.g. Cagg-cre). In addition, it is possible that cyst initiation/growth upon stromal Nphp2 loss occurs substantially slower compared to epithelial Nphp2 loss. However, it seems the authors did not look at kidney phenotypes beyond 28 days of age. Publications from the ADPKD field suggest, that stromal Pkd1 loss initiates cystogenesis much slower than epithelial Pkd1 loss. Further, while the authors suggest that kidney fibrosis precedes cyst development, the results supporting this conclusion are limited to one time point, analyzing IF staining of a single marker that can be compared between non-cystic and cystic time points. These analyses need to be extended to make any firm conclusions.

The most interesting finding of the manuscript, and likely most impactful to the field, is, that loss of cilia within the setting of epithelial Nphp2 loss reduces PKD severity. This finding parallels published findings for Pkd1 and Pkd2 which are suggested to function in a cilia-dependent cyst-activation mechanism. Unfortunately, the here shown studies, do not add to the mechanistic insight beyond showing the descriptive finding. Most importantly, it remains unclear whether NPHP2 functions in the same pathway as polycystin-1 or -2 (the Pkd1, Pkd2 gene products) or in a separate independent pathway.

With respect to the HDAC preclinical studies, the authors show supporting data that a broad-spectrum HDAC inhibitor may be suitable for slowing cyst growth in their model of NPHP. Overall, these studies are not novel to the field, as HDAC inhibition has been shown to slow PKD progression in various models of PKD al while not in NPHP specifically. Further, the studies seem like an add-on, which does not directly link to the prior cell type-specific studies of NPHP2, and no mechanisms linking the two concepts are provided.

Reviewer #1 (Public Review):

The paper reports important work in which the Fub-1 boundary of the Drosophila bithorax complex is characterized in detail. Fub-1 separates the bxd/pbx regulatory domain, which is active in PS6/A1, from the abx/bx regulatory domain, which is active in PS5/T3. The work presented provides compelling evidence that Fub-1 consists of two key elements: an insulating boundary region called HS1, which is regulated by an adjacent region called HS2. HS2 contains a promoter that is activated in PS6/A1 by enhancers in the bxd/pbx region. Read-through of HS1 by transcripts from the HS2 promoter blocks the insulating activity of HS1, allowing the bxd/pbx regulatory regions to activate Ubx transcription in PS6/A1. It has long been appreciated that boundary elements within the BX-C are regulated in a segment-specific fashion. The work presented in the Ibragimov manuscript provides a very nice example of how this segment-specific regulation can take place. For the most part, the work is very thorough and the conclusions are well-supported. However, there are a few important issues that should be addressed.

First, throughout the manuscript, it is stated that the read-through transcription of HS1 eliminates its blocking activity. Missing, however, is a test of whether the direction of transcription of HS1 is important. That is, no construct is tested in which HS1 is inverted so that RNAs from the HS2 promoter are transcribed from the opposite strand of HS1. If read-through transcription of HS1 is all that is required to abrogate its blocking activity, such a construct should behave identically to constructs in which HS1 is not inverted. However, if the structure of the F1HS2 RNA is critical to preventing the blocking activity of HS1, inversion of HS1 relative to HS2 may render it immune to inactivation by transcripts initiated at HS2.

Second, the terminology used to designate the constructs tested is very hard to follow and needs simplification. Since the orientation of HS1 in isolation is unimportant, perhaps just HS1 HS2, HS1 Inv(HS2), HS2 HS1, and Inv(HS2) HS1 could be used.

Third, in many places in the manuscript genotypes are shown in which the HS1 insulator is placed into F7attP50. For these genotypes, H1 is said to block the interaction between iab-6 and iab-7, but not to support bypass activity. Readers need some help here, as they will not understand why A5 and A6 tergites are black in these genotypes, as this implies that iab-5 is able to act over the HS1 element to activate Abd-B. One explanation may be that iab-5 can promote pigmentation by acting on abd-A.

Fourth, a more complete description of the HS1248 HS2505R genotype is needed. In this genotype, the H1 insulator is constitutively active, as H2 is inverted. Do animals of this genotype show a bxd phenotype in the larval cuticle? Do adults show a transformation of the halteres like that shown by classical bxd mutations? Answers to these questions would shed light on when H1 is active as an insulator, and whether it is active throughout PS6/A1.

Reviewer #2 (Public Review):

The work presented in the manuscript addresses regulatory mechanisms in a complex genome locus, the Bithorax-Complex (BX-C) in Drosophila. Here three homeotic genes are controlled by multiple regulatory domains, each of which comprises distinct sets of cis-regulatory elements including insulators, enhancers, Polycomb responsive elements, and promoters for coding and non-coding transcripts. Despite such complexity, the authors have made good efforts to explain the context for the study and the question that they are interested in, what is the function of an evolutionarily conserved but newly defined cis-element, Fub-1?

Fub-1 localizes at the chromatin boundary between the homeotic gene Ubx and the bxd/pbx regulatory domain, which thus predicts it is a chromatin insulator. To dissect the function of Fub-1, the authors utilized powerful and versatile gene exchange cassettes (phiC31/attp; FRT/FLP; Cre/Loxp) to engineer both the endogenous locus of Fub-1 and another insulator site Fab-7 to introduce exogenous Fub-1. Using these transgenic tools, they tested the insulator activity of Fub-1. They first confirmed that deleting Fub-1 causes changes in chromosomal configuration in the flanking region using Micro-C. However, unexpectedly, they found that Fub-1 depletion does not cause homeotic transformation, a phenotype that is expected to occur when the expression of the homeotic gene is changed due to the loss of chromatin insulators. Instead, they observed that only a sub-element within Fub-1 has an insulator function while the other sub-element that contains an active promoter suppresses insulator activity. They further demonstrated that although there is no detectable phenotype when both sub-elements are deleted, changing the direction of the promoter or stopping transcription by adding an SV40 terminator in between the two sub-elements could relieve the suppression of insulator activity. From this evidence, the authors conclude that transcriptional read-through from the active promoter of a non-coding transcript regulates the insulator activity of Fub-1.

The finding provides a new angle to examine regulation by insulators and reveals a new function of active promoters of non-coding transcripts. The work also leaves further questions, for example, how general is such a mechanism in the genome organization of Drosophila and other organisms, and what is the significance of the mechanism given that deleting the Fub-1 insulator does not cause phenotypic outcomes in Drosophila? In the discussion, the authors elaborated on possibilities to discuss these questions.

Reviewer #1 (Public Review):

Maksim Kleverov et al. developed the tool called Phantasus, a web application for matrix visualization and analysis of gene expression data generated by either microarray or RNA-seq technologies. By Phantasus, the users can load, normalize, and plot their own data or those available in public databases and investigate the samples to remove outliers before the differential expression analysis.

Phantasus can be accessed on-line or can be installed locally from Bioconductor.<br /> One of the advantages of the web application is that it combines an interactive graphical user interface with access to various R-based analysis methods. For the methods that rely on functions that are already available in the existing R packages, for such practices, only wrapper R functions are implemented. The tool was developed focusing on being helpful to both expert and non-expert users in bioinformatic gene expression analysis.

Reviewer #2 (Public Review):

Maksim et al. present Phantasus, a web application for interactive gene expression analysis. The tool allows the user to load microarrays and RNA-Seq data from NCBI GEO.<br /> The user is able to explore, normalize, filter and perform differential expression analysis using limma or DESeq2 pipelines for microarray and RNA-Seq data, respectively. The web tool is capable of generating figures such as PCA and volcano plots and performing gene set enrichment analysis. Phantasus has some advantages when compared to the set of tools already available, showing a good trade-off between ease of use, access to data and different functions. Furthermore, the application is open source and the pre-processed cache files are provided by the authors. Thus, the more experienced user can install the tool on a local computer.

Finally, Phantasus is limited to standardized analyzes available in its internal methods and databases, which may not meet the needs of researchers who wish to apply different types of quantification and normalization. However, this is the ideal tool for the non-bioinformatics user who wants to reanalyze public data or perform simple differential expression analyzes on their own data.

Reviewer #3 (Public Review):

Software UX design is not a trivial task and a point-and-click interface may become difficult to use or misleading when such design is not very well crafted. While Phantasus is a laudable effort to bring some of the out-of-the box transcriptomics workflows closer to the broader community of point-and-click users, there are a number of shortcomings that the authors may want to consider improving. Here I list the ones I found running Phantasus locally through the available Bioconductor package:

1. The feature of loading in one click one of the thousands of available GEO datasets is great. However, one important use of any such interfaces is the possibility for the users to analyze his/her own data. One of the standard formats for storing tables of RNA-seq counts are CSV files. However, if we try to upload from the computer a CSV file with expression data, such as the counts stored in the file GSE120660_PCamerge_hg38.csv.gz from https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE120660, a first problem is that the system does not recognize that the CSV file is compressed. A second problem is that it does not recognize that values are separated by commas, the very original CSV format, giving a cryptic error "columnVector is undefined". If we transform the CSV format into tab-separated values (TSV) format, then it works, but this constitutes already a first barrier for the target user of Phantasus.

2. Many RNA-seq processing pipelines use Ensembl annotations, which for the purpose of downstream interpretation of the analysis, need to be translated into HUGO gene symbols. When I try to annotate the rows to translate the<br /> Ensembl gene identifiers, I get the error

"There is no AnnotationDB on server. Ask administrator to put AnnotationDB sqlite databases in cacheDir/annotationdb folder"

3. When trying to normalize the RNA-seq counts, there are no standard options such as within-library (RPKM, FPKM) or between-library (TMM) normalization procedures. If I take log2(1+x) a new tab is created with the normalized data, but it's not easy to realize what happened because the tab has the same name as the previous one and while the colors of the heatmap changed to reflect the new scale of the data, this is quite subtle. This may cause that an unexperienced user to apply the same normalization step again on the normalized data. Ideally, the interface should lead the user through a pipeline, reducing unnecessary degrees of freedom associated with each step.

4. 4. Phantasus allows one to filter out lowly-expressed genes by averaging expression of genes across samples and discarding/selecting genes using some cutoff value on that average. This strategy is fine, but to make an informed decision on that cutoff it would be useful to see a density plot of those averages that would allow one to identify the modes of low and high expression and decide the cutoff value that separates them. It would be also nice to have an interface to the filterByExpr() function from the edgeR package, which provides more control on how to filter out lowly-expressed genes.

5. When attempting a differential expression (DE) analysis, a popup window appears saying:

"Your dataset is filtered. Limma will apply to unfiltered dataset. Consider using New Heat Map tool."

One of the main purposes of filtering lowly-expressed genes is mainly to conduct a DE analysis afterwards, so it does not make sense that the tool says that such an analysis will be done on the unfiltered dataset. The reference to the "New Heat Map tool" is vague and unclear where should the user look for that other tool, without any further information or link.

6. The DE analysis only allows for a two-sample group comparison, which is an important limitation in the question we may want to address. The construction of more complex designs could be graphically aided by using the ExploreModelMatrix Bioconductor package (Soneson et al, F1000Research, 2020).

7. When trying to perform a pathway analysis with FGSEA, I get the following error:

"Couldn't load FGSEA meta information. Please try again in a moment. Error: cannot open the connection In call: file(file, "rt")

Finally, there have been already some efforts to approach R and Bioconductor transcriptomics pipelines to point-and-click users, such as iSEE (Rue-Albrecht et al, 2018) and GeneTonic (Marini et al, 2021) but they are not compared or at least cited in the present work. One nice features of these two tools that I missed in Phantasus is the possibility of generating the R code that produces the analysis performed through the interface. This is important to provide a way to ensure the reproducibility of the analyses performed.

Reviewer #1 (Public Review):

In this work, the authors propose a "transfer learning" approach for modeling the properties of sequences that are selected from larger sequence pools on the basis of biophysical or functional properties, where the source populations may themselves be biased in composition. Examples include the set of immunogenic peptides, considered as a subset of all HLA-presented peptides, or the set of TCRs that are specific for a given peptide epitope, as selected from within the much larger pool of all peripheral TCRs. The motivation for transfer learning is that there may only be small numbers of selected sequences available for training and many more examples of the background sequences. Rather than directly fitting a single model on the selected sequences, the idea is to first fit a background model that captures the properties of the source/background population of sequences, using the many examples available for training, and then train a "differential" model that specifically seeks to capture the differences between the selected and background populations. This differential model is trained using the subset of selected sequences, by optimizing their likelihood under a composite model that combines the background model (whose parameters are frozen) and the differential model. The specific architecture used here is the "restricted Boltzmann machine" (RBM), which can be thought of as a generalization of the position-weight matrix approach that can capture pairwise and higher-order interactions between positions. The applications are the two mentioned above, prediction of immunogenic peptides and prediction of TCRs specific for a given peptide-MHC epitope. This work builds on previous work by the authors applying the RBM architecture to peptide-MHC binding [Bravi et al., 2021b] and T-cell responses [Bravi et al., 2021a]. The advance here is in formalizing the "differential" framework and testing immunogenicity prediction and epitope specificity. Considering the field and the current state of the art, the main contributions of the manuscript appear to be theoretical/conceptual, in introducing the "diffRBM" method and providing a range of evaluations of its performance, for example, the use of contact prediction to assess the model. For TCR-epitope prediction, it does not look like the method improves over methods like TCRex or TCRdist, though an advantage is that the parameters may be more interpretable than some black box machine learning approaches. Also for epitope prediction, as noted by the authors, the model may be learning features that differentiate TCRs expressed by CD8+ T cells from the background of all TCRs (which is probably weighted toward CD4+ T cells). This would explain the poorer performance discriminating TCRs specific for one MHC class I epitope from those specified for a different class I epitope. For immunogenicity prediction, evaluations are so dependent on the specifics of the datasets, and the feature itself is so murky, that it's hard to say whether there is a performance advance here.

One nice feature of the diffRBM model is that scores ("single-site factors") can be assigned to individual amino acids in a peptide (or TCR) sequence that captures the contribution of that amino acid at that position to the overall score of the sequence, taking into account the sequence context. The authors show that these single-site factors, for the diffRBM model trained on immunogenic peptides, highlight positions that tend to be involved in TCR contacts as well as specific amino acids, such as "W at position 5", that have been found in previous studies to enhance TCR recognition. The single-site factors for a diffRBM model trained on epitope-specific TCRs appear to do a reasonable job of predicting CDR3 positions that contact the peptide.

Overall, the conclusions of the study are well-supported and the descriptions of the method's performance are balanced. The manuscript is well-written, and the supporting information nicely addresses minor questions that come up in reading the main text. One minor quibble I have is with the description of the method as "unsupervised", especially in the TCR-epitope prediction setting, since the sequences provided to the diffRBM for training, and which the model is tasked with learning differences between, is exactly the positive and negative sequences for the AUROC calculations (up to train/test sampling). It is also confusing to me that the overall selection factors for TCR-epitope binding are so very modest (0.19 for Flu M158, for example; Figure S20D, this is the "effective fraction of sequences retained in selected data compared to background ones"). This doesn't seem like it can be correct, given how focused some of these epitope-specific repertoires are. Overall, though, the study and associated software tools are likely to be useful contributions to the field.

Reviewer #2 (Public Review):

The work by Bravi et al. introduces a learning technique based on Restricted Boltzmann machines, that uses analog to differential learning to model two distinct datasets being part of a common biophysical framework but that behave differently depending on a set of parameters with "background" and "select" features. The biological problem tackled by the authors is the prediction of immunogenetic peptides versus non-immunogenetic ones, as well as determining the sequence features related to binding recognition.

My assessment of the strengths and weaknesses of this work is the following:

Strengths

The authors propose a novel and technically robust solution to a significant and currently unsolved problem in molecular immunology. They are detailed and exhaustive in the description of the formulation of their model as well as in the assessment analysis. Being this a hard problem, the results presented seem a very important step forward not only to solve some of these problems but also to provide convincing arguments that this methodology is more general than other previous approaches; that it can be applied to both immunogenicity prediction as well as binding specificity and is of generative nature. This can have a significant use in therapeutic applications. Another strength of this work is that their methodology could be easily applicable to other biological problems that deal with general versus selected features. For instance, specificity in recognition of other protein-protein interactions, protein-RNA recognition as well as the analysis of ever-growing SELEX and in vitro evolution datasets. Finally, I thought that the efforts of this work to provide "interpretable" learning models are important and definitely a strength of this work.

Weaknesses

As stated before, this work is detailed in nature and contains technical details to make it reproducible. However, in the attempt of the authors to compare against the large number of alternative approaches to this model, I felt that the readability of the article is affected. If this article is meant to be read by broader audiences that might utilize this framework in immunology research, at points the manuscript feels lost in comparison and descriptions of other methods. This is due to the fact that every time a new technical method is introduced, readers want to know about a comparison with other methods, but I feel that the manuscript can be rewritten in such a way that those technical comparisons don't become the major point of the paper and focuses more on how the predictive results of the model can be then applied in immunology. A similar point can also be raised about the methods section, although it has the advantage of being exhaustive and detailed, it also makes it hard for the reader to focus on the most important parts of the work. Perhaps, a better distribution of the methods and SI methods could help streamline the readability of this interesting work.

Reviewer #3 (Public Review):

The authors present in great detail a novel transfer of learning AI model architecture called diffRBM, which is based on the original RBM papers [Hinton, 2002, Hinton and Salakhutdinov, 2006]. They further show how this tool can be used to assess the immunogenicity of TCR positions and the importance of different by-position amino acid usages in creating this immunogenicity. They show that this novel method identifies all known important positions at least as well as existing analytical and structural methods, potentially in a more explanatory way.

Reviewer #1 (Public Review):

CD73 is a promising biomarker in cancer and has been characterized as having an immunosuppressive role in the tumor microenvironment. However, many cancer cell-intrinsic roles of CD73 are still under investigation. In this work, the authors explore the immune-independent roles of CD73 in cancer and demonstrate a function in maintaining metabolic fitness in cancer cells. The authors utilize genetic and pharmacological inhibition of CD73 to characterize metabolic changes in a panel of cancer cell lines and assess tumor growth in vivo. Furthermore, the authors demonstrate that the impaired metabolic fitness due to CD73 inhibition rendered cancer cells more susceptible to DNA-damaging agents. Overall, this work demonstrates the new roles of CD73 in cancer and provides a rationale for combination therapies including CD73 inhibition.

Reviewer #2 (Public Review):

This manuscript describes the involvement of CD73 in tumor cell metabolism by inhibiting CD73 expression in a CD73-positive tumor cell line. The authors demonstrated that CD73 deletion decreases aspartate synthesis via the alteration of mitochondrial respiration. The study is well-designed and the data are convincing.

Reviewer #1 (Public Review):

This important study by Bonnet et al addresses the question of how AMPA receptor numbers at the synapse are regulated during basal conditions and during chemically induced Long Term Potentiation (cLTP). Specifically, the study aims to determine which molecular mechanisms contribute to export from Golgi/the ER, intracellular trafficking of AMPA receptors, and insertion into the synaptic plasma membrane, respectively. The authors had previously established an approach to separately measure these distinct events: to enable a high-fidelity measurement of the Golgi/ER release and subsequent speed of GluA1-containing vesicles, the release of vesicles is synchronized. Finally, the insertion into the plasma membrane is measured by immunolabelling.<br /> The authors set out to specifically understand the contributions of two auxiliary proteins in AMPA receptor expression: 4.1N and SAP97. Namely, the authors find that under basal conditions, binding of SAP97 to GluA1 is necessary for the GluA1 release from the Golgi/ER and intracellular trafficking. In turn, binding of 4.1N to GluA1 is necessary for the exocytosis of the receptor at the plasma membrane at basal conditions. Following induction of cLTP, the authors find that the role of SAP97 remains similar to that observed under basal conditions but, interestingly, 4.1N significantly grows in influence and is required for all stages of GluA1 expression - from release from the Golgi/ER to exocytosis and insertion into the plasma membrane.

In summary, using convincing methodology, the authors are able to dissect the distinct roles of two proteins that bind to the C-terminal domain of the AMPA receptor subunit GluA1: 4.1N and SAP97.<br /> The scientific rigor is high in this work. For example, the question of whether the expression of GluA1 depends on physical interaction with 4.1N and/or SAP97 is nicely addressed by several, well-considered experiments. Overall, the authors' claims are well justified by the data presented.

I did not find any major scientific weaknesses in this manuscript. The approach developed by the group appears to be a good tool for studying the molecular choreography at the synapse under different conditions and the results will be of interest to a wide range of neuroscientists.

Reviewer #2 (Public Review):

The study of Bonnet et al. focuses on how proteins 4.1N and SAP97 affect intracellular trafficking (IT) and externalisation of AMPA receptors (AMPARs) in cultured rat hippocampal neurons. To specifically look at IT, the authors combine the so-called Ariad approach with confocal spinning disc microscopy and photobleaching of dendritic regions, developed in their previous paper (Hangen et al., 2018). This allowed them to synchronously release newly synthesized AMPARs from the ER (upon addition of a synthetic ligand) and measure the number of vesicles carrying AMPARs, their velocity as well as time spent moving and pausing. To detect the insertion of AMPARs at the plasma membrane, live immunolabelling was used. Using RNA-based knock-outs of 4.1N and SAP97 proteins as well as mutants of the AMPAR C-terminus which mediates interactions with these two proteins, in basal conditions and during chemically induced long-term potentiation (cLTP), they could show that the two proteins play different roles in AMPAR trafficking, with SAP97 more profoundly affecting IT compared to 4.1N in basal conditions.

The unique approach allowing observation of IT of AMPARs and a series of tested mutants in basal and cLTP conditions are the main strengths of the paper and also result in the main new finding which is differential regulation of AMPAR IT by 4.1N and SAP97. The measurements of IT parameters and externalisation of unmodified AMPARs across different conditions (and the previous publication) are very reproducible and that makes the whole approach very reassuring.

However, a few points regarding the methodology and analysis remained after reading the manuscript:<br /> Due to the tested mutants, I find the data for the 4.1N-AMPAR interaction particularly strong, but less so for SAP97. For SAP97, sh-RNA experiments are performed and the delta7 mutant is tested. In the case of 4.1N, sh-RNA knockouts were found to be affected by interactions other than AMPAR-4.1N, so the same might be the case for SAP97. Delta4.1N mutant was found to be less reliable than the S816A S818A mutant, in which the AMPAR C-terminus length was retained and 4.1N binding abolished via two mutations. Although only 4 amino acids were removed in the delta7 mutant, this still changes the length of the AMPAR C-terminus. It would be good to acknowledge these aspects of SAP97 experiments.

As there is a number of conditions tested in the paper and to make the conclusions clearer, it might be useful to provide a summary table. It seems to me there are conditions where IT parameters remain unchanged, but no condition where externalisation is not reduced compared to the relevant control condition. Hence, I would agree that 4.1N might be less relevant than SAP97 for IT, but I am not sure it is clear that 4.1N plays a bigger role in externalisation than SAP97, which is what the conclusion figure (Fig. 7) seems to be implying.

Reviewer #3 (Public Review):

This manuscript uses novel techniques to examine the intracellular trafficking and membrane insertion of AMPA receptors to dissect the molecular mechanism involved in regulating these processes in neuronal cultures under basal conditions and during the induction of a chemical form of long-term potentiation (LTP). Specifically, they examine the role of the interaction of the GluA1 subunit with two neuronal proteins SAP97 and 4.1N. The manuscript uses a novel approach to synchronize and temporally control the release of GluA1-containing receptors from the ER and examine its trafficking through the Golgi and dendrites to the plasma membrane. This assay can measure the number of GluA1-containing intracellular vesicles, their speed of trafficking, and the delivery of newly synthesized GluA1 to the surface.

First, the authors use shRNA knockdown (KD) techniques to decrease the expression of SAP97 and 4.1 and found dramatic effects on the number of GluA1-containing vesicles and plasma membrane insertion of GluA1. SAP97 had a larger effect on trafficking while 4.1N had a larger effect on plasma membrane insertion. The authors then went on to use mutants of GluA1 that lack the whole C-terminal domain or mutations in the SAP97 and 4.1N biding sites in GluA1 C-termini and examine the trafficking of these mutants. These mutations decreased the intracellular trafficking and the membrane insertion of GluA1. In addition, the authors mutated phosphorylation sites that have been reported to regulate the interaction of GluA1 with 4.1N. Mutations in these sites that eliminated phosphorylation inhibits membrane insertion while the phosphomimetic mutations did not affect membrane insertion. Finally, mutations in the SAP97 and 4.1N binding sites including mutations in the phosphorylation sites also inhibited chemical-induced LTP increases in the regulation of GluA1 ER-Golgi exit, intracellular transport, and membrane insertion.

These studies are well done and novel and provide support for the role of the GluA1 C-termini and its protein interactors in the trafficking of the AMPA receptor under basal and plasticity conditions. This contributes new data using a novel approach to the controversy over the role of the C-termini of AMPA receptors in the regulation of AMPA receptor function. It supports the role of these interactions in AMPA receptor function.

Reviewer #1 (Public Review):

The authors used viral replication assays to select for and define the resistance pathways against ten developmental Protease Inhibitors (PIs) and their parent drug, Darunavir (DRV), which is one of the leading antiretrovirals used to treat people living with HIV/AIDS. There are two specific regions of the small molecule inhibitors that are actively being modified to increase potency against drug-resistant mutants, the P1' region, and the P2' region, which protrude into pockets of PR occupied by I84 / I50, and a neighboring region containing D29-D30, respectively. Selections using drugs containing small modifications of the P1' region led to primary mutations at PR position I184V, but not I150V. In contrast, selections using drugs containing larger modifications at the P1' region led to primary mutations at PR position I150V, a pathway that is less fit. Furthermore, having modifications at the P2' position added additional potency to the inhibitors, most evident within the I184V pathway. The authors rationalize their findings using previously published structural biology data. These results provide the first evidence for de novo pathway selection using state-of-the-art drugs based on the DRV scaffold and provide an atomic basis for designing compounds that are highly active against DRMs. The comprehensive nature of the analysis of drug resistance to the latest generation PIs, and the insights gained that can be rationalized based on atomic structure, are the major strengths of the paper. The weakness is the lack of commentary on the accessory mutations, which frequently arise in the selections but are not well-explained. It would also be useful to provide some concrete suggestions for minimizing drug resistance using 5th generation PIs, as part of a discussion.

Reviewer #2 (Public Review):

Spielvogel and colleagues report in vitro studies investigating the development of de novo resistance of HIV to Darunavir. Darunavir is one of the most widely used protease inhibitors worldwide, but pathways for the development of de novo resistance are uncertain, as many individuals have had prior protease inhibitor experience prior to treatment with darunavir. As such studies of the kind reported here are essential. The authors have performed foundational studies using compelling and complementary approaches to characterize the emergence of protease drug resistance. They have investigated darunavir, as well as a series of 10 structurally related compounds to provide a clear picture of the role of side chains in the development of resistance. They have complemented these studies with precise structural studies of the interactions of drug with WT and mutant viruses. These data are relevant to the understanding of clinical responses to darunavir and are important in developing new protease inhibitors.

Reviewer #3 (Public Review):

Darunavir (DRV) has been shown to be a potent HIV-1 protease inhibitor in individuals, has pM binding to the protease active site, has activity to protease inhibitor resistant HIV-1s, and has been reported to be difficult to develop resistance to individuals and in tissue culture. The authors argue that given published studies of generating HIV-1 resistance to DRV in tissue culture was not accomplished and all published studies started with either a drug-resistant virus or a combination of drug-resistant viruses for selection, new information can be gleaned as to the viral mutational pathways leading to drug-resistant viruses from HIV-1 wild type (no pre-existing drug mutations) NL4-3.

To better understand the development of HIV-1 wild-type DRV resistance, Spielvogel and colleagues detail their studies on characterizing HIV-1 protease genomic and structural alterations and viral fitness before and during the development of tissue culture resistance to DRV, as well as 10 new compounds (UMass compound series) based on DRV. The UMass compounds have distinct R1 and R2 groups as compared to DRV, which provides for a comprehensive chemical toolset to probe protease genetics and structural changes and alterations in viral fitness resulting during HIV protease drug resistance development in tissue culture. Differences in HIV protease resistance patterns developing over time combined with the potency of the protease inhibitors to HIV mutants resulting from inhibitor selections provide insights as to how DRV chemical groups impact resistance development. The manuscript is comprehensive, well-written, and informative, yet dense and with some figures that readers may not find informative.

Protease inhibitor tissue culture selection of wild-type NL4-3 was based on increasing protease inhibitor concentrations over time. Generally, the DRV resistance mutations that came up early de novo from wild-type NL4-3 virus were, 84V, followed by the acquisition of accessory mutations, predominately 54L and 82I, with 84V, 85V, 46I, 47V, 63P, and others as well, which became entrenched over time. The 84V mutational series have been reported for DRV as the authors noted. To determine the DRV selection pattern from pre-existing HIV single drug-resistant population a pool of 26 single mutant viruses was used for selection. Similar patterns were seen as for wild-type viruses, starting with 84V.

Interestingly, when the UMass compound series was used to select wild-type NL4-3 in tissue culture, 3 mutational series resulted, a protease mutational pattern similar to DRV (UMass 1, and 4, a protease mutational pattern starting with 50V, and followed by the predominate accessory mutations 10F, 13V, 33F, 46I, 63P, and 71V, but not 84V (UMass 3,6,7,8,9, and 10) and a mixture of both populations (UMass 2 and 5). When the HIV single drug-resistant population pool was used, which didn't contain 50V, was used for selection, UMass 2,4,7, and 8 retained the same mutational patterns as the original wild-type HIV selection, where, interestingly, UMass 6 utilized the 84V mutational pathway, rather than 50V, when the 84V mutation was pre-existing.

The results pointed out that modification of the DRV R2 and R1 groups alters selection patterns. It appears that a smaller hydrophobic side chain at the P1' position appears to drive towards 84V selection, whereas a larger side chain selects for the 50V pathway. UMass compounds 2, 5, 7, and 10 demonstrate the highest potency to both 50V/71V and 84V mutant viruses. Interestingly, UMass 2 and 5 were selected for both 50V/71V and 84V resistance mutational pathways, whereas 7 and 10 were selected for 50V/71V pathways.

Based on entry/replication studies, the authors argue that pushing viruses to select 50V/71V mutational pathways in protease, vs 84V mutational pathways in protease, promotes a higher genetic barrier to overcome resistance. This would be due to the reduction in fitness for the 50V/71V protease mutant and the large number of accessory mutants required to regain fitness. However, more in-depth analyses of the various mutants are warranted to support this point, such as head-to-head viral replication studies. A further limitation to the general conclusions is whether mutations in Gag provide for compensatory mutations to augment protease (and viral) fitness for the UMass compound findings.

Reviewer #1 (Public Review):

Thakkar et al describe the immune effects of 3rd and 4th doses of COVID-19 monovalent vaccines in a diverse cohort of immunocompromised cancer patients. They describe augmentation of anti-Spike antibodies after dose 3, especially seroconversion in 57% of patients, followed by a durable response over six months. The fourth dose was associated with increased anti-Spike antibodies in 67% of patients. T-cell responses were seen in 74% and 94% of patients after the third and fourth doses respectively. Strikingly, neutralization of Omicron was absent in all patients after the third dose but increased to 33% after the fourth dose.

Strengths:<br /> Diverse cohort (34% Caucasian, 31% AA, 25% Hispanic 8% Asian) including 106 cancer patients after dose 3, of which 47 patients were longitudinally assessed for six months, as well as eighteen patients assessed after the fourth dose.<br /> Seronegative as well as seropositive patients benefit from a third dose of vaccination.<br /> Assessment of cellular (T cell) immune responses and viral neutralization against wild-type as well as Omicron variant is commendable.

Weaknesses:<br /> The efficacy of the bivalent vaccine (Omicron specific) is not studied here, since the fourth dose of vaccine was a monovalent vaccine. This should be clarified in the discussion.<br /> The authors describe an increase in anti-S titers after monoclonal antibodies. Were any of the patients receiving IVIG, and what was the effect, if any on Anti-S antibodies?<br /> Characteristics of breakthrough infections, particularly if they had prolonged duration, would be important to include.

Reviewer #2 (Public Review):

In this manuscript, Thakkar and colleagues evaluate the immunogenicity of 3rd and 4th doses of SARS-CoV2 vaccinations in patients with cancer. The authors find that additional vaccine doses are able to seroconvert a subset of patients and that antibody levels correlate with T-cell responses and viral neutralization.

The main strengths of this manuscript are:<br /> 1) The authors systemically performed a broad array of immunological assessments, including assessments of antibody levels, T cell activity, and neutralization assays, in a large cohort of patients with cancer receiving 3rd and 4th doses of COVID vaccines.<br /> 2) The authors recruited an ethnically diverse cohort of patients with diverse cancer types, though enrolled participants were enriched for hematological malignancies.<br /> 3) Prior to FDA/CDC guidance supporting a 4th vaccine dose, the authors recruited participants with no or inadequate responses into a prospective clinical trial of a 4th dose, the results of which are outlined here.<br /> 4) The authors' findings that patients with hematologic malignancies and those receiving anti-CD20/BTK inhibitors have lower immunological responses to SARS-CoV-2 vaccines are consistent with multiple prior studies, including prior studies from these authors.<br /> 5) The authors also find that 3rd and 4th COVID vaccine doses are able to seroconvert a subset of patients with no or "inadequate" responses, though it's unclear whether seroconversion is enough for true protection from SARS-CoV-2 infection.

The main weaknesses of the manuscript include:<br /> 1) The study cohorts disproportionately enrolled patients with hematological malignancies who have been previously shown to mount lower immunological responses to COVID-19 vaccines; thus, the findings may not be representative of a typical oncology patient population.<br /> 2) The subgroup analyses were relatively small.<br /> 3) The nomenclature used in the manuscript was confusing when it came to "baseline" assessments and boosters versus additional doses of vaccines.<br /> 4) Ultimately, the major limitation of this manuscript is that antibody levels/T-cell responses/neutralization are surrogates for immune protection against SARS-CoV-2, but it's unclear what defines the ideal cutoffs for protection. Simply seroconverting may still be insufficient. The authors don't provide data showing antibody levels as relates to breakthrough infection, likely because they are underpowered for this analysis.

Reviewer #1 (Public Review):

In this work, Diekmann and Cheng have proposed a new computational model for hippocampal replay. The new model is based on the linear RL work by Piray and Daw 2021, and addresses a fundamental problem in the seminal replay model of Mattar and Daw 2018 (M&D). The new model is based on the default representation, which is a realistic account for state closeness in model-based RL.

This study addresses an important problem in neuroscience at the computational level. The proposed theory is a significant normative computational model that captures important aspects of experimental data in the replay literature. The paper is very well-written (a difficult task for a pure computational work) and figures illustrate the main concepts very well. I have only one question/suggestion:

I believe that there is important data in the literature that cannot be explained by the current model, especially regarding representation of the goal. That is fine; no model is complete, but it is important that authors discuss those caveats in the discussion.

Reviewer #2 (Public Review):

In their paper, Diekmann and Cheng describe a model for the generation of so-called hippocampal replay sequences - a process thought to play a central role in planning, decision making and the consolidation of new memories. Given the diversity of functions replay has been purported to support coming up with a single mechanism for it has remained a challenge. Diekmann and Cheng are able to achieve this with a relatively simple and intuitive model. Specifically, in their model replay is determined based on a finite number of factors; namely, the likelihood and reward-association of an experience, how similar an experience is with an agent's/animal's current state and whether an experience matches *too* much the current state (so to avoid replaying persistently the same state). With these few ingredients the authors are able to replicate important replay findings. Further, the authors emphasise that their model has the significant advantage of being more biologically feasible than other contemporary models in the field.

The model achieves its objectives broadly however the authors have not sufficiently explained the advantage of their model over other models - i.e. how they address the limitations of previous models - nor have they attempted to replicate multiple important features of replay - such as that it can often be non-local. Finally, the details of the biological implementation of their model, particularly with regard to the two modes it can operate in, have not been fleshed out. These points limit the potential impact of the model.

Reviewer #3 (Public Review):

This manuscript provides a remarkably simple, yet effective, model of hippocampal replay. A replay event is stitched together as a chain of reactivated experiences. Individual experiences are prioritized for reactivation according to three intuitive measures: the spatial proximity of an experience to that previously reactivated, the frequency of and reward associated with an experience, and an inhibitory term that propagates the replay across space. Under certain conditions, their model can produce replays that are nearly as optimal--in terms of teaching a reinforcement learning agent to successfully navigate to a reward--as those produced by Mattar and Daw's 2018 model which, by design, generates the most behaviorally useful replays.

The authors assert that their model can recapitulate the replay statistics observed in a subset of experimental works, including the ability of replay to generate novel 'short cuts' from segments of past experience, the resemblance of replay to Brownian diffusion following random exploration, the ability of replay to steer around environmental barriers, and the observation of pre-play. These claims are generally well supported by the data presented (in particular, the model seems to be quite robust to different parameters).

One important caveat is that the proposed model requires two modes ('default' and 'reverse') to simultaneously account for empirical findings and provide behavioral utility (the performance of the agent is poor when using the default mode, but comparable with that of Mattar and Daw in the reverse mode). The authors suggest that the brain could dynamically switch between modes (dubbed the 'dynamic' mode). I feel that the paper would be strengthened by focusing on this dynamic mode throughout and demonstrating that it produces replays with statistics matching empirical data. For example, what is the distribution of forward and reverse replays produced by the default model (figure 3D)? Since neither mode by itself is adequately consistent with experimental findings, showing that the model appropriately switches between modes would strengthen its plausibility.

The authors state that their model is able to recapitulate the finding that replay in sleep following random exploration can be described by Brownian diffusion. A key point in that paper was that the preceding behavior was not diffusive. The authors go some way to address this point by showing that their model produces diffusive replays even if the strength of experience across space is not uniform. However, it isn't clear to me that modeling non-uniform experience strength is equivalent to modeling non-diffusive behaviorally trajectories. A more convincing test would have been to simulate realistic behavioral trajectories and show that subsequent replay events are still diffusive.

In my view, the fact that the model can generate 'pre-play' (in this case, replay of a visually cued, but unvisited arm of the maze) is not particularly informative. In order to generate pre-play, the authors allow the agent to 'visually explore' the cued arm. The implementation of this visual exploration is equivalent to allowing the agent a limited amount of real physical experience on the cued arm. Thus, the finding of replay for the cued arm is unsurprising. It would have been more useful to show that the model over-represents the rewarded arm on a T-maze, given equal exploration of the arms (as in Mattar and Daw).

Also debatable is the authors' assertion that their model is biologically plausible, while that of Mattar and Daw is not. While the former model is certainly computationally less expensive, little experimental data exists that could definitively point to the biological plausibility or implausibility of either model.

Overall, this model is impressive in its ability to generate replay events with realistic and varied statistics, using only a few simple rules. It will be a welcomed addition to the fields of replay, learning and memory, and reinforcement learning.

Reviewer #1 (Public Review):

The manuscript by Zhang et al. titled "Retinal microvascular and neuronal pathologies probed in vivo by adaptive optical two-photon fluorescence microscopy" reports a custom-designed two-photon fluorescence microscope coupled with adaptive optics (AO-2PFM) that allows in vivo imaging of mouse retinal structures at a lateral resolution of ~0.8 μm and axial resolution of ~6.7 μm. The authors provided two examples of applications for in vivo imaging of mouse retinal structure and function. In the first example, AO-2PFM has been used to visualize capillary lesions in a mouse model of retinal angiomatous proliferation (RAP), a form of age-related macular degeneration characterized by capillary proliferation and focal vascular leakage. Using AO-2PFM, the authors observed capillary disruption, with which dye leakage was associated. In the second example, the authors performed in vivo functional imaging of Ca2+ signals in RGCs of the rd1 mouse - a model of retinal degeneration with a mutation in the Pde6B gene. They interpreted the elevated Ca2+ signals in RGCs of rd1 mouse as an indication of RGC hyperactivity that has been reported in ex vivo electrophysiological recordings. They further observed dampened Ca2+ signals in RGCs of rd1 mouse upon retro-orbital injection of lidocaine.

The authors carefully documented the technical features of this state-of-the-art in vivo mouse retina imaging system. The manuscript is very well written and, needless to say, the images presented are of superb quality. There is no doubt that the system will be of great value to many retinal researchers studying the normal structure and function of the retina as well as tracking the pathophysiology of retinal disease models longitudinally.

Reviewer #2 (Public Review):

This is a technical study by Ji and colleagues that uses adaptive optics to correct for the intrinsic aberrations of the mouse eye to improve the quality of in vivo two-photon retinal imaging. Currently, the most common approach to retinal imaging is to use isolated ex vivo retina preparations for direct access to the tissue. However, in vivo retinal imaging offers the unique advantage of tracking long-term changes in vascular/cellular structure and function in disease or development. The authors describe an optimized adaptive optical two-photon microscope setup for imaging fluorescent markers through the mouse eye and evaluate the effect of the wavefront sensing area on the imaging quality. They further demonstrate the power of this setup by monitoring the focal vascular leakage in a mouse model of proliferative vascular retinopathy and by monitoring drug-induced population activity changes using GCaMP6s in a mouse model of photoreceptor degeneration. Together, these results provide a valuable, enabling technical resource for applying AO-two-photo imaging to study outstanding questions in retinal biology that require long-term in vivo imaging. Overall, this is an important development with a broad impact on the investigation of neuronal and vascular functions in the retina.

Reviewer #3 (Public Review):

Zhang, Q. et al. developed a two-photon fluorescence microscope (2PFM) by incorporating direct wavefront sensing adaptive optics (AO), which is optimized for mouse in vivo retinal imaging. By using the same 2PFM with the option of using or not using the incorporated AO system, this team compared the in vivo retinal images and convincingly demonstrated that AO correction acquired brighter and higher resolution images of retinal ganglion cells (RGCs) and their axons in both densely and sparse labeled transgenic mouse lines, normal and defected capillary vasculatures, and RGC spontaneous activities detected by genetic Ca2+ sensor. Interestingly and importantly, this team found that a global correction by removing the common aberration from the entire FOV enhances imaging signals throughout the entire large FOV, indicating a preferable AO imaging strategy for large FOVs. The potential applications of the in vivo retinal imaging techniques and strategies developed by this study will certainly inspire further investigation of the dynamic morphological and functional changes of retinal vasculatures and neurons during disease progression and before and after treatments.

It would be beneficial to the manuscript and the readers if the authors can elaborate on optic design a little bit more. For example, whether the incorporation of AO adversely affects the 2PFM optic design? If the 2PFM can be further optimized by uncompromised optic design without incorporating AO, the quality of in vivo images will comparable to the AO-2PFM or not?

Reviewer #1 (Public Review):

This manuscript investigates the question of how polylysogeny impacts competition with a sensitive non-lysogen, and how this is shaped by phage resistance. This is an important and timely question, as lysogeny can be a strategy to invade new niches, and prophages are important vehicles for the acquisition of a range of virulence factors by pathogens including Klebsiella. The authors use a polylysogenic Klebsiella clone in competition with a non-lysogen that is sensitive to at least some of the prophages produced by the polylysogen. They compete these strains over a 30-day period and measure host population dynamics and evolution of phage resistance and lysogenic conversion in the (initially) sensitive competitor. Overall, the experiment shows that lysogen formation is relatively rare and short-lived. Instead, phage resistance through complete loss of the capsule is the primary mechanism evolving, but other resistant capsule mutants, with more subtle mutations affecting capsule expression, emerge as well. The authors have collected a very impressive amount of data and made some very interesting observations.

My main problem with this paper is that the manuscript lacks a clear narrative, making it very hard to extract the key message this paper wants to convey. Related to this, (some of) the conclusions that the authors make do not appear to be well supported by the data. For example, the authors conclude that selection favours more subtle capsule mutations because they are less costly than capsule-loss mutants (lines 497-500). However, there are no data to support this conclusion, as fitness costs of the various resistance phenotypes analysed were not measured. Apart from the genotypes, the data that are presented in this show that these subtle mutants have more subtle decreases in capsule production compared to the mutants that show a complete loss of capsule. But this does not tell us their relative cost. It also doesn't tell us how the emergence of these different mutants relates to phage pressure, because whilst bacterial population dynamics data are monitored meticulously, phage dynamics data are missing (I have not found them in the supplemental information either). This makes it impossible to directly relate the emergence of the various resistance mechanisms to phage infection pressure during the coevolution experiment, even though this appears to be a hypothesis the authors wish to test.

Overall I think the overarching question of the manuscript is important and the model system is a very relevant one to study this question, but in my view, the current data don't support the conclusions of the paper. Apart from these criticisms, the manuscript is very well written and the figures are overall easy to interpret.

Reviewer #2 (Public Review):

This manuscript presents data on multiple experiments regarding the co-evolution of poly-lysogenic and phage-susceptible Klebsiella pneumoniae strains. In particular, the manuscript aimed to determine the mechanisms of resistance that would shape bacterial competition over co-evolutionary timescales. The major finding is that the potential for lysogenization as a phage resistance mechanism is narrow and only likely to occur given certain circumstances. Moreover, the manuscript again reinforces the importance of receptor changes -initially loss, but modification in structure or expression over longer time scales- as a major mechanism of phage resistance that influences bacterial competition.

Strengths<br /> A major strength of this manuscript is the care in designing experiments and conducting follow-up experiments to isolate the essential elements to support each of the conclusions. This includes using orthogonal methods such as sequencing and modeling to support or expand the findings from culturing and experimental evolution. The study features results that were beautifully replicated (e.g. Figure 3) lending confidence to the findings.

Weaknesses<br /> Two weaknesses of the manuscript in its current form are: 1) a need to discuss other studies that also have found context-dependent results and 2) more focus on delivering the key overall "message" of the paper to the reader. Finally, not a weakness, but a (necessary) limitation is the study system, but this manuscript sets a bar for other groups to test in their systems to probe the generality of the findings.

The support for the conclusions is compelling. The findings were counter to the initial expectation (lysogenization as a major feature) and the manuscript does an admirable job of supporting the unexpected conclusion with thorough experimental work, supplemented with modeling.

This manuscript will be of great significance in microbial evolution, both for its implications in limiting the scope of lysogenization as a viable phage resistance mechanism in the long term and for its significant experimental rigor, particularly with regard to the co-evolutionary timescale studied. The study has very important implications for the evolution of antimicrobial resistance and phage therapy.

Reviewer #1 (Public Review):

Rab27 is a major regulator of insulin granule exocytosis from beta cells, and it acts via (at least) three distinct effector proteins; Granuphilin, Melanophilin and Exophilin-8. Although the role of each of these three Rab-effectors in the regulation of insulin secretion is fairly well-established from studies of KO mice, the functional hierarchy between the effectors remains largely unknown. This study by Zhao et al addresses this question by investigating how simultaneous loss of two these effectors influence insulin granule exocytosis and also provide an explanation for their differential regulation of this process. They propose that Exophilin-8 acts upstream of Melanophilin, which in turn is involved in crash-fusion of granules with the plasma membrane, and that the interaction between these two effectors require the exocyst complex. This mode of exocytosis is relatively rare and only accounts for around 20% of all fusion events. The majority of fusion events instead involves exocytosis of granules stably docked at the plasma membrane. The authors propose that this mode of exocytosis also depends on Exophilin-8, now acting by removal of a Granuphilin-mediated exocytic clamp.

Technically, this is a superb study where the authors use primary mouse islets isolated from both single and double KO mice and perform both bulk secretion assays and single-cell granule imaging to elucidate the role of Rab27 effectors in glucose-stimulated insulin secretion. Unfortunately, while visualization of granule dynamics is performed in living cells, visualization of the Rab27 effectors and the ecoxyst components is restricted to static immunofluorescence imaging. It is therefore difficult to reconcile granule dynamics with effector action. While the results are clearly presented and largely consistent with previous work, I feel that many of the conclusions are based on over-interpretation of data and that important control experiments are missing. The authors are able to confirm their and others' previous observations that each of the three Rab27 effectors have distinct functions during insulin secretion. A connection between insulin granule exocytosis and the Exocyst complex has also been established in previous studies. The most intriguing finding in this study is that the Exocyst complex function in cooperation with Rab27 and its effectors, thus connecting these two pathways, and that there appears to be a functional hierarchy amongst the Rab27 effectors where Exophilin-8 act upstream of the other two. What remains unclear to me is how this entire process is regulated and how it relates to prevailing models of insulin granule pools and modes of exocytosis.

Understanding the mechanism that regulate insulin secretion is imperative for understanding how this process fails in certain types of diabetes. This study reinforces the concept that the secretion of insulin granules is a very heterogenous process that involves multiple pools of granules and modes of exocytosis and provides important new information on how cross-talk between these pathways help to shape the secretory response and give it robustness.

Reviewer #2 (Public Review):

Insulin exocytosis is a tightly orchestrated process that involves proteins acting in complexes near the plasma membrane. The authors have contributed much of the field's knowledge on how exophilin anchors insulin granules in cortical actin and works with other effectors to prepare granules for exocytosis. Here they find that, while both exophilin and melanophilin localize to the exocyst, functionally they are not equivalent. TIRF imaging of monolayer dispersed beta cells, although a non-physiologic model to study islet cell secretion (which requires homotypic and heterotypic cell coupling), is nonetheless an established method that the authors have used with expert proficiency. The imaging and quantitative methods here should be broadly applicable to those studying secretory events at cellular resolution, and practical details e.g. the need for double transfection in RNAi experiments, are helpful and appreciated.

Reviewer #3 (Public Review):

In this manuscript Zhao et al investigated how multiple Rab27 effectors work to regulate insulin secretion by murine pancreatic b-cells. They do this by comparing the phenotypes of b-cells/islets lacking effectors doubly or singly. Their main findings/contributions are that:

Mlph works downstream of Myrip/exophilin-8 to mobilise granules for fusion from the actin network to the plasma membrane after stimulation.

Mlph and exophilin-8 interact via the exocyst

Down-regulation of exocyst affects exocytosis in cells expressing exophilin-8

Exophilin-8 promotes fusion of granules docked by granuphilin at the membrane

Exophilin-8 not required for Grph related granule docking at the plasma membrane

A model for how the three effectors coordinate ISG secretion. According to this model there are 2 insulin secretion pathways in b-cells; a) where Exo8 acts upstream of Mlph and with actin/Myosin Va/VIIa, exocyst and syntaxin 4 to move dock granules in actin and promote exocytosis, and b) where Exo8 works in an antagonistic manner with Grph promoting secretion of granules docked at the membrane by Grph.

This is an interesting/important question and the authors make important contributions (above). In general experiments are well designed and controlled but there are some questions that remain open that could have been included to make the study a more comprehensive analysis of Rab27 effectors in insulin secretion.

Reviewer #1 (Public Review):

Decidualization, denoting the transformation of endometrial stromal cells into specialized decidual cells, is a prerequisite for normal embryo implantation and successful pregnancy in humans. Abnormal cytokine-associated inflammation during decidualization can alter the endometrium's receptivity to healthy embryo implantation. Jiang and colleagues present an important analysis of the role and function of the Gaq axis on the inflammatory response during decidualization essential for early pregnancy, and present preliminary data on its clinical relevance.

The data narrative provides solid evidence of the mechanisms suggested by Jiang and colleagues. The study is highlighted both by the in vitro analysis and also by the study of human samples and subjects impacted by Recurrent Pregnancy Loss (RPL). Overall, the data seems to justify the conclusions taken, although some of the methodology and data interpretation require further clarification and justification.

Reviewer #2 (Public Review):

This manuscript provided evidence that Gaq is a key regulator of the expression of inflammatory cytokines to maintain the proper progress of decidualization of human endometrial stromal cells for successful implantation and pregnancy. The major strength of the manuscript is the experimental design to answer sequential scientific questions regarding the function of Gaq during decidualization in the human endometrium using various molecular and pharmacologic tools. A weak point of this manuscript is that the author did not provide a reason to focus on HDAC5 among various downstream targets for the study of Gaq. In addition, if the authors make a knockout mouse of Gaq and characterize its phenotypes to support what they found in human stromal cells, the findings in this manuscript could become a piece of compelling evidence for the importance of Gaq during decidualization in the human endometrium for a successful pregnancy. This could be the next scientific topic for the authors to pursue this project.

Reviewer #1 (Public Review):

The expression and localization of Foxc2 strongly suggest that its role is mainly confined to As undifferentiated spermatogonia (uSPGs). Lineage tracing demonstrated that all germ cells were derived from the FOXC2+ uSPGs. Specific ablation of the FOXC2+ uSPGs led to the depletion of all uSPG populations. Full spermatogenesis can be achieved through the transplantation of Foxc2+ uSPGs. Male germ cell-specific ablation of Foxc2 caused Sertoli-only testes in mice. CUT&Tag sequencing revealed that FOXC2 regulates the factors that inhibit the mitotic cell cycle, consistent with its potential role in maintaining a quiescent state in As spermatogonia. These data made the authors conclude that the FOXC2+ uSPG may be the true SSCs, essential for maintaining spermatogenesis. The conclusion is largely supported by the data presented, but two concerns should be addressed: 1) terminology used is confusing: primitive SSCs, primitive uSPGs, transit amplifying SSCs... 2) the GFP+ cells used for germ cell transplantation should be better controlled using THY1+ cells.

Reviewer #2 (Public Review):

The authors found FOXC2 is mainly expressed in As of mouse undifferentiated spermatogonia (uSPG). About 60% of As uSPG were FOXC2+ MKI67-, indicating that FOXC2 uSPG were quiescent. Similar spermatogonia (ZBTB16+ FOXC2+ MKI67-) were also found in human testis.

The lineage tracing experiment using Foxc2CRE/+;R26T/Gf/f mice demonstrated that all germ cells were derived from the FOXC2+ uSPG. Furthermore, specific ablation of the FOXC2+ uSPGs using Foxc2Cre/+;R26DTA/+ mice resulted in the depletion of all uSPG population. In the regenerative condition created by busulfan injection, all FOXC2+ uSPG survived and began to proliferate at around 30 days after busulfan injection. The survived FOXC2+ uSPGs generated all germ cells eventually. To examine the role of FOXC2 in the adult testis, spermatogenesis of Foxc2f/-;Ddx4-cre mice was analyzed. From a 2-month-old, the degenerative seminiferous tubules were increased and became Sertoli cell-only seminiferous tubules, indicating FOXC2 is required to maintain normal spermatogenesis in adult testes. To get insight into the role of FOXC2 in the uSPG, CUT&Tag sequencing was performed in sorted FOXC2+ uSPG from Foxc2CRE/+;R26T/Gf/f mice 3 days after TAM diet feeding. The results showed some unique biological processes, including negative regulation of the mitotic cell cycle, were enriched, suggesting the FOXC2 maintains a quiescent state in spermatogonia.

Lineage tracing experiments using transgenic mice of the TAM-inducing system was well-designed and demonstrated interesting results. Based on all data presented, the authors concluded that the FOXC2+ uSPG are primitive SSCs, an indispensable subpopulation to maintain adult spermatogenesis.

The conclusion of the mouse study is mostly supported by the data presented, but to accept some of the authors' claims needs additional information and explanation. Several terminologies define cell populations used in the paper may mislead readers.

1) "primitive spermatogonial stem cell (SSC)" is confusing. SSCs are considered the most immature subpopulation of uSPG. Thus, primitive uSPGs are likely SSCs. The naming, primitive SSCs, and transit-amplifying SSCs (Fig. 7K) are weird. In general, the transit-amplifying cell is progenitor, not stem cell. In human and even mouse, there are several models for the classification of uSPG and SSCs, such as reserved stem cells and active stem cells. The area is highly controversial. The authors' definition of stem cells and progenitor cells should be clarified rigorously and should compare to existing models.

2) scRNA seq data analysis and an image of FOXC2+ ZBTB16+ MKI67- cells by fluorescent immunohistochemistry are not sufficient to conclude that they are human primitive SSCs as described in the Abstract. The identity of human SSCs is controversial. Although Adark spermatogonia are a candidate population of human SSCs, the molecular profile of the Adark spermatogonia seems to be heterogeneous. None of the molecular profiles was defined by a specific cell cycle phase. Thus, more rigorous analysis is required to demonstrate the identity of FOXC2+ ZBTB16+ MKI67- cells and Adark spermatogonia.

3) FACS-sorted GFP+ cells and MACS-THY1 cells were used for functional transplantation assay to evaluate SSC activity. In general, the purity of MACS is significantly lower than that of FACS. Therefore, FACS-sorted THY1 cells must be used for the comparative analysis. As uSPGs in adult testes express THY1, the percentage of GFP+ cells in THY1+ cells determined by flow cytometry is important information to support the transplantation data.

4) The lineage tracing experiments of FOXC2+-SSCs in Foxc2CRE/+;R26T/Gf/f showed ~95% of spermatogenic cells and 100% progeny were derived from the FOXC2+ (GFP+) spermatogonia (Fig. 2I, J) at month 4 post-TAM induction, although FOXC2+ uSPG were quiescent and a very small subpopulation (~ 60% of As, ~0.03% in all cells). This means that 40% of As spermatogonia and most of Apr/Aal spermatogonia, which were FOXC2 negative, did not contribute to spermatogenesis at all eventually. This is a striking result. There is a possibility that FOXC2CRE expresses more widely in the uSPG population although immunohistochemistry could not detect them.

5) The CUT&Tag_FOXC2 analysis on the FACS-sorted FOXC2+ showed functional enrichment in biological processes such as DNA repair and mitotic cell cycle regulation (Fig.7D). The cells sorted were induced Cre recombinase expression by TAM diet and cut the tdTomato cassette out. DNA repair process and negative regulation of the mitotic cell cycle could be induced by the Cre/lox recombination process. The cells analyzed were not FOXC2+ uSPG in a normal physiological state.

6) Wei et al (Stem Cells Dev 27, 624-636) have published that FOXC2 is expressed predominately in As and Apr spermatogonia and requires self-renewal of mouse SSCs; however, the authors did not mention this study in Introduction, but referred shortly this at the end of Discussion. Their finding should be referred to and evaluated in advance in the Introduction.

Reviewer #3 (Public Review):

By popular single-cell RNA-seq, the authors identified FOXC2 as an undifferentiated spermatogonia-specific expressed gene. The FOXC2+-SSCs can sufficiently initiate and sustain spermatogenesis, the ablation of this subgroup results in the depletion of the uSPG pool. The authors provide further evidence to show that this gene is essential for SSCs maintenance by negatively regulating the cell cycle in adult mice, thus well-established FOXC2 as a key regulator of SSCs quiescent state.

The experiments are well-designed and conducted, the overall conclusions are convincing. This work will be of interest to stem cell and reproductive biologists.

Reviewer #1 (Public Review):

The fields of ancient and environmental DNA have many similarities. Practitioners are constantly tinkering with methods to extract as much information from biological samples as possible. Both fields of research also have to deal with the fact that only a tiny fraction of the DNA is 'on target' and that the background DNA (largely bacterial) is often immense.

In this research Urban et al tackle the question of individual identification of a flightless New Zealand parrot (the kakapo) using shotgun eDNA (from soil) within a study system where reference genomes exist for most of the animals within a population. Most eDNA studies stay in the relative safety of metabarcoding (typically on mitochondrial DNA) thus Urban et al are breaking new ground.

In this small-scale (and highly controlled) study, Urban et al. use shotgun eDNA from a gram of soil and then match kakapo reads to reference genomes. Using some innovative Bayesian inference the researchers are able to identify individuals within the populations.

There are a number of innovations in this study that have relevance to the conservation sector. The idea that we can identify individuals in a population in a non-invasive manner is an exciting prospect. It immediately conjures up the possibility of genetic mark-recapture applications. In the case of highly endangered populations, the work shows the value of building reference genomes for the whole population.

At its core, this is a proof-of-principal study that arguably leaves the reader with more questions than answers. I was left wondering (i) why didn't nanopore's adaptive sampling function enrich targets? (ii) how would short-read platforms compare (iii) could genomic signatures of other taxa (e.g. bats) identified by metabarcoding be detected in shotgun data? And (iv) is sediment the best substrate for this work?

Sedimentary DNA methods have been around for ~20 years and it is exciting to see the field continue to innovate. The speed and portability of nanopore devices may, with time, see real-time genotyping become a reality in conservation biology. I welcome these innovations as, on the global stage, we need all the tools we can get to battle the biodiversity crisis.

Reviewer #2 (Public Review):

This study uses DNA metabarcoding to identify vertebrates and kākāpō DNA in soils from sites where they are known to occur and from control sites housing related birds. The authors then attempt to identify individual kākāpō birds that have contributed DNA into just three samples with high kākāpō DNA content. For this, they use Oxford Nanopore adaptive sequencing, haplotype identification, and two statistical approaches to determine the number of individuals that contributed to a sample and which specific individuals contributed. This study builds on recent developments in the field that move eDNA into population genomics and individual surveillance.

The manuscript introduction does a satisfactory job of contextualizing the need for this study and the state of the field. It does not detail the challenges of applying adaptive ONT to eDNA samples and the kinds of choices such as selective assays available. I think the authors are using confusing language in the abstract and throughout that is not clear enough to be useful to a reader community that is interested in adopting ONT but not already using it.

As for the methods chosen for this study, I found it peculiar that the authors did not use qPCR specific to kākāpō to estimate the relative proportion of kākāpō eDNA to other vertebrate DNA in the total sample. A fair comparison of methods would make this study more useful to guide the field forward. qPCR should be more sensitive than metabarcoding and is the standard approach for the relative abundance that the terrestrial eDNA community uses for targeted studies.

There is a lot of work done in this study that would be useful to the eDNA community if it were presented clearly. Paragraphs are written often without topic sentences, headings are vague, specific objectives are not clearly outlined, and too many questions remain about why certain approaches were used. For example, there is a selective and non-selective approach used for ONT sequencing. In some places, is not clear what exactly the authors did, and it's not clear why the non-selective approach was preferred by the authors (as they describe in the discussion). The ONT portion of the methods seems written out of order and with frivolous choices about what details to include and omit. No mention of the pore destruction of selective/adaptive sequencing is described, so this study creates hyperbole about the promise of ONT unblocking pores for future research. There are drawbacks! Further, there surely is going to be a lot of interest in the statistical approaches to infer individuals and the number of individuals that shed DNA into a sample but this is not clearly explained. An effort to improve the writing quality throughout is needed prior to publication.

The study fails to describe the scale of the sites and how they are managed. As such, we cannot assess the distance from the site and why kākāpō DNA was found at an abandoned nest site. Maybe it was clear but the names of the sites are inconsistent throughout the ms, and there are assumptions that readers know about this field setting already, which is not a good assumption to make.

The discussion cites nobody and does not put the results back into the broader context of where the science is today. It is a weak discussion that just reiterates the results, but then boasts about the significance of the results when those results referred to were insufficiently described in the manuscript.

Altogether, I think this study has potential if the paper can be improved in clarity and quality. The science is solid and the topic is of great interest to a broad community.

Reviewer #1 (Public Review):

Medwig-Kinney et al perform the latest in a series of studies unraveling the genetic and physical mechanisms involved in the formation of C. elegans gonad. They have paid particular attention to how two different cell fates are specified, the ventral uterine (VU) or anchor cell (AC), and the behaviors of these two cell types. This cell fate choice is interesting because the anchor cell performs an invasive migration through a basement membrane. A process that is required for correct C. elegans gonad formation and that can act as a model for other invasive processes, such as malignant cancer progression. The authors have identified a range of genes that are involved in the AC/VC fate choice, and that imparts the AC cell with its ability to arrest the cell cycle and perform an invasive migration. Taking advantage of a range of genetic tools, the authors show that the transcription factor NHR-63 is strongly expressed in the AC cell. The authors also present evidence that NHR-63 is could function as a transcriptional repressor through interactions with a Groucho and also a TCF homolog, and they also suggest that these proteins are forming repressive condensates through phase separation.

The authors have produced an extensive dataset to support their two primary claims: that NHR-67 expression levels determine whether a cell is invasive or proliferative, and also that NHR-67 forms a repressive complex through interactions with other proteins. The authors should be commended for clearly and honestly conveying what is already known in this area of study with exhaustive references. But absent data unambiguously linking the formation and dissolution of NHR-67 condensates with the activation of downstream genes that NHR-67 is actively repressing, the novelty of these findings is limited.

Reviewer #2 (Public Review):

Medwig-Kinney et al. explore the role of the transcription factor NHR-67 in distinguishing between AC and VU cell identity in the C. elegans gonad. NHR-67 is expressed at high levels in AC cells where it induces G1 arrest, a requirement for the AC fate invasion program (Matus et al., 2015). NHR-67 is also present at low levels in the non-invasive VU cells and, in this new study, the authors suggest a role for this residual NHR-67 in maintaining VU cell fate. What this new role entails, however, is not clear. The model in Figure 7E shows NHR-67 switching from a transcriptional activator in ACs to a transcriptional repressor in VUs by virtue of recruiting translational repressors. In this model, NHR-67 actively suppresses AC differentiation in VU cells by binding to its normal targets and acting as a repressor rather than an activator. Elsewhere in the text, however, the authors suggest that NHR-67 is "post-translationally sequestered" (line 450) in nuclear condensates in VU cells. In that model, the low levels of NHR-67 in VU cells are not functional because inactivated by sequestration in condensates away from DNA. Neither model is fully supported by the data, which may explain why the authors seem to imply both possibilities. This uncertainty is confusing and prevents the paper from arriving at a compelling conclusion. What is the function, if any, of NHR-67 and so-called "repressive condensates" in VU cells?

Below we list problems with data interpretation and key missing experiments:

1) The authors report that NHR-67 forms "repressive condensates" (aka. puncta) in the nuclei of VU cells and imply that these condensates prevent VU cells from becoming ACs. Fig. 3A, however, shows an example of an AC that also assemble NHR-67 puncta (these are less obvious simply due to the higher levels of NHR-67 in ACs). The presence of NHR-67 puncta in the AC seems to directly contradict the author's assumption that the puncta repress the AC fate program. Similarly, Figure 5-figure supplement 1A shows that UNC-37 and LSY-22 also form puncta in ACs. The authors need to analyze both AC and VU cells to demonstrate that NHR-67 puncta only form in VUs, as implied by their model.

2) While a pool of NHR-67 localizes to "repressive condensates", it appears that a substantial portion of NHR-67 also exists diffusively in the nucleoplasm. This would appear to contradict a "sequestration model" since, for such a model to work, a majority of NHR-67 should be in puncta. What proportion of NHR-67 is in puncta? Is the concentration of NHR-67 in the nucleoplasm lower in VUs compared to ACs and does this depend on the puncta?

3) The authors do not report whether NHR-67, UNC-37, LSY-22, or POP-1 localization to puncta is interdependent, as implied in the model shown in Fig. 7.

4) The evidence that the "repressor condensates" suppress AC fate in VUs is presented in Fig. 4D where the authors deplete the presumed repressor LSY-22. First, the authors do not examine whether NHR-67 forms puncta under these conditions. Second, the authors rely on a single marker (cdh-3p::mCherry::moeABD) to score AC fate: this marker shows weak expression in cells flanking one bright cell (presumably the AC) which the authors interpret as a VU AC transformation. The authors, however, do not identify the cells that express the marker by lineage analyses and dismiss the possibility that the marker-positive cells could arise from the division of an AC-committed cell. Finally, the authors did not test whether marker expression was dependent on NHR-67, as predicted by the model shown in Fig. 7.

5) Interaction between NHR-67 and UNC-37 is shown using Y2H, but not verified in vivo. Furthermore, the functional significance of the NHR-67/UNC-37 interaction is not tested.

6) Throughout the manuscript, the authors do not use lineage analysis to confirm fate transformation as is the standard in the field. There are 4 multipotential gonadal cells with the potential to differentiate into VUs or ACs. Which ones contribute to the extra ACs in the different genetic backgrounds examined was not determined, which complicates interpretation. The authors should consider and test the following possibilities: disruption of NHR-67 regulation causes 1) extra pluripotent cells to directly become ACs early in development, 2) causes VU cells to gradually trans-fate to an AC-like fate after VU fate specification (as implied by the authors), or 3) causes an AC to undergo extra cell division(s)?? In Fig. 1F, 5 cells are designated as ACs, which is one more that the 4 precursors depicted in Fig. 1A, implying that some of the "ACs" were derived from progenitors that divided.

In conclusion, while the authors report on interesting observations, in particular the co-localization of NHR-67 with UNC-37/Groucho and POP-1 in nuclear puncta, the functional significance of these observations remains unclear. The authors have not demonstrated that the "repressive condensates" are functional and play a role in the suppression of AC fate in VU cells as claimed. The colocalization data suggest that NHR-67 interacts with repressors, but additional experiments are needed to demonstrate that these interactions are specific to VUs, impact VU fate, and sequester NHR-67 from its targets or transform NHR-67 into a transcriptional repressor.

Reviewer #1 (Public Review):

Vaparanta et al propose a new bioinformatic algorithm for pathway discovery from multi-omics data sources at one time point, and validate some of their algorithm's predictions using functional experiments. The authors should be commended for their detailed experimental work and comprehensive data collection around TYRO3 signaling in melanoma, which will likely be of value to that field. They also provide a mature software package that is well documented for implementing their bioinformatic methods. The reviewer's experience with the software was that it is computationally efficient/fast with well written code. The biological data (both multiomics and functional validation studies) will be of interest to melanoma research as well as scientists interested in TYRO3 signaling.

At this time, however, the bioinformatics algorithm proposed is of unclear utility to the broader multiomics community for the following reasons:

First, the algorithm itself has numerous hyperparameters, which can make it challenging to use and potentially highly sensitive to these user inputs. Just the regulatory complex inference step has 10 hyperparameters/settings required to be selected.

Second, the algorithm is presented in an ad hoc manner without mathematical/statistical justifications of the many design decisions and steps in the analysis. For example, the authors write "The inference of regulatory complexes from the combined score follows the nearest neighbor principle, assuming that while a single high combined score can be random chance, the combination of combined scores between 3 cell signaling molecules would be predictive". It is mathematically unclear that this is true, and thus this reviewer attempted to test the algorithm using simulated uncorrelated Gaussian noise (see code/outputs at end of the review) in 10K genes and 10 samples using a best attempt at hyperparameter selection per the code comments and documentation. It appears that nearly 1/3 of all genes (i.e., 3205 of 10K) were erroneously grouped into complexes (assuming no mistakes in reviewer's usage of the code). In general, "unbiased" pathway analysis in multiomics that is not relying on prior knowledge will require solving the extraordinarily challenging task of estimating a very large covariance matrix from statistically small sample sizes. This puts the method at high risk of producing spurious results.

Third, pathway analysis has long been a bioinformatic goal in the literature, with the authors citing a landmark paper for the WGCNA method from 2008. As such, there are numerous and long-standing discussions in the literature regarding challenges of pathway analysis (i.e., omics data often has dimensionality D far larger than sample size N, and correlation matrix estimation requires D^2 >> N parameters to be estimated) and its potential for spurious correlations. Some authors use sophisticated statistical tools (e.g., "Biological network inference using low order partial correlation" 2014, "Learning Large‐Scale Graphical Gaussian Models from Genomic Data" 2005, "Incorporating prior knowledge into Gene Network Study" 2013) to attempt to deal with this issue. Furthermore, the authors indicate that their approach is the first to attempt pathway analysis in multi-omics setting, stating "Integrative approaches combining more than one robust molecular association measure, however, have not been explored", but one can find attempts such as "An Integrative Transcriptomic and Metabolomic Study of Lung Function in Children With Asthma" to build on WGCNA for work in multiomics datasets. The 2020 review paper "Metabolomics and Multi-Omics Integration: A Survey of Computational Methods and Resources" seems to identify multiple published methods dealing with pathway estimation in multiomics datasets. As the paper stands, this reviewer cannot adequately assess the impact of the proposed bioinformatic algorithm and its results against the existing body of literature for pathway inference.

Reviewer #2 (Public Review):

The authors describe a bioinformatic platform that allows for unbiased pathway analysis from multiomics data. The concept is based on correlation, stoichiometry scores and their combination to evidence interaction between two proteins, transcripts or phosphosites in an omic dataset. This platform was developed and validated on both previously published and in house omics data. I really appreciate that the paper is well written and clear, and I would like to acknowledge the amount of work generated to produce the in-house dataset.

Reviewer #1 (Public Review):

Siegfried et al. study a very interesting and timely topic in cell biology: the connection between ER-PM membrane contact sites (MCS) and cell migration. In brief, the authors use the polarized epithelial model cell line (CACO-2) to study this process. They routinely compare parental cells (Control) with a clonal CACO-2 cell line knocked out (KO) for the ER tether protein VAP-A. They convincingly show that KO cells move faster but in a less directional manner, leading to slower monolayer migration. Interestingly, they showed that KO cells have larger focal adhesions (FAs), a phenotype that was reverted upon expression of the wild-type VAP-A but not of a VAP-A mutant (VAP-A-KDMD, mutation in the MSP domain) defective in binding to FFAT-containing partner proteins. Some observations regarding the role of VAP-A's MSP domain on the regulation of the actin cytoskeleton, although the evidence for this was incomplete. Furthermore, VAP-A depletion was shown to have an impact on PI(4,5)P2 levels at the PM (but not on PI(4)P levels at the Golgi membranes), to stabilize the dynamics of ER-PM MCS, and to increase FA lifetime by decreasing FA disassembly rate. Finally, they showed that there is a correlation between the appearance of ER-PM MCS at FAs with FA disassembly, however, how VAP-A plays a role in this effect is unclear. The authors put their findings in the context of the literature in the field to propose a working hypothesis by which VAP-A at ER-PM MCS could impact FA dynamics and cell motility.

Reviewer #2 (Public Review):

In this study, the authors assessed the role of the ER protein VAPA in cell migration and regulation of focal adhesions dynamics. The authors used CRISPR/Cas9 knock-out of VAPA in Caco-2 cells. They demonstrate that VAPA KO cells have slower migration capacity which is linked to a slower FA disassembly rate. Interestingly, the VAPA KO cells don't show any defects of PI4P level at endosomes nor at the Golgi complex but have a decreased PI4,5P2 level, probably linked to the redundant function of VAPB at endosomes and Golgi while VAPA might be solely responsible for effects on migration.

The results provided by the authors support their conclusions. The experiments performed are well carried out. The VAPA KO cells used in this study are originating from a clonal population but the authors used rescue experiments expressing the VAPA wild-type of the KDMD mutant to demonstrate the role of VAPA in the phenotype. In addition, appropriate and careful quantifications are provided with the different experiments, strengthening the conclusions. The data provided in this manuscript suggest a role for the ER-resident membrane contact protein VAPA in cell migration potentially independent of lipid homeostasis.

Reviewer #3 (Public Review):

In this manuscript, the authors examine the role of VAPA in focal adhesion (FA) turnover and cell motility via effects on ER-PM contact site functions. The authors show that VAPA KO CaCo2 cells form larger FA and have aberrant migration behavior and spreading. Those cells show lower levels of PI(4,5)P2 at PM, but no change in PI(4)P at Golgi and endosomes. PI(4)P is not tested at the PM. The authors show that VAPA KO cells have a similar number but less stable GFP-MAPPER positive ER-PM contact sites as compared to control cells. In contrast, FA are more stable over time in VAPA KO. The authors also aimed to evaluate GFP-MAPPER proximity with vinculin spots and concluded that ER-PM contacts partially overlap with FA, whereas they are more distant in VAPA KO. Thus, a correlation between stable contact sites near FA and FA disassembly likely exists. From this set of data, the authors suggest that VAPA has a key role at ER-PM contacts near FA by mediating lipid transfer, which ultimately enables internalization of integrins and FA disassembly.

The approach in the paper is innovative and interesting because VAPA is a major tether at contact sites and the link between contact sites and cytoskeleton dynamics and cell motility remains little explored. This can potentially lead to significant advances in the field. The experiments presented are technically well executed, but most of the results and hypotheses arising from VAPA KO cells are not tested by rescue experiments with exogenous VAPA and VAPA mutants. Although the proposed role for VAPA might fit with the data, the final model is not experimentally tested and is thus highly speculative. The role of VAPA at ER-PM contact sites near FA, and the direct link between VAPA, PI(4,5P)2, and FA disassembly, are not established. VAPA is not shown at ER-PM contacts in the manuscript. Some controls are missing and statistics must be improved. In summary, this work seems to be on the right track, but looks quite preliminary.

Reviewer #1 (Public Review):

The authors' conclusions presented herein are supported by a well-established mouse genetic conditional approach and an extensive array of phenotypic analyses.

Strengths:

1. The authors utilized well-described genetic tools, AdipoQCre, to target preadipocyte-like progenitor cell populations in bone marrow, as well as Csf1 floxed alleles. They further sifted through the cell population by showing that mature lipid-laden adipocytes express Csf1 at a much lower level, and determined that AdipoQCre-marked progenitor cell population presents a major cellular source of M-CSF,

2. The reanalysis of published scRNAseq datasets in Figure 1, as well as the following phenotypic analyses of the mutant mice are well-conducted. The analyses include a broad range of experiments both in vivo (3DmicroCT, histology, flow cytometry) and ex vivo (osteoclastogenesis assay in bone marrow cell culture). The confidence of the reported findings is high.

3. The data presented in this manuscript are of very high quality.

Weaknesses:

1. The role of AdipoQ-lineage progenitors as a source of M-CSF is overstated. The authors claim in many instances that "mature bone adipocytes do not express M-CSF", "These cells however do not produce Csf1", "...these peripheral AdipoQ+ cells nearly do not produce M-CSF". However, the authors' qPCR experiments only show four times differences in Csf1 expression. Therefore, the claim that AdipoQ-lineage progenitors are an exclusive source of M-CSF is not well substantiated. In line with this, some of the recent literature reporting conditional deletion of M-CSF in other bone cells (JBMR Plus. 4:e10080., Nature. 590:457-462) are not included.

2. Some of the phenotypic analyses are still incomplete. The authors did not report whether CHet (AdipoQCre Csf1(flox/+)) showed any bone phenotype. Further, the authors did not show that Csf1 mRNA or M-CSF protein is expressed in AdipoQ-lineage progenitors using histological methods. Current evidence is only based on scRNAseq and qPCR of isolated cells. Whether there was any change in circulating bone resorption markers in CKO mice was not shown. Cortical bone parameters were not included in the 3D-microCT analyses. These missing pieces of information would be important to correctly interpret the phenotypes.

3. Which bone marrow cell population(s) are marked by AdipoQCre remain largely unclear. It is possible that AdipoQCre also marks at least part of MSPC-osteo cluster in addition to MSPC-adipo. Adipo-lineage progenitors may not stay entirely as adipoprogenitors and drift toward osteoblasts or their precursor cells.

4. The OVX data in Figure 5 are not very well explained. The data do not seem to support the authors' conclusion that M-CSF deficiency in AdipoQ-lineage progenitors alleviates estrogen-deficiency induced osteoporosis. The CKO mice lose bone mass almost to the same extent as WT mice upon OVX.

Reviewer #2 (Public Review):

This study demonstrates that AdipoQ+ cells, which constitute approximately 0.8% of bone marrow mesenchymal cells, are major producers of M-CSF (Csf1) in murine bone marrow. The initial finding was discovered in scRNA seq datasets and studied in depth here with animal models and cellular assays. Deletion of Csf1 with AdipoQ-Cre increased trabecular bone mass in long bones and reduced the number of osteoclasts on trabecular bone surfaces. Cd11b+ F4/80+ macrophage numbers were also reduced in bone marrow. Bone loss from ovariectomy was prevented in Csf1∆AdipoQ female mice. Strengths of this study include use of a tissue-directed knock out (Adipo-Cre) model system to understand the relative contribution of AdipoQ+ cells to Csf1 levels and trabecular bone mass, careful examination of other adipose tissues for Csf1 expression, challenging bone responses in Csf1∆AdipoQ female mice with ovariectomy, and studying the effect of Csf1 deletion in macrophage levels. Mechanical studies of bone strength were not included but would be necessary to determine if deletion of Csf1 in AdipoQ+ cells is sufficient to cause osteopetrosis as concluded by the authors. Additional information on other molecular changes Csf1∆AdipoQ mice would provide insights into how deletion of Csf1 in AdipoQ+ cells affects bone remodeling. Overall, this is a very important study that has a lot of merit. It's impact on the field will be high because it is challenging the paradigm that osteoblasts and osteocytes are the major sources of M-CSF in the bone marrow.

Reviewer #3 (Public Review):

Macrophage colony-stimulating factor (M-CSF) plays key roles in the differentiation of myeloid-lineage cells, including monocytes, macrophages and osteoclasts. The latter mediate bone resorption, which is important for physiological bone remodelling but, unrestrained, contributes to bone loss in conditions such as in post-menopausal osteoporosis. M-CSF production within the bone marrow is implicated in the maintenance of myeloid and skeletal homeostasis, but the cellular source of bone marrow M-CSF has remained elusive. In this study, Inoue et al address this issue through advanced transcriptomic and gene targeting approaches. They conclude that a population of Adipoq-expressing progenitors within the bone marrow, designated "AdipoQ-lineage progenitors", is the key cellular source of M-CSF. Consistent with this, they find that transgenic deletion of M-CSF from these cells disrupts macrophage and osteoclast development, leading to osteopetrosis and possibly preventing bone loss following ovariectomy. However, they have not adequately addressed the possibility that M-CSF production from other cell types, particularly adipocytes in peripheral adipose tissues, may also be influencing these phenotypes. Specific strengths and weaknesses are as follows:

Strengths:

1. The manuscript is written in a clear, succinct manner and the data are generally nicely presented. It is therefore a pleasure to read.

2. The analysis of single-cell transcriptomic data is clear and convincing, and the skeletal phenotyping has been done to a high standard.

Weaknesses:

1. The authors underplay the potential contribution of M-CSF production from other cell types, particularly from adipocytes in peripheral adipose tissues. They show that M-CSF expression from these cells is lower than from the bone marrow progenitors that they focus on; however, based on this they allude to "no expression" of M-CSF from these other adipocytes. This overlooks the findings of other studies showing that peripheral adipocytes produce M-CSF and that this has biological functions. Whether their knockout model alters M-CSF expression in peripheral adipose tissue, whether for whole tissue or for isolated adipocytes, has not been tested.

2. The decreases in M-CSF have been assessed at the transcript level, but not for M-CSF protein. Whether their knockout model

3. It is also unclear if the Adipoq-lineage progenitors consist exclusively of adipogenic cells, or if osteogenic progenitors are also part of this population.

If these weaknesses are addressed then this work has potential to yield firm conclusions and new insights into the regulation of myeloid and skeletal homeostasis, both in normal physiology and in clinically relevant conditions.

Reviewer #1 (Public Review):

Although COVID-19 primarily causes an inflammatory response in the lungs, there is growing evidence that other organs are also affected by SARS-Cov-2, and that some patients continue to receive long-term effects of the disease sequelae even after treatment. We are not clear at this time about the effects of COVID-19 in organs other than the lungs. In this study, the authors presented the COVID Tissue Atlas (CTA) that comprises scRNA-seq data across six human organs of severe COVID-19. This study provides a valuable data resource to study the systemic effects of severe COVID-19, especially the common and specific transcriptional response to COVID-19 in multiple organs. Specifically, the authors identified dysregulated insulin and HIF signaling and prominent macrophage-endothelial interactions. This study will obviously help us to understand the pathogenesis of long-COVID.

Reviewer #2 (Public Review):

This cell atlassing study used single nuclei RNA-sequencing to profile cell type-specific transcriptional response to COVID-19 across multiple organs. The authors surveyed a cohort of 20 patients including 15 COVID-19 donors and 6 organs including the lung, liver and heart. They then annotated major cell types across these tissues and performed systematic differential gene expression analysis to propose cell type-specific shared transcriptional responses in macrophages and endothelial cells across multiple tissues. Finally, they inferred COVID-19 enriched cell interactions between macrophages and endothelia across multiple organs.

The strengths of the study include cross organ profiling from COVID-19 patients beyond the lungs, the immediate availability of this snRNAseq dataset as a resource and the systematic gene expression analysis that compares cell type specific disease programs across the body. There are several novel observations including dysregulation of insulin signalling in the liver and the heart. Most notable are the putative receptor-ligand interactions identified between macrophages and endothelial cells, an understudied aspect of COVID-19 tissue pathology.

However, the study presents weaknesses that diminish the impact of the resource. First, tissue profiling depth/coverage is lower than existing resources with relatively few number of cells per tissue and, more importantly, a very coarse grained cellular annotation. Second, the extent of coordinated gene expression changes across different organs is not very clear from the analysis presented in the paper, especially for macrophages. Finally, the comparisons to existing resources are not very strong and it would be more impactful to see the orthogonal (IHC or smFISH) validation of the novel snRNASeq observations in this study (e.g. endothelial-macrophage interactions).

Major comments:

1. While multiple organs have been profiled, the overall cell numbers are low (~85k nuclei across six organs) compared to existing studies (Delorey study from broad with ~100k nuclei from lung alone). There is also cell # and type bias towards certain donors - 6 donors (donors 15-20) have significantly more cells than others and majority of certain cell types come from a handful of donors (e.g. fibroblasts in covid lung). There is no analysis or discussion to compare the statistical power of this study to other resources - I expect it is limited in recovering DE genes compared to other resources, especially given patient heterogeneity in COVID-19.

2. The results on ABI/Transitional AT2 and PATS cells in the lung are not clear. While the increased basal cells are presented as likely ABIs, the label transfer seems to map most of this signature to AT1 cells (Fig 2E). Fig 2F presents gene expression similarities - but it is difficult to see them on the heatmap (there are few cells and this reviewer is color blind). A more quantitative approach or clear visualisation of shared definitive marker gene expression is needed. Regarding PATS, with the limited number of nuclei & patients profiled here, I am not confident in the label transfer based comparison to the Broad study.

3. More granular annotation of endothelial and macrophage subtypes would improve the utility of the resource. For example, lymphatic vs vascular endothelial cells in the lung show different responses to COVID-19 with the former population increasing in abundance in disease while the latter population diminishes (e.g. Broad delorey study). Such phenotypes cannot be extracted from the current annotation.

4. The extent of the cross organ coordinated response is not very clear. Fig 5A and Fig 5 sup fig suggest common DEG genes in macrophages and endothelial cells respectively across organs, but Fig 5F and G seem to suggest that DE coordination is close to random or not significant (except endothelial cells). Fig 5B-E correlations also seem limited. Fig 6C-E finds few cell-cell interactions conserved between macrophages and endothelial cells. In addition, endothelial cells change in abundance in opposite directions in the lung vs heart, suggesting divergent responses.

5. How many STR genes are there and are they conserved across different cell types?

6. Orthogonal validation of some of the novel findings with IHC or smFISH would confirm the robustness of novel findings and utility as a resource. The validation of hepatocyte insulin dysregulation or the vascular-macrophage cell interactions would add great value.

Reviewer #1 (Public Review):

In this study, Tanentzapf and colleagues have developed a new live-imaging technique for the lymph gland hematopoiesis over 12 hours, which is enough to visualize changes in the cell state or cell division by tracking the same cell. With the new method, the authors successfully cultured the lymph gland for a long period without modifying cell viability or stress responses and detected a continuous cell cycle and division. Moreover, the authors showed that lymph gland progenitors divide when they reach a certain size and regrow upon division, supporting previous findings and providing a new concept in lymph gland biology. The authors moved on to resolve the spatial distribution of progenitor mitosis in 3D and found that progenitors divide in a polarized manner which contributes to the typical shape of lymph glands. In addition to developing lymph glands, the authors observed the lymph gland following oral infection and found that progenitors divide less upon infection but significantly increase the number of differentiations at the MZ-CZ boundary. Furthermore, the authors found two different modes of differentiation in the lymph gland: sigmoid and linear, which are altered during infection.

Studies in the lymph gland hematopoiesis have heavily relied on snapshots of the lymph gland phenotypes although stem-progenitor differentiation is a continuous process. In this regard, the method developed in this study is extremely valuable to the fly community and will help improve the ex vivo culture and analysis techniques of fly organs as well as the lymph gland. The authors rigorously took advantage of numerous measures to validate the new method, including cell death, oxidative stress response, cell viability, and cell cycle, and observed biologically significant phenomena of the correlation between cell size and cell division, cell division polarity, and changes in the mode of differentiation during development or infection. This study provides a useful system for Drosophila immunologists and developmental biologists and will help explore the real-time mechanisms underlying blood development and immune reactions.

Reviewer #2 (Public Review):

The authors sought to be able to examine what cellular mechanisms underlie increases in mature blood cell production upon immune challenge. To this end they devised a new in vitro organ culturing system for the lymph gland, the main hematopoetic organ of the fruit fly Drosophila melanogaster; the fly serves as an excellent model for studying fundamental questions in immunology, as it allows live imaging combined with genetic manipulation, and the molecular pathways and cellular functions of its innate immune system are highly conserved with vertebrates.

The authors provide compelling evidence that the cultured lymph gland shows a similar time scale, dynamics, and capacity for cell division as was observed in vivo, and does not undergo undue oxidative stress in their optimized culture conditions. This technique will prove extremely useful to the large community studying the fly lymph gland, and potentially vertebrate immunologists seeking to expand the models they utilize.

In these cultured glands, the authors identify progenitors undergoing symmetric cell divisions and provide some evidence that is consistent with, but does not prove, that these two cells maintain their proliferative capacity. They detect equivalent levels in the two equally sized daughter cells of dome-Meso-GFP, a marker for JAK-STAT activity; however, this could be due to an equal inheritance of the protein from the mother, not an equivalent maintenance of a proliferative capacity.

The authors develop a technique to conduct tracking of progenitor cell size over time in the cultured lymph glands and identify a switch increase in growth after division, as well as two orientations of the divisions, with the main one occurring 90% of the time.

They show that bacterial infection results in a significant decrease in the division of Blood progenitors and the elimination of the minor orientation of division, but no obvious change in the rate of division.

By imaging two markers, Dome-GFP for the progenitor state and Eater dsRed for the differentiated one, they examine the trajectories by which differentiation occurs in the wild-type lymph gland. They describe two main categories of fate transitions. In one that they call linear, the blood cells express high levels of the differentiation marker along with the progenitor marker before turning off the progenitor marker. The dynamics of how these progenitor cells get to the state of expressing both the differentiation and progenitor marker at high levels is not described. In the other, which they call sigmoidal, cells express only high levels of the progenitor marker, and the differentiation marker increases after or as the progenitor marker decreases. The authors show that upon infection there is a large increase in the amount of the linear type of differentiation.<br /> But how this change in the type of differentiation upon infection explains the increased amount of differentiation is not clear.

A potential explanation comes from an aspect of their data that the authors don't comment upon. In their live analysis of lymph glands at a distinct time point in the uninfected state (Fig 7M-N), 95% of the cells they analyze traversing the sigmoidal path are in the intermediate step. This would predict that the cells on this path spend a much longer time stuck in this intermediate state before traversing to the final differentiated one, or that only a small fraction of the cells that become sigmoidal intermediate cell progress onwards to full differentiation. But this does not match the trajectories observed in the real-time analysis for uninfected cultured lymph glands (Fig 7A'-D'). marker. Perhaps their algorithm discarded traces from the live imaging in which the differentiation marker did not come up quickly and was thus not analyzed in the trajectories. If my interpretation of the single time point analysis is true, this would argue that the linear path is actually much faster/more fruitful than the sigmoidal one and this would explain why a higher level of total progenitor differentiation infection is the result of infection-inducing more differentiation by the linear path. Otherwise, I don't understand how their data explains that observation.

This work provides a very useful new system for Drosophila immunologists and could provide an important new perspective on the systems-level mechanisms that an organism utilizes to enable increased differentiation of immune cells upon infection.

Reviewer #3 (Public Review):

In this study, the authors sought to develop an ex vivo organ culture system that would allow for long-term (>12 hours) live imaging of the lymph gland (LG), the hematopoietic organ in Drosophila, in order to gain insights into the process of differentiation during hematopoiesis. The authors successfully built such a system through trial and error and showed that the LG could survive for over 12 hours and that it recapitulated many of the aspects seen in in vivo LGs.

The authors also developed sophisticated quantitative image analysis tools that allowed them to identify new modes of differentiation that may help explain the cellular heterogeneity previously seen by other groups. Furthermore, they were able to follow mitosis in real-time and showed evidence that not only can progenitors undergo symmetric cell division but that mitosis shows some orientation bias which may help explain the overall structure of the organ. The authors went on to show that upon infection, modes of differentiation and mitosis orientation seem to shift, but they did not provide any mechanistic insight into how this may occur or whether these shifts would impact the final cell fate or function of the mature hemocytes. Nevertheless, the identification and description of these patterns are in itself helpful and informative and provide a basis for future studies delving into these mechanistic questions.

The major strengths of the methods include the advancement in live-imaging technology and the development of quantitative image analysis tools. Weaknesses of the results include small sample sizes (and relatively high p-values), which limit the strength and breadth of some conclusions. This is to be expected as there is a trade-off between long-term live imaging of individual samples and sample number, nevertheless, it represents a minor weakness. Overall this weakness is overshadowed by the strength of the advancements afforded by live imaging and following in real-time the process of differentiation and mitosis. Furthermore, the quantitative analysis tools developed and used in this study can be applied across multiple subfields and represents an important step forward in the field.

The evidence presented here is generally solid and the results tend to support their conclusions although some specific conclusions are supported by data with no p-values noted or relatively high p-values and low correlation coefficients, and so should be interpreted with this in mind.

This study represents a compelling and convincing theoretical and technical advance in efforts to understand hematopoiesis in flies. This is a powerful and versatile system that will allow for not only genetic manipulation of the LG but also of the tissues co-cultured with the LG to elucidate the mechanisms that control various signaling pathways during homeostasis. In fact, which additional tissues (like the fat body and brain) that had to be included in the co-culture system in order for the LG to survive recapitulate what past studies have shown about where key signals come from that help maintain homeostasis in the LG.

One caveat of the work is that because the authors used Eater-DsRed to follow differentiation, these modes may only apply to the formation of plasmatocytes and not necessarily crystal cells, which the authors noted do not tend to go through an Eater-DsRed intermediate state. Future work using this live-imaging system and image analysis tools to study the formation of the various mature cell types in flies will be a valuable addition to the field.

The methods developed here will be highly useful to both the specific subfield and to the general scientific community and will likely spark new insights into the process of hematopoiesis when combined with different markers and genetic manipulations, as outlined by the authors in the discussion. Future studies that explore whether the different modes of differentiation identified here ultimately result in divergent cell fates for the mature hemocytes will be important for understanding the significance of the findings more generally. But the identification of changes in the ratios and rates of the modes of differentiation upon infection with E.coli suggests functional ramifications of the different modes. It will be interesting to see if other types of infection or systemic stresses cause similar or different changes in differentiation modes.

Reviewer #1 (Public Review):

Single-cell sequencing technologies such as 10x, in conjunction with DNA barcoded multimeric peptide MHCs (pMHCs) has enabled high throughput paring of T cell receptor transcript with antigen specificity. However, the data generated through this method often suffers from the relatively high background due to ambient DNA barcodes and TCR transcripts leaking into "productive" GEMs that contain a 10X bead and a T cell decorated with antigen-specific barcoded proteins. Such contaminations can affect data analysis and interpretation and have the potential to lead to spurious results such as an incorrect assessment of antigen-TCR pairs or TCR cross-reactivity. To address this problem, Povelsen and colleagues have described a data-driven algorithm called "Accurate T cell Receptor Antigen Pairing through data-driven filtering of sequencing information from single-cells" (ATRAP) that supplies a set of filtering approaches that significantly reduces background and allows for accurate pairing of T cell clonotypes with cognate pMHC antigens.

This paper is rigorously conducted and will be useful for the field - there are some areas where further clarifications and comparisons will benefit the reader.

Strengths:<br /> 1. Povelsen and colleagues have systematically evaluated the extent to which parameters in the experimental metadata can be used to assess the likelihood of a GEM to correctly identify the antigen specificity of the associated T cell clonotype.<br /> 2. Povelsen and colleagues have provided elegant data-driven scoring metrics in the form of concordance score, specificity score, and an optimal ratio of pMHC UMI counts between different pMHCs on a GEM, which allows for easy identification of poor quality data points.<br /> 3. Based on the experimental goals, ATRAP allows for customizable filters that could achieve appropriate data quality while maximizing data retention.

Weakness:<br /> 1. The authors mention that 100% of the 6,073 "productive" GEMs contained more than one sample hashing barcode, and 65% contained pMHC multiplets. While the rest of the paper elaborates on the steps taken to deal with pMHC multiplets issue, not much is said about the extent of multiplet hashing issue and how was it dealt with when assigning cells to individual donors. How is this accounted for? Even a brief explanation would be beneficial.

2. It would be helpful for the authors to describe how experimental factors such as the quality of the input MHC protein may affect the outputted data (where different proteins may have different degrees of non-specific binding), and to what degree the ATRAP approach is robust to these changes. As an example, the authors mention that RVR/ A03 was present at high UMI counts across all GEMs and RPH/ B07 was consistently detected at low levels. Are these observations the property of the pMHCs or the barcoded dextran reagent? Furthermore, are there differences in the frequency of each of these multimers in the starting staining library which manifests in consistent high vs low read counts for the pMHC barcodes?

3. It would be helpful for the authors to further explain how ATRAP handles TCRs that may be present in only one (or a small number) of GEMs, as seen in Figure 7b, and potentially for the large number of relatively small clonotypes observed for the RVR/A03 peptide in Figure 6 (it is difficult to know if the long tail of clonotypes for RVR is in the range of 1 or 10 GEMs based on the scale bar). Beyond that, is there any effect on expected (or observed) clonal expansion on these data analyses, for example, if samples are previously expanded with a peptide antigen ex vivo or not?

4. The authors mention a second method, ICON, for conducting these types of analyses, and that the approach leads to significantly more data loss. However, given there could be differences in dataset quality themselves, and given the dataset, ICON is publicly available, it would be helpful for a more explicit cross-comparison to be conducted and presented as a figure in the paper.

Reviewer #2 (Public Review):

The study by Povlsen, Bentzen et al. describes certain computational pipelines authors used to analyze the results from a single-cell sequencing experiment of pMHC-multimer stained T cells. DNA-barcoded pMHC multimers and single-cell sequencing technologies provide an opportunity for the high-throughput discovery of novel antigen-specific TCRs and profiling antigen-specific T-cell responses to multiple epitopes in parallel from a single sample. The authors' goal was to develop a computational pipeline that eliminates potential noise in TCR-pMHC assignments from single-cell sequencing data. With several reasonable biological assumptions about underlying data (absence of cross-reactivity between these epitopes, same specificity for different T-cells within a clonotype, more similarity for TCRs recognizing the same epitope, HLA-restriction of T cell response) authors identify the optimal strategy and thresholds to filter out artifacts from their data.

It is not clear If the identified thresholds are optimal for other experiments of this kind, and how the violation of authors' assumptions (for example, inclusion of several highly similar pMHC-multimers recognized by the same clone of cross-reactive T cells) will impact the algorithm performance and threshold selection by the algorithm. The authors do not discuss several recent papers featuring highly similar experimental techniques and the same data filtering challenges:<br /> https://www.science.org/doi/10.1126/sciimmunol.abk3070<br /> https://www.nature.com/articles/s41590-022-01184-4<br /> https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9184244/

Unfortunately, I was unable to validate the method on other datasets or apply other approaches to the authors' data because neither code nor raw or processed data were available at the moment of the review.

One of the weaknesses of this study is that the motivation for the experiment and underlying hypothesis is unclear from the manuscript. Why these particular epitopes were selected, why these donors were selected, are any of the donors seropositive for EBV/CMV/influenza is unclear. Without particular research questions, it is hard to evaluate pipeline performance and justify a particular filtering strategy: for some applications, maximum specificity (i.e. no incorrect TCR specificity assignments) is crucial, while for others the main goal is to retain as many cells as possible.

Reviewer #3 (Public Review):

The method of ATRAP provides a useful workflow for processing and analysing single-cell sequencing data of TCRs and barcoded pMHC. The method addresses an important subfield of research, as the availability of these datasets is increasing substantially due to the wider availability of commercial reagents and tools.

Overall the study is highly technical and can be considered almost a "user manual" to assist researchers who pursue this TCR-pMHC specificity experiments by single-cell sequencing. Convincing experimental work, data analysis, appropriate controls, and technical details are provided throughout.

Reviewer #1 (Public Review):

George and Levine present in their manuscript a mathematical framework describing the evolution of tumor cells under immune surveillance. The adaptive immune system recognizes tumor associate antigens (TAAs) to eliminate the cancer cells, while the tumor evades it through an evolutionary process of clonal selection. Their framework describes how the TAAs are gained and lost from the tumor, as a discreet time-stochastic process. The authors construct and parametrize their model to fit different known regimes of tumor and its microenvironment and explore the consequences of different tumor behaviors. Specifically, they suggest that tumor cells might sense the action of the immune system and adapt their escape probability.

The mathematical analysis is clear and is an impressive attempt to find governing principles behind a complicated and messy process. While the model cannot give specific predictions at this point, it facilitates understanding real-world observations, like high and low mutation tumors. As such it can motivate further modeling of more realistic situations. In its current form, however, the manuscript is difficult to follow, with the many mathematical details and regimes confounding the message. Also, since the model simplifies the clonal nature of the evolution processes considerably, in its current form it has limited capability to make predictions or be more than supporting evidence to empirically known observations.

Reviewer #2 (Public Review):

The manuscript "Optimal Cancer Evasion in a Dynamic Immune Microenvironment Generates Diverse Post-Escape Tumor Antigenicity Profiles" by George and Levine describes TEAL - a mathematical model for the dynamics of cancer evolution in response to immune recognition. The authors consider a process in which tumor cells from one clone are characterized by a set of neoantigens that may be recognized by the immune system with a certain probability. In response to the recognition, the tumor may adapt to evade immune recognition, by effective removal of recognizable neoantigens. The authors characterize the statistics of this adaptive process, considering, in particular, the evasion probability parameter, and a possibility of an adaptive strategy when this parameter is optimized in each step of the evolution. The dynamics of the latter process are solved with a dynamic programming approach. In the optimal case, the model captures the tradeoff between a cancer population's need for adaptability in hostile immune microenvironments and the cost of such adaptability to that population. Additionally, immune recognition of neoantigens is incorporated. These two factors, anti-tumor vs pro-tumor IME as quantified by the Beta penalty term, and the level of immune recognition as quantified by the rate q, form the basis of a characterization of tumors as 'hot' or 'cold'.

I think this framework is a valuable attempt to formally characterize the processes and conditions that result in immunologically hot vs cold tumors. The model and the analytical work are sound and potentially interesting to a major audience. However, certain points require clarification for evaluation of the relevance of the model:

1) Tumor clonality

My main concern is about the lack of representation of the evolutionary process in the model and that the heterogeneity of the tumor is just glossed over.

The single mention of the problem occurs in Section 2, p2: "Our focus is on a clonal population, recognizing that subclonal TAA distributions in this model may be studied by considering independent processes in parallel for each clone."

I don't think this assumption resolves the impact of tumor heterogeneity on the immune evasion process. Furthermore, I would claim that the process depicted in Fig 1A is very rare and that cancers rarely lose recognizable neoantigens - typically it would be realized via subclonal evolution, with an already present cancer clone without the neoantigens picking up. Similarly, the adaptation of a tumor clone is an evolutionary process - supposedly the subclones that manage to escape recognition via genetic or epigenetic changes are the ones that persist. It is not clear what the authors assume about the heterogeneity of the adapting/adapted population between different generations, n->(n+1). Is the implicit assumption that the n+1 generation is again clonal, i.e. that the fitness advantage of the resulting subclone was such that the remaining clones were eliminated? Or does the model just focuses on the fittest subclone? A discussion on whether these considerations are relevant to the result would clarify the relevance of the result.

2) Time scales

Section 2, p2: "We assume henceforth that the recognition-evasion pair consists of the T cell repertoire of the adaptive immune system and a cancer cell population, recognizable by a minimal collection of s_n TAAs present on the surface of cancer cells in sufficient abundance for recognition to occur over some time interval n.".

How do the results depend on the duration of interval n? The duration should be long enough to allow for recognition and, up to some limiting duration, proportional to the TAA recognition probability q. However, it should not be so long that the state of the system can change significantly. A clarification on this point is needed.

Reviewer #3 (Public Review):

Cancer cell populations co-evolve under the pressure exerted by the recognition of tumor-associated antigens by the adaptive immune system. Here, George and Levine analyze how cancers could dynamically adapt the rate of tumor-associated antigen loss to optimize their probability of escape. This is an interesting hypothesis that if confirmed experimentally could potentially inform treatments. The authors analyze mathematically how such optimally adapting tumors gain and lose tumor-associated antigens over time. By simplifying the complex interplay of immune recognition and tumor evolution in a toy model, the authors are able to study questions of practical interest analytically or through stochastic simulations. They show how different model parameters relating to the tumor microenvironment and immune surveillance lead to different dynamics of tumor immunogenicity, and more immunologically hot or cold tumors.

Simple models are important because they allow an exhaustive study of dynamical regimes for different parameters, such as has been done elegantly in this study. However, in this quest for simplification, the authors have not considered biological features that are likely to be of importance for understanding the process of cancer immune co-evolution in generality: tumor heterogeneity and immune recognition that only stochastically results in cancer elimination. In this sense, this paper might be seen as the opening act in a series of more sophisticated models, and the authors discuss avenues towards such further developments.

Reviewer #1 (Public Review):

N1-methyladenosine (m1A) is a rather intriguing RNA modification that can affect gene expression and RNA stability etc. The manuscript presented the exploration of RNAs m1A modification in normal and OGD/R-treated neurons and the effects of m1A on diverse RNAs. The authors showed that m1 modification can mediate circRNA/LncRNA-miRNA-mRNA mechanism and 3'UTR methylation of mRNAs can disturb miRNA-mRNA binding.

The manuscript provides evidence for the following,<br /> 1. The OGD/R can have impacts on various functions of m1A mRNAs and neuron fates.<br /> 2. The m1A methylation of mRNA 3'UTRs disturbs the miRNA-mRNA binding.<br /> 3. The authors identified three possible patterns of m1A modification regulation in neurons.

The main merit of the manuscript is that the authors identified some critical features and patterns of m1A modification and in neurons and OGD/R-treated neurons. Moreover, the authors identified m1A modifications on different RNAs and explored the possible effects of m1A modification on the functions of different RNAs and the overall posttranscriptional regulation mechanism via an integrated approach of omics and bioinformatics. The major weakness of the manuscript is that technique details for many results are missing. Moreover, language inconsistences can be found throughout the manuscript. My general feeling about the manuscript is that some conclusions are rather superficial and therefore require validation and discussion.

Reviewer #2 (Public Review):

In this manuscript, investigators explore the m1A modification, an important post-transcriptional regulatory mechanism, in primary normal neuron and OGD/R treated neuron. As far as I know, the regulatory m1A modification remains poorly characterized in neuron. This is an interesting topic in the context of epitranscriptomics. This paper not only provided us with a landscape of m1A modifications in neuron, but also explored the impact of m1A modifications on the biological functions of different RNA (mRNA, lncRNA, circRNA). In addition, the argument that m1A modification affects miRNA binding to other RNAs is of interest to reader, and the authors have performed a dual luciferase validation here to add feasibility to this conclusion.

Reviewer #3 (Public Review):

Overall, this is an interesting and well performed study that described a comprehensive landscape of m1A modification in primary neuron and investigated the role of m1A in the circRNA/lncRNA‒miRNA-mRNA regulatory network following OGD/R. The focus on the two different complex regulatory networks for differential expression and differential methylation is important and it will be a valuable resource for the research community that focuses on epitranscriptomics and central nerve system diseases. Collectively, the authors present an exciting piece of work that certainly adds to the literature regarding epitranscriptomic features in neuron. While interesting results obtained and the paper is nicely written, I have the following suggestions for minor revisions to improve the paper.

1. The authors have explored the role of m1A modification in neuron, but it would have been helpful if the authors described the significance of these findings in depth in some sections (Figure 5 and Figure 6) to enhance the value of the article.<br /> 2. The authors should describe in detail the current research state of m1A modification and the significance of this study to the field of epitranscriptomics in the introduction and discussion section.

Reviewer #1 (Public Review):

Here, the authors generated a CSAS-LexA driver line to investigate the expression pattern of CSAS and showed that CSAS expression is confined to glia and does not overlap with DSiaT expression. DSiaT expression is presumed to be in neurons, but this was not evaluated with specific markers in this study.

The authors showed that restoring CSAS expression specifically in glia but not neurons could rescue the mutant phenotype of temperature sensitive paralysis and confirmed that glial (and not neuronal) CSAS expression could rescue excitatory junction potentials at neuromuscular junctions in CSAS mutants. In addition to rescue experiments, the authors also performed RNAi knockdowns in glia vs. neurons to show that CSAS function is required in glia and DSiaT in neurons for the same paralysis phenotype.

Next, the authors performed mass spec to analyse sialylated proteins in larval brains and found that sialylated proteins could not be detected in DSiaT and CSAS mutants. However, sialylated proteins were only barely detectable in wildtypes.

Of note, the authors show that CSAS functions normally in glia and cannot function in neurons due to low endogenous NANS activity (sialic acid synthase).

Finally, the authors explore the hypothesis that the temperature-sensitive paralysis CSAS phenotype is due to oxidative stress with a paraquat exposure paradigm. This could be strengthened by examining ROS levels in vivo in CSAS or DSiaT mutants. The specific genetic background of these experiments seemed to be a major factor in the results obtained and more stringent controls or backcrossing to isogenize the genetic background would be required to be fully confident in the conclusions drawn from these experiments.

The authors also demonstrate a link between sialylation and Para (protein) expression. Although intriguing, there is very little data provided on this aspect of the story, though it does not detract from the broader message of the manuscript.

Reviewer #2 (Public Review):

The function of many proteins depends on posttranslational modifications. Protein glycosylation is widespread and glycosylated proteins are mostly found on the outer surface of cells, where it is frequently implicated in cell-to-cell adhesion. It involves the addition of often complex and branched sugar chains to a protein backbone. Sialic acid is a particular relevant sugar as it is negatively charged and occupies terminal positions at the glycan chain. The enzymatic cascade leading to sialylated proteins is known. Unlike mammals, flies have only one sialyltransferase (SiaT), thus, Drosophila is a particularly well-suited model to study protein sialylation. The penultimate enzymatic steps in sialylation are mediated by N-acetlyneuraminic acid synthetase (NANS) and sialic acid synthetase (CSAS).

Scott et al., start with careful and state-of-the-art dissection of the expression patterns of the relevant genes. They first generated transgenic flies harboring a BAC covering the CSAS gene - which was able to rescue the mutant phenotype. They then replaced the CSAS coding sequence with LexA and demonstrated that LexA expression was sufficient to drive LexAop-CSAS to a full rescue of the CSAS mutant. CSAS-LexA was found to be active only in Repo expressing glial cells. The authors performed further experiments employing another BAC harboring an HA-tagged SiaT gene and found complementary expression in neurons (here I missed a comment on why the endogenously tagged SiaT gene (Repnikova 2010) was not used).

To study cell-type specific requirements UAS-based rescue experiments were conducted. The CSAS mutant phenotype could be rescued not only by panglial expression of CSAS but also by expression exclusively in subperineurial or ensheathing glial cells. Whether astrocytes or cortex glial cells are similarly able to rescue the mutant phenotype has not been addressed. No rescue was observed when CSAS was expressed in neurons, but co-expression of CSAS and NANS led to a partial rescue, further validating the split of the biosynthetic pathway leading to sialylated proteins to glial and neuronal cells.<br /> In addition to the rescue experiment, the authors also performed RNAi-based knockdown experiments for both, CSAS and SiaT which together support the conclusion that sialylation requires a split of the biosynthesis pathway.

In a subsequent mass spec approach, the authors analyzed sialylated proteins in larval brains. Whereas in wild type brains sialylated proteins were barely detected, they could not be seen in SiaT or CSAS mutant brains. However, according to Flybase, the highest expression of both genes is in adult flies. Why not look at these stages? It would also be good to use the cell type-specific knockdown flies for such experiments to fully support the notion that sialylation requires a glia-neuron transfer of intermediates. Possibly, low (and thus undetected) levels of SiaT in glia could be sufficient for function. In this respect it is interesting that the presence of a UAS-SiaT element is sufficient to rescue the SiaT mutant phenotype, suggesting that only very low levels of SiaT are needed for function.

Subsequently, Scott et al., demonstrate that the paralysis phenotype of CSAS mutants is sensitive to gene dose and that CSAS activity protects flies from oxidative stress. Quite interesting, they also demonstrate that sialylation is required - directly or indirectly - to maintain protein expression of the voltage gate sodium ion channel Para.

Reviewer #3 (Public Review):

In this work, the authors find that similar to mammals, sialylation is critical in neurons within flies, yet in flies the critical substrate for sialylation, CMP-Neu5Ac, is 'outsourced' to glial cells. These findings are shown through an extensive array of knockout, knockdown, and transgenic flies where CMP-Neu5Ac biosynthesis and sialyltransferase expression is modulated in either glial cells or neurons. The importance of sialylation in neurons is demonstrated by showing that sialylation impacts the expression levels of a critical voltage-gated ion channel.

This elegant work dissecting sialylation in the fly brain convincingly demonstrates the requirement for glial cells in the process of sialylation of neurons and deserves to be published. The major unaddressed question remaining is precisely how the CMP-Neu5Ac is delivered from the glial cells to neurons with several possibilities that merit further discussion including (but not limited to): extracellular vesicles, receptor-mediated uptake (unlikely but can't be ruled out), or exocytosis. The authors could make the point stronger that CMP-Neu5Ac should not be able to cross the neuronal membrane (or the Golgi membrane for that matter), requiring specific transport mechanisms.

Reviewer #1 (Public Review):

High-throughput genetic screening is a powerful approach to elucidate genes and gene networks involved in a variety of biological events. Such screens are well established in single-celled organisms (i.e. CRISPR-based K/O in tissue culture or unicellular organisms; screens of natural variants in response to drugs). It is desirable to extend such methodology, for example to Arabidopsis where more than 1000 ecotypes from around the Northern hemisphere are available for study. These ecotypes may be locally adapted and are fully sequenced, so the system is set up for powerful exploration of GxE. But to do so, establishing consistent "in vitro" conditions that mimic ecologically relevant conditions like drought is essential.

The authors note that previous attempts to mimic drought response have shortcomings, many of which are revealed by 'omics type analysis. For example, three treatments thought to induce osmotic stress; the addition of PEG, mannitol, or NaCl, fail to elicit a transcriptional response that is comparable to that of bonafide drought. As an alternative, the authors suggest using a low water-agar assay, which in the things they measure, does a better job of mimicking osmotic stress responses. The major issues with this assay are, however, that it introduces another set of issues, for example, changing agar concentration can lead to mechanical effects, as illustrated nicely in the work of Olivier Hamant's group (e.g., https://elifesciences.org/articles/34460).

Reviewer #2 (Public Review):

The authors aim to make a reliable plate-based system for imposing drought stress (which for experiments like this would be better referred to as low water potential stress). This is an admirable goal as a reliable experimental system is key to conducting successful low-water potential experiments and some of the experimental systems in use have problems. They compare several treatments but seem to be unaware that such comparisons need to be based on the measurement of water potential as the fundamental measure of how severe the level of water limitation is. Only by comparing things at the same water potential can one determine if the methods used to impose the low water potential are introducing confounding factors. In this manuscript, they compare several agar-plate-based treatments to what they view as a baseline experiment of plants subjected to soil drying. However, that baseline soil drying (vermiculite drying, to be precise) experiment illustrates many of the problems present in the molecular drought literature in that they give no information on plant or soil water potential or water content. Thus, there is no way to know how severe the drought stress was in that experiment and no way for any other lab to reproduce it. It is directly akin to doing a heat stress experiment and not reporting the actual temperature.

They compare transcriptome data from this soil drying experiment to transcriptome data from agar plates with PEG, mannitol or salt added. However, this comparison is problematic, because none of the treatments being compared are at the same water potential (as mentioned above). Also, the PEG-infused agar plates have limitations in that no buffer is added and it is not clear that anything is done to check or control the pH. Adding PEG to the solution will reduce the pH. Thus, in their unbuffered PEG plates, the plants are almost certainly exposed to low pH stress and this can explain the supposed difference they observe between PEG and other treatments, especially since the plants are left on such stressful pH conditions for a relatively long period. It is also problematic that the comparison between soil drying and plate-based treatments is at different times (5 vs 14 days). They also show an over-reliance on the GO annotations of drought-induced gene expression. This GO annotation is based on experiments using very severe stress for a short time period. It is notorious for not accurately reflecting what happens on longer-term exposure to more moderate levels of low water potential stress. Thus, for example, we would not expect many of the canonical drought regulation genes (RD29A and similar genes) to be upregulated in the longer-term treatments as its expression is induced rapidly but also rapidly declines back to near baseline at the plant acclimates to the low water potential stress.

The authors have not always considered literature that would be relevant to their topic. For example, there is a number of studies that have reported (and deposited in the public database) transcriptome analysis of plants on PEG-plates or plants exposed to well-controlled, moderate severity soil drying assays (for the latter, check the paper of Des Marais et al. and others, for the former, Verslues and colleagues have published a series of studies using PEG-agar plates). They also overlook studies that have recorded growth responses of wild type and a range of mutants on properly prepared PEG plates and found that those results agree well with results when plants are exposed to a controlled, partial soil drying to impose a similar low water potential stress. In short, the authors need to make such comparisons to other data and think more about what may be wrong with their own experimental designs before making any sweeping conclusions about what is suitable or not suitable for imposing low water potential stress.

To solve the problem of using these other systems to impose low water potential stress, the authors propose the seemingly logical (but overly simplistic) idea of adding less water to the same mix of nutrients and agar. Because the increased agar concentration does not substantially influence water potential (the agar polymerizes and thus is not osmotically active), what they are essentially doing is using a concentrated solution of macronutrients in the growth media to impose stress. This is a rediscovery of an old proposal that concentrated macronutrient solutions could be used to study the osmotic component of salt stress (see older papers of Rana Munns). There are also effects of using very hard agar that is of unclear relationship to actual drought stress and low water potential. Thus, I see no reason to think that this would be a better method to impose low water potential.

Reviewer #3 (Public Review):

This work compares transcriptional responses of shoots and roots harvested from four plate-based assays that simulate drought and from plants subjected to water deficit in pots using the model plant Arabidopsis thaliana with the aim to select a plate-based assay that best recapitulates transcriptional changes that are observed during water-deficit in pots. Polyethylene glycol (PEG), mannitol, and sodium chloride (salt) treatments that are commonly used by molecular biologists to simulate drought were used for the plate-based assays as well as a new assay that uses increased concentrations of agar and nutrients to elicit drought which was developed by the authors and termed a 'low-water agar' assay since the amount of water added to the media mix and plates was lowered. Plants in pots were grown on vermiculite with the same nutrient mix as used in the plates and drought was induced by withholding watering for five days. Additionally, treatment with abscisic acid was conducted to study whether growth on plates itself led to artifacts compared to water deficit in pots. Shoot and root samples were harvested from all treatments for RNA sequencing analysis and differentially expressed genes were called against control samples.

The authors observed that gene expression responses of roots in their 'low-water agar' assay resembled more closely the water deficit in pots compared to the PEG, mannitol, and salt treatments (all at the highest dose). In particular, 28 % of PEG led to the down-regulation of many genes that were up-regulated under drought in pots. Through GO term analysis, it was pointed out that this may be due to the negative effect of PEG on oxygen solubility since downregulated genes were over-represented in oxygen-related categories. The data also shows that the treatment with abscisic acid on plates was very good at simulating drought in roots. Gene expression changes in shoots showed generally a high concordance between all treatments at the highest dose and water deficit in pots, with mannitol being the closest match. This is surprising, since plants grow in plates under non-transpiring conditions, while a mismatch between water loss by transpiration on water supply via the roots leads to drought symptoms such as wilting in pot and field-grown plants. The authors concluded that their 'low-water agar' assay provides a better alternative to simulate drought on plates.

Strengths:

The development of a more robust assay to simulate drought on plates to allow for high-throughput screening is certainly an important goal since many phenotypes that are discovered on plates cannot be recapitulated on the soil. Adding less water to the media mix and thereby increasing agar strength and nutrient concentration appears to be a good approach since nutrients are also concentrated in soils during water deficit, as pointed out by the authors. To my knowledge, this approach has not specifically been used to simulate drought on plates previously. Comparing their new 'low-water agar' assay to popular treatments with PEG, mannitol, salt, and abscisic acid, as well as plants grown in pots on vermiculite led to a comprehensive overview of how these treatments affect gene expression changes that surpass previous studies. It is promising that the impact of 'low-water agar' on the shoot size of 20 diverse Arabidopsis accessions shows some association with plant fitness under drought in the field. Their methodology could be powerful in identifying a better substitute for plate-based high-throughput drought assays that have an emphasis on gene expression changes.

Weaknesses:

While the authors use a good methodological framework to compare the different drought treatments, gene expression changes were only compared between the highest dose of each stress assay (Fig. 2B, 3B). From Fig. 1F it appears that gene expression changes depend significantly on the level of stress that is imposed. Therefore, their conclusion that the 'low-water agar' assay is better at simulating drought is only valid when comparing the highest dose of each treatment and only for gene expression changes in roots. Considering how comparable different levels of stress were in this study leads to another weakness. The authors correctly point out that PEG, mannitol, and salt are used due to their ability to lower the water potential through an increase in osmotic strength (L. 45/46). In soils, water deficit leads to lower water potential, due to the concentration of nutrients (as pointed out in L. 171), as well as higher adhesion forces of water molecules to soil particles and a decline in soil hydraulic conductivity for water, which causes an imbalance between supply and demand (see Juenger and Verslues, The Plant Cell 2022 for a recent review). While the authors selected three different doses for each treatment that are commonly used in the literature, these are not necessarily comparable on a physiological level. For example, 200 mM mannitol has an approximate osmotic potential of around -5 bar (Michel et al. Plant Physiol. 1983) whereas 28 % PEG has an osmotic potential closer to -10 bar (Michel et al. Plant Physiol. 1973). It also remains unclear how the increase in agar concentration versus the increase in nutrient concentration in the 'low-water agar' affect water potentials. For these reasons it cannot be known whether a better match of the 'low-water agar' at the 28% dose to water deficit in pots for roots in comparison to the other treatments is due to a good match in stress levels with the 'low-water agar' or adverse side-effect of PEG, mannitol, or and salt on gene regulation. Lastly, since only two biological replicates for RNA sequencing were collected per treatment, it is not possible to know how much variance exists and if this variance is greater than the treatments themselves.

Reviewer #1 (Public Review):

In this study, the authors found that the chromatin remodeling complex mutant isw1Δ of the fungal pathogen Cryptococcus neoformans is resistant to multiple different antifungal drugs. The mutant, however, is fully virulent in a mouse model. By comparing transcript changes of the wild type and the mutant when treated with antifungal fluconazole, they found that many transporter genes are differentially expressed in the isw1Δ mutant. Consistently, they showed reduced expression of genes involved metabolism of another antifungal 5-FC and a lower level of cellular accumulation of 5-FC in the isw1Δ mutant, which likely contributes to its 5-FC resistance. They found that the Isw1 protein is degraded mostly through ubiquitination and identified K97 deacetylation as being critical for drug resistance/protein degradation. Then they mutated nine E3 ubiquitination ligase genes and identified Cdc4 to be responsible for Isw1 degradation. Lastly, they showed that Isw1 is low in some clinical isolates that are modestly resistant to antifungals. The evidence of the interplay between acetylation status and ubiquitination of Isw1 is strong. The finding that reduced Isw1 increases drug resistance also fits the growing interest in studying epigenetic regulation of drug resistance in fungal pathogens. One area that needs to be strengthened is the potential clinical relevance of Isw1 reduction in drug resistance.

Reviewer #2 (Public Review):

Cryptococcus neoformans is an important human pathogen, particularly in immunocompromised individuals. Like many fungal pathogens, resistance to antifungal drugs can emerge quickly in Cryptococcus. Understanding the mechanisms by which fungi develop resistance to antifungals will support new treatment strategies and, potentially, identify new drug targets. In this manuscript, Meng et al. describe a novel role for the conserved ATP-dependent chromatin remodeling factor, Imitation Switch (Isw1) in responding to antifungals in Cryptococcus. The authors first find that loss of Isw1 increases resistance to multiple antifungals and changes expression levels of genes potentially involved in antifungal resistance using functional genetics and cell growth assays. Next, the authors use mass spectrometry data (data generated in this study and public data) to identify ubiquitinated and acetylated sites of Isw1. The authors use this information to carry out an extensive series of western blot experiments using point mutations and chemical perturbations to dissect the contribution of specific modified sites of Isw1. Here, they identify important roles for the acetylation of K97 and ubiquitination of K113 and K441 in Isw1 stability. Lastly, the authors present evidence that clinical isolates of Cryptococcus that have increased antifungal resistance may have defects in Isw1 stability and that overexpressing ISW1 reduces antifungal resistance.

Strengths:

The authors present novel data that Isw1 is involved in responding to antifungals and that changes in Isw1 stability may lead to antifungal resistance. These results are of particular interest to the fungal pathogen research community and add to the general understanding of antifungal resistance.

The authors present exciting data on post-translation modification (i.e., acetylation and ubiquitination) of Isw1, how those modifications contribute to Isw1 stability, and the regulatory interplay between modifications. Considering that Isw1 is broadly conserved across eukaryotes, these results are, potentially, of broad interest and raise questions outside of pathogen biology to be addressed in future research. For example, are the residues characterized in this study conserved in other Isw1 homologs, are they similarly modified, and is regulating the stability of Isw1 (or other chromatin remodeling factors) a general strategy for responding to external signals?

Weaknesses:

The authors demonstrate that Isw1 has a role in responding to antifungals in Cryptococcus. However, it is not clear if changes in Isw1 stability represent a general response to stress. This study would have benefited from experiments to test: (1) if levels of Isw1 change in response to other stressors (e.g., heat, osmotic, or oxidative stress) and (2) if loss of Isw1 impacts resistance to other stressors.

The authors demonstrate a critical role in the acetylation of K97 and ubiquitination of K441 in regulating Isw1 stability. Additionally, this study shows that K113 is also likely involved in this process. However, it appears that K113 can be either acetylated or ubiquitinated, and it is, thus, less clear if one of the two modifications or both modifications is critical at this residue. Additional experiments may be required to answer this question. This study would have benefited from an additional discussion on the results related to the modification of K113.

The authors demonstrate that overexpression of ISW1 in select clinical isolates of Cryptococcus increases sensitivity to antifungals. However, these experiments would have benefited from additional controls, such as including overexpression of ISW1 in the wild-type strain (H99) and antifungal-sensitive isolate (CDLC120).

Reviewer #3 (Public Review):

This study focuses on the role of the chromatin remodeller ISWI in Cryptococcus. The authors show that a) ISWI modulates Cryptococcus' ability to grow in the presence of antifungal drugs and b) ISWI post-translational modifications (Acetylation and Ubiquitination) regulate ISWI protein stability. The observation that post-translational modifications regulate ISWI activity and stability is exciting and it could unveil novel mechanisms to rapidly and reversibly regulate the response to antifungal drug treatments. However, the study lacks a fundamental characterisation of ISWI. This information is essential to understand the mechanistic regulations of ISWI in Cryptococcus and how it mediates drug response. The following are questions that should be addressed:

1. ISWI chromatin remodellers are well-characterised in many organisms. How many ISWI proteins does Cryptococcus contain? Why did the authors focus on ISWI?<br /> 2. What is the ISWI protein complex(es)? The Mass-Spec analysis should reveal this.<br /> 3. Is Cryptococcus ISWI a transcriptional activator or repressor?<br /> 4. Is ISWI function in drug resistance linked to its chromatin remodelling activity?<br /> 5. Does ISWI interact with chromatin? If so, which are ISWI-target genes? Does drug treatment modulate chromatin binding?

Reviewer #1 (Public Review):

In this manuscript the authors overproduce two M. smegmatis DNA polymerases, DinB2 and DinB3, as a way to determine whether they may contribute to DNA damage tolerance and/or mutagenesis; the roles of these DNA polymerases in DNA damage tolerance and mutagenesis is currently unknown. The authors show that overproduced levels of DinB2, but not DinB3, impeded growth, and this inhibition was relieved by the disruption of DinB2 catalytic activity using the DinB-D109A mutation. They further demonstrate that the overproduction of DinB2 contributed to frameshift mutagenesis, while DinB3 did not. The contribution of overproduced levels of DinB2 to frameshift mutagenesis was studied in a careful and systematic way, convincingly showing that frameshifts correlated with DinB2 slipping while replicating homopolymeric nucleotide runs during dNTP and not rNTP incorporation. The authors also show that the metal cofactor (Mn vs Mg) contributes to the mutagenic behavior of DinB2. While this work is mostly compelling, the major concern is it fails to address the contribution of DinB2 and DinB3 to DNA damage tolerance and mutagenesis when they are expressed at normal levels from their respective chromosomal loci.

Reviewer #2 (Public Review):

The role of the family IV polymerases in mycobacteria is only partly understood. In this work, the authors investigate the role of the M. smegmatis DinB2 and DinB3 polymerases by a combination of biochemical analysis of enzyme activity in vitro and mutational and phenotypic characterization of M. smegmatis strains during induced over-expression of these proteins. They show both polymerases to be mutagenic and uncover a distinct role for DinB2 in slippage on homopolymeric tracts that is dependent on manganese.

Previous work showed that DinB1 overexpression resulted in SOS induction. This work shows that DinB2 and DinB3 similarly increase RecA levels. Previous work also showed that DinB1 overexpression resulted in growth inhibition and loss of viability which was independent of its polymerase activity. In this work, overexpression on DinB2 but not DinB3 inhibits growth along with a loss in viability but in contrast to DinB1, this inhibitory effect is only seen with a polymerase-proficient enzyme and is even more enhanced in a steric gate mutant. Overexpression of DinB3 and DinB2 increases the frequency of Rif-resistant mutants independent of the SOS response and DnaE2. The mutation spectrum in DinB2-overexpressing cells was distinct from that caused by DinB1 or DinB3 overexpression. In vitro and in vivo experiments clearly demonstrate that DinB2 catalyzes frameshift mutagenesis on substrates with homopolymeric nucleotide stretches demonstrating enhanced slippage compared to the recent data with DinB1. Remarkably, this slippage is enhanced on homopolymeric runs of purines than pyrimidines in vitro. In vivo slippage by DinB2 was not enhanced by long G runs. The slippage in vitro was only evident in its DNA-dependent DNA polymerase mode and not during ribonucleotide incorporation. In addition, while magnesium alone was associated with mis-addition, the presence of manganese shifted the enzyme to slippage mode in vitro. The detrimental effect of DnaB2 over-expression on viability is, however, not related to its slippage activity since conditions that enhance slippage in vitro (specifically manganese) are associated with a greater detrimental effect on viability in vivo despite a lack of evidence of slippage using reporter constructs.

Reviewer #3 (Public Review):

The work from Dupuy et al aims to characterize the mutagenic effects of two DinB homologs of Mycobacteria, DinB2, and DinB3. The manuscript shows solid and convincing biochemical data about slippage promoted by DinB 2 on various homopolymeric templates. Overall, this study makes a solid contribution to the understanding of the properties of polymerases from the different DinB subfamilies of bacteria, although some points of the in vivo experiments should be critically evaluated by the readers as described below.

In vivo DinB2 is the more mutagenic of the two and is toxic when overexpressed. Nevertheless, these results are obtained with the overexpression of the polymerases and should be interpreted with caution. In this sense, it would have been interesting to have a quantification of how much overexpression the plasmids constructs achieve in the conditions used in the experiments, for a better assessment of the relevance of the data. For example, a physiological 10-fold increase in the expression of DinB2 is mentioned in the discussion - would that be close to what is achieved with plasmid-based overexpression?

The finding of kanR CFUs without any detectable mutations in the kan marker is worrisome and should be better discussed in the text. The same for sacB data in supplementary material. The explanation given in lines 216-218 does not make sense. Markers 7G and 8G clearly are barely measuring any mutagenesis. I think that the experiments in which most of the supposed KanR revertants actually have no Kan mutation should either be removed from the manuscript or better discussed, because it is uncertain what they are measuring, therefore no conclusion can be drawn from them. For the Kan markers, one possible explanation is that translational frameshifts are occurring and allow residual growth of some of the cells. Gene amplification as seen in the lac system of Cairns and Foster in E. coli could also promote growth without actual mutations. Is the KanR phenotype of these colonies heritable and stable?

Also, spontaneous mutagenesis should have been more precisely measured by using fluctuation analysis of larger sample sizes. In many instances, the results shown are the means of a few cultures with very large differences in mutant frequencies (several hundred-fold - e.g. Figures 4C, D and E, 5C and F, S3). Authors could discuss/explain their choice of statistical analysis and sample sizes.

Reviewer #1 (Public Review):

Damon-Soubeyrand and colleagues use 3DISCO tissue clearing and light-sheet microscopy to provide a detailed atlas of the blood and lymphatic circulating networks of the mouse epididymis. While this manuscript does not address the function of these networks during the development or homeostasis of the epididymis, it is an outstanding example of a descriptive study that paves the way towards functional investigations of the role of epididymal vasculature in the post-testicular maturation of spermatozoa.

Strengths: The authors used a wide range of markers to carefully assess the differential patterns of epididymal blood and lymphatic vasculature, and elegantly describe each image in great detail. Where possible, the authors used appropriate quantitative methods to support their descriptive data, which are useful metrics for readers seeking to characterize vascular and lymphatic networks in disease models.

Weaknesses: In its current form, it is unclear which of the elements presented in the manuscript are novel discoveries about the blood and lymphatic networks of the epididymis, as the text lacks concise and precise statements about the major findings of the study. In addition, the authors frame this study of the vasculature as a way to understand the immune context of spermatozoa in the epididymis but do not integrate their data on blood and lymphatic networks with the immune system.

Reviewer #2 (Public Review):

This manuscript illustrates a vascular network in the postnatal developing and adult epididymis using high-resolution three-dimensional (3D) imaging and organ clearing coupled with multiplex immunodetections of lymphatic and blood markers.

Strengths:<br /> The cutting-edge imaging technique to visualize the three-dimensional vascular network.<br /> The images and videos were of great quality.<br /> The authors were very cautious and careful when interpreting the results of marker immunostaining.

Weaknesses:<br /> 1. Although the images and videos were of great quality, the results derived from them provided little new knowledge and few conceptual insights into male reproductive tract biology and basically confirmed what has been published using traditional methods. For example, the high intensity of the vascular network in the initial segment was previously reported by Abe in 1984 and Suzuki in 1982; the pattern of the major lymphatic vessel and drainage was beautifully depicted by Perez-Clavier, 1982.

2. The authors were very cautious when interpreting the results of marker immunostaining however these markers were not specific for a definite cell type. For example, as the authors stated, VEGFR3 marks both lymphatic vessels and fenestrated blood vessels. how could the authors claim the VEGFR3+ network was lymphatic? The authors claimed that they used three markers for the lymphatic vessel. But staining results of the networks were very different. How could the author make conclusions about the network of lymphatic vessels in the epididymis?

3. To understand the vascular network development in the epididymis, would the authors please look at the fetal stage when the vascular network is established in the first place? Wolffian duct tissues are much smaller and thinner and would be amenable for 3D imaging probably even without clearing.

4. Immunofluorescence staining of VEGF factors was not convincing. As a secreted factor, VEGF will be secreted out of the cells, would it be detected more in the interstitium? I am always skeptical about the results of immunostaining secreted growth factors. Would it be possible to perform in situ or RNAscope to confirm the spatial expression pattern of VEGFs?

5. The study is descriptive and does not provide functional and mechanistic insights. Maybe, the combination of 3D imaging with lineage tracing of endothelium cells or ligation study (removal/ligation of the certain vessel) would help better understand how the vascular network is established and their functional significance.

6. Immune response is among many physiological processes in which vascular networks play significant roles. Discussion would be needed in other physiological processes, such as tissue metabolism and stem/progenitor cell niche microenvironment.

7. How could the author determine the Cd-A labeled vessel in Fig 1 was an artery, not a vein? This leads to another critical question. Would it be possible to stain with artery and vein markers to help illustrate the blood flow directions of the vessel?

Reviewer #1 (Public Review):

This study aims at investigating temporal variation in patterns of germline mutation during the evolution of human populations. For this purpose, the authors analyzed polymorphism data from the 1000 Genomes project. They inferred the age of each derived variant using Relate, a newly developed method that reconstructs local genealogies based on phased haplotype sequences and estimates allele ages (Speidel et al. 2019).

Speidel et al. (2019) already had used their method to explore temporal variation in mutation patterns. Their analysis had confirmed the transient elevation in non-CpG C>T mutations in Europeans compared to African and Est Asians previously described by Harris (2015). However, Speidel et al. did not push their study very far, notably because of the difficulty of distinguishing the effects of changes in mutation patterns from those of GC-biased gene conversion (gBGC).

Here Gao et al. carefully accounted for gBGC to further explore variation in mutation patterns. As expected, they confirmed the previously described European-specific mutational shift. In addition, they identified two novel interpopulation differences in the mutation spectrum. This suggests that shifts in mutation spectra occur frequently, over a few thousands of generations. The reasons (environmental or genetic) for these recurrent shifts are not known, but the authors convincingly show that they cannot be explained by changes in the age of reproduction over time.

I found this manuscript very well written and very interesting. There is however an important point in their results that seems very puzzling. Indeed, the authors report that among mutations that are estimated to be old (>28800 generations), the ratio of T>C over T>G differs significantly in African samples compared to non-African samples (Fig. 2A). This difference is unexpected given that these old mutations largely predate the out-of-Africa migration (<3000 generations), and hence are a priori expected to be largely shared across populations. Curiously, this pattern is driven by variants for which the derived allele is observed in both Neanderthals and Denisovans (ND11 variants) (while ND01 and ND10 variants do not contribute to this pattern; Fig. 2D, SupFig 2.8). The authors hypothesize that the T>C/T>G ratio was higher in one or more populations in the remote past and those ancient groups contribute variable amounts of ancestry to contemporary populations. However, I do not understand how this model can account for the fact that ND10 and ND01 variants behave differently from ND11 variants (ND10 and ND01 variants are also expected to be emerged prior to the split of modern humans and archaic hominins).

It is possible that I misunderstood something, but in any case, there are several points in the methodology that have to be clarified. Notably, it is not clear to me if the reported pattern is driven by variants that are specific to the African samples, or if it is also observed among variants that are shared across populations. Furthermore, I suspect that polarization errors (notably at CpG sites) might be responsible for this pattern.

In summary, this manuscript reports very interesting observations, but several additional tests have to be done to check whether they are real or if they might result from methodological artefacts.

Reviewer #2 (Public Review):

This manuscript reassesses the strength of evidence for rapid human germline mutation spectrum evolution, using high coverage whole genome sequencing data and paying particular attention to the potential impact of confounders like biased gene conversion. The authors also refute some recently published arguments that historical changes in the age of reproduction might explain the existence of such mutation spectrum changes. My overall impression is that the paper presents a useful new angle for studying mutation spectrum evolution, and the analysis is nicely suited to addressing whether a particular model such as the parental age model can explain a set of observed polymorphism data. My main criticism is that the paper overstates certain weaknesses of previously published papers on mutation spectrum evolution as well as the generation time hypothesis; correcting these oversimplifications would more accurately capture what the paper's new analyses add to the state of knowledge in these areas.

As part of the motivation for the current study, the introduction states in lines 97-99 that "it thus remains unclear if the numerous observed [mutation spectrum] differences across human populations stem from rapid evolution of the mutation process itself, other evolutionary processes, or technical factors." This seems to overstate the uncertainty that existed prior to this study, given that Speidel, et al. 2021 found elevated TCC>TTC fractions in ancient genomes from a specific ancient European population, which seems like pretty airtight evidence that this historical mutation rate increase really happened. In addition, earlier papers (Harris 2015, Mathieson & Reich 2016, Harris & Pritchard 2017) already presented analyses rejecting the hypothesis that biased gene conversion or genetic drift could explain the reported patterns-in fact, the Mathieson & Reich paper reports one mutation spectrum difference between populations that they conclude is an artifact caused by the Native American population bottleneck, but they conclude that other mutation spectrum differences appear more robust. As the authors acknowledge in the discussion of their own results, biased gene conversion and non-equilibrium demography are difficult confounders to deal with, and neither previous papers nor the current paper are able to do this in a way that is 100% foolproof. The current manuscript makes a valuable contribution by presenting new ways of dealing with these issues, particularly since previous papers' work on this topic was often confined to supplementary material, but it seems appropriate to acknowledge that earlier papers discussed the potential impacts of biased gene conversion and demographic complexity and presented their own analyses arguing that these phenomena were poor explanations for the existence of mutation spectrum differences between populations.

For the most part, I found the paper's introduction to be a useful summary of previous work, but there are a few additional places where the limitations of previous work could be described more clearly. I'd suggest noting that the data artifacts discovered by Anderson-Trocmé, et al. were restricted to a few old samples and that the large differences the current manuscript focuses on were never implicated as potential cell line artifacts. In addition, when the authors mention that their new approach includes "minimiz[ing] confounding effects of selection by removing constrained regions and known targets of selection" (lines 106-107), they should note that earlier papers like Harris & Pritchard 2017 also excluded conserved regions and exons.

One innovative aspect of the current paper's approach is the use of allele ages inferred by Relate, which certainly has advantages over using allele frequencies as a proxy for allele age. Though the authors of Relate previously used this approach to study mutation spectrum evolution, they did not perform such a thorough investigation of ancient alleles and collapsed mutation type ratios. I like the authors' approach of building uncertainty into the use of Relate's age estimates, but I wonder about the validity of assuming that the allele age posterior probability is distributed uniformly between the upper and lower confidence bounds. Can the authors address why this is more appropriate than some kind of peaked distribution like a beta distribution?<br /> I would also argue that the statement on line 104 about Relate's reliability is not yet supported by data-there is certainly value in using Relate ages to investigate mutation spectrum change over time and compare this to what has been seen using allele frequencies, but I don't think we know enough yet to say that the Relate ages are definitely more reliable. Relate's estimates might be biased by the same processes like selection and demography that make allele frequencies challenging to interpret. The paper's statements about the limitations of allele frequencies are fair, but there is always a tradeoff between the clear drawbacks of simple summary statistics and the more cryptic possible blind spots of complicated "black box" algorithms (in the case of Relate, an MCMC that needs to converge properly). DeWitt, et al. 2021 noted that the demographic history inferred by Relate doesn't accurately predict the underlying data's site frequency spectrum, indicating that the associated allele ages might have some problems that need to be better characterized. While testing Relate for biases is beyond the scope of this work, the introduction should acknowledge that the accuracy and precision of its time estimates are still somewhat uncertain.

The paper's results on C>T mutations in Europeans versus Africans are a nice confirmation of previous results, including the observation from Mathieson & Reich that neither SBS7 nor SBS11 is a good match for the mutational signature at play. More novel is the ancient mutational signature enriched in Africa and the interrogation of the ability of parental age to explain the observed patterns. I just have a few minor suggestions regarding these analyses:

1. I like the idea of using maternal age C>G hotspots to test the plausibility of the maternal age as an explanatory factor, but I think this would be more convincing with the addition of a power analysis. Given two populations that have average maternal ages of 20 and 40, and the same population sample sizes available from 1000 Genomes, can the authors calculate whether the results they'd predict are any different from what is observed (i.e. no significant differences within the maternal hotspots and significant differences outside of these regions)?

2. Is it possible that the T>C/T>G ratio is elevated in all variants above a certain age but shows up as an African-specific signal because the African population retains more segregating variation in this age range, whereas non-African populations have fixed or lost more of this variation? Since Durvasula & Sankararaman identified putative tracts of of super-archaic introgression within Africans, is it possible to test whether the mutation spectrum signal is enriched within those tracts?

3. Although Coll Macià, et al. argued that generation time is capable of explaining all mutation spectrum differences between populations, including the excess of TCC>TTC in Europeans, Wang et al. argue something slightly different. They exclude TCC>TTC and the other major components of the European signature from their analysis and then argue that parental age can explain the rest of the differences between populations. I think the analysis in this paper convincingly refutes the Coll Macià, et al. argument, but refuting the Wang, et al. version would require excluding the same mutation types that are excluded in that paper.

Reviewer #1 (Public Review):

Inter-cellular mitochondria transfer has been observed in many systems but the role or relevance of transferred mitochondria in recipient cells is poorly defined in contexts where recipient cells have intact functional networks. This manuscript directly addresses this important question and present a model in which transferred mitochondria act as signaling organelles to increase cancer cell proliferation.

The authors present compelling evidence that macrophages transfer mitochondria to cancer cells. Activated macrophages transfer mitochondria more effectively than non-activated macrophages, and this increased transfer is at least in part due to enhanced mitochondrial fragmentation in activated macrophages. Probing the significance of mitochondrial transfer, the authors find that transferred mitochondria remain distinct from endogenous mitochondrial networks and do not exhibit the polarization that traditionally characterizes functional mitochondria. The transferred mitochondria have features consistent with elevated oxidative stress and/or ROS production. A series of elegant imaging experiments demonstrate that mitochondrial transfer is associated with increased growth in daughter cells that inherit transferred mitochondria. Mechanistically, the authors propose that ROS produced by transferred mitochondria stimulate ERK signaling to induce a proliferation advantage.

Overall the work addresses an important question regarding the functional role of mitochondria transferred to cancer cells. The data largely support the model that transferred mitochondria are defective and induce proliferation in recipient cells. Some clarification on the effect timescales and the role role of ROS and ERK signaling in cell proliferation in cells that do not receive mitochondria is warranted. Overall this work provides an important new view for how mitochondrial transfer affects cell biology and provides a suite of tools and protocols for quantifying the impact of mitochondrial transfer on recipient cells.

Reviewer #2 (Public Review):

In these studies, the authors make the observation that macrophages transfer their mitochondria to cancer cells. The authors claim that these mitochondria are dysfunctional and release reactive oxygen species (ROS) in the recipient cancer cells. Further, the authors illustrate that the mitochondrial-derived ROS activates proliferative ERK signaling. Macrophage mitochondria exhibit fragmentation, the extent of which promotes their transfer to cancer cells resulting in a functional increase in cancer cell proliferation. The authors initiated this work based on their previous findings where they illustrated the ability of macrophages to transfer cytosolic contents to recipient cancer cells.

The observations made in this manuscript, if further substantiated, are of interest in the field of cancer immunotherapy, metabolism, and basic cancer biology.

Reviewer #3 (Public Review):

In this manuscript, Kidwell & Casalini, et al. use cell biology and functional approaches to investigate the dynamics and consequences mitochondrial transfer from macrophages to breast cancer cells. Unlike prior studies that emphasize the metabolic benefits of mitochondrial reconstitution in cells with defective mitochondrial DNA, they ask how mitochondrial transfer affects breast cancer cells with intact mitochondria. They observe that macrophage co-culture or "bathing" breast cancer cells in isolated mitochondria from macrophages results in low frequency mitochondrial transfer, which increases cell cycling, ERK signaling, and cell proliferation rate of recipient cells. Interestingly, fluorescent dyes and sensors were used to determine that transferred mitochondria had low mitochondrial membrane potential and were highly oxidized, suggesting dysfunctional mitochondria with elevated ROS. In addition, activation of mitochondrial ROS by photobleaching a region of mitochondria in cells expressing mito-KillerRed was sufficient to similarly increase cell cycling, and mitochondrial targeted antioxidants could mitigate the proliferative benefits of mitochondrial transfer. Finally, the authors used several in vitro and in vivo models to demonstrate that M2-like macrophages had more fragmented mitochondria, had higher mitochondrial transfer rates, and promoted cell cycling in tumors.

Overall, a strength of the study is the usage of creative cell biology techniques and rigorous mouse models to provide compelling support for their primary claims, many of which go against the grain of current thinking in mitochondrial transfer research. While the discrepancies with the literature are by no means the fault of the authors, this study could nonetheless improve its reach by directly seeking resolution to these differences. In addition, the study raises some important questions how mitochondrial ROS from transferred dysfunctional mitochondria might be beneficial and at what doses, which should be further investigated to contextualize the findings.

Reviewer #1 (Public Review):

The present study investigates the anatomical connectivity between Mu opioid receptor (MOR) expressing neurons of the pontine respiratory group with down-stream targets of the respiratory network in the medulla oblongata. The study employs a variety of viral tracing approaches, optogenetic stimulation of pre-synapses of descending pontine projection neurons, and patch clamp electrophysiology. Overall the study is well conducted and the authors show that MOR expressing excitatory glutamatergic pontine neurons project to the medullary respiratory rhythm generator and adjacent ventral respiratory group. The study implies that opioids act on MOR-located somata and dendrites of the pontine and medullary respiratory groups. Importantly MOR are expressed on the pre-synapses of the descending pontine projections neurons. The authors, therefore, propose that opioids mediate respiratory depression via distinct pre- and post- synaptic mechanisms across inter-connected ponto-medullary respiratory neurons. The study advances our knowledge of network mechanisms that mediate opioid respiratory depression and may provide interesting frameworks for the development of therapies to counteract or prevent opioid respiratory depression. The study is of broad interest to the respiratory control research community, as well as medically relevant.

Reviewer #2 (Public Review):

This study identifies the neural circuits inhibited by activation of opioid receptors using complex experimental approaches such as electrophysiology, pharmacology, and optogenetics and combined them with retrograde and anterograde tracings. The authors characterize two key regions of the brainstem, the preBötzinger Complex, and the Kolliker-Fuse, and how these neuronal populations interact. Understanding the interactions of these circuits substantially increases our understanding of the neural circuits sensitive to opioid drugs which are critical to understand how opioids act on breathing and potentially design new therapies.

Major strengths.<br /> This study maps the excitatory projections from the Kolliker-Fuse to the preBötzinger Complex and rostral ventral respiratory group and shows that these projections are inhibited by opioid drugs. These Kolliker-Fuse neurons express FoxP2, but not the calcitonin gene-related peptide, which distinguishes them from parabrachial neurons. In addition, the preBötzinger Complex is also hyperpolarized by opioid drugs. The experiments performed by the authors are challenging, complex, and the most appropriate types of approaches to understanding pre- and post-synaptic mechanisms, which cannot be studied in vivo. These experiments also used complex tracing methods using adenoassociated virus and cre-lox recombinase approaches.

Limitations.<br /> (1) The roles of the mechanisms identified in this study have not been established in models recording opioid-induced respiratory depression or respiratory activity. This study does not record, modulate, or assess respiratory activity in-vitro or in-vivo, without or with opioid drugs such as fentanyl or morphine.<br /> (2) Experiments are performed in-vitro which do not mimic the effects of opioids observed in-vivo or in freely-moving animals. However, identification of pre- and post- synaptic mechanisms, as well as projections, cannot be performed in-vivo, so the authors use the right approaches for their experiments.<br /> (3) The type of neurons projecting from KP to preBötzinger Complex or ventral respiratory group have not been identified. Although some of these cells are glutamatergic, optogenetic experiments could have been performed in other cre-expressing cell populations, such as neurokinin-1 receptors.

This study provides new insights into the types of circuits inhibited by opioid drugs, and the site of actions of inhibition, such as pre- or post-synaptic, and proposes how inhibition by opioids acts at multiple sites in the brainstem through various mechanisms.

Although many studies have recently explored the types of neurons and sites in the brain sensitive to opioids, the present study is the first to provide a clear picture of the neuronal mechanism underlying inhibition by opioids. Importantly, it provides a link between two sites known to inhibit breathing when inhibited by opioids. The results provided here combined with a complex methodology support the various conclusions reached by the authors.

Reviewer #3 (Public Review):

This manuscript reveals opioid suppression of breathing could occur via multiple mechanisms and at multiple sites in the pontomedullary respiratory network. The authors show that opioids inhibit an excitatory pontomedullary respiratory circuit via three mechanisms: 1) postsynaptic MOR-mediated hyperpolarization of KF neurons that project to the ventrolateral medulla, 2) presynaptic MOR mediated inhibition of glutamate release from dorsolateral pontine terminals onto excitatory preBötC and rVRG neurons, and 3) postsynaptic MOR-mediated hyperpolarization of the preBötC and rVRG neurons that receive pontine glutamatergic input.

This manuscript describes in detail a useful method for dissecting the relationship between the dorsolateral pons and the rostral medulla, which will be useful for various researchers. It's also great to see how many different methods have been applied to improve the accuracy of the results.

1. Relationship between the dorsolateral pons and rostral ventrolateral medulla.

The method of this paper is a good paper to show a very precise relationship between the presence of opioid receptors and the dorsolateral pons and rostral ventrolateral medulla, and for opioid receptors, based on the expression of Oprm1, the use of genetically modified mice with anterograde or retrograde viruses with additional fluorescent colors showed both anterograde and retrograde projections, revealing a relationship between the dorsolateral pons and rostral ventrolateral medulla.

For example, to visualize dorsal pontine neurons expressing Oprm1, Oprm1Cre/Cre mice were crossed with Ai9tdTomato Cre reporter mice to generate Ai9tdT/+ oprm1Cre/+ mice (Oprm1Cre/tdT mice) expressing tdTomato on neurons that also express MOR at any point during development, and the retrograde virus encoding Cre-dependent expression of GFP (retrograde AAV-hSIN-DIO-eGFP was injected into the respiratory center of Oprm1Cre/+ mice and into the ventral respiratory neuron group, showing that KF neurons expressing Oprm1 project to the respiration-related nucleus of the ventrolateral medulla.

However, although the authors have also corrected it, the virus may spread to other places as well as where they thought it would be injected, and it is important to note that it is injected accordingly to mark the injection site with an anterograde virus encoding a different fluorescent color mCherry, and the extent of the injection is quantified, which is excellent as a control experiment.

In addition, the respiratory center seems to be related not only to preBötC but also to pFRG recently, so if the relation with it is described, it is important from the viewpoint of the effect on the respiratory center and the effect on the rhythm.

2. Electrophysiological approaches and useful methods for target neurons

Oprm1Cre/+ mice), the authors found abundant Oprm1 + projections in the preBötC region of the medulla oblongata (respiratory center) and sought to determine whether presynaptic opioid receptors inhibit glutamate release from KF terminals to excitatory preBötC and rVRG neurons, since KF neurons in the dorsolateral pons projecting to the ventrolateral medulla oblongata had been shown to be glutamatergic and to have opioid receptors. The authors injected a channelrhodopsin-2-encoding virus (AAV2-hSin-hChR2 (H134R) -EYFP-WPRE-PA) into the dorsolateral pontine KF of vglu2Cre / tdT mice and performed whole-cell voltage-clamp recordings from td tomato-expressing, excitatory vglu2-expressing preBötC and rVRG neurons, contained in acute brain slices. Moreover, both opioid-sensitive and opioid-insensitive KF neurons that project to preBötC and rVRG were visible and recorded using FluoSpheres which are much more visible in acute brain sections than retrograde tracers of viruses.

1) Optogenetic stimulation of the KF terminus was blocked by the AMPA-type glutamate receptor antagonist DNQX. In excitatory pre-BötC and rVRG neurons, the terminals from the dorsal pontine KF were activated by optogenetic stimulation, and the KF synapses to the medullary respiratory neurons were found to be monosynaptic because oEPSCs(optical stimulated EPSCs) were removed by TTX but were subsequently restored by the application of K-channel blocker 4AP. Thus, KF neurons have been shown to send monosynaptic glutamatergic projections to excitatory ventrolateral medullary neurons using terminal optogenetic stimulation and receptor and channel inhibitors.

2) To determine whether opioids inhibit glutamate release from KF terminals to medullary respiratory neurons, we recorded a pair of oEPSCs (50 ms stimulus interval) from excitatory preBötC and rVRG neurons and applied an endogenous opioid agonist, [Met5] enkephalin (ME), to the perfusion solution. ME is preBötC and rVRG neurons, indicating inhibition of glutamate release by presynaptic MOR PPR. Thus, presynaptic opioid receptors have been shown electrophysiologically to inhibit glutamate release from KF terminals to excitatory pre-BötC and rVRG neurons.

3) Whether excitatory pre-BötC or rVRG neurons themselves receiving opioid-sensitive glutamatergic synaptic inputs from KF are hyperpolarized by opioids can be determined by monitoring their retention currents.

4) Since FluoSpheres are much more visible in acute brain sections than retrograde tracers of viruses and do not spread to injection sites, they chose to record from retrogradely labeled KF neurons with FluoSpheres injected into preBötC or rVRG in wild-type mice, allowing us to label KF neurons regardless of Oprm1 expression status and determine the projection patterns of both Oprm1 + and Oprm1- neurons. Whole-cell voltage-clamp recordings from fluorescent KF neurons contained in acute brain slices show that the presence of ME-mediated outward current can identify KF neurons that express functional MORs and are opioid-sensitive compared to neurons that lack ME-mediated outward current (insensitive). This suggests that both opioid-sensitive and opioid-insensitive KF neurons project to preBötC and rVRG.

Although much has been written about the relationship between KF neurons and medulla oblongata neurons and their being glutaminergic neurons, detailed descriptions of the recorded neuronal firing patterns are lacking. You should describe what firing pattern the recorded neurons had. If we don't do that, we won't be able to tell whether it's a respiratory neuron or another tonic firing neuron, so I don't think we can discuss whether it's involved in the respiratory rhythm.

3. Compare the distribution of neurons

To examine the distribution of Oprm1 + and Oprm1- dorsolateral pontine neurons projecting to the ventrolateral medulla, we injected retrograde AAV-hSin-DIO-eGFP and retrograde AAV-hSin-mCherry into preBötC and rVRG of Oprm1Cre/+ mice and found a neuronal distribution in which Oprm1-expressing projection neurons expressed GFP and mCherry, but not Oprm1-expressing projection neurons expressed only mCherry.

In addition, rostral glutamatergic KF neurons express FoxP2, while MOR-expressing glutamatergic neurons in the lateral parabrachial region that project to the forebrain express the CGRP-encoding gene, Calca. In view of this, the authors performed immunohistochemistry for FoxP2 and CGRP on Oprm1 + KF neurons projecting to the ventrolateral medulla, and Oprm1 + medulla oblongata projecting KF neurons expressed FoxP2 but not CGRP. The expression of CGRP was not observed in rostral KF and medullary projection Oprm1 + neurons and neurites but was strong in lateral parabrachial neurons and their axonal fiber projections. Can you describe the relationship between CGRP and FoxP2 and recorded neurons?

Reviewer #1 (Public Review):

Many previous studies have examined the regulation of hyphal growth in vitro, and have identified about 1,000 genes capable of influencing this process. However, a weakness is that most of these genes have weak effects and are not important in vivo. Therefore, it is very significant that this is the first large-scale study to examine the regulation of hyphal growth in vivo by analyzing a set of 156 transcription factor mutants in mice. A strength of these innovative studies is that mutant strains were injected into a mouse ear, which permitted the use of high-resolution microscopy to quantify the fraction of cells forming hyphae and the rate of hyphal elongation. Furthermore, wild-type cells were co-injected to serve as an internal control, which enhanced the rigor of these studies.

One major conclusion is that three core transcription factors were identified as being important in vivo (Rob1, Brg1, and Efg1) and two negative regulators (Tup1 and Efg1). Previously, many transcription factors were found to be important in vitro, so this is important for focusing future studies on the key regulators. Nanostring gene expression studies verified that these core factors regulate overlapping but distinct sets of genes in vitro and in vivo, which reinforces the importance of carrying out studies in vivo. Additional mutants were discovered to have minor defects in filamentous growth and were considered to be ancillary factors that act in concert with the core regulators.

Another innovative aspect of the manuscript is that they examined the rate of hyphal elongation in vivo. This is an understudied area both in vitro and in vivo. Transcription factors UME6, LYS14, and HMS1 were shown to regulate the elongation rate, which opens up new opportunities to study the mechanisms. Consistent with this, these transcription factors were shown to regulate a set of genes that is distinct from those regulated by transcription factors that control the initiation of hyphal growth.

Genetic approaches (complex haploinsufficiency) were used to examine the relationship between the core factors and the ancillary factor TEC1. Interestingly, these results revealed genetic interactions between TEC1 and the core factors EFG1 and BRG1, including their ability to regulate other transcription factors. This shows how these complex networks are functioning in vivo.

Another major advance was that the in vivo analysis of the two negative regulators of hyphal growth (Nrg1 and Tup1) revealed a new model for how they interact with the master transcriptional regulator Efg1. The results indicate that the major function of Efg1 in vivo is to mediate relief of Nrg1 repression. It was not needed to regulate the expression of hypha-induced genes.

Reviewer #2 (Public Review):

This manuscript is focused on the identification and characterization of transcriptional networks that control the major Candida albicans virulence property of filamentation during infection in vivo. Using an intravital imaging assay, the authors have screened a C. albicans transcription factor mutant library to identify factors important for controlling both filament initiation and elongation in vivo. They also perform Nanostring experiments to identify the in vivo transcriptional profiles of genes controlled by specific key factors in the network. Overall, the authors identify three positive and two negative core factors important for the initiation of filamentation and several factors specifically important for filament elongation (including 4 factors whose mutants have no in vitro elongation phenotypes). Target genes associated with filament initiation and elongation were shown to be mostly distinct. Unexpectedly, the authors also show that the main role of Efg1, a major positive regulator of filamentation, is to mediate relief of repression by Nrg1.

Overall, the manuscript is well-written and the data are clearly presented. In addition, the authors clearly appear to have achieved their Aim of identifying and characterizing transcriptional networks that regulate C. albicans morphogenesis during infection in vivo. In general, the conclusions of this paper are well-supported by the results. The results of this study are likely to have a significant impact on the field for several reasons: 1) new and valuable information will be provided about transcriptional networks that control C. albicans filamentation in vivo, 2) this study describes an important distinction between genes associated with filament initiation and elongation and will be the first to systematically analyze C. albicans genes associated with filament elongation, 3) while there are similarities, the authors also observe several important differences between transcriptional networks that control C. albicans filamentation in vivo vs. in vitro, which will help to clarify regulation that actually occurs during infection, 4) as indicated above, a new and surprising role for the C. albicans master regulator of filamentation, Efg1, is reported, 5) because filamentation is an important C. albicans virulence property, several of the target genes of transcription factor networks identified by this study (and the factors themselves) could serve as potential targets for new antifungals. As a consequence, this study is likely to provide information that opens up new and useful lines of research for the field.

Strengths:<br /> 1. Intravital imaging allows for the identification of transcription factors specifically important for C. albicans filamentation during infection.<br /> 2. Distinct sets of C. albicans genes and factors associated with filament initiation vs. elongation are identified.<br /> 3. Key differences between in vivo and in vitro transcriptional regulation of C. albicans filamentation are demonstrated, which in some cases challenge current paradigms. This also highlights the effect of the environment in determining target genes.<br /> 4. Evidence is presented to suggest that Efg1 promotes C. albicans filamentation primarily through relief of Nrg1 repression.

Weaknesses:<br /> 1. Nanostring does not profile the complete set of C. albicans genes, but rather a subset that is pre-selected. Therefore, defining proportions of genes and gene classes controlled by specific transcription factors may not give the complete picture and may not be accurate with respect to the transcriptome as a whole.<br /> 2. As the authors have noticed, transcription factors and target genes associated with C. albicans filamentation may vary significantly depending on the environment. It is therefore unclear whether the in vivo gene expression patterns observed in this study apply to other host niches besides the ear.<br /> 3. Similarly, variations in filamentation-associated transcription factors and target genes may occur in the "in vitro" conditions used by the authors. RPMI + 10% serum is the main "in vitro" condition but many other conditions are known to drive C. albicans filamentation.<br /> 4. Lines 361-366: A clear rationale for additional TFs to study in more detail was not provided.<br /> 5. Post-translational mechanisms, particularly septin phosphorylation, are likely to have an important effect on filament elongation (see work from Yue Wang's lab), which was not discussed.<br /> 6. Many Nrg1 targets are known to also be Tup1 targets (Kadosh & Johnson, 2005), which counters the argument that this corepressor and DNA-binding protein function separately.<br /> 7. While useful, examining genetic interactions using haploinsufficiency has several limitations and certain interactions may escape detection.

Reviewer #1 (Public Review):

Taking advantage of a publicly available dataset, neuronal responses in both the visual and hippocampal areas to passive presentation of a movie are analyzed in this manuscript. Since the visual responses have been described in a number of previous studies (e.g., see Refs. 11-13), the value of this manuscript lies mostly on the hippocampal responses, especially in the context of how hippocampal neurons encode episodic memories. Previous human studies show that hippocampal neurons display selective responses to short (5 s) video clips (e.g. see Gelbard-Sagiv et al, Science 322: 96-101, 2008). The hippocampal responses in head-fixed mice to a longer (30 s) movie as studied in this manuscript could potentially offer important evidence that the rodent hippocampus encodes visual episodes.

The analysis strategy is mostly well designed and executed. A number of factors and controls, including baseline firing, locomotion, frame-to-frame visual content variation, are carefully considered. The inclusion of neuronal responses to scrambled movie frames in the analysis is a powerful method to reveal the modulation of a key element in episodic events, temporal continuity, on the hippocampal activity. The properties of movie fields are comprehensively characterized in the manuscript.

Although the hippocampal movie fields appear to be weaker than the visual ones (Fig. 2g, Ext. Fig. 6b), the existence of consistent hippocampal responses to movie frames is supported by the data shown. Interestingly, in my opinion, a strong piece of evidence for this is a "negative" result presented in Ext. Fig. 13c, which shows higher than chance-level correlations in hippocampal responses to same scrambled frames between even and odd trials (and higher than correlations with neighboring scrambled frames). The conclusion that hippocampal movie fields depend on continuous movie frames, rather than a pure visual response to visual contents in individual frames, is supported to some degree by their changed properties after the frame scrambling (Fig. 4). However, there are two potential issues that could complicate this main conclusion.

One issue is related to the effect of behavioral variation or brain state. First, although the authors show that the movie fields are still present during low-speed stationary periods, there is a large drop in the movie tuning score (Z), especially in the hippocampal areas, as shown in Ext. Fig. 3b (compared to Ext. Fig. 2d). This result suggests a potentially significant enhancement by active behavior.

Second, a general, hard-to-tackle concern is that neuronal responses could be greatly affected by changes in arousal or brain state (including drowsy or occasional brief slow-wave sleep state) in head-fixed animals without a task. Without the analysis of pupil size or local field potentials (LFPs), the arousal states during the experiment are difficult to know. Many example movie fields in the presented raw data (e.g., Fig. 1c, Ext. Fig. 4) are broad with low-quality tuning, which could be due to broad changes in brain states. This concern is especially important for hippocampal responses, since the hippocampus can enter an offline mode indicated by the occurrence of LFP sharp-wave ripples (SWRs) while animals simply stay immobile. It is believed that the ripple-associated hippocampal activity is driven mainly by internal processing, not a direct response to external input (e.g., Foster and Wilson, Nature 440: 680, 2006). The "actual" hippocampal movie fields during a true active hippocampal network state, after the removal of SWR time periods, could have different quantifications that impact the main conclusion in the manuscript.

Another issue is related to the relative contribution of direct visual response versus the response to temporal continuity in movie fields. First, the data in Ext. Fig. 8 show that rapid frame-to-frame changes in visual contents contribute largely to hippocampal movie fields (similarly to visual movie fields). Interestingly, the data show that movie-field responses are correlated across all brain areas including the hippocampal ones. This could be due to heightened behavioral arousal caused by the changing frames as mentioned above, or due to enhanced neuronal responses to visual transients, which supports a component of direct visual response in hippocampal movie fields. Second, the data in Ext. Fig. 13c show a significant correlation in hippocampal responses to same scrambled frames between even and odd trials, which also suggests a significant component of direct visual response.

Is there a significant component purely due to the temporal continuity of movie frames in hippocampal movie fields? To support that this is indeed the case, the authors have presented data that hippocampal movie fields largely disappear after movie frames are scrambled. However, this could be caused by the movie-field detection method (it is unclear whether single-frame field could be detected). Another concern in the analysis is that movie-fields are not analyzed on re-arranged neural responses to scrambled movie frames. The raw data in Fig. 4e seem quite convincing. Unfortunately, the quantifications of movie fields in this case are not compared to those with the original movie.

Reviewer #2 (Public Review):

Purandare and Mehta investigated the neural activities modulated by continuous and sequential visual stimuli composed of natural images, termed "movie-tuning," measured along the visuo-hippocampal network when the animals passively viewed a movie without any task demand. Neurons selectively responded to some specific parts of the movie, and their activity timescales ranged from tens of milliseconds to seconds and tiled the entire movie with their movie-fields. The movie-tuning was lost in the hippocampus but not in the visual cortices when the image frames were temporally scrambled, implying that the rodent hippocampus encoded the specific sequence of images.

The authors have concluded that the neurons in the thalamo-cortical visual areas and the hippocampus commonly encode continuous visual stimuli with their firing fields spanning the mega-scale, but they respond to different aspects of the visual stimuli (i.e., visual contents of the image versus a sequence of the images). The conclusion of the study is fairly supported by the data, but some remaining concerns should be addressed.

1) Care should be taken in interpreting the results since the animal's behavior was not controlled during the physiological recording. It has been reported that some hippocampal neuronal activities are modulated by locomotion, which may still contribute to some of the results in the current study. Although the authors claimed that the animal's locomotion did not influence the movie-tuning by showing the unaltered proportion of movie-tuned cells with stationary epochs only, the effects of locomotion should be tested in a more specific way (e.g., comparing changes in the strength of movie-tuning under certain locomotion conditions at the single-cell level).

2) The mega-scale spanning of movie-fields needs to be further examined with a more controlled stimulus for reasonable comparison with the traditional place fields. This is because the movie used in the current study consists of a fast-changing first half and a slow-changing second half, and such varying and ununified composition of the movie might have largely affected the formation of movie-fields. According to Fig. 3, the mega-scale spanning appears to be driven by the changes in frame-to-frame correlation within the movie. That is, visual stimuli changing quickly induced several short fields while persisting stimuli with fewer changes elongated the fields. The presentation of persisting visual input for a long time is thought to be similar to staying in one place for a long time, and the hippocampal activities have been reported to manifest in different ways between running and standing still (i.e., theta-modulated vs. sharp wave ripple-based). Therefore, it should be further examined whether the broad movie-fields are broadly tuned to the continuous visual inputs or caused by other brain states.

3) The population activities of the hippocampal movie-tuned cells in Fig. 3a-b look like those of time cells, tiling the movie playback period. It needs to be clarified whether the hippocampal cells are actively coding the visual inputs or just filling the duration. The scrambled condition in which the sequence of the images was randomly permutated made the hippocampal neurons totally lose their selective responses, failing to reconstruct the neural responses to the original sequence by rearrangement of the scrambled sequence. This result indirectly addressed that the substantial portion of the hippocampal cells did not just fill the duration but represented the contents and temporal order of the images. However, it should be directly confirmed whether the tiling pattern disappeared with the population activities in the scrambled condition (as shown in Extended Data Fig. 11, but data were not shown for the hippocampus).

Reviewer #3 (Public Review):

In their study, Purandare & Mehta analyze large-scale single unit recordings from the visual system (LGN, V1, extrastriate regions AM and PM) and hippocampal system (DG, CA3, CA1 and subiculum) while mice monocularly viewed repeats of a 30s movie clip. The data were part of a larger release of publicly available recordings from the Allen Brian Observatory. The authors found that cells in all regions exhibited tuning to specific segments of the movie (i.e. "movie fields") ranging in duration from 20ms to 20s. The largest fractions of movie-responsive cells were in visual regions, though analyses of scrambled movie frames indicated that visual neurons were driven more strongly by visual features of the movie images themselves. Cells in the hippocampal system, on the other hand, tended to exhibit fewer "movie fields", which on average were a few seconds in duration, but could range from >50ms to as long as 20s. Unlike the visual system "movie fields" in the hippocampal system disappeared when the frames of the movie were scrambled, indicating that the cells encoded more complex (episodic) content, rather than merely passively reading out visual input.

The paper is conceptually novel since it specifically aims to remove any behavioral or task engagement whatsoever in the head-fixed mice, a setup typically used as an open-loop control condition in virtual reality-based navigational or decision making tasks (e.g. Harvey et al., 2012). Because the study specifically addresses this aspect of encoding (i.e. exploring effects of pure visual content rather than something task-related), and because of the widespread use of video-based virtual reality paradigms in different sub-fields, the paper should be of interest to those studying visual processing as well as those studying visual and spatial coding in the hippocampal system. However, the task-free approach of the experiments (including closely controlling for movement-related effects) presents a Catch-22, since there is no way that the animal subjects can report actually recognizing or remembering any of the visual content we are to believe they do. We must rely on above-chance-level decoding of movie segments, and the requirement that the movie is played in order rather than scrambled, to indicate that the hippocampal system encodes episodic content of the movie. So the study represents an interesting conceptual advance, and the analyses appear solid and support the conclusion, but there are methodological limitations.

Major concerns:

1) A lot hinges on hinges on the cells having a z-scored sparsity >2, the cutoff for a cell to be counted as significantly modulated by the movie. What is the justification of this criterion? It should be stated in the Results. Relatedly, it appears the formula used for calculating sparseness in the present study is not the same as that used to calculate lifetime sparseness in de Vries et al. 2020 quoted in the results (see the formula in the Methods of the de Vries 2020 paper immediately under the sentence: "Lifetime sparseness was computed using the definition in Vinje and Gallant").

To rule out systematic differences between studies beyond differences in neural sampling (single units vs. calcium imaging), it would be nice to see whether calculating lifetime sparseness per de Vries et al. changed the fraction "movie" cells in the visual and hippocampal systems.

2) In Figures 1, 2 and the supplementary figures-the sparseness scores should be reported along with the raw data for each cell, so the readers can be apprised of what types of firing selectivity are associated with which sparseness scores-as would be shown for metrics like gridness or Raleigh vector lengths for head direction cells. It would be helpful to include this wherever there are plots showing spike rasters arranged by frame number & the trial-averaged mean rate.

3) The examples shown on the right in Figures 1b and c are not especially compelling examples of movie-specific tuning; it would be helpful in making the case for "movie" cells if cleaner / more robust cells are shown (like the examples on the left in 1b and c).

4) The scrambled movie condition is an essential control which, along with the stability checks in Supplementary Figure 7, provide the most persuasive evidence that the movie fields reflect more than a passive readout of visual images on a screen. However, in reference to Figure 4c, can the authors offer an explanation as to why V1 is substantially less affected by the movie scrambling than it's main input (LGN) and the cortical areas immediately downstream of it? This seems to defy the interpretation that "movie coding" follows the visual processing hierarchy. Relatedly, the hippocampal data do not quite fit with visual hierarchical ordering either, with CA3 being less sensitive to scrambling than DG. Since the data (especially in V1) seem to defy hierarchical visual processing, why not drop that interpretation? It is not particularly convincing as is.

5) In the Discussion, the authors argue that the mice encode episodic content from the movie clip as a human or monkey would. This is supported by the (crucial) data from the scrambled movie condition, but is nevertheless difficult to prove empirically since the animals cannot give a behavioral report of recognition and, without some kind of reinforcement, why should a segment from a movie mean anything to a head-fixed, passively viewing mouse? Would the authors also argue that hippocampal cells would exhibit "song" fields if segments of a radio song-equally arbitrary for a mouse-were presented repeatedly? (reminiscent of the study by Aronov et al. 2017, but if sound were presented outside the context of a task). How can one distinguish between mere sequence coding vs. encoding of episodically meaningful content? One or a few sentences on this should be added in the Discussion.

I don't have an intuitive understanding of logs, I should spend sometime revisiting this.

Consensus Public Review:

Ottenheimer et al., present an interesting study looking at the neural representation of value in mice performing a pavlovian association task. The task is repeated in the same animals using two odor sets, allowing a distinction between odor identity coding and value coding. The authors use state-of-the-art electrophysiological techniques to record thousands of neurons from 11 frontal cortical regions to conclude that 1) licking is represented more strongly in dorsal frontal regions, 2) odor cues are represented more strongly in ventral frontal regions, 3) cue values are evenly distributed across regions. They separately perform a calcium imaging study to track coding across days and conclude that the representation of task features increments with learning and remains stable thereafter.

Overall, these conclusions are interesting and mostly well supported by the data, although there are some doubts about their definition of value coding. One limitation is the lack of focus on population-level dynamics from the perspective of decoding, with the analysis focusing primarily on encoding analyses within individual neurons.

Some specific comments:

The authors use reduced-rank kernel regression to characterize the 5332 recorded neurons on a cell-by-cell basis in terms of their responses to cues, licks, and reward, with a cell characterized as encoding one of these parameters if it accounts for at least 2% of the observed variance. At least 50% of cells met this inclusion criterion in each recorded area. 2% feels like a lenient cutoff, and it is unclear how sensitive the results are to this cutoff, though the authors argue that this cutoff should still only allow a false positive rate of 0.02% (determined by randomly shuffling the onset time of each trial).

Having identified lick, reward, and cue cells, the authors next select the 24% of "cue-only" neurons and look for cells that specifically encode cue value. Because the animal's perception of stimulus value can't be measured directly, the authors created a linear model that predicts the amount of anticipatory licking in the interval between odor cue and reward presentations. The session-average-predicted lick rate by this model is used as an estimate of cue value and is used in the regression analysis that identified value cells. (Hence, the authors' definition of value is dependent on the average amount of anticipatory behavior ahead of a reward, which indicates that compared to the CS+, mice licked around 70% as much to the CS50 and 10% as much to the CS-.) The claim that this is an encoding of value is strengthened by the fact that cells show similar scaling of responses to two odor sets tested. Whereas the authors found more "lick" cells in motor regions and more "cue" cells in sensory regions, they find a consistent percentage of "value" cells (that is, cells found to be cue-only in the initial round of analysis that is subsequently found to encode anticipatory lick rate) across all 11 recorded regions, leading to their claim of a distributed code of value.

In subsequent sections, the authors expand their model of anticipatory-licking-as-value by incorporating trial and stimulus history terms into the model, allowing them to predict the anticipatory lick rate on individual trials within a session. They also use 2-photon imaging in PFC to demonstrate that neural coding of cue and lick are stable across three days of imaging, supported by two lines of evidence. First, they show that the correlation between cell responses on all periods except for the start of day 1 is more correlated with day 3 responses than expected by chance (although the correlation is still quite low, for example, 0.2 on day 2). Second, they show that cue identity is able to capture the highest unique fraction of variance (around 8%) in day 3 cue cells across three days of imaging, and similarly for lick behavior in lick cells and cue+lick in cue+lick cells. Nonetheless, their sample rasters for all imaged cells also indicate that representations are not perfectly stable, and it will be interesting to see what *does* change across the three days of imaging.

Importantly, the authors do not present evidence that value itself is stably encoded across days, despite the paper's title. The more conservative in its claims in the Discussion seems more appropriate: "these results demonstrate a lack of regional specialization in value coding and the stability of cue and lick [(not value)] codes in PFC."

Reviewer #1 (Public Review):

The authors use a model of neonatal E.coli pneumonia to study differences between early neonates ad juvenile animals. They observe increased monocyte derived macrophage recruitment in juveniles compared to neonates as well as an increase in IFNG related genes. The data are of potential interest but in its current form it is unclear how well the experiments were controlled for confounders, such as sex and CFU.

1. This paper conducted research to identify the window of susceptibility to pneumonia due to E. coli, a bacteria that most often causes pneumonia in the neonatal period. This is an understudied area and thus the research is significant.

2. The paper provides evidence of differences in immune response in neonatal mice vs juvenile mice. However, it is unclear if the data are controlled adequately for the bacterial burden in the lung, which would be a crucial control to control for epi-phenomena. Additionally, it is unclear if the molecules that regulate macrophage recruitment are defective in neonatal mice or if it is an issue of macrophage progenitor cells.

Reviewer #2 (Public Review):

The authors have provided important detailed information on the inflammatory response to live E. coli infection in neonatal and juvenile mouse lungs. They have delineated key distinctions in these two periods and the potential impact on lung development. The study will inform future lines of investigation on the impact of bacterial infections on lung development.

Reviewer #1 (Public Review):

Synapses are modulated by neural activity on a variety of timescales. Typical neural network models primarily consider long-lasting changes to synaptic strengths, applied while the network is learning, with synaptic strengths then being fixed after learning. However, shorter-term plasticity mechanisms are ubiquitous in the brain and have been shown to have significant computational and information-storage capabilities. Here the authors study these mechanisms in the context of the integration of information tasks. Their two primary contributions are to analyze these short-term mechanisms separately from recurrent connections to isolate the specific ways these might be useful and to apply ideas from population data analysis to dissect how their networks solve the tasks.

I thought this was a clear, well-written, and well-organized paper, tackling an important problem. I also found that the conclusions were adequately supported by the simulations and analyses shown. I particularly appreciated the careful analysis of how the different networks solved the task and found the distinction between hidden neurons reflecting accumulated evidence (attractor architecture) vs. reflecting inputs (MPN architecture) very interesting and potentially very useful for thinking about experimental observations. My comments are primarily about the connection to biology/biological interpretability as well as how this study relates to prior work.

1) I was confused about the nature of the short-term plasticity mechanism being modeled. In the Introduction, the contrast drawn is between synaptic rewiring and various plasticity mechanisms at existing synapses, including long-term potentiation/depression, and shorter-term facilitation and depression. And the synaptic modulation mechanism introduced is modeled on STDP (which is a natural fit for an associative/Hebbian rule, especially given that short-term plasticity mechanisms are more often non-Hebbian). On the other hand, in the network models the weights being altered by backpropagation are changes in strength (since the network layers are all-to-all), corresponding more closely to LTP/LTD. And in general, standard supervised artificial neural network training more closely resembles LTP/LTD than changing which neurons are connected to which (and even if there is rewiring, these networks primarily rely on persistent weight changes at existing synapses). Moreover, given the timescales of typical systems neuroscience tasks with input coming in on the 100s of ms timescale, the need for multiple repetitions to induce long-term plasticity, and the transient nature/short decay times of the synaptic modulations in the SM matrix, the SM matrix seems to be changing on a timescale faster than LTP/LTD and closer to STP mechanisms like facilitation/depression. So it was not clear to me what mechanism this was supposed to correspond to.

2) A number of studies have explored using short-term plasticity mechanisms to store information over time and have found that these mechanisms are useful for general information integration over time. While many of these are briefly cited, I think they need to be further discussed and the current work situated in the context of these prior studies. In particular, it was not clear to me when and how the authors' assumptions differed from those in previous studies, which specific conclusions were novel to this study, and which conclusions are true for this specific mechanism as opposed to being generally true when using STP mechanisms for integration tasks.

Reviewer #2 (Public Review):

Most neuronal computations require keeping track of the inputs over temporal windows that exceed the typical time scales of single neurons. A standard and relatively well-understood way of obtaining time scales longer than those of the "microscopic" elements (here, the single neurons) is to have appropriate recurrent synaptic connectivity. Another possibility is to have a transient, input-dependent modulation of some neuronal and/or synaptic properties, with the appropriate time scale. Indeed, there is ample experimental evidence that both neurons and synapses modify their dynamics on multiple time scales, depending on the previous history of activation. There is, however, little understanding of the computational implications of these modifications, in particular for short-term memory.

Here, the authors have investigated the suitability of a class of transient synaptic modulations for storing and processing information over short-time scales. They use a purely feed-forward network architecture so that "synaptic modulation" is the only mechanism available for temporarily storing the information. The network is called Multi-Plasticity Network (MPN), in reference to the fact that the synaptic connectivity being transiently modulated is adjusted via standard supervised learning. They find that, in a series of integration-based tasks of varying difficulty, the MPN exhibits performances that are comparable with those of (trained) recurrent neuronal networks (RNNs). Interestingly, the MPN consistently outperforms the RNNs when only the read-out is being learned, that is in a minimal-training condition.

The conclusions of the paper are convincingly supported by the careful numerical experiments and the analysis performed by the authors, mostly to compare the performances of the MPN against various RNN architectures. The results are intriguing from a "classic" neuroscience perspective, providing a computational point of view to rationalize the various synaptic dynamics observed experimentally on largely different time scales, and are of certain interest to the machine learning community.

On the other hand, the general principle appears (perhaps naively) very general: any stimulus-dependent, sufficiently long-lived change in neuronal/synaptic properties is a potential memory buffer. For instance, one might wonder whether some non-associative form of synaptic plasticity (unlike the Hebbian-like form studied in the paper), such as short-term synaptic plasticity which depends only on the pre-synaptic activity (and is better motivated experimentally), would be equally effective. Or, for that matter, one might wonder whether just neuronal adaptation, in the hidden layer, for instance, would be sufficient. In this sense, a weakness of this work is that there is little attempt at understanding when and how the proposed mechanism fails.

Reviewer #3 (Public Review):

The authors study the performance, generalization, and dynamics of artificial neural networks trained on integration tasks. These types of tasks were studied theoretically in the past, and comparisons have also been made between artificial and biological networks. The authors focus on the effect of short-term plasticity on the networks. This is modeled as a multiplicative modulation of synaptic strengths that decays over time. When not decaying, this modulation is driven by Hebbian (or anti-Hebbian) activity-dependent terms. To isolate the effects of this component of the networks, the authors study a feedforward architecture, thereby rendering the synaptic modulations the only dynamical variables in the system. The authors also compare their network (MPN) with RNNs (gated and vanilla).

Perhaps not surprisingly, the information on the integration task is encoded in the dynamic variables of the networks - which are hidden units for RNNs and synaptic modulations for MPNs. The authors also study the dynamics of MPNs in the presence of noise or longer-than-trained input sequences. Finally, context-dependent integration is also studied.<br /> Biological neurons are far more complex than their artificial counterparts. This implies that there are computations that can be "outsourced" to these complexities, instead of being handled by a vanilla-rnn-like network that only has connectivity and hidden states. Given the recent rise in applications of trained RNNs as models of biological systems, it is thus timely to ask what are the consequences of integrating some of these complexities. The current study falls under this broad question, with a focus on short-term synaptic plasticity.<br /> I am worried, however, by two issues: the relation between integration tasks and the plasticity mechanism introduced, and the relation to existing work.

Because the MPN is essentially a low-pass filter of the activity, and the activity is the input - it seems that integration is almost automatically satisfied by the dynamics. Are these networks able to perform non-integration tasks? Decision-making (which involves saddle points), for instance, is often studied with RNNs.

The current work has some resemblance to reservoir computing models. Because the M matrix decays to zero eventually, this is reminiscent of the fading memory property of reservoir models. Specifically, the dynamic variables encode a decaying memory of the input, and - given large enough networks - almost any function of the input can be simply read out. Within this context, there were works that studied how introducing different time scales changes performance (e.g., Schrauwen et al 2007).

Another point is the interaction of the proposed plasticity rule with hidden-unit dynamics. What will happen for RNNs with these plasticity rules? I see why introducing short-term plasticity in a "clean" setting can help understand it, but it would be nice to see that nothing breaks when moving to a complete setting. Here, too, there are existing works that tackle this issue (e.g., Orhan & Ma, Ballintyn et al, Rodriguez et al).

One point regarding biological plausibility - although the model is abstract, the fact that the MPN increases without bounds are hard to reconcile with physical processes.<br /> To summarize, the authors show that plastic synapses can perform integration tasks in a manner that is dynamically distinct from RNNs - thereby strengthening the argument to include such synapses in models. This can be of interest to researchers interested in biologically plausible models of neural circuits.

Schrauwen, Benjamin, Jeroen Defour, David Verstraeten, and Jan Van Campenhout. "The Introduction of Time-Scales in Reservoir Computing, Applied to Isolated Digits Recognition." In Artificial Neural Networks - ICANN 2007, edited by Joaquim Marques de Sá, Luís A. Alexandre, Włodzisław Duch, and Danilo Mandic, 471-79. Lecture Notes in Computer Science 4668. Springer Berlin Heidelberg, 2007. http://link.springer.com/chapter/10.1007/978-3-540-74690-4_48.

Orhan, A. Emin, and Wei Ji Ma. "A Diverse Range of Factors Affect the Nature of Neural Representations Underlying Short-Term Memory." Nature Neuroscience 22, no. 2 (February 2019): 275-83. https://doi.org/10.1038/s41593-018-0314-y.

Ballintyn, B., Shlaer, B. & Miller, P. Spatiotemporal discrimination in attractor networks with short-term synaptic plasticity. J Comput Neurosci 46, 279-297 (2019). https://doi.org/10.1007/s10827-019-00717-5

Rodriguez, H.G., Guo, Q. & Moraitis, T.. (2022). Short-Term Plasticity Neurons Learning to Learn and Forget. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:18704-18722 Available from https://proceedings.mlr.press/v162/rodriguez22b.html.

Reviewer #1 (Public Review):

This paper identifies an intracellular O-GlcNAc glycosylation of specific proteins in the control of bone formation and bone marrow adiposity. Compelling evidence is provided for the role of OGT-mediated O-GlcNAc glycosylation of RUNX2 in osteogenic differentiation versus OGT-mediated O-GlcNAc glycosylation of C/EBPβ in bone marrow adipogenesis.

Overall, the experiments have been done with great rigor, and sufficient details are provided for reproducibility. The authors developed a novel concept in the control of bone formation and bone marrow adiposity.

Reviewer #2 (Public Review):

Here I will mainly comment on the biology of adipocytes, which is my specialty.

In this manuscript, it has been very convincingly shown that O-GlcNAc acts as an important regulator of MSC differentiation in mice, and given previous studies in which O-GlcNAc is regulated by aging and nutritional status, it makes sense that this PTM determines differentiation and BM niche.

The point that O-GlcNAc regulates adipocyte differentiation is convincing, but there are already previous studies using 3T3-L1 (e.g., Biochemical and Biophysical Research Communications 417 (2012) 1158-1163), and a more step-by-step demonstration of the molecular mechanism would make this an excellent paper that can be extended to adipocyte research in general, not just BM.

It is somewhat unclear whether or not the authors' in vitro experiments using 10T1/2 cells accurately reflect what is happening in vivo in knockout mice. The PDGFRa+VCAM1+ population of adipocyte progenitors shown by the authors is upregulated by about 30% by knockout of Ogt (Figure 4C). How significant is this difference? Rather, might the expression of Pparg, which indicates lineage commitment, be the underlying mechanism? In any case, this manuscript is highly impactful in the sense that the differentiation of adipocytes forming the BM niche can be controlled using tissue-specific knockouts of the Ogt gene.

Reviewer #3 (Public Review):

This study has the strengths of novelty and significance across multiple fields, including bone marrow biology, skeletal health, hematopoiesis, and protein posttranslational modification (PTM). It establishes the role of protein O-GlcNAcylation in bone development and bone marrow niche. The cooperative O-GlcNAcylation on Runx2 and C/EBPb to prime BMSCs toward osteoblast differentiation over adipogenesis is a very interesting and sounding molecular mechanism. The employment of an inducible OGT conditional knockout mouse model with appropriate Osx-Cre controls is conclusive and rigorous. The in vitro experiments were carefully designed in support of strong rationales. The overall flow of the story is logical and clear. Last, the conclusions are drawn from concrete evidence in an accurate way.

Reviewer #1 (Public Review):

The authors made some biologically reasonable approximations of the Pump and Leak model. e.g., assuming the alpha_0 parameter to be zero. These approximations significantly simplify the model and make the results much more intuitive, e.g., Eq. 4 in the main text. The authors proposed an interesting and simple model of amino acid production, which is argued to be the primary determinant of cell volume. Combined with the gene expression model proposed recently by Lin and Amir, their model can nicely explain the homeostasis of protein density. Furthermore, by considering the saturation of DNA and mRNA by RNA polymerase and ribosome, the authors extended Lin and Amir's model by introducing protein degradation, which I think is the key to explaining cytoplasm dilution. The authors also discussed other applications of their model, including mitotic swelling and nuclear scaling. Below are my major comments:

1. Eq. 2 is valid for stationary states where the cell volume is constant with time. However, many cells grow and divide, including yeast cells. I think the authors have implicitly neglected the effects of cell growth. The authors may want to mention this explicitly to avoid confusion.

2. It's unclear how the authors go from Eq. S.21 to Eq. 2, although the authors mentioned it is straightforward. I think the dilute solution assumption is used without explicit mention, at least in section A of the SI.

3. A slight deviation from equilibrium is implicitly assumed in Eq. S.22 I think since the flow is linearly proportional to the chemical potential difference. The authors may want to mention this explicitly since the linear assumption is not necessarily true for biological systems.

4. A more general gene expression model is recently proposed by some of the authors of Ref. 30, in which the saturation of DNA by RNAPs is due to a high free RNAP concentration near the promoter (Wang and Lin, Nature Communications, 2021). I think the exact saturation mechanism is not very important to the conclusions. Still, I think it's good to let readers be aware that there are biologically more realistic saturation mechanisms.

5. The success of the fitting in Figure 2E is intriguing but may not be a smoking gun evidence of the model's validity. All one needs is a protein number proportional to cell volume for tt**, as far as I understand. Alternative models incorporating the above features will be able to reproduce the fitting of Figure 2E as well, I think. For example, instead of adding protein degradation, one can alternatively assume that protein translation becomes much slower for t>t**, but amino acids are still produced at a constant rate. The time-dependences of amino acids and cell volume may not be important if one just wants to fit the data in Figure 2E since the cell volume dynamics are extracted from Figure 2B. The authors may want to discuss this point.

6. On line 752, the estimation of the average charge of proteins is unclear to me. How did the authors obtain z_p = 0.8?

Reviewer #2 (Public Review):

The manuscript proposes a theoretical framework for the size scaling of cells. The main predictions are (1) the application of a nested pump-leak model to explain cell size scaling through an osmotic balance, (2) the role of metabolites in maintaining electroneutrality, and (3) the breakdown of this scaling law during specific phases of cell growth and senescence.

Although the overall topic and approach are of significant interest, there are several issues with the presentation and claimed scope, detailed below.

Major comments:

1. The manuscript claims to provide a unified theory of cell size scaling, but quantitative agreement is only shown in a few specific cases (non-dividing yeast cells, mitotic swelling in mammalian cells, nuclear size scaling). Given the significant number of adjustable parameters in the model, the claim of a unified theory seems to be somewhat of a stretch. In addition, many of the approximations used (such as turgor pressure being negligible on p. 5) are valid in mammalian cells, but not in plant or yeast cells. For example, in walled cells, the rate of volume growth is dictated largely by cell-wall synthesis and turgor pressure (Rojas and Huang, 2018).

2. The paper claims to supersede previous work: "Many theoretical papers have assumed a priori a linear phenomenological relation between volume and protein number in order to study cell size [30],[31],[32]. Our results instead emphasize that the proportionality is indirect, only arising from the scaling between amino-acid and protein numbers." However, the conclusions reached (e.g. NC1 in eq. 15) appear to recover those of previous work, at least in certain limiting cases. Moreover, this is not a fully accurate description of the previous work, since in some of the previous works the osmotic balance is given in terms of general macromolecules, not necessarily proteins, and the linear relationship was not assumed but rather derived based on osmotic balance. The authors should carefully explain the relationship of their work to the previous studies.

3. The role of metabolites is an important point that should be further clarified. The authors state that "As a key consequence, we find that the NC ratio would be four times larger in the absence of metabolites". However, the formula obtained in the metabolite-dominated limit for NC1 in eq. 15 recovers previous results which were based solely on osmotic balance, without accounting for electroneutrality via metabolites. Why is electroneutrality violated in the absence of metabolites? Does this remain true if the chromatin and counterions are considered to be polyelectrolytes?

4. Appendix H on the extension to scaling of other organelles contains no comparison to data. Is the size control of all membrane-bound organelles expected to behave according to the same principles, or is the theory applicable to a particular subset of organelles?

5. It is stated several times that the size cell is "tightly regulated by active processes". The authors should define what they mean by "control" and "active" in this context. For example, one interpretation of the NC ratio size scaling result is that it is not under direct control, but rather is a consequence of the ratio of nuclear-bound proteins and is only controlled indirectly. (The authors themselves state that the relationship between volume and protein number is indirect.) If the NC ratio is actively controlled, this suggests that its maintenance at a certain value is important for the proper functioning of the cell. Is there evidence of this, or would the cell continue to function if the nuclear size could hypothetically be perturbed independently of the protein ratio?

Reviewer #1 (Public Review):

This work applies duplex sequencing to study point mutations in mice across tissues in young (4.5 months) and old mice (26 months). In this study, they identified 89,000 independent somatic mtDNA mutations representing the largest collection of somatic 'point' mtDNA mutation (not considering mtDNA deletions). They find that mtDNA mutations accumulate linearly with age in a clock-like manner but are not uniformly represented in all tissues. This indicates a likely constant 'clock-like' accumulation analogous to what is seen in the nuclear genome. This part of the paper is a comprehensive extension of work done by Arbeithuber et al., 2020. They also find variability between tissues of the ROS-linked (transversions) mutations. Similar to prior work by Kennedy and Loeb (2013 Plos Genetics) they conclude that ROS-linked mutations do not accumulate significantly with age. Lastly, the authors apply this knowledge and technique to interrogate whether mtDNA mutations are affected by two known treatments, elimipretide and nicotinamide mononucleotide, that have been shown to improve mitochondrial function and reverse apparent aging phenotypes. Here they demonstrate that these treatments reduced the low level of ROS accumulated mtDNA mutations seen in untreated tissues.

Comments:<br /> The paper states that they observed a combined total of 77,017 single-nucleotide variants (SNVs) and 12,031 insertion/deletions (In/Dels) across all tissue, age, and intervention groups. Collectively, these data represent the largest collection of somatic mtDNA mutations obtained in a single study to date. However, A study with more somatic mtDNA mutations by the LostArc method (PMID 32943091) revealed 35 million deletions (~ 470,000 unique spans) in skeletal muscle from 22 individuals with and 19 individuals without pathogenic variants in POLG. Thus, the authors should reword this part to say that this study represents the largest collections of mouse mtDNA point mutations detected, but not the largest amount of mutations (deletions exceed this number).

What is the theoretical limit of pt mutations in the mitochondrial genome, assuming only one pt mutation per genome? Doesn't 77000 detected independent pt mutations approach that limit? Can the authors estimate how many molecules contained two or more pt mutations? Did the analysis reveal any un-mutated regions implying an essential function? For example, on p.9 can the authors provide an explanation of why OriL and other G/C-rich regions were not uniformly covered as compared to the rest of the genome?

Given that mitochondrial disease usually doesn't present until >60% of the genomes are affected, the very low level of detected pt mutations observed in the mouse (and presumably similar to human) would mean that they are well below a physiological level. Thus, these low-level pt mutations are well tolerated. Can the authors estimate a theoretical age of the mouse (well beyond their life span) where over 50% of the genomes carry at least one pt mutation?

Also, the problem with this low level of pt mutations is that they are not physiological, the effect of the drug treatment causing a reduction in ROS-mediated transversions would not be expected to have a detectable effect on mitochondria. The improvement on mitochondrial seen by others is most likely independent of the mutations in the genome. There needs to be a cause and effect here and I don't see one.

There's no mention in this paper and methodology about how point mutations in nuclear-encoded mtDNA (NUMTs) are excluded from the reads and I'm worried that these errors are being read as rare errors in the mtDNA genome. While NUMTs have been documented for decades, a recent report in Science (PMID: 36198798) documents how frequently and fluidly NUMTs occur. Can the authors provide a clear explanation of how mutations in NUMTs are excluded?

Reviewer #2 (Public Review):

A common problem in mutation analysis is that DNA damage (present in one strand) is difficult to separate from real mutations (present in both strands). One of the approaches to solve this problem based on independent tagging of the two strands by different unique molecular identifiers was developed by the authors about 10 years ago. This study summarizes the application of this method to a wide range of mouse tissues, ages, and drug treatment regimes. Much of the results confirm previous conclusions from this laboratory. This involves overall mutational levels of somatic mtDNA mutations (~10-6-10-5), their accumulation with age, the prevalence of GA/CT transitions, and their clonality. Although these results were not new, it is important that these were confirmed in a single study with high confidence in a huge number of independent mutations.

What really sets this study apart from other studies is the detection of a large proportion of transversion mutations, primarily of the C>A/G>T and C>G/G>C types. Transversions are traditionally considered 'persona non grata' in mtDNA mutational spectra and are typically associated with errors of mutational analysis (which they in fact are). The presence of these mutations in both strands of the duplex makes a good case that these mutations are real, rather than converted damage. However, because this is such a novel discovery and because regular controls do not work (I mean, for example, that these mutations never clonally expand. If there is a clonal expansion, then the mutation is real, only real mutation can expand. But in the case of non-expandable C>A/G>T and C>G/G>C this control does not help to validate these mutations), it would be nice to provide extra assurances that this is not some kind of artifact that somehow slipped through the ds sequencing procedure. I would recommend including in the supplement the data on the abundance of single-stranded base changes as detected by ds sequencing (i.e., changes confirmed in one and not in the other strand of a given molecule). An unusually high presence of such single-stranded changes of the C>A/G>T and C>G/G>C type would be a red flag for me. If ratios of single and double-stranded mutations were similar for transitions and transversions - that would reassure me and hopefully the reader.

Furthermore, a similar excess of C>A/G>T and C>G/G>C has been observed in a recent paper by Abascal 2021 (cited in the manuscript). In that paper, a UMI- free, but otherwise very similar ds sequencing approach in nuclear DNA (BotSeqS) was demonstrated to suffer from an artifact causing (among other effects) an excess of C>A/G>T and C>G/G>C transversions. This artifact is related to end repair and nick-translation of DNA fragments during library preparation. Because BotSeqS is very similar to ds sequencing, we expect that same artifact may be taking place in the study under review. We recommend running checks similar to those undertaken by Abascal et al (which include, at the very minimum, checking the distribution of the C>A/G>T and C>G/G>C transversions within the reads (artifacts tend to be concentrated towards the ends of the reads).

Of note, even if transversions detected in this study prove to be artifacts of the Abascal type (likely) they still may reflect real ss damage in mtDNA (not instrumental artifacts, like sequencing errors or in vitro DNA damage). This is supported by the strong variation in the levels of transversions across tissues and as a result of the ameliorating drug intervention. Artifacts, in contrast, would be expected to be at a constant level. This logic, however, does not differentiate between real ds mutations and ss damage. So UMI-based ds sequencing evidence remains the only (though very strong) independent proof. So, in my view, whereas the jury may be still out on whether the observed transversions are true ds mutations or some kind of single-stranded damage, this is a critically important observation. The evidence of ss damage greatly varied between tissues and detected with such precision on a single molecule level is a very important finding as well.

Out of caution, I would recommend mentioning the above-stated uncertainty and noting that more research is needed to fully confirm that C>A/G>T and C>G/G>C changes detected in this study are indeed double-stranded mutations.

Reviewer #1 (Public Review):

The manuscript by Hekselman et al presents analyses linking cell-types to monogenic disorders using over-expression of monogenic disease genes as the signal. The manuscript analyses data from 6 tissues (bone marrow, lung, muscle, spleen, tongue and trachea) together with ~1,000 rare diseases from OMIM (with ~2,000 associated genes) to identify cell-type of interest for specific disease of choice. The signal used by the approach is the relative expression of OMIM-genes in a particular cell type relative to the expression of the gene in the tissue of interest identifying cell-type-disease pairs that are then investigated through literature review and recapitulated using mouse expression. A potentially interesting finding is that disease genes manifesting in multiple tissues seem to hit same cell-types. Overall this important study combines multiple data analyses to quantify the connection between cell types and human disorders. However whereas some of the analyses are compelling, the statistical analyses are incomplete as they don't provide full treatment of type I error.

Reviewer #2 (Public Review):

This study identifies 110 disease-affected cell types for 714 Mendelian diseases, based on preferential expression of known disease-associated genes in single-cell data. It is likely that many or most of the results are real, and the results are biologically interesting and provide a valuable resource. However, updates to the method are needed to ensure that inference of statistical significance is appropriately stringent and rigorous.

Strengths: a systematic evaluation of disease-affected cell types across Mendelian diseases is a valuable addition to the literature, complementing systematic evaluations of common disease and targeted analyses of individual Mendelian diseases. The validation via excess overlap with disease-cell type pairs from literature co-appearance provides compelling evidence that many or most of the results are real. In addition, many of the results are biologically interesting. In particular, it is interesting that diseases with multiple affected tissues tend to affect similar cell types in the respective tissues.

Limitations: the main limitation of the study is that, although many or most of the results are likely to be real, the criteria for statistical significance is probably not stringent enough, and is not well-justified. For diseases with only 1 disease-associated gene, the threshold is a z-score>2 for preferential expression in the cell type, but this threshold is likely to be often exceeded by chance. (For diseases with many disease-associated genes, the threshold is a median (across genes) z-score>2 for preferential expression in the cell type, which is less likely to occur by chance but still an arbitrary threshold.) Thus, there is a good chance that a sizable proportion of the reported disease-affected cell types might be false positives. The best solution would be to assess statistical significance via empirical comparison with results for non-disease-associated control genes, and assess the statistical significance of the resulting P-values using FDR.

The re-analysis using mouse single-cell data adds an interesting additional dimension to the study, with the small caveat that mouse single-cell data does not provide statistically independent information across genes (for the same reason that adding data from independent human individuals would not provide statistically independent information across genes, given that human and mouse expression are partially correlated).

Reviewer #3 (Public Review):

The authors describe the method, PrEDiCT, which helps identify disease affected cell types based on gene sets. As I understand it, the method is based on finding which "disease genes" (from an annotation) are relatively highly expressed. The idea is nice, however, I have concerns about how "significance" is assessed and the relative controls.

Overall, I find the idea interesting, but the execution raises some concerns.

1. From a causal perspective, there is an association of high expression of these genes within these cell types, but without also assessing individuals with those specific diseases, I do not it is fair to say "disease affected" cell types. It is possible that these genes might behave completely fine but are highly expressed in those cell types while being affected another in other cell types.

2. It is unclear to me what the "null" comparison is in the method and if there is one. For example, by chance, would I expect this gene to be highly expressed because other genes are also highly expressed in this cell type? Some way to assess "significance" or "enrichment" beyond simply using ranks and thresholds would be helpful in deciding whether these associations are robust.

3. Additionally, it is unclear to me, but I suspect that there are unequal cell numbers in the scores computed as well as between relevant tissues. This is related to point (2) above, but as a result, the estimates of the scores will inherently have different variances, thus making comparisons between them difficult/unreliable unless accounted for. If I understand correctly, the score is first the average expression within a tissue, _then_, the Z-score? If so, my comment applies.

4. There is a large set of work done in gene enrichment sets which appears to not be mentioned (e.g. GSEA and other works by the Price group). It would be helpful for the authors to summarize these methods and how their method differs.

5. Additionally, it should be noted that a caveat of this analysis is that the comparisons are all done only relative to the cell types sampled and the diseases which have Mendelian genes associated with them. I would expect these results to change, possibly drastically, if the sampled cell types and diseases were to be changed.

6. Finally, I would appreciate a more detailed explanation in the methods of how the score is computed. Some equations and the data they are calculated from would be helpful here.

In summary, the general idea is an interesting one, but I do think the issues above should be addressed to make the results convincing.

Reviewer #1 (Public Review):

This is a carefully performed and well-documented study to indicate that the FUS protein interacts with the GGGGCC repeat sequence in Drosophila fly models, and the mechanism appears to include modulating the repeat structure and mitigating RAN translation. They suggest FUS, as well as a number of other G-quadruplex binding RNA proteins, are RNA chaperones, meaning they can alter the structure of the expanded repeat sequence to modulate its biological activities.

Overall this is a nicely done study with nice quantitation. It remains somewhat unclear from the data and discussions in exactly what way the authors mean that FUS is an RNA chaperone: is FUS changing the structure of the repeat or does FUS binding prevent it from folding into alternative in vivo structure?

Reviewer #2 (Public Review):

Fuijino et al. provide interesting data describing the RNA-binding protein, FUS, for its ability to bind the RNA produced from the hexanucleotide repeat expansion of GGGGCC (G4C2). This binding correlates with reductions in the production of toxic dipeptides and reductions in toxic phenotypes seen in (G4C2)30+ expressing Drosophila. Both FUS and G4C2 repeats of >25 are associated with ALS/FTD spectrum disorders. Thus, these data are important for increasing our understanding of potential interactions between multiple disease genes. However, further validation of some aspects of the provided data is needed, especially the expression data.

Some points to consider when reading the work:

The broadly expressed GMR-GAL4 driver leads to variable tissue loss in different genotypes, potentially confounding downstream analyses dependent on viable tissue/mRNA levels.

The relationship between FUS and foci formation is unclear and should be interpreted carefully.

Reviewer #3 (Public Review):

In this manuscript Fujino and colleagues used C9-ALS/FTD fly models to demonstrate that FUS modulates the structure of (G4C2) repeat RNA as an RNA chaperone, and regulates RAN translation, resulting in the suppression of neurodegeneration in C9-ALS/FTD. They also confirmed that FUS preferentially binds to and modulates the G-quadruplex structure of (G4C2) repeat RNA, followed by the suppression of RAN translation. The potential significance of these findings is high since C9ORF72 repeat expansion is the most common genetic cause of ALS/FTD, especially in Caucasian populations and the DPR proteins have been considered the major cause of the neurodegenerations.

1) While the effect of RBP as an RNA chaperone on (G4C2) repeat expansion is supposed to be dose-dependent according to (G4C2)n RNA expression, the first experiment of the screening for RBPs in C9-ALS/FTD flies lacks this concept. It is uncertain if the RBPs of the groups "suppression (weak)" and "no effect" were less or no ability of RNA chaperone or if the expression of the RBP was not sufficient, and if the RBPs of the group "enhancement" exacerbated the toxicity derived from (G4C2)89 RNA or the expression of the RBP was excessive. The optimal dose of any RBPs that bind to (G4C2) repeats may be able to neutralize the toxicity without the reduction of (G4C2)n RNA.

2) In relation to issue 1, the rescue effect of FUS on the fly expressing (G4C2)89 (FUS-4) in Figure 4-figure supplement 1 seems weaker than the other flies expressing both FUS and (G4C2)89 in Figure 1 and Figure 1-figure supplement 2. The expression level of both FUS protein and (G4C2)89 RNA in each line is important from the viewpoint of therapeutic strategy for C9-ALS/FTD.

3) While hallmarks of C9ORF72 are the presence of DPRs and the repeat-containing RNA foci, the loss of function of C9ORF72 is also considered to somehow contribute to neurodegeneration. It is unclear if FUS reduces not only the DPRs but also the protein expression of C9ORF72 itself.

4) In Figure 5E-F, it cannot be distinguished whether FUS binds to GGGGCC repeats or the 5' flanking region. The same experiment should be done by using FUS-RRMmut to elucidate whether FUS binding is the major mechanism for this translational control. Authors should show that FUS binding to long GGGGCC repeats is important for RAN translation.

5) It is not possible to conclude, as the authors have, that G-quadruplex-targeting RBPs are generally important for RAN translation (Figure 6), without showing whether RBPs that do not affect (G4C2)89 RNA levels lead to decreased DPR protein level or RNA foci.

Reviewer #1 (Public Review):

This manuscript by Mahlandt, et al. presents a significant advance in the manipulation of endothelial barriers with spatiotemporal precision, and in the use of optogenetics to manipulate cell signaling in vascular biology more generally. The authors establish the role of Rho-family GTPases in controlling the cytoskeletal-plasma membrane interface as it relates to endothelial barrier integrity and function and adequately motivate the need for optogenetic tools for global and local signaling manipulation to study endothelial barriers.

Throughout the work, the optogenetic assays are conceptualized, described, and executed with exceptional attention to detail, particularly as it relates to potential confounding factors in data analysis and interpretation. Comparison across experimental setups in optogenetics is notoriously fraught, and the authors' control experiments and measurements to ensure equal light delivery and pathway activation levels across applications are very thorough. In demonstrating how these new opto-GEFs can be used to alter vascular barrier strength, the authors cleverly use fluorescent-labeled dextran polymers of different sizes and ECIS experiments to demonstrate the physiological relevance of BOEC monolayers to in vivo blood vessels. Of particular note, the resiliency of the system to multiple stimulation cycles and longer time course experiments is promising for use in vascular leakage studies.

Given that dozens of Rho GTPase-activating GEFs exist, an expanded rationale for the selection of p63, ITSN1, and TIAM1 in the form of discussion and literature citations would be helpful to motivate their selection as protein effectors in the engineered tools. Extensive tool engineering studies demonstrate the superiority of iLID over optogenetic eMags or rapamycin-based chemogenetic tools for these purposes. However, as the utility of iLID and eMags has been demonstrated for the manipulation of a variety of signaling pathways, the iSH-Akt demonstration does not seem necessary for these systems.

The demonstration of orthogonality in GTPase- and VE-cadherin-blocking antibody-mediated barrier function decreases and is compelling, even without full elucidation of the role of cell size or overlap in barrier strength. The discussion section presents a mature and thoughtful description of the limitations, remaining questions, and potential opportunities for the tools and technology developed in this work. Importantly, this manuscript demonstrates a commitment to scientific transparency in the ways in which the data are visualized, the methods descriptions, and the reagent and code sharing it presents, allowing others to utilize these tools to their full potential.

Reviewer #2 (Public Review):

This manuscript reports on the use of Optogenetics to influence endothelial barrier integrity by light. Light-induced membrane recruitment of GTPase GEFs is known to stimulate GTPases and modulate cell shape, and here this principle is used to modulate endothelial barrier function. It shows that Rac and CDc42 activating constructs enhance barrier function and do this even when a major junctional adhesion molecule, VE-cadherin, is blocked. Activation of Rac and Cdc42 enhanced lamellipodia formation and cellular overlaps, which could be the basis for the increase in barrier integrity.

The authors aimed at developing a light-driven technique with which endothelial barrier integrity can be modulated on the basis of activating certain GTPases. They succeeded in using optogenetic tools that recruit GEF exchange domains to membranes upon light induction in endothelial cell monolayers. Similar tools were in principle known before to modulate cell shape/morphology upon light induction but were used here for the first time as regulators of endothelial barrier integrity. In this way, it was shown that the activation of Cdc42 and Rac can increase barrier integrity even if VE-cadherin, a major adhesion molecule of endothelial junctions, is blocked. Although it was shown before that stimulation of the S1P1 receptor or of Tie-2 can enhance endothelial barrier integrity in dependence on Cdc42 or Rac1 and can do this independent of VE-cadherin, the current study shows this with tools directly targeting these GTPases.

Furthermore, this study presents very valuable tools. The immediate and repeatable responses of barrier integrity changes upon light-on and light-off switches are fascinating and impressive. It will be interesting to use these tools in the future in the context of analyzing other mechanisms which also affect endothelial barrier function and modulate the formation of endothelial adherens junctions.

Reviewer #3 (Public Review):

Mahlandt et al. report the design and proof of concept of Opto-RhoGEF, a new set of molecular tools to control the activation by light of the three best-known members of the Rho GTPase family, RhoA, Rac1, and Cdc42.

The study is based on the optogenetically-controlled activation of chimeric proteins that target the plasma membrane guanine nucleotide exchange factors (GEFs) domains, which are natural activators specific for each of these three Rho GTPases. Membrane-targeted GEFs encounter and activate endogenous Rho proteins. Further investigation into the effect of these tools on RhoGTPase signaling would have strengthened the report.

These three Opto-RhoGEFs are reversible and enable the precise spatiotemporal control of Rho-regulated processes, such as endothelial barrier function, cell contraction, and plasma membrane extension. Hence, these molecular tools will be of broad interest to cell biologists interested in this family of GTPases.

Mahlandt et al. design and characterize three new optogenetic tools to artificially control the activation of the RhoA, Rac1, and Cdc42 by light. These three Rho GTPases are master regulators of the actin cytoskeleton, thereby regulating cell-cell contact stability or actin-mediated contraction and membrane protrusions.

The main strength of this new experimental resource lies in the fact that, to date, few tools controlling Rho activation by reversibly targeting Rho GEFs to the plasma membrane are available. In addition, a comparative analysis of the three Opto-RhoGEFs adds value and further strengthens the results, given the fact that each Opto-GEF produces different (and somehow expected) effects, which suggest specific GTPase activation. The design of the tools is correct, although the membrane targeting could be improved, since the Lck N-terminus used to construct the recombinant proteins contains myristoylation and palmitoylation sites, which have the potential to target the chimeric protein to lipid rafts. As a consequence, this may not evenly translocate these Rho-activating domains.

An additional technical feature that must be highlighted is an elegant method to activate Opto-RhoGEFs in cultured cells, independent of laser and microscopes, by using led strips, which notably expands the possibilities of this resource, potentially allowing biochemical analyses in large numbers of cells.

The experimental evidence clearly indicates that the authors have achieved their aim and designed very useful tools. However, they should have taken more advantage of this remarkable technical advance and investigated in further detail the spatiotemporal dynamics of Rho-mediated signaling. Although the manuscript is a "tool and resource", readers may have better grasped the potential benefits of tuning GTPase activity with this tool by learning about some original and quantitative insights of RhoA, Rac1, and Cdc42 function.

One of such insights may have come from the set of data regarding the contribution of adherens junctions. The effect of other endothelial cell-cell junctions, such as tight junctions, may also contribute to barrier function, as well as junctional independent, cell-substratum adhesion. These optogenetic tools will undoubtedly impact these future studies and help decipher whether these other adhesion events that are important for endothelial barrier integrity are also under the control of these three GTPases. Overall, the manuscript is sound and presents new and convincing experimental strategies to apply optogenetics to the field of Rho GTPases.

Reviewer #1 (Public Review):

This is a well-conceived and well-executed investigation of how activation loop autophosphorylation and IN-box autophosphorylation synergistically activate AURKB/INCENP. An elegant chemical ligation strategy allowed construction of the intermediate phospho-forms so that the contributions of each phosphorylation event to structure, dynamics, and activity could be dissected. Autophosphorylation at both sites serves to rigidify both AURKB and the IN-box, and to coordinate opening, twisting, and activation loop movements. Consistent with previous findings, both sites are necessary for enzymatic activity; further, this work finds that activation loop autophosphorylation occurs slowly in cis while IN-box autophosphorylation occurs quickly in trans.

Due to abundant previous work in the field, many of the conclusions of this paper were expected. However, that does not diminish the quality of the work, and the addition of how kinase dynamics contribute to activation is important for AURKB and many other kinases. The experimental results are clear and interpreted appropriately, with good controls. The computational work is also clearly explained and directly tied to the function of the enzyme, making it highly complementary to the experimental findings and to previously published structures.

Some minor limitations of the study:

1. Of note when interpreting the HDX data, there is no coverage of the peptide containing the activation loop autophosphorylation site T248 (Fig S2A), and as mentioned in the Discussion, the time scale of HDX is not able to capture differences in exchange in very flexible regions like the activation loop.

2. Some data lack robust statistical analysis, which would make the findings more compelling.

3. One point that might be clarified is how the occupancy of T248 was confirmed to be either fully phosphorylated in the [AURKB/IN-box]IN-deltaC or fully dephosphorylated in the IN-box K846N/R827Q mutant. Especially because T248 autophosphorylation is found to occur in cis, it is unclear how incubating the [AURKB/IN-box]IN-deltaC with traces of wild-type [AURKB/IN-box]all-P would ensure that T248 is phosphorylated.

Reviewer #2 (Public Review):

This study presents a dynamic, multi-step model for the activation of Aurora-B kinase through the interaction with INCENP and autophosphorylation. This interaction is critical to the proper execution of chromosome segregation, and key details of the mechanism are not resolved. The study is an advance on previous studies on Aurora-B and the related kinase Aurora-C, primarily because it clarifies the roles of the different phosphorylation sites. However, major differences in the details of the molecular interactions are presented that are not clearly backed up by the evidence due to limitations in the approach, when compared to previous work based on crystal structures.

Strengths. The experimental approach to the analysis of the Aurora-B/INCENP interaction is sound and novel and it is striking example of preparation of proteins in specific phosphorylation states, and of using HDX to characterise localised changes in the structural dynamics of a protein complex. The authors have generated two intermediate phosphorylation states of the complex, enabling them to dissect their contributions to the regulation of structural dynamics and activity of the complex.

Weaknesses. The major weakness of the study is the molecular dynamics simulation. The resulting model of the complex differs from the crystal structure of the Aurora-C/IN-box structure in key details, and these are neither described clearly nor explained. The challenges/limitations of simulation of phosphorylated proteins should be described.

Reviewer #3 (Public Review):

The chromosomal passenger complex (CPC) is an important regulator of mitotic progression, e.g. controlling kinetochore-microtubule attachment and cytokinesis. In this manuscript, Segura-Peña and colleagues investigated how the enzymatic core complex of the CPC, Aurora B and IN-box (the C-terminal part of INCENP), is structurally and functionally regulated by multiple (auto)phosphorylations. By doing so they are providing an insightful, dynamic picture of how the coordinated phosphorylations of the Aurora B T-loop and two serines in IN-box act cooperatively in order to fully activate the kinase.

Previously, several structures of Aurora B/IN-Box (missing the C-terminus of IN-box with two important phosphorylation sites or being unstructured, Sessa et al. 2005, Sessa and Villa et al. 2015, Elkins et al., 2012) and phosphorylated Aurora C/IN-Box (Abdul Azeez et al., 2019) had provided numerous structural insights and highlighted the role of the phosphorylated residues in T-loop and IN-box. Here, the authors now reveal the dynamic dimension of how the activity of this complex is regulated by using a compelling combination of H/D exchange mass spectrometry (HDX), molecular dynamics simulation and elegant biochemistry. Using HDX they demonstrate that upon Aurora B/IN-box autophosphorylation several regions of the complex become more structured. Using molecular dynamics, they explore the different conformational states of the complex and in particular how the phosphorylation and interactions of the phosphorylated C-terminal tail of IN-box coordinates and rigidifies Aurora B. To dissect the contributions of the phosphorylations on T-loop and IN-box, the authors create differentially phosphorylated versions of the complex using a sophisticated, intein-based protein engineering approach. The biochemical assays performed with these versions reveal not only the synergistic nature of these phosphorylation sites but also establish the nature of the autophosphorylation (cis for Aurora B, trans for IN-box) and show that Aurora B autophosphorylation in cis is rate-limiting. The data is convincing and intriguing, and remaining criticisms have been addressed extensively during the rewriting of the manuscript. In my opinion no additional experiments are required.

In summary, this is a well-executed study that provides new detailed molecular insights into the regulation of an important cell cycle complex. The findings and approaches will be of great interest to both the kinase and the cell cycle community.

Reviewer #1 (Public Review):

This paper by Melo et al. is a technically elegant study investigating the important emerging hypothesis that the brainstem preBötzinger complex (preBötC) region - a critical nuclear structure where the rhythm of breathing in mammals originates - has segregated subgroups of output neurons that modulate specific behaviors coordinated with breathing, in this study the orofacial muscle activity. The preBötC has been under intense investigation for several decades but the subregional neuronal subtype composition and organization are not fully understood. Understanding this organization and how breathing modulates specific behaviors has many implications for normal brain function and pathophysiology.

Strengths of the paper include:<br /> 1) The authors use an effective combinatorial dual viral transgenic approach for Cre-dependent expression of the chloride channel (GtACR2) and labeling of neurons projecting to the facial motor nucleus controlling orofacial muscle activity, for optogenetic photoinhibition of these preBötC neurons in vivo.<br /> 2) The experimental results presented convincingly support the authors' conclusion that a subgroup of preBötC neurons provides inspiratory modulation of facial motoneurons that appear to be distinct from other output neurons that drive inspiratory activity to bulbospinal neurons and neurons projecting to autonomic nervous system circuits.<br /> 3) These results advance our understanding of preBötC circuit organization that coordinates and integrates breathing with different motor and physiological behaviors.

Weaknesses:<br /> There are a few technical issues related to the photoinhibition paradigm used and the patterns of neuronal transduction with the dual viral transgenic approach used that the authors need to clarify.

Reviewer #2 (Public Review):

By using elegant optogenetic viral transgenic approaches the authors show that subgroups of neurons located in the preBötzinger region of the brainstem and projecting to the facial nucleus are involved in controlling orofacial activity while being minimally implicated in breathing behavior. The experiments are properly performed, and technically challenging with several physiological parameters measured in vivo allowing the monitoring of several functions simultaneously (breathing, heart rate, blood pressure, orofacial muscle activity). They also demonstrate that the type of anesthetic used and the state of consciousness are important for the effects of their photoinhibition. While this study is particularly interesting for a better understanding of the coordination between breathing and other behaviours controlled by neurons located in the brainstem, the identification of the neurons of interest here as components of the preBötC network requests clarification and the interpretation of the effects of photo-inhibiting both excitatory and inhibitory neurons remain difficult.

Reviewer #3 (Public Review):

Melo et. al. sought to characterize the neuronal basis for the breathing modulation of nasal dilation (mystacial pad activity). The hypothesis is that a subset of breathing pacemaker neurons (preBötC) are specialized to relay a breathing signal to modulate the nares instead of contributing to pacing breathing. The authors identify that a subset of neurons within the anatomical region of the preBötC project to the facial motor nucleus and are required for the respiratory modulation of the nares. Furthermore, they show these neurons are partially required for breathing. The authors do this by using an intersectional genetic approach to selectively inhibit the preBötC neurons that project to the facial motor nucleus while measuring the impact of this manipulation on the breathing-related movement of the nares and breathing. As a control, the authors broadly silence the preBötC. The simplicity of the experiments makes the results robust and the correct positive control is used. The manuscript's conclusion contributes to the logic for the breathing modulation of the nares and the notion that subsets of neurons in the preBötC play distinct roles in breathing-related behaviors. Although the data are compelling for this conclusion, alternative models cannot be completely ruled out, like that these neurons are important for breathing rhythm generation and a secondary cell type from other premotor centers (Kurnikova 2019) are those that relay this signal to the motor neurons for the nares. The role of the preBötC as a "master clock" for orofacial activity (nose movement, swallowing, chewing, vocalizing; Kurnikova 2017) is an important line of research and this work contributes to understanding the cellular mechanisms.

Reviewer #1 (Public Review):

This research tackles an important question in evolutionary biology that has long stood on theory, with little experimental evidence to support this big idea. This paper provides a large natural dataset on several morphometric factors that allow a robust testing of the "handicap principle". The strength in this dataset comes from extensive field observations not only on morphology, but also fecundity and pairing behavior. The manuscript could use a little tightening up in prose, but the statistics and results are well explained. As the discussion mostly focuses on shrimp, generalizable principles are somewhat unclear. Overall, the research is an important finding that could one day be incorporated into undergraduate textbooks.

Reviewer #2 (Public Review):

This study presents important findings on trade-offs in investment in costly traits related to survival and reproduction. The evidence supporting the claims of the authors is convincing with an exceptional sample size, the inclusion of three species, and measurement of numerous traits. The authors do not incorporate genetics or use experimentation, but they do use an elegant observational approach to glean the likely presence of trade-offs and improve understanding of investment in crucial life-history traits. The work will be of interest to evolutionary biologists, researchers working in the field of animal behavior, and those specializing in sexual selection.

The extent to which individuals should invest in costly traits is an ongoing puzzle to evolutionary biologists. Why is there a limit to investment in traits that enhance survival or mating? Why do some individuals invest so much less than others in traits that should boost fitness? In this manuscript, Dinh and Patek use a strong sample size of snapping shrimp to investigate this question. They examine three species and measure numerous traits. The approach they use to deduce trade-offs is to examine residuals. Specifically, they plot the traits of interest against body size generating a regression for the population. Then, for each individual, they extract a residual value that is how much more or less they invest in a trait for a given body size. For example, some individuals might grow a big claw, but also express a small abdomen relative to others of the same size. The authors measure the extent to which each individual invests in a number of traits to investigate resource allocation trade-offs and reproductive benefits and costs.

This is an elegant and thorough study that thoughtfully examines how animals invest in their bodies and with what potential costs. They even look at male pairing success and the size of his mate to better understand the reproductive benefits of growing a larger claw in snapping shrimp. For females, they examine if growing a larger claw might lead to reduced reproduction because such females cannot care for as many eggs. The strengths of this study are many. It would, of course, be helpful to more thoroughly understand the costs and benefits of investment in claws, but the authors did an excellent job with what was possible. The current version of the manuscript would benefit from a discussion of the pros and cons of their approach of using residuals versus other approaches to measure resource allocation trade-offs.

Overall, this is such a nice study with excellent writing, and it will likely inspire others to examine trait investment in a myriad of other animals. It helps the field of sexual selection better understand the costs and benefits of growing a big (or small) weapon. And, more generally, it addresses the important question of why animals cannot have it all.

Reviewer #1 (Public Review):

Kim et al carried out a genetic screening using Drosophila lines to identify genetic modifiers of ubiquilin 2 mutations associated with ALS/FTD. They generated Drosophila lines expressing wild-type or various mutations of ubiquilin 2 and used the rough eye phenotype as the primary screening criterion. They used the deficiency library in the screening and subsequently attempted to narrow down to single genes. They identified multiple suppressors and enhancers from the deficiency lines and carried out further studies on an endosomal gene rab5, an axon guidance gene unc5 and its co-receptor frazzled, and another axon guidance gene beat-1b. Critical findings were also confirmed in iPSC and induced motor neurons (iMNs), supporting the relevance of the findings in human neurons. The study is important as it provides compelling evidence linking axon guidance/synaptic maintenance to ubiquilin 2-mediated neurotoxicity.

With the above strengths and impact, there are several weaknesses. First, the heat shock effect in the drosophila lines was not understood in the study. Why did some lines show phenotypes only at 29C but not 22C? The study showed data that ubiquilin 2 expression was not impacted by 29C, then what caused the phenotypic differences? In addition, the method section did not describe clearly whether a temperature sensitive promoter was used in the flies. Second, the study showed data on male and female flies separately in some but not all experiments. In addition, the manuscript largely avoided discussing whether there was a sex difference in those experiments. Third, some data appear to be peripheral with no significant contribution to the main findings. Moreover, some data were introduced but were not explained. For instance, the RNA-Seq analysis (Fig 2) did not contribute much to the study. The rescue effect of UBA* (F594A mutant) in Fig 1-Supplemental 1B was interesting but was not elaborated or followed up. FUS flies in Fig 6-Supplement 2 were abrupted introduced with little discussion. Fourth, the main quadrupole (4xALS) mutation used in the study was not found in patients. The relevance of the findings needs to be thoroughly justified. Lastly, ALS and FTD are age-related neurodegenerative diseases, whereas the involvement of axon guidance genes in indicative of disruptions during the developmental stage. The manuscript did not discuss this potential caveat.

Overall, this study identified several potential genetic modifiers of ubiquilin 2 in the context of ALS/FTD. It represents a significant advancement of our understanding of ubiquilin 2-mediated ALS/FTD and related neurodegenerative diseases.

Reviewer #2 (Public Review):

In the present article, the author aimed at finding disease-modifier for a disease that still nowadays is incurable. To do so the authors decided to employ a drosophila model of ALS, bearing four mutations on the Ubiquilin gene. The model displays eye and motoneuron phenotypes serving as a valuable platform for genetic screenings. The screening performed in the present work shows many suppressors and enhancers of the toxicity associated with the presence of the 4 Ubiquilin mutations. The authors then strengthen the findings of the screening by validating some hits and by studying more in details one involved in the axon guidance signaling. They found that suppressing Unc5 and DCC leads to a less severe phenotype in the flies. They then suppress the ligand of the Unc5 receptor and found that also this approach relieves the phenotype. They then confirmed this results in iPSCs by creating a new cell line harboring the four mutations. They found that the neurites defects found in the mutated UBQLN iPSC was rescued by suppressing Unc5 and DCC. This study has relevance to the ALS field as many of the findings can be harnessed to develop drugs suited for ALS patient bearing Ubiquilin mutations. I think that the major weaknesses of this paper are (i) the fact that they focus on just one mutation, which is pretty rare, while probably most of findings should be also validated in models of sporadic ALS (iPSCs lines). (ii) The amount of data presented, for as much as it is technically well-performed, does not help the reader to focus the attention of the main point which is Unc5 signaling relevance in Ubiquilin associated ALS.

Reviewer #1 (Public Review):

The manuscript by Kschonsak et al. describes the rational structure-based design of novel hybrid inhibitors targeting human Nav1.7 channel. CryoEM structure of arylsulfonamide (GNE-3565) - VSD4 NaV1.7-NaVPas channel complex confirmed binding pose observed in x-ray structure GX-936 - VSD4 Nav1.7-NavAb channel. Remarkably, cryoEM structure of acylsulfonamide (GDC-0310) - VSD4 NaV1.7-NaVPas channel complex revealed a novel binding pocket between the S3 and S4 helices, with the S3 segment adopting a distinct conformation compared to the arylsulfonamide (GNE-3565) - VSD4 NaV1.7-NaVPas channel complex. Creatively, the authors designed a novel class of hybrid inhibitors that simultaneously occupy both the aryl- and acylsulfonamide binding pockets. This study underscores the power of structure-guided drug design to target transmembrane proteins and will be useful to develop safer and more effective therapeutics.

Reviewer #2 (Public Review):

In this manuscript, the authors identify a critical unmet need for the (structure-based) drug design of human Nav channels, which are of clinical interest. They cleverly rationalized a hybrid strategy for developing target-specific small molecule inhibitors, which integrate binding mechanisms of two drug candidates that act orthogonally on the VSD4 of Nav 1.7. Thus, the authors illustrate a promising outlook on pharmaceutical intervention on Nav channels.

Overall, the cryo-EM structures of the ligand-bound Nav channels are convincing, with a clear indication of the site-specific, distinct density of the small molecules. At the moment, it is difficult to tell how innovative the pipeline is compared to conventional cryo-EM structure determination.

Reviewer #3 (Public Review):

This is an excellent manuscript, describing a few lines of discoveries:<br /> 1. Establishment of a structural biological pipeline for iterative structural determination of an engineered Nav1.7;<br /> 2. Illumination of the novel compound binding mode;<br /> 3. Structure-based development of the hybrid compounds, which led to the novel Nav1.7 inhibitor;

The cryo-EM study on the engineered Nav1.7 consistently reveals the map at the mid to low 2 Å range, which is unprecedented and impressive, thus, demonstrating the high value of this workflow. The further strength of this study is that the authors were able to develop a new compound by combining structural information gained from the two Nav1.7 structures complexed to two different compounds with different binding modes. Overall, the depth and quality of this study are excellent.

Reviewer #1 (Public Review):

In this manuscript, Chure and Cremer first provide a broad panorama of the different sector models for resource allocation in biosynthesis and how they provide an explanation of cellular growth physiology; then they formalise how optimal flux balance (flux parity) can reproduce many different physiological observables in a quantitative manner.

The first part of this study comprises a valuable synthesis of many literature results, which are here gathered together and clearly reformulated. The authors also assembled a rich and impressive collection of experimental published datasets in E. coli from several sources, which are then extensively compared with the outcomes of the models. In my view, these points are the main strengths of the manuscript.

The flux-parity regulation introduced in the second part emerges from the balance of metabolic and biosynthesis fluxes, which have to be mutually optimised in the authors' framework. Those ingredients are often found in the literature, and the reader has sometimes the impression that novelty is lacking. Although flux balance and optimisation are often assumed in modelling resource allocation, the authors have the merit of formalising the approach in a clearer way than was done before, making an extensive comparison with data.

Reviewer #2 (Public Review):

The authors propose a proteome allocation model which includes a ribosomal and metabolic sector (and an additional sector in the case of nutrient upshift or downshift), and they consider the effect of tRNA charging on translation. It appears that the rate of protein generation via translation by ribosomes and the rate of tRNA charging via metabolic proteins are mutually maximized (the so-called "flux-parity regulation"). Based on this principle, one can reproduce many aspects of bacterial growth both in and out of a steady state, without having to consider other processes.

A major strength of this article is that the authors include many different E. coli datasets. From the figures presented, the model appears to agree well with the data. If the model can indeed predict bacterial growth out of a steady state, then it will be useful in understanding how tRNA charging affects the bacterial response to environmental fluctuations.

To improve the manuscript, units and typical values in E. coli should be provided in the main text as parameters are introduced, to give the reader some benchmark numbers and physical intuition. Furthermore, how proteins are assigned to metabolic, ribosomal, or other proteome sectors can be better explained in the main text, i.e. based on the dependence of their respective abundances on the growth rate. It would also help the reader to explicitly state which parameters are being adjusted and which are fixed (four are mentioned in Section 8 of the appendix but there are many others defined in the text). Finally, whether v_max (max metabolic rate) and tau (uncharged-to-charged tRNA ratio) take on physically reasonable values is not clear, e.g. values for v_max span 4 orders of magnitude. These are essential parameters to the model, and without a sense of how they compare to real values, it is difficult to judge the robustness of the results.

Some specific questions follow:

- Are there experimental data to verify the charging sensitivity parameter tau?<br /> - Which molecules, other than charged tRNAs, are considered 'precursors', and are these neglected or accounted for in the model? For example, the other components of the ternary complex, e.g. GTP and EF-Tu, are not mentioned.<br /> - What is the yield coefficient Y in Eqs. 10, 55, Fig. S2,A(iii)? No value appears in the text or supplemental tables.<br /> - Why is the inactive fraction of ribosomes considered a puzzle? Bremer & Dennis and Metzl-Raz et al. have provided polysomal profiling data in E. coli and in S. cerevisiae, respectively. In E. coli it is ~85% but can be considerably lower in S. cerevisiae. Furthermore, it seems unphysical that 100% of ribosomes would be active at all times; it takes time for a ribosome to find and bind to mRNA.<br /> - (p)ppGpp binds to molecules other than tRNAs, e.g. RNA polymerase. Shouldn't this be accounted for in, e.g., Eq. 3?

Reviewer #1 (Public Review):

Pyrin domains (PYD) in inflammasome proteins oligomerize into filamentous assemblies and mediate inflammasome formation. Mammalian pyrin-only-proteins (POPs) exert inhibitory effects on inflammasome as they mimic the pyrin domains while lacking the effector domain. In this manuscript, Mazanek and colleagues combined computational prediction with cellular and in vitro experiments to investigate the mechanism and target specificity for three POPs, POP1, POP2, and POP3, in inflammasome activation.

The authors first modeled the structures of complex formed by POPs with inflammasomal PYDs, including ASCPYD, AIM2PYD, IFI16PYD, NLRP6PYD, and NLRP3PYD, then calculated their Rosetta interface energies(∆Gs). By comparing the ∆Gs of inflammasomal PYD(∆GPYD•PYD) with inflammasomal PYD/POPs complex (∆GPOP•PYD), they defined favorable and unfavorable interaction surfaces (∆∆G = ∆GPYD•PYD- ∆GPOP•PYD ). Their initial computational model indicates POP1 may have the strongest inhibitory effect on ASC, as it exhibits the most favorable interfaces. But the experiment results showed otherwise, with POP2 and POP3, which contain both favorable and unfavorable interfaces, exhibiting stronger inhibitory effects. They then revised the model and proposed the combination of favorable (recognition) and unfavorable interfaces (repulsion) is necessary for POPs to interfere with the assembly of inflammasome PYDs, which was further tested by other inflammasomal PYDs.

This is a timely study that enhanced our current understanding of inflammasome regulation by POPs, it is also interesting as it combined the newest computational prediction method with biological experimental validation. The explanations on 1.) sequence homology may not dictate the target specificity of POPs, and 2.) excess POPs are required to inhibit the polymerization of inflammasome assembly, are well supported; however, some questions about the target specificity need to be addressed/clarified:

1. The authors showed MBP tag affected the oligomerization of POPs, while the POPs used in Figures 2A, 3A, and 4A contain a GFP tag. It should be considered GFP may affect the property of POPs, such may change the inhibitory effect of POPs on ASC filament formation.

2. The authors take the reduction of PYD filamentation as an indication of inhibition, but it was not clear how they ruled out the possibility that POP1 co-assembles into ASCPYD filaments and inhibits inflammasome formation by repressing the recruitment of Caspase-1, as it lacks CARD the effector domain. Especially the model predicted comparable energy between POP1 and ASC, which could indicate POP1 co-assembled into ASC filament.

3. Further computational analysis should be performed to evaluate the interpretation of Rosetta interface energies. Could the "combination of favorable and unfavorable interfaces" theory apply to other PYD/PYD interactions and CARD/CARD interactions?

Reviewer #2 (Public Review):

In this manuscript, Mazanek et al use Rosetta to calculate the relative binding energies of the six distinct PYD/PYD interactions between the pyrin-only proteins (POPs) and the pyrin domains (PYD) of various inflammasome components. Following these calculations, the authors measure the ability of the POPs to disrupt PYD spec formation or disrupt PYD oligomerization. From these experiments the authors propose that the POPs do not simply disrupt ASC oligomerization, but instead that each POP has unique specificity for the various PYDs and can thusly act upstream of ASC filamentation through their direct interactions with the inflammasome PYDs. Furthermore, the authors propose the ability of the POPs to inhibit PYD filament formation is not solely dictated by sequence similarity between the POP and the PYDs, but instead that a combination of both strong and weak interactions between the POP and PYD is required to disrupt PYD filament formation. These observations help to elucidate the individual roles of the different POPs.

In total this manuscript presents a rigorous and careful biochemical analysis of how the POPs act to modulate PYD oligomerization. However, there are several weaknesses that need to be addressed. First, while the authors propose that the combination of strong and weak interactions dictates the ability of the POPs to disrupt PYD oligomerization this hypothesis is not directly tested. Second, while the author's careful examination demonstrates the ability of the POPs to disrupt PYD spec formation in a reconstituted system, they do not confirm that their in vitro measurements correlate with the ability to restrict inflammasome activity in an endogenous system and as such the physiological consequences of their measurements remain unclear.

Reviewer #3 (Public Review):

The authors use a combination of computational and experimental analyses to study how Pyrin-only proteins (POPs) could regulate either the abundant ASC effector protein or the PYDs of ALRs AIM2 and IFI16 or NLRs NLRP3 and NLRP6. This systematic approach shows differences in the free energy of binding interfaces within the potential filament assemblies. Fluorescence anisotropy experiments are performed on PYD filament formation, using FRET-donor and -acceptor labeled recombinant PYDs (e.g., ASC) and increasing concentrations of unlabeled POPs. These experiments indicate how the lag phase of PYD nucleation and the kinetics of the filament elongation phase is perturbed. Fluorescence microscopy images of HEK cells co-transfected with, e.g., mCherry-tagged ASC-PYD and eGFP-labelled POPs indicate co-localization and overall filament content (as % puncta). Finally, negative stain EM imaging shows assemblies into ordered filaments or aggregates for the recombinant PYD proteins in the presence or absence of POPs. In conclusion, the authors propose a decoy receptor mechanism for the POPs and NLRs/ALRs with different specificities for each individual PYD.

Reviewer #1 (Public Review):

In this manuscript, McQuate et al. use serial block face SEM to provide a high resolution, 3D analysis of mitochondrial structure in hair cells and surrounding supporting cells of the zebrafish lateral line. They first demonstrate that hair cells have a higher mitochondrial volume as compared to supporting cells, which likely reflects the high metabolic load of these sensory cells. Their deeper analysis of mitochondrial morphology in hair cells reveals that the base of the hair cell - near the presynapse is dominated by a large, networked mitochondrion, while the apex of the cell is dominated by many small mitochondria. By examining hair cells at different stages of development, the authors show that specialized features of hair cell mitochondria are gradually established over the course of development. Finally, by examining hair cells in mutants that lack mechanosensation or presynaptic calcium responses, McQuate et al. reveal that cellular activity contributes to the development of appropriate mitochondrial morphology and localization within hair cells. This dataset, which will be made publicly available, is an immense resource to the community and will facilitate the generation of novel hypotheses about hair cell mitochondrial function in health and disease.

Strengths:<br /> 1. The painstaking acquisition and analysis of hair cell EM data in a genetically tractable system that is easily accessible for in vivo functional experiments to address hypotheses that emerge from this work.<br /> 2. The use of multiple datasets and analysis methods to cross-validate results.<br /> 3. The thoughtful, careful analysis of the data highlights the richness of the dataset.<br /> 4. The use of both wild-type and mutant animals substantially adds to the manuscript, providing significantly more insight than wild-type data alone.

Weaknesses:<br /> 1. The manuscript could more strongly highlight the utility of this dataset and facilitate its future use by providing a summary table that lists each sample together with salient details.<br /> 2. The authors examine an opa-1 mutant with altered mitochondrial fission (which consequently has changes in mitochondrial morphology and organization) to suggest that aberrant mitochondrial architecture negatively impacts mitochondrial function. However, mitochondrial fusion is thought to be critical for mitochondrial health beyond just altered architecture. Because fusion has other roles, it is difficult to use this manipulation to conclude that it is simply disruptions in mitochondrial architecture that alters function.

3. Although the work of acquiring and reconstructing EM data is labor-intensive, ideally, multiple fish would be examined for each genotype. Readers should take into consideration that one of the mutant datasets is derived from just one animal.

Reviewer #2 (Public Review):

Sensory hair cells have high metabolic demands and rely on mitochondria to provide energy as well as regulate homeostatic levels of intracellular calcium. Using high-resolution serial block face SEM, the authors examined the influences of both developmental age and hair cell activity on hair cell mitochondrial morphology. They show that hair cell mitochondria develop a regionally specific architecture, with the highest volume mitochondria localized to the basolateral presynaptic region of hair cells. Data obtained from mutants lacking either mechanotransduction or presynaptic calcium influx provide evidence that hair cell activity shapes regional mitochondrial morphology. These observed specializations in mitochondrial morphology may play an important role in mitochondrial function, as mutants showing disrupted hair cell mitochondrial architecture showed depolarized mitochondrial potentials and impaired evoked mitochondrial calcium influx.

This work provides novel and intriguing evidence that mechanotransduction and presynaptic calcium influx play important roles in shaping subcellular mitochondrial morphology in sensory hair cells. Yet there was a lack of consistency in the analysis and presentation of the data which made it difficult to contextualize and interpret the results. This study would be greatly strengthened by i) consistent definitions for hair cell maturation, ii) comparable data analysis of cav1.3a mutant and cdh23 mutant mitochondrial morphologies, and iii) more detailed descriptions and interpretations of the UMAP analysis.

Reviewer #3 (Public Review):

McQuate et al have succeeded in reconstructing 3D images of mitochondria and discovered unique structural features of mitochondria in zebrafish hair cells. Compared to the other cell types, such as central and peripheral support cells, Hair cells have many elongated and connected mitochondria and they seem to be involved in hair cell and ribbon synapses development. These findings will contribute to understanding the mechanisms for mitochondrial network regulation.

Using the SBFSEM technique, the authors provide clear 3D images of hair cells and the technique improves the resolution of the image to understand the structural parameters of not only mitochondria but also ribbon synapses compared to typical fluorescent imaging. These results are very attractive and have the high potential to broadly apply to 3D imaging of any type of organelles, cells, and tissues. On the other hand, however, the authors provide the data from a small sample size, and the functional experiments to make a conclusion are lacking. Some missing representative images and the nonunified methods of grouping for the analysis make the reviewer concerned.

Reviewer #1 (Public Review):

The article from Dumoux et al. shows the use of plasma-based focused ion beams for volume imaging on cryo-preserved samples. This exciting application can potentially increase the throughput and quality of the data acquired through serial FIB-SEM tomography on cryo-preserved and unstained biological samples. The article is well-written, and it is easy to follow. I like the structure and the experimental description, but I miss some points in the analyses, without which the conclusions are not adequately supported.

The authors state the following:<br /> "the application of serial FIB/SEM imaging of non-stained cryogenic biological samples is limited due to low contrast, curtaining, and charging artefacts. We address these challenges using a cryogenic plasma FIB/SEM (cryo-pFIB/SEM)".<br /> Reading the article, I do not find that the challenges are addressed; it appears that some of these are evaluated when the samples are prepared using plasma-based beams. To support the fact that charging, contrast, and curtaining are addressed, a comparison should be made with the current state of the art, or it is otherwise impossible to determine whether these systems bring any advantage.

Charging is an issue that is not described in detail, nor has it been adequately analysed. The effect of using plasma beams is independent of the presented algorithm for charging suppression, which is purely image processing based, although very interesting. Given that the focus of the work is on introducing the benefit of using plasma ion beams (from the title) and given that a great deal of data is presented on the effect of the multiple ion sources, one would expect to have comparable images acquired after the surfaces have been prepared with the different beams. This should also be compared against the current state-of-the-art (gallium) to provide a baseline for different beams' benefits. I realise that this requires access to another microscope and that this also imposes controls on the detector responses on each instrument to have a normalised analysis. Still, it also provides the opportunity to quantify the benefits of each instrumentation.

The curtaining scores. This is a good way to explain the problem, though a few aspects need to be validated. For example, curtains appear over time when milling, and it would be useful to understand how different sources behave over time in FIB/SEM tomography sessions. The score is currently done from individual windows milled, which gives a good indication of the performance. However, it would make sense to check that the behaviour remains identical in an imaging setting and with the moving milling windows (or lines). This will show the counteracting effect to the redeposition and etching effect reported when imaging with the E-beam the milled face.

No detail about the milling resolution has been reported. Since different currents and beams have different cross-sections, it is expected to affect the z-resolution achievable during an imaging session. It would be useful to have a description of the beam cross-sections at the various conditions used and how or whether these interfere with the preparation.

Contrast. No analysis of plasma FIBs' benefits on image contrast compared to the current state of the art has been provided. Measuring contrast is complex, especially when this value can change in response to the detector settings. Still, attempts can be made to quantify it through the FRC and through the analysis of the image MTF (amplitude and fall off), given that membranes are the only most prominent and visible features in cryoFIB/SEM images of biological samples.

Figure S4 points out that electrons that hit the sample at normal incidence give better signal/contrast or imaging quality than when the sample is imaged at a tilt. This fact is expected to significantly affect large areas as the collection efficiency will vary across the sample, particularly as regions get further away from the optimal location. The dynamic focusing option available on all SEM will compensate for the focal change but not the collection efficiency. Even though this is a fact, the authors show a loss of resolution, which is not explained by the tilt itself. In particular, the generation of secondary electrons is known to increase with the increased tilt, and to consider that the curtains (that are the prominent feature on the surface) are running along the tilt direction, it would be expected to see no contrast difference between the background and the edge of each curtain as the generation of secondary electrons will increase with tilt for both the edges and the background. Therefore, the contrast should be invariant, at least on the curtains.

Looking at the images presented in the figure, they appear astigmatic and not properly focused when imaged at a tilt. As evidence of this claim, the cellular features do not measure the same, and the sharpness of the edge of the curtains is gone when tilted. This experience comes from improper astigmatism correction, which in turn, in scanning systems, leads to the impossibility of focusing. The tilt correction provides not only dynamic focusing but also corrects for the anisotropy in the sampling due to the tilt. If all imaging is set up correctly, the two images should show the imaged features with the exact sizes regardless of the resolution (which, in the presented case, is sufficient), and the sharpness of the curtain edges should be invariant regardless of the tilt, at least while or where in focus. Only at that point, the comparison will be fair.

Finally, the resolution measurements presented in the last supplementary figures have no impact or relation to the use of plasma FIB/SEM. It is an effect related to the imaging conditions used in the SEM regardless of the ion beam nature. The distribution of the resolution within images appears predominantly linked to local charging and the local sample composition (from fig8). Given the focus is aimed at introducing or presenting the use of the plasma-based beams the results should be presented in that optic in mind with a comparison between beams.

Reviewer #2 (Public Review):

The authors present a manuscript highlighting recent advancements in cryo-focused ion beam/scanning electron microscopy (cryo-FIB) using plasma ion sources as an alternative to positively-charged gallium sources for cryo-FIB milling and volumetric SEM (cryo-FIB/SEM) imaging. The authors benchmark several sources of plasma and determine argon gas is the most suitable source for reducing undesirable curtaining effects during milling. The authors demonstrate that milling with an argon source enables volumetric imaging of vitrified cells and tissue with sufficient contrast to gleam biological insight into the spatial localization of organelles and large macromolecular complexes in both vitrified human cells and in high-pressure frozen mouse brain tissue slices. The authors also show that altering the sample angle from 52 to 90 degrees relative to the SEM beam enhances the contrast and resolution of biological features imaged within the vitrified samples. Importantly, the authors also demonstrate that the resolution of SEM images after serial milling with argon and nitrogen plasma sources does not appear to significantly affect resolution, suggesting that resolution does not vary over an acquisition series. Finally, the authors test and apply a neural network-based approach for mitigating image artifacts caused by charging due to SEM imaging of biological features with high lipid content, such as lipid droplets in yeast, thereby increasing the clarity and interpretability of images of samples susceptible to charging.

Strengths and Weaknesses:<br /> The authors do a fantastic job demonstrating the utility of plasma sources for increased contrast of biological features for cryo-FIB/SEM images. However, they do not specifically address the lingering question of whether or not it is possible to use this plasma source cryo-FIB/SEM volumetric imaging for the specific application of localizing features for downstream cryo-ET imaging and structural analyses. As a reader, I was left wondering whether this technique is ideally suited solely for volumetric imaging of cryogenic samples, or if it can be incorporated as a step in the cellular cryo-ET workflow for localization and perhaps structure determination. Another biorxiv paper (doi.org/10.1101/2022.08.01.502333) from the same group establishes a plasma cryo-FIB milling workflow to generate lamella of sufficient quality to elucidate sub-nanometer reconstructions of cellular ribosomes. However, I anticipate the real impact on the field will be from the synergistic benefits of combining both approaches of volumetric cryo-FIB/SEM imaging to localize regions of interest and cryo-ET imaging for high-resolution structural analyses.

Another weakness is the lack of demonstration that the contrast gained from plasma cryo-FIB/SEM is sufficient to apply neural network-based approaches for automated segmentation of biological features. The ability to image vitrified samples with enhanced contrast is huge, but our interpretation of these reconstructions is still fundamentally limited in our ability to efficiently analyze subcellular architecture.

Reviewer #3 (Public Review):

The authors present analyses of cryo-plasma FIB/SEM hardware for practical use in the field of cell and tissue biology at microscopic resolutions. The results include several practical analyses and considerations for structural biologists when imaging their specimens; details are provided for optimizing imaging parameters and some image processing. Several examples of pFIB-milling cells and tissues are shown. The authors also introduce a method for quantifying curtaining, one of the major artifacts in FIB/SEM imaging, and software for reducing streaking artifacts in images. The analyses in the manuscript appear to come to conclusions that are experimentally justified. I see no major weaknesses in this manuscript.

Reviewer #1 (Public Review):

Taliani et al. have studied the role of the lncRNA pCharme during cardiac development. pCharme knockout-mice present hyperplastic hearts and the authors attempt to decipher whether this cardiac phenotype result from a developmental alteration during heart formation. They showed that pCharme is specifically expressed in the heart from early stage of development at heart tube stage and persists until birth while its expression decreases after birth. The expression of pCharme in early cardiac progenitors is regulated by the transcription factor Tbx5. Several genes and signaling pathways are differentially affected in pCharme mutant hearts and most of them affected cell cycle activity and cardiac differentiation. pCharme is required to form chromatin aggregates including the MATR3 protein and this sounds important to regulate cardiac gene transcription.

One issue concerns the description of the cardiac phenotype in pCharme mutant embryos as immunofluorescent data are difficult to interpret. A deeper investigation of the level of compaction and hypotrabeculation is required to affirm that pCharme plays a role in the ventricular wall differentiation/maturation.

Another issue is that the cardiac phenotype in pCharme is not directly related to that observed in MATR3 mutants.

Reviewer #2 (Public Review):

Charme is a long non-coding RNA reported by the authors in their previous studies. Their previous work, mainly using skeletal muscles as a model, showed the functional relevance of Charme, and presented data demonstrating its nuclear role, primarily via modulating the sub-nuclear localization of Matrin 3 (MATR3). Their data from skeletal muscles suggested that loss of the intronic region of Charme affects the local 3D genome organization, affecting MATR3 occupancy and this gene expression. Loss of Charme in vivo leads to cardiac defects. In this manuscript, they characterize the cardiac developmental defects and present molecular data supporting how the loss of Charme affects the cardiac transcriptome repertoire. Specifically, by performing whole transcriptome analysis in E12.5 hearts, they identify gene expression changes affected in developing hearts due to loss of Charme. Based on their previous study in skeletal muscles, they assume that Charme regulates cardiac gene expression primarily via MATR3 also in developing cardiomyocytes. They provide CLIP-seq data for MATR3 (transcriptome-wide footprinting of MATR3) in wild-type E15.5 hearts and connect the binding of MATR3 to gene expression changes observed in Charme knockout hearts. I credit the authors for providing CLIP seq data from in vivo embryonic samples, which is technically demanding.

Major strengths:

Although, as previously indicated by the authors in Charme knockout mice, the major strength is the effect of Charme on cardiac development. While the phenotype might be subtle, the functional data indicate that the role of Charme is essential for cardiac development and function. The combinatorial analysis of MATR3 CLIP-seq and transcriptional changes in the absence of Charme suggests a role of Charme that could be dependent on MATR3.

Weakness:

(i) Nuclear lncRNAs often affect local gene expression by influencing the local chromatin. Charme locus is in close proximity to MYBPC2, which is essential for cardiac function, sarcomerogenesis, and sarcomere maintenance. It is important to rule out that the cardiac-specific developmental defects due to Charme loss are not due to (a) the influence of Charme on MYBPC2 or, of that matter, other neighboring genes, (b) local chromatin changes or enhancer-promoter contacts of MYBPC2 and other immediate neighbors (both aspects in the developmental time window when Charme expression is prominent in the heart, ideally from E11 to E15)

(ii) The authors provide data indicating cardiac developmental defects in Charme knockouts. Detailed developmental phenotyping is missing, which is necessary to pinpoint the exact developmental milestones affected by Charme. This is critical when reporting the cell type/ organ-specific developmental function of a newly identified regulator.

(iii) Along the same line, at the molecular level, the authors provide evidence indicating a change in the expression of genes involved in cardiogenesis and cardiac function. Based on changes in mRNA levels of the genes affected due to loss of Charme and based on immunofluorescence analysis of a handful of markers, they propose a role of Charme in cell cycle and maturation. Such claims could be toned down or warrant detailed experimental validation.

(iv) Authors extrapolate the mechanistic finding in skeletal muscle they reported for Charme to the developing heart. While the data support this hypothesis, it falls short in extending the mechanistic understanding of Charme beyond the papers previously published by the authors. CLIP-seq data is a step in the right direction. MATR3 is a relatively abundant RBP, binding transcriptome-wide, mainly in the intronic region, based on currently available CLIP-seq data, as well as shown by the authors' own CLIP seq in cardiomyocytes. It is also shown to regulate pre-mRNA splicing/ alternative splicing along with PTB (PMID: 25599992) and 3D genome organization (PMID: 34716321). In addition, the authors propose a MATR3 depending molecular function for Charme primarily dependent on the intronic region of Charme and due to the binding of MATR3. Answering the following question would enable a better mechanistic understanding of how Charme controls cardiac development. (i) what are the proximal genomic regions in the 3D space to Charme locus in embryonic cardiomyocytes? Authors can re-analysis published Hi-C data sets from embryonic cardiomyocytes or perform a 4-C experiment using Charme locus for this purpose. (ii) does the loss of Charme affect the splicing landscape of MATR3 bound pre-mRNAs in E12.5 ventricles in general and those arising from the NCTC region specifically? (iii) MATR3 binds DNA, as also shown by authors in previous studies. Is the MATR3 genomic binding altered by Charme loss in cardiomyocytes globally, as well as on the loci differentially expressed in Charme knockout heart? Overlapping MATR3 genomic binding changes and transcriptome binding changes to differentially expressed genes in the absence of Charme would better clarify the MATR3-centric mechanisms proposed here. Further connecting that to 3D genome changes due to Charme loss could provide needed clarity to the mechanistic model proposed here.

Reviewer #3 (Public Review):

With this work, the authors build on their previous findings on the role of the long non-coding RNA, Charme. Here, the authors show that the nuclear isoform of Charme ncRNA, pCharme, is specifically expressed in cardiac myocytes from the earliest stages of cardiac development and persists in postnatal life too. The authors perform phenotypic and molecular analysis on Charme knockout hearts to demonstrate abnormal cardiogenesis in the form of cardiac hyperplasia during development which persists postnatally. pCharme also localizes with the nuclear matrix protein MATR3 to form puncta in cardiomyocytes during development, similar to what was observed in skeletal muscle and the authors provide data to show that this punctated form of MATR3 is lost in Charme KO hearts. Finally, by CLIP-seq, the authors identify other transcripts that can interact with MATR3, including pCharme, and a percentage of these are involved in cardiac development. This paper is of interest since it highlights a new non-coding player in cardiac development which could further inform how non-coding RNAs govern gene expression during specific developmental processes. However, the authors have previously shown similar studies identifying the role of pCharme and its interaction with MATR3 in skeletal muscle. While it is important to show that a similar process is occurring in a different muscle cell-type, a more in-depth analysis and discussion especially of the CLIP-seq data would further elevate the paper. Overall, these findings do extend the authors' previous work. However, the manuscript would greatly benefit from a more nuanced and in-depth discussion of their findings as to how this non-coding RNA is regulating cardiac development at a more mechanistic level.

Reviewer #1 (Public Review):

Plasmodium falciparum must decide how much resources it will invest in within-host proliferation, or divert into gametocytogenesis to ensure onward transmission to the mosquito. The authors present here an interesting new perspective to this question using longitudinal data from a single study site over 18 years which covers three distinct transmission phases: pre-decline, decline and post-decline (which reflects a high malaria transmission setting declining to a low transmission setting). Laboratory studies in gametocyte commitment are having a renaissance in recent years, however in vivo studies have lagged behind. To address this knowledge gap, the authors have quantified the transcript levels in patient samples of ap2-g (a key player in gametocyte commitment), and PfSir2a (a gene hypothesised to sense various cellular processes via metabolic regulation). They found that transcripts of both genes increase (indicating increased investment in gametocyte production) as transmission declines. Using the Luminex platform, they were then able to link gene expression directly to key inflammatory markers within the patient, showing that as malaria transmission declines, host inflammatory response changes. Adding greater depth, using unbiased lipidomics the authors then went on to link identified inflammatory response phenotypes to specific lipid species. Excitingly, they link depleted levels of host lysophosphatidylcholine (LPC) with a defined immune response state and increased gametocyte commitment in the low transmission setting. Taken together, this gives strong in vivo support for LPC as a key modulator in both parasite development and host immune response, something that to date has been mostly characterised in vitro.

Reviewer #2 (Public Review):

The manuscript by Abdirahman I. Abdi et al. examines markers of host immunity and metabolism and markers of the malaria parasite (Plasmodium falciparum) growth and transmission. As the transmission of the malaria disease is governed by the sexual forms, (gametocytes), understating the commitment process represents a major step towards the global elimination of malaria. While the study focuses on a sound, very important topic in malaria research, its findings are partially based on rather weak evidence. In particular, in some parts there is a lack of adequate correlations, inaccurate statistics and misleading statistical tests. Moreover, these analyses are poorly explained, to a degree that some conclusions seem a bit enforced. In addition, the multitude of terms used makes it hard for the reader to follow the text. The appeal of this study lies in its potential relevance to the global public health drive to eliminate malaria.

Reviewer #3 (Public Review):

The authors present an association study geared to examine how epigenetic regulation of sexual commitment, immune responses and parasite growth change within a region that has undergone dramatic changes in transmission patterns over time. The work builds on previous epidemiological studies suggesting lower transmission settings result in parasites increasing sexual commitment, and most notably, examines mechanisms underlying these trends. The work shows the first in vivo association between LysoPC and gametocyte commitment (previously shown in vitro) in a large patient cohort. It also shows some very interesting trends relating LPC and parasite epigenetic markers to patient immune reactions.

The strengths of this paper include the use of a large patient cohort from a single geographic region, across distinct transmission intensities - an intrinsically exciting way of studying.<br /> The combination and integration of Luminex, RT-PCR, lipidomics, and clinical data provide a rich dataset for understanding host and parasite factors and provide novel in vivo evidence to support a role for LysoPC in commitment to gametocytogenesis.

In terms of weaknesses, by its nature as an association study it is difficult to ascribe causation to the patterns of seen. However, the work is built around testing of clearly defined hypothesis (based on both in vitro and clinical data) and has enabled the development of sound and exciting models for testing in future work.

The work is well-designed and written, and the conclusions fully align with the data presented. The one minor contention with the description of data is the discussion of Fig 4C-E. The manuscript states "Indeed, LPC species showed a negative association with both ap2-g and Pfsir2a transcription levels (Fig.4C-E). The association was only significant in our data when inflammation is highest (and LPC level lowest), which is at low transmission (i.e., post decline)." There is in fact only an association in post decline samples and very clearly no association pre decline. This could be made clearer here and also in the discussion (L217). This is a minor point of clarity - the work remains a compelling addition to our understanding of sexual commitment of malaria parasites.

Reviewer #1 (Public Review):

In this manuscript, Wei & Robles et al seek to estimate the heritability contribution of Neanderthal Informative Markers (NIM) relative to SNPs that arose in modern humans (MH). This is a question that has received a fair amount of attention in recent studies, but persistent statistical limitations have made some prior results difficult to interpret. Of particular concern is the possibility that heritability (h^2) attributed to Neanderthal markers might be tagging linked variants that arose in modern humans, resulting in overestimation of h^2 due to Neanderthal variants. Neanderthal variants also tend to be rare, and estimating the contribution of rare alleles to h^2 is challenging. In some previous studies, rare alleles have been excluded from h^2 estimates.

Wei & Robles et al develop and assess a method that estimates both total heritability and per-SNP heritability of NIMs, allowing them to test whether NIM contributions to variation in human traits are similar or substantially different than modern human SNPs. They find an overall depletion of heritability across the traits that they studied, and found no traits with enrichment of heritability due to NIMs. They also developed a 'fine-mapping' procedure that aims to find potential causal alleles and report several potentially interesting associations with putatively functional variants.

Strengths of this study include rigorous assessment of the statistical methods employed with simulations and careful design of the statistical approaches to overcome previous limitations due to LD and frequency differences between MH and NIM variants. I found the manuscript interesting and I think it makes a solid contribution to the literature that addresses limitations of some earlier studies.

My main questions for the authors concern potential limitations of their simulation approach. In particular, they describe varying genetic architectures corresponding to the enrichment of effects among rare alleles or common alleles. I agree with the authors that it is important to assess the impact of (unknown) architecture on the inference, but the models employed here are ad hoc and unlikely to correspond to any mechanistic evolutionary model. It is unclear to me whether the contributions of rare and common alleles (and how these correspond with levels of LD) in real data will be close enough to these simulated schemes to ensure good performance of the inference.

In particular, the common allele model employed makes 90% of effect variants have frequencies above 5% -- I am not aware of any evolutionary model that would result in this outcome, which would suggest that more recent mutations are depleted for effects on traits (of course, it is true that common alleles explain much more h^2 under neutral models than rare alleles, but this is driven largely by the effect of frequency on h^2, not the proportion of alleles that are effect alleles). Likewise, the rare allele model has the opposite pattern, with 90% of effect alleles having frequencies under 5%. Since most alleles have frequencies under 5% anyway (~58% of MH SNPs and ~73% of NIM SNPs) this only modestly boosts the prevalence of low frequency effect alleles relative to their proportion. Some selection models suggest that rare alleles should have much bigger effects and a substantially higher likelihood of being effect alleles than common alleles. I'm not sure this situation is well-captured by the simulations performed. With LD and MAF annotations being applied in relatively wide quintile bins, do the authors think their inference procedure will do a good job of capturing such rare allele effects? This seems particularly important to me in the context of this paper, since the claim is that Neanderthal alleles are depleted for overall h^2, but Neanderthal alleles are also disproportionately rare, meaning they could suffer a bigger penalty. This concern could be easily addressed by including some simulations with additional architectures to those considered in the manuscript.

Reviewer #2 (Public Review):

The goal of the work described in this paper is to comprehensively describe the contribution of Neanderthal-informative mutations (NIMs) to complex traits in modern human populations. There are some known challenges in studying these variants, namely that they are often uncommon, and have unusually long haplotype structures. To overcome these, the authors customized a genotyping array to specifically assay putative Neanderthal haplotypes, and used a recent method of estimating heritability that can explicitly account for differences in MAF and LD.

This study is well thought-out, and the ability to specifically target the genotyping array to the variants in question and then use that information to properly control for population structure is a massive benefit. The methodology also allowed them to include rarer alleles that were generally excluded from previous studies. The simulations are thorough and convincingly show the importance of accounting for both MAF and LD in addition to ancestry. The fine-mapping done to disentangle effects between actual Neanderthal variants and Modern human ones on the same haplotype also seems reasonable. They also strike a good balance between highlighting potentially interesting examples of Neanderthal variants having an effect on phenotype without overinterpreting association-based findings.

The main weakness of the paper is in its description of the work, not the work itself. The paper currently places a lot of emphasis on comparing these results to prior studies, particularly on its disagreement with McArthur, et al. (2021), a study on introgressed variant heritability that was also done primarily in UK Biobank. While they do show that the method used in that study (LDSR) does not account for MAF and LD as effectively as this analysis, this work does not support the conclusion that this is a major problem with previous heritability studies. McArthur et al. in fact largely replicate these results that Neanderthal variants (and more generally regions with Neanderthal variants) are depleted of heritability, and agree with the interpretation that this is likely due to selection against Neanderthal alleles. I actually find this a reassuring point, given the differences between the variant sets and methods used by the two studies, but it isn't mentioned in the text. Where the two studies differ is in specifics, mainly which loci have some association with human phenotypes; McArthur et al. also identified a couple groups of traits that were exceptions to the general rule of depleted heritability. While this work shows that not accounting for MAF and LD can lead to underestimating NIM heritability, I don't follow the logic behind the claim that this could lead to a false positive in heritability enrichment (a false negative would be more likely, surely?). There are also more differences between this and previous heritability studies than just the method used to estimate heritability, and the comparisons done here do not sufficiently account for these. A more detailed discussion to reconcile how, despite its weaknesses, LDSR picks up similar broad patterns while disagreeing in specifics is merited.

In general this work agrees with the growing consensus in the field that introgressed Neanderthal variants were selected against, such that those that still remain in human populations do not generally have large effects on phenotypes. There are exceptions to this, but for the most part observed phenotypic associations depend on the exact set of variants being considered, and, like those highlighted in this study, still lack more concrete validation. While this paper does not make a significant advance in this general understanding of introgressed regions in modern populations, it does increase our knowledge in how best to study them, and makes a good attempt at addressing issues that are often just mentioned as caveats in other studies. It includes a nice quantification of how important these variables are in interpreting heritability estimates, and will be useful for heritability studies going forward.

Reviewer #1 (Public Review):

Overall, this is a well-written and well-executed study that addresses the in vivo and in vitro functions of PCM1, a key component and regulator of centriolar satellites previously implicated in centrosome and ciliary biogenesis and function. The authors first generated mice lacking PCM1 and through careful phenotypic characterization, they demonstrate a tissue- and cell-type specific role for PCM1 in ciliogenesis in vivo, including a role in ciliogenesis in multiciliated ependymal cells but not airway epithelial cells. Consistently, Pcm1-/- mice were demonstrated to display perinatal lethality and select ciliopathy phenotypes such as hydrocephalus. Using high resolution immunofluorescence imaging and electron microscopy, the authors provide evidence that PCM1 promotes early stages of ciliogenesis, specifically removal of the CP110 capping protein from the distal end of (mother) centrioles. They go on to investigate this in more detail using cultured mouse embryonic fibroblasts (MEFs) and RPE1 cells lacking PCM1. Intriguingly, they find that PCM1 is required for ciliogenesis in RPE1 cells but not in MEFs, even though CP110 levels at the mother centriole are elevated in both cell types when PCM1 is depleted. The authors propose that PCM1 promotes ciliogenesis in select cell types by "wicking away" CP110 from the mother centriole at the onset of ciliogenesis, and provide some additional evidence (e.g. co-immunoprecipitation and live cell imaging analysis) to support this model. The manuscript represents a significant amount of high-quality work, and most of the claims are justified by the data. However, the manuscript would be strengthened by addressing the following points:

1) Based on their results, including the observation that CP110 and CEP97 centrosomal levels are increased in PCM1-/- cells, the authors propose that PCM1 promotes ciliogenesis by mediating removal/"wicking away" of CEP97 and CP110 from the mother centriole at the onset of ciliogenesis (Figure 9). Although this model could explain the authors' observations, alternative models should be considered. For example, an equally plausible mechanism is that PCM1 promotes centrosome/mother centriole recruitment of an E3 ligase that (negatively) regulates CP110. Indeed, the authors show in Fig. 4 that MEFs lacking PCM1 display reduced centrosome levels of the E3 ligase MIB1. This raises the question if MIB1 is also reduced at the centrosome in RPE1 cells lacking PMC1, and whether other E3 ligases known to promote CP110 removal/degradation are also decreased at the mother centriole of PCM1-/- cells. This includes EDD1/UBR5, which was previously implicated in CP110 removal from the mother centriole of RPE1 cells (Hossain et al. 2017; Goncalves et al., 2021), and which may be linked to centriolar satellites via CSPP-L (Shearer et al. 2018). Other relevant CP110 regulators to check include LUBAC and PRPF8, which may act in parallel with UBR5 to mediate CP110 removal from the mother centriole (Shen et al., 2021). The authors should at least discuss the possibility that PCM1 might affect the centrosome localization of these known CP110 regulators, if not address it experimentally. Finally, to confirm that reduced ciliogenesis in PCM1-/- cells is indeed due to increased levels of CP110 at the mother centriole, the authors could (partially) deplete CP110 from PCM1-/- RPE1 cells to investigate if this rescues the ciliogenesis phenotype of the mutant cells, e.g. as done recently by Goncalves et al. for CEP78-/- cells.

2) Figure 5 supplement 1A, B; lines 232-242; 430-439: the authors report that Talpid3 localization at the centrosome in PCM1 mutant cells is equivalent to that of controls. However, when looking at Figure 5 supplement 1B it seems that Talpid3 levels at the centrosome may be slightly elevated at the centrosome in the mutant cells although the change is not statistically significant. I suggest the authors specifically state this in the text, given that previous work by Wang et al. (2016) indicated that PCM1 does have an effect on centrosomal Talpid3 levels. A change in Talpid3 centrosomal level could be very small, requiring larger sample size to reach statistical significance, and different experimental conditions (fixation, permeabilization, antibody dilution etc.) could also influence the results and explain the discrepancy between the authors' observations and those of Wang et al. (2016).

3) Figure 5 supplement 1C, D: given that the authors´ results are in contrast to those of Wang et al. (2016), they should measure the actual fluorescence intensity of Centrobin at the mother centriole rather than just counting number of Centrobin foci, as they have done for e.g. CP110.

4) The observed requirement for PCM1 in promoting ciliogenesis in RPE1 cells and not MEFs is puzzling, given that the authors still observed increased CP110 levels at the mother centriole in the Pcm1-/- MEFs. In the discussion (lines 464-473), the authors suggest that CP110 removal from the mother centriole may be more important for ciliogenesis in cells using the "extracellular" pathway of ciliogenesis compared to cells forming cilia via the "intracellular" pathway. However, mouse fibroblasts and RPE1 cells were shown to both form cilia via the "intracellular" pathway (e.g. see Ganga et al. 2021) thus this explanation seems insufficient to explain the observed differences between RPE1 cells and MEFs lacking PCM1. It would be helpful if the authors could comment on this.

Reviewer #2 (Public Review):

Centriole satellites are membraneless granules that surround the centrosome. Some proteins localize exclusively to centriole satellites, while others are present at both satellites and the centrosome. The function of centriole satellites is somewhat mysterious, but they have been implicated in ciliogenesis, autophagy, and mediating cellular stress responses. PCM1 is a core scaffolding protein essential for the assembly of centriole satellite and many studies have examined the role of centriole satellites in PCM1 depleted cell lines. However, the role of centrosome satellites at the organismal level has not been examined, and it remains unclear if the effects observed in cell lines are present across diverse cell types found in vivo.

In this manuscript, Hall et al., examine the effect of PCM1 knockout in mice. Surprisingly, Pcm1-/- mice are viable but exhibit increased perinatal lethality. Mice lacking PCM1 also have many interesting phenotypes, including dwarfism, male infertility, hydrocephaly, and hydronephrosis. These phenotypes are consistent with defects occurring in both primary and motile cilia. The ciliogenesis deficits in Pcm1-/- mice must be relatively mild, as severe defects in cilia assembly result in embryonic lethality. Thus, centriole satellites are not required for cilia assembly in most cell types. Consistently, the authors show that Pcm1-/- MEFs have no apparent phenotypes in cilia assembly. Pcm1-/- multiciliated ependymal cells have a delay in ciliogenesis and defects in cilia beating. Surprisingly, given the array of interesting phenotypes to examine in the mice, the authors switch to characterizing PCM1-/- RPE1 cells. Unlike primary MEFs, PCM1-/- RPE1 cells show reduced ciliogenesis. The authors show that in RPE1 cells, PCM1 promotes the recruitment of preciliary vesicles to the mother centriole and helps remove the CP110/CEP97 centriole capping complex. The authors propose that CP110 and CEP97 are transported away from mother centrioles by centriole satellites. However, Pcm1-/- MEFs also fail to remove CP110 from the mother centriole, despite having no defects in ciliogenesis. Thus, CP110 removal is not universally required for ciliogenesis.

This is an excellent manuscript that thoroughly examines the role of PCM1 both in vivo and in vitro. In my view, the major strength of this work lies in the examination of the impact of PCM1 loss in vivo. As a result, I was a little surprised the authors didn't focus more attention on the interesting phenotypes that arise in the Pcm1-/- mouse. The switch over to RPE1 cells is abrupt. Moreover, the phenotypes observed in this cell line are likely not occurring in most cell types in vivo, or else the expected organismal phenotypes would probably be even more severe. That notwithstanding, the RPE1 cell biology is rigorous, high quality, and the conclusions are well-justified. Overall, the work will be of broad interest to the centrosome/cilia community.

Reviewer #3 (Public Review):

The manuscript by Hall et al., first describes the global and multi-organs phenotype of PCM1-/- mice and then focus on the role of PCM1 in the process of basal body production/maturation in multiciliated cells and finally on the role of PCM1 in primary ciliogenesis on RPE1 and MEF cells. In multiciliated cells, they show that the absence of PCM1 delays basal body formation and that PCM1 is required for the formation of structurally normal cilia, and for their consecutive coordinated beating. As regards to primary ciliogenesis, they show that PCM1 is required to allow efficient ciliation in RPE1 but not in MEF cells. Notably, they reveal defects in the formation of the preciliary vesicle in RPE1 cells and propose that PCM1 restricts CP110 and Cep97 at the centrosomal centriole in both MEFs and RPE1.

The study presented here represents a lot of nice work and highlights original data. However, in its present form, the study, which covers many aspects of the PCM1 mouse phenotype, is too fragmentary and does not allow to have, either a global view of the diversity of the phenotypes, or give mechanistic insight into one of the phenotypes. I would recommend the authors make two different papers on multiciliation and primary ciliogenesis, or try to test whether both type of ciliation are affected in a common way by the absence of PCM1. For instance, the title focuses only on the last part of the paper. Below are my comments.

Global phenotype

The authors convincingly show that the absence of PCM1 during development leads to perinatal lethality, hydrocephalus, cerebellar hypoplasia, oligospermia and cystic kidneys.

Role of PCM1 in multiciliation

The authors convincingly show that the absence of PCM1delays centriole amplification and therefore multiciliation which has never been shown before to my knowledge.

They also propose that the basal bodies produced in absence of PCM1 show a problem of rotational polarity. This is not fully supported by the data. To confirm this observation, the authors should look at later time points as P3 is very early and the rotational polarity is progressively established after BB docking and the beginning of cilia beating. Also many more cells should be analyzed. Since this is a lot of work by EM, one should consider doing it by immunostainings as done in some other papers. Same comment for the absence of ciliary pocket in PCM1 KO. P3 is too early and since some cilia do not show a clear ciliary pocket, one should look in a sufficient number of EM sections.

The defect in translational polarity is interesting and has never been described before. This phenotype is analyzed at P5 and should also be confirmed at later time point since the delay in multiciliation in the PCM1 KO may affect the number of cells with a terminal differentiated state and therefore bias the result. In fact, migration of BB is the last event occurring during multiciliation.

The phenotype of cilia beating uncoordination is convincing and confirms what has been also described by Zhao et al., in 2021. The authors seem to propose a causality link between this phenotype and the proteomic study between WT and PCM1 KO in another MCC cell type: mTEC at ALID7. Since the difference resolve in these mTEC at ALID21, do the authors think the delay in cilia motility protein expression could explain a consecutive permanent problem of cilia beating coordination seen at later stages ? Also it is difficult to link these results with motility since motility is assessed in ependymal cilia and proteomic study in mTEC. One would like to know if motility is also affected in mTEC. And to use the proteomic study to propose an additional explanation of the one proposed by Zhao et al. showing that PCM1 depletion also deregulates the centriolar and ciliary targeting of satellites client proteins, a process that could affect cilia beating. The structural defects of cilia seen by the authors and by Zhao et al., are also one important piece of explanation.

In vitro, MCC in PCM1 KO seem to display less cilia. Is this true in vivo in the brain? Since it is not obvious in vivo in the trachea, it would be nice to just address qualitatively whether this is the case in vivo in the brain. Also, are the number of BB affected ? Zhao et al., counted the number of BB in PCM1 siRNA treated cells and show no difference. If one would address how PCM1 affect the number of cilia, this is important to know whether less centrioles are produced or whether they fail to dock correctly at the plasma membrane. Since formation of the preciliary vesicle is affected in in RPE1 cells, it is tempting to speculate that a similar defect could arise in MCC and affect motile ciliogenesis. If the « number of cilia » phenotype is not true in vivo, one should also consider a culture artefact.

Altogether, the phenotype on multiciliation needs to be strengthened to confirm the original results and to be put into the context of the previous study done in vitro (Zhao et al., 2021).

Role of PCM1 in primary ciliogenesis

Knockdown of different satellite components have been shown to affect primary ciliogenesis (Conkar et al., 2017; Kim et al., 2008; Klinger et al., 2014; Lee and Stearns, 2013; Mikule et al., 2007; Staples et al., 2014, Kurtulmus et al., 2016). More particularly cell type dependent variability of PCM1 suppression on ciliogenesis has previously been described (Odabasi et al., 2019; Wang et al., 2016). It appears necessary to clarify in one paragraph in the introduction this bibliographic context and to put forward the unresolved questions the present study proposes to address as well as the new insights it provides on the question.

First, the two main phenotypes described here, e.g. defect in ciliary vesicle formation and defect in CP110 and Cep97 removal from the mother centrioles, are very similar to the phenotype described in WDR8 knock down (Kurtulmus et al., 2016). Is there any reason why the authors did not cite this study ? If not, and since WDR8 and PCM1 are interacting partners and are interdependent for their localization, I would suggest assessing whether PCM1 acts upstream or downstream of the WDR8-Cep135 axis. For example, I would suggest testing if WDR8 expression in PCM1 KO rescue the ciliary vesicle and CPP110/Cep97 phenotypes.

The phenotype of preciliary vesicle formation defect in PCM1 KO is convincing in RPE1 cells. I would suggest to reproduce the MyoVa staining in MEFs to detect whether, in cells forming cilia in the absence of PCM1, the ciliary vesicles are forming properly. It may be a good control and also give insight into how PCM1 affects differentially ciliogenesis in different cell types. Also, the extent of TEM analysis is difficult to assess (I did not find the « n »). TEM is important to confirm the phenotype since MyoVa is an actin-based molecular motor that plays several roles in the final stages of secretory pathways.

Then the authors propose that PCM1 promotes the transition zone formation and IFT recruitment. The data presented here support that PCM1 promotes TZ formation. However, since PCM1 absence compromises preciliary vesicle formation, one could conclude that TZ alterations are just a consequence of this defect. This needs to be discussed. Regarding recruitment of IFT and TZ components, the data presented here do not support that PCM1 promotes TZ components and IFT recruitment. In fact, TZ components are not absent in non ciliated RPE1 KO cells, just decreased, and they are present at normal levels in ciliated MEFs in absence of PCM1.

The authors propose that centriolar satellites restrict CP110 and Cep97 levels at centrioles, which promotes ciliogenesis. Defect in the removal of CP110 and Cep97 from the mother centriole are very convincing in PCM1 KO both in RPE1 and MEFs. However, the causality link between this mother centriole maturation and ciliogenesis still needs to be tested since MEFs are able to ciliate in the absence of PCM1 and in the presence of CP110. Knock down of CP110 in PCM1 KO would be needed to accurately test this hypothesis. For example, in absence of WDR8, CP110 knock down does not rescue ciliogenesis defect probably because of the upstream defect of preciliary vesicle docking (Kurtulmus et al., 2016). This could be the case also here.

Finally, the authors propose that PCM1 satellites transport CP110 and Cep97 away from the centriole. They nicely show that CP110 colocalize with satellites. By IP, they suggest that PCM1 and CP110 coIP which need to be further confirmed by another IP since the signal is really weak. They show that CP110 does not colocalize anymore to the satellites as soon as 1h after serum deprivation. If satellites were involved in removing CP110 from the mother centriole for ciliation, I would expect to see an increase in CP110 localization to the satellites, and not a decrease at this time point. The authors also measure an increase of CP110 and Cep97 at the centrioles in PCM1 KO, which would go in line with their hypothesis. However, this phenotype is the opposite of what was shown in Quarantotti 2019 in the same cell type where they show that upon PCM1 loss, CP110 was decreased at the centrosome. Together with the fact that the overaccumulation of CP110 and Cep97 illustrated by IF and measured is weak, more data are needed to support this phenotype. Altogether, the hypothesis that satellites are transporting CP110 and Cep97 away from the centrioles needs more data to be convincing.

Reviewer #1 (Public Review):

In the present manuscript, the authors investigated circuits mechanisms that underlie habituation of visually evoked escape behaviors in larval zebrafish. For eliciting escape behaviors, the authors used dark looming stimuli. Larvae habituate to repeated stimulation with dark looming stimuli. The authors decomposed a dark looming stimulus into two independent components: one that is characterized by an overall spatial expansion, and the other that represents an overall dimming within the whole visual field. The authors found that pre-exposure to just the dimming component habituates responsiveness to dark looming in a comparable fashion than repeated exposure to the full dark looming. They investigated neural mechanisms that account for this using two photon calcium imaging experiments. Based on the results, the authors propose a circuits model where a subset of inhibitory DS (dimming sensitive) neurons are incrementally potentiated by repetitive stimulation and where these neurons serve to locally depress the looming selective relay pathway.

There are two caveats in the present study. First, there exists another independent habituation pathway as habituation also occurs for spatial expansion stimuli that do not accompany dimming (checkerboard stimuli). This manuscript does not investigate neural mechanisms of this habituation pathway at all. Second, the authors performed no experiment that supports the validity of the model (i.e., no ablation experiment). These two caveats reduce the impact of the manuscript. Nevertheless, I think the manuscript is worth publishing, as the model the authors propose is interesting. The model generates a series of predictions about behavior, neural response properties and synaptic connectivity, which, I hope, will be tested in future experiments.

Reviewer #2 (Public Review):

Overall, the greatest value of this article lies in the discovery and statistics of the inhibitory components that increased in response to continuous repetitive visual stimuli and suppressed responses of the critical neurons that transmit looming information to elicit escape. Although the author proposes a possible mechanism for visual habituation in larva zebrafish, there are still some shortcomings in the circuitry level proof and data interpretation, most conclusions in Figures 1-5 have been drawn in other work and lack certain innovations. In general, the overall logic of this article is relatively complete and the content is substantial, many data are very interesting and worth further interpretation.

Reviewer #1 (Public Review):

The study by Scinicariello et al. set out to identify novel factors that controlled TTP stability and identified HUWE1 by CRISPR screening in macrophages. HUWE1 phosphorylated TTP on residues distinct from those phosphorylated by MAPKs and regulated TTP protein stability. Overall, the biochemical and cellular signaling experiments were thoughtfully designed and well executed, leading to the discovery of HUWE1 as a TTP regulator.

Reviewer #2 (Public Review):

TPP is critical for regulating the mRNA abundance of proinflammatory cytokines. Sara Scinicariello et al., identified ubiquitin E3 ligase HUWE1 function as a key regulator of the TPP degradation, which could direct the related immune responses. However, the physiological importance and their major conclusions were not fully clarified or supported by the experimental data.

Reviewer #3 (Public Review):

The manuscript by Scinicariello and collaborators examines the mechanisms regulating the cellular accumulation of the RNA-binding protein Tristetraprolin (TTP). This factor is a well-described regulator of mRNA stability. TTP binds to RNA AU-rich sequences localized in mRNA 3'Untranslated regions. As AU-rich elements are abundant in mRNA encoding pro-inflammatory factors, TTP has been described as a negative regulator of the inflammatory response.

Previous reports have described that the cellular level of TTP is modulated by phosphorylation and proteasome-dependent process (see several references in the introduction of the manuscript). Non-degradative phosphorylation-dependent ubiquitination of TTP has also been reported (Schichl et al. 2011 JBC 286:38466). This publication is not cited in the current version of the manuscript. The results of Schichl et al. seem particularly relevant for the interpretation of some of the results presented here and should be considered in the final discussion and conclusions of the present work.

In the first part of the results section, Scinicariello et al. evaluate the degradation and ubiquitination of TTP and conclude that TTP is degraded in a ubiquitin-dependent manner. By a pharmacological approach, they observed, as previously shown, that endogenous TTP is degraded by the proteasome (Fig1a). They also show that an overexpressed tagged version of TTP is degraded by the proteasome and ubiquitinated on lysine residues (Fig. 1B, C). The general conclusion of this paragraph seems premature in relation to the results presented. The ubiquitination of endogenous TTP has not been demonstrated. The type of ubiquitination detected on the overexpressed version of TTP is not characterized. This seems important in view of the results of Schichl et al. who showed non-degradative ubiquitination (K63) of TTP. The half-life of the non-ubiquitinated mutant of TTP (K→R) was not precisely compared to the half-life of the wild-type TTP protein (similar to the experiment presented in 1B). The effect of the E1 ubiquitin ligase TAk-243 on endogenous TTP levels was not tested.

In the second part, the authors identified the E3 ligase HUWE1 as a major determinant of cellular TTP protein abundance. This demonstration is first based on the identification of HUWE1 in an unbiased CRISPR/cas9 screen to identify modulators of mCherry-TTP fusion reporter accumulation upon activation of RAW 264.7 cells by LPS. While they demonstrate that TTP-HA is efficiently degraded after 3 to 7h of LPS stimulation (Fig 1B) and that the stronger decrease in mCherry-TTP fusion level occurs between 4 and 6h of LPS stimulation the screen for identification of TTP modulators is performed 16h of LPS stimulation (Fig 2A). The rationale behind this experimental setting is not explicitly described. Nevertheless, the authors convincingly demonstrate that HUWE1 is involved in the controls of TTP cellular abundance. This demonstration mainly relies on the fact that HUWE1 inactivation induced a strong increase of both mCherry-TTP fusion and endogenous TTP (Fig. 2B and C). Ablation of HUWE1 selectively decreases the abundance of a limited number of proteins including TTP (Fig. 5A). The specificity of Huwe1 effect is confirmed by the detection of a constant level of the co-expressed BFP protein upon HUWE1 depletion (fig sup. 2E). The effect of HUWE1 depletion on TTP accumulation is observed in different cell lines and primary cells (murine, human) (Fig. sup. 2G, Fig2F).<br /> In this paragraph, the demonstration that Huwe1 specifically affects the stability of TTP protein appears less robust. The authors did not directly test the effect of HUWE1 inactivation on endogenous TTP accumulation after blocking protein synthesis. This control seems important as data presented in figure 2E could result both from an effect of Huwe1 level on LPS-induced TTP synthesis and TTP degradation.

In the data presented in figure 2, it is not entirely clear what exactly the authors are referring to as "endogenous TTP". In Figure 2C endogenous TTP is detected by western blot on cells transfected with an mCherry-TTP fusion. In this case, the size difference allows unambiguous identification of the endogenous form of TTP (although one could not exclude that overexpressing a TTP fusion protein might affect the level of the endogenous protein). However, TTP and mCherry-TTP cannot be distinguished by FACS (Fig2 D and E). If cells used in the experiments shown in 2C and 2D-E are distinct, this should be mentioned more explicitly in the legend of Fig. 2. Otherwise, the detection of endogenous TTP should be performed on cells that do not express mCherry-TTP.

The third part of the manuscript aims to demonstrate that loss of Huwe1 decreases the half-life of pro-inflammatory mRNAs controlled by TTP. In my opinion, this conclusion is reliably supported by the data presented in Figure 3 and Supplementary Figure 3. As the conclusion of this paragraph refers to the effect of TTP on the stability of these mRNAs, the measurement of TNF mRNA stability (Fig. sup. 3C) should be presented in the main part of Fig. 3.

The authors then aim to demonstrate that HUWE1 regulates TTP phosphorylation and its increase is responsible for increased TTP stability. Taken together, data from fig. 1F, 2C, and 2F clearly show that a phosphorylated form of TTP is accumulated in Huwe1 deficient cells. The authors state that Fig 4E aims to identify kinases and phosphatases potentially involved in TTP stability (line 277, line 298). However, the approach used here (a measure of intracellular TTP level) cannot distinguish between increased production of TTP or a decrease in TTP degradation. Also, the result presented in fig. 4E, are not totally consistent with the results presented in 4A. Fig4D shows a similar level of endogenous TTP accumulating after 2h of LPS stimulation in Huwe1 KO and control cells while a clear difference in TTP level is observable in the same condition in fig. 4A. Could the difference in the TTP detection method (Western vs intracellular FACS) be responsible for this discrepancy? In addition, the absence of positive control for the various pharmacological treatments renders difficult the interpretation of these results, especially when the inhibitor shows no effect on TTP level (ex: CalyculinA). On this basis, the authors' conclusions for this paragraph seem partially over-interpreted.

From the data presented in figure 5, the authors conclude that HUWE1 controls only a small fraction of proteasome targets and regulates the stability of TTP paralog ZFP36L1.<br /> A comparison of protein levels in Huwe1 and Psmb7 Ko cells reveals that Huwe1 ablation significantly changes the concentrations of only a limited number of proteins (Fig. 5A). The reliability of these data is confirmed by the identification as increased proteins in the huwe1 ko of factors previously identified as targets of HUWE1 (Fig. sup. 5C). These experiments and data presented in Fig.5D show that the level of the TTP paralog ZFP36L1 accumulates in huwe1 KO cells but do not demonstrate that HUWE1 affects ZFP36L1 protein stability.

The next conclusion of the manuscript describes residues in the TTP234-278 region as important for their stability. Based on data presented in fig. 6 B and sup. 6B the authors conclude that residues S52 and 178, previously identified as regulators of TTP stability, are unlikely to be involved in HUWE1-dependent TTP accumulation. The data are only based on 2 independent experiments, one of which (fig 6B) shows a difference in TTP S52/S178 mutant in Huwe1 deficient cells as compared to wt TTP. These results seem therefore too preliminary to reliably exclude the implication of S52 and 178 on the HUWE1 accumulation of TTP.

Other data from Fig. 6 further analyze the effect of deleting different regions of the TTP protein on the accumulation of this factor in HUWE1 KO and control cells. From these data, the authors conclude (line 416) that N-terminal deletion does not affect the TTP protein level. However, TTP accumulation in Huwe1 KO cells seems mostly lost in mutant N4. As mentioned above the limited number of replicates (n=2) and the absence of a statistical test makes the interpretation of this result difficult.

Several TTP C-terminal mutants show a HUWE1-independent accumulation when compared to the wt protein (Fig6. D). Is this region identical to the unstructured region identified by Ngoc (line 1255) as a potent regulator of TTP degradation? If relevant this point should be discussed.

Reviewer #1 (Public Review):

Understanding the evolution of nitrogenases is a very important problem in the field of evolutionary biogeochemistry. Ancestral sequence reconstruction at least in theory could offer insights into how this planet alerting activity evolved from ancestors that did not reduce nitrogen. But the very many components of the nitrogenase enzyme system make this a very challenging question to answer.

This paper now demonstrates the first empirical resurrection of functional ancestral nitrogenases both in vivo and in vitro. The nodes that are resurrected are very shallow in the nitrogenase tree and do not help answer how these proteins evolved. The authors' reasoning for choosing these nodes is that they are likely compatible with the metal cluster assembly machinery of their chosen host organism, A. vinelandii. The reader is left to wonder if deeper, more interesting nodes were tried but didn't yield any activity. As the paper stands, it proves that relatively shallow nitrogenase ancestors can be resurrected, but these nodes do not yet teach us anything very fundamental about how these enzymes evolved.

Technically, this work was no doubt challenging. Genome engineering in A vinelandii is very difficult and time-consuming. This organism was chosen because it is an obligate aerobe, which makes it easier to handle than the many anaerobic bacteria and archaea that harbor nitrogenases. It does make one wonder if this choice of organism is wise: the authors themselves note that it probably has a set of specialized proteins that allow the nitrogenase to be assembled and function in the presence of oxygen. This may limit A. vinelandii's potential future ancestral reconstructions deeper in the tree, which according to the authors' reasoning probably requires different assembly machinery.

The ancestral sequence reconstruction is done in two different ways: Two out of three reconstructions are carried out with what appears to be an incorrect algorithm implemented in older versions of RaxML. This algorithm is not a full marginal reconstruction, because it only considers the descendants of the node of interest for the reconstruction. The full algorithm (implemented e.g. in PAML and the newest versions of RaxML) considers all tips for a marginal reconstruction. The fact that this was called a marginal ancestral sequence reconstruction in RaxML's manual is unfortunate - as far as I understand it is in fact just the internal labelling of nodes produced by the pruning algorithm, which is not equivalent to a marginal reconstruction. In this specific case, it is unlikely that this has led to any fundamental issues with the reconstructions (as all are functional nitrogenases, which is to be expected in this part of the tree). For the shallower of the two nodes, the authors in fact verify that they get the same experimental results if they use PAML's full implementation of a marginal reconstruction (which yields a somewhat different sequence for this node). It would have been helpful to point this RaxML-related issue out in the methods, so as to prevent others from using this incorrect implementation of the ASR algorithm.

One other slightly confusing aspect of the paper is that it contains two different maximum likelihood trees, which were apparently inferred using the same dataset, model, and version of RaxML. It is unclear why they have different topologies. This probably indicates a lack of convergence. Again, this does not cast any doubt on the uncontroversial findings of this paper that shallow nodes within the nitrogenases are also nitrogenases.

Reviewer #2 (Public Review):

The authors convincingly show that their reconstructed ancestral nitrogenases are active both in vivo and in vitro, and show similar inhibitory effects as extant/wild-type enzymes.

The conclusion that, evolutionarily, there is a "single available mechanism for dinitrogen reduction" is not well explored in the paper. This suggests a limitation of using ancestral sequence reconstruction in this instance.

Reviewer #3 (Public Review):

In this work, the authors attempt to probe the constraints on the early evolution of nitrogen fixation, the development of which presented a key metabolic transition. Given that life on Earth evolved only once (to our knowledge) which aspects were necessary and which may have taken a different course are open questions. Are there alternative forms of life, metabolic networks, or even enzymatic mechanisms that could have replaced the ones we see today, or is the space of possible biologies limited? This manuscript tests the ability of ancestrally-reconstructed molybdenum-dependent nitrogenase complexes to support diazotrophic growth in Azotobacter vinelandii, as well as in vivo and in vitro activity, which all point towards a conserved mechanism for nitrogen reduction at least since proteobacteria divergence.

This is an ambitious project, requiring multiple techniques, systems, and approaches, and the successful combination of these is one of the major strengths of this work. Using parallel techniques is an important way to be certain that the overall results are robust, and an appropriate mix of in vivo and in vitro experiments is chosen here. The manuscript should serve as a useful model for how to combine phylogenetics and biochemistry.

The nature of ASR means that a solid phylogeny and/or understanding of how robust the results are to uncertainty in reconstructed states is essential since all results flow from there. The overall phylogenetic methods used are appropriate and the system is an apt one for the technique, but there is not quite enough detail in the methods to be certain of the results. Given that only the single maximum a posteriori sequence is assayed at every 3 nodes, this may have compounding results in that the sensitivity to uncertainty in the reconstruction is increased. The authors appropriately make qualitative rather than quantitative inferences, but some hesitation towards the overall results still exists.

The assumption that the Anc1A/B and Anc2 nodes correspond to ancestral states might be undermined by horizontal gene transmission, which has been reported for nif clusters. In particular, there may be different patterns of transmission for each element of the cluster. By performing reconstruction with a concatenated alignment, the phylogenetic signal is potentially maximized, but with the assumption that each gene has an identical history. Discordant transmission may cause an incorrect topology to be recovered.

Finally, I am unsure if ASR is the most appropriate approach to answer questions of contingency and alternative pathways for protein evolution. ASR may tell what nitrogenase millions or billions of years ago looked like, but it can only say what has already existed. If there are different mechanisms or metabolic pathways enabling nitrogen fixation that simply never came to pass, via contingency and entrenchment or simple chance, ASR would say nothing about them. It is true that a conserved mechanism would point towards a constrained space for evolving nitrogen fixation, but that does not directly address it.

Overall, despite these issues, the manuscript is compellingly written and the figures are attractive and clear, and help get the major narrative across. This work will be of interest to protein biochemists of evolutionary bent and microbial physiologists with an interest in the origins of life.

Reviewer #1 (Public Review):

Because of the importance of brain and cognitive traits in human evolution, brain morphology and neural phenotypes have been the subject of considerable attention. However, work on the molecular basis of brain evolution has tended to focus on only a handful of species (i.e., human, chimp, rhesus macaque, mouse), whereas work that adopts a phylogenetic comparative approach (e.g., to identify the ecological correlates of brain evolution) has not been concerned with molecular mechanism. In this study, Kliesmete, Wange, and colleagues attempt to bridge this gap by studying protein and cis-regulatory element evolution for the gene TRNP1, across up to 45 mammals. They provide evidence that TRNP1 protein evolution rates and its ability to drive neural stem cell proliferation are correlated with brain size and/or cortical folding in mammals, and that activity of one TRNP1 cis-regulatory element may also predict cortical folding.

There is a lot to like about this manuscript. Its broad evolutionary scope represents an important advance over the narrower comparisons that dominate the literature on the genetics of primate brain evolution. The integration of molecular evolution with experimental tests for function is also a strength. For example, showing that TRNP1 from five different mammals drives differences in neural stem cell proliferation, which in turn correlate with brain size and cortical folding, is a very nice result. At the same time, the paper is a good reminder of the difficulty of conclusively linking macroevolutionary patterns of trait evolution to molecular function. While TRNP1 is a moderate outlier in the correlation between rate of protein evolution and brain morphology compared to 125 other genes, this result is likely sensitive to how the comparison set is chosen; additionally, it's not clear that a correlation with evolutionary rate is what should be expected. Further, while the authors show that changes in TRNP1 sequence have functional consequences, they cannot show that these changes are directly responsible for size or folding differences, or that positive selection on TRNP1 is because of selection on brain morphology (high bars to clear). Nevertheless, their findings contribute strong evidence that TRNP1 is an interesting candidate gene for studying brain evolution. They also provide a model for how functional follow-up can enrich sequence-based comparative analysis.

Reviewer #2 (Public Review):

In this paper, Kliesmete et al. analyze the protein and regulatory evolution of TRNP1, linking it to the evolution of brain size in mammals. We feel that this is very interesting and the conclusions are generally supported, with one concern.

The comparison of dN/dS (omega) values to 125 control proteins is helpful, but an important factor was not controlled. The fraction of a protein in an intrinsically disordered region (IDR) is potentially even more important in affecting dN/dS than the protein length or number of exons. We suggest comparing dN/dS of TRNP1 to another control set, preferably at least ~500 proteins, which have similar % IDR.

Reviewer #3 (Public Review):

In this work, Z. Kliesmete, L. Wange and colleagues investigate TRNP1 as a gene of potential interest for the evolution of the mammalian cortex. Previous evidence suggests that TRNP1 is involved in self-renewal, proliferation and expansion in cortical cells in mouse and ferret, making this gene a good candidate for evolutionary investigation. The authors designed an experimental scheme to test two non-exclusive hypotheses: first, that evolution of the TRNP1 protein is involved in the apparition of larger and more convoluted brains; and second, that regulation of the TRNP1 gene also plays a role in this process alongside protein evolution.

The authors report that the rate of TRNP1 protein evolution is strongly correlated to brain size and gyrification, with species with larger and more convoluted brains having more divergent sequences at this gene locus. The correlation with body mass was not as strong, suggesting a functional link between TRNP1 and brain evolution. The authors directly tested the effects of sequence changes by transfecting the TRNP1 sequences from 5 different species in mouse neural stem cells and quantifying cell proliferation. They show that both human and dolphin sequences induce higher proliferation, consistent with larger brain sizes and gyrifications in these two species. Then, the authors identified six potential cis-regulatory elements around the TRNP1 gene that are active in human fetal brain, and that may be involved in its regulation. To investigate whether sequence evolution at these sites results in changes in TRNP1 expression, the authors performed a massively parallel reporter assay using sequences from 75 mammals at these six loci. The authors report that one of the cis-regulatory elements drives reporter expression levels that are somewhat correlated to gyrification in catarrhine monkeys. Consistent with the activity of this cis-regulatory sequence in the fetal brain, the authors report that this element contains binding sites for TFs active in brain development, and contains stronger binding sites for CTCF in catarrhine monkeys than in other species. However, the specificity or functional relevance of this signal is unclear.

Altogether, this is an interesting study that combines evolutionary analysis and molecular validation in cell cultures using a variety of well-designed assays. The main conclusions - that TRNP1 is likely involved in brain evolution in mammals - are mostly well supported, although the involvement of gene regulation in this process remains inconclusive.

Strengths:<br /> - The authors have done a good deal of resequencing and data polishing to ensure that they obtained high-quality sequences for the TRNP1 gene in each species, which enabled a higher confidence investigation of this locus.<br /> - The statistical design is generally well done and appears robust.<br /> - The combination of evolutionary analysis and in vivo validation in neural precursor cells is interesting and powerful, and goes beyond the majority of studies in the field. I also appreciated that the authors investigated both protein and regulatory evolution at this locus in significant detail, including performing a MPRA assay across species, which is an interesting strategy in this context.

Weaknesses:<br /> - The authors report that TRNP1 evolves under positive selection, however this seems to be the case for many of the control proteins as well, which suggests that the signal is non-specific and possibly due to misspecifications in the model.<br /> - The evidence for a higher regulatory activity of the intronic cis-regulatory element highlighted by the authors is fairly weak: correlation across species is only 0.07, consistent with the rapid evolution of enhancers in mammals, and the correlation in catarrhine monkeys is seems driven by a couple of outlier datapoints across the 10 species. It is unclear whether false discovery rates were controlled for in this analysis.<br /> - The analysis of the regulatory content in this putative enhancer provides some tangential evidence but no reliable conclusions regarding the involvement of regulatory changes at this locus in brain evolution.