Author Response

Reviewer #1 (Public Review):

Overview

In this work, the authors set to study the effects of topographic connectivity in a hierarchical model of neural networks. They hypothesize that the topographic connectivity, often observed in cortical networks, is essential for signal propagation and allows faithful transmission of signals. To study the effects of topographic connectivity on the dynamics, the authors consider a network composed of several layers. Each layer is a recurrent neural network with excitatory and inhibitory sub-populations. The excitatory neurons in each layer enervate a sub-population of the following layer. The receiving excitatory sub-population targets a specific group in the next layer and so on. This procedure leads to separate channels that carry the inputs through the network. The authors study how the degree of specificity in each targeted projection, called ’modularity,’ affects signal propagation through the network.

The authors find that the network reduces noise above a critical level of network modularity: the deep layers show a clear separation of an active channel and inactive channels, despite the noisy input signal. They study how different dynamical and structural properties affect the signal propagation through the network layers and suggest that the dynamics can implement a winnertakes-all computation.

We thank the reviewer for the concise summary of our work.

Strengths and novelty

Topographic projections, in which sub-populations of neurons target specific cells in efferent populations, are common in the central nervous system. The dynamic and computation benefits of this organization are not fully understood. With their simple model, the authors were able to quantify the amount of topographic structure and selectivity in the network and study its impact on the network’s steady-state. In particular, a bifurcation point suggests a qualitative difference between networks with and without sufficient topographic modularity. The theoretical analysis in the paper is rigorous, and the mean-field study shows good agreement with computer simulations of the model.

We thank the reviewer for acknowledging the rigor of our work both in terms of theory and simulations.

The authors describe simulation results of networks with different dynamical properties, including rate-based networks, integrate-and-fire neurons, and more realistic conductancebased spiking neurons. All simulations exhibit similar qualitative behavior, supporting the conclusion that the behavior due to structural modularity will carry to more complex and biologically relevant neural dynamics.

Overall, the authors convince that the topographic structure of the network can lead to noise reduction, given that the input to the network is provided as distinct channels.

Weaknesses

The authors support their hypothesis and show a relation between topographic connection and noise reduction in their model. However, I find the study limited and struggle to see the impact it will have on the field. The paper is purely theoretical; it does not provide any physiological evidence that supports the conclusion. On the other hand, and this is the key issue, I do not find real theoretical insights in this work. In the following, I elaborate on why I hold this opinion.

We understand the reviewer’s point and therefore significantly extended our theoretical results and their conclusions in the revised manuscript (see below). We are confident that the revised manuscript provides the theoretical insights that the reviewer was asking for.

The hypothesis is that topographic projections in cortical areas allow faithful signal propagation. However, as the authors point out, reliable transmission can be achieved in other ways, such as by direct routing of information (lines 17-19). Furthermore, denoising can be accomplished by a simple feedforward network (e.g., ref 38) without E/I balance and with plasticity rules that do not require topographic connectivity. Thus, I find the computational model not well motivated.

The reviewer mentions an important point that has not been sufficiently addressed in the previous version, namely the distinguishing feature of our model. Direct routing is indeed a simple way to transmit signals, but without the possibility of denoising them. The reviewer is also right that the denoising solution in the work by Kadmon and Sompolinsky (ref 38) does not require any topographic connectivity. However, their model does not constrain feedforward connections between layers in any way. In particular, neurons can excite and inhibit other neurons (i.e., ignoring Dale’s law) in downstream layers so that feedforward input covers a much wider range, thereby extending the activity range of the target neurons and generating fixed points more easily. In the biologically more plausible setting that we study (excitatory and inhibitory populations, excitatory background input and excitatory feedforward connectivity), we find that recurrent inhibition is crucial to compensate the excitation from previous layers and the external input. Only if the recurrent inhibition is sufficiently strong does the topographic organization of feedforward connections enable denoising. This is addressed in a new section ”Critical modularity for denoising” of the revised manuscript, where we also study the case of no recurrent connectivity and excitatory recurrent connectivity (for further details, see answers below). We further extended our discussion on other forms of signal transmission and denoising (see lines 489-498).

The task studied here is a simple classification of static inputs: the efferent readout needs to identify the active channel. Again, this could be achieved by a single layer of simple binary neurons [Babadi and Sompolinsky 2014]. The recurrent connectivity and E/I balance suggest that dynamics should play an essential part in the model. However, the task is not well suited for understanding the role of dynamics.

We appreciate the reviewer’s comments and completely agree. The simple classification task we explored can certainly be performed by simpler network architectures, such as the one studied in Babadi and Sompolinsky. However, as discussed above, this only works if the feedforward connectivity is unconstrained. In the case of Babadi and Sompolinsky, there is an expansion of inputs into a higher dimensional space through random connectivity drawn from a centered Gaussian distribution and appropriately chosen readout weights. This scenario is not compatible with the well-established biological constraints mentioned above that our model takes into account. In the new section ”Critical modularity for denoising” of the revised manuscript we show that recurrent inhibition is necessary to enable signal transmission and denoising under these constraints. The inhibition thereby not only generates competition between input channels but it also allows the modules to track their input very rapidly (as originally demonstrated by van Vreeswijk and Sompolinsky in 1996). To demonstrate this point and emphasize the relevance of dynamics, we added a new signal reconstruction task in the new section ”Reconstruction and denoising of dynamic inputs”, where we show that our model can faithfully track and denoise spatially encoded time-varying inputs.

The authors perform a mean-field study to explain how modularity affects signal propagation. At the heart of their argument is that the E/I network exhibit bistability. However, bistability can be achieved by an excitatory population with a threshold [Renart et al., 2013]. The role of the inhibitory population does not seem crucial for the task and questions the motivations for this analysis.

We thank the reviewer for raising this important point which we address in the section ”Critical modularity for denoising” of the revised manuscript. The reviewer is correct that bistability can be obtained in a purely excitatory network, and the modular topographic connectivity in our work essentially renders the stimulated pathway excitatory. The important feature of our model, however, is that the non-stimulated pathways remain inhibitory to get a distinction between stimulated and non-stimulated populations and the denoising feature. This is only achieved by recurrent inhibition that causes competition between pathways. Our analyses show that, for networks without recurrent connections or even excitatory recurrent connections, the network lacks mechanisms to compensate the excitatory feedforward and external background input. In these cases, all populations show high (and synchronous) activity and no classification and denoising can be achieved. Therefore, the revised manuscript unambiguously demonstrates the critical role of recurrent inhibition.

Active and inactive channels are decided by the two stable states of the network: the high and the low activity regimes. However, noise fluctuations and their propagation through the network may have a prominent role in the overall dynamics. I find that noise fluctuation analysis is bluntly missing in this work.

Fig. 7b of the previous version showed the stability of theoretically predicted fixed points using numerical fluctuation analysis around the fixed points. We apologize for not having made this sufficiently clear, and have therefore updated the caption of Fig. 7 to emphasize this point and extended the subsection ”Fixed point analysis” of the Methods detailing our approach. Furthermore, we fully agree with the reviewer that fluctuation analyses are important to understand the dynamics of our system. Therefore, we performed a theoretical fluctuation analysis in the new Figure 8 and the extended Appendix B of the revised version. This extended theory shows that competition induced by recurrent inhibition stabilizes the low activity state of non-stimulated sub-populations such that fluctuations cannot build up and propagate across layers, in line with the previously presented numerical simulation results.

The main finding is a critical level of modularity, m= 0.83, above which the network shows denoising properties of silencing inactive channels and increasing the mean activity of active ones. However, the critical modularity is numerically demonstrated and is not derived theoretically. For a theoretical insight into this transition between denoising and mixing properties of the network, I would have liked to see a more rigorous discussion on the critical value. What does the critical point depend on? The authors show that the single-neuron dynamics do not affect the critical value, but what about other structural elements such as the relative efficacies of the E/I and the feedforward connectivity matrices? Do the authors suggest that m=0.83 is a universal number? I expect a more detailed analysis and discussion of this core issue in a theoretical paper.

We fully agree with the reviewer and are grateful that this point was brought up. The initial submission did not provide a sufficent or deep enough discussion on which features determine the critical modularity and it certainly is important to do so. We also apologize that our presentation was misleading and suggested a universal number for the critical modularity. Unfortunately, there is no closed form expression for the critical modularity for the non-linear activation functions shown in the previous version. We therefore added a new analysis with a fully tractable piecewise linear activation function that allows us to derive a closed-form solution for the critical modularity. The new section ”Critical modularity for denoising” and Appendix B show the results of this analysis and discuss the various parameters that affect the value of the critical modularity. In short, the reviewer was completely right that the critical modularity depends on a number of connectivity parameters as well as single-neuron properties. In particular, our theoretical results show that recurrent inhibition is crucial for denoising.

To conclude my main criticism, I believe that a theoretical paper should offer a more in-depth analysis and discussion of the core ideas presented and not rely mainly on simulations. For example, to provide theoretical insight, the authors should address central questions such as the origin of the critical modularity, the role of the recurrent balance connectivity, and how the network can facilitate computations other than winner-takes-all among channels. Alternatively, if the authors aim to describe a neural dynamics model without deep theoretical insights, I would expect to see physiological evidence supporting the suggested dynamics.

We are very grateful for the reviewer’s criticism and believe the manuscript has substantially improved as a consequence. We are confident that our revised manuscript, by addressing these issues and extending the theoretical insights, now provides a much more thorough and comprehensive understanding.

Conclusions

The model studied by the authors is novel and provides a valuable way of exploring the effects of modularity and topographic connectivity on signal propagation through hierarchical recurrent neural networks. However, the study lacks theoretical insights into cortical circuit functions in its current version. I believe that for this work to impact the field, it needs to show further analysis and not rely on a numerical study of the model with limited theoretical derivations.

Reviewer #2 (Public Review):

This manuscript puts forward a new idea that topography in neural networks helps to remove noise from inputs. The neural network consists of multiple stages. At each stage, the network is structured to be balanced in terms of the strength of inhibitory and excitatory signals. Because of topography, the networks become ”dis-balanced” and receive more recurrent excitatory signals locally for those regions that receive strong initial inputs. This leads to error correction. The main weakness in the manuscript is that the approach will only work for inputs that are constant-in-time. It is important to acknowledge this limitation in both the title and throughout the manuscript.

We thank the reviewer for the concise summary of our work and for acknowledging its novelty. Given the importance of the issue raised by the reviewer regarding the nature of the input signals, in the revised manuscript we added a new section ”Reconstruction and denoising of dynamic inputs” in which we investigate more complex, time-varying inputs and demonstrate that the model, due to the balance between excitation and inhibition, is able to quickly follow, process and denoise the external inputs. There are of course limits to the signal frequencies which can be successfully denoised, which we discuss in the Supplementary Materials (see Figure 10 - supplement 1) and elaborate on in the Discussion, but these are roughly within the ranges found in Human psychophysics studies.

Author Response

Reviewer #1 (Public Review):

In the article "Neuroendocrinology of the lung revealed by single cell RNA sequencing", Kuo et. al. described various aspects of pulmonary neuroendocrine cells (PNECs) including the scRNA-seq profile of one human lung carcinoid sample. Overall, although this manuscript does not have any specific storyline, it is informative and would be an asset for researchers exploring various new roles of PNECs.

Thank you for appreciating the significance of the data presented. Our storyline focuses on the newly uncovered molecular diversity of PNECs and the extraordinary repertoire of peptidergic signals they express and cell types these signals can directly target in (and outside) the lung, in mice and human, and in health and disease (human carcinoid tumor).

Major comments:

The major concern about the work is most results are preliminary, and at a descriptive level, conclusions or sub-conclusions are derived from scRNA-seq analysis only, lacking in-depth functional analysis and validation in other methods or systems. There are many open-end results that have been predicted by the authors based on their scRNA-seq data analysis without functional validation. In order to give them a constructive roadmap, it would be better to investigate literature and put them in a potential or probable hypothesis by citing the available literature. This should be done in each section of the result part. The paper lacks a main theme or specific biology question to address. In addition, the description about the human lung carcinoid by scRNA-seq is somehow disconnected from the main study line. Also, these results are derived from the study on only one single patient, lacking statistical power.

We agree that much of the data and analysis presented in the paper is descriptive and hypothesis-generating for PNECs, however we do not consider it preliminary. We focused on validating two key conclusions from the scRNA-seq analysis: PNECs are extraordinarily diverse molecularly (as validated by multiplex in situ hybridization and immunostaining) and they express many different combinations of peptidergic signals (and appear to package them in separate vesicles). From the lung expression profiles of the cognate receptors, we also predicted the direct lung targets of the dozens of new PNEC peptidergic signals we uncovered, and validated the cell target (PSN4, a recently identified subtype of pulmonary sensory neuron) of one of the newly identified PNEC signals (the classic hormone angiotensin) by confirming expression of the cognate receptor gene in PSN4 neurons that innervate PNECs and showing that the hormone can directly activate PSN4 neurons. The characterized human carcinoid provided evidence that during tumorigenesis, the amplified PNECs retain a memory (albeit imperfect) of the molecular subtype of PNEC from which they originated. As suggested by the Reviewer, we have provided more background in Results by adding additional citations from the literature to clarify the rationale for each analysis and what was known prior to the analysis. We feel that our paper provides a broad foundation for exploring the diversity and signaling functions of PNECs, and although each molecular type of PNEC and new PNEC peptidergic signal we uncovered and potential target cell in (and outside) the lung warrants follow up (as do the sensory and other properties of PNECs we inferred from their expression profiles), such studies will require the effort of many individuals in many labs studying both normal and disease physiology in mouse and human, and exploiting the data, hypotheses, approaches, and framework we provide.

Reviewer #2 (Public Review):

Pulmonary neuroendocrine cells (PNECs) are known to monitor oxygen levels in the airway and can serve as stem cells that repair the lung epithelium after injury. Due to their rarity, however, their functions are still poorly understood. To identify potential sensory functions of PNECs, the authors have used single-cell RNA-sequencing (scRNA-seq) to profile hundreds of mouse and human PNECs. They report that PNECs express over 40 distinct peptidergic genes, and over 150 distinct combinations of these genes can be detected. Receptors for these neuropeptides and peptide hormones are expressed in a wide range of lung cell types, suggesting that PNECs may have mechanical, thermal, acid, and oxygen sensory roles, among others. However, since some of these cognate receptors are not expressed in the lung, PNECs may also have systemic endocrine functions. Although these data are largely descriptive, the results represent a significant resource for understanding the potential roles of PNECs in normal biology as well as in pulmonary diseases and cancer and are likely to be relevant for understanding neuroendocrine cells in other tissue contexts.

However, there are several aspects of the data analysis that are unclear and require clarification, most notably the definition of a neuroendocrine cell (points #1 and #2 below).

1) Figure S1 shows the sorting strategy used for isolation of putative PNECs from Ascl1CreER/+; Rosa26ZsGreen/+ mice, and distinguishes neuroendocrine cells defined as ZsGreen+ EpCAM+ and "neural" cells defined as ZsGreen+ EpCAM-; the figure legend also refers to the ZsGreen+ EpCAM- cells as "control" cells. However, the table shown in panel D indicates that the NE population combines 112 ZsGreen+ EpCAM+ cells together with 64 ZsGreen+ EpCAM- cells to generate the 176 cells used for subsequent analyses. Why are these ZsGreen+ EpCAM- cells initially labeled as neural or control, but are then defined as neuroendocrine? If these do not express an epithelial marker, can they be rigorously considered as neuroendocrine?

As explained above in the response to Essential Revision point 1, we define pulmonary neuroendocrine cells (PNECs) throughout the paper by their transcriptomic clustering and signatures, which includes the dozens of newly identified PNEC markers as well as the few extant marker genes available before this study (listed in Table S2). The confusion here arises from the two previously known markers (Ascl1 lineage marker ZsGreen, EpCAM) we used for flow sorting to enrich for these rare cells for transcriptomic profiling (Fig. S1). Although most of the cells with PNEC transcriptomic profiles were from the ZsGreenhi EpCAMhi sorted population (as expected), some were from the ZsGreenhi EpCAMlo sorted population. The latter resulted from the high EpCAM gating threshold we used during flow sorting, which excluded some PNECs with intermediate levels of surface EpCAM. Indeed, nearly all PNECs (> 95%) expressed EpCAM by scRNAseq, and there was no difference in EpCAM transcript levels or transcriptomic clustering of PNECs that were from the ZsGreenhi EpCAMhi vs. ZsGreenhi EpCAMlo sorted populations, as we now show in the new panels (C', C'') added to Fig S1C. This point is now clarified in the legend to Fig. S1C, and it nicely demonstrates that transcriptomic profiling is a more robust method of identifying PNECs than flow sorting based on two classical markers.

2) Similarly, in the human scRNA-seq analysis, how were PNECs defined? The methods description states that these cells were identified by their expression of CALCA and ASCL1, but does not indicate whether they also expressed epithelial markers.

Human PNECs were identified in the single cell transcriptomic analysis by the same strategy described above for mouse PNECs: by their transcriptomic clustering and signatures, which includes the dozens of newly identified PNEC markers as well as the few extant marker genes available before this study (listed in Table S2). In addition to expression of classic and new markers, the human PNEC cluster defined by scRNA-seq indeed showed the expected expressed of epithelial markers (e.g, EPCAM, see dotplot below), like other epithelial cells.

3) The presentation of sensitivity and specificity in Figure 1 is confusing and potentially misleading. According to Figure 1B, Psck1 and Nov are two of the top-ranked differentially expressed genes in PNECs with respect to both sensitivity and specificity. However, the specificity of these two genes appears to be lower than that of Scg5, Chgb, and several other genes, as suggested in Figure 1C and Figure S1E. In contrast, Chgb appears to have higher specificity and sensitivity than Psck1 in Figures 1C and E but is not shown in the list of markers in Figure 1B.

As explained above in the response to Essential Revision point 2, because different marker features are important for different applications, we have provided several different graphical formats (Figs. 1B,C, Fig. S1E) and a table (Table S1) to aid in selection of the optimal markers for each application. Fig. 1B shows the most sensitive and specific PNEC markers identified by ratio of the natural logs of the average expression of the marker in PNECs vs. non-PNEC epithelial cells (Table S1), and we have added a two-dimensional plot of this sensitivity and specificity for a large set of PNEC markers (new panel E of Fig. S1). The violin plots in Fig. 1C allow visual comparison of expression of selected markers across PNECs and 40 other lung cell types including non-epithelial cells (from our extensive mouse lung atlas in Travaglini, Nabhan et al, Nature 2020). Pcsk1 and Nov score high in the analysis of Fig. 1B because they are highly sensitive and specific markers within the pulmonary epithelium, and they are also valuable markers because they are highly expressed in PNECs. However, they appear slightly less specific in the violon plots of Fig. 1C (Pcsk1) and Fig. S1F (Nov) because of expression (though at much lower levels) in individual lung cell types outside the epithelium: Pcsk1 is expressed also at low levels in some Alox5+ lymphocytes, and Nov is expressed at low levels in some smooth muscle cells. Chgb is a new PNEC marker that did not make the cutoff for the list in Fig. 1B because it is expressed in a slightly higher percentage of non-PNEC epithelial cells than the markers shown, which ranked slightly above it by this metric (see Table S1).

4) The expression of serotonin biosynthetic genes in mouse versus human PNECs deserves some comment. The authors fail to detect the expression of Tph1 and Tph2 in any of the mouse PNECs analyzed, but TPH1 is expressed in 76% of the human PNECs (Table S8). Is it possible that Tph1 and Tph2 are not detected in the mouse scRNA-seq data due to gene drop-out? If serotonin signaling by mouse PNECs is due to protein reuptake, as implied on p. 5, is there a discrepancy between serotonin expression as detected by smFISH versus immunostaining?

It is always possible that the failure to detect expression of Tph1 and Tph2 in the mouse scRNA-seq dataset is due to technical dropout, however when we analyzed this in our other mouse PNEC scRNA-seq dataset obtained using a microfluidic platform and also deeply-sequenced (Ouadah et al, Cell 2019), we found similar values as in the previously analyzed dataset: no Tph2 expression was detected and only 3% (3 of 92) of PNECs had detected Tph1 expression, whereas 24% (22 of 92) had detected expression of serotonin re-uptake transporter Slc6a4. Because our mouse and human scRNA-seq datasets were prepared similarly and sequenced to a similar depth (105 to 106 reads/cell), the difference observed in Tph1/TPH1 expression between mouse (0-3% PNECs) and human (76% PNECs) is more likely a true biological difference. We also analyzed serotonin levels in mouse PNECs by immunohistochemistry (not shown) and detected serotonin in nearly all (~90%) embryonic PNECs but only ~10% of adult PNECs. Systematic follow up studies will be necessary to resolve the mechanism of serotonin biogenesis and uptake in PNECs, and the potential stage and species-specific differences in these processes suggested by this initial data.

5) The smFISH and immunostaining analyses are often presented without any indication of the number of independent replicate samples analyzed (e.g., Figure 2B, Figure 3F, G).

The number of samples analyzed have been added (the values for Fig. 2B are given in legend to Fig. 2C, the quantification of Fig. 2B).

6) It would be helpful to provide a statistical analysis of the similarities and differences shown in the graphs in Figures 1E and G.

We added a statistical analysis (Fisher's exact test, two-sided) of Fig. 1E comparing expression of each examined gene in the two scRNA-seq datasets (Table S4). We added a similar statistical analysis of Fig. 1G comparing the expression values of each examined gene by scRNA-seq vs smFISH (see Fig. 1G legend).

Author Response

Reviewer #2 (Public Review):

SIGNIFICANCE: Movement is based on the coordinated activation and deactivation of muscle groups that depend on the timing and strength of firing of the motoneurons connected to them. Motoneuron recruitment ultimately depends on the activity of local interneurons. By difference to other CNS regions, the interneurons in the spinal cord controlling motor output display a very high diversity in genetics, anatomy, localization, and electrophysiological properties. Making sense of the interneuronal circuits that modulate motor output to the different muscles of the body has revealed to be quite complex. One technique proposed over 10 years ago is the use of retrograde transsynaptic-monosynaptic tracing with modified rabies virus injected in single muscles to define premotor connections to individual motor pools controlling single muscles. Using this technique, the original authors suggested that interneurons controlling flexors and extensors occupied different locations in the spinal cord. This idea was an extension of pioneering work from the Jessell lab at Columbia University demonstrating that positional identity determined input connectivity of motoneurons, at least from Ia afferents. This principle, if extended to premotor spinal interneurons would simplify mechanisms by which extensor and flexor interneuron networks could be connected and controlled. In this paper, the authors combine data from four independent groups to show this principle might not be correct. In other words, interneurons connected to individual motor pools are highly intermingled in the spinal cord. This raises the bar for understanding both the intrinsic organization principles of interneuron microcircuits in the spinal cord (if any) and how they develop their specific connectivity.

STRENGTHS AND WEAKNESSES: The authors propose that the conflicting conclusions occur because technical differences. The technique is based on complementation of rabies virus glycoprotein (G) in specific targeted motoneurons infected with a glycoprotein deficient rabies virus (RVdG). The way G and RVdG are delivered to specific motoneurons controlling one muscle differ. Originally this was accomplished by co-injecting RVdG and an AAV-G vectors in the same muscle simultaneously. However as previously published by a different group and now confirmed by the authors, this approach also infects muscle sensory afferents capable of transynaptically labeling populations of interneurons in the spinal cord anterogradely. This results in the labeling of mixed interneuron populations through their output to specific motor pools and/or their input from primary afferents of the same muscle. To avoid this problem the authors used transgenic approaches to induce expression of G in all motoneurons (not sensory neurons) and obtain muscle specifity by injecting RVdG in single muscles. Unfortunately, there is no single gene that selects only motoneurons for transgenic expression and tools for intersectional approaches were not available. Therefore, G is unavoidably expressed in some interneurons, in addition to motoneurons. These interneurons could be an additional source of transsynaptic jumps if they receive the RVdG from the motoneurons, raising the possibility that some labeling is the result of disynaptic, not monosynaptic, connections. The authors try to control for this possibility by comparing two different cre lines to direct G expression in motoneurons and each with different types of additional interneurons targeted. The results in both lines are similar raising confidence in the main conclusions. Moreover, the authors indicate that some motoneurons outside the intended pools are also labeled because motoneuron-to-motoneuron connections. In other words, the starter neurons for tracing monosynaptic connections are not as homogeneous or specific to a single motor pool as desired. This is acknowledged as a current limitation and is addressed in the discussion by proposing possible alternative approaches. Despite this weakness, the main conclusion of the study remains strong.

A second technical issue raised by the authors is that of possible leakage during injection in the muscle. To reduce this possibility the authors reduced the volume injected compared to previous studies from 5 to 1 microliter and checked post-hoc the injection site for possible leakage (these are neonatal pups with muscles volumes of 2-3 microliters or less). In addition, they make a nice comparison injecting different titers of RVdG to demonstrate that variations in the number of infected motoneurons of one or two orders of magnitude does not alter the main conclusion on the topographic positioning of the interneurons connected to different motor pools. One weakness is that the exact numbers of motoneurons that start the tracing is impossible to evaluate and this prevents accurate comparisons across experiments. This is because cell death induced by the rabies virus is to be expected and only a variable subset of surviving neurons can be identified. Currently, this is an unavoidable characteristic of the technique. Nevertheless, there is a nice correlation between titer, surviving motoneuron numbers and interneurons labeled in number and location. The large number of replicates and their consistency further raises confidence in the authors claim of high specificity and replicability during injection, despite variable numbers of recovered motoneurons. The authors conclude that it is very important to check for the number and localization of starter motoneurons to confirm specificity after the injections. This reviewer totally agrees and is surprised this was not done in the experiment in which they try to replicate previous experiments by co-injecting in muscle AAV-G and RVdG.

We agree with the reviewer that ideally the starter cells should have been identified in all the experiments. However, data were collected independently, at very different times in each of the labs involved, with different initial aims and there was no prior agreement on the details of injection and post-processing. The realization that we had similar experiments, performed with different techniques, led us to pool our observation together in order to give a picture of the distribution of premotor interneurons, the leitmotif of this paper, and a great effort has been devoted to ensure that all the cell counting procedures were uniform across labs. The lack of initial coordination is the reason why in some datasets the starter cells have not been quantified. Furthermore, in the previous version we had wrongly indicated that motor neurons analysed at Glasgow University were identified by ChAT expression. We have corrected this in the current version, since for those experiments motor neurons were only identified by location within any of the motor nuclei and size (diameter greater than 30 µm). On the other hand, since we have started comparing results, we have agreed on a uniform way of analysing and representing the data using the same normalization criteria. Therefore, while we cannot compare quantities like the ratio of secondary and primary infected cells for all the experiments (but we do it for the subset in which this is possible, see new Figure 4-Figure supplement 3 and comment number 3 below), the positional analysis has been done following the exact same criteria.

One final problem with interpretation is that, for yet unknown reasons, the technique is dependent on the age of the animal and cannot be implemented in mature animals. Therefore, the connectivity revealed here is the one present during the first few days of life in the mouse. This is a period of significant synaptogenesis and synaptic selection and de-selection. The authors are encouraged to discuss further this limitation when interpreting interneuron connectivity in adult from studies in neonates.

A very important point, see detailed answer to comment number 10 below.

In summary, the authors have introduced new technical variations to trace premotor interneurons and challenge a major idea in the field, that is that interneuron connected to flexors and extensors occupy different positions in the spinal cord. The technique has still some weaknesses. 1) possibility of disynaptic jumps, 2) accurate identification of starter neurons, 3) restriction to neonates. However, the authors strengthen their conclusions considering alternatives and introducing a large number of controls (two cre lines, different titers, large number of animals analyzed, large numbers and consistency of replicates, independent counting in different labs... etc). This is an important and very useful study that suggests topographic localization is not a major organizing principle for interneuron connections with motor pools. It remains to be investigated then what are the organizational mechanisms that couple interneurons to functional distinct motor pools.

The weaknesses summarized in the paragraph above are addressed in detail below in the answers to the specific comments.

Reviewer #3 (Public Review):

The manuscript by Ronzano et al presents a rigorous neuroanatomical study to convincingly demonstrate that there is no difference in the medio-lateral organization of flexor and extensor premotor interneurons. The study uses monosynaptic restricted transsynaptic tracing from ankle flexor and extensor muscles with several (4) strategies for delivery of the G protein complement and delta G Rabies virus, and additional (2) variations that consider titer and cre line. The authors went to great lengths here in attempt to replicate prior studies for which they had initial conflicting findings. Further, the experiments are performed in laboratories in four different locations. The variations on the Rabies and complement delivery, regardless of lab performing the experiment and analysis, all converge on the same conclusion. Aside from the primary conclusion, the paper can be used as a manual for anyone considering transsynaptic tracing as it details the benefits and caveats of each strategy with examples.

The initial conflicting results put the onus on the authors to demonstrate where the divergence occurred. The authors took a highly comprehensive approach, which is a clear strength of the paper. All of the data is fully and transparently presented. Standardizations and differences between experiments run or analyzed in each lab are well laid out. Figure 1 and Table 2 provide a great summary of the techniques and their limitations. These are also well thought out and discussed within each section of results.

The only thing missing is a likely explanation for the differences seen. Although the authors made several attempts to provide such explanation, the question remains - how did two groups who published independent studies using different strategies demonstrate flexor and extensor separation in the dorsal horn, when this study, using several strategies in multiple labs, show that the premotor neurons are in complete overlap? Additional small differences in methodologies could be identified which are not discussed and may provide potential explanations, but only for discrepancies in results of single techniques, not for all of the strategies used. The lack of reason for the discrepancy with prior studies despite the extensive efforts is unsatisfying, but, most importantly, the experiments were rigorously performed and the data support the conclusions presented.

We thank the reviewer for the positive comments and we share the opinion that the discrepancy is unsatisfying. While we propose possible explanations, despite the extensive efforts, we could not provide a definite answer, but we hope that making our work public and all the data available, will trigger even more efforts from the rest of the community.

Author Response

Reviewer #1 (Public Review):

This is a very interesting paper. In this manuscript Hendi et al. examined how two independent mechanisms, Wnt signalling and gap junction control two critical aspects of neuronal tiling. Here they have quite elegantly used two neighboring GABAergic motor neurons to show while one specific C. elegans Wnt-homolog, EGL-20, regulates the axonal tiling; innexin UNC-9-mediated gap junction at a very specific position on these axons regulate the chemical synapse tiling on these axons. They also performed multiple experiments to show that the UNC-9 gap junctions controls chemical synapse tiling independent of their channel activity.

Overall, the paper is interesting and would be of general interest for many neuroscience researchers, specifically to those who are studying neuronal tiling and the role of gap junctions. However, there are some concerns with this study.

Major concerns:

1) Authors here only looked at the tiling of axons and presynaptic clusters in DD5/DD6 axons. However, these neurites get transformed in L1 from dendrite to axon and subsequently the nature of the synaptic termini also changes from postsynaptic to presynaptic. To say that egl-20/UNC-9 specifically control axonal tiling and GABAergic presynaptic tiling the authors must check the dendritic tiling and tiling of postsynaptic termini. Specifically, a) does UNC-9 channels also affect the postsynaptic patterning in L1? b) what is the time of unc-9 puncta formation? Is it present in the L1 stage or appears at L2 stage only after the fate switch from dendrite to axon? c) does egl-20 also control dendritic tiling in L1?

We thank the reviewer for their insightful comments. As described in our original manuscript, we could not check the dendritic tiling between DD5 and DD6 at L4 stage due to the inconsistent labeling of DD6 dendrite with our fluorescent marker. As an alternative method, we measured the length of the (ventral) posterior dendrite of DD5 and showed that it is significantly longer in the egl-20(n585) mutant than in wild type at L4 stage. We also measured the length of postsynaptic domains in the DD5 posterior dendrite and showed that it was also longer in the egl-20(n585) mutant than wild type. Furthermore, we show that the UNC-9 localization at the tip of DD6 dendrite is unaffected in the egl-20(n585) mutant, despite the extension of postsynaptic domains. From these observations, we suggested that postsynaptic spines are distributed throughout the dendrite of DD5 in the egl-20(n585) mutant, and it is not regulated by unc-9.

In the revised manuscript, we included images of wild type and egl-20(n585) animal in which ACR-12::GFP is co-labeled with mCherry::CAAX. In these strains, the expression of mCherry::CAAX and ACR-12::GFP is not detectable in DD6 in most animals. Using these strains, we confirmed that the DD5 postsynaptic sites are present throughout the dendrite of DD5 in both wild type and egl-20(n585) mutant backgrounds (Figure 1- figure supplement 1).

a) Unfortunately we were not able to quantify postsynaptic patterning at L1 due to the low expression of ACR-12::GFP and mCherry::CAAX at L1 stage.

b) UNC-9::7×GFP puncta are present at the tiling border of DD neurons on both ventral and dorsal sides throughout the development. In the original manuscript, we only showed the UNC-9 localization at the dorsal side. We believe our limited description of UNC-9 in the dendrites has caused confusion regarding the phenotypes of DD5 posterior dendrite and postsynaptic sites. In the revised manuscript, we have updated the images of UNC-9::7×GFP to show that the puncta are present in both axons and dendrites (Figures 2F-H).

In the revised manuscript we also show that UNC-9 puncta are present at DD tiling border in L1 animals. We have included images of UNC-9::7×GFP at L1 at the axonal and dendritic tiling borders of DD5 and DD6 in both wild type and egl-20(n585) animals in Figure 2- figure supplement 5.

c) As described above, we could not quantify dendritic tiling at L1 due to the low expression of our fluorescent makers at the L1 stage.

2) Authors have shown that the previously known regulators for gap junction formation, NLR-1 and ZOO-1, do not regulate UNC-9 gap junction puncta on DD5/DD6 axons. Since they are cell adhesion molecule and tight junction component, respectively, presynaptic tiling should be checked in these mutants as well. Also, it is not clear whether these proteins are expressed in DD5/DD6 neurons at all. Since, NLR-1 has previously been shown to regulate unc-9 puncta in nerve ring, expression of these genes in DD5/DD6-neurons should be checked before making these conclusions.

In the revised manuscript, we have included the presynaptic tiling quantification in zoo-1(tm4133); egl-20(n585) and nlr-1(miz202) egl-20(n585) mutants which showed no significant presynaptic tiling defects (Figure 2- figure supplement 1). We also cited a paper (Taylor et al., 2021) that described the expression of zoo-1 and nlr-1 in the DD neurons.

3) Authors assumed that the relevant gap junction to be an UNC-9 homotypic homomeric channel, but DD neurons also express several other innexins (inx-1, inx-2, inx-10, inx-14 and unc-7). This raises the possibility that unc-9 channel could be heteromeric in nature. Effect of some other expressed innexins on synaptic tiling apart from unc-7 should also be tested.

We thank the reviewer for their comment. As per their advice, we tested four additional innexins (inx-1, inx-2, inx-10, and inx-14) which have been reported to be expressed in DD neurons and examined their potential role in presynaptic tiling in egl-20(n585) mutant background. We found that none of them showed significant presynaptic tiling defect. In the revised manuscript, we have included this data in Figure 2E.

4) Effect of unc-9(Del18) and unc-1 double mutant should be tested.

We knocked out unc-1 using CRISPR/Cas9 genome editing in the egl-20(n585); unc-9(syb3236 [unc-9(ΔN18)]) mutant background and observed no significant presynaptic tiling defect compared with egl-20(n585); unc-9(syb3236 [unc-9(ΔN18)]), which further strengthen our model that the gap junction channel activity of UNC-9 is dispensable for its function in presynaptic tiling. We have included this data in Figure 5D.

5) Authors have acknowledged the need to study the role of UNC-9 gap junction channels in maintaining the presynaptic pattering. This reviewer appreciates that idea and suggests the authors check whether late expression of UNC-9 is sufficient to rescue the presynaptic pattering defect observed in egl-20; unc-9 double mutant animals.

We thank the reviewer for their comment. We conducted late rescue experiment using a heat shock promoter to express unc-9 at L2 stage after the presynaptic tiling competes. We did not observe significant rescue in presynaptic tiling defect in two independent transgenic lines of Phsp::unc-9. While we understand that this does not deny the function of unc-9 for the maintenance of presynaptic tiling, this result is consistent with the idea that unc-9 is required for the establishment of presynaptic tiling. We have included this data in Figure 2- figure supplement 4.

Reviewer #3 (Public Review):

This interesting paper from Hendi et al. describes a novel mechanism governing synaptic tiling that depends on expression of a gap junction protein at the border between adjacent presynaptic domains of neighboring neurons. The authors define the role of innexin UNC-9 in establishing the spatial arrangement of synapses in adjacent C. elegans GABA motor neurons. They show that axonal tiling is controlled by Wnt signaling. However, synaptic tiling is preserved when axonal tiling is disrupted in egl-20/Wnt mutants. Synaptic and axonal tiling are both disrupted in egl-20; unc-9 double mutants, suggesting these two processes are controlled through distinct molecular mechanisms. The authors find that UNC-9 is localized to the border between axons of adjacent GABA neurons and provide evidence that the function of UNC-9 in tiling does not require its channel function. The experiments are made possible by the development of a new system for labeling adjacent GABA motor neurons that will also be of general use to the field. The studies rule out requirements for either gap junction activity or several other genes previously implicated in gap junction function/localization, but fall short of clearly defining mechanism. Instead, the study provides additional support for channel-independent structural roles of gap junctions in the nervous system.

The study's conclusions are generally well-supported by the data but more clarification is required in some areas:

1) Overlaps between DD5 and DD6 dendrites are not evaluated directly. The authors show the extent of labeling in the DD5 dendrite. This should be clarified.

We thank the reviewer for their comment. As described above, we could not directly quantify dendritic tiling defect between DD5 and DD6 neurons due to the inconsistent expression of mCherry in the dendrite of DD6. Alternatively, we measured the length of DD5 posterior dendrite in wild type and the egl-20(n585) mutant, and found a significant increase in the DD5 posterior dendrite length in the egl-20(n585) mutants. In the revised manuscript, we have edited the text to more clearly explain the defect of DD5 posterior dendrite.

2) The authors suggest UNC-9 establishes axonal tiling as early as L2 stage, immediately following DD remodeling. However, no data is shown for UNC-9 localization at this developmental stage. It would also be interesting to know whether UNC-9 performs a similar role prior to remodeling, or if UNC-9 itself undergoes redistribution during the remodeling process.

We thank the reviewer for their comment. As described above, we acknowledge our initial description of UNC-9 localization in the DD neurons was not sufficient. UNC-9 is present at both the axonal and dendritic tiling borders between DD5 and DD6 neurons throughout the larval development.

In the revised manuscript, we included UNC-9 localization at the axonal and dendritic tiling borders between DD5 and DD6 in both wild type and egl-20(n585) animals at the L1 stage (Figure 2- supplement figure 5). However, we could not determine whether egl-20(n585); unc-9(e101) mutant exhibits presynaptic patterning defect in the ventral axons prior to remodeling at the L1 stage due to the low expression of our axonal and presynaptic markers at L1 stage.

3) Based on the representative image, UNC-9 abundance appears reduced in unc-104. The authors should quantify.

We thank the reviewer for their comment. In the revised manuscript, we quantified the signal intensity of UNC-9::7×GFP at the DD5-DD6 axonal tiling border in wild type, egl-20(n585), unc-104(e1265), zoo-1(tm4133) and nlr-1(gk366849). We found that the fluorescent intensity of UNC-9::7×GFP was indeed slightly lower in egl-20(n585) and unc-104(e1265) mutants compared with wild type animals. This result implies that egl-20 and unc-104 have a minor role in UNC-9 localization. Nevertheless, the UNC-9 puncta are always present in all genotypes we examined. The quantification is included in Figure 2- figure supplement 6, and we suggest that the weak presynaptic tiling defect in the egl-20 single mutant could be explained by this reduction of UNC-9 localization (lines 284-285).

4) The authors show the distribution of muscle NLG-1 mirrors that of RAB-3. While this suggests the altered distribution of RAB-3 reports on synaptic rearrangement, this conclusion would be strengthened by analysis of an active zone marker.

We agree with the reviewer that examining the co-localization of RAB-3 with an active zone protein would strengthen our conclusion. As such, we expressed BFP::RAB-3 under the DD specific promoter, flp-13, in a transgenic marker strain (wyIs292) that expresses the active zone protein, UNC-10::tdTomato under the GABAergic promoter, unc-25, and NLG-1::YFP expressed under the body wall muscle promoter, unc-129dm (Maro et al., 2015). Using this strain, we show that RAB-3 co-localized with UNC-10 and apposed to the postsynaptic NLG-1 in both wild type and the egl-20(n585); unc-9(e101) mutant. The representative images are included in Figure 2- figure supplement 2.

Author Response

Reviewer #1 (Public Review):

The stated goal of this research was to look for interactions between metabolism, (manipulated by glucose starvation) and the circadian clock. This is a hot topic currently, as bi-directional links between metabolism and rhythmicity are found in several organisms and this connection has important implications for human health. The authors work with the model organism Neurospora crassa, a filamentous fungus that has many advantages for this type of research.

The authors' first approach was to assay the effects of glucose starvation on the levels of the RNA and protein products of the key clock genes frq, wc-1, and wc-2. The WC-1 and WC-2 proteins form a complex, WCC, that activates frq transcription. The surprising finding was that WC-1 and WC-2 protein levels and WCC transcriptional activity were drastically reduced but frq RNA and protein levels remained the same. Under conditions where rhythmicity is expressed, the rhythms of frq RNA, FRQ protein, and expression of clock-driven "output" genes were also unaffected by starvation. The standard model for the molecular clock is a transcription/translation feedback loop dependent on the levels and activity of these clock gene products, so this disconnect between the starvation-induced changes in the stoichiometry of the loop components and the lack of effects of starvation on rhythmicity calls into question our understanding of the molecular mechanism of the clock. This is yet another example of the inadequacy of the TTFL model to explain rhythmicity. For me, the most significant sentence in the paper was this: "...an unknown mechanism must recalibrate the central clockwork to keep frq transcript levels and oscillation glucose-compensated despite the decline in WCC levels."

The author's second approach was to try to identify mechanisms for the response to starvation by focussing on frq and its regulators, using mutations in the frq gene and strains with alterations in the activity of kinases and phosphatases known to modify FRQ protein. The finding that all of these manipulations have some effect on the starvation-induced changes in WC protein level is taken by the authors to indicate a role for FRQ itself in the response to starvation. This conclusion is subject to the caveat that manipulations of the activity of multifunctional kinases and phosphatases will certainly have pleiotropic effects on many cellular processes beyond FRQ protein activity.

Because of the sometimes-speculative nature of our conclusions and based on the suggestion of the editor, we restructured the Discussion and discuss now the mechanism addressed by the Reviewer in the subsection "Ideas and Speculation". We added a sentence to the section about the possible pleiotropic effects of the tested signaling pathways: "Starvation triggers characteristic changes in the activity of signaling routes that affect basic components of the circadian clock. Although the multifunctional pathways might act via pleiotropic mechanisms as well, based on their earlier characterized role in the control of the Neurospora clock, their action can be inserted into a model describing the glucose-dependent reorganization of the oscillator."

The third section of the paper is a major transcriptomic study of the effects of starvation on global gene expression. Two strains are compared under two conditions: wc wild-type and the wc-1 knockout strain, under fed and starved conditions. The hypothesis is that WCC has a role in the starvation response. The results of starvation on the wild-type are unsurprising and predictable: the expression of many genes involved in metabolic processes is affected. There are no new insights that come from these results and no new testable hypotheses are generated by the data.

We agree with the reviewer that it is not surprising that glucose depletion strongly affects genes involved in metabolic processes and monosaccharide transport. These data obtained in wt served rather as a control for our experimental conditions. As a new aspect, our analysis focused on the differences between wt and wc-1 in the transcriptomic response to altered glucose availability.

The authors refer to the wc-1 mutant strain as "clockless" and discuss its effects on the transcriptome only in terms of WC-1's function in the clock mechanism. However, WCC is known to be a major transcriptional regulator, controlling a number of genes beyond the TTFL. As acknowledged earlier in the paper, WC-1 is also the major light receptor in Neurospora. The transcriptomics experiments were carried out in a light/dark cycle, with cultures harvested at the end of the light period, when "an adapted state for light-dependent genes can be expected" according to the authors. However, wc-1 mutants are essentially blind, and so those samples are equivalent to being harvested in the dark. The multifunctional nature of WCC complicates the interpretation of the transcriptomics data. The differences in the transcriptome between wild-type and wc-1 may not be due to loss of clock function, but rather the loss of a major multifunctional transcription factor, or the difference between light and "dark".

The reviewer is right, when we discussed the difference between wt and wc-1 in the transcriptional response to glucose, we did not emphasize the possible contribution of the photoreceptor function of the WCC. We added the following sentence to the revised version of the discussion: "Further investigations could differentiate between the clock and photoreceptor functions of the WCC in the glucose-dependent control of the transcriptome." Furthermore, we more specifically indicate that in wc-1 the lack of the WCC (and not the lack of a functional clock) results in the altered transcriptomic response to starvation when compared to wt (P15 L14-17).

In the final set of experiments, the authors tested the hypothesis that the changes in the transcriptome between wild type and wc-1 might make wc-1 less competent to recover growth after starvation. They also test the recovery of frq9, a "clockless" mutant. The very surprising result is that the growth rates of these two mutants are slower than the wild type after transfer from starvation media to high glucose. This is surprising because there will be several generations of nuclear division and doublings of mass within a few hours and the transcriptome should have recovered fully fairly rapidly. A mechanism for this apparent "after-effect" is suggested with evidence concerning differences in expression of a glucose transporter, but it is not clear why this expression should not change rapidly with re-feeding on high glucose. As with previous experiments, the cultures were grown in light/dark cycles, which results in different conditions for the mutants, both of which have very low or absent WC-1 and are therefore blind to light. The potential effects of light have been disregarded.

The reviewer is right that several generations of nuclear divisions occur within a few hours and lead to a number of doublings of the biomass. However, when the first phase of regeneration is delayed in one or more strains compared to the control, until the stationary phase a substantial difference in the biomass can be expected.

To the expression change of the glucose transporter: In order to emphasize the different tendency of how glt-1 levels respond to glucose in the different strains, in the previous version of the manuscript we normalized the expression levels to the beginning of recovery (time point of glucose addition). Thus, expression differences between the strains were not shown. To give a more comprehensive picture, in the revised version of the manuscript expression levels without normalization are depicted (Fig 5F). The mutants did not adapt efficiently to changes in the glucose levels, i.e. expression of the transporter was relatively high in both wc-1 and frq10 during starvation and did not further increase upon glucose addition. On the other hand, 24 hours after glucose resupply, glt-1 levels were similar in all strains which might contribute to the similar growth rates observed under steady-state conditions in the standard medium.

To the photoreceptor-independent function of the WCC during growth recovery: In the revised version of the manuscript we present additional data suggesting the importance of the photoreceptor-independent function of the WCC for efficient recovery from starvation. Fig. 5C and Fig. 5D show now that upon resupply of glucose, wt grows faster than the clock-deficient strains Δwc-1 and frq10 in both LD cycles and constant darkness, indicating that the role of the WCC in growth regeneration is at least partially independent of its photoreceptor function. To the function of the WCC in frq10: frq10 can not be considered blind. Although both Δwc-1 and frq10 lack a functional clock and WC levels are reduced in frq10, these strains show significant differences in WCC activity. While Δwc-1 is considered blind, in frq10 lack of the negative feedback results in high activity of the WCC in both DD and LL and expression levels of all examined, light-sensitive or light-dependent genes were found comparable in wt and in frq-less mutants (Schafmeier et al., 2005; Hunt et al., 2007; own unpublished data).

The title of the paper refers to a "flexible circadian clock" but this concept of flexibility is not developed in the paper. I would substitute "the White Collar Complex" for this phrase: "Adaptation to starvation requires a functional White Collar Complex in Neurospora crassa" would be more accurate. Some experiments are also conducted using an frq null "clockless" strain, but because WC expression is very low in frq null mutants, any effects of frq null could also be attributed to WC depletion.

As detailed above, low level of the WCC in the frq-less mutant does not mean low transcriptional activity and accordingly, the two clock mutants, wc-1 and frq10 show important functional differences. We used the word "flexible" to indicate that the molecular clock is able to operate under critical nutrient conditions and with a significantly changed stoichiometry of its key components. Results of our new experiments performed in DD (mentioned above) indicate that growth regeneration is rather independent of the photoreceptor function of the WCC. Nevertheless, we accepted the criticism of the reviewer and changed the title to "Adaptation to glucose starvation is associated with molecular reorganization of the circadian clock in Neurospora crassa".

The major conclusion I took away from this paper is the multifunctional nature of the WCC as a transcription factor complex. It has been known for a long time that WCC controls the expression of many genes beyond the frq gene at the core of the circadian transcription/translation feedback loop. WC-1 is also the major blue light photoreceptor in Neurospora, controlling the expression of light-regulated genes, and this fact is barely touched on in the paper. These new data now extend the role of WCC in the regulation of metabolic networks as well.

Reviewer #2 (Public Review):

The authors have performed an interesting study addressing a topical question in considering how circadian oscillators remain accurate in changing environmental conditions and these circadian oscillators contribute to responses to environmental changes. The authors have performed their studies in Neurospora crassa. The authors have made a very interesting finding that starvation causes a profound decrease in white collar 1 WC-1 abundance, yet the circadian system continues to run despite this decrease in the abundance of a core oscillator component. The study of chronic glucose starvation in a Δwc-1 mutant is interesting and provides the opportunity to investigate the role of the WHITE COLLAR COMPLEX (WCC) and the clock system in adaption to starvation.

Strengths:

The authors have used a range of techniques to measure clock behaviour, including qPCR, phosphorylation, protein abundance, and subcellular localisation studies.

An frq9 mutant was used to test the effects of FRQ on WC1 abundance since WC1 decreased during starvation. This is elegant, though it is not quite clear the logic of this experiment because FRQ did not change abundance during starvation, so why did the author think this experiment was needed?

We regret that the examination of frq9 was not clearly justified in the previous version of the manuscript. It is true that FRQ levels did not change during starvation, only phosphorylation of the protein was affected, i.e. FRQ became more phosphorylated (displayed by an electrophoretic mobility shift on the Western blot (Garceau N, Liu Y, Loros J J, Dunlap J C. Cell. 1997;89:469–476.)) under low glucose conditions. We tested the starvation response in the FRQ-less strain because WCC level changed significantly in wt upon glucose depletion and expression of WC proteins is known to be controlled by FRQ. In the revised version of the manuscript we tried to introduce and explain the experiments performed with frq9 more thoroughly (P7 L22-P8 L14; P16 L21 – P17 L6).

An interesting experiment was performed to test whether CK1a-dependent phosphorylation and inactivation of the WCC are involved in the starvation response. An FRQΔFCD1-2 mutant is used in which FRQ cannot interact with CK1a and therefore CK1a cannot phosphorylate and inactivate WC. This experiment suggested that CK1a is not involved in the response to starvation, again leading to the conclusion that FRQ is not involved in the starvation regulation of WC.

The referee is right, effect of FRQ-bound CK-1a seems to be minor on the adaptation of the molecular clock to starvation, and this is also our conclusion in the manuscript. The major message of this experiment was that FRQ became phosphorylated in response to starvation without stably interacting with CK1a, probably via another mechanism. We agree with the notion that the behavior of WCC levels upon starvation was similar to that in the FRQ-less mutant.

PKA is shown to be involved in the starvation-induced reduction of WC because the starvation-induced reduction in abundances of WC-1 was absent in the mcb strain in which the regulatory subunit of PKA is defective and hence, PKA is constitutively active.

The authors have found an interesting potential link between glucose levels and WCC phosphorylation, they demonstrated that starvation reduces PP2A activity and that in a regulatory mutant of PP2A, which has reduced PP2A activity, there is little effect of starvation on WCC levels, suggesting the hypothesis that glucose-dependent PP2A dephosphorylation stabilises WCC.

Analysis of starvation-regulated transcriptome in Δwc-1 and wild type found strong evidence that the transcriptomic response to starvation is in part dependent on WCC. Much of the misregulated transcriptome appears to be associated with metabolism.

In a series of growth studies in wild-type frq and wc-1 mutants the authors provide strong evidence that FRQ and WC are involved in growth and survival following starvation, and recovery from starvation.

Weaknesses:

The authors describe Neurospora crassa as a model for circadian biology and apparently make the assumption that the findings are indicative of the behaviour of clock systems in other kingdoms. This is not the case. Neurospora crassa is a wonderful model for studying fungal clocks and is a great tool for studying basic circadian dynamics, but the interesting findings here are of a detailed molecular nature and therefore are applicable for fungal clocks, but not other kingdoms.

We agree that we still do not know whether the described mechanism is specific for only fungal clocks. However, besides the basic feedback loop, overlapping mechanisms (controlled by e.g. casein kinases, glycogen synthase kinase, PKA, PP2A) are involved in the regulation of circadian timekeeping in different eukaryotic systems (reviewed in Reischl and Kramer, 2011, FEBS Lett; Brenna and Albrecht, 2020, Front Physiol). Our results suggest that some of these common factors (PKA, GSK, PP2A) are involved in the reorganization of the Neurospora clock in response to changes in glucose availability. Therefore, it is possible that analogous changes occur in the time keeping mechanisms of other eukaryotic systems when they face serious environmental challenges.

We included a short section into the Discussion which gives a short overview about known interactions between glucose availability and circadian timekeeping at different levels of the phylogenetic hierarchy (P15 L18 – P16 L7).

The authors assume that the reader is intimate with the intricacies of Neurospora crassa circadian studies and the significance of differences between LL and DD investigations. More background on the logic of the experiments would be helpful for readers from other fields.

Thank you for the comment. In the revised version of the manuscript we tried to introduce the molecular clock of Neurospora more thoroughly and completed the description of the experimental conditions with detailed explanations.

The data in Figure 2 are essential for the interpretation of the findings, demonstrating the presence of free-running rhythms. However, the data are entirely qualitative, making it hard to fully assess the authors' interpretations, a more quantitative assessment of the data would improve clarity.

We quantified the Western blot signals and show the results in Fig 1E in the new version of the manuscript (according to the reviewer's suggestion Fig 2 of the old version is now part of Fig 1). Our data indicate that oscillation of FRQ levels is similar under both nutrient conditions.

The conclusion that FRQ contributes to the regulation of WC1 abundance in response to starvation does not seem to be supported by the data because FRQ RNA does not change upon starvation. Furthermore, the authors conclude that the starvation-induced decrease in WC-1 and WC-2 protein levels are due to FRQ because a lack of reduction in an frq9 mutant is open to misinterpretation because this mutant makes WC levels low and therefore starvation might not lower already low levels of WC. Indeed WC-1 is lower in the frq9 mutant under any condition than in the WT under starvation and WC-2 does decrease in abundance in the frq9 mutant in starvation. The data strongly suggest to this reader that FRQ does not participate in the regulation of WC abundance in response to starvation.

After rereading the criticized section, we admit that the text was not well structured and we carried out several modifications. We intended to emphasize that upon drastic changes of the glucose availability frq RNA levels remained compensated in wt, but this compensation was affected when functional FRQ was not present. We agree with the reviewer's opinion that the low expression of the WCC in frq9 makes it difficult to compare the glucose-dependence of WCC expression in frq9 and wt. We modified the conclusion by adding this information and now mainly focus on the strain-dependent difference in the changes of frq RNA expression. (P7 L22-P8 L14)

The discussion accurately summarises the results and provides an interpretation but lacking is a comparison to other circadian systems in other kingdoms. How do the data compare with the effects of glucose and other sugars on the mammalian, plant, and insect clocks?

We included a short section into the Discussion which gives a short overview about known interactions between glucose availability and circadian timekeeping in different organisms (P15 L18-P16 L7).

How changes in WCC might result in changes in transcription is not explained. This might be very obvious to the authors but to the reader, it is not. Are the transcriptional outputs direct targets of WCC? Has WCC CHIPseq been performed by the authors or others, are the regulated transcripts directly bound by WCC? What are the enriched promoter sequences in the regulated genes, is it possible to identify the network by which these changes in transcription occur?

We now show the list of genes (Figure 4 – Figure supplement 2) that changed in a strain-specific manner in response to glucose starvation and, based on Chip-Seq results, were earlier described as direct targets of the WCC (Smith et al., 2010; Hurley et al., 2014). Based on the literature data showing that the WCC affects the expression of several other transcription factors and controls basic cellular functions which might affect the expression of further genes, it was not surprising that only 90 out of the 1377 genes were reported to be direct targets of the WCC.

Whilst the authors claim it is the circadian clock that is involved in the starvation response, in my view a more precise interpretation of the data is that WCC is involved in the response. Since WCC is a photoreceptor with dual function in the clock, is it yet possible to conclude that the effects discovered are due to the clock role of WCC? Or do the data support the role of light signalling in regulating the starvation response through WCC?

We thank you for the comment. In the revised version of the manuscript we more specifically indicate that in wc-1 the lack of the WCC (and not the lack of a functional clock) results in the altered transcriptomic response to starvation compared to wt. In addition, in the revised version we present a new experiment (Fig. 5D.) which shows that upon resupply of glucose wt grows faster also in constant darkness than the clock-deficient strains wc-1 and frq10 do. This indicates that the role of the WCC in growth regeneration is largely independent of its photoreceptor function.

The authors do not apparently reconcile that the effect of starvation is to hugely decreases WCC levels, but they find the transcriptional and growth response to starvation requires WCC?

We agree with the reviewer that the problem of how low levels of WCC could sufficiently support the transcription of frq and different output genes under starvation conditions was not discussed properly. Our results suggest a model in which the maintained level of nuclear WCC and the weakened inhibition by both FRQ (the hyperphosphorylated form is less active in the negative feedback) and PKA (its activity lowered upon glucose depletion) together might ensure that transcriptional activity of the WCC is preserved upon glucose withdrawal in both DD and LL despite the decrease of the overall level of the complex. In the revised version these aspects are discussed more thoroughly (P16-18).

This study contributes to the increased focus of the circadian community on the regulation of outputs by circadian oscillators. The manuscript will be of interest to many in the field. There needs to be less assumption of knowledge about the N. Crassa circadian system, and better discussion in a broader context of clocks in other kingdoms.

We added a new section to the Discussion with data concerning interrelationships between glucose availability and the circadian clock in other organisms.

Author Response

Reviewer #1 (Public Review):

Drosophila ovarian follicle cells have been utilized as a model system to study organogenesis and tumorigenesis of epithelia. Studies have found that lack of proper cell polarity causes invasive delamination of cells and formation of multilayered epithelia, reminiscent of Epithelial-Mesenchymal Transition (EMT). Using this system, the authors analyzed the single-cell transcriptome of follicle cells and show that distinct cell populations emerge shortly after induction of polarity loss. Authors identified dynamic activation of Keap1-Nrf2 pathway Finally, subpopulation classification and analysis of regulon activity identified that Keap1-Nrf2 pathway is responsible for epithelial multilayering caused by polarity loss.

Strengths:

The authors characterized the single-cell transcriptome of follicle cell subpopulations after induction of polarity loss. Using temperature-inducible driver, they can induce the polarity loss in a short period of time, which enables detection of epithelial populations in various transition stages. Detected cell-heterogeneity could be caused intrinsically or by environmental cues within in vivo tissue. Therefore, it is likely well recapitulating tumorigenesis in vivo.

Weaknesses:

1) Authors should show cells corresponding to identified key cell clusters within the tissue by immunostaining, GFP-trap, or RNA FISH.

We thank the reviewer for their comment. However, for this particular case, we would like to underscore the observation that the clusters derived from our integrated analysis do not exhibit mutually exclusive gene expression. This is unlike other studies where different clusters exhibit unique markers. The different clusters in this study represent distinguishable cell states and not distinct cell types. Even though the Lgl-KD follicle cells transcriptomically deviate from their corresponding cells of origin to form their own clusters, they continue to express several markers that show gene-expression overlap with normal follicle cells. This overlap exacerbates the problem of identifying distinct cells using differentially-enriched markers.

However, we have shown the antibody staining against Drpr to identify cluster 8 follicle cells that associate with Dcp1+ dying germline cells. We have used GstD-lacZ reporter (cluster 7 marker, specifically cluster 7_3) to show pathway activity within the multilayer. Besides GstD-lacZ, we also show F-Actin enrichment in cluster 7 (specifically 7_3) cells, that is significantly enriched in invasive cells. Additionally, we now have added images depicting the cell/stage specific expression pattern of JNK pathway components pJNK and puc, as well as that of Thor (4E-BP) which is expressed at high levels in cluster 8 and medium levels in cluster 7, and Xbp1-GFP (UPR stress sensor) that marks late stages of Lgl-KD cells.

2) Images are low magnification and difficult to see individual cells.

We have replaced several such images in the revised manuscript. Specifically, the revised manuscript has entirely new (or improved versions of) image panels in figure 5. In figure 1A, the focus is the entire ovariole and therefore, we have only highlighted the enrichment of Hnt and pH3 antibody staining separately for a subsetted region of interest (ROI). The ROI panels are included within the larger image itself. For figure 6, we have converted the LUTs of panels showing distinct channels for RFP and Shg/Arm antibody stainings to grayscale.

3) Manuscript is written weighted toward the technical aspect and more biology behind this study has to be discussed.

We have added new paragraphs to discuss the evidence supporting the loss of polarity, specifically that of Lgl, in human cancers. Additionally, we have also discussed how our results regarding Keap1 relates to what is already known about it and the implications of our results in context to cancer progression and metastasis.

Reviewer #2 (Public Review):

Chatterjee et al. perform extensive image and single-cell RNA sequencing (scRNA-seq) analysis of Drosophila ovaries with and without knockdown of a gene, Lethal giant larvae (Lgl), which is known to establish apical-basal polarity as well as controlling proliferation of epithelial tissues. The goal of the study is to characterize the effect of apicobasal-polarity loss in epithelial cells via Lgl knockdown on Drosophila ovaries at the phenotypic, cellular, single-cell gene expression and regulatory level. By focusing on single-cell gene expression clusters that are unique to Lgl-KD compared to those from flies without the knockdown, they were able to identify a highly transient cluster (cluster 7) which consists of tumorigenic cells. Differential markers within a sub-cluster (cluster 7_3) of this cluster followed by validation using a GstD-lac-Z enhancer-trap reporter assay lead to their conclusion that cluster 7 represents the cells of multilayering phenotype (i.e., the major Lgl-KD phenotype observed from image analysis) where activation of Keap1-Nrf2 signaling was observed. The KEAP1-NRF2 pathway is associated with protecting cells from oxidative stress. KEAP1 forms part of an E3 ubiquitin ligase, which controls NRF2, a transcription factor, by targeting it for ubiquitin-mediated proteasomal degradation. Surprisingly, inducing loss of function of both Keap1 and separately NRF2 (cnc in Drosophila) in Lgl-KD cells resulted in the same phenotype/rescue, loss of the multilayering phenotype. Over expression of Keap1 in Lgl-KD induced increased multilayer volume compared to Lgl-KD alone further supporting the role of Keap1 in cellular invasion and possibly early stages of tumorigenesis when epithelial cells start losing their polarity.

The strengths of this paper are:

The mutually reinforcing advanced imaging, scRNA-seq and genetic manipulation (knockdown and over expression) experiments/analyses that largely support the major conclusions of the manuscript which are summarized above as well as more minor observations that the authors make.

The systems biology flow of the study from broad to a specific gene/pathway implicated in the phenotype. The authors start with a clear phenotypic characterization of Lgl-KD and genome-wide scRNA-seq analysis. This leads to regulatory factor enrichment and further identification of a cluster (cluster 7) and then to a sub-cluster (cluster 7_3). This is followed by the identification of the KEAP1-NRF2 pathway and demonstration that KEAP1 knockdown and overexpression in Lgl-KD rescues and aggravates the cell multilayering phenotype, respectively.

The multilayering phenotype, genes and regulatory factors associated with loss of polarity are known to play an important role in the epithelial to mesenchymal transition (EMT). For example, this includes the enrichment of AP-1 family members, which are known to regulate EMT, in the regulon analyses as well as identification of KEAP1-NRF2.

The weaknesses of the paper are:

The framing/motivation of the study could be improved especially for those who study EMT/metastasis in humans. Given that loss of polarity is one of many events associated with tumorigenesis and metastatic progression, the claims made that studying Lgl-KD in Drosophila ovaries directly leads to insights into tumor cell invasiveness, early stages of tumorigenesis and EMT may leave some readers doubtful if they are not familiar with Lgl. Reviewing major findings that show that Lgl is a tumor suppressor as is its human homologue Hugl-1 as well as making a stronger case that studying Lgl-KD in Drosophila is relevant for tumorigenesis and EMT would be helpful.

We thank the reviewer for these suggestions. Accordingly, we have added new paragraphs to the Discussion section, where how the Lgl-KD mediated polarity loss links to mammalian tumorigenesis, as well as the implications of our results, have been discussed.

Given that Keap1 antagonizes NRF2, the apparent contradictory result that inducing loss of function of both Keap1 and separately NRF2 (cnc in Drosophila) in Lgl-KD cells resulted in the same phenotype/rescue (loss of the multilayering phenotype) is not fully addressed. Keap1 over expression revealed it aggravates multilayering. NRF2 over expression experiments were not performed. In addition, it was shown that over expression and knockdown of Keap1 did not affect NRF2 gene expression (Figure 5C); however, Keap1 regulates Nrf2 at the protein level directly via ubiquitin-mediated proteasomal degradation. Nrf2 protein levels in flies with and without Lgl-KD with various manipulations of Keap1 including control, KD and OE were not measured.

As the Keap1-Nrf2 pathway is widely studied in context of oxidative-stress response signaling, Keap1 is widely accepted as a negative regulator of Nrf2-driven transcription. However, Nrf2 has been found to positively drive the expression of Keap1 (Sykiotis and Bohmann, 2008), and that manipulating Keap1 did not change Nrf2 expression (Fig.5C). In response to this comment however, we performed additional experiments driving the ectopic expression of Nrf2 (CncC-OE) in Lgl-KD cells, which increased the invasiveness of Lgl-KD cells, similar to that by Keap1-OE. Since the UAS-CncC line has been shown to upregulate Keap1 expression (Sykiotis and Bohmann, 2008), we concluded that this increase in invasiveness is indirectly due to the increase in Keap1 expression itself.

Given that the antagonizing relationship of Keap1 and Nrf2 is only relevant to oxidative-stress response pathway, the genetic epistasis experiments in this study render that relationship irrelevant in context to the observed phenotype, as KD or OE of both components result in comparable phenotypes. Previous studies showing that Keap1 plays a role in cytoskeletal regulation (which is in agreement with our observation) also add weight to the argument that the observed phenotype is likely an indirect consequence of Keap1-Nrf2 signaling activation.

Many of the conclusions in early Results paragraphs are purely technical and not biological. For example, "These observations highlight the limitations of marker validation to identify specific cells of the differential Lgl-KD phenotype" and "SCENIC was able to detect the common as well as distinct transcriptomic states of the cells in unique Lgl-KD clusters, while also highlighting the heterogeneity among them". Some of these technical conclusions could be part of brief discussions in the Methods section.

For those not familiar with various detailed scRNA-seq analysis approaches (e.g., RNA velocity analysis), a brief description of how they should be interpreted biologically in Methods would be helpful. This might help resolve what appear to be contradictory/confusing results. First, the upper branch of cluster 7 (which is a focus of the study) shown in Fig. 3B is in a "late" stage based on Velocity Pseudotime analysis (left panel) and a "root" or an early stage based on Terminal end-points of differential analysis (right panel). The bottom branch of cluster 7 is "late"/"stable end point" based on these two analyses which is now consistent. Second, given these differences between the upper and lower branch of cluster 7, how is cluster 7 biologically the same cluster? Third, the bottom branch of cluster 7 bleeds into cluster 8 and while Ets21C is uniquely expressed in the bottom branch of 7, important markers of the study including Jra, kay (AP-1 family members), grnd, cnc (NRF2), Keap1, and the genes shown in Fig. 6F are all robustly expressed in clusters 7 (bottom branch) and 8. The biologically relevant distinction between the bottom branch of cluster 7 and 8 is not clear. Is cluster 8 important/relevant to the phenotypes observed as well?

We have now added the following paragraph elaborating the logical choices made within the analytical pipeline in our Methods section:

In this study, we have highlighted RNA velocity-derived interpretations that strictly agree with the other analytical perspectives pursued in this study. We applied scVelo to obtain information on the underlying lineage for (1) all unique Lgl-KD clusters, and (2) cluster-7 cells. The cells of the unique Lgl-KD clusters represent a mixed population of mitotic, post-mitotic, border-follicle cells and dying germline-cell associating cells that depict inconsistent transcriptional lineages. In this group of cells, the true developmental end-point of the observed Lgl-KD lineage is cluster 8 (germline-cell death occurs at the end of Lgl-KD follicular development), which likely consists of a mixed population of cells from the lateral epithelia as well as the multilayered epithelia, all responding to germline-cell death. Indeed, certain sections of cluster 7 appear more similar to cluster 8 and others seem comparable to that of cluster 13. These observations underscore our conclusions that the unique Lgl-KD clusters exhibit distinguishable gene expression, representing different cell states. For cluster 7, the state of transcriptomic heterogeneity is what defines its unique state of gene expression and we have assessed this heterogeneity by specifically sub-setting those cells.

For a comprehensive interpretation of the results of the RNA-velocity based analysis, more information can be found in the scVelo tutorial (https://scvelo.readthedocs.io/).

Author Response

Reviewer #1 (Public Review):

Gu et al. examine how activity in the substantia nigra pars reticulata (SNr) contributes to proactive inhibition - the suppression of upcoming actions - by recording SNr activity in rats performing a task requiring them to be prepared to cancel a planned movement. This task was developed in a previous study by the same authors in which they examined how globus pallidus pars externa (GPe) activity depends on proactive inhibition (Gu et al., 2020), which motivated the present focus on SNr. The task is rich and the complementary analyses of how the neural activity relates to the behavior, at the level of individual neurons and populations, are appropriate and illuminating. Overall, this study is well done and has the potential to be a nice contribution to our understanding of how the SNr, and therefore the basal ganglia, mediate behavioral inhibition. Addressing a few questions, however, would improve the paper.

We appreciate both the positive comments and constructive criticism.

• It is not obvious why the presence or absence of proactive inhibition should be determined on a session-by-session basis. It seems quite possible that proactive inhibition is not an all-or-none phenomenon, and also that it might be exhibited to a greater or lesser extent across a session (e.g., due to changes in motivational drive). It would therefore strengthen the paper to better explain the rationale for comparing neural activity across entire sessions "with" and "without" proactive inhibition. Within-session variation in proactive inhibition could be quite advantageous, allowing for within-neuron comparisons. It is even possible that the differences in neural activity that the authors report here using session-by-session analysis are an underestimate of the true effect of proactive inhibition.

It is true that some of our analyses compare whole sessions with- and without- overall behavioral evidence for proactive inhibition. But our primary results come from within-session comparisons of Maybe-Stop to No-Stop trials. For this purpose, the session-wide assessment of proactive inhibition is primarily a screen for which sessions to use for within-session analysis.

It would be desirable if we could use behavior to determine the degree of proactive inhibition on each individual trial, and then compare this to neural measures. Unfortunately, this is not generally feasible in our experiments. Our key evidence for proactive inhibition is the prolongation of reaction times (RTs). However, RTs are famously highly variable over trials. This variability likely reflects a variety of factors, not simply proactive inhibition. For example, in our previous paper (Gu et al. 2020) we showed that dividing trials into slower and faster RTs did not reproduce the same neural differences as comparing Maybe-Stop to No-Stop trials.

An alternative approach to investigating proactive inhibition is to focus on the increased restraint that typically follows over-hasty responses. We found that when rats fail to Stop, on the next trial the degree of SNr variability increases (Fig. 6). We have now expanded this analysis to include additional types of errors. We find that another form of over-hasty action, premature responses before the Go cue, are also followed on the next trial by increased SNr variability (Fig. 6- supp1). By contrast, other error types (wrong choices; failure to respond quickly enough) do not provoke greater variability. These additional within-session analyses provide convergent evidence for increased variability as an adaptive response to failures evoked by excessive haste.

• It is difficult to rule out alternative explanations for the observed differences in SNr activity. While the authors acknowledge this point in the 3rd paragraph of the discussion, they only discuss one potential alternative - reward expectation. Another difference between maybe-stop and no-stop trials is the likelihood that a particular target should be selected, which has also been shown to modulate SNr activity (Basso & Wurtz, 2002). As is often the case with complex behavioral tasks, there may be many other differences between trial types that may contribute to differences in neural activity. It would be helpful for the authors to more fully explain how their results relate to contextual modulation of SNr activity, and why the dependence of SNr activity on proactive inhibition may be a novel finding.

We have expanded the Discussion to include additional alternative explanations.

• A natural question arising from this study, as with most studies of neural recordings during behavior, is the causal nature of the neural activity. It would be non-trivial and beyond the scope of the current study to perform the sort of perturbations that could determine whether population variability causally relates to preparation to suppress actions. But it would be useful to discuss future experiments that might be able to test causality.

We added in Discussion the possibility of using optogenetic manipulations of specific inputs to SNr, to help determine their distinct contributions to SNr firing patterns and proactive behavior.

Reviewer #2 (Public Review):

The authors have recorded the activity of neurons in the rat substancia nigra pars reticulata (SNr) while animals performed a version of a stop-signal task. The goal of this study is to investigate and describe the contribution of SNr in proactive inhibitory control. By examining single-cell responses as well as population activity, the authors show that increasing the probability of stop signal trials induces several changes in SNr responses. First, specific populations of SNr neurons increase their activity during proactive, direction-specific inhibition. At the population level, neurons are biased away from the side of the movement that has to be potentially inhibited. Second, during proactive inhibition, neuron activity is more variable, both at the single-cell and population levels. Finally, the authors show that animals' outcome history influences both firing rates and variability of neuron responses in the current trial. Especially, neural variability is increased following a failure to inhibit a movement.

Strengths

The manuscript provides an interesting and timely insight into the role of the basal ganglia output nucleus in movement initiation control. The paper is often clearly and concisely written (although see one issue related to this below). One of the main strengths of the work is to allow an interesting comparison with recent work by the same team, aimed at investigating the responses of another basal ganglia nucleus (GPe) in the same task, using similar analyses (this comparison is not extensively exploited in the discussion section though). Another potential strength is the use of different analysis scales. The authors investigated single-unit responses as well as population "trajectories" in the neural state space. This is an interesting option that could have been better motivated, given that the two approaches assume quite different brain operations.

Thank you for the interest and careful comments.

Weaknesses

The analyses and results sometimes lack clarity and details. For instance, and unless I missed the information, it is not clearly stated whether "maybe-stop" trial analyses only include Go trials or if (failed) Stop trials are also considered. Moreover, quite complicated figures are often described very briefly in the main text. Methods are also often too succinctly described, and sometimes refer to a previous publication (Gu et al., 2020) that readers did not necessarily read.

We have made a range of changes to make the analyses and their rationale more clear. This includes specifying that Maybe-Stop trials include both Go and Stop trials (and why). We have also added more details in both main text and Methods.

There are some points that the authors might need to discuss more. Especially, a global picture of the role of the different basal ganglia nuclei during movement control would have been appreciated. Also, the authors monitored the activity of the rat basal ganglia output. We would have appreciated more information regarding the impact of this output activity on SNr target areas, as compared to their previous work that focused on GPe for instance. Another example concerns the observation that SNr activity is elevated during active inhibition regardless of the firing rate pattern before movement (increase or decrease). As noted by the authors themselves, this is inconsistent with the classical role assigned to the basal ganglia output nucleus (i.e. a decrease in activity promotes movement). Despite that this observation is of potential interest to readers working on the basal ganglia, it is not discussed.

The revised Discussion includes a section on how altered basal ganglia output may affect targets to alter behavior.

Author Response

Reviewer #3 (Public Review):

In the submitted manuscript, Eliazer et. al. conclude that Dll4 and Mib present on myofibers maintain a continuum of SC fates providing SCs capable of regenerating muscle and repopulatin the SC niche. The data provide new insights into the maintenance of SCs, demonstrating niche-derived factors are responsible for regulating SC behavior. Loss of either Dll4 or Mib from the myofiber reduces SC numbers and impairs muscle regeneration. Overall the data provide compelling evidence that niche-derived Dll4 and Mib regulate SC fate, however, whether the interaction maintains a continuum of SC fates as concluded by the authors is insufficiently supported by the data provided.

We thank the reviewer for their comments.

One significant issue with the manuscript is the "discovery" of an SC continuum related to the relative levels of Pax7 expression. A similar continuum was established nearly a decade ago by Zammit et al., 2004 and Olguin et al., 2004 and thus, is not new. The authors need to reference the work and discuss the prior published data with regard to the observations in the current manuscript. The data establishing a continuum of SCs and the relationship to Pax7 protein levels can largely be eliminated and referenced by the two former manuscripts. For example, these manuscripts establish that elevated Pax7 levels drive quiescence and low Pax7 levels correlate with differentiation. The data from these manuscripts establish that SCs with modest Pax7 protein levels can acquire quiescence accompanied by increases in Pax7 protein

The omission of these two seminal papers was a massive oversight on our behalf. They have now been included. In the original manuscript we acknowledged that SCs exist on a continuum-a gradual transition from one state to another, based on scRNA-seq studies and the present data (Dll4, Pax7 and Ddx6 expression). The references for the sequencing data were included. But with all due respect to the reviewer, the Zammit and Olguin papers binned Pax7 into discrete classes once satellite cells had activated. This is not a demonstration of a continuum. Moreover, we do not make any statements about Pax7 levels in activated conditions. Therefore, the reviewer is drawing comparisons between two different contexts. The statements we have made as they pertain to a continuum under homeostatic conditions are accurate with publications to date.

The data relating the level of Pax7 expression with Dll4a and Mib are intriguing but the authors do not establish a direct relationship, demonstrating that Dll4 or Mib regulate Pax7 levels. An alternative explanation is that Dll4 and Mib inhibit differentiation and thus promote SC quiescence indirectly. This is a critical distinction, as the authors could be correct and Dll4 via Mib regulate SC fate.

We don’t make the claim that Dll4/Mib1 regulates Pax7 directly. We would side with the majority of publications showing that Notch signaling directly regulates Pax7. We have now added further experiments to examine whether Dll4 regulates Notch signaling. We crossed a transgenic mouse line harboring a Notch reporter with MF-Dll4 mice to analyze Notch signaling in SCs. The first experiment we performed with this reporter was to correlate the levels of Pax7 and Notch signaling on a cell-by-cell basis. In control mice, we found a linear positive relationship between levels of Pax7 and the Notch reporter. Next, we compared Notch reporter levels in control versus Dll4-null. We observed that Notch reporter levels decreased to below detectable levels in Dll4 null muscle. Therefore, Dll4 acts non-autonomously to regulate Notch signaling in SCs during homeostasis (refer to Reviewer 1 comment 1, and Essential revisions #3).

The reviewer raises an important point: Does Notch regulate quiescence directly or a differentiation/commitment program when SCs are in a quiescent state. We never claimed that Dll4/Mib1 regulates quiescence. The only way to conclude anything about quiescence would be to examine expression of proliferative markers in vivo. Rather, throughout the manuscript we referred to Dll4 regulating the state of the quiescent SC pool, as measured by changes in Pax7 and Ddx6 expression. In the discussion section we had discussed that Notch signaling may regulate differentiation/commitment of cells in a quiescent state.

It is unclear that the loss of Dll4 or Mib1 reduce diversity of SCs. If these repress differentiation then their loss would be expected to enhance differentiation and reduce SC numbers, which is what the data demonstrate.

Diversity can be restated as the variability across a population. We demonstrate that the variance of Pax7 and Ddx6 expression decreases after Dll4 deletion. Important to note that we are analyzing the SCs that are not lost through differentiation. The fact that some of the SCs are lost through differentiation is not inconsistent with a shift in the continuum. We expect SCs to be lost through differentiation as they shift along the continuum towards a Dll4/Notch/Pax7 low state.

We observe reduced number of Dll4/Pax7 high cells, which is consistent with a shift in continuum. The counterpoint would be that Dll4/Notch/Pax7 high cells commit to differentiation. There is no evidence for that conclusion in this work or any other work published to date. We discussed this issue in the results section.

We have also performed an experiment where mice were treated with a lower dose of TMX to reduce rather than delete Dll4. We find that the total number of SCs does not change, while the relative number of Dll4/Pax7 high cells is reduced while mid and low are increased (Figure 4). This is consistent with a shift in a continuum of states.

Finally, the injury data provided are for 4d post injury and thus, the data may represent a delay in regeneration as opposed to a failure to regenerate. At 30 d post injury regeneration is typically considered complete. How do wild type and Dll4 null as well as Mib null muscle compare at 30d post injury.

We analyzed muscle regeneration of MF-Dll4fl/fl tissue, 40 days after injury. The mean CSA of muscle fibers are significantly smaller than the control fibers, suggesting a defect in tissue regeneration. This is now included in Figure 5-figure supplement 2. Due to time constraints, we have not performed the same experiment with Mib1 mutants.

Author Response

Reviewer #1 (Public Review):

Weaknesses:

The data presented here is, on the whole, descriptive. Whilst the descriptive elements are strong and important, more analysis and quantification is required to support the conclusions made in the paper. For example, in contrast to their analysis of the rail-MIP, their assertion that the ciliary vane orientation is linked to the CPC orientation is not backed up by quantification. In addition, this paper does not extensively discuss proteins within the MIP densities and central pair complex in detail, to the extent they can be discussed using the recent structures from Chlamydomonas.

We thank the reviewer for pointing out these areas for improvement, which are addressed. We are grateful for their helpful suggestions, which we have incorporated to the best of our ability to improve the quality of the manuscript.

Author Response

Reviewer #2 (Public Review):

Wang et al. elegantly exploit single-cell RNA-seq datasets to question the putative involvement of lncRNAs in human germ cell development. In the first part of the study, the authors use computational approaches to identify and characterize, from existing data, lncRNAs expressed in the germline. Of note, the scRNA-seq data used were generated from polyA+ RNAs, and thus non-polyadenylated lncRNAs could not be retrieved. Most of the lncRNAs identified in the germ cells and in the somatic cells of the gonads were previously unannotated. While this increases the catalog of lncRNA genes in the human genome, further characterization is needed to determine which fraction of these newly identified lncRNAs represent bona fide transcripts or transcriptional noise.

Differential expression analysis between developmental stages, sexes, or cell types led to several observations: (i) whatever the stage of development, the number of expressed lncRNAs is higher in fetal germ cells compared to gonadal somatic cells; (ii) there is a continuous increase in the number of expressed lncRNA during the development of the germline; of note, a similar, although the more subtle trend is observed for protein-coding genes; (iii) the developmental stage at which there is the highest number of lncRNA expressed differs between male and female germ cells. While convincing, the significance of these observations is difficult to assess. However, the authors remain prudent with their conclusion and are not over-interpreting their findings.

We appreciate Reviewer #2 precise summary of our analysis and highlighting the significances of these datasets for other researchers and future studies.

Interestingly, integrating lncRNA expression to classify cell types led to the identification of a novel population of cells in the female germline that had not been revealed by protein-coding gene only-based classification. The biological relevance of this population, which cluster with mitotic populations, remains to be demonstrated. Finally, by examining lncRNA biotype, the authors could demonstrate an enrichment, in the germ cells, of the antisense head-to-head organization (in relation to the nearby protein-coding gene) compared to other biotypes. Whether this is different from the general distribution of lncRNA should be discussed.

We analyzed the lncRNAs in NONCODEv5 database (human genome), and the result showed that XH type occupied 21.73% of the intragenic lncRNA-mRNA pairs in NONCODEv5 database (human genome), which is lower than 26.58% in fGC and 26.23% in mGC (Response Figure 1).

Response Figure 1. Genomic distribution and biotypes of the lncRNAs in NONCODEv5 database and lncRNAs expressed in human gonad.

In the second part of the manuscript, Wang et al focus on one pair of divergent lncRNA-protein coding genes (LNC1845-LHX8). To document the choice of this particular pair, it would be informative to have its correlation score indicated in Figure 3C. he existence of this transcript was validated using female fetal ovaries, and its function was addressed in late primordial germ cells like cells (PGCLC) derived from human embryonic stem cells (hESCs). The authors have used an admirable set of orthogonal approaches that led them to conclude as to a role for LNC1845 in regulating in cis the nearby gene LHX8. They further went on to identify the underlying mechanisms, which involve modification of the chromatin landscape through direct interaction of LNC1845 with a histone modifier. Among the different strategies used (KO, stop transcription, overexpression), the shRNA-mediated knock-down is the only one to specifically address the function of the transcript itself, as opposed to the active transcription. The result of this experiment led the authors to conclude that the LNC1845 RNA is functional, a conclusion that is reinforced by the demonstration of physical interaction between the LNC1845 RNA and WDR5, a component of MLL methyltransferase complexes. The result of the KD experiment is however puzzling as RNAi has been shown not to be the method of choice for targeting nuclear lncRNAs (Lennox et al. NAR 2016).

We thank the Reviewer #2’s suggestion to add the correlation score of LNC1845-LHX8 pair and the Pearson Correlation of this pair is 0.3268. We have added the number to Figure 4C because which the expression correlation of LNC1845 and LHX8 was first mentioned. We have compared many other similar studies, shRNA knockdown has been widely used to target nuclear lncRNAs (Guttman et al. Nature 2011; Luo et al. Cell Stem Cell 2016; Subhash et al. Nucleic Acids Res. 2018; Li et al. Genome Res 2021), and the knockdown efficiency seemed to be feasible and acceptable to be used. The knockdown results are consistent with the deletion mutation and stop transcription approaches, all three showed that LNC1845 transcriptional expression is required for proper LHX8 expression in late PGCLCs.

Overall, the functional investigation is convincing and strengthened by the inclusion of multiple clones for each approach, and by the convergence in the outcome of each individual approach. The depth of characterization is also remarkable. The analyses of the mechanisms at stake are somehow less solid, as there is less evidence demonstrating the involvement of the LNC1845 RNA and its interaction with WDR5.

We have added more experimental evidence to strengthen the model especially the interaction of LNC1845 and WDR5. Apart from the RIP-qPCR results of WDR5 demonstrating the enrichment of LNC1845 by WDR5 pulldown (Figure S8D), we performed chromatin isolation by RNA purification (ChIRP) assay using antisense oligos along the entire LNC1845 transcript sequence. ChIRP results confirmed that WDR5 protein were enriched when anti-LNC1845 oligo probes were used to isolate the complex but not the controls without the probes or without overexpression of LNC1845 transcript (Response Figure 2). Taken together, the findings of both approaches support the model that LNC1845 directly interacts with WDR5 to modulate the H3K4me3 modification for LHX8 transcriptional activation. (Related to supplementary figure 8D and 8E.)

Response Figure 2. LNC1845 binding for WDR5 was verified by CHIRP-western blot.

Altogether, this study provides a convincing demonstration of the role of a lncRNA on the regulation of a nearby gene in the context of the germline. However, to have a better understanding of the functionality of lncRNA genes in general, it would be interesting to know whether other pairs of lncRNA-PC genes have been functionally investigated in this context, where no function for the lncRNA gene could be demonstrated. Negative results are highly informative and if so, these could be included in the manuscript.

We appreciate Reviewer #2 suggestion to add other lncRNA-PC gene pairs results. In fact, we have analyzed and presented the results of another 2 pairs in figure 7D. LncRNAs LNC3346 and LNC15266 were also transcriptionally regulated by FOXP3, and they may regulate their neighbor genes TMCO1 and MPP5, as figure 7D showed. Our analysis showed that other lncRNA-PC gene pairs may also have the similar transcriptional regulation as LNC1845-LHX8 during germ cell development.

Author Response

Reviewer #1 (Public Review):

Einarsson et al have produced CAGE data from EBV-immortalised lymphoblastoid cells from more than a hundred individuals from two genetically diverse African populations (YRI and LWK), and used it to study how sequence variation affects the activity of promoters at the level of expression variability and at the level of transcription start site usage within promoters across individuals.

The dataset is very exciting, and the analyses were performed carefully and described well. The results show that promoters in the genome vary a lot with respect to their expression variability across individuals and that their level of variability is closely associated with their biological function and their sequence and architectural features. These results are often confirmatory - it is well established that promoters have different architectures associated with different sequence elements, different types of gene regulation and even differences across individual cells. In general, the multifarious observations boil down to one key distinction:

• Regulated genes have promoters that look and act differently from those of housekeeping genes.

We are pleased that the reviewer is as excited as we are about the unique dataset, the rigorous analyses performed, and the biological results. While we agree that housekeeping and regulated genes show apparent differences in terms of promoter variability, our analyses were not informed or guided by the expression of promoters/genes across cell types or tissues, but rather by their variability within the same cell type across individuals. It is indeed interesting that the same underlying mechanisms that cause stable expression across cell types also attenuates variability across individuals. And, similarly, promoters that display cell-type restricted (regulated) expression levels tend to also be more variable within the same cell type across individuals. While one may argue that these relationships are unsurprising they have, to the best of our knowledge, not been demonstrated before. Of note, however, while most low variable promoters regulate housekeeping genes and highly variable ones regulate regulatory genes, this is not always the case.

While this is unsurprising, the authors then proceed to analyse other underlying differences between low variability (mostly housekeeping) and high variability (overwhelmingly regulated) promoters. Several observations have alternative and sometimes more elegant explanations if some of the previously worked out properties of housekeeping vs regulated promoters are taken into consideration:

• The authors are keen to interpret the architectural features of ubiquitously expressed (housekeeping) promoters as selected for robustness against mutations in ensuring stable and steady expression levels.

However, there are some known facts about both housekeeping and regulated promoters that make alternative interpretations plausible.

• When discussing broad promoters, the authors disregard the well known fact that the most commonly used transcription start positions are those with YR sequence at (-1,+1) position. Any mutation within the span of broad promoter cluster that removes an existing YR or introduces a new one has the capacity to change both the TSS distribution pattern and overall level of expression of that promoter - but only slightly. This way, broad promoters can be viewed as adaptation not for robustness but for ability to take many mutations with small effect size that will drive any positive selection smoothly across a changing fitness landscape.

We thank the reviewer for these remarks. We fully agree with the scenario described by the reviewer, that disruptions of TSSs may have different consequences depending on whether this would be in a broad promoter with multiple YR sequences or within a sharp promoter. However, we argue that the observation that promoters containing such flexible TSSs are not affected much upon genetic perturbations reveals robustness. Per definition, robustness is the ability to produce a persistent phenotype (in our case the molecular phenotype of promoter expression) even in perturbed conditions (e.g. under the influence of natural genetic variation affecting TSS usage). The very fact that TSS disruptions will only have small effect sizes in certain promoters but not in others, tells us that the unaffected or only mildly affected promoters have architectural properties that minimize the effect sizes of these disruptions and thereby cause robustness in overall promoter expression. Hence, we do not see our explanations and those of the reviewer to contradict each other.

• Indeed, the main property of low variability promoters is that there isn't a single nucleotide change (either substitution or indel) that can substantially change their activity. (In that they are clearly different from e.g. TATA-dependent promoters, where one change can abolish TBP binding or deprive the promoter of a YR dinucleotide at a suitable distance from the TATA box.) This is achieved by their dependence on broad and weak sequence signatures such as GC composition and nucleosome positioning signal. However, most such genes are not known to have a strict requirement for dosage control. On the contrary, dosage seems to be much more critical for the functional classes that in the authors' analysis show variable expression.

• Whether it is a removal of YR dinucleotide, introduction of a new one, or the change of nucleosome positioning, it seems that the transcription level from housekeeping, low variability promoters is unaffected, or at least affected mildly enough that it is not within the statistical power of the CAGE data across different individuals to detect the difference. Rather than robustness, it can be interpreted as competition - the architecture recruits preinitiation complex at a fairly constant rate, and it is the different YR positions that "compete" for serving as transcription initiation position, with the CAGE signal reflecting the relative effectiveness of each position in that competition. If one of the YR dinucleotides is removed, often the other, neighbouring ones will be used instead. The same might happen for potential multiple nucleosome positioning signals - if one becomes less efficient at stopping a nucleosome, another will be used more often.

• The fact that decomposed parts of housekeeping promoters add up to approximately the same expression level across individuals even when they are uncorrelated point that they might actually be anticorrelated - indeed, the UFSP2 plot in Figure 4E looks like the two decomposed promoters are anticorrelated. That would argue against the independence of the decomposed promoters - indeed it may again point to "competition" where the decrease in use of one will simply shift most initiation events to the other.

We thank the reviewer for these thoughts. The reviewer has made an excellent observation regarding the correlation between decomposed promoters within low variable promoters. While decomposed promoter pairs of highly variable promoters frequently have correlated expression levels, low variable multi-modal promoters often contain decomposed promoters that have low or even negative expression correlation across individuals. We agree that negative correlation points to the possibility that these decomposed promoters are competing for the transcriptional machinery. Indeed, nucleosome positioning analysis (described below), suggests the existence of diverse configurations of chromatin accessibilities within low variable multi-modal promoters with low or negatively correlated decomposed promoters. This may suggest a competition between the usage of their decomposed promoters. We have revised the manuscript to better reflect this aspect, discussed the potential for YR shifts encoded within the promoter sequence, and also toned down the independence of decomposed promoters. However, regardless of whether decomposed promoters are independent (low correlation) or competing for the transcriptional machinery (negative correlation), we do not agree that this violates our conclusion of robustness. A competition between decomposed promoters within a low variable multi-modal promoter would favor the strongest decomposed promoter, and if the strongest decomposed promoter is affected by genetic perturbation (for instance though disruption of YRs or proximal TF binding) this will affect the competition and shift the dominant usage to another decomposed promoter, as suggested by the frQTL analysis, leading to minimal change in total promoter expression, i.e. a robust molecular phenotype.

• In general, not everything is a result of direct evolutionary selection, and that is what should have clear landmarks of purifying selection. On the contrary, promoters, especially housekeeping promoters, have vastly different nucleotide and dinucleotide compositions across Metazoa, both at large and at relatively short distances, which means they can undergo concerted evolution as a group, which means they should be "robust" to mutations in a way that allows them to change much more and more rapidly than some other promoter architectures - especially TATA-dependent architectures whose key elements and spacing between them haven't substantially changed for more than a billion years, and possibly longer.

We fully agree with the reviewer and have revised the manuscript to remove the evolutionary aspect of robustness. We believe our results are better interpreted with regards to the existence of inherent mechanisms of low variable multi-modal promoters to provide regulatory robustness. Indeed, the vastly different sequence composition of housekeeping promoters between species makes these properties even more interesting. We do not believe that the robustness for perturbations need to be encoded by a specific sequence signature. Rather, we observe that multimodal promoters with low variability require broad initiation regions and a flexibility in the usage of TSSs. This fits well with observations in flies (Schor et al, 2017, DOI: 10.1038/ng.3791) of shifts in the shape of the promoter, which we believe to reflect shifts in decomposed promoter usage, upon genetic perturbation.

• While housekeeping promoters are broad but mostly not among the broadest, regulated promoters can be either broad or narrow. This is also known - while narrow promoters are overrepresented for tissue-specific and non-CGI promoters, promoters of Polycomb-bound developmental genes are often broad and have large CpG islands; the latter may account for some of the broadest CAGE clusters observed in the data. It would be an interesting finding if both TATA-dependent and developmental promoters were found to be variable across individuals in a non-trivial way (the trivial way being the variability due to larger dynamic range of their expression - e.g. the expression of SIX3 in many cell types is basically zero, while the dynamic range of RPL26L1 is very limited) - this should be checked by analysing them separately.

We agree that an analysis of the variability of developmental, Polycomb-bound promoters would be very interesting and thank the reviewer for ideas for a follow-up study. We do not feel that LCLs are the best model system for analyzing developmental promoters and therefore argue that this is out of scope in this study.

• While broad promoters can be decomposed into subclusters with differential expression across individuals, the authors do not seem to allow for the decomposition of intertwined TSS positions within the cluster, but rather postulate hard boundaries between subclusters. This is different from e.g. overlapping maternal and zygotic promoter use (Haberle et al Nature 2014), where the distribution of the used TSS positions is different but the clusters can overlap.

This is correct, we do not allow for overlapping decomposed promoters. We agree that the work by Haberle et al (2014, DOI: 10.1038/nature12974) on switches between maternal and zygotic TSSs is an excellent demonstration of how intertwined promoters can occur and be of biological relevance. Our analysis is based on the observation that low variable promoters are often multimodal and can not be well-explained by simply the width of promoters. This led us to decompose multimodal promoters into their sub-peak constituents. We believe that the frQTL analysis and the new decomposed promoter QTL (dprQTL) analysis clearly demonstrate the value of our approach. While it would indeed be interesting to see the results of an alternative approach for decomposition, we feel this is out of scope in this study but acknowledge that additional determinants of promoter variability may possibly be discovered using alternative strategies.

• Both Dreos et al (PLOS Comp Biol 2016) and Haberle et al. (2014) show that one stable element of a broad promoter is the positioning signal of its first downstream nucleosome. As seen very convincingly in both Drosophila and zebrafish, the dominant TSS position of the broad promoter is highly predictive of the position of first downstream nucleosome and its underlying positioning sequence, and the most plausible interpretation is that there is an "optimal" distance from nucleosome for transcriptional initiation, resulting in the dominant (i.e. most often used) TSS position. In mammals, broad promoters are even broader than in those two species and might have multiple nucleosome positioning signals they can use. In such cases, mutations in one of the nucleosome positioning signals, or indels changing the spacing between the nucleosome and the part of sequence that contains TSS, might lead to differential use of one nucleosome signal vs other. This would be compatible with the authors' observations in low variability promoters that decomposed promoters are used to different extents in different individuals.

We thank the reviewer for this excellent suggestion. In the revised manuscript, we have analyzed both the preference of the distance between the dominant TSS and the downstream (+1) nucleosome and the positional fuzzyness of that nucleosome. We observe a clear separation between low variable multimodal promoters with highly correlated decomposed promoters and those with low correlated decomposed promoters. Interestingly, those with low correlated decomposed promoters show a much less restrictive +1 nucleosome positioning with higher fuzziness, in contrast to what we would expect from broad CGI promoters having a reported fixed +1 nucleosome positioning. While this may be unexpected, it fits well with a model on how a flexible nucleosome positioning architecture can allow differential usage of decomposed promoters. Our results suggest that an array of underlying nucleosome positioning configurations exists for these promoters across single cells, which causes fuzzy nucleosome positioning and may allow for a competition between initiation sites, which provide robustness through their compensatory usage. Interestingly, we find that these results are consistent when analyzing the relationship between transcription initiation and nucleosome positioning within a single individual. This suggests that there is an inherent mechanism of flexibility in TSS usage in these robust promoters even when there is no differential influence of genetic variants. However, to which extent TSS preference is affected by nucleosome positioning or whether nucleosome positioning reflects TSS usage remains unclear. We believe these results further strengthen our general conclusions and thank the reviewer for this constructive suggestion of new analysis.

• If we were to look for sources of difference other than the actual sequence architecture, some differences between regulated and unregulated promoters can be explained by the key difference: the regulation of regulated genes comes from outside the core promoter; the regulation of housekeeping genes is largely dependent on the intrinsic activity of the core promoter itself. This way, for example, in the absence of a causative variant in the promoter itself, the observed variability in the SIX3 promoter might not be encoded in the promoter itself - instead, enhancer responsiveness might be encoded in the promoter, and the variability itself could be due an enhancer that can be hundreds of kilobases away. Such a scenario combined with broad promoter would likely result in decomposed promoters that are highly correlated across individuals - because they are both externally controlled by the same regulatory inputs.

These thoughts are very much in line with our own ideas on how enhancers may influence expression variation. Here, we aimed to investigate variability from a promoter perspective and we are confident that we observe several promoter features associated with low variability. Describing these, we agree that it is important to speculate also on the added contributions by distal elements. We now acknowledge the likely added contribution by enhancers in the Discussion:

“The promoter sequence may also encode a promoter’s intrinsic enhancer responsiveness (Arnold et al., 2017), which may influence its expression variability. Although current data cannot distinguish between direct or secondary effects, an increased variability mediated via enhancers is supported by a higher dependency on enhancer-promoter interactions for cell-type specific genes compared to housekeeping genes (Furlong and Levine, 2018; Schoenfelder and Fraser, 2019). However, compatibility differences between human promoter classes and enhancers only result in subtle effects in vitro (Bergman et al., 2022), suggesting that measurable promoter variability is likely a result of both intrinsic promoter variability and additive or synergistic contributions from enhancers. Directly modeling the influence and context-dependency of enhancers on promoter variability would therefore be important to further characterize regulatory features that may amplify gene expression variability.”

Reviewer #2 (Public Review):

This manuscript by Einarsson and colleagues in the Andersson lab examined how genetic variability across a population impacts both gene expression and promoter architecture in a human population. The authors generate new CAGE data in 108 lymphoblastoid cell lines (LCLs). The authors' analysis is focused on defining how DNA sequence and promoter architecture correlate with population-variation in expression across this cohort. In general, there is a lot that I like about this manuscript: The dataset will be an extremely valuable resource for the genomics community. Furthermore, the biological findings are often thoughtful and potentially interesting and significant for the community. The analysis is generally very strong and is clearly conducted by a lab that has a lot of expertise in this area. My main concerns are centered around the often unwarranted implication that DNA sequence or promoter features cause differences in variation at different genes.

We are pleased that the reviewer is as excited as we are about the unique dataset, the rigorous analyses performed, and the biological results. In our revised manuscript we have followed the recommendations by the reviewer and:

● Toned down implied causal relationships and added additional interpretations to our results, including YR positional preferences

● Performed additional analyses on nucleosome positioning of low variable promoters, as well as genetic association testing for decomposed promoter expression

In all, we believe these revisions substantially improved our manuscript and even strengthened our previous conclusions.

Author Response

Reviewer #1 (Public Review):

The manuscript "Centrally expressed Cav3.2 T-type calcium channel is critical for the initiation and maintenance of neuropathic pain" identifies a subset of parvalbumin-expressing GABAergic neurons in the anterior pretectum (APT) that co-express Cav2.3 T-type calcium channels. The firing frequency and burst patterns are potentiated in these neurons following spared nerve injury (SNI) and the development of neuropathic pain. Deletion of the channels in these cells reduced both the development and maintenance of mechanical and cold allodynia. Studies show nice co-expression of the PV and GFP in the Cav2.3.2eGFP-flox KI mouse line.

Multi-unit recordings from PV-Cre X Ai32 mice show that PV neurons in the APT are fast-spiking and that the mean firing rate and frequency of spikes in bursts are potentiated in SNI animals. The graphs in Fig. 2, panel F show compiled data of 18-20 cells from 6-8 animals depending on the treatment. The statistical design for the in vivo experiments (and actually all of the studies) are not clearly stated with degrees of freedom. It is important to know if recordings from a single animal are considered independent observations, and if so, what the rationale for that is. This information should be included in the Quantification and Statistical Analysis section. In addition, it would be interesting to determine if T-type calcium channel blockers can reverse this behavior in these recordings.

We considered each unit as an independent observation. We did so since the number of recorded PV+-units per animal (identified with the PINP method) was small and varied greatly between animals, from 2 to 6 units. We are not aware of statistical methods using a nested design for multiple cells in animals that could be used in such condition.

Since the measured variables did not follow normal distributions, we performed unpaired comparison with the Wilcoxon sum rank test. This is now stated in the section ‘quantification and statistical analysis’ (page 19, line 683). P-values are now included in the figures and the result section when appropriate.

In vitro electrophysiological studies show that the PV-expressing APT neurons exhibit fast-spiking to depolarization and single-cell RT-PCR shows that Cav3.2 is expressed in APT neurons that also express GABA. These cells show an after-hyperpolarization burst of APs that is reduced by blockers of Cav3.2 channels. There are no statistics displayed on panels C-E in Fig. 3, although they are reported in the text. Again, the test used and degrees of freedom, etc. should also be reported as it allows for evaluation of the experimental design.

We apologize for the lack of statistics in Fig. 3 (now Fig. 4). Statistics are now clearly presented on each figure panel and the statistical tests are stated in the figure legends and in the results (page 4, line 144-159).

As it is now stated in the “Quantification and Analysis section” (page 19, line 682), each neuron was considered as an independent observation since 1 to 3 neurons were recorded per mouse. The number of mice and the mean number of neurons per mouse are indicated in the data-set for each experimental condition in order to allow for a clear evaluation of the experimental design. Note that in the experiments with application of T-channel antagonist, only one neuron was recorded per slice. This is now specified in the Method Details (page 17, line 579).

It is also noted in the Discussion (lines 185-186) that "Our in vitro data indicate that 92% of APT-PV+ neurons are able to discharge bursts of action potentials at high frequency underpinned by a large transient depolarization due to the activation of T channels." It would be more clear to refer to the rebound as the figure also shows the fast-spiking properties due to depolarization as well as the transient depolarization due to the rebound but only an effect of the Cav2.3 on the rebound.

We agree and have changed the sentence accordingly (page 5, line 202).

Behavioral studies of mechanical and cold allodynia in male and female naïve and SNI-treated KI and KO mice were performed. These results show a clear contribution of the Cav3.2 channels in APT in both the development and maintenance of neuropathic pain. Again, the statistical design is not clearly defined and it is extremely difficult to resolve what comparisons are delineated in panels B-E of Fig. 4.

We fully agree with the reviewer that the rationale for the choice of statistical tests used to analyze the behavioral data was lacking. We have rewritten the relevant paragraph in the Quantification and Analysis section (page 19, lines 699-714). The statistical results presented in the Fig 4 and its supplemental figure (now Fig 5 and Figure 5 – Figure supplement 1) are now clearly stated in the legends.

Reviewer #3 (Public Review):

The authors used state-of-the-art techniques to investigate the role of centrally located (GABAergic APT neurons) CaV3.2 isoform of T-channels in an animal model of neuropathic pain using speared nerve injury model. This is generally an excellent and very rigorous study. The data is very compelling and it is likely going to have a major impact in the field of ion channels and pain transmission. The data presentation is superb and major conclusions are highly justified. Major strengths include the use of powerful complementary techniques such as molecular (single-cell PCR), mouse genetics, and pain testing in vivo, as well as sophisticated ex vivo (slice physiology) and in vivo recordings (burst analysis using tetrodes). This study may explain recent clinical studies that failed to show the efficacy of peripherally acting Cav3.2 channel blockers in patients with neuropathic pain. Hence, this study has the potential to change the focus from peripheral to supraspinal Cav3.2 channels in various pain pathologies.

Some moderate weaknesses are identified and should be addressed:

1) The data showing the effect of T-channel deletion on the excitability of GABAergic neurons of APT is very convincing. However, what is missing is a discussion of how changes in the excitability of inhibitory APT neurons impact the circuitry that is involved. Without knowing the circuitry involved, one could speculate that blocking inhibitory drive may do just the opposite effect of what is proposed and increase hyperalgesia.

We agree with the reviewer that discussing this issue is essential and it has now been added (page 6, lines 270-295).

2) Methods should clearly state if any experiments were done in a blinded fashion.

Behavioral experiments were performed blind. This has been added in the method section (page 18, line 624). For in vivo electrophysiological experiments, we cannot say that we performed blind experiments (although we tried). Indeed, under anesthesia, the forelimb of SNI animals presents a slight but observable withdrawal.

3) There is no mention anywhere of how was selective Cav3.2 knock-out achieved, nor how was this assessed. It would be very helpful if authors could perform recordings of T-channel amplitudes in sham animals, animals after SNI and after selective knock-out in the SNI group.

The efficiency of the Cav3.2 deletion after Cre virus injection was assessed by immunolabeling of GFP in APT slices. As shown in Figure 5 – Figure supplement 1, we checked that unilateral injection of AAV8-hSyn-Cre-mCherry virus induced a drastic reduction in the number of GFP+ neurons when compared to the non-injected hemisphere. The absence of Cav3.2 expression in Cre injected APTs was systematically checked in each mouse at the end of the behavioral tests (Figure 5A). This is now added in the Method Details (page 18, line 655).

4) It should be discussed that global Cav3.2 animals had only minimal neuropathic pain phenotype (Choi et al., 2007).

This point is now discussed (page 6, line 251).

Author Response

Reviewer #3 (Public Review):

Dingus et al. have developed an innovative and powerful approach for improving the intracellular stability of nanobodies. Nanobodies are single chain antibodies that are typically generated in select species such as llamas or alpacas. Because nanobodies are secreted and are present in general in the extracellular environment, they often become unstable when expressed in the reduced intracellular environment. Dingus et al. investigated 75 nanobodies from the Protein Data Bank and found that 42 were unstable when expressed intracellularly. In order to improve stability of these nanobodies, they first determined consensus residues that were present within the framework region, which does not include the CDR regions, in over 80% of the stable nanobodies. Mutating residues within the framework of unstable nanobodies to match consensus residues in the stable nanobodies stabilized 26 of 42 nanobodies. Mutating consensus unstable residues stabilized another 11. Thus 37/42 unstable nanobodies were stabilized using this mutational approach. Further experiments provided evidence that some of the stabilized nanobodies still had some affinity for their targets. Furthermore, one stabilized nanobody was stable when expressed in the retina in vivo and 3 of 5 were stable when expressed in bacteria.

1) This study provides a straightforward approach to improving the intracellular stability of nanobodies that could prove to be very useful for solving a common and vexing problem.

Thanks!

2) From the data provided, it was difficult to determine whether the binding affinity of the mutated nanobodies had been diminished by the mutations that increased stability, and if so, by how much. Furthermore, target binding affinity was assessed for just 5 nanobodies, which calls into question whether this strategy will be useful.

It is the case that we are unable to guarantee that any nanobody stabilized by our consensus-based approach will retain full target-binding affinity. It is additionally not guaranteed that a given nanobody will be able to bind its target in cells in the absence of any mutagenesis, as paratope structure may be influenced/compromised in the intracellular environment. We are additionally limited in what we can test intracellularly, as the majority of current nanobodies target extracellular factors that cannot be effectively expressed intracellularly. What we provide is a rationale for limited impact on target binding via partial consensus mutagenesis, which excludes highly variable framework positions, most likely to contribute directly to binding, from mutagenesis. While our approach to generalizable intracellular stabilization may not be perfect for every nanobody, we believe it is likely to be a simple and useful approach in a variety of cases, and likely the majority of cases.

3) Ultimately, the goal of expressing most nanobodies intracellularly is to bind to endogenous targets. It is difficult to assess how useful the stabilization strategy will be since it was not determined whether any of the stabilized nanobodies could bind their endogenous targets intracellularly.

We are limited in the number of intracellular targets we are currently able to test, as most current nanobodies target extracellular antigens. Endogenous intracellular targets are even more limited. However, we agree that targeting endogenous targets is ultimately the goal. We have included an in vivo experiment against the endogenous target GFAP in our revised manuscript, where we show that binding is preserved following mutagenesis (Figure 6).

Author Response

Reviewer #1 (Public Review):

Kohler and Murray present high-throughput image-based measurements of how low-copy F plasmids move (segregate) inside E. coli cell. This active segregation ensures that each daughter cell inherit equal share of the plasmids. Previous work by different labs has shown that faithful F-plasmid segregation (as well as segregation of many other low-copy plasmids, segregation of chromosomes in many bacterial species and segregation of come supramolecular complexes) require ParA and ParB proteins (or proteins similar to them) and is achieved by an active transport mechanism. ParB is known to bind to the cargo (plasmid) and ParA forms a dimer upon ATP binding that binds to DNA (chromosome) non-specifically and also can bind to ParB (associated with cargo). After ATP hydrolysis (stimulated by the interaction with ParB), ParA dimer dissociates to monomers and from ParB and the chromosome. While different mechanisms of the ParA-dependent active transport had been proposed, recently two mechanisms become most popular - one based on the elastic dynamics of the chromatin (Lim et al. eLife 2014, Surovtsev PNAS 2016, Hu et al Biophys.J 2017, Schumaher Dev.Cell 2017) and the other based on a theoretically-derived "chemophoretic" force (Sugawara & Kaneko Biophysics 2011, Walter et al. Phys.Rev.Lett. 2017).

It is a minor comment, but we would like to point out that we do not consider these two model types as alternatives but rather as models with different levels of coarse-graining. Our interest is in the molecular-level (stochastic) models (Lim et al. eLife 2014, Surovtsev PNAS 2016, Hu et al PNAS 2015, Hu et al Biophys.J 2017, Schumacher Dev.Cell 2017).

The authors start by following motion of F plasmid with one or two plasmids per cell and by analyzing plasmid spatial distribution, plasmid displacement (referred to as velocity) as a function of their relative position, and autocorrelations of the position and the displacement. They concluded that these metrics are consistent with 'true positioning' (i.e. average displacement is biased toward the target position - center for one plasmid and 1/4 and 3/4 positions for two plasmids ) but not with 'approximate positioning' (i.e. when plasmid moves around target position, for example, in near-oscillatory fashion). This 'true positioning' can be described as a particle moving on the over-dampened spring. They reproduce this behavior by expanding the previous model for 'DNA-relay' mechanism (Lim et al. eLife 2014, Surovtsev PNAS 2016), in which plasmid is actively moved by the elastic force from the chromosome and ParA serves to transmit this force from the chromosome to the plasmid. Now, the authors explicitly consider in the model that the chromosome-bound ParA can diffuse (which the authors refer as 'hopping') and this allows the model to achieve 'true plasmid positioning' for some combination of model parameters in addition to oscillatory dynamics reported in the original paper (Surovtsev PNAS 2016).

Based on their computational model, the authors proposed that two parameters, diffusion scale of ParA = 2(2Dh/kd)1/2/L (typical length diffused by ParA before dissociation) and ratio of ParB-dependent and independent hydrolysis rates = kh/kd are key control parameters defining what qualitative behavior is observed - random diffusion, near-oscillatory behavior, or overdamped spring ('true positioning'). They vary this two parameters ~30- fold and ~200-fold range by changing Dh and kh respectively, to illustrate how dynamics of the system changes between these 3 modes of motion. While these parameters clearly play important role, the drawback is that the authors did not put either theoretical reasoning why these parameters are truly governing or showed it by varying other model parameters (kh, number of ParA NParA, spring constant of chromosome k, diffusion coefficient of the plasmid Dp) to show that only these combinations define the type of the system behavior. The authors qualitative analysis on importance of relies on the steady state solution for the diffusion equation for ParA. It is really unfortunate that no ParA distribution was measured simultaneously with the plasmid motion, as this would allow to compare experimental ParA profiles to expected quasi-steady-state solutions.

We spend almost an entire section and a figure explaining the theoretical reasoning behind the identification of the $\lambda=s/(L/2n)$ as an important system parameter (section “Hopping of ParA-ATP on the nucleoid as an explanation of regular positioning” and Figure 2) and predicted that regular positioning could only occur for $\lambda>1$. This was confirmed by parameter sweeps for the cases of 1 (Figure 3I) and multiple plasmids (Figure 5-figure supplement 1), indicating that $\lambda$ is indeed an important system parameter and that our conceptual understanding of this aspect of the system is correct. This point has now been made clearer.

However, we agree that the reasoning for $\epsilon$ (varied through the hydrolysis rate $k_h$) was not clear. It was chosen to allow us to modulate the ParA concentration at the plasmid compared to elsewhere, motivated by the differences between different ParABS systems. We originally had also considered a third quantity related to the number of nucleoid-bound ParA but we found that this had little effect on the nature of the dynamics. All three quantities describe how the timescale of a reaction/process (ParA hopping/diffusion across the nucleoid, ParB induced hydrolsysis, ParA association to the nucleoid) compares to the timescale of basal hydrolysis, which we use as a reference timescale.

We have now made this clearer as well as adding supplementary figures showing the effect of varying other system parameters at several locations in the phase diagram (Figure 3-figure supplement 3 and 4). These sweeps justify our identification of $\epsilon$ and $\lambda$ as a useful/important set of quantities for determining the dynamics of the system.

Additionally, we now add example kymographs showing the ParA distribution (Figure 3-figure supplement 2C).

The authors also show by simulations that overdamped spring dynamics can transition into oscillatory behavior when decreases, for example by cell growth. Indeed, they observed more oscillatory behavior when they compared single-plasmid dynamics in the longer cells compared to the shorter cells. This was not the case in double-plasmid cells, in eprfect agreement with their analysis. They also calculated ATP consumption in the model and concluded that the system operates close but below (perhaps, "above" should be used as it refers to bigger ) the threshold to oscillatory regime which minimize ATP consumption. While ATP consumption analysis is very intriguing, this statement (Abstract Ln24-25) seems at odds with the authors own analysis that another ParA-dependent plasmid system, pB171, operates mostly in oscillatory regime, and it is actually for this regime the authors' analysis suggest minimal ATP-consumption (Fig. 8).

To clarify, we found that pB171 (which in our hands has a copy number of 2-3 in the SR1 reduced-copy-number strain) is only clearly oscillatory in cells with a single plasmid (and only mildly so in cells with two plasmids). Otherwise, it behaves very similarly to F plasmid. We therefore believe that these two distantly related ParABS systems exhibit, overall, similar dynamics and differ only in how close the systems are to the threshold of oscillatory instability. This was not clear as we did not specify the copy number of pB171. We now provide this in Figure 7–figure supplement 1.

We refer to these systems as lying just below, rather than above, the threshold of the oscillatory instability because, on average, plasmids do not oscillate but only do so in cells with the lowest plasmid concentration.

I think the real strength of the paper is that it can potentially to show that if one considers that the intracellular cargo can be moved by the fluctuating chromosome via ParA-mediated attachments, then various dynamics can be achieved depending on combinations of several control parameters (plasmid diffusion coefficient, ParA diffusion coefficient, rate of hydrolysis and so on) including previously reported 'oscillations' (Surovtsev PNAS 2016), 'local excursions' (Hu et al Biophys.J 2017) and 'true positioning' (Schumaher Dev.Cell 2017). The main drawback (in this reviewer opinion) that this is obscured by the current presentation and discussion of this work and previous modelling work on ParA-dependent systems. For example, instead of using "unifying" potential of the presented model, yet another name 'relay and hopping' is used in addition to previously used 'DNA-relay', 'Brownian ratchet', 'Flux-based positioning', …

In the abstract and discussion, we already refer to developing a “unified” model (p1 L21, p15 L22 of the original manuscript) and in the discussion we explain how our model contains other models as limiting cases. But we agree with this recommendation - the unifying nature of our model is its main strength. We now emphasise this more.

Regarding the model name, we felt obliged to refer to the previous named models (DNA-relay and Brownian ratchet) and simply gave our model a name to avoid confusion when making comparisons. We have now removed almost all mention of ‘hopping and relay’ and just refer to ‘our model’. However, our gitlab repository with the code must have a name and therefore is still called ‘Hopping and relay’ and so the same term is used in Table 3.

… and it appears that the presented model is an alternative to these previously published work. And only in model description (in Methods section) one can find that the "... model is an extension of the previous DNA-relay model (Surovtsev et al., 2016a) that incorporates hopping and basal hydrolysis of ParA and uses analytic expressions for the fluctuations rather than a second order approximation"(p.17, ln15-17).

We are sorry that this reviewer felt that the fact that our model is an extension of DNA relay is hidden in the methods. However, we wrote in the main text:

“Motivated by the previous discussion, we decided to develop our own minimal molecular model (‘hopping and relay’) of ParABS positioning, taking the DNA relay model as a starting point … The original scheme is as follows… We supplemented this scheme with two additional components: diffusion (hopping) of DNA-bound ParA-ATP dimers across the nucleoid (with diffusion coefficient Dh, where the subscript indicates diffusion of the home position) and plasmid-independent ATP hydrolysis and dissociation (with rate kd). See Material and Methods for further details of the model. “

We now make this clearer.

However, we would argue that as models of the same system, there are naturally overlaps and the models of Hu et al and Schumacher et al could also be thought of as extensions of the DNA relay model.

While it is of course the authors right to decide how to name their model, it should be explicitly clear to the reader what is a real conceptual difference between presented and previous models from the abstract, introduction and discussion section of the paper, not from the "fine-print" details in the supplementary materials.

The main conceptual difference is that we have identified the importance of having a finite diffusive length scale for ParA diffusion/hopping on the nucleoid. This allows both oscillations and regular positioning to occur for biologically relevant parameter values and reproduces the length dependent transition from mid-cell positioning to confined oscillations that we observe for F plasmid. The DNA relay model does not have this behaviour as the ParA diffusive length scale in zero while it is infinite in the models of Ietswaart et al 2014 and Schumacher et al 2017. The model of Hu et al 2017 does have a finite length scale but the authors appear not to have realised its importance and never discovered the regular positioning regime at \lambda >1. While we make these points in the discussion in the context of Figure 8A, where we compare our model to the others, we agree with this reviewer that we should have been more explicit in the abstract and introduction. We have now corrected this.

This would allow to avoid unnecessary confusion (especially for the readers not directly involved into the modelling of ParA/B system) and clarify that all these models rely on the elastic behavior of fluctuating chromosome to drive active transport of the cargo. This reviewer believes that more explicit discussion on the models (one from the authors and previously published) differences and similarities will help with our understanding of how ParA-dependent system operate. This discussion should also include works on PomXYZ system, in which it was shown that similar dynamic system can lead to specific positioning within the cell (Schumaher Dev.Cell 2017, Kober et al. Biophys.J 2019). This will may it explicit that the models results have direct impact beyond the ParA-dependent plasmid segregation.

To further clarify the differences between the models (beyond the second and third sections of the main text and the discussion), we have now added a section to the methods and a new table (Table 3). We have also included the mentioned PomXYZ model. However, we would like this was not the first stochastic model to have ‘true’ positioning as this reviewer cites above. Though they did not include the mechanism of force generation, the model of Ietswaart et al 2014 produces regularly positioned plasmids and is referenced repeatedly in Schumacher et al. 2017.

I think that expanded parameter analysis, and explicit model comparison/discussion will make the contribution of this work to the field more clear and with the potential to advance our general understanding of how the same underlying mechanism can lead to various modes of intracellular dynamics and patterning depending on parameters combination.

Reviewer #2 (Public Review):

The work presented in this manuscript details an analysis of the partitioning of low copy plasmids under the control of the ParABS system in bacteria. Using a high throughput imaging set up they were able to track the dynamics of the partition complex of one to a few plasmids over many cell cycles. The work provides an impressive amount of quantitative data for this chemo-mechanical system. Using this data, the paper sought to clarify whether the dynamics of plasmids is due to regular positioning or noisy oscillations around a mean position. They supplement their experimental work with an intuitive model that combines elements of previous modelling efforts. Their model relies on diffusion of the ParA substrate on the nucleoid with the dynamics of the ParB partition complex being driven by the underlying elastic force due to the nucleoid on which the substrate is tethered. Their model dynamics depend on two parameters, the ratio of the length over which the substrate can explore to the characteristic length of the space and the ratio of stimulated to non-stimulated hydrolysis rates of the substrate. If the length ratio is large, ParA can fully explore the space before interacting with the ParB complex leading to balanced fluxes and regular positioning. If it gets reduced, for example by lengthening the cell, oscillations can emerge as fluxes of substrates become imbalanced and a net force can pull the partition complex.

Strengths:

Given the large amount of data, the observations unambiguously show that one particular ParABS system under the conditions studied is carrying out regular positioning of plasmids. The model synthesizes prior work into a nice intuitive picture. These model parameters can be fit to the data leading to estimates of molecular kinetic parameters that are reasonable and in line with other observations. Lining up the experimental observations with the phase space of the model suggests that the system is poised on the edge of oscillations, allowing for the system to have regular positioning with low resource consumption.

Weaknesses:

However, despite the correspondence of the simulated results with the experimental findings, other explanations are not completely ruled out. The paper emphasizes that ParA diffusion/hopping on the nucleoid is essential for the establishment of regular positioning and that without it, only oscillations were possible. Prior simulation efforts, that the paper cites, which include ParA diffusion and mixing in the cytosol but no diffusion on the nucleoid have shown that regular positioning is possible and that oscillations could get triggered as the system lengthened. Thus ParA hopping is not a necessity for regular positioning (as claimed in the paper), but very well might be needed for the given kinetic parameters of the system studied here.

We now comment on this result. In short, we believe that the mentioned model/regime is not relevant due to stochastic effects. We are not able to produce, with biological relevant parameters, regular positioning without ParA hopping.

The paper also presents experimental results for a second ParABS system (pB171) that is more likely to show oscillations. They attribute the greater likelihood of oscillations for pB1717 being due to ParA exploring a smaller space than the F plasmid system that showed regular positioning. This is pure conjecture and the paper does not provide any evidence that this is the reason. Thus it is hard to conclude if oscillations may not be due to other factors.

We do not explicitly make that claim. We did have a point in the phase diagram of Figure 8A representing pB171 with a lower value of lambda than F plasmid and stated “The location of pB171 is an estimate based on a qualitative comparison of its dynamics”. We agree this was unclear.

We now indicate the region that has oscillations with roughly the same period as single plasmids of pB171. We also make it clear that we speculate, but have not shown, that the length scale of ParA hopping is smaller than for F plasmid.

An important point here is that we can explain both oscillations and regular positioning in the same model with the same kinetic parameters, the regimes being determined by the cell length and plasmid number in a manner consistent with experimental observations.

Author Response

Reviewer #1 (Public Review):

This work sheds light on the adverse effects of Bacillus thuringiensis, a strong pathogenic bacteria used as a microbial pesticide to kill lepidopteran larvae that threaten crops, on gut homeostasis of non-susceptible organisms. By using the Drosophila melanogaster as a non-susceptible organism model, this paper reveals the mechanisms by which the bacteria disrupt gut homeostasis. Authors combined the use of different genetic tools and Western blot experiments to successfully demonstrate that bacterial protoxins are released and activated throughout the fly gut after ingestion and influence intestinal stem cell proliferation and intestinal cell differentiation. This phenomenon relies on the interaction of activated protoxins with specific components of adherens junctions within the intestinal epithelium. Due to conserved mechanisms governing intestinal cell differentiation, this work could be the starting point for further studies in mammals.

The conclusions proposed by the authors are in general well supported by the data. However, some improvements in data representation, as well as additional key control experiments, would be needed to further reinforce some key points of the paper.

We thank reviewer1 for her appreciation of the work and in depth analysis of the data. We agree with all her comments and believe the suggestions significantly improved the manuscript.

1) Figure 1 and others: Several graphs in the manuscript show the number of cells/20000µm2. How is the shape of the gut in the different conditions studied in this manuscript? The gut shape (shrunk gut versus normal gut for example) could influence the number of cells seen in a small area. For example, the number of total cells quantified in a small area (here 20000µm2) of a shrunk gut can be increased while their size decrease. As a result, the quantification of a specific cell type in a small region (here 20000µm2) can be biased and not represent the real number of cells present in the whole posterior part of the R4 region. Would it make sense to calculate a ratio "number of X cells/number of DAPI positive cells per 20000µm2"?

We provided a suitable answer in the "Essential Revisions point 1" corresponding to this reviewer's concern. To summarize, we have now added whole posterior midgut images in the different conditions to highlight the intestinal morphology (Figure 1-figure supplement 1A). The whole gut morphology was not affected by the different challenges we performed. Indeed, we used low doses of spores and/or toxins in order to mimic "natural" amounts of spores/toxins the fly can eat in the environment and in order to avoid drastic gut lining disturbances.

We have also added the cell type ratio in figure 1- figure supplement 2.

2) Figure 4: Is it possible that Arm staining is less intense between ISC and progenitors after ingestion of the bacteria due to the fact there is a high rate of stem cell proliferation? Could it be an indirect effect of stem cell proliferation rather than the binding of the toxins to Cadherins?

We thank the reviewer for this pertinent comment. Indeed, for this reason, we compared the intensity of Arm expression at the junction between neighboring progenitors with the Arm intensity around the rest of the cellular membranes and calculated the ratio between both values (see Figure 4-figure supplement 1F-G for an illustration of how we proceeded and the new section in the Material and Methods 736-742). Using this method, even if the whole Arm staining intensity is different (in all the midgut), the ratio reflects the internal cell-cell interaction changes between the two neighboring cells. Moreover, we have observed that Arm staining (using the usual monoclonal antibody N2 7A1 from the DSHB) was very variable from one midgut to another in the same feeding/intoxication condition. So, we do not want to draw conclusion about the whole Arm intensity due to this variability whatever are the intoxication conditions. Finally, the challenged guts always displayed a more disorganized epithelium due to cell proliferation and differentiation. Consequently, Arm staining in ECs and progenitor cells are found in the same focal plane while in unchallenged and well-organized guts, Arm staining in ECs is above the focal plane of Arm staining in progenitor cells. This likely leads to the impression that Arm staining is more intense in challenged midguts. This method description is now added in the Material and methods section (lines 736-742).

Could the authors use the ReDDM system to distinguish between "old" and newly formed cells? This could be a good control to make sure that the signal is quantified in similar cells between the control and the different conditions.

We have analyzed intensity of Arm expression between pairs of GFP cells. Most of these pairs arose from de novo divisions. Indeed, as shown in control conditions (water) with Dl-ReDDM (for example see figure 1-figure supplement 1D), pairs of GFP cells (ISC-ISC) are rare. Most pairs correspond to ISC-EB or ISC-EEP pairs with the progenitor marked by the RFP, meaning that it just arises from the GFP+ mother ISC. Therefore we assume, that in the esg>GFP genotype, pairs of GFP+ cells correspond to one ISC and one progenitor (see Figure 4 – figure supplement 1A-A'). Therefore, when we analyzed the Arm intensity between pairs of GFP cells after intoxication, these cells are very likely "newborn" cells. Even if we suppose there are ISCs and progenitors that remain stuck together for a long time (for instance several days), Cry1A toxins can also be able to disrupt their cell junction. In the context of Cry1A toxin activity, it seems important to analyze the whole impact on cell-cell junctions without discriminating old and new cell-cell interactions.

We tried to use anti-Arm and anti-Pros double staining to mark new EEPs. Unfortunately, anti-Arm and anti-Prospero antibodies were both raised in mice. Co-staining with both antibodies give rise to bad labelling either for Arm or for Prospero or for both. Our first author spent lot of energy trying to set up good conditions but unfortunately this was unsuccessful.

Here is an example of what we got (this was the best image we got) with esg>GFP flies fed with water (control) and labelled for Arm and Pros in red. White arrows point two EEPs. Red arrows points the Arm staining between two precursors (ISC/ISC or ISC/EB or EB/EB). It was extremely hard to identify junctions marked by Arm between EEPs and ISCs because the Pros staining was too strong.

Another example with flies fed with spores of SA11 (increasing the number of EEs). In green is the esg>GFP and in Red Arm and Prospero. The right panel correspond to the single red channel (Arm/Prospero).

Nevertheless, we have now performed a similar analysis in an esg>GFP, Shg::RFP background and analyzed Shg::RFP (Tomato::DE-Cadherin) labelling intensity. We found similar results that are presented in the new Figure 4 (data we Arm have been moved in Figure 4-figure supplement 1). This last analysis have been included in the text lines 285-299.

Figure 4E' and 4G': Arm staining seems more intense when looking at the whole membrane levels of cells compared to control. Is it possible that the measured ratio contact intensity/membrane intensity presented in Figure 4I could be impacted and not reflect the real contact intensity between ISC and progenitor cells?

Please check our answer just above: "…//… we have observed that Arm staining (using the usual monoclonal antibody N2 7A1 from the DSHB) was very variable from one midgut to another in the same feeding/intoxication condition. So, we do not want to draw conclusion about the whole Arm intensity due to this variability whatever are the intoxication conditions".

See also our intensity measurement method described above to avoid bias: "…//… we compared the intensity of Arm expression at the junction between neighboring progenitors with the Arm intensity around the rest of the cellular membranes and calculated the ratio between both values (see Figure 4-figure supplement 1F-G for an illustration of how we proceeded and the new section in the Material and Methods 736-742). Using this method, even if the whole Arm staining intensity is different (in all the midgut), the ratio reflects the internal cell-cell interaction changes between the two neighboring cells."

What is the hypothesis of the authors about the decrease of Arm or DE-Cad seen after bacterial/crystal ingestion? Does the interaction between the toxins and DE-Cad induce a relocation of DE-Cad?

It has been shown that E-Cadherin could be recycled when adherens junctions are destabilized both in Drosophila and mammals(Buchon et al., 2010; O'Keefe et al., 2007; Tiwari et al., 2018). To investigate this possibility, we tried to analyze DE-Cad cytoplasmic relocalization using anti-DE-Cad immunostaining (DCAD2 antibody from DSHB) as well as Shg::RFP (Bloomington stock #58789) or Shg::GFP (Bloomington stock #60584) endogenous fusion. Unfortunately, we did not see obvious differences. Nevertheless, we have now added the split channels of the Shg::RFP labelling in the different conditions in Figure 4A-D'. Nevertheless, we are still interested in the behavior of the DE-cadherin (and signaling, see (Liang et al., 2017)) upon binding of the Cry1A toxin. N. Zucchini-Pascal (author in this article) are currently investigating this question.

The authors should add more details about the way to quantify in the Material and methods section. How many cells have been quantified per intestine? How did they choose the cells where they quantified the contact intensity?..etc

These details were missing in the methods and we thank the reviewer for highlighting this issue. We added these information to the methods (lines 725-742). The number of cell pairs analyzed was present in the raw data related to figure 4 but absent from the main figure and legend. It is now rectified. We only measured the intensity in isolated pairs of cells.

Figure 4B, D, F and H: How did the authors recognize the ISCs?

We agree with the reviewer comment. We cannot recognize ICS per se. Green cells correspond either to ISCs or to EBs. We modified the text accordingly (lines 285-287).

Could the authors do quantifications of DE-Cad signal?

This has been done. It is shown now in figure 4E and in Table 1. We also adapted the text (lines 289-299) to fine-tune our interpretation in light of this new analysis. Indeed, what we have now defined as "mild" adherens junction intensity is between the ratio 1.4 and 1.6 instead of the previous ratio (1.3 to 1.6), because we observed most of the EEP progenitors arising from cell displaying a junction intensity with their mother cells below the 1.4 ratio (see Table 1).

Like Arm staining, the staining seems stronger at the whole membrane level in F and H compared to the control.

As we described above for Arm staining, the intensity of Tomato::DE-Cad labelling can differ from one posterior midgut to another one. One simple explanation would be related to changes in the structure of midgut epithelium which is well organized in unchallenged conditions, while in challenged midguts the epithelial cells are not well-arranged anymore due to rapid cell proliferation and differentiation. Consequently, DE-Cad labelling in ECs is at the same level as that in ISC/progenitors cells, giving the impression that the labelling is stronger.

3) Figure 5: How is the stem cell proliferation upon overexpression of DE-Cad in control or upon bacteria/crystals ingestion? Do the authors think that the decrease of Pros+RFP+ new cells upon overexpression of DE-Cad could result from a decrease of stem cell proliferation?

Great suggestion. Thereby, we chose to count the progenitor cells (GFP+ cells) reflecting the ISC division during the last 3 days. Moreover, this also has the advantage of working on the same pictures (samples) used for all the analyzes shown in figure 5 and Figure 5-figure supplement 1. Hence, If we consider the number of GFP+ cells (esg expressing cells corresponding to ISC, EB or EEP) in challenged midguts, the overexpression of the DE-Cad did not seem to alter ISC division. In addition, we still observed more GFP+ cells when the midguts were challenged with SA11 or crystals than with BtkCry, in agreement with the rate of ISC division observed in the WT genetic background shown in figure 1B.

We have now added the counting of GFP+ cells in Figure 5-figure supplement 1E. The text has been modified to integrate this results (lines 306-308).

Did the authors quantify the % of new ECs in the context of overexpression of DE-Cad?

The data has been added in figure 5F. The text has been modified to integrate this result lines 312-313.

Figure 5F: As asked before, did the authors distinguish the signal between newly born cells and the signal between older cells?

In the new figure 5G: we used the esg-ReDDM system that is very efficient. Almost all ISC and progenitors express the GFP. The counting have been done between cell pairs that express both the GFP and RFP. It is specified in the text lines 310-311. Nevertheless, we cannot distinguish between new and old cells here. Indeed, the esg-ReDDM system induce both the GFP and the RFP in all esg+ cells (the old ones and the new ones). Hence, if a division has occurred just before the induction of the system to give birth for instance to an ISC and an EB, both cells will express the GFP and the RFP. But should we consider those pairs of cells as old cells or new cells? Noteworthy, as we analyzed the intensity of junctions 3 days after intoxication and induction of the ReDDM system, we assume that the pairs of GFP+/RFP+ cells arose after the induction of the system. Indeed, to our knowledge, nobody has shown in the posterior midgut, that a progenitor remains stuck to its mother ISC as long as 3 days. Even if we assume that this event can occur, Cry1A toxins can also be able to disrupt their cell junction.

We now have removed the DAPI channel and added the RFP+ channel in Figure 5-figure supplement 1A-D' (previously the Figure S4A-D) to illustrate this explanation and to facilitate the interpretation by the reader.

It would be interesting to compare the junction intensity between mother ISCs and their daughter progenitors before and after intoxication in a same intestine. But we think that this event is quite rare because of the experimental conditions we used (i.e. analyses 3 days after the induction of the ReDDM/intoxication).

The same experiments (stem cell proliferation + quantification of the % of new ECs) could be also done when authors overexpress of the Connectin, supplemental figure 5. This would be another control to conclude that the effects on cell differentiation are specific due to the interaction between DE-Cad and the toxins.

We have added the analyses in Figure 5 - figure supplement 2J and K.

The text has been completed lines 317-320.

In the "crystals" condition, the overexpression of Connection seems to partially rescue the increase % of new Pros+RFP+ new cells observed in Figure 3F (Figure S5I compared to Figure 3F).

Yes, we agree with the reviewer comment. In an esg-ReDDM background (figure 3F), crystals induced a much greater increase in EE numbers than did SA11 spores. However, in a WT or esg>GFP background, crystals induced a similar increase in EE/EEP to that induced by SA11 spores. So we do not yet have explanation excepted the genetic background of the esg-ReDDM.

Author Response

Reviewer #1 (Public Review):

The authors use the nanobody tools generated in the companion manuscript and have combined them with DNA-Paint oligonucleotide labeling to generate super-resolution images of indirect flight muscles. Using this approach, they could map the precise organization of the different domains from the two giant titin-like fly homologs called Sallimus and Projectin against which the nanobodies had been raised with a precision ranging from 1 nm to 4 nm, depending on the distance between them. They show that in indirect flight muscles the N-ter of Sallimus is located within 50 nm of the Z-disc, and that its C-ter reaches the A-band roughly 100 nm away from the Z-disc. Likewise, they show that the N-ter of Projectin colocalizes with the C-ter of Sallimus at the edge of the A-band, whereas its C-ter is located about 250 nm away in the A-band and 350 nm from the Z-disc. It overall suggests a staggered and linear organization of both proteins with a potential area of overlap spanning 10-12 nm, that Sallimus could bridge the Z-disc to the A-band acting as a ruler, while Projectin should only overlap with 15% of the A-band and possibly a 10 nm of the I-band.

Thanks for this nice summary of our findings.

The value of this work comes from its use of advanced technologies (DNA-Paint + superresolution). The biological conclusions confirm and refine earlier and recent papers, especially EM papers and the impressive and very comprehensive JCB paper by Szikora et al in 2020, although the conclusions of the present work differ somewhat from those of Szikora who had predicted that Sallimus does not reach the A-band. That aspect could have been better discussed.

We have further extended our discussions of the results from Szikora et al. 2020, in particular regarding Sallimus in this revised version.

Reviewer #2 (Public Review):

Taking advantage of the high molecular order of the Drosophila flight muscle, Schueder, Mangeol et al. leverage small (<4 nm) original nanobodies, tailored coupling to fluorophores, and DNA-PAINT resolution capabilities, to map the nanoarchitecture of two titin homologs, Sallismus and Projectin.

Using a toolkit of nanobodies designed to bind to specific domains of the two proteins (described in the companion article "A nanobody toolbox to investigate localisation and dynamics of Drosophila titins" ), Schueder, Mangeol et al position these domains within the sarcomere with <5nm resolution, and demonstrate that the N-ter of Sallismus overlaps with the C-ter of Projectin at the A-band/I-band interface. They propose this architecture may help to anchor Sallismus to the muscle, thus supporting flight muscle function while ensuring muscle integrity.

This study nicely extends previous work by Szikora et al, and precisely dissect the the sarcomeric geography of Sallismus and Projectin. From these results, the authors formulate specific functional hypotheses regarding the organization of flight muscles and how these are tuned to the mechanical constraints they undergo.

Although they remain descriptive in essence, the conclusions of the paper are well supported by the experimental results.

We thank this reviewer for the nice summary of our results.

Reviewer #3 (Public Review):

This manuscript by Schueder et al. provides new insight into an important question in muscle biology: how can the smaller titin-like molecules of the much larger sarcomeres of invertebrate muscle perform the same function as the larger titin of vertebrate muscles which have smaller sarcomeres? These functions include the assembly, stability and elasticity of the sarcomere. Using two state of the art methods--nanobodies and DNA-PAINT superresolution microscopy, the authors definitively show that in the highly ordered indirect flight muscle of Drosophila, the elongated proteins Sallimus and Projectin are arranged such that the N-terminus of Sallimus is embedded in the Z-disk, and the C-terminus is embedded in the outer portion of the A-band, and that in this outer portion of the A-band is also embedded the C-terminus of Projectin; thus, if the C-terminus of Sallimus can bind to thick filaments, and/or these overlapping portions of Sallimus and Projectin interact, there would be a linkage of the Z-disk and/or thin filament to the thick filaments to help determine the length and stability of the sarcomere.

The strengths of this paper include the implementation of nanobody and DNA-PAINT superresolution microscopy for the first time for muscle. The extraordinary 5-10 nm resolution of this method alloiws imaging for definitive localization of the termini of these elongated proteins in the Drosophila flight muscle sarcomere. In addition, the manuscript is well written with sufficient background information and rationale presented, is easy to read, complex new methods are well-described, the figures are of high quality, and the conclusions are well-justified. A minor weakness is that despite the authors demonstrating that the Cterminus of Sallimus is located at the outer edge of the A-band, and that the N-terminus of Projectin is located also in the outer edge of the A-band, the authors provide no data to show whether, for example, these portions of these titin-like molecules interact, or whether Sallimus might interact with thick filaments. Such data would be required to prove their model. However, I can understand that this would require extensive additional study, and the authors have already provided a tremendous amount of data for this first step in supporting the model. Nevertheless, the authors should cite a relevant previous study on the Sallimus homolog in C. elegans called TTN-1, which is also a 2 MDa polypeptide of similar domain organization to at least the large isoforms of Salliums found in fly synchronous muscles. In the study by Forbes et al. (2010), immunostaining, albeit not to the impressive resolution achieved in the present paper, showed that TTN-1 was also localized to the I-band with extension into the outer edge of the A-band. More importantly, that study also showed that "fragment 11/12", Ig38-40, which is located fairly close to the C-terminus of TTN-1 binds to myosin with nanomolar affinity (Kd= 1.5 nM), making plausible the idea that TTN-1 may bind to the thick filament in vivo.

We thank this reviewer for sharing his enthusiasm about our results and methodology, and also about the way the data are presented. This is one more argument for us to leave a shortened Figure 1 in the PAINT manuscript.

We are particularly thankful for pointing out the important C. elegans data that we had missed and that, as the reviewer said, perfectly fit with the model we propose for flight muscle (and also the larval muscle data, as the C-term of Sls is the same). Hence, we highlight this paper now in our discussion and compare to our findings.

Reviewer #4 (Public Review):

This manuscript reports combining recently developed and described in the accompanying paper nanobodies against Sallimus and Projectin with DNA-Paint technology that allows super-resolution imaging. Presented data prove that such a combination provides a powerful system for imaging at a nano-scale the large and protein-dense structures such as Drosophila flight muscle. The main outcome is the observation that in flight muscle sarcomeres Salimus and Projectin overlap at the I/A band border. This was elegantly achieved using double color DNA-Paint with Sls and Projectin nanobodies.

We thank the reviewer for appreciating the quality of our work.

Overall, as it stands, this manuscript even if of high technological value, remains entirely descriptive and short in providing new insights into muscle structure and architecture. The main finding, an overlap between short Sls isoform and Proj in flight muscle sarcomeres, is redundant with the author's observation (described in the companion paper "A nanobody toolbox to investigate localisation and dynamics of Drosophila titins") that in larval muscles expressing a long Sls isoform, Sls and Proj overlap as well.

Alternatively, combination of Sls and Proj nanobodies with DNA-Paint represents an interesting example of technological development that could strengthen the accompanying nanobodies toolkit manuscript.

Every structural paper reports the structure and is thus by definition descriptive. This is the aim of our manuscript. We do not think that the other nanobody resource paper reports an overlap of Sls and Projectin in the larvae. To resolve such a possible overlap, super resolution would be needed. The other paper does report that larval Sls isoform is dramatically stretched, more than 2 µm, and that Projectin is decorating the thick filament, likely in an oriented manner. If N-term of Projectin overlaps with C-term of Sallimus in this muscle is an open question that needs DNA-PAINT imaging of larval muscle. This requires a TIRF setting that is technically not trivial to achieve for larval muscle and hence has not been done by anybody.

Author Response

Reviewer #2 (Public Review):

Point 1: The transcriptomic analysis of E12.5 endocardial cushion cells in the various mouse models is informative in the extraction of Igf2- and H19-specific gene functions. In Fig. 6D, a huge sex effect is obvious with many more DEGs in female embryos compared to males. How can this be explained given that Igf2/H19 reside on Chr7 and do not primarily affect gene expression on the X chromosome? Is any chromosomal bias observed in the genomic distribution of DEGs?

We examined chromosomal distribution of DEGs between WT and +/hIC1 (Supplemental Figure 6D) and did not see any bias on X chromosome. We described this result on lines 278-280: “Although the number of +/hIC1-specific DEGs largely differed between males and females, there was no sex-specific bias on the X chromosome (Supplemental Figure 6D).” Additionally, we agree with the reviewer that it is noteworthy that the dysregulated H19/Igf2 expression affected transcriptome in a sex-specific manner, especially when the mutation is located on a somatic chromosome. Although investigating the role of hormones versus sex chromosome in these effects would be quite interesting, it is beyond the scope of current study.

Point 2: A separate issue is raised by Fig. 6E that shows a most dramatic dysregulation of a single gene in the delta3.8/hIC1 "rescue" model. Interestingly, this gene is Shh. Hence, these embryos should exhibit some dramatic skeletal abnormalities or other defects linked to sonic hedgehog function.

The reason why Shh appeared to be differentially expressed between wild-type and d3.8/hIC1 samples was that Shh expression was 0 across all the samples except for two wild-type samples. In order to detect all the DEGs that might be lowly expressed, we did not want to filter DEGs based on the level of total expression. As a result, Shh was represented as significantly differently expressed in d3.8/hIC1 samples, although its expression in our samples appears to be too low to have any significant effect on development. This explanation was added to lines 310-312. To confirm that this was an exceptional case, we analyzed the expression of DEGs obtained from other pairwise comparisons. In the volcano plots below, genes of which expression is not statistically different between two groups are marked grey. Genes of which expression is statistically different and detected in both groups are marked red. Genes with statistically different but not detected in one group at all, such as Shh, are marked blue (Figure G). It is clear that that almost all of our DEGs are expressed consistently across the groups, and genes with no expression detected in one group are very rare.

Point 3: The placental analysis needs to be strengthened. Placentas should be consistently positioned with the decidua facing up, and the chorionic plate down. The placentas in Fig. 3F are sectioned at an angle and the chorionic plate is missing. These images must be replaced with better histological sections.

As requested, we have replaced placental images with better representative sections (Figure 3F and 4E). In addition, we have improved alignment of placental histology figures.

Point 4: The CD34 staining has not worked and does not show any fetal vasculature, in particular not in the WT sample.

As requested, we have replaced the CD34 vascular stained images with those that better represent fetal vasculature (Figure 3G).

Point 5: The "thrombi" highlighted in Fig. 4E are well within the normal range, to make the point that these are persistent abnormalities more thorough measurements would need to be performed (number, size, etc).

As requested, we measured the number and relative size of the thrombi that are found in dH19/hIC1 placentas with lesions. No thrombi were found in wild-type placentas whereas an average of 1.3 thrombi were found in six dH19/hIC1 placentas. The size of the thrombi widely varied, but occupied average of 2.58% of the labyrinth zone where these lesions were found (Supplemental Figure 4D). Additionally, we replaced the image in Figure 4E into the section that better represents the lesion.

Point 6: The statement that H19 is disproportionately contributing to the labyrinth phenotype (lines 154/155) is not warranted as Igf2 expression is reduced to virtually nothing in these mice. Even though there is more H19 in the labyrinth than in the junctional zone, the phenotype may still be driven by a loss of Igf2. Given the quasi Igf2-null situation in +/hIC1 mice, is the glycogen cell type phenotype recapitulated in these mice, and how do glycogen numbers compare in the other mouse models?

The sentence was edited in line 157. We performed Periodic acid Schiff (PAS) staining on +/hIC1 placentas to address if glycogen cells are affected by abnormal H19/Igf2 expression (Supplemental Figure 1E). In contrary to previous reports where Igf2-null mice had lower placental glycogen concentration (Lopez et al., 1996) and H19 deletion led to increased placental glycogen storage (Esquiliano et al., 2009), our quantification on PAS-stained images showed that the glycogen content is not significantly different between wild-type and +/hIC1 placentas. We have described this result in lines 166-168.

Point 7: How do delta3.8/+ and delta3.8/hIC1 mice with a VSD survive? Is it resolved some time after birth such that heart function is compatible with postnatal viability? And more importantly, do H19 expression levels correlate with phenotype severity on an individual basis?

Our study was limited to phenotypes prior to birth, thus postnatal/adult phenotypes were not examined. Because the VSD showed only partial penetrance in these mice, we cannot state that the d3.8/+ or d3.8/hlC1 mice with VSDs survive. It has also been previously reported in another mouse model with incomplete penetrance of a VSD that the mice which survived to adulthood did not have the VSDs (Sakata et al., 2002). We find it highly unlikely that either mouse model would survive significantly past the postnatal timepoint with a VSD. We have examined two PN0 d3.8/hIC1 neonates, and both did not have VSD.

Regarding the second point, the only way to quantitatively address this question would be to do qPCR or RNA-seq on individual hearts, which then makes it impossible for those hearts to be examined for histology to confirm the VSD. Thus, hearts used to identify VSDs via histology could not also be used for quantitative H19 measurements. One thing to note is that the H19/Igf2 expression in independent replicates of d3.8/hIC1 cardiac ECs used in our RNA-seq experiment is quite variable, not clustering together in contrast to other mouse models used in this study (Fig. 6A). Such wide range of variability in the extent of H19/Igf2 dysregulation suggests that H19/Igf2 levels could have an impact on the penetrance or the severity of the VSD phenotype in d3.8/hIC1 embryos.

Author Response

Reviewer #2 (Public Review):

Zylbertal and Bianco propose a new model of trial-to-trial neuronal variability that incorporates the spatial distance between neurons. The 7-parameter model is attractive because of its simplicity: A neuron's activity is a function of stimulus drive, neighboring neurons, and global inhibition. A neuroscientist studying almost any brain area in any model organism could make use of this model, provided that they have access to 1) simultaneously-recorded neurons and 2) the spatial locations of those neurons. I could foresee this model being the de-facto model to compare to all future models, as it is easy to code up and interpret. The paper explores the effectiveness of this distance model by modeling neural activity in the zebrafish optic tectum. They find that this distance-based model can capture 1) bursting found in spontaneous activity, 2) ongoing co-fluctuations during stimulus-evoked activity, and 3) adaptation effects during prey-catching behavior.

Strengths:

The main strength of the paper is the interpretability of the distance-based model. This model is agnostic to the brain area from which the population of neurons is recorded, making the model broadly applicable to many neuroscientists. I would certainly use this model for any baseline comparisons of trial-to-trial variability.

The model is assessed in three different contexts, including spontaneous activity and behavior. That the model provides some prediction in all three contexts is a strong indicator that this model will be useful in other contexts, including other model organisms. The model could reasonably be extended to other cognitive states (e.g., spatial attention) or accounting for other neuron properties (such as feature tuning, as mentioned in the manuscript).

The analyses and intuition to show how the distance-based model explains adaptation were insightful and concise.

We thank the reviewer for these supportive comments.

Weaknesses:

Model evaluation and comparison: The paper does not fully evaluate the model or its assumptions; here, I note details in which evaluation is needed. A key assumption of the model - that correlations fall off in a gaussian manner (Fig. 1C-E - is not supported by Fig. 1C, which appears to have an exponential fall-off. Functions other than gaussian may provide better fits.

A key feature of our model is that connection strengths smoothly decrease with distance. However, we did not intend to make strong claims about the exact function parametrizing this distance relationship. In light of the reviewer’s comment, we have additionally tested an exponential function and find that it too can describe activity correlations in OT with a negligible decrease in r2 (Figure 1 – figure supplement 1A-C). The main purpose of the analysis was to show that the correlation is maximal around the seed and decays uniformly with distance from it (i.e. no sub-networks or cliques are detected). We have emphasized this in a revised conclusion paragraph and note that while multiple functions can be used to parameterize the relationship, they are nonetheless certainly simplifications. Secondly, we also ran a version of the network simulation where the connections decay in space according to an exponential rather than Gaussian function and show that, as expected, tectal bursting is robust to this change.

Furthermore, it is not clear whether the r^2s in Fig. 1E are computed in a held-out manner (more details about what goes into computing r^2 are needed).

These values are computed by fitting the 2-d Gaussian (or exponential function) to all neurons excluding the seed itself (added a short clarification in the Methods).

Assessing the model based on peak location alone (Fig. 1E) is not sufficient, as other smooth monotonically-decreasing functions may perform similarly.

As discussed above, an exponential function indeed performs similarly to a Gaussian. However, goodness of fit is secondary to the main aim of Fig 1E, which is to show that the correlation peak tends to fall near the seed cell.

Simulating from the model greatly improves the reader's understanding (Fig. 2D), but no explanation is given for why the simulations (Fig. 2D) have almost no background spikes and much fewer, non-co-occurring bursts than those of real data (Fig. 2E).

In part this is because the simulation results depicted in Fig 2D were derived from the ‘baseline model’, prior to optimizing to match biological bursting statistics. It is thus expected that activity will differ from experimental observation and was our main motive to tune the model parameters (now emphasized in the text). However, the model will certainly not account for all aspects of tectal activity; rather, it was designed to reproduce bursting as a prominent feature of ongoing activity and in the second part of the paper we explore the extent to which it can account for other phenomena. As noted above, in the revised abstract, introduction and discussion we have tried to clarify the motivation for developing the model and how it was used to gain insight into activity-dependent changes in network excitability.

A key assumption of the distance model (Fig. 2A) is that each neuron has the same gaussian fall-off (i.e., sigma_excitation and sigma_inhibition), but it is unclear if the data support this assumption.

We intentionally opted for a simple model (i.e. described by few parameters), in part due to the lack of connectivity data and additionally to set a lower bound on the extent to which multiple features of tectal activity could be accounted for. More complex models with additional degrees of freedom (such as cell-specific connectivity) may well describe the data better, but likely at the cost of interpretability. We consider such extensions are beyond the scope of the present study but might be fruitful avenues for future research.

Although an excitatory and inhibitory gain is assumed (Fig. 2A), it is not clear from the data (Fig. 1C) that an inhibitory gain is needed (no negative correlations are observed in Fig. 1C-D).

This is now explored in the revised Figure 3A which includes the condition of zero inhibition gain. See also response to reviewer 1.

After optimization (Fig. 3), the model is evaluated on predicting burst properties but not evaluated on predicting held-out responses (R^2s or likelihoods), and no other model (e.g., fitting a GLM or a model with only an excitatory gain) is considered. In particular, one may consider a model in which "assemblies" do exist - does such an assembly model lead to better held-out prediction performance?

The model we developed is a mechanistic, generative model. In contrast to Pillow et al 2008, we did not fit the model to data but rather we used it to simulate network activity and tuned the seven parameters (using EMOO) to best match biological observations. Thus, rather than assessing goodness-of-fit using cross-validation, our approach involved comparison of summary statistics related to the target emergent phenomenon (tectal bursting). This was necessary as bursting appears highly stochastic. Further to the comments above, we have expanded the parameter space to include instances with only an excitatory gain (where bursting failed) and no distance-dependence (again, busting failed). Introducing assemblies into the model will inevitably support bursting (and introduce many more free parameters), but one of our key observations is that such assemblies are not required for this aspect of spontaneous activity. Again, our aim was not to produce a detailed picture of tectal connectivity, but rather to develop a minimal model and estimate the extent to which it can account for observed features of activity. Note that the second half of the paper (Figure 4 onwards) shows the model can explain phenomena that were not considered during parameter tuning.

It is unclear why a genetic algorithm (Fig. 1A-C) is necessary versus a grid search; it appears that solutions in Generation 2 (Fig. 3C, leftmost plot, points close to the origin) are as good as solutions in Generation 30 and that the spreads of points across generations do not shrink (as one would expect from better mutations). Given the small number of parameters (7), a grid search is reasonable, computationally tractable, and easier to understand for all readers (Fig. 3A).

Perhaps in hindsight a grid search would have worked, but at increased computational cost (each instantiation of the model is computationally expansive). At the time we chose EMOO, and since it produced satisfactory results, we kept it. As often happens with multi-objective optimization, an improvement in one objective usually happens at the expense of other objectives, so the spread of the points does not shrink much but they move closer to the axes (i.e. reduced error). The final parameter combination is closer to the origin than any point in generation 2, though admittedly not by much. Importantly, however, optimizing the model using the training features generalized to other burst-related statistics.

It is unclear why the excitatory and inhibitory gains of the temporal profiles (Fig. 3I) appear to be gaussian but are formulated as exponential (formula for I_ij^X in Methods).

The interactions indeed have exponential decay in time. These might appear Gaussian because the axis scale is logarithmic.

Overall, comparing this model to other possible (similar) models and reporting held-out prediction performance will support the claim that the distance model is a good explanation for trial-to-trial variability.

See comments above. A key point we want to stress is that we intentionally explored a minimal network model and found that, despite obvious simplifications of the biology, it was nonetheless able to explain multiple aspects of tectal physiology and behaviour. We hope that it inspires future studies and can be extended, in parallel to experimental findings, to more accurately represent the cell-type diversity and cell-specific connectivity of the tectal network.

Data results: Data results were clear and straightforward. However, the explanation was not given for certain results. For example, the relationship between pre-stimulus linear drive and delta R was weak; the examples in Fig. 4C do not appear to be representative of the other sessions. The example sessions in Fig. 4C have R^2=0.17 and 0.19, the two outliers in the R^2 histogram (Fig. 4D).

The revised figure 4 is based on new data and new analysis (see below), and the presented examples no longer represent the extreme tail of the distribution (they still, however, represent strong examples, as is now explicitly indicated in the figure legend).

The black trace in Fig. 4D has large variations (e.g., a linear drive of 25 and 30 have a change in delta R of ~0.1 - greater than the overall change of the dashed line at both ends, ~0.08) but the SEMs are very tight. This suggests that either this last fluctuation is real and a major effect of the data (although not present in Fig. 4C) or the SEM is not conservative enough. No null distribution or statistics were computed on the R^2 distribution (Fig. 4C, blue distribution) to confirm the R^2s are statistically significant and not due to random fluctuations.

We agree that this was not sufficiently robust and in response to this comment we undertook a significant revision to figure 4 and the associated text:

i) The revised figure is based on an entirely new dataset, allowing us to verify the results on independent data. We used 5 min ISI for all stimulus presentations, regardless of stimulus type (high or low elevation), thus ensuring that we are only examining differences in state brought about by previous ongoing activity, without risk of ‘contamination’ by evoked activity.

ii) As per the reviewer’s suggestion, we compared model-estimated pre-stimulus state to a null estimate using randomly sampled time-points. We additionally compared the optimised model with the baseline model. Whereas the null (random times) estimates had no predictive power, both models using pre-stimulus activity were able to explain a fraction of the response residuals with the optimised model performing better.

iii) We refined the binning process by first computing, for each response, the mean of response residuals across neurons for each bin of estimated linear drive, and then averaging across responses. This prevents the relationship being skewed by rare instances involving unusually large numbers of neurons for a particular linear drive bin, and thereby eliminates the fluctuations the reviewer was referring to.

The absence of any background activity in Fig. 6B (e.g., during the rest blocks) is confusing, given that in spontaneous activity many bursts and background activity are present (Fig. 2E).

The raster only presents evoked responses and no background activity is shown. This has been clarified in the revised figure and legend.

Finally, it appears that the anterior optic tectum contributes to convergent saccades (CS) (Fig. 7E) but no post-saccadic activity is shown to assess how activity changes after the saccade (e.g., plotting activity from 0 to 60).

Activity before and after the saccade is shown in Fig 7A. Fig 7E shows the ‘linear drive’ (or ‘excitability’), and how it changes leading up to the saccade. Since we were interested in the association between pre-saccade state and saccade-associated activity, we did not plot post-saccadic linear drive. However, as can be seen in the below figure for the reviewer, linear drive is strongly suppressed by the saccade, as expected due to CS-associated activity.

No explanation is given why activity drops ~30 seconds before a convergent saccade (Fig. 7E).

This is no longer shown after we trimmed the history data in Fig 7E in accordance with a comment from reviewer 1. We speculate, however, that the mean linear drive of a compact population of neurons would be somewhat periodical, since a high linear drive leads to a burst which results in a prolonged inhibition (low linear drive) with a slow recovery and so on.

No statistical test is performed on the R^2 distribution (Fig. 7H) to confirm the R^2s (with a mean close to R^2=0.01) are meaningful and not due to random fluctuations.

We revised the analysis in Fig 7 along the same lines as the revision of Fig 4. Model-estimated linear drive predicts CS-associated activity whereas a null estimate (random times) shows no such relationship.

Presentation: A disjointed part of the paper is that for the first part (Figs. 1-3), the focus is on capturing burst activity, but for the second part (Figs. 4-7), the focus is on trial-to-trial variability with no mention of bursts. It is unclear how the reader should relate the two and if bursts serve a purpose for stimulus-evoked activity.

In the first part of the paper (Figs. 1-3), we use ongoing activity to develop an understanding (formulated as a network model) of how activity modulates the network state. In the second part, we test this understanding in the context of evoked responses and show that model-estimated network state explains a fraction of visual response variability and experience-dependent changes in activity and behaviour. In the revised MS we further emphasize this idea and have edited the results text to strengthen the connections between these parts of the study. See also comments above.

Citations: The manuscript may cite other relevant studies in electrophysiology that have investigated noise correlations, such as:

• Luczak et al., Neuron 2009 (comparing spontaneous and evoked activity).

• Cohen and Kohn, Nat Neuro 2011 (review on noise correlations).

• Smith and Kohn, JNeurosci 2008 (looking at correlations over distance).

• Lin et al., Neuron 2015 (modeling shared variability).

• Goris et al., Nat Neuro 2014 (check out Fig. 4).

• Umakantha et al., Neuron 2021 (links noise correlation and dim reduction; includes other recent references to noise correlations).

We agree that the manuscript could benefit from citing some of these suggested studies and have added citations accordingly.

Author Response

Reviewer #1 (Public Review):

This manuscript by McCafferty et al. presents the integrative computational structural modelling of the IFT-A complex, which is important to proper cilium organelle formation in eukaryotic cells. Recent advances in protein structure prediction (AlphaFold) allowed the authors to model the structures of the 6 individual subunits of the IFT-A complex. Interactions between IFT-A proteins were experimentally investigated by purifying Tetrahymena cilia, isolating IFT complexes, and utilizing chemical crosslinking and mass spectrometry (MS). In addition, the authors present a somewhat improved 23Å cryo-electron tomography (cryo-ET) map of the IFT-A complex (previously determined cryo-ET structures of IFT trains have resolutions of 24 - 40 Å). Integrative modelling using the predicted structures of the 6 IFT-A proteins and the experimental data as restraints allows the authors to present a structural model for the entire IFT-A complex. This model is analysed in the context of the polymeric IFT train structure, interactions with the IFT-B complex, and the structural position of ciliopathy disease variants.

This is in principle a timely and interesting study that attempts to push the limits of structural modelling of large protein complexes using structure prediction in combination with experimental data. Unfortunately, the study has several shortcomings and the data providing restraints for the integrative modelling are not optimal.

1) Chemical crosslinking and MS were used to obtain both intra-molecular crosslinks used to validate the structural models of the individual IFT-A proteins as well as inter-molecular crosslinks used as restraints in the structural modelling of the hexameric IFT-A complex. It is mentioned on p. 4, line 9, that IFT-A complexes were enriched from the flagellar lysate M+M fractions using SEC and that fractions from SEC containing IFT-A complexes were crosslinked for MS analysis. However, the authors do not show the data for this sample, neither SEC profiles, SDS-PAGE nor data of the cross-linked samples. On p. 7 the authors write that their SEC profile corresponds to monomeric IFT-A, but this is not shown anywhere in the manuscript. The reason this is so important is that the IFT-A complex assembles into linear polymeric structures together with the IFT-B complex as so-called IFT trains in cilia. Data obtained from isolated IFT trains would thus have additional crosslinks between subunits in neighbouring IFT-A complexes that, if used to restrain the position of subunits within a hexameric IFT-A complex, would likely result in a wrong architecture. The fact that the authors also observe crosslinks between IFT-A and IFT-B proteins strongly suggests that they indeed carried out the crosslinking experiment on polymeric rather than monomeric IFT complexes.

These are excellent points, and we apologize for previously omitting these data. In the new Figure 1—figure supplement 2, we now include size exclusion chromatography elution profiles for IFT-A along with molecular weight calibrants, plotting the mass spectrometrically-determined abundances of IFT-A subunits. Based on these data, we experimentally determined the molecular weight of the IFT-A particles that we analyzed to lie between 720 kDa and 1.1 MDa, consistent with the expected monomeric molecular weight of 772 kDa.

These samples were isolated directly from Tetrahymena cilia and were composed of ~3% each of IFT-A and IFT-B. However, as we now note on p. 11, the samples were subsequently concentrated before crosslinking. We speculate that concentrating the particles could have induced some degree of oligomerization and interactions with IFT-B, which may in turn explain the small number of crosslinks consistent with IFT-A/IFT-A and IFT-A/IFT-B interactions. However, we have now removed all discussion of specific IFTA/B contacts in the paper and present only the general orientation of the two complexes as determined by cryo-ET.

2) Given that the crosslink/MS data are unlikely to provide sufficient restraints for IFT-A structure assembly (and may even be misleading), the cryo-ET data become increasingly important. Unfortunately, the 23Å cryo-ET map does not provide sufficient detail to unambiguously fit domains of the IFT-A subunits as several of these have similar architectures consisting of WD-repeats followed by TPRs.

We now address this comment using a different approach, which we describe in full on p. 5, 14-15, and Figure 2—figure supplements 1-4 of the paper.

In particular, we used AlphaFold-Multimer (AF-Multimer) to identify confidently-modeled rigid-body domains and domain-domain interactions for directly contacting protein pairs (see Figure 2—figure supplements 1-2), which we used as starting models for integrative modeling (see Figure 2—figure supplements 3-4). We incorporated our cross-links as distance restraints for the modeling. This approach allowed us to model the entire IFT-A complex in a manner compatible both with our experimental structural data and the computationally derived restraints. We suspect this will be a very useful strategy for others to adopt, as the approach should be generalizable to many other large molecular assemblies that are too big to predict using AF-Multimer alone. Importantly, we see high concordance between the AlphaFold intermolecular constraints and our crosslinks (as plotted in the new Figure 2—figure supplement 4), and the models produced by this strategy agree well with the two structures presented in the newly posted preprints, which were arrived at using very distinct methodologies.

This approach allowed us to withhold the cryo-ET tomogram from the modeling altogether in order to generate a fully independent model. We could then compare the final model to the subtomogram average and, by docking the model into the cryo-ET tomogram, to build a model of polymeric IFT-A, as described on p. 6 and presented in the new Figure 4, Figure 4—figure supplement 1, and Figure 4—animation 1.

3) Two preprints of the IFT-A structure appeared over the last few weeks. Hesketh et al., (https://www.biorxiv.org/content/10.1101/2022.08.09.503213v1) have obtained a single particle cryo-EM structure of the human IFT-A complex at 3.5Å resolution for the IFT121/122/139 part of the complex providing amino acids side-chain information. In addition, Lacey et al. (https://www.biorxiv.org/content/10.1101/2022.08.01.502329v1) provide a 10-18Å resolution cryo-ET structure of the Chlamydomonas IFT trains containing both IFT-A and IFT-B. It is noteworthy that the model outlined in the current manuscript is very different from the IFT-A models of Hesketh et al., and Lacey et al. (the Lacey et al. manuscript by the way shares an author with the McCafferty et al., manuscript). In both Hesketh et al., and Lacey et al. the IFT121 and IFT122 subunits interact via the N-terminal WD-repeats and the C-terminal TPRs with the beta-propellers (WD-repeats) positioned parallel and in close contact. In the model proposed by McCafferty, the beta-propellers of IFT121 and IFT122 are positioned far away from each other (>50Å) and are perpendicular to each other. Several other large discrepancies are found in the relative positions of IFT-A subunits. This suggests serious problems with the structural model of IFT-A proposed by McCafferty and needs to be addressed with great care.

This is an important point that we have indeed considered with great care. Our new model now positions the WD-40 domains of IFT121 and IFT122 proximal to each other and broadly matches the 2 preprints in general placement and orientation of all subunits, including the placement of IFT43, for which only we and Hesketh provide models.

We now include an extensive comparison to the structures reported in the other two preprints. Note that a direct 3D alignment of the structures was not possible, as we were the only group to deposit our atomic coordinates. However, we now include a new Figure 4—figure supplement 2 orienting our structure to match figures appearing in those preprints, and use this as the basis for comparison, which can be found on p. 10. While it is not possible to calculate a quantitative measure of agreement (e.g. RMSD), our IFT43/120/121/139 structure visually agrees with the structure of Hesketh et al., even to the placement of IFT43, which is highly disordered for the most part, and which is omitted from Lacey et al. Our structure also generally agrees with that of Lacey et al. in this region, with the exception of what appears to be a re-orientation of the N-terminus of IFT139 in the Lacey structure relative to that of ours and Hesketh, which appear to be concordant with each other (again, with the caveat that we are limited in the comparisons we can make without having access to atomic coordinates.) Most importantly, all three structures agree with respect to the nature of the IFT-A monomer-monomer interactions in the polymeric train, with IFT140 acting to bridge adjacent monomers. Differences in the resolutions of the cryo-ET subtomogram averages (which range from 18 to 30 Å) are relatively small across the 3 studies and, at least as best we can tell by this necessarily crude comparison at this stage, do not obviously lead to any major changes in the structures of the polymeric assemblies.

4) The authors observe crosslinks between the IFT-A proteins (IFT122 and IFT140) and IFT-B proteins (IFT70, IFT88, and IFT172) as discussed on pg. 6 and shown in figure 5A. To accommodate these crosslinks into the structural model of the IFT train shown in Figure 5A, the authors place the IFT-B subunits IFT70 and IFT88 far apart in the IFT-B complex. However, these subunits are known to interact directly (Taschner et al. JCB 2014) and indeed sit in proximity to the IFT train structure as observed by Lacey et al. While the crosslinking data may well be correct, the incorrect structural model of IFT-A likely forces an incorrect positioning of IFT-B proteins to fulfill the crosslinking data.

It is now clearly evident that the earlier segmentation of the monomeric unit from within the polymeric IFT-A chain, which we based on the published segmentation of Jordan et al. Nat Cell Biol. 2018, did not properly capture the true boundaries of an IFT-A monomer, especially with regard to IFT140, which extends outward to connect adjacent monomers. The use of this artificially truncated monomer as a molecular envelope in our initial modeling effectively forced the IFT-A subunits to pack in a reversed orientation in order to fit the truncated density.

In order to address this issue, we omitted the cryo-ET data from the modeling altogether and instead incorporated evidence capturing domain-domain structures of interacting protein pairs from AlphaFold-Multimer. This substantially reduced the number of degrees of freedom to be explored by the integrative modeling process in order to satisfy the available structural restraints, leading in turn to significantly better convergence of independent modeling runs and high concordance with the input data (Figure 2—figure supplement 3 and Figure 2—figure supplement 4), and a significantly improved structural model of the IFT-A monomer. Docking this refined monomer structure into the (now fully independent) cryoET tomogram produced a model of the polymer that fit well into the cryo-ET density (Figure 4 and Figure 4—figure supplement 1) and agreed in large part with those derived by Hesketh and Lacey, as described above and visualized in Figure 4—figure supplement 2.

Author Resonse

Reviewer #1 (Public Review):

The manuscript by Himmel et al is an interesting study representing a topic of substantial interest to the somatosensory neurobiology community. Here, the authors use CIII peripheral neurons to investigate polymodality of sensory neurons. From vertebrates to invertebrates, this is a long-standing question in the field: how is it that the same class of sensory neurons that express receptors for myriad sensory modalities encode different behavioral responses. This system in Drosophila seems to be an intriguing system to study this question, making use of the genetic toolkit in the fly and ease of behavioral assays. In this study, the authors identify a number of channels that are important for cold nociception, and they showed that some of these do not appear to also encode mechanosensation. Despite my initial enthusiasm for this paper, halfway through, it felt as if I were reading two different papers that were loosely tied together. This lack of cohesion significantly reduced my enthusiasm for this work. Below are some of my criticisms:

We thank Reviewer #1 for their feedback. In addition to the points below, and in accordance with the reviewer’s overall criticisms, we have revised the body text to make it more cohesive. Our main goal with this revision was to better explain to the reader the shift from anoctamins to SLC12 cotransporters.

1) The first half of the paper is about a role for Anoctamins in cold nociception, but the second half switched somewhat abruptly to ncc69 and kcc. I assumed the authors would connect these genes in a genetic pathway, performing some kind of epistatic genetic interaction studies or even biochemical assays, and that this was the reason to switch the focus of the paper midway through. But this was not the case. Moreover, they performed a different constellation of experiments for the genes in the first half vs the second half of the paper (eg. Showed a role in cold nociception vs mechanosensation or showing phenotype from overexpression). This lack of cohesion made it difficult to follow the work.

We have edited the text to better explain this shift. Two notable changes are: (1) moving the phylogenetics to Figure 1, to more immediately present and demonstrate that subdued is part of the ANO1/ANO2 family of calcium-activated chloride channels; and (2) a new cartoon schematic in Figure 6 to more strongly communicate to a reader that chloride is a hypothetical mechanism of cold discrimination.

In short, previous work and our phylogenetic analyses indicate that subdued is a Cl- channel (we have moved the phylogeny earlier in the paper to make this clear from the onset). We were therefore surprised that knockdown/mutation resulted in reduced CT behavior, as neural Cl- currents are often inhibitory. Thus, we looked to known mechanisms of Cl- homeostasis to try to formulate an informed hypothesis about the function of anoctamins in this system; hence the shift in focus to SLC12.

In response to the second half of the comment: We have in fact performed cold nociception and mechanosensation experiments for both the anoctamins and the SLC12 cotransporters, although the SLC12 mechanosensation results were in a supplemental figure. We have moved the mechanaosensation results to the main Figure 6 to make this clearer. With respect to simple overexpression, the goal of the anoctamin experiments was to test the necessity of anoctamins to cold-evoked behavior, whereas the goal of the SLC12 experiments was to differentially modulate Cl- homeostasis, and this could hypothetically be accomplished by both knockdown and overexpression (hence we performed both knockdown and overexpression).

2) In Fig1B,C how does one confirm a CIII neuron is being analyzed. It might help the reader if there were at least some zoomed out photos where all the cell types are labeled and potentially compared to a schematic. Moreover, is there a CIII specific marker to use to co-stain for confirmation of neuron type?

Our CIII fusion is a specific marker for CIII neurons. To better demonstrate this, we have added images of the new CIII fusion expression patterns overlapping with a previously described CIII GAL4 driver (i.e. nompC-GAL4), and provided text describing how the CIII fusion transgene was discovered and generated. Please see the new Figure 1-Figure supplement 1.

3) As this paper is predicated on detecting differences by behavioral phenotype, the scoring analysis is not as robust as it could be, especially considering the wealth of tools in Drosophila for mapping behaviors. The "CT" phenotype is begging for a richer behavioral quantification. This critique becomes relevant here when considering the optogenetic induced CT behavior in Fig5. If the authors were to use unbiased quantitative metrics to measure behavior, they could show how similar the opto behavior is to the natural cold evoked behavior. Perhaps the two are not the same, although loosely fitting under the umbrella of "CT".

In accordance with our response above to necessary revisions, we have added one additional metric and reorganized the figures to better demonstrate the complexity of the behavior. We have no further data or new tools at this time.

To improve our optogenetic analyses, we have added data for Channelrhodopsin-dependent CIII activation, which has been previously shown to induce cold-like behaviors at high levels of activation and innocuous touch-like behaviors at low levels of activation (Turner, Armengol et al 2016). Further, we have added videos (Figure 5—videos 1-3) showing behavior in response to both Channelrhodopsin and Aurora activation.

With respect to differences in behavior, we have pointed out some differences in the Aurora-evoked behavior from the cold-evoked behavior: chloride optogenetics induces innocuous touch-like behaviors following CT. Please see lines 296-299.

4) Following on from the last comment, the touch assays in Fig3 have a different measurement system from the other figures. Perhaps touch deficits would be identified with richer behavioral quantification. Moreover, do these RNAi larvae show any responses to noxious mechanical stimulation?

The touch assays necessarily have different metrics from cold assays, as the touch-evoked behaviors are quite different from cold-evoked change in length (which are relatively simple, prima facie).

With respect to noxious mechanical stimulation, while Class III neurons have been shown to facilitate this modality and be connected to relevant circuitry (please see Hu et al 2017 https://doi.org/10.1038/nn.4580 and Takagi et al 2017 https://doi.org/10.1016/j.neuron.2017.10.030), Class IV neurons are the primary sensory neuron which initiate the noxious mechanical-induced rolling response. Although this is an interesting question, we believe it is outside the scope of this study.

Reviewer #2 (Public Review):

Himmel and colleagues study how individual sensory neurons can be tuned to detect noxious vs. gentle touch stimuli. Functional studies of Drosophila class III dendritic arborization neurons characterized roles in gentle touch and identified a receptor, NompC, and other factors that mediate these responses. Subsequent work primarily from the authors of the current study focused on roles for the same sensory neurons in cold nociception. The two proposed sensory inputs lead to quite distinct sets of behaviors, with touch leading to halting, head turning and reverse peristalsis, and noxious cold leading to whole body contraction. How activity of one type of sensory neuron could lead to such different responses remains an outstanding question, both at the levels of reception and circuitry.

The cIII responses to noxious cold and innocuous touch raises questions that the authors address here, proposing that studies of this system could advance the understanding of chronic neuropathic pain. A candidate approach inspired by studies in vertebrate nociceptors led the authors to study anoctamin/TMEM16 channels subdued, and CG15270, termed wwk by the authors. The authors focus on a pathway for gentle touch vs. cold nociception discrimination through anoctamins. Several of the experiments in this manuscript are well done, in particular, the electrophysiological recordings provide a substantial advance. However, the genetic and expression analysis has several gaps and should be strengthened. The data also do not provide strong support for some key aspects of the proposed model, namely the importance of relative levels of Cl co-transporters.

Major comments:

1) Knockout studies are accomplished using two MiMIC insertions whose effects on subdued or CG15270/wwk are not characterized by the authors. This needs to be established. The MiMIC system is also not well explained in the text for readers.

We have modified the text to better explain MiMICs (Lines 137-140) and we have verified the mutagenic effects of these MiMIC insertions via RT-PCR (Figure 2 – supplement 1). We believe these data, in conjunction with other converging lines of evidence (e.g. rescue) demonstrate necessity of these genes in cold nociception.

2) Subdued expression is inferred by a Gal4 enhancer trap. This can be a hazardous way of determining expression patterns given the uncertain relevance of the local enhancers driving the expression. According to microarray analysis subdued is strongly expressed in cIII neurons, but c240-Gal4 is barely present compared to nearby neurons, raising questions about whether this line reflects the expression pattern, including levels, even though the authors suggest that the line is previously validated (line 95; it is unclear what previously validated means). Figure 1B should not be labeled "subdued > GFP" since it is not clear that this is the case. Another more direct method of assessing expression in cIII is necessary. Confidence is higher for wwk using a T2A-Gal4 line, however, Figure 1C might be misleading to readers and indicate that wwk-T2A-Gal4 is cIII specific whereas in supplemental data the authors show how it is much more broadly expressed. The expression pattern in the supplemental figures should be moved to the main figures.

We have removed the phrase “previously validated” and we have modified Figure 1 to change how we refer to the GFP expression (removed “subdued > GFP”).

In accordance with the response to necessary revisions above, we make use of several converging lines of evidence to infer expression, including GAL4 expression patterns, microarray, and qPCR (the two latter experiments from isolated CIII samples). That subdued and wwk are expressed in CIII is clearly the most parsimonious hypothesis.

We have also carefully reviewed our body text to be certain we do not make claims of differential expression between different neural subtypes based on differences in fluorescence in the GAL4-driven GFP imaging. We do not believe that this would be a reasonable way to infer differences in expression levels in any instance.

With respect to the design of Figure 1, the intent is not to mislead the reader, and we state in the text that wwk is not solely expressed in CIII (lines 120-125). As eLife makes supplemental figures available directly alongside the main figures, we have left the relevant supplemental figures as supplements – we simply think this makes more sense from a standpoint of readability and style.

3) In figure 8 the authors propose a model in which the relative levels of K-Cl cotransporters Kcc (outward) and Ncc69 (inward) in cIII neurons determine high intracellular Cl- levels and a Cl- dependent depolarizing current in cIII neurons. They test this model using overexpression and loss of function data, but the results do not support their model since for most of the overexpression and LOF of kcc and ncc69 do not significantly affect cold nociception, the exception being ncc69 RNAi. The authors suggest that this could be due to Cl homeostasis regulated by other cotransporters. Nonetheless, it leaves a significant unexplained gap in the model that needs to be addressed.

We respectfully disagree that our results are not consistent with the stated hypothesis. In fact, it is the lack of change under certain conditions which lend evidence against the alternative hypothesis that CIII neurons maintain relatively low intracellular Cl-. The hypothesis we are testing is that ncc69 expression is driving relatively high intracellular Cl- concentrations, thus resulting in depolarizing Cl- currents.

Under this hypothesis, we would predict that knockdown of ncc69 and overexpression of kcc would reduce cold sensitivity at 5˚C. That knockdown of ncc69 and overexpression of kcc reduces cold sensitivity is consistent with this hypothesis (and we point out in text that the evidence for kcc is less convincing) – at the least, these results do not disprove it.

Under this hypothesis, we would also predict that knockdown of kcc and overexpression of ncc69 would not result in reduced cold sensitivity at 5˚C. As there was no phenotype at 5C, our results are likewise consistent with the hypothesis (at the least, they do not disprove it).

We did find it curious that ncc69 RNAi did not affect neural activity at 10˚C, but speculate that our inability to detect physiological effects for ncc69 knockdown are limitations of our electrophysiology methodology (and we discuss this in the manuscript).

The only piece of data inconsistent with the hypothesis may be that kcc overexpression may not have affected cold nociception at 5˚C – the data aren’t overwhelmingly convincing. However, this is only one experiment among many, and we believe the preponderance of evidence is consistent with the hypothesis. That is not to say we believe this hypothesis has complete explanatory power, however, as noted by our discussion of both the ncc69 electrophysiological and kcc behavioral data, and by our suggestion that there may be other regulatory mechanisms at work. This latter suggestion is wholly speculative, and we believe appropriate for the discussion section. We agree (and state in the discussion) that this would require further experimentation.

4) Related to the #3, the authors should verify the microarray data that form the basis for their differential expression model.

We have performed qPCR for ncc69 and kcc. Although qPCR is semiquantitative when comparing between genes, the Ct value for ncc69 was lower than for kcc, indicating more transcripts were present at the onset (assuming identical efficacy). These data (although semi-quantitative), the microarray, and our behavioral and electrophysiological data are consistent with the stated hypothesis.

Reviewer #3 (Public Review):

There are also several modest weaknesses in the paper:

1) A notable gap remains in the evidence for the hypothesized mechanisms that enhance electrical activity during cold stimulation and the proposed role of anoctamins (Fig. 8) - the lack of evidence for Ca2+-dependent activation of Cl- current. The recording methods used in the fillet preparation should enable direct tests of this important part of the model.

We have performed an additional experiment at the reviewer’s suggestion. Please see above (in essential revisions) and below (in recommendations for authors).

2) The behavioral and electrophysiological consequences of knocking down either of the two anoctamins are incomplete (Fig.2), raising the significant question of whether combined knock-down of both anoctamins in the CIII neurons would largely eliminate the cold-specific responses.

While the results of this experiment would certainly be interesting, we are unsure of how it would be acutely informative in this context and are not convinced that any possible outcomes would disprove any particular hypothesis. In part, this is because we know that blocking synaptic transmission in CIII neurons (via tetanus toxin) does not completely ablate cold-evoked behavior (Turner & Armengol et al 2016 https://doi.org/10.1016/j.cub.2016.09.038). This is also the case for combinatorial mutation of other genes associated with cold nociception (please see Turner & Armengol et al 2016; and more recently, Patel et al 2022 https://doi.org/10.3389/fnmol.2022.942548). Further, the husbandry required to generate the double knockdowns would be quite challenging and might result in GAL4 titration (hypothetically less strongly knocking down each gene). For these reasons, we have not performed this suggested experiment.

3) Blind procedures were not used to minimize unconscious bias in the analyses of video-recorded behavior, although some of the analyses were partially automated.

This is correct and a relative weakness of the study. We note it in our methods section. The use of semi-automated data analyses of the behavioral videos is designed to minimize experimenter-specific variability.

4) The term "hypersensitization" is confusing. Pain physiologists typically use "sensitization" when behavioral or neural responses are increased from normal. In the case of increased neuronal sensitivity, if the mechanism involves an increase in responsiveness to depolarizing inputs or an increased probability of spontaneous discharge, the term "hyperexcitability" is appropriate. Hypersensitization connotes an extreme sensitization state compared to a known normal sensitization state (which already signifies increased sensitivity). In contrast, the effects of ncc69 overexpression in this manuscript are best described simply as sensitization (increased reflexive and neuronal sensitivity to cooling) and hyperexcitability (expressed as increased spontaneous activity at room temperature).

We have modified the text in accordance with the reviewer’s suggestions (see recommendations for authors section). We have also changed the title of the paper to “Chloride-dependent mechanisms of multimodal sensory discrimination and nociceptive sensitization in Drosophila”

Author Response

Reviewer #3 (Public Review):

This paper focuses on characterizing differences between D. suzukii and D. melanogaster preferences for laying eggs on substrates of varying sugar content and stiffness. The authors demonstrate that D. suzukii show a weaker preference for multiple sugars in oviposition choice assays, that D. suzukii show a loss of sugar responsiveness in some labellar sensilla, and that some GR-encoding genes are expressed at much lower levels compared to. D. melanogaster in the legs and labellum. Intriguingly, a number of mechanosensory channel genes are upregulated In D. suzukii legs and labellum. The authors show that D. suzukii females prefer stiffer oviposition substrates compared to D. melanogaster and the balance of sweetness/texture preference differs between the two species. This is consistent with their ecological niches, with D. suzukii generally preferring to lay eggs in ripe fruit and D. melanogaster generally preferring overripe fruit.

This paper builds on previous work from this group (Dweck et al., 2021) and others (Karageorgi et al., 2017 and others) that previously demonstrated that D. suzukii prefer to lay eggs on stiffer substrates compared to D. melanogaster, will tolerate more bitter substrates and show reduced expression of several bitter GR genes. This manuscript appropriately acknowledges this work and the findings are consistent with these studies.

The manuscript is well-written, the experiments are well-controlled, the figures clearly convey the experimental findings, the data support the authors conclusions, and the statistical analysis is appropriate.

The weakest point of the paper is the lack of connection drawn between the sequencing, electrophysiological, and behavioral data. For example, the electrophysiological responses to glucose appear to be the same in both species in Figure 3 but the behavioral responses in Figure 2 are different between the two species. The authors do not provide any speculation as to what could account for this seeming discrepancy.

The revised ms. contains the following statement: " The weaker behavioral responses to glucose observed in D. suzukii could derive from weaker responses of untested taste neurons. Multiple taste organs, including the pharynx as well as the labellum and legs, contribute to oviposition behavior; sensory neurons of the ovipositor appear to play an important role as well (Yang et al., 2008; Joseph et al., 2012; Chen et al., 2022). The weaker behavioral response to glucose in D. suzukii could also arise from differences in central processing of glucose signals. It will be interesting to determine if there are differences in the connectivity of taste circuits in the two species. Alternatively, taste projection neurons in D. suzukii could have a reduced dynamic range, saturate at lower levels of receptor neuron firing, and be less able to distinguish among higher sugar concentrations."

Additionally, although Gr64d transcript is almost completely absent in D. suzukii leg RNA seq data in Figure 4B, there are no differences in the electrophysiological responses in leg sensilla in Figure 3.

This seems to imply that, although there are differences gene expression of some Grs that this does not necessarily lead to a functional difference.

We have added to the Discussion a statement to clarify that although similar, the sugar responses of leg sensilla are in fact not the same in the two species: "Leg sensilla of D. suzukii responded to sucrose, but dose-response analysis of the f5s sensillum of the leg showed that the response was lower than in its D. melanogaster counterpart to higher concentrations of sucrose (Figure 3—figure supplement 1E) ."

The authors identify mechanosensory genes that are upregulated in D. suzukii compared to D. melanogaster and suggest that these changes underlie the difference in substrate stiffness. However, it is not immediately clear that high levels of these mechanosensors would impart a new oviposition preference. Although the authors acknowledge that there are likely circuit-level differences between the two species, they do not directly test the role of any of these mechanosensors in oviposition preference in either species.

See response below to the point about nompC.

In Figure 3 there are clear differences in some of labellar responses but the leg responses look similar overall. This suggests that the labellum is playing a special role in oviposition evaluation. The paper would be strengthened by providing more insight into which tissues (labellum, legs, wings, ovipositor, etc...) are likely used to sample potential egg laying substrates.

We agree and have added to the Discussion the following: "Multiple taste organs, including the pharynx as well as the labellum and legs, contribute to oviposition behavior; sensory neurons of the ovipositor appear to play an important role as well (Yang Science 2008; Joseph Genetics 2012; Chen PNAS 2022)."

Author Response

Reviewer #3 (Public Review):

Main results:

1) TCR convergence is different from publicity: The authors look at CDR3 sequence features of convergent TCRs in the large Emerson CMV cohort. Amino usage does not perfectly correlate with codon degeneracy, for example, arginine (which has 6 codons) is less common in convergent TCRs, whereas leucine and serine are elevated. It's argued that there's more to convergence than just recombination biases, which makes sense. (I wonder if the trends for charged amino acids could be explained by the enrichment of convergent TCRs in CD8 T cells, which tend to have more acidic CDR3 loops). There's also a claim that the overlap between convergent and public TCRs is lower in tumors with a high mutational burden (TMB), but this part is sketchy: the definition of public TCRs is murky and hard to interpret, and the correlation between TMB and convergence-publicity overlap is modest (two cohorts with low TMB have higher overlap, and the other three have lower, but there is no association over those three, if anything the trend is in the other direction). It's also not clear why the overlap between COVID19 cohort convergent TCRs and public TCRs defined by the pre-2019 Emerson cohort should be high. A confounder here is the potential association between convergence and clonal expansion since expanded clonotypes can spawn apparently convergent TCRs due to sequencing errors. The paper "TCR Convergence in Individuals Treated With Immune Checkpoint Inhibition for Cancer" (Ref#5 here) gives evidence that sequencing errors may be inflating convergence in this specific dataset.

We really appreciate the reviewer’s feedback. We respond to each of the reviewer’s points below:

(1) Amino acid preference of convergent TCRs might be caused by CD8+ T cell enrichment. To test this hypothesis, we performed the same analysis using only CD8+ T cells (using the Cader 2019 lymphoma cohort). The results are shown below. We do not observe significant changes after excluding CD4+ T cells, indicating that this enrichment might be caused by factors other than CD4/CD8 differences.

(2) Definition of public TCRs. We have changed the definition of public TCRs. Instead of mixing the Emerson cohort into each group and using the mixed cohort to define the public TCRs, we just used the 666 samples of the Emerson cohort to define the same set of public TCRs and applied them to each cohort. Both the dataset and the approach used in this manuscript is consistent with a previous study on the same topic (Madi et al., 2014, elife).

(3) Convergence-publicity overlap: We agree with the reviewer that some high TMB tumors did not show further decrease of convergence-publicity overlap. One potential explanation is that the correlation between the two is not linear. By adding additional cohorts in this revision (healthy and recovered COVID-19 patients), we confirmed the previously observed overall trend between TMB and the overlap, which supported our conclusions (see figure below). On the other hand, we believe that the high overlap of convergent TCRs among healthy cohorts might result from exposure to common antigens. In the cancer patients, while still exposed, private antigens derived from tumor cells are expected to compete for resources, thus reducing the proportion of these public TCRs in the blood repertoire. The above discussion has been added to the revised manuscript:

“Healthy individuals are expected to be exposed to common pathogens, which might induce public T cell responses. On the other hand, cancer patients have more neoantigens due to the accumulative mutation, which drives their antigen-specific T cells to recognize these 'private' antigens. This reduces the proportion of public TCRs in antigen-specific TCRs. Furthermore, a higher tumor mutation burden (TMB) would indicate a higher abundance of neoantigens, resulting in a lower ratio of public TCRs.”

2) Convergent TCRs are more likely to be antigen-specific: This is nicely shown on two datasets: the large dextramer dataset from 10x genomics, and the COVID19 datasets from Adaptive biotech. But given previous work on TCR convergence, for example, the Pogorelyy ALICE paper, and many others, this is also not super-surprising.

We thank the reviewer for bringing up this related work. In the Pogorelyy ALICE paper, the authors defined TCR neighbors based on one nucleotide difference of a given CDR3, which included both synonymous and non-synonymous changes. In other words, ALICE combines both convergence and mismatched (with hamming distance 1) sequences as neighbors. Although highly relevant, our approach is different by focusing only on the convergence, as mismatch has been extensively investigated by previous studies. We have now added this paper as Ref 27, and discussed the difference between ALICE and our method in the revised manuscript.

3) Convergent T cells exhibit a CD8+ cytotoxic gene signature: This is based on a nice analysis of mouse and human single-cell datasets. One striking finding is that convergent TCRs are WAY more common in CD8+ T cells than in CD4+ T cells. It would be interesting to know how much of this could be explained by greater clonal expansion of CD8+ T cells, together with sequencing errors. A subtle point here is that some of the P values are probably inflated by the presence of expanded clonotypes: a group of cells belonging to the same expanded clonotype will tend to have similar gene expression (and therefore similar cluster membership), and will necessarily all be either convergent or not convergent collectively since they share the same TCR. So it's probably not quite right to treat them as independent for the purposes of assessing associations between gene expression clusters and convergence (or any other TCR-defined feature). You can see evidence for clonal expansion in Figure 3C, where TRAV genes are among the most enriched, suggesting that Cluster 04 may contain expanded clones.

(1) We agree with the reviewer that a possible explanation of the CD8/CD4 difference is the larger cell expansion of CD8+ T cells. We tested this hypothesis by counting the number of T cell clones instead of cell number to remove the effect that would have been caused by CD8 T cell expansion. We first investigated the bulk TCR repertoire sequencing samples as Figure 3 - figure supplement 2C-2D (see figure below). We observed higher convergence levels for the CD8+ T cell clones compared to CD4+ T cells. The additional description of this topic was added at the last paragraph of the result section of “Convergent T cells exhibit a CD8+ cytotoxic gene signature” as follows:

“The results may be explained by larger cell expansions of CD8+ T cells than CD4+ T cells. Therefore, we calculated the number of convergent clones within CD8+ T cells and CD4+ T cells from the above datasets to exclude the effects of cell expansion. As a result, in the scRNA-seq mouse data, while only 1.54% of the CD4+ clones were convergent, 3.76% of the CD8+ clones showed convergence. Likewise, 0.17% of convergent CD4+ T cell clones and 1.03% of convergent CD8+ T cell clones were found in human scRNA-seq data. In the bulk TCR-seq lymphoma data, similar results were also observed, where the gap between the convergent levels of CD4+ and CD8+ T cells narrowed but remained significant (Figure 3—figure supplement 2C-2D). In conclusion, these results suggest that CD8+ T cells show higher levels of convergence than CD4+ T cells, which substantiated our hypothesis that convergent T cells are more likely antigen-experienced. This observation has been tested using multiple datasets with diverse sequencing platforms and sequencing depth to minimize the impact of batch or other technical artifacts.”

(2) We next investigated the effect of cell expansion in the single cell analysis. We agree with the reviewer that some highly-expanded convergent clones could inflate the p-value. Therefore, we revised the calculation of TCR convergence by using the T cell clone instead of individual cells. We observed that the clusters of interest mentioned in the paper (for both mouse and human data) remain at the top convergent level among all clusters (see table below), with p values estimated using Binomial exact test. These results supported our hypothesis that TCR convergence is enriched for T cell clusters that are more likely antigen-experienced.

4) TCR convergence is associated with the clinical outcome of ICB treatment: The associations for the first analysis are described as significant in the text, and they are, but just barely (0.045 and 0.047, but you have to check the figure to see that).

As suggested by the reviewer, we have added the p-value to the test so that it is easier to see. In this revision, we adopted another definition of convergent level, changing from the ratio of convergent TCR to the actual number of convergent T cell clones within each sample. The p-values were more significant using this new indicator (0.02 and 0.00038). To avoid the effect of other variables that might be correlative with convergent levels, especially the sequencing depth, the multivariate Cox model was used for both datasets tested in the paper, correcting for TCR clonality, TCR diversity and sequencing depth (and different treatment methods for melanomas data). As a result, convergence remains significantly prognostic after adjusting for the additional variables.

5) Introduction/Discussion: Overall, the authors could do a better job citing previous work on convergence, for example, papers from Venturi on convergent recombination and the work from Mora and Walczak (ALICE, another recombination modeling). They also present the use of convergence as an ICB biomarker as a novel finding, but Ref 5 introduces this concept and validates it in another cohort. Ref 5 also has a careful analysis of the link between sequencing errors and convergence, which could have been more carefully considered here.

We thank the reviewer for this excellent suggestion. We have added the citation of Venturi on convergent recombination as Ref 43 and we cited it at the last paragraph of the result selection:

“Convergent recombination was claimed to be the mechanistic basis for public TCR response in many previous studies(Quigley et al., 2010; Venturi et al., 2006).”

We also included work from Mora and Walczak in the fourth paragraph of the introduction and the third paragraph of the discussion as Ref 27 to introduce this TCR similarity-based clustering method as well as its application in predicting ICB response:

“This idea has led several TCR similarity-based clustering algorithms, such as ALICE (Pogorelyy et al., 2019), TCRdist (Dash et al., 2017), GLIPH2 (Huang et al., 2020), iSMART (Zhang et al., 2020), and GIANA (Zhang et al., 2021), to be developed for studying antigen-driven T cell expansion during viral infection or tumorigenesis.”

“In addition, the potential prognostic value of TCR convergence and TCR similarity-based clustering was testified in other studies(Looney et al., 2019; Pogorelyy et al., 2019).”

Ref 5 was recited while discussing the effect of sequencing error on TCR convergence in the fourth paragraph of discussion:

“Improper handling of sequencing errors may result in the overestimation of TCR convergence (Looney et al., 2019).”

Author Response

Reviewer #1 (Public Review):

This manuscript by Borsatto et al describes atomic-level structural details of the central core domain of non-structural protein 1 (Nsp1) of SARS-CoV-2, the virus responsible for the ongoing COVID-19 pandemic. Authors combined X-ray crystallography, fragment screening, computational modelling, and molecular dynamics simulation approaches to characterize potentially druggable pockets in Nsp1 core (aa 10-126). This study presents several notable strengths. For example, authors screened and tested 60 fragments from the Maybridge Ro3 library and solved a co-crystal structure of Nsp1 core with one such fragment 2E10 (N-(2,3dihydro-1H-inden-5-yl) acetamide) to 1.1Å resolution. The molecular dynamics simulation and other computational experiments were performed rigorously.

Nsp1 blocks the path of mRNA in ribosomes to modulate protein synthesis in the host cell. Nsp1 also binds the first stem-loop (SL1) of SARS-CoV-2 mRNA. The authors used a molecular docking program (HADDOCK) to build models of the Nsp1/RNA complex and predicted two modes of Nsp1 binding to SL1 RNA. A comparative structural analysis of Nsp1/2E10 experimental structure with Nsp1/SL1 (model) reveals that small molecule compounds occupying this site may block RNA binding of Nsp1. Given the established role of this interface in modulating the host and viral gene expression programs, this finding provides an important framework for designing the small molecules capable of completely blocking this interface.

A weakness of this study is the lack of experimental validation of the two modes of Nsp1 binding to SL1 RNA.

The mechanism of binding, in particular whether Nsp1 binds to the ribosome first and then to the SL1 or the other way round, is still debated. Moreover, to the best of our knowledge, to this day there is no structure of the N-terminal region Nsp1 bound to the ribosome. Thus, we expect that obtaining a structure of the binary and possibly ternary complex to validate the predicted binding mode will necessitate considerable time and efforts and will hopefully be the focus of a follow up study.

Reviewer #2 (Public Review):

In this manuscript, the authors have identified cryptic pockets in the Nsp1 protein of the SARSCoV-2 virus. The authors used computational methods to identify these pockets and demonstrate drug binding via simulation studies. The authors also show that such cryptic pockets exist in other beta-coronaviruses as well.

The authors carried out fragment-based screening using macromolecular crystallography and confirmed the presence of drug bound in one of the pockets identified. However, the binding assays showed a weak binding with high error.

The weak binding is typical for fragments, however we agree that the error was high, therefore, we re-measured the data (both for Nsp1N and full-length Nsp1) to bring the error down. The new values can be found in Figure 6 – Figure supplement 2.

Further, the authors perform Nsp1-mRNA simulation studies to identify how Nsp1 binds to the 5'UTR of SARS-CoV-2 mRNA and mention that targeting the identified pocket in Nsp1-N could disrupt the SARS-CoV-2 Nsp1-mRNA complex. However, there are conflicting reports on direct binding between the SARS-CoV-2 Nsp1-mRNA (See references 17 & 29).<br /> Nsp1 helps establish viral infection in the host, and hence identifying the druggable site in this protein is important. Therefore, this study is important and exciting.

Author Response

Reviewer #1 (Public Review):

Yanis Zekri et al have addressed an important question of the possible role of thyroid hormone (T3) and its nuclear receptor (TR) on local BAT thermogenesis and energy expenditure. In this well-written manuscript and well-carried work, the authors address the above question by A) by generating the BATKO mice by selectively eliminating TR signaling in BAT by knocking-in a TRα1L400R, a dominant negative version of the TRa1 receptor, and by floxing the ThRb gene. They characterized this mouse thoroughly to show that they totally lacked T3 responsiveness. Using qPCR they evaluated the selective abrogation of Thrb and Hr expression in BAT tissue relative to other tissue sites. B) Using time-course transcriptome analysis they then go on to enlist all the T3/TR direct target genes using well-defined criteria and further linking with their ChipSeq data they identified 639 putative target genes which are under the direct control of T3/TR signaling. Interestingly their gene analysis lead them to some target genes directly involved with UCP1 and PGC1α in addition to genes of many other metabolic processes related to BAT thermogenesis. The experiments on denervated BAT on wild-type PTU-fed was a rather neat experiment to eliminate the influence of noradrenergic terminal BAT target genes. Furthermore, the cold exposure experiments and the high-fat diets feeding with the series of complex analyses led them to the conclusion that BAT KO animals suffered from reduced efficiency of BAT adaptive thermogenesis. By comparing the BAT transcriptome of BATKO and CTRL mice after 24h at 4{degree sign}C, the authors further go on to show how BAT TR signaling controls other subsets of genes, especially a wide variety of metabolic regulations, especially lipolysis/fatty acid oxidation. Finally, EdU injection experiments showed a direct effect of T3 on BAT proliferation.

I think it was well thought and well-designed study for understanding the complex action of cell-autonomous T3 regulation of adaptive thermogenesis. The conclusions of this paper are well supported by the data provided.

Thank you very much for this very pertinent and kind summary of our work.

Reviewer #2 (Public Review):

The authors designed this study to identify the direct T3 target genes that underlie the T3 actions in the brown adipose tissue (BAT). The unique model used (dominant negative TRa knock-in and a TRb knock-out) allows for the isolation of BAT-specific actions from other well-known systemic effects on thermogenesis, including the central nervous system. The strengths of the study reside in the novel methodological approach. Previous studies of T3 actions in the BAT used animal models that did not allow for full isolation of BAT-specific effects of T3. A limitation however is the combination of TRa knock-in (which causes permanent suppression of TRa-dependent genes) with the TRb knockout, which only prevents T3 induction of TRb-dependent genes. Nonetheless, the results were impressive with the identification of about 1,500 genes differentially expressed in the BAT, among which UCP1 and PGC1a were the two main ones. Although it has been known that both UCP1 and PGC1a are downstream targets of T3, the work establishes through an ingenious approach the critical direct role played by T3 in BAT thermogenesis. In addition, the genetic approach utilized here is of great value and could be easily expanded to other tissues and systems.

Thank you for this very pertinent summary of our work. We just want to clarify one point: we do not believe that Ucp1 is, quantitatively, one of the main genes regulated by T3 in the BAT. First, it does not belong to the set of the most induced genes after T3/T4 injection in mice. Most importantly, Ucp1 expression was not altered in BATKO mice exposed at 4°C according to unbiased RNAseq analysis. Only targeted qRT-PCR analysis could evidence a modest change. We do not call into question the crucial role of Ucp1 in BAT thermogenesis. However, we think that our approach put into perspective the relevance of Ucp1 in the T3-dependent control of BAT thermogenesis, suggesting that other mechanisms might be more directly linked to T3 activity.

Reviewer #3 (Public Review):

This paper details the importance of thyroid hormone signaling in BAT in response to environmental and nutritional stress. The authors utilize a novel genetic model to specifically target BAT and impair thyroid hormone signaling. The physiologic insight is of great interest. The role of the sympathetic nervous system in the BAT response is not fully addressed but it appears that cell-autonomous signaling mediates TH signaling in response to hyperthyroidism. The link cistromically between the TR and PGC1 is also novel and of interest.

Thank you very much for your kind comments that are highly appreciated.

Author Response

Reviewer #1 (Public Review):

This work addresses a long-standing question about how tolerance develops at the presynaptic level. That the number of receptors is unchanged following the treatment of animals with opioids was known since the early work using receptor binding assays. The conclusion was that receptor/effector coupling was disrupted was thought to be the primary mechanism underlying tolerance. This work indicates that the location of receptors is critically important in the development of tolerance. This work is groundbreaking and a game changer in the understanding of tolerance at the cellular level.

We appreciate that the Reviewer is positive about the potential impact of our study.

Reviewer #2 (Public Review):

Jullie et al addressed the long-standing question of how presynaptic desensitization of opioid receptor signaling can occur on the timescale of hours despite the fact that it does not occur on the timescale of minutes. They also compared the mu and delta opioid receptors in this context and asked whether their desensitization occurs in a homologous or heterologous manner when co-expressed in the same neurons.

A major strength of the work is the use of a relatively high-volume imaging assay of synaptic transmission based on VAMP2-SEP to detect exocytosis of synaptic vesicles and its modulation by heterologously expressed opioid receptors in cultured neurons. This allowed for large data sets to be acquired and analyzed with good statistical power. It also reports on a validated metric of synaptic transmission.

A significant weakness arises from the need to overexpress opioid receptors in cultured striatal neurons in order to conduct the experiments with high reliability. Because the authors did not attempt to address receptor expression levels and relate overexpression to endogenous receptor expression levels in axons, the physiological significance of the findings remains, to some extent, in doubt.

Using heterologously expressed receptors, the primary finding that slow desensitization (of presynaptic suppression of neurotransmission) occurs via endocytosis of membrane-localized opioid receptors, is well supported by multiple lines of evidence. 1) Blocking receptor endocytosis, either via mutation of GRK2/3 phosphorylation sites or pharmacological block with compound 101 prevents slow desensitization of MOR. ) SEP-MOR and SEP-DOR fluorescence (indicative of plasma membrane localization) is reduced by chronic agonist treatment.

The secondary findings that MOR and DOR do not desensitize or undergo endocytosis in a heterologous manner, and that DOR-depletion from the plasma membrane is more facile than MOR and independent of C-terminus phosphorylation, are well supported by the data and analyses.

Despite the reliance on heterologously expressed opioid receptors, the findings are likely to have a high impact on the fields of GPCR trafficking and opioid signaling, as they address a major outstanding question with direct relevance to opioid drug tolerance and may generalize to other GPCRs.

The findings also evoke new questions that will spur further work in the field. For example, just focusing on DOR, by what mechanism does agonist-driven DOR endocytosis occur not via GRK2/3 phosphorylation? By extension, would G protein-biased DOR agonists be expected to produce less tolerance? To be clear these are not to be addressed in this manuscript.

We appreciate that this Reviewer found that the current manuscript addresses long standing questions in the field and that our results are well supported by the data, acknowledging the strength of the presented method. We agree that our methods and results do have some associated limitations, particularly with respect to linking the present mechanistic findings to true physiology, and that the question of receptor expression level is pertinent to this link. We have attempted to address this to the best of our ability in the revised manuscript, as summarized below. We agree with the Reviewer that there remain many interesting questions for further study, and have modified the Discussion to more clearly point this out.

Reviewer #3 (Public Review):

The studies in the manuscript "Endocytic trafficking determines cellular tolerance of presynaptic opioid signaling" use a novel approach to assess the signaling of presynaptic opioid receptors that inhibit the release of neurotransmitters. Historically, studies have used whole-cell patch-clamp electrophysiology studies of spontaneous and evoked neurotransmitter release to measure the presynaptic effects of opioid receptors. Since the recordings were made in postsynaptic cells that expressed receptors for the released neurotransmitter, the electrophysiological measurements are indirect with respect to the presynaptic receptors under study. The technique used in this manuscript is based on a pHlorin-based unquenching assay that is a measure of synaptic vesicle exocytosis. In this case, the super-ecliptic pHluorin (SEP) is a pH-sensitive GFP that increases fluorescence as the synaptic vesicle protein that it is attached to (VAMP2-SEP) relocates from the acidic synaptic vesicle to the plasma membrane. Opioid agonists inhibit this activity with acute administration and this inhibition is reduced with prolonged, or chronic administration (hours), demonstrating tolerance. The SEP protein can also be conjugated to opioid receptors and used to measure the proportion of receptors on the plasma membrane compared to internalized receptors. The studies show that agonist activation of mu-opioid receptors (MORs) induces endocytosis that is dependent on phosphorylation of the C-terminus and that the development of tolerance is correlated with the loss of MORs at the surface. The results are different for the delta-opioid receptor (DOR) which is also internalized with acute agonist administration but that loss of receptors on the membrane occurs more rapidly and is not dependent on phosphorylation of the C-terminus.

The results in the studies are clearly presented and clearly substantiate the prior work using electrophysiology to show the late development of tolerance of presynaptic opioid receptor signaling. The studies extend prior work by showing that endocytosis of both MOR and DOR occurs in presynaptic locations but that the cellular mechanisms underlying the maintenance of these receptors on the plasma membrane differ. The imaging results show convincing effect sizes, even with genetic and pharmacological manipulations, that will allow for even further investigation into the cellular mechanisms underlying the development of tolerance. Since these studies transfected the cultured striatal neurons with both the opioid receptors and the VAMP2-SEP, one question that remains is whether imaging of the VAMP2-SEP has the resolution to detect inhibition of endocytosis by endogenous opioid receptors. Since the authors make the point that this technique has advantages over traditional electrophysiological approaches, it is important that the technique allows for the measurement of endogenous levels of receptors. There are minor questions about the statistics used in some of the graphs, and the utility of the presentation of p values on the right-hand axis but these concerns do not alter the overall significance of the studies, which are high impact.

We are pleased that this Reviewer found our results generally convincing and impactful. We are grateful for the critical comments and suggestions, particularly with regard to improving the statistical analysis and simplifying / removing speculation from our model. We have done our best to address both important aspects in the revised manuscript, as detailed below.

Author Response

Reviewer #3 (Public Review):

1) This work focuses exclusively on excitatory input. However, as the authors mention, LGMD neurons also receive inhibitory inputs, and these inputs also appear to segregate to different areas of the dendritic tree depending on the pathway. The contribution of inhibition is mostly ignored throughout the manuscript, but I think that it would be beneficial to discuss how inhibitory inputs fit into the story. For example, if OFF inhibition maps onto the C field, then presumably when there is mixed ON/OFF stimulation there is inhibition of the ON excitation onto the C field? If so, how much excitation of the C field is left? How much does the retainment of spatial coherence sensitivity with mixed stimuli arise from the fact that OFF excitation might dominate because it inhibits the C field? I don't think that additional experiments are needed, but a discussion would be useful. Related, does the model include inhibitory synapses?

We have not elaborated more specifically on inhibition, as the experimental characterization of its interaction with excitation has not yet been investigated experimentally. We agree that the interaction between excitation and inhibition for mixed ON/OFF stimuli in field C is an interesting topic, but it is unlikely to affect substantially responses to ON stimuli alone. We added a paragraph on E-I integration to the discussion (lines 461-473). The model does include inhibitory synapses which are now more clearly described.

2) The argument that the cellular organization found here is good because it allows grasshoppers to be sensitive to white approaching stimuli while disregarding spatial coherence and saving energy seems plausible. But it's not clear to me why this is 'optimal' (from the title - 'optimizes neuronal computation'). What exactly is being optimized here? And why is it good that grasshoppers can't discriminate the spatial coherence of ON looming stimuli? Is everything that approaches a grasshopper fast and white always a bad thing, but not the case if the approaching thing is black? Some further placement of these findings into an ecological setting might be helpful here.

Our thinking is not that there is an advantage to responding to incoherent white looms (on the contrary), but that white looming stimuli in nature are likely less frequent than black/white mixtures or than all dark stimuli. Thus, the inability to discriminate white spatial coherence might have been sacrificed to decrease energy expenditure. We agree that ‘optimal’ might be too strong a wording and we have modified the title and text accordingly. Hopefully the text is now clearer on this point.

Author Response

Reviewer #1 (Public Review):

Abdel-Hag, Reem et al. investigated the beneficial effects of a fiber-rich diet in the pathology of α-synuclein overexpressing (ASO) mice, a preclinical model of Parkinson's disease. They found that a prebiotic intervention attenuates motor deficits and reduces microglial reactivity in the substantia nigra and striatum. They extended these findings by doing scRNA sequencing, and they identified the expansion of a protective disease-associated microglia (DAM), a microglial subset previously described during the early stages of disease in several mouse models. Interestingly, the data indicate that microglia do not influence the behavior of ASO mice in the early stages of disease progression. However, microglia are the key mediators of the protective effects of prebiotic treatment in ASO mice. Overall, the conclusions of this paper are well supported by data, but some aspects should be considered to improve the manuscript.

1) Colony-stimulating factor 1 receptor (CSF1R) inhibition has been widely used as a method for microglia depletion, however, the impact of this approach on peripheral immune cells is controversial. The authors elegantly showed that most gut-associated immune cell populations were unaffected by PLX5622. However, CSF1R signaling has been implicated in the maintenance of gut homeostasis. Could it be possible that PLX5622 treatment affects directly the gut microbiome composition? Are the beneficial changes in the gut microbiome composition of a prebiotic diet still maintained in combination with PLX5622? CSF1R inhibitors with low brain penetration such as PLX73086 and therefore unable to deplete resident microglia (Bellver- Landete, Victor et al., Nat Commun, 2019) would be helpful to rule out peripheral off-target effects.

We agree that loss of benefits by the prebiotic diet following PLX5622 treatment is possibly due to changes to the microbiome, and cannot exclude this possibility. The mechanism of action of PLX5662 in reshaping the microbiome would most likely involve effects through changes in immune (or other) cell types in the gut, as the drug is not known to have direct effects on the microbiome. As described by the referee, we carefully profiled the mucosal immune system of mice treated with PLX5622 and control chow, and show minor changes associated with the drug. These are control experiments that very few previous studies using PLX5622 have performed, and suggest immune-mediated microbiome changes may be subtle. Further, we do not suggest in the manuscript that microbiome changes, in the first place, mediated the benefits of the prebiotic diet but rather focus the current study on the well-known effect of microglia depletion by PLX5622.

Microbiome profiling and additional experiments transferring microbiota from diet-treated animals, with and without PLX5622, to naïve mice would be needed to determine the functional effects of gut bacteria on microglial activation and motor symptoms. The use of PLX73086 is also an excellent way to address this point, as are several additional approaches. Comprehensively investigating the indirect contributions of the microbiome to motor symptoms in ASO mice represent a separate series of studies, in our respectful opinion. Nonetheless, this is an important caveat of our work and we now include the following text in the Discussion section to address this point: “Our study does not rule out indirect effects of PLX5622 that include reshaping the microbiome to promote motor symptoms in prebiotic diet-fed mice”. We thank the referee for this comment.

2) The authors claimed that microglial depletion eliminates the protective effects of the prebiotic diet in ASO mice by showing increased levels of aggregated aSyn in the SN (Fig 5G). However, microglial depletion also has the same effect on WT mice. How do authors interpret this result?

The referee raises an astute point. Microglia appear to play a complex role in PD and mouse models, with both positive and negative effects demonstrated in various context (for example, PMIDSs: 29401614, 32086763). A primary and non-exclusive function of microglia is the removal of -synuclein accumulations (PMIDs: 32170061, 34555357). Importantly, there is no change in motor behavior in prebiotic-fed WT mice with or without PLX5622 treatment, as expected (see Figs. 5D-F). We have been careful in the manuscript to not suggest that microglia effects on motor symptoms are via a process that include -synuclein aggregates, as this has not been convincingly shown in this mouse model at the time point we are studying (ie., 22 weeks of age). While it would be straightforward to add a statement suggesting why -synuclein levels increase in WT mice on drug, our preferred remedy here is to point out this observation so it does not go unnoticed, but refrain from speculation in the absence of data since this is not a major point of the study. We have now inserted the statement “However, in prebiotic-fed WT and ASO mice, depletion of microglia significantly increased levels of aggregated αSyn in the SN, while levels in the STR remained unchanged (Figure 5G-H).” We thank the referee for this important comment.

3) What is the rationale for doing a long-term (17 weeks) prebiotic intervention? Have the authors considered doing a short-term intervention? The prebiotic diet should change quickly the gut microbiome composition within a few days or weeks.

We have previously shown that long-term microbiome depletion is required to impact motor performance in ASO mice (similar timeline as current prebiotic study) (Sampson et al., Cell, 2016). In unpublished data, short-term antibiotic treatment (4 weeks before motor testing) is unable to improve motor symptoms in ASO mice. Thus, we chose a timeframe for the current prebiotic studies guided by empiric data, but further details on dose intervals remain unknown. We agree that the microbiome should rapidly respond to the prebiotic diet, but it is unknown if this response is durable or would the ‘pre-treated’ microbiome profile re-establish at some time after removal of the experimental diet. We respectfully suggest that these more specialized studies are better suited for future projects.

Author Response

Reviewer #1 (Public Review):

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

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

The channel capacity of a cell signaling network depends critically on the distributions of the downstream signaling molecules in question: see the original paper by Cheong et al. (2011, Science 334 (6054), 354-8) and subsequent papers (notably Selimkhanov et al. (2014) Science 346 (6215), 1370-3 and Suderman et al. (2018) Interface Focus 8 (6), 20180039). The U937 cells considered here clearly don't serve the physiological function of detecting the glycans considered by the authors; despite the fact that this is an artificial cell line, the fact the authors have to exogenously express the relevant receptors indicates that these cells are not necessarily a good model for the types of cells in the body that actually have evolved to sense these glycan molecules.

Signaling molecules readily exhibit cell-type-specific expression levels that influence cellular responses to external stimuli (Rowland et al.(2017) Nat Commun 8, 16009). So it is unclear that the distributions of downstream signaling molecules in U937 cells mirror those that would be observed in the immune cell types relevant to this response. As such, the physiological relevance of the differences between dectin-2 channel capacities and those exhibited by the other receptors are currently unclear.

We appreciate Reviewer #1’s in-depth comments related to physiological relevance of the U937 cell. A big benefit of using information theory to investigate a biological communication channel is the realization of quantitative measurement of information that the channel transmits without having detailed measurement of spatiotemporal dynamics of receptors and downstream signaling cascades. In addition, the quantity of measured information itself in turn gives us a decent prediction about detailed signaling mechanisms by comparing the information quantity difference. For example, we investigated how transmission of glycan information from dectin-2 is synergistically modulated in the presence of either dectin-1, DC-SIGN or mincle. Our approach allows to investigate how individual lectins on immune cells contribute to glycan information transmission and be integrated in the presence other type of lectins. Therefore, the findings describe how physiologically relevant lectins are integrating the extracellular signal in a more defined way. Furthermore, we found that our model cell line has one order of magnitude higher expression of dectin-2 compared with primary human monocytes and exhibits a similar zymosan binding pattern (will be described in Recommendations for the authors and Figure R8).

We fully agree that acquiring more information on the information transmission capability of primary immune cells would increase physiological relevance. In the revised manuscript we addressed this concern by comparing the receptor expression levels of our model cell lines with primary monocytes, for which we find an agreement of cellular heterogeneity. However, we would also like to point out that the very basic nature of our question, of how information stored in glycans is processed by lectins, is not tightly bound to these difference of primary cells and cell lines.

Line 382: Finally, it is important to take into consideration that our conclusions came from model cell lines, which were used as a surrogate for cell-type-specific lectin expression patterns of primary immune cells. Human monocytes and dectin-2 positive U937 cells have comparable receptor densities and respond similar to stimulation with zymosan particles (SI Fig. 6A and B).

2) Another issue that readers might want to keep in mind is that the details of the channel capacity calculation are a bit unclear as the manuscript is currently written. The authors indicate that their channel capacity calculations follow the approach of Cheong et al. (2011) Science 334 (6054), 354-8. However, the extent to which they follow that previous approach is not obvious. For instance, the calculations presented in the 2011 work use a combined bootstrapping/linear extrapolation approach to estimate the mutual information at infinite population size in order to deal with known inaccuracies in the calculation that arise from finite-size effects. The Cheong approach also deals with the question of how many bins to use in order to estimate the joint probability distribution across signal and response.

They do this by comparing the mutual information they calculate for the real data with that calculated for random data to ensure that they are not calculating spuriously high mutual information based on having too many bins. While the Cheong et al. paper does a great job explaining why these steps need to be undertaken, a subsequent paper by Suderman et al. (2017, PNAS 114 (22), 5755-60) explains the approach in even greater detail in the supporting information. Those authors also implemented several improvements to the general approach, including a bootstrap method for more accurately estimating the error in the mutual information and channel capacity estimates.

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

We thank Reviewer #1’s recommendation for increasing the legitimacy of the calculation. In the revised manuscript we tried to explain channel capacity calculation procedures in more detail with statistical approaches that adopted from Cheong et al. (2011) and Suderman et al. (2018) (SI section 1 and 2). Furthermore, we decide the number of binning from not only random dataset but also the number of total samples as shown below:

Figure R1. A) Extrapolated channel capacity values of random dataset at infinitely subsampled distribution under various total number of samples and output binning. The white line in the heatmap represents the channel capacity value at 0.01 bit. B) Extrapolated channel capacity values at infinite subsample size of U937 cells’ input (TNF-a doses) and output (GFP reporter) response.

Figure R1 describes channel capacity values from random (A) and experimental dataset (B, TNFAR + TNF-a). The channel capacity values from random data indicates the dependence of channel capacity on the number of the output binning and total number sample. According to this heatmap, we decided the allowed bias as 0.01 bits as shown in contour line shown in Figure R1A. Since our minimum dataset that used for channel capacity calculation in the absence of labelled input is near 90,000, the expected bias in channel capacity calculation is therefore less than 0.01 bits in binning range from 10 to 1000 as shown in Figure R1A.

Furthermore, we demonstrated mutual information maximization procedure using predefined unibimodal input distribution and compared with the systematic method that we used in the work. We found that there is no noticeable difference in channel capacity value between two approaches (SI Figure 3M).

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

We understand Reviewer #1’s concern regarding the choice of log binning. We found that if the number of binning is higher than 200, no matter the binning methods, including linear, logarithmic or equal frequency, the estimated channel capacities in each binning number are converged into the same value. The only difference is how quickly the values approach the converged channel capacity as increasing the binning number (shown in Figure R2). In the revised manuscript, we used linear binning to represent more relevant protein signaling as the Reviewer mentioned. Note that the channel capacity values calculated from linear binning do not show noticeable different from our previously calculated channel capacity values.

On the other hand, linear binning generates significant bias, if we consider labelled input (i.e., continuous input) into channel capacity calculation, due to the increase of binning in input region.

Figure R2. Output binning number and binning method dependence of channel capacity value for experimental dataset. The inset plots show the relative difference of channel capacity value to the maximum channel capacity value in the entire binning range (i.e., from 10 to 1000) of the corresponding binning method.

According to Reviewer #1’s comment we have changed the binning method from logarithmic binning to linear binning in the whole experimental dataset except in the presence of labelled input (i.e., dectin-2 antibody). If we consider channel capacity between labelled input and NF-kB reporter, equal frequency binning is used for every layer of the channel capacity (i.e., labelled input-binding, binding-GFP, labelled input-GFP)

Reviewer #2 (Public Review):

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

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

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

We understand reviewer #2’s comment related with rigorousness and correctness of the theoretical results of this work. In the revised manuscript, we added following contents to help the reader to better understand the channel capacity calculation procedures.

• General illustrative introduction regarding how we measured input and output dataset and how we handle those data to prepare joint probability distribution shown in SI section 1.1 and 1.2.

• Exemplified mutual information maximization procedure using experimental and arbitrary dataset shown in SI section 1.3.

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

We thank Reviewer #2’s comment regarding the arbitrariness of the weightings. We realize there was a lack of explanation on the weighting values in the original manuscript. 𝑃x(𝑖) is a marginal probability distribution of input from the original dataset and 𝑃x'(𝑖) is the marginal probability distribution of modified input that maximize the mutual information. In usual case 𝑃x(𝑖) is not equal to 𝑃x'(𝑖) and therefore one needs to find 𝑃x'(𝑖) from 𝑃x(𝑖). Because 𝑃x'(𝑖) is a linear combination of 𝑃x(𝑖), it can be expressed as 𝑤(𝑖)𝑃x(𝑖) , where 𝑤(𝑖) is the weightings, under constraint ∑input/i 𝑤(𝑖)𝑃x (𝑖) = 1 . The changed input distribution, in turn, modifies the joint probability distribution as 𝑃'xy (𝑖, 𝑗) = 𝑤(𝑖)𝑃xy)(𝑖, 𝑗). To help readers understand of this work we expanded the Appendix with illustrative descriptions.

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

We agree with Reviewer’s suggestion for better description of input and output in the manuscript. Therefore, we have modified Figure 1 A and B and the main text to describe the source of input and output much clearly, as follows:

Line 92: Accounting for the stochastic behavior of cellular signaling, information theory provides robust and quantitative tools to analyze complex communication channels. A fundamental metric of information theory is entropy, which determines the amount of disorder or uncertainty of variables. In this respect, cellular signaling pathways having high variability of the initiating input signals (e.g. stimulants) and the corresponding highly variable output response (i.e. cellular signaling) can be characterized as a high entropy. Importantly, input and output can have mutual dependence and therefore knowing the input distribution can partly provide the information of output distribution. If noise is present in the communication channel, input and output have reduced mutual dependence. This mutual dependence between input and output is called mutual information. Mutual information is, therefore, a function of input distribution and the upper bound of mutual information is called channel capacity (SI section 1) (Cover and Thomas, 2012). In this report, a communication channel describes signal transduction pathway of C-type lectin receptor, which ultimately lead to NF-κB translocation and finally GFP expression in the reporter model (Fig. 1A). To quantify the signaling information of the communication channels, we used channel capacity. Importantly, the channel capacity isn’t merely describing the resulting maximum intensity of the reporter cells. The channel capacity takes cellular variation and activation across a whole range of incoming stimulus of single cell resolved data into account and quantifies all of that data into a single number.

Author Response

Reviewer #1 (Public Review):

The software presented in this paper is well documented and represents a significant achievement that breaks new ground in terms of what is possible to render and explore in the web browser. This tool is essential for the exploration of SC2 data, but equally useful for the tree of life and other tree-like data sets.

Thank you for reviewing my work and for this generous assessment.

Reviewer #2 (Public Review):

This manuscript describes a web-based tool (Taxonium) for interactively visualizing large trees that can be annotated with metadata. Having worked on similar problems in the analysis and visualization of enormous SARS-CoV-2 data sets, I am quite impressed with the performance and "look and feel" of the Taxonium-powered cov2tree web interface, particularly its speed at rendering trees (or at least a subgraph of the tree).

Thank you for the kind words.

The manuscript is written well, although it uses some technical "web 2.0" terminology that may not be accessible to a general scientific readership, e.g., "protobuf" (presumably protocol buffer) and "autoscaling Kubernetes cluster". The latter is like referring to a piece of lab equipment, so the author should provide some sort of reference to the manufacturer, i.e., https://kubernetes.io/.

Thank you for flagging this. I have now replaced the colloquial "protobuf" with "protocol buffer". I have now provided a URL for Kubernetes. It is always difficult to judge how much to explain technical terms. I certainly agree that many people will be unfamiliar with, for instance, protocol buffers, but an explanation of what they are (which may not be particularly important for understanding Taxonium) can sometimes overshadow more important details. So my preference in that particular case is for an interested reader to research the unfamiliar term.

In other respects, the manuscript lacks some methodological details, such as exactly how the tree is "sparsified" to reduce the number of branches being displayed for a given range of coordinates.

This is an important point also raised by Reviewer 3. I have added a new section in the Materials and Methods which discusses this in some detail.

Some statements are inaccurate or not supported by current knowledge in the field. For instance, it is not true that the phylogeny "closely approximates" the transmission tree for RNA viruses.

I agree that this was an overly broad claim, and have softened it, now saying:

"The fundamental representation of a viral epidemic for genomic epidemiology is a phylogenetic tree, which approximates the transmission tree and can allow insights into the direction of migration of viral lineages."

Mutations are not associated with a "point in the phylogeny", but rather the branch that is associated with that internal node.

I have changed this as suggested.

A major limitation of displaying a single phylogenetic tree (albeit an enormous one) is that the uncertainty in reconstructing specific branches is not readily communicated to the user. This problem is exacerbated for large trees where the number of observations far exceeds the amount of data (alignment length). Hence, it would be very helpful to have some means of annotating the tree display with levels of uncertainty, e.g., "we actually have no idea if this is the correct subtree". DensiTree endeavours to do this by drawing a joint representation of a posterior sample of trees, but it would be onerous to map a user interface to this display. I'm raising this point as something for the developers to consider as a feature addition, and not a required revision for this manuscript.

I entirely agree with this point. I have added a sentence in the discussion:

"Even where sequences are accurate, phylogenetic topology is often uncertain, and finding ways to communicate this at scale, building on prior work [Densitree citation] would be valuable."

The authors make multiple claims of novelty - e.g., "[...] existing web-based tools [...] do not scale to the size of data sets now available for SARS-CoV-2" and "Taxonium is the only tool that readily displays the number of independent times a given mutation has occurred [...]" - that are not entirely accurate. For example, RASCL (https://observablehq.com/@aglucaci/rascl) allows users to annotate phylogenies to examine the repeated occurrence of specific mutations. Our own system, CoVizu, also enables users to visualize and explore the evolutionary relationships among millions of SARS-CoV-2 genomes, although it takes a very different approach from Taxonium. Taxonium is an excellent and innovative tool, and it should not be necessary to claim priority.

I agree that comparisons with existing tools are difficult and often provide a sense of unnecessary competition. I attempted to be quite careful in the specific section focused on comparison, but may have been less careful earlier on. The intent with this first sentence in the abstract was to provide a succinct description of the gap that Taxonium was developed to fill with "however, existing web-based tools for analysing and exploring phylogenies do not scale to the size of datasets now available for SARS-CoV-2". I have now removed the words "analysing and", focusing on the exploration of phylogenies. I think this new sentence is defensible in that valuable tools such as CoVizu intentionally do not explore a phylogeny directly but instead take a multi-level approach, and this new sentence better matches the comparisons in the paper. In the second sentence, I have removed the phrase "is the only tool that", which I agree adds little and may not be accurate, depending on one's interpretation of "readily". Thank you for these points.

Although the source code (largely JavaScript with some Python) is quite clean and has a consistent style, there is a surprising lack of documentation in the code. This makes me concerned about whether Taxonium can be a maintainable and extensible open-source project since this complex system has been almost entirely written by a single developer. For example, usher_to_taxonium.py has a single inline comment (a command-line example) and no docstring for the main function. JBrowsePanel.jsx has a single inline comment for 293 lines of code. There is some external documentation (e.g., DEVELOPMENT.md) that provides instructions for installing a development build, but it would be very helpful to extend this documentation to describe the relationships among the different files and their specific roles. Again, this is something for the developers to consider for future work and not the current manuscript.

This is an entirely fair comment. The version of Taxonium presented in the manuscript is "2.0", which is a new version built from scratch with considerably less technical debt than the version that preceded it. Its technical strengths are that (with the exception of the backend) it is relatively well-modularised into functional components. But the limitations that the reviewer notes with respect to commenting are entirely fair. What I would say is that in the time since this manuscript was submitted, several important features have been added by an external collaborator, Alex Kramer, most notably the Treenome Browser (https://www.biorxiv.org/content/10.1101/2022.09.28.509985v1). I hope that the ability of Alex to add these features with little need for support provides some evidence of Taxonium's extensibility. But I acknowledge there is room for improvement.

Reviewer #3 (Public Review):

The paper succinctly provides an overview of the current approaches to generating and displaying super-large phylogenies (>10,000 tips). The results presented here provide a comprehensive set of tools to address the display and exploration of such phylogenies. The tools are well-described and comprehensive, and additional online documentation is welcome.

The technical work to display such large datasets in a responsive fashion is impressive and this is aptly described in the paper. The author rightly decides that displaying large phylogenies is not simply a matter of rendering "more nodes", and so in my eyes, the major advancement is the approach used to downsample trees on-the-fly so that the number of nodes displayed at one time is manageable. This is detailed only briefly (Results section, 1st paragraph, 2 sentences). I would like to see more discussion about the details of this approach.

Thank you for this point, also raised by Reviewer 2. I have now added a lengthy section on this in the Materials and Methods, which I hope is helpful. The approach is not especially sophisticated, but it does the job and runs quickly.

Examples that came up while exploring the tool: the (well implemented) search functionality reports results from the entire tree (e.g. in Figure 4, the number of red circles is not a function of zoom level), how does this interact with a tree showing only a subset of nodes?

Yes, this is an important feature which I perhaps did not do justice to in the write-up. I have included in the new section in the Materials and Methods a paragraph discussing search results:

"In order to ensure that search results are always comprehensive, but at the same time to avoid overplotting, we take the following approach::

● Searches are performed across every single node on the tree to select a set of nodes that match the search. The total number of matches is displayed in the client.

● If fewer than 10,000 matches are detected, these are simply displayed in the client as circles

● If more than 10,000 matches are detected, the results are sparsified using the method above, and then displayed.

● Upon zooming or panning, the sparsification is repeated for the new bounding box."

How is the node order chosen with regards to "nodes that would be hidden by other nodes are excluded" and could this affect interpretations depending on the colouring used?

This perhaps was slightly sloppy language which did not directly describe the implementation. I have now rephrased this to "only nodes that overlap other nodes are excluded", as we don't in fact consider a notion of z-index when doing this. The way the sparsification works (now better described) means that the nodes excluded are determined essentially by position and I don't foresee this introducing particular biases, but this was an insightful point to raise.

Taxonium takes the approach of displaying all available data (sparsification of nodes notwithstanding). Biases in the generation of sequences, especially geographical, will therefore be present (especially so in the two main datasets discussed here - SARS-CoV-2 and monkeypox). This caveat should be made explicit.

This is certainly true. I have added this new paragraph in the Discussion:

"A further challenge is the vastly different densities of sampling in different geographic regions. Because Cov2Tree does not downsample sequences from countries which are able to sequence a greater proportion of their cases, the number of tips on a tree is not indicative of the size of an outbreak and in some cases even inferences of the directionality of migration may be confounded. There would be value in the development of techniques that allow visual normalisation of trees for sampling biases, which might allow for less biased phylogenetic representations without downsampling."

Has the author considered choosing which nodes to exclude for sparsified trees in such a way as to minimise known sampling biases?

The last sentence of the new paragraph above alludes to a sort-of-similar approach. I hadn't directly considered the approach the reviewer suggests. It is an interesting idea. The downsampling algorithm has to be very computationally inexpensive but it would be interesting to explore ways to do this. I am tracking this in https://github.com/theosanderson/taxonium/issues/437.

Interoperability between different software tools is discussed in a technical sense but not as it pertains to discovering the questions to ask of the data. As an example, spotting the specific mutations shown in figure 3 + 4 is not feasible by checking every position iteratively; instead, the ability to have mutations flagged elsewhere and then seamlessly explore them in Taxonium is a much more powerful workflow. This kind of interoperability (which Taxonium supports) enhances the claim of "providing insights into the evolution of the virus".

Thank you for flagging this point -- I am very excited by the growing ecosystem of interoperable tools. You are absolutely right that most of the insights Taxonium can bring into evolution rely also on this broader ecosystem. I have added a florid sentence in the concluding paragraph: "It forms part of an ecosystem of open-source tools that together turn an avalanche of sequencing data into actionable insights into ongoing evolution."

The prosaic reason I don't discuss Taxonium's interoperability features in more detail in this manuscript is that it aims to describe the version of Taxonium I initially developed, and these features were developed collaboratively by a broader group later on (and after deposition of this preprint).

Taxonium has been a fantastic resource for the analysis of SARS-CoV-2 and this paper fluently presents the tool in the context of the wider ecosystem of bioinformatic tools in use today, with the interoperability of the different pieces being a welcome direction.

Author Response

Reviewer #1 (Public Review):

“This manuscript reports the results of studies on the effects of an ActRIIB-Fc ligand trap inhibitor of myostatin on muscle contractures that develop when brachial plexus nerve roots are severed at 6 after birth. One component of this pathological response seems to be a failure to add sarcomeres as the skeleton grows resulting in short muscles. The authors use a carefully performed set of animal studies to test the effects of the ligand trap on denervation-induced limitations in range of motion in young mice. They also investigate several biochemical mechanisms that might contribute to contractures and be modified by the ligand trap. Finally, the test for gender discordance in the protective effect of a proteasome inhibitor against contractures. The major finding of these studies is that the ligand trap improved the range of motion at the elbow and shoulder in female mice but not in males. The major caveat to interpreting the data is that group sizes are relatively small such that the study may have been underpowered to detect smaller effects on a range of motion and biochemical endpoints.”

Thank you very much for your thoughtful review of our manuscript. We have taken your feedback regarding the interpretation of our data into consideration, and revised our manuscript accordingly.

We appreciate the reviewer’s careful scrutiny of our group sizes. As mentioned in the Statistical analysis section of our Materials and Methods, we included at least 6 mice per group for all range of motion and physiological endpoints. Based on an a priori power analysis, this is the number of mice per group necessary to detect a 10° difference in contractures and a 0.2 µm difference in sarcomere lengths at 80% power between experimental conditions. However, the small size of the forelimb muscles, especially following denervation, precluded the investigation of all biochemical parameters in each muscle. Therefore, we used we used smaller subgroup sizes for certain biochemical endpoints (Akt, Smad2/3, and Atrogin-1). In our revised Discussion, we acknowledge our study may be underpowered to detect smaller effects in these parameters of protein dynamics.

Discussion (lines #461-468): First, the small size of our denervated muscles precluded the use of the same muscles for all analyses, instead requiring smaller subgroup sizes as well as different muscles for certain biochemical endpoints (Akt, Smad2/3, MuRF1, and Atrogin-1). We therefore acknowledge that our study may be underpowered to detect smaller effects in certain parameters of protein dynamics, specifically signaling proteins and ubiquitin ligases. We also acknowledge that the precision of our findings would be further enhanced with the use of the same muscle type across all of our morphological, physiological, and biochemical analyses.

Reviewer #2 (Public Review):

“The manuscript by Emmert et al. describes an original and straightforward study demonstrating the utility of targeted therapy in a neonatal brachial plexus injury (NBPI) mouse model. The authors sought to investigate whether pharmacologic inhibition of MSTN signaling using a soluble decoy receptor (ACVR2B-Fc) could preserve longitudinal muscle growth and prevent contractures after NBPI. More specifically, through in vivo experiments using wild-type female and male mice, the authors assessed the impact of inhibiting the MSTN signaling in basal and pathophysiological conditions, on developmental, morphological, and biomechanical parameters, and on several biochemical markers of protein synthesis, protein degradation, and their associated signaling pathways, in forelimb skeletal muscle.

The authors provide multiple lines of compelling evidence that ACVR2B-Fc improves skeletal muscle biology and function in NBPI mice, provokes hypertrophy, rescues longitudinal growth, and impedes neuromuscular contractures in denervated muscles. Rather than improving the condition independently of the sex, it appears selective to the muscles of female mice showing thus a sex-specific improvement, and therefore the discovery of a sex dimorphism. The experiments also try to provide a mechanistic explanation, though it is incompletely clear why and how it is happening at the end.

Overall, the study details a promising intervention in NBPI mice and begins to highlight a pathway that can be exploited for this goal. While the reviewer did enjoy the manuscript, and the conclusions of this paper are mostly well supported by data, there are certain deficiencies that cannot be overlooked.

Strengths:

A) This study includes a clear-cut demonstration leading to a coherent narrative of a potential intervention for children affected by NBPI, which is well supported by prior literature mentioning the effects of palliative mechanical solutions and investigating the effects of pharmacologic strategies for the prevention of muscle contractures.

B) This study uses a pharmacologic chronic treatment, in vivo, on female and male neonatal mice to investigate the effects and relevant mechanisms of the MSTN signaling inhibition, using a soluble decoy receptor (ACVR2B-Fc), from the whole organism into the skeletal muscle and further into cellular signaling pathways.

C) This study provides promising data about the effects of the MSTN signaling inhibition on developmental, morphological, and biomechanical parameters, as well as biochemical markers in the NBPI mice.

D) This study underlines the importance of using female and male mice during experimental procedures, clearly showing that sex dimorphism can produce very different results.

E) The manuscript is well written, well organized, and cogent.

Weaknesses and Limitations:

A) This study attempts to provide mechanistic information to support and explain the results observed. However, the analysis remains superficial and should go further into detail especially in investigating completely the different molecular pathways considered, and the non-canonical alternatives.

B) The use of different muscles for biochemical analyses compared to the muscles used for developmental, morphological, and biomechanical parameters limits the interpretation of data, which could be due to muscle differences instead for example.

C) The interpretation of the findings should be done carefully, knowing that it is an MSTN/Activin A signaling blockade and not an MSTN inhibition alone.

D) The conclusion would be reinforced with data obtained at later time points (8 and/or 12 weeks).”

Thank you very much for your comprehensive and insightful review. Your detailed comments and suggestions have not only allowed us to improve our current manuscript with greater clarity and additional data, but they also reinforce our plan to elucidate biochemical mechanisms more completely in future studies. In this revision, we have provided additional experiments to strengthen our analysis of known pathways downstream of MSTN signaling, addressed the use of different muscles as well as the four-week time point, and discussed the potential implications stemming from the broad specificity of the ligand trap. We certainly share your enthusiasm about dissecting the different molecular pathways and non-canonical alternatives. Indeed, we intend to interrogate these mechanistic underpinnings with the same rigor with which we obtained our physiological and translational findings, which cannot be completed within the scope of the current study. Follow-up studies will focus on exploring non-canonical alternatives and investigating long-term effects at skeletal maturity and beyond.

Reviewer #3 (Public Review):

“This timely manuscript describes the sex dimorphisms in neonatal development as it applies to muscle injury and denervation. More and more studies are identifying sex differences in skeletal muscle function and dysfunction. This is one more study to point out differences. A missing piece to the field and this study are the mechanistic links between skeletal muscle function/dysfunction and sex differences. This paper starts to point to a mechanism highlighting the non-canonical AKT pathway. This is a very wellwritten manuscript with a clear experimental plan and workflow. I have no major concerns.

My biggest question is the molecular mechanism linking sex differences and skeletal muscle function and dysfunction. However, this is perhaps a follow-up study to the already complete study the authors present.”

Thank you very much for your kind words and enthusiasm! We likewise find it important to improve our understanding of sex differences in muscle function/dysfunction, and are committed to unraveling the molecular mechanism(s) that link them in future studies.

Author Response

Reviewer #1 (Public Review):

According to the space-time wiring hypothesis proposed by (Kim, Greene et al. 2014), the BC-off SAC circuit mimics the structure of a Reichardt detector; BCs closer to SAC soma have slower dynamics (they can be more sustained, have a delay in activation or slower rise time), while BCs further away are more transient. Later studies confirmed the connectivity and expanded the model on SACs (Ding, Smith et al. 2016, Greene, Kim et al. 2016). However, physiological studies that used somatic recordings to assess the BC properties at different dendritic distances were inconclusive (Stincic, Smith et al. 2016, Fransen and Borghuis 2017). Here, the authors used iGluSnFR, a glutamate sensor to measure the signals impinging on SAC dendrites. Their experimental findings align with the space-time wiring hypothesis, revealing sustained responses closer to SAC soma (mediated by prolonged release from type 7 BCs, and only slightly affected by amacrine cells), which according to their simulated SAC should produce a substantial increase in direction selectivity (DS).

I find the work to be clear and well presented. However, I do have some reservations with the findings:

Main points:

1) Very low number of cells examined in the key experiment presented in the first figure. The authors used a viral approach to express flex- iGluSnFR in SACs in Chat-Cre mice. Sometimes (apparently twice) the construct was expressed in individual SACs - this is a very underpowered experiment! The low number of successes precludes adequately judging the validity of the findings.

We agree with the reviewer that measuring iGluSnFR signals from single starburst dendrites is a powerful approach to confirm space-time wiring hypothesis. To bolster our data, we doubled our n number (updated Figure 1C and D, n = 66 ROIs; 20 dendrites and 4 retinas/FOVs). It should be noted that the results from these experiments are also validated on a larger scale across the starburst plexus (Figure 2).

2) The model doesn't represent key known properties of BC-SACs and the interactions within SAC dendrites. First, the authors decided to construct a ball and stick model that doesn't consider the dendritic morphology of the starburst cell. A stimulus moving over a SAC is expected to engage multiple dendrites with complex spatiotemporal patterns that are expected to have a substantial effect on the voltages recorded on the investigated dendrite (Koren, Grove et al. 2017). For example, the dendrites in the orthogonal orientation will be activated at about the same time as the proximal dendrites; how such strong input will affect dendritic integration is unclear but should be taken into account in the model. Second, the authors assume a similar peak BC drive between proximal and distal inputs. However, a recent study found an enhanced glutamate release from proximal BCs, mediated by cholinergic SAC drive ((Hellmer, Hall et al. 2021); not cited). How different release amplitude would affect the conclusions of the model?

It is well established that individual starburst dendritic sectors are relatively electrically isolated from each other (Miller & Bloomfield, 1983; Euler et al., 2002; Tukker et al., 2004, Poleg-Polsky et al., 2018) and thus we used a simple ball and stick to model direction selectivity in starburst dendrites.

Related to the Reviewer’s point, in the paper we explicitly acknowledge that the simple ball and stick model will not capture important network interactions that are expected to impact direction selectivity (e.g. SAC-SAC inhibition). We suggest this as a future line of investigation.

The idea of different synaptic weights across the starburst dendrite is an interesting one. If the proximal inputs are stronger relative to distal ones as the Reviewer suggests, it might be expected that the direction selectivity will be further enhanced. However, in a preliminary analysis, we did not find strong evidence for directionselectivity or sensitivity to MLA, to support the idea of cholinergic modulation.

3) Another reason for including an accurate dendritic morphology is in the differences in the number of BCs that target a cell. Because SAC dendrites cover the entire receptive field area, type 7 BCs, which occupy the proximal third of the dendrites (Ding, Smith et al. 2016, Greene, Kim et al. 2016), are expected to cover only 11% of the area covered by SAC dendrites (1/3 x 1/3 = 1/9) and correspondingly mediate just 11% of the BC drive. A nonbifurcating model presented here would dramatically overrepresent their contribution to SAC responses. ??

We have estimated BC numbers directly from the connectomics data which takes into account starburst morphology (Ding et al., 2016). To capture the heterogeneity of BCs that might be encountered at the level of single dendrites, in the revised manuscript, we have averaged responses over many trials in which the precise BC numbers varied according to the probability density functions observed in the connectomics data set (Ding et al., 2016). The details of the model parameters are now provided in the Methods section.

4) (Fransen and Borghuis 2017) found that off-SACs have a more pronounced distinction in the time to peak than on-SACs. I found it surprising that given the large body of work demonstrating the effectivity of the viral approach in expressing iGluSnFR in off BC (Borghuis, Marvin et al. 2013, Franke, Berens et al. 2017, Szatko, Korympidou et al. 2020, Gaynes, Budoff et al. 2021, Strauss, Korympidou et al. 2021), that the authors did not compare between on and off SAC populations.

It is possible that the kinetic differences are more pronounced for inputs to OFF starbursts. However, we observed a weaker iGluSnFR expression in the OFF starburst layer and the S/N was below what was required for our analysis. Therefore, we focused on the ON starburst.

5) Recent work (Gaynes, Budoff et al. 2021) suggests that BCs' responses to motion and to static flashes have distinct dynamics. However, the current manuscript tests responses to flashed stationary stimuli experimentally, and then combines them in a simulation modeling a moving stimulus. This potential limitation of the study should at least be discussed.

The Reviewer correctly points out that static and motion stimuli might have distinct dynamics (especially ‘emerging’ stimuli). We now describe this limitation of our study and discuss the findings of Gaynes et al. (2022)

We have revised our model to take BC release rates truncated according to stimulus velocity and size to more appropriately represent the duration of the stimulus.

Reviewer #2 (Public Review):

The authors present a nice series of imaging experiments confirming previous anatomical and electrophysiological evidence for the "space-time wiring" model for directionally selective responses in SAC dendrites. Fluorescence measurements with a genetically encoded glutamate indicator show that excitatory inputs to proximal SAC dendrites are more sustained than distal dendrites. Although the signals are shaped by surround inhibition, the fundamental differences persist with inhibition blocked, suggesting intrinsic differences in the synaptic release processes in different cone bipolar cell types.

The authors examine iGluSnFR dynamics in individual SACs (Figure 1) and in a population of SACs (Figure 2). The latter is possible because distal inputs to all SACs occur deeper in the IPL and so can be imaged separately from the proximal inputs, and it permits the measurement of many more synapses in each experiment. The former approach is particularly powerful, however, because it allows careful mapping of the different types of inputs along the dendritic axis of individual SACs. This experiment was performed in only seven dendrites in two retinas, however; consequently, the confidence intervals for any spatial fitting would be quite broad. This experiment would be strengthened with additional data from more dendrites.

We have now the increased n number for Figure 1 in the revised manuscript (updated Figure 1C and D, n = 66 ROIs; 20 dendrites and 4 retinas/FOVs). Please see response to the Reviewer #1.

It is very interesting that white noise stimuli do not pull out the kinetic differences - interesting enough to merit inclusion in the primary figures rather than as a supplement. These results seem valuable to our understanding of DS processing, but the implications remain unclear. Is it really the case that DS is eliminated - or even substantially degraded - when motion stimuli are presented atop some background (i.e., conditions in which the circuit is continuously stimulated)? Are the distinct kinetics are brought about by abrupt, large changes in luminance - if so, wouldn't one expect much weaker DS in response to drifting sinusoidal gratings?

The question of how direction is encoded in the natural scene is very interesting but beyond the scope of this study. We presented a preliminary white noise analysis to show that our recordings are consistent with other recent reports (e.g. Strauss et al., 2022) which cast doubt on the space-time wiring model, rather than to directy address this specific issue. It should also be noted that complementary inhibitory and/or other intrinsic dendritic mechanisms may ensure that dendrites continue to remain DS in regimes in which BC mechanisms appear to be ineffective.

In the introduction (p. 3A) the authors suggest that space clamp errors could distort EPSC kinetics, causing EPSCs arriving distally to appear more transient than those arriving more proximally. This seems contrary to what one would typically expect: cable theory would predict that more distal inputs ought to be filtered more, therefore appearing more prolonged, not more transient, than proximal inputs. It does not seem necessary to cast doubt on the previous results (Fransen and Borghuis, 2017) to motivate sufficiently the present experiments. One might simply point out that electrophysiological recordings do not provide precise information regarding the anatomical location of synaptic inputs.

In the revised manuscript, we changed the text according to the Reviewer’s suggestion.

Reviewer #3 (Public Review):

In the study "Spatiotemporal properties of glutamate input support direction selectivity in the dendrites of retinal starburst amacrine cells", Srivastava, deRosenroll, and colleagues study the role of excitatory inputs in generating direction selectivity in the mouse retina. Computational and anatomical studies have suggested that the "space-time-wiring" model contributes to direction-selective responses in the mammalian retina. This model relies on temporally distinct excitatory inputs that are offset in space, thereby yielding stronger responses for motion in one versus the other direction. Conceptually, this is similar to the Reichardt detector of motion detection proposed many decades ago. So far, however, there is little functional evidence for the implementation of the space-time-wiring model. Here, Srivastava, deRosenroll and colleagues use local glutamate imaging in the ex-vivo mouse retina combined with biophysical modeling to test whether temporally distinct and spatially offset excitatory inputs might generate direction-selective responses in starburst amacrine cells (SACs). Consistent with the space-time-wiring model, they find that glutamatergic inputs at proximal SAC dendrites are more sustained than inputs at distal dendrites. This finding was consistent across different sizes of stationary, flashed stimuli. They further linked the sustained input component to the genetically identified type 7 bipolar cell and showed that the difference in temporal responses across proximal and distal inputs was independent of inhibition, but rather relied on excitatory interactions. By estimating vesicle release rates and building a simple biophysical model, the authors suggest that next to already established mechanisms like asymmetric inhibition, excitatory inputs with distinct kinetics contribute to direction-selective responses in SACs for slow and relatively large stimuli.

In general, this study is well-written, the data is clearly presented and the conclusion that (i) the temporal kinetics of excitatory inputs varies along SAC dendrites and that (ii) this might then contribute to direction selectivity is supported by the data. The study addresses the important question of how excitation contributes to the generation of direction-selective responses. There have been several other studies published on this topic recently, and I believe that the results will be of great interest to the visual neuroscience community.

However, the authors should address the following concerns:

• They should demonstrate that differences in response kinetics between proximal and distal dendrites are unrelated to differences in signal-to-noise ratio.

In response to the Reviewer’s comment, we have now added new plots to supplementary Figure S1 (A, B) that show that the response kinetics are not strongly related to signal strength.

• To demonstrate consistency across recordings/mice, the authors should indicate data points from different recordings (e.g. Fig. 2C).

In the updated Figure 2C-E, we have now added the average values for each recording to indicate the consistency/variability in the data.

• The authors mention in the introduction that the space-time-wiring model is conceptually similar to other correlation-type motion detectors that have been experimentally verified in different species. It would be great to expand on the similarity and differences of the different mechanisms in the Discussion, especially focusing on Drosophila where experimental evidence at the synaptic level exists.

It should be noted that the results of the influential Nature paper describing the spacetime wiring of inputs to T4 DS neurons in the fly system were not reproduced by the same group. A new paper from Axel Borst’s group, however, shows a distinct source of spatially offset excitation (glutamate-mediated by disinhibition) may underlie the multiplicative operation. Nevertheless, to our knowledge, no studies have mapped out the spatiotemporal properties of inputs across single T4/T5 DS neurons as we have done for the starburst. In the revised manuscript we briefly summarize the fly literature in response to the Reviewer’s suggestion.

• The authors use stationary spot stimuli of different sizes to characterize the response kinetics of excitatory inputs to SACs. I suggest the authors add an explanation for choosing only stationary stimuli for studying the role of excitatory inputs in direction selectivity/motion processing.

Please see the above response to Reviewer #1.

In addition, the authors use simulated moving edges to stimulate the model bipolar cells. They should provide details about the size of the stimulus and the rationale behind using this size, given their previous results.

For simulation experiments, bipolar cell inputs were triggered by a 400 µm wide bar moving over a range of velocities (0.1 – 2 mm/s). We have now added more details in the Methods section and main text in the revised manuscript.

• Using the biophysical model, the authors show that converting sustained bipolar cell inputs to transient ones reduces direction selectivity in SACs. I suggest the authors also do the opposite manipulation/flip the proximal and distal inputs or provide a rationale why they performed this specific manipulation.

Thanks for the suggestion. We have now updated Figure 6B showing DSi vs velocity plots for several different bipolar cell input distributions – (i) sustained-transient, (ii) transient-sustained, (iii) all transient, and (iv) all sustained.

• In each figure, the authors should note whether traces show single trial responses or mean across how many trials. If the mean is presented (e.g. Suppl. Fig. 2a), the authors should include a measure of variability - either show single ROIs in addition and/or add an s.d. shading to the mean traces.

In the revised manuscript, we have now indicated the mean and number of trials for each figure. We have also added S.E.M values to the mean traces in the figures.

Author Response

Reviewer #1 (Public Review):

In this manuscript by Kim et al., the authors use live-cell imaging of transcription in the Drosophila blastoderm to motivate quantitative models of gene regulation. Specifically, they focus on the role of repressors and use a 'thermodynamic' model as the conceptual framework for understanding the addition and placement of the repressor Runt, i.e. synthetic insertion of Runt repressor sites into the Bicoid-activated hunchback P2 enhancer. Coupled with kinetic modeling and live-cell imaging, this study is a sort of mathematical enhancer bashing experiment. The overarching theme is measuring the input/output relationship between an activator (bicoid), repressor (runt), and mRNA synthesis. Transcriptional repression is understudied in my opinion. One finding is that the inclusion of cooperativity between trans-acting factors is necessary for understanding transcriptional regulation. Most, if not all, of the tools used in this paper have been published elsewhere, but the real contribution is a deep, quantitative dissection of transcriptional regulation during development. As such, the only real questions for this referee are whether the modeling was done rigorously to produce some general biological conclusions. By and large, I think the answer is yes.

We thank the reviewer for this thoughtful evaluation of our work. We agree with the reviewer’s assessment that transcriptional repression, especially the quantitative dissection of transcriptional repression, is understudied compared to transcriptional activation.

Comments:

Fig. 6 was the most striking figure for this referee, specifically that different placements of Runt molecules on the enhancer lead to distinct higher order interactions. I am wondering if the middle data column in Fig. 6 represents a real difference from the other two, and if so, it seems that the positioning - as opposed to simply the stoichiometry - is essential in cooperativity. This conclusion implies that transcriptional regulation is more precise than what some claim is just a mushy ball of factors close to a promoter. In other words, orientation may matter. Proximity may matter. Interactions in trans matter.

We thank the reviewer for pointing out a feature of our data that we did not emphasize enough originally. Indeed, the construct in the middle column, which we termed [101], could be better recapitulated with the simplest model of zero free parameters than the other two constructs. As the reviewer pointed out, this raises an interesting question about the “grammar” of an enhancer: the placement and orientation of binding sites for transcription factors might matter yet we do not have a clear understanding of the logic. We have now incorporated a discussion of this topic in the Discussion section.

There needs to be at least one prediction which is validated at the level of smFISH / mRNA in the embryo. Without detracting from the effort the authors have expended in looking directly at transcription, if the effects can't be felt by the blastoderm at the level of mRNA/cell, it becomes difficult to argue for the relevance to development. Also, I feel there is little chance that these measurements can be quantitatively replicated unless translated to the level of total protein or mRNA. Such a measurement (orthogonal quantitative confirmation of the repressor cooperativity result) would also assuage my concern about the time averaging as shown in Fig. S3.

Our study focused on predicting the initial rate of transcription because it is the measurable quantity that most directly relates to the binding and action of the transcriptional activators and repressors used in this study. We argue that the action of transcription factors would be more accurately assessed by monitoring the rate of transcription, rather than the accumulated mRNA, which could be confounded by the dynamics of the whole transcription cycle—initiation, elongation and termination—as well as nuclear export, diffusion and degradation of transcripts. We are, of course, excited to eventually be able to predict a whole pattern of cytoplasmic mRNA over space and time from knowledge of the enhancer sequence. However, if we cannot predict the initial rate of RNA polymerase loading dictated by an enhancer, we argue that there is little hope in predicting such cytoplasmic patterns. We emphasized this point in the Discussion (Line XX-YY). Regardless, to assuage the reviewer’s concern, we have performed additional analyses to assess the effect of repression at the level of accumulated mRNA.

First, we have quantified the accumulated mRNA during nuclear cycle 14, which is the time window that we have focused on in this study. To make this possible, we have integrated the area under the curve of MS2 time traces which has been already shown to be a reporter of the total amount of mRNA produced by FISH (Garcia et al., Current Biology 23:2140, 2013;Lammers et al., PNAS 17:836, 2020). This integration reporting on accumulated mRNA is now shown for all constructs in the presence and absence of Runt protein in the new Figure S17. This figure clearly shows that the consequences of repression are present in the blastoderm, not just at the level of transcriptional initiation, but also at the level of accumulated mRNA.

We then compared the accumulated mRNA profiles shown in Figure S17 to the initial rate of RNAP loading at each position of the embryo along the anterior-posterior axis for all constructs in the presence and absence of Runt protein. These new results are shown in a new figure, Figure S19. Interestingly, we saw a good correlation (Pearson correlation coefficient of 0.90) between these two metrics. Thus, we argue that our conclusion that higher-order cooperativity is necessary to account for the initial rate of RNA polymerase loading would still hold for predicting the accumulated mRNA.

Reviewer #3 (Public Review):

The authors have presented results from carefully planned and executed experiments that probe enhancer-drive expression patterns in varying cellular conditions (of the early Drosophila embryo) and test whether standard models of cis-regulatory encoding suffice to explain the data. They show that this is not the case, and propose a mechanistic aspect (higher order cooperativity) that ought to be explored more carefully in future studies. The presentation (especially the figures and schematics) are excellent, and the narrative is crisp and well organized. The work is significant because it challenges our current understanding of how enhancers encode the combinatorial action of multiple transcription factors through multiple binding sites. The work will motivate additional modeling of the presented data, and experimental follow-up studies to explore the proposed mechanisms of higher order cooperativity. The work is an excellent example of iterative experimentation and quantitative modeling in the context of cis-regulatory grammar. At the same time, the work as it stands currently raises some doubts regarding the statistical interpretation of results and modeling, as outlined below.

We thank the reviewer for noting the significance of our work. We tried our best to address the concerns of the reviewer regarding the statistical interpretation of results and theoretical modeling throughout our responses below.

The results presented in Figure 5 are used to claim that the data support (i) an unchanging K_R regardless of the position of the Runt site in the enhancer and (ii) an \omega_RP that decreases as the site goes further away from the promoter, as might be expected from a direct repression model. This claim is based on only testing the specific model that the authors hypothesize and no alternative model is compared. For instance, are the fits significantly worse if \omega_RP is kept constant and the K_R allowed to vary across the three sites. If different placements of the Runt site can result in puzzling differences in RNAP-promoter interaction, it seems entirely possible that the different site placements might result in different K_R, perhaps due to unmodeled interference from bicoid binding. Due to these considerations, it is not clear if the data indeed argue for a fixed K_R and distance-dependent \omega_RP.

We apologize for the lack of justification in assuming that Kr remains constant and wrp varies depending on the position of the Runt binding sites. Following the reviewer’s suggestion, we tested the alternative scenarios where we either fix or vary different combinations of wrp and Kr for our one-Runt binding site constructs. The result is now shown in a new figure, Figure S16. In short, as reported by the Akaike Information Criterion (AIC) in Figure S16F, the MCMC fit explains the data best in the scenario of fixed Kr and different wrp values for one-Runt binding site constructs. Furthermore, we also performed the MCMC inference in the case where we varied both Kr and wrp values across constructs. From this analysis, we obtained similar values of Kr while having different values of wrp across constructs as shown in Figure S16G. Overall, we believe that this evidence strongly supports our assumption of having consistent Kr values but different wrp values for the one-Runt binding site constructs.

Results presented in Figure 6 make the case that higher order cooperativity involving two DNA-bound molecules of Runt and the RNAP is sufficient to explain the data. The trained values of such cooperativity in the three tested enhancers appear orders of magnitude different. As a result, it is hard to assess the evidence (from model fits) in a statistical sense. Indeed, if all of the assumptions of the model are correct, then using the high-order cooperativity is better than not using it. To some extent, this sounds statistically uninteresting (one additional parameter, better fits). It is not the case that the new parameter explains the data perfectly, so some form of statistical assessment is essential.

The inferred cooperativity values are indeed orders of magnitude different. However, the cooperativity terms can be also written as “w = exp(-E/(kBT))”, where the E is the interaction energy, kB is the Boltzmann constant, and T is the temperature. As a result, we should compare the magnitude of the different cooperativities on a log-scale. In brief, the interaction energies wrr from the three two-Runt binding site constructs range between 0 and 1kBT, and the higher-order cooperativity wrrp has an energy between -2 and 4kBT. Interestingly, these energies are of the same order of magnitude as the interaction energies typically reported for bacterial transcription factors (e.g., Dodd et al., Genes and Development 18:344-54, 2004). It is important to note that our inferred interaction energies could be either positive or negative, suggesting that both cooperativity and anti-cooperativity can be at play depending on the architecture of the two Runt binding sites. We now report on these interactions in the language of energies Table S1 and elaborate on this in the Discussion section (Line XX-YY).

Finally, following the reviewer’s suggestion on statistical assessment of whether addition of parameters indeed explains the data better, we adopted the Akaike Information Criterion (AIC) as a metric to compare different models used in Figure 6 and now show the results in a new panel, panel G. Briefly, AIC is calculated by assessing the model’s ability to explain the data while penalizing for having more parameters. The smaller the AIC value is, the better the model explains the data. As we have claimed, the AIC showed a dramatic decrease when adopting the higher-order cooperativity as shown in Figure 6G. Thus we argue that the addition of higher-order cooperativity, while not being able to completely explain the data, is indeed capable of increasing the agreement between experiments and theory across all our two-Runt site constructs.

Moreover, it is not the case that the model structure being tested is the only obvious biophysics-driven choice: since this is the first time that such higher order effects are being tested, one has to be careful about testing alternative model structures, e.g., repression models that go beyond direct repression and pairwise cooperativity that goes beyond the traditional approach of a single (pseudo)energy term.

We agree with the reviewer that alternative models with different mechanisms of repression should be mentioned. We have clarified this point further in Discussion (Line XX -YY). In summary, we tested both “competition” and “quenching” models of repression as proposed in Gray et al, (Genes and Development 8:1829, 1994). Interestingly, Figure S5 shows that the “competition” model gives a worse fit compared to the “direct repression” and “quenching” models for the one-Runt binding site cases. We further tried to test these alternative models in the case of two-Runt binding sites constructs. The result is shown in Figure S7 (competition) and S8 (quenching). These figures also reveal that the “competition” model underperformed compared to the “direct repression” or “quenching” models. For the “quenching” model to fit the data, we also had to invoke higher-order cooperativity that is beyond pairwise cooperativity. Thus, we believe that the requirement of higher-order cooperativity holds regardless of the choice of the specific model. Of course, our models of repression are very likely an oversimplification of how repressors actually work. However, given that these simple models have been a prevalent choice of proposed mechanisms for repression in the field of transcriptional repression for the past decades, we believe that the significance of our work lies in the fact that we challenged these models by turning them into precise mathematical statements (in the form of widespread thermodynamics models) and confronting them with quantitative data.

The general theme seen in Figure 6 is seen again in Figure 7, when a 3-site construct is tested: model complexities inferred from all of the previous analyses are insufficient at explaining the new data, and new parameters have to be trained to explain the results. The authors do not seem to claim that the higher order cooperativity terms (two parameters) explain the data, rather that such terms may be useful.

We agree that our previous approach was confusing. Figure 7A indeed incorporated all inferred parameters from the previous rounds of inference (Kb, wbp, p, R, as well as Kr, wrp, wrr, and wrrp). However, it is clear that this set of parameters, even including the higher-order cooperativity from two-Runt binding sites cases, was not enough to explain the data from three-Runt binding sites case. Thus, we had to invoke another free parameter, which we termed wrrrp, to explain the data. We have revised Figure 7B such that it is now showing the “best” MCMC fit which explains the data quite well (instead of just showing the “improvement” of fits).

Author Response

Reviewer #1 (Public Review):

Xu et al show that mutants in three DNA replication proteins, Mcm2, Pole3, and Pole4 have defects in differentiation in a mouse embryonic stem cell (ESC) model. The Mcm2 mutant (called Mcm2-2A), which specifically blocks the interaction of Mcm2 with histones, has defects in multilineage differentiation and neural differentiation, despite having minimal effect on ESC proliferation or gene expression. Mcm2-2A fails to fully silence ESC genes and activate appropriate differentiation genes. Chromatin profiling analyses show Mcm2 binds many promoters. During differentiation, the Mcm2-2 mutant retains K3K27me3 at differentiation gene promoters and reduced accessibility, consistent with the observed defects in gene expression.

The findings that Mcm2-2A has minimal effect on proliferation and gene expression in ESCs, but impairs differentiation are interesting, particularly since this mutant seems to separate the histone binding roles of Mcm2 and its roles in DNA replication. Furthermore, the fact the histone binding function is only necessary when cells exit the pluripotent state is of interest. The studies were reasonably thorough and generally support the conclusions that Mcm2 is important for reshaping histone modifications during differentiation, although the details by which this occurs are not clear. Although the authors used two different strategies for identifying the direct binding sites of Mcm2 on chromatin, Mcm2 enrichment at individual loci was relatively weak, suggesting Mcm2 may localize somewhat diffusely. This somewhat weakens the conclusions about the direct vs indirect effects of Mcm2 on chromatin structure and gene expression.

Overall, this paper reports an interesting set of findings that have a few caveats/limitations regarding how Mcm2 mediates these effects on chromatin during ESC differentiation.

My biggest question is about the Mcm2 CUT&RUN data, which appears to have low signal-to-noise. The authors appear to be aware of this issue, as they also used an Mcm2-FLAG line for CUT&RUN studies, with similarly low signal to noise. To be clear, this may be due to the binding properties of Mcm2, which may bind chromatin relatively broadly, causing few highly enriched peaks to be observed (similar to cohesin complex in the absence of CTCF). However, it makes the Mcm2 binding data difficult to interpret. First, most Mcm2 peaks seem to be near promoters. Promoters often have a small amount of signal in negative control (IgG or irrelevant antibody) CUT&RUN experiments, presumably due to their MNase accessibility. It is not clear to what extent Mcm2 peaks exceed background because no negative control CUT&RUN was performed. The high correlation of FLAG and Mcm2 CUT&RUN libraries might still be evident if some of this signal is due to background at TSSs. Second, the authors call 13,742 peaks, but browser tracks of some example peaks at the Pax6 and Nanog promoters show minimal enrichment relative to surrounding regions (Fig. 5I, 5S1B). I have concerns that some of these peaks called statistically significant are not biologically meaningful.

We thank the reviewer for his/her time to review this story and for his/her positive comments. We shared the reviewers’ concern about low signal to noise for Mcm2 CUT&RUN. However, the Mcm2 CUT&RUN signals most likely reflect Mcm2 binding.

Reviewer #2 (Public Review):

It is established that different histone chaperones not only facilitate the assembly of DNA into nucleosomes following DNA replication and transcription but also are essential to stem cell maintenance and differentiation. Here the authors Xiaowei Xu et al. propose a novel role for Mcm2 DNA helicase, a subunit of the origin licensing complex Mcm2-7 in stem cell differentiation in addition to or in connection to maintaining genomic integrity in DNA replication. This study is a continuation of the authors' previously published work implicating Mcm2-Ctf4-Polα axis in the parental histone H3-H4 transfer to lagging strands. The present study is elegantly executed with a systemic analysis of the role of Mcm2 in the ES differentiation to neuronal lineage.

We thank the reviewer for his/her time to review the manuscript and for his/her positive comments.

Major questions

1) Mouse ES cells with a mutation at the histone binding motif of Mcm2 (Mcm2-2A) grew normally, but exhibited defects in differentiation. Also, the Mcm2-2A mutation linked global changes in gene expression, chromatin accessibility and histone modifications were not apparent to the similar degree in mouse ES cells compared to NPCs. The authors suggest that the excessive amount of Mcm2 in ES cells, similar to DNA replication, safeguards the chromatin accessibility and gene expression in mouse ES cells resulting in Mcm2-2A mutant ES cells being able to restore the symmetric distribution of parental histones before cell division. What is underlying the mechanism of this difference since overabundant Mcm2 is present in both ES cells and NPCs?

This is an excellent good question that we can only speculate. As discussed above and below, our results indicate that Mcm2 functions with Asf1a to resolve the bivalent chromatin domains during pluripotency exit. Therefore, it is highly likely that Mcm2’s role in differentiation is independent of its role in DNA replication. Therefore, in the revised manuscript, we downplayed this possibility and suggested that the differentiation defects in Mcm2-2A mutant cells may arise from the involvement of Mcm2 in resolving bivalent chromatin domains (p24).

2) CAF-1, Asf1a, and Mcm2 partake in similar or redundant chromatin regulation during differentiation with silencing of pluripotent genes and induction of lineage-specific genes. These processes were found commonly dysregulated in both Mcm2-2A cells and Asf1a KO ES cells, albeit with varying degrees. How can authors exclude the possibility of Mcm2 affecting the differentiation via Asf1 with which it forms a complex, as a potentially redundant mechanism in the deposition of newly synthesized or recycled histones?

To address this question, we performed the following experiment. First, we overexpressed Asf1a in both WT and Mcm2-2A mutant ES cells and determined whether Asf1a overexpression suppress the differential defects in Mcm2-2A mutant cells (Figure 2- figure supplement 2A). We observed that Asf1a overexpression did not rescue the differential defects of Mcm2-2A mutant cells based on analysis of cell morphology (Figure 2- figure supplement 2B) as well as the expression of Oct4 and lineage specific genes during differentiation (Figure 2- figure supplement 2C-E).

Second, we knocked out Asf1a in both WT cells and Mcm2-2A mutant cells using CRISPR/Cas9 (Figure 2- figure supplement 2F and 2G). and compared the effects of Asf1a KO, Mcm2-2A and Mcm2-2A Asf1a KO double mutation on differentiation. As detailed above, these results indicate that Mcm2’s function in the induction of lineage specific genes is dependent on Asf1a. However, Mcm2 also has independent role on the regulation of pluripotency genes which might through its unique roles on parental histone deposition and gene expression regulation. We discussed these points in the results (p10-13) and discussion (p22-23).

It is known that CAF-1 and Mcm2 are involved in deposition of new H3-H4 and parental H3-H4, respectively. Further, there is little evidence that CAF-1 interacts with Mcm2 in the literature. Therefore, we did not analyze the relationship between CAF-1 and Mcm2 during differentiation. In the revised manuscript, we discussed these points to address the reviewer’s concern.

Can authors test potential redundancy between Mcm2 and other histone chaperones and modifiers? Can the authors rescue the NPC phenotype induced by Mcm2 -2A mutant? Can the authors rescue the Mcm2-2A phenotype by overexpression of another histone chaperone or modifier?

As stated above, we overexpressed Asf1a, which is known to interact with Mcm2, and found that overexpression of Asf1a did not rescue differentiation defects of Mcm2-2A mutant cells. On the other hand, overexpression of Mcm2 in Mcm2-2A cells did rescue defects in differentiation (Figure 2E-G). As discussed above, our results indicate that Mcm2 and Asf1a function in the same pathway for resolving bivalent chromatin domains based on analysis of Asf1a KO Mcm2-2A double mutant as well as RNA-seq datasets of Asf1a KO and Mcm2-2A during differentiation. However, the defects of Mcm2-2A on silencing of Oct4 was not observed in Asf1a mutant cells. Together, these results indicate that the defects in differentiation of Mcm2-2A cells are, at least in part, due to a reduced interaction with Asf1a. Furthermore, Mcm2 also has its unique role in promoting the silencing of pluripotency genes.

3) Authors argue that Mcm2 may regulate the deposition of newly synthesized or recycled histones via the ability to recycle 1. parental H3.1 and H3.3, 2. via binding directly H3-H4, and/or via 3. Pol II transcription. Which of these mechanisms may be more unique to Mcm2 compared to the other histone chaperones and modifiers?

This is a very interesting, but challenging question to address for the following reasons. First, while Mcm2-2A mutant showed defects in binding to both H3.1 and H3.3, it is almost impossible to identify a Mcm2 mutant that bind H3.1 and H3.3 differently. Based on our recent studies, our results indicate that the defects in the induction of lineage specific genes are likely due to a loss of Asf1a interaction. However, the defects in silencing of pluripotent gene such as Oct4 is unlikely due to a loss of interaction with Asf1a. Therefore, we suggest that defects in silencing of pluripotent genes in Mcm2-2A mutant cells are likely due to Mcm2’s role in parental histone transfer and/or gene transcription. In the revised manuscript, we dramatically modified the discussion section to reflect the new results as well as to further mitigate the concerns of the reviewer.

4) Authors observed that in the ES cells the majority of Mcm2 CUT&RUN peaks were enriched with H3K4me3 CUT&RUN signals and ATAC-seq peaks and a small fraction of Mcm2 CUT&RUN peaks were engaged at the bivalent chromatin domains (H3K4me3+ and H3K27me3+). In contrast, in wild-type NPCs all the Mcm2 peaks co-localized with H3K4me3 and ATAC-seq peaks (H3K4me3+, H3K27me3-). The authors thus argued that Mcm2 binding to chromatin is rewired during differentiation citing this differential engagement of Mcm2 with the bivalent chromatin domains in ES and NPCs. What is the mechanism of Mcm2 differential engagement with the bivalent chromatin domains?

As stated above, the original discussion may be misleading. In the revised manuscript, we dramatically rewrote the discussion based on the new results indicating that Mcm2 and Asf1a function similarly for the induction of lineage specific genes marked by bivalent promoters during pluripotency exit.

5) Authors indicated that in mouse ES cells Mcm2 CUT&RUN peaks exhibited low densities at the origins. DNA replication origins are licensed by the MCM2-7 complexes, with most of them remaining dormant. Dormant origins rescue replication fork stalling in S phase and ensure genome integrity. It is reported that ESs contain more dormant origins than progenitor cells such as NPCs and that may prevent the replication stress. Also, partial depletion of dormant origins does not affect ECs self-renewal but impairs their differentiation, including toward the neural lineage. Moreover, reduction of dormant origins in NPCs impairs their self-renewal due to accumulation of DNA damage and apoptosis. Can authors exclude the role of reduced dormant origins reflected in the reduced density of Mcm2 at the origins in the differentiation to neuronal lineages?

Thank the reviewer for excellent suggestions. We have now discussed these points about the potential role of Mcm2 in dormant origins and differentiation defects in the discussion (p24). However, I would like to point out that based on the new results, this is an unlikely mechanism. Supporting this idea, it is known that Mcm2-2A mutant cells from yeast and mouse ES cells are not sensitive to replication stress, such as HU (Foltman, Evrin et al. 2013, Huang, Stromme et al. 2015).

Author Response

Reviewer #1 (Public Review):

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

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

We thank the reviewer for calling our attention to the lack of accessible explanations on the significant terms and quantities in this framework. Following the suggestion in the comments below, we added Box 1, providing intuitive and plain explanations on the terms of fitness, fitness landscape, selection, selection strength, and cumulants. In each section, we explain the standard usage of these terms in evolutionary biology and clarify the similarities and differences in this framework. We also added a figure to Box 1 and provided a schematic explanation of the relationships among chronological and retrospective distributions, fitness landscapes, and selection strength. We believe that these explanations and a figure would better clarify the meanings and functions of these quantities.

Major comments :

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

We appreciate the suggestion of making a glossary of the terms. Following the suggestion, we added Box 1 to provide intuitive and plain explanations of the terms used in this framework.

In Box 1, we explain why we called h(x) a fitness landscape, referring to its standard usage in evolutionary biology. In evolutionary biology, fitness landscapes (also called adaptive landscapes) are visual representations of relationships between reproductive abilities (fitness) and genotypes. The height of landscapes corresponds to fitness. Since constructing "genotype space" is usually difficult, fitness is often mapped on an allele frequency or phenotype (trait) space to depict a "landscape." Fitness landscapes introduced in our framework are analogous to those in evolutionary biology in that fitness differences are mapped on trait spaces. Although fitness landscapes in evolutionary biology are usually metaphorical or conceptual tools for understanding evolutionary processes, the landscapes in our framework are directly measurable from division count and trait dynamics on cellular lineages.

We also explain "selection" and "selection strength" in Box 1. As pointed out, we define three kinds of selection strength measures. These three measures share a similar property of reporting the overall correlations between traits and fitness. However, they also have critical differences regarding additional selection effects they represent: S_KL^((1)) for growth rate gain, S_KL^((2)) for additional loss of growth rate under perturbations, and their difference S_KL^((2))-S_KL^((1)) for the effect of selection on fitness variance. We restructured the sections in Results and clarified these important meanings of the different selection strength measures.

We removed the term "fitness cumulants" as this is non-general and might cause confusion to readers. We now rephrased this more precisely as "cumulants of a fitness landscape (with respect to chronological distribution)." Besides, we added a general explanation of "cumulants" to Box 1 and clarified what first, second, and third-order cumulants represent about distributions.

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

In addition to Box 1, we included more intuitive explanations on fitness landscapes and selection strength where they first appear in the Theoretical background section. As pointed out, descriptions of the linkage between the selection strength measures and Kullback-Leibler divergence were only in the Supplemental Information in the original manuscript. We now explicitly show this linkage where we first define the selection strength.

Following this comment, we also changed the definition of a fitness landscape from the original one to h(x)≔τΛ+ln⁡〖Q_rs (x)/Q_cl (x)〗 (Eq. 1), using the chronological and retrospective distributions introduced in the preceding paragraph. This definition is mathematically equivalent to the previous one, but we believe it is more intuitive.

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

We thank the reviewer for pointing out our insufficient explanation of an essential feature of this framework. As we now explain in the "Examples of biological questions" section (L62-65) and Discussion (L492-493), this framework does not assume stationary phenotypes (traits) on cellular lineages. On the contrary, we developed this framework so that one can quantify fitness and selection strength even for non-stationary phenotypes (traits) due to factors such as non-constant environments and inherent stochasticity.

In fact, if traits are stationary in cellular lineages, this framework becomes essentially identical to the individual-based evolutionary biology framework (see ref. 26, for example). Our framework assumes a cell lineage as a unit of selection and any measurable quantities along cellular lineages as lineage traits, whether they are stationary or non-stationary. Therefore, our framework can evaluate fitness landscapes and selection strength without explicitly taking the environmental conditions around cells into account. This means that h(x) and S[X] in this framework extract the correlations between the traits of interest and division counts among various factors that could potentially influence division counts. On the other hand, this framework has a limitation due to this design: it cannot say anything about the influence of factors such as non-quantified traits and potential variations in environmental conditions. We now explain these important points explicitly in the revised manuscript (L493-496).

Likewise, stochasticity in division rate does affect division count distributions, and its influence appears as differences in the selection strength of division count S[D]. As stated in the text, S[D] sets the maximum bound for the selection strength of any lineage trait (L143-145). Therefore, S_rel [X]≔S[X]/S[D] reports the relative strength of the correlation between the trait X and lineage fitness in a given level of S[D] in each condition.

To clarify the influence of stochasticity in division rate, we present a cell population model in which cells divide stochastically according to generation time (interdivision time) distributions in Appendix 2 (we moved this section from the Supplemental Information with modifications). We can confirm from this model that the shapes of generation time distributions influence the selection strength S[D]. Importantly, one can understand from this model that stochasticity in generation times constantly introduces selection to cell populations and modulates the growth rate and selection strength even in the long-term limit. We now clarify this important point in the Discussion (L519-526).

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

When a decrease in fitness variance by selection is mentioned in evolutionary biology, an upper bound and inheritance of fitness across the generations of individuals are usually assumed. In such circumstances, selection drives the fitness distribution toward the maximum value, and the selection eventually causes fitness variance to decrease. However, even in this process, a decrease is not assured for every step; whether selection reduces fitness variance at each step depends on the fitness distribution at that time.

In our argument, we compared fitness variances between chronological and retrospective distributions. We showed both theoretically and experimentally that there are cases where the variances of the retrospective distributions (distributions after selection) become larger than those of the chronological distributions (distributions before selection). The direction of variance change depends on the shape of chronological distributions, primarily on the skewness of the distributions (positive skew for increasing the variance and negative skew for decreasing the variance). The direction of variance changes can also be probed by the difference between the two selection strength measures S_KL^((2))-S_KL^((1)). Notably, we can demonstrate that there are cases where retrospective fitness variances are larger than chronological fitness variances even in the long-term limit, as shown by a cell population model in Appendix 2.

We now explain what kind of situations are usually premised when reduction of fitness variance is mentioned and clarify that, in our framework, we compare the fitness variances between chronological and retrospective distributions (L542-548). We also explain that a selection effect on fitness variance generally depends on fitness distribution and that a larger fitness variance in retrospective distribution is possible even in the long-term limit (L548-557).

Reviewer #2 (Public Review):

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

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

We appreciate the reviewer for the positive evaluation. We will reply to your specific comments below.

Reviewer #3 (Public Review):

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

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

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

We thank the reviewer for the comment on the point of which our explanation was insufficient in the original manuscript. Intrinsic stochasticity in interdivision time (generation time) is, in fact, critical for selection. For example, if a cell divides with a generation time shorter than the average due to stochasticity, this cell is likely to have more descendant cells in the future population on average than the other cells born at the same timing, even if the descendants follow identical statistics. Therefore, the properties of intrinsic stochasticity, including shapes of generation time distributions and transgenerational correlations, significantly affect the overall selection strength S_KL^((1)) [D] (and also S_KL^((2)) [D]). We now explain this important point in the Results section, referring to the analytical model in Appendix 2 (L327-334), and also in Discussion (L519-524).

Importantly, even when cell division processes seem purely stochastic, different states in some traits might underlie these variations in generation times. In such cases, evaluating h(x) and S_rel [X] can still unravel the correlations between the trait values and fitness. Especially, the relative selection strength S_rel [X]≔S_KL^((1) ) [X]/S_KL^((1) ) [D] extracts the correlation of the trait values in a given level of division count heterogeneity in each condition. We now clarify this important aspect of the framework in Discussion (L524-526).

When a cell population is composed of heterogeneous subpopulations each of which follows a distinct statistical rule, our framework evaluates the combined effects from the heterogeneous rules and the inherent stochasticity of each subpopulation. Untangling these two contributions is generally challenging unless we have appropriate markers for distinguishing the subpopulations. However, when the subpopulations follow significantly distinct statistics, the division count distribution should become skewed or multimodal, and the difference between the two selection strength measures S_KL^((2) ) [D]-S_KL^((1) ) [D] can suggest the existence of such subpopulations. Therefore, detailed analyses using all the selection strength measures and the fitness landscapes can provide insights into cell populations’ internal structures and selection.

We now explain the effect of inherent stochasticity in generation times (L327-334 and L519-524) and discuss how we can probe the existence of subpopulations based on the selection strength measures (L508-512). Please also refer to our reply to the comment 3 of reviewer #1.

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

We dealt with the situation of 2) in this model. As noted by the reviewer, we can calculate the chronological and retrospective mean fitness and the population growth rate by a simpler individual-based age-structured population model (see ref. 10, for example). However, applying this framework to this model can clarify the utility of the cumulant generating function, the meaning of the differences between these fitness measures, and the effect of statistical properties of intrinsic stochasticity on long-term growth rate and selection. Therefore, we kept this model in Appendix 2 (the section is moved from Supplemental Information) with additional clarification of our motivation for analysis and the implication of the results.

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

We thank the reviewer for an insightful comment. It is true that the duration of observation influences the results. In the regrowing experiments with E. coli, we aimed to compare the two cell populations regrowing from different stages of the stationary phase. Therefore, it is appropriate to fix the time windows between the two conditions. Even though a significant fraction of cell lineages remains undivided, the regrowing cells already divide several times within this time window. Therefore, the results are valid if we compare and discuss the selection levels in this time scale. However, clarification of the selection in the longer time scales requires a more detailed characterization of lag time distributions under both conditions.

We now clarify the range of validity of the results and the limitations on prediction for the long-term selection without knowing the details of the lag time distributions in Discussion (L536-539).

Author Response

Reviewer #2 (Public Review):

There is emerging evidence that connexin43 hemichannels localized to mitochondria can influence their function. Here the authors demonstrated using an osteocyte cell model that connexin43 is localized to mitochondria and that this is enhanced in response to oxidative stress. Several lines of evidence were presented showing that mitochondrial connexin43 forms functional hemichannels and that connexin43 is required for optimal mitochondrial respiration and ATP generation. These aspects were major strengths of the study.

The authors also show that connexin43 is recruited to mitochondria in response to oxidant stress, as a cell protective mechanism. This was primarily done using hydrogen peroxide to generate oxidant stress; primary osteocytes from Csf-1+/- mice, which are prone to Nox4 induced oxidant stress, also show enhanced mitochondrial connexin43 when compared with wild type osteocytes.

Several approaches were used to demonstrate that connexin43 interacts with the ATP synthase subunit, ATP5J2, suggesting a direct role for connexin43 in the control of ATP synthesis by mediating mitochondrial ion homeostasis. Several experiments were done using a series of pHluorin fusion protein constructs as a proton sensor, these experiments hint at a potential role for connexin43 in regulating H+ permeability to support ATP production. However, the effects of inhibiting connexin43 on pH were modest, suggesting that additional roles for mitochondrial connexin43 in ATP generation should be considered.

Thank you for your positive and thoughtful comments. We agree that additional roles for mitochondrial Cx43 may be possible. As an example, we consider that there may be a change in the stability of ATP synthase that occurs after mtCx43 deficiency. This and other possible roles of mtCx43 ought to be investigated in the future.

Reviewer #3 (Public Review):

This manuscript should be of broad interest to readers not only in the field of gap junction (GJ) mediated cell-to-cell communication but also to scientists and clinicians working on the function of mitochondria and metabolism. Their data elucidates a new function of Cx43 in regulating the energy (ATP) generation of mitochondria, e.g., under oxidative stress.

The canonical function of gap junctions is in direct cell-to-cell communication by forming plasma membrane traversing channels that electrically and chemically connect the cytoplasms of adjacent cells. These channels are assembled from connexin proteins, connexin 43 (Cx43). However, more recently new, non-canonical cellular locations and functions of Cx43 have been discovered, e.g. mitochondrial Cx43 (mtCx43). However, very little is known about where Cx43 transported into mitochondria is derived from, how Cx43 is transported into mitochondria, where it is located in mitochondria, in which form Cx43 is present in mitochondria, (polypeptides, hemi-channels (HCs), complete GJ channels), and what the function of mtCx43 is. The authors addressed the latter question. The authors provide convincing evidence that mtCx43 modulates mitochondrial homeostasis and function in bone osteocytes under oxidative stress. Together, their study suggests that mtCx43 hemi-channels regulate mitochondrial ATP generation by mediating K+, H+, and ATP transfer across the mitochondrial inner membrane by directly interacting with mitochondrial ATP synthase (ATP5J2), leading to an enhanced protection of osteocytes against oxidative insult. These findings provide important information of a role of Cx43 functioning directly in mitochondria and not at the canonical location in the plasma membrane. While most of the functional assays presented in Figures 2-8 appear solid, the mitochondrial localization of Cx43, its translocation into mitochondria under oxidative stress, and its configuration as hemi-channels (Figure 1) is less convincing. I have five general comments that should be addressed:

1) This study was performed in MLO-Y4 osteocyte cells. Is the H2O2 induced increase of mitochondrial Cx43 MLO-Y4 cell type or osteocyte specific, or is Cx43 playing a more general role in mitochondrial function, e.g. under oxidative stress? Osteoblasts such as MC3T3-E1 and MG63, and many other cell types endogenously express Cx43, and oxidative stress is a general physiological stressor, not only for osteocytes and bone cells. Attending to this question would address the generality of the findings for mitochondrial function.

We thank the reviewer for bringing up these valid points; seeing the phenotype displayed in secondary cell types, such as osteoblasts, would be of great relevance and interest. To address this, we conducted new experiments on MC3T3-E1 cells (Figure 1-figure supplement 2). After 2 hrs of H2O2 treatment, Cx43 accumulated on the mitochondria, marked by Mitotracker. Statistical analysis also showed a significant increase of the localization between Cx43 and Mitotracker (Figure 1-figure supplement 2B). The colocalization coefficient is higher in the Ctrl group in MC3T3-E1 cells when compared with the MLO-Y4 Ctrl group, indicating a different response level in other cell lines. Osteoblasts seemed to be more sensitive to redox interference. Overall, proving the point that under oxidative stress, mtCx43 may display a similar phenotype, across multiple cell lines, although the degree of sensitivity may differ.

2) The images of MLO-Y4 cells (Figure 1A) and the primary osteocytes isolated from Csf-1+/- and control mice (Figure 8) do not show visible gap junctions. I guess this is due to the fact that slides were stained with the Cx43(E2) antibody. I feel, staining of these cells in addition with the Cx43(CT) antibody would be helpful to get a better understanding on the distribution of Cx43 in gap junctions and undocked/un-oligomerized Cx43 in these cells.

Thank you for the suggestion. To get a better understanding of the distribution of Cx43, either in GJ or HC form, we performed additional experiments in MLO-Y4 cells using the Cx43(CT) antibody and data are shown below. With Cx43(CT) staining, we observed more signals in the cells and on the plasma membrane. After H2O2 treatment, we observed increased and stronger signals localized on the mitochondria compared with the untreated control group. Stronger signals observed in the plasma membrane indicate the gap junction stained by Cx43(CT) antibody.

3) The images of cells presented in Figure 1A are quite fussy. No mitochondria are visible, and the Cx43 staining is hazy and does not localize to any subcellular structures. Also, it is not clear if the higher resolution image presented in Figure 1C actually represents a mitochondrion. A good DIC image, or co-staining with another mitochondrial marker such as MitoTracker (as shown in Figure 4-S1) would make the localization and translocation of Cx43 into mitochondria upon oxidative stress more convincing. This is especially important as the translocation, although statistically significant, increases only by about 10% or less (Figure 1B). Such a small difference (also represented in the Western analyses presented in Figure 1D) could easily be artefactual, depending on how the correlation coefficient was generated. Of note in this respect is that control cells in Figure 1A appear larger (compare the size of the nuclei) and are spread out more than the H2O2 treated cells. Better, more clear images would make the mitochondrial localization/translocation more convincing.

The reviewer made great points. To improve the image clarity, we redid the staining/imaging and determined the colocalization of SDHA and MitoTracker Deepred. The result (shown below) suggested that under normal conditions without H2O2 treatment, SDHA and MitoTracker merged perfectly, while after H2O2 treatment for 2 hrs, mitochondria became fragmented and the SDHA signal exhibited a more dotted pattern compared to the MitoTracker. Overall, we feel that MitoTracker represents the distribution of mitochondria better. SDHA is a subunit of mitochondrial complex II, and the images we presented in Figure 1C were captured from isolated mitochondria under a confocal microscope with SDHA and Cx43(CT) co-staining. Considering the specificity of SDHA (see images below), we believe the Cx43 signal we captured demonstrates the mitochondrial localization/translocation. After using MitoTracker as a mitochondrial marker and higher magnificent images, the correlation coefficient increased from 0.35 to 0.47, a 32% increment with statistical significance. As to the nuclei size, some cells indeed have smaller sizes, which may be affected by varied local cell density. The new images represented in Figure 1A are much more consistent in the nuclei size.

4) How pure are the mitochondria that were probed for Cx43 by Western shown in Figure 1D? The preparation method described is relatively simple, collecting the 10,000xg supernatant (here 9,000xg supernatant) as mitochondrial fraction. Is it possible that the Cx43 signal, at least in part, is derived from other, contaminating membranes, such as PM, Golgi, or ER? Testing the mitochondrial preparation by Western with marker proteins specific for these compartments would strengthen the author's results.

The reviewer made a great suggestion. To address this, we did a western blot to test the mitochondrial purity. Indeed, this method using centrifugation is simple, and as expected there were some contamination of ER (marked by PDI) and Golgi (marked by STX6). However, to further confirm the purity of the mitochondrial fraction, fluorescent dyes for mitochondria (MitoTracker Deepred), ER (ER-Tracker Blue-White), and nuclei (Hochest) were used. The organelle-specific dyes indicated most parts of the fraction were mitochondria. There were some contaminations with ER fragments and minimal nuclear contamination. Combining our western blot and immunofluorescence data, it can be concluded that our Cx43 signal is primarily derived from mitochondria.

5) The authors rely on previous studies to postulate that Cx43 in mitochondria forms hemichannels in their system, is localized in the inner membrane, and is oriented with the Cx43 C-termini facing the inter-membrane space (as schemed in Figure 8C). The authors use lucifer yellow (LY) dye transfer and carbenoxolone, but both are not hemi-channel specific probes. They are transferred by, and block GJ channels as well. Experiments, using hemi-channel specific probes would be more convincing. This is important, as the information cited is based on only two references (Boengler et al., 2009; Miro-Casas et al., 2009), and it still is highly unclear how a membrane protein that is co-translationally inserted into the ER membrane, then traffics through the Golgi to be inserted into the plasma membrane is actually imported into mitochondria and in which state (monomeric, hexameric). Why the Cx43(CT) specific antibody traverses the outer mitochondrial membrane and reaches the Cx43CT while the Cx43(E2) specific antibody is not described and clear either. Where are these mitochondria permeabilized with Triton X-100 as described in M&M?

We edited the Methods section. We did not use Triton X-100 to permeate mitochondria. PMP appeared to preserve mitochondrial inner membrane integrity allowing us to assess the localization of Cx43(CT) antibody on mitochondria. We showed these new immunofluorescence images in Figure 5- figure supplement 2. PMP used as a plasma membrane permeabilizer has a 6x affinity with MOM compared with MIM. Meanwhile, no Cx43(E2) Ab signal was detected in mitochondria, suggesting the extracellular loop of Cx43 faces the matrix and cannot be accessed by Cx43(E2) antibody.

The translocation of Cx43 to mitochondria was reported to involve the chaperone Hsp90-dependent TOM complex pathway (Rodriguez-Sinovas et al., 2006). After the translocation, if mtCx43 forms gap junctions in mitochondria is unclear. Lucifer yellow is widely used in hemichannel-mediated dye uptake or gap junction-mediated dye transfer. In our case, considering the channel orientation, mtCx43 should form hemichannels, and Cx43(CT) Ab could be used as a specific Cx43 HCs blocker like the study reported in cardiomyocytes (Lillo et al., 2019).

Author Response:

Reviewer #1 (Public Review):

Here, Servello et al explore the role of temperature and the temperature-sensing neuron AFD in promoting protection against peroxide damage. Unlike many other environmental threats, peroxide toxicity is expected to be temperature-dependent, since its chemical reactivity should be enhanced by higher temperatures. The authors convincingly and rigorously show that transient exposure to 25C, a condition of mild heat stress in C. elegans, activates animals' defenses against peroxides but potentially not other agents. Interestingly, this response requires the temperature-sensing AFD neurons, though whether temperature-dependent AFD activity is itself involved in this regulation is not explored. Further, the authors find that temperature regulates AFD's expression of the insulin ins-39 and provide evidence supporting the idea that repression of ins-39 at 25C contributes to enhanced peroxide defense. The authors use transcriptomic approaches to explore gene expression changes in animals in which AFD neurons are ablated, providing evidence that the FoxO-family transcription factor DAF-16 potentiates AFD signaling. However, because AFD ablation triggers effects broader than transient 25C exposure, the significance of these findings for temperature-dependent peroxide defense is somewhat unclear. Additionally, the possibility that DAF-16 (as well as another protective factor, SKN-1) function in parallel to temperature stress is consistent with many of the results shown but is not as thoroughly considered. Together, these studies identify a fascinating example of pre-emptive threat response triggered by the detection of a potentiator of that threat, a phenomenon they term "enhancer sensing." While some predictions of the specificity of this phenomenon remain untested, the paper provides intriguing insight into the potential mechanisms by which it may occur.

Major issues:

The dependence of the enhancer-sensing phenomenon on AFD leads the authors to conclude that the 25C stimulus is sensed by AFD itself, but this needs to be directly tested. To do this, they could ask whether tax-4 function is required in AFD, or use mutants in which AFD's thermosensory function is compromised.

We thank the reviewer for suggesting these experiments. As requested, we determined whether previously identified mechanisms for temperature perception by the AFD neurons were required for the temperature-dependent regulation of peroxide resistance using gcy-18 gcy-8 gcy-23 triple mutants and the respective single mutants. The findings from the new experiments lead us to conclude that temperature perception by AFD via the GCY-8, GCY-18, and GCY-23 receptor guanylate cyclases, which are exclusively expressed in the AFD neurons, contributes to the temperature-dependent regulation of peroxide resistance in C. elegans. These experiments are detailed in the following new paragraph in the results section:

“Last, we determined whether previously identified mechanisms for temperature perception by the AFD neurons were required for the temperature-dependent regulation of peroxide resistance. The AFD neurons sense temperature using receptor guanylate cyclases, which catalyze cGMP production, leading to the opening of TAX-4 channels (Goodman and Sengupta, 2019). Three receptor guanylate cyclases are expressed exclusively in AFD neurons: GCY-8, GCY-18, and GCY-23 (Inada et al., 2006; Yu et al., 1997) and are thought to act as temperature sensors (Takeishi et al., 2016). Triple mutants lacking gcy-8, gcy-18, and gcy-23 function are behaviorally atactic on thermal gradients and fail to display changes in intracellular calcium or thermoreceptor current in the AFD neurons in response to temperature changes (Inada et al., 2006; Ramot et al., 2008; Takeishi et al., 2016; Wang et al., 2013; Wasserman et al., 2011). We found that when grown and assayed at 20°C, gcy-23(oy150) gcy-8(oy44) gcy-18(nj38) triple null mutants survived 43% longer in the presence of tBuOOH than wild-type controls (Figure 3J). In contrast, at 25°C, the gcy-23 gcy-8 gcy-18 triple mutants showed a 12% decrease in peroxide resistance relative to wild-type controls (Figure 3K). Therefore, the three AFD-specific receptor guanylate cyclases influenced the temperature dependence of peroxide resistance, lowering peroxide resistance at 20°C and slightly increasing it at 25°C. At 20°C, the gcy-8(oy44), gcy-18(nj38), and gcy-23(oy150) single mutants increased peroxide resistance by 10%, 51%, and 21%, respectively, relative to wild-type controls (Figure 3L). Therefore, each of the three AFD-specific receptor guanylate cyclases regulates peroxide resistance. We conclude that temperature perception by AFD via GCY-8, GCY-18, and GCY-23 enables C. elegans to lower their peroxide resistance at the lower cultivation temperature.”

The enhancer-sensing model is fascinating, but as it stands it is somewhat oversold. The authors could tone down the writing, indicating that this model is suggested rather than shown. Alternatively, they could more carefully test some of its predictions - for example by exploring the response to other threats (e.g. some of the toxicants described in Fig. S5) at 20C and 25C in WT and AFD-ablated animals.

We edited the manuscript and expanded the manuscript’s discussion to address these concerns as well as similar concerns from reviewer #3. In the paper we show that the regulation of the induction of H2O2 defenses in C. elegans is coupled to the perception of temperature (an inherent enhancer of the reactivity of H2O2). To understand the significance of this finding in an evolutionary context, and to explain why such a regulatory system would evolve, we introduced in the discussion a new conceptual framework, “enhancer sensing,” and devoted a section of the discussion to demonstrating that the phenomenon that we observed could not be adequately explained by existing frameworks used to understand the evolutionary origins of the regulatory systems for defense responses.

We now realize that we did not sufficiently and clearly explain the scope for the criterion for establishing a phenomenon represents enhancer sensing, leading to incorrect predictions by reviewer’s 1 and 3 about (a) whether what we observed in C. elegans is an instance of enhancer sensing (or more proof is needed) and (b) what the enhancer sensing model for the coupling of temperature perception to H2O2 defense would predict about how temperature and the AFD neurons would affect resilience to other chemicals. We regret failing to adequately explain the model’s scope and predictions and believe that we have now explicitly addressed the scope of what constitutes enhancer sensing and the predictions of the model. In particular, we previously did not spell out (a) the distinction between the enhancer sensing strategy and the mechanistic implementation of that strategy; and, importantly, (b) we did not discuss what the enhancer sensing strategy coupling temperature perception to H2O2 defense in C. elegans predicted (and did not predict) about whether a similar strategy would be expected to be used by C. elegans to deal with other temperature-dependent threats. We now address these issues in two new paragraphs in the discussion that read:

“We show here that C. elegans uses an enhancer sensing strategy that couples H2O2 defense to the perception of high temperature. We expect this strategy’s output (the level of H2O2 defense) to provide the nematodes with an evolutionarily optimal strategy across ecologically relevant inputs (cultivation temperatures) (Kussell and Leibler, 2005; Maynard Smith, 1982; Wolf et al., 2005). This strategy is implemented at the organismic level through the division of labor between the AFD neurons, which sense and broadcast temperature information, and the intestine, which responds to that information by providing H2O2 defense (Figure 9D). Ascertaining that C. elegans relies on this enhancer sensing strategy does not depend on the temperature information broadcast by AFD exclusively regulating defense responses to temperature-dependent threats, because the regulation of defenses towards temperature-insensitive threats could affect defenses towards temperature-dependent threats; for example, suppressing defenses towards a temperature-insensitive threat would be beneficial if those defenses interfered with H2O2 defense or depleted energy resources contributing to H2O2 defense.

As with any sensing strategy, enhancer sensing strategies are more likely to evolve when sensing is informative and responding is beneficial. In their natural habitat, C. elegans encounter many environmental chemicals that, like H2O2, are inherently more reactive at higher temperatures. It will be interesting to determine the extent to which C. elegans uses enhancer sensing strategies coupling temperature perception to the induction of defenses towards those chemicals, and whether those strategies rely on temperature perception and broadcasting by the AFD neurons. We expect that sensing strategies regulating defense towards those chemicals would be more likely to evolve when those chemicals are common, reactive, and cause consequential damage.”

We note that our ability to predict survival to other toxicants, such as those that trigger specific gene-expression responses that are AFD-dependent but are unaffected between 20C and 25C (as proposed by the reviewer), is limited not only by our lack of knowledge about the specific mechanisms that protect worms from those toxicants, but also by our lack of knowledge about whether defense towards hydrogen peroxide interferes (or synergizes) with defense towards each of those toxicants and whether defense towards those toxicants interferes (or synergizes) with H2O2 defense. We therefore think that those experiments would be better addressed in future studies.

The role of ins-39 remains somewhat speculative. Fig 4F shows that ins-39 mutants have a reduced induction of peroxide defense, but it seems that this could be the result of a ceiling effect. The authors' model predicts that overexpression of ins-39, particularly at 25C, should sensitize animals to peroxide damage, a prediction that should be tested directly. Further, the authors seem to assume that AFD is the relevant site of ins-39 function, but this needs to be better supported.

As requested by all three reviewers, we determined whether ins-39 gene expression in AFD was sufficient to lower peroxide resistance by restoring ins-39(+) gene expression only in the AFD neurons using the AFD-specific gcy-8 promoter. As predicted by the reviewer, these worms were more sensitive to peroxide than wild-type worms. The findings from this experiment lead us to conclude that expression of ins-39 in the AFD neurons was sufficient to regulate the nematode’s peroxide resistance. The new section reads:

“Next, we determined whether the INS-39 signal from AFD regulated the nematode’s peroxide resistance. The tm6467 null mutation in ins-39 deletes 520 bases, removing almost all the ins-39 coding sequence (Figure 5A), and inserts in that location 142-bases identical to an intervening sequence located between ins-39 and its adjacent gene. In nematodes grown and assayed at 20°C, ins-39(tm6467) increased peroxide resistance by 26% relative to wild-type controls (Figure 5F). To determine whether ins-39 gene expression in AFD was sufficient to lower peroxide resistance, we restored ins-39(+) expression only in the AFD neurons using the AFD-specific gcy-8 promoter (Inada et al., 2006; Yu et al., 1997) in ins-39(tm6467) mutants. Expression of ins-39(+) only in AFD eliminated the increase in peroxide resistance of ins-39(tm6467) mutants (Figure 5F). Notably, the peroxide resistance of the two independent transgenic lines was 28% and 30% lower than that of wild-type controls, likely due to overexpression of the gene beyond wild-type levels. We conclude that the gene dose-dependent expression of ins-39 in the AFD neurons regulated the nematode’s peroxide resistance.”

The temperature-shift experiments in figure 5G (formerly 4F) indicated that the effect on peroxide resistance at 20C of growth at 25C and of the ins-39 mutation were non additive. We interpreted this epistatic interaction to be due to action in a common pathway. It is possible that while growth at 25C increases the subsequent peroxide resistance at 20C, it could limit the nematodes’ subsequent peroxide resistance at 20C (beyond those peroxide-resistance increasing effects) when in combination with another intervention, even if those interventions acted via parallel mechanisms—a ceiling effect, as proposed by the reviewer. We favor the alternative interpretation, that the mechanisms act sequentially, because of our findings that ins-39 gene expression within AFD was lower at 25C than at 20C, leading us to propose the sequential model in figure 5H (formerly 4G).

Most of the daf-16 and skn-1 experiments are carried out in AFD-ablated animals, making the relevance of these findings for the 25C-dependent induction of peroxide defense somewhat unclear. As the authors show, AFD ablation causes much more extensive changes than transient 25C exposure, clearly seen in slope of the line in 3C. Further, unlike 25C exposure, AFD ablation is a chronic and non-physiological state. It would be useful for the authors to be cautious in their interpretation of these findings and to be clearer about how strongly they can connect them to the "enhancer sensing" phenomenon. Along these lines, the potentiation idea could be toned down a bit. Much of the data is consistent with parallel function for daf-16 (and skn-1) - for example, Fig 5C indicates additive effects of daf-16 and 25C exposure; 6C shows that AFD ablation still has a clear effect on peroxide sensitivity in the absence of both daf-16 and skn-1; and Fig S8a shows that much of the transcriptional response to AFD ablation (along PC1) is intact in daf-16 animals.

We have made several adjustments in the text to address these concerns. As the reviewer noted, the experiments with skn-1 were performed only in AFD ablated worms. We have renamed the section heading to “SKN-1/NRF and DAF-16/FOXO collaborate to increase the nematodes’ peroxide resistance in response to AFD ablation” to make that clear.

In contrast, the peroxide resistance experiments with daf-16 were done also in worms grown at 25C and then shifted to 20C during the peroxide resistance assay. The connection of daf-16 with the temperature dependent regulation of peroxide resistance was established in temperature shifts experiments in daf-16 single mutants (Figure 6C, formerly 5C) and in transgenic worms rescuing the daf-16 mutant only in the intestine (Figure 6F). In the revised text we make it clearer that the effect of the daf-16 mutation is bigger when the nematodes are shifted from 25C to 20C: “The daf-16(mu86) null mutation decreased peroxide resistance in nematodes grown at 25°C and assayed at 20°C by 35%, a greater extent than the 21% reduction in peroxide resistance induced by that mutation in nematodes grown and assayed at 20°C (Figure 6C).”

As the reviewer noted, daf-16 and skn-1 have a role in peroxide resistance when the AFD neurons are not ablated (albeit a smaller one than when those neurons are ablated). We have made several changes and additions to the text to make that explicit. Most notably, the revised last paragraph of the SKN-1 section now reads: “We propose that when nematodes are cultured at 20°C, the AFD neurons promote signaling by the DAF-2/insulin/IGF1 receptor in target tissues, which subsequently lowers the nematode’s peroxide resistance by repressing transcriptional activation by SKN-1/NRF and DAF-16/FOXO. However, this repression is not complete, because both daf-16(mu86) and skn-1(RNAi) lowered peroxide resistance at 20°C when the AFD neurons were present. It is also likely that DAF-16 and SKN-1 are not the only factors that contribute to peroxide resistance in AFD-ablated nematodes at 20°C, because AFD ablation increased peroxide resistance in daf-16(mu86); skn-1(RNAi) nematodes, albeit to a lesser extent than in daf-16(+) or skn-1(+) backgrounds.”

The potentiation idea was specific to the effects of DAF-16 on gene expression. As the reviewer noted, much of the transcriptional response to AFD ablation is intact (albeit reduced in magnitude) in AFD-ablated daf-16 mutants, leading to a shift in the PC1 score for the mutant. At the level of the expression of individual genes, we quantified those effects in Figure 8G (formerly 7D). When we did the RNAseq experiments we had expected that lack of daf-16 would eliminate either all the changes in gene expression induced by AFD ablation or eliminate those changes for a subset of genes. Instead, what we found was much more subtle, and unexpected: the size of the gene expression change induced by AFD ablation was reduced by the daf-16 mutation, and that reduction was systematic. Specifically, we found that the bigger the change in gene expression induced by AFD ablation, the bigger the effect of daf-16 in the AFD ablated animals (that is, potentiation), leading to a change in the slope in the regression line in Figure 8G. We revised the paper to ensure we only used the word potentiation in this context (gene expression), even though formally DAF-16 also potentiated the effects of AFD ablation (and temperature shift from 25C to 20C) on peroxide resistance.

Reviewer #3 (Public Review):

This paper offers novel mechanistic insights into how pre-exposure to warm temperature increases the resistance of C. elegans to peroxides, which are more toxic at warmer temperature. The temperature range tested in this study lies within the animal's living conditions and is much lower than that of heat shock. Therefore, this study expands our understanding of how past thermosensory experience shapes physiological fitness under chemical stress. The paper is technically sound with most experiments or analyses carried out rigorously, and therefore the conclusions are solid. However, it challenges our current understanding of the role of the C. elegans thermosensory system in coping with stress. The traditional view is that the AFD thermosensory neuron is activated upon sensing temperature rise, and that temperature sensation through AFD positively regulates systemic heat shock response and promotes longevity in C. elegans. Thus, it is quite unexpected that AFD ablation activates DAF-16 and improves peroxide resistance. It also appears counterintuitive that genes upregulated at 25 degrees overlap extensively with those upregulated by AFD ablation at 20 degrees. I feel that it is premature to coin the term "enhancer sensing" for such a phenomenon, as their work does not rule out the possibility that AFD ablation increases resistance to other stresses that are independent of temperature regarding their toxicity or magnitude of hazard. Additional work is necessary to clarify these issues.

1. Whether the role of AFD in inhibiting peroxide resistance is related to AFD activity needs further clarification. AFD activity depends on the animal's thermosensory experience. As animals in this study are maintained at 20 degrees unless indicated specifically, the AFD displays activities starting around 17 degrees and peaks around 20 degrees. Under such condition, the AFD displays little or no activity to thermal stimuli around 15 degrees. It will be important to test whether cultivation of animals at 20 degrees improves peroxide resistance at 15 degrees, compared to 15 degrees-cultivation/15 degrees peroxide testing. The authors should also test whether AFD ablation further improves survival under peroxides at 15 degrees for animals grown at 20 degrees, whose AFD should show little or no activities at 15 degrees.

The reviewer raises an interesting point about the relation between the mechanisms that determine AFD activity in response to temperature and those that enable AFD to regulate peroxide resistance. In the revised manuscript we tested whether known mechanisms enabling AFD to sense changes in temperature acutely (receptor guanylate cyclases GCY-8, GCY-18, and GCY-23) played a role in the temperature dependence of peroxide resistance. We found that they did, as detailed in our response to reviewer #1’s point 1.

As noted by reviewer #2 in their point 1, and in our reply to that comment (and in a new discussion paragraph in the revised manuscript), the relationship between the known mechanisms the acutely regulate the activity of AFD in response to temperature and the mechanisms by which constant cultivation temperature regulates gene expression in AFD (and therefore the expression of peroxide resistance regulating signals like INS-39) is not well understood. Therefore, it is difficult to predict which temperatures will cause induction of peroxide defenses via AFD-dependent mechanisms, or via other mechanisms. While we agree with the reviewer that it will be interesting to characterize the extent to which other cultivation temperatures besides 25C lead to increased peroxide resistance at lower temperatures (including the proposed shifts from 20C to 15C), we think that those questions will be better addressed in future studies.

2. The importance of the thermosensory function of AFD should be verified. In the current study, the tax-4 mutation was used to infer AFD activity, but tax-4 is expressed in sensory neurons other than AFD. In addition to AFD, AWC can sense temperature and it also expresses tax-4. Therefore, influence on AFD from other tax-4-expressing neurons cannot be excluded. On the other hand, ablation of AFD removes all AFD functions, including those that are constitutive and temperature-independent. Therefore, the authors should test the gcy-18 gcy-8 gcy-23 triple mutant, in which the AFD neurons are fully differentiated but completely insensitive to thermal stimuli. These three thermosensor genes are exclusively expressed in AFD. Compared to the tax-4 mutant that is broadly defective in multiple sensory modalities, this triple gcy mutant shows defects specifically in thermosensation. They should see whether results obtained from the AFD ablated animals could be reproduced by experiments using the gcy-18 gcy-8 gcy-23 triple mutant. The authors are also recommended to investigate ins-39 expression in AFD and profile gene expression patterns in the gcy-18 gcy-8 gcy-23 triple mutant.

We thank the reviewer for this suggestion. We have performed the requested experiments, as detailed in our response to reviewer #1’s point 1. Briefly, we determined found that gcy-18 gcy-8 gcy-23 triple mutants increased peroxide resistance at 20C but not at 25C, and found that the respective gcy single mutants affected peroxide resistance at 20C. In light of these findings, we concluded that temperature perception by AFD via GCY-8, GCY-18, and GCY-23 enables C. elegans to lower their peroxide defenses at the lower cultivation temperature.

3. The literature suggests that AFD promotes longevity likely in part through daf-16 (Chen at al., 2016) or independent of daf-16 (Lee & Kenyon, 2009). Whatever it is, various studies show that activation of AFD and daf-16 promote a normal lifespan at higher temperature, and AFD ablation shortens lifespan at either 20 or 25 degrees. Therefore, the finding that DAF-16-upregulated genes overlap extensively with those upregulated by AFD ablation is quite unexpected (Figure 5B). The authors should perform further gene ontology (GO) analysis to identify subsets of genes co-regulated by DAF-16 and AFD ablation, whether these genes are reported to be involved in longevity regulation, immunity, stress response, etc.

We thank the reviewer for this interesting comment about the complex mechanisms by which AFD regulates longevity. We note that AFD also has additional temperature-dependent roles in lifespan regulation, as Murphy et al. 2003 found that RNAi of gcy-18 increased lifespan in wild-type worms at 20C but not at 25C. Therefore, AFD-specific interventions can also be lifespan extending at 20C.

We performed WormCat analysis, which is similar to gene ontology, in Figure 8-figure supplement 2 (formerly Figure S8G), which we described in the results section: “we found that the extent to which AFD ablation affected the average expression of sets of genes with related functions (Higgins et al., 2022; Holdorf et al., 2020) was systematically lower in daf-16(mu86) mutants than in daf-16(+) nematodes (R_2 = 86%, slope = 0.67, _P < 0.0001, Figure 8—figure supplement 2).” Visual inspection of the plot and the very high coefficient of determination of 86% indicate that the size of the effect of AFD ablation on gene expression was systematically smaller when the contribution of DAF-16 to gene expression was removed.

In the revised manuscript we also moved the three panels quantifying the expression of DAF-16 targets and daf-16-regulated genes from the supplement to the main figure. One of those panels (Figure 8F) shows that genes upregulated by daf-16(+) in daf-2 mutants were disproportionally affected by lack of daf-16 in AFD-ablated worms, as we described in the results section: “In addition, in AFD ablated nematodes, lack of daf-16 lowered the expression of genes upregulated in a daf-16-dependent manner in daf-2(-) mutants (Murphy et al., 2003) to a greater degree than in unablated nematodes (Figure 8F).”

4. I feel that "enhancer sensing" is an overstatement, or at least a premature term that is not sufficiently supported without further investigations. The authors should explore whether AFD ablation or pre-exposure to warm temperature specifically enhances resistance to a stressor the toxicity of which is increased at higher temperature, but does not affect the resistance to other temperature-insensitive threats.

We edited the manuscript and expanded the manuscript’s discussion to address these concerns as well as similar concerns from reviewer #1. For clarity, we repeat much of our response to reviewer #1’s point 2 here, with the last paragraph of this response specific to this reviewer’s comment.

In the paper we show that in C. elegans the regulation of the induction of H2O2 defenses is coupled to the perception of temperature (an inherent enhancer of the reactivity of H2O2). To understand the significance of this finding in an evolutionary context, and to explain why such a regulatory system would evolve, we introduced in the discussion a new conceptual framework, “enhancer sensing,” and devoted a section of the discussion to demonstrating that the phenomenon that we observed could not be adequately explained by existing frameworks used to understand the evolutionary origins of the regulatory systems for defense responses.

We now realize that we did not sufficiently and clearly explain the scope for the criterion for establishing a phenomenon represents enhancer sensing, leading to incorrect predictions by reviewer’s 1 and 3 about (a) whether what we observed in C. elegans is an instance of enhancer sensing (or more proof is needed) and (b) what the enhancer sensing model for the coupling of temperature perception to H2O2 defense would predict about how temperature and the AFD neurons would affect resilience to other chemicals. We regret failing to adequately explain the model’s scope and predictions and believe that we have now explicitly addressed the scope of what constitutes enhancer sensing and the predictions of the model. In particular, we previously did not spell out (a) the distinction between the enhancer sensing strategy and the mechanistic implementation of that strategy; and, importantly, (b) we did not discuss what the enhancer sensing strategy coupling temperature perception to H2O2 defense in C. elegans predicted (and did not predict) about whether a similar strategy would be expected to be used by C. elegans to deal with other temperature-dependent threats. We now address these issues in two new paragraphs in the discussion that read:

“We show here that C. elegans uses an enhancer sensing strategy that couples H2O2 defense to the perception of high temperature. We expect this strategy’s output (the level of H2O2 defense) to provide the nematodes with an evolutionarily optimal strategy across ecologically relevant inputs (cultivation temperatures) (Kussell and Leibler, 2005; Maynard Smith, 1982; Wolf et al., 2005). This strategy is implemented at the organismic level through the division of labor between the AFD neurons, which sense and broadcast temperature information, and the intestine, which responds to that information by providing H2O2 defense (Figure 9D). Ascertaining that C. elegans relies on this enhancer sensing strategy does not depend on the temperature information broadcast by AFD exclusively regulating defense responses to temperature-dependent threats, because the regulation of defense towards temperature-insensitive threats could affect defenses towards temperature-dependent threats; for example, suppressing defenses towards a temperature-insensitive threat would be beneficial if those defenses interfered with H2O2 defense or depleted energy resources contributing to H2O2 defense.

As with any sensing strategy, enhancer sensing strategies are more likely to evolve when sensing is informative and responding is beneficial. In their natural habitat, C. elegans encounter many environmental chemicals that, like H2O2, are inherently more reactive at higher temperatures. It will be interesting to determine the extent to which C. elegans uses enhancer sensing strategies coupling temperature perception to the induction of defenses towards those chemicals, and whether those strategies rely on temperature perception and broadcasting by the AFD neurons. We expect that sensing strategies regulating defense towards those chemicals would be more likely to evolve when those chemicals are common, reactive, and cause consequential damage.”

We note, in the first of the new discussion paragraphs, that the existence of an enhancer sensing strategy is not contingent on whether the AFD neurons (that implement the temperature sensing and temperature-information broadcasting functions regulating peroxide defenses) also do not regulate defense responses to temperature-insensitive threats. For example, it may be beneficial to an animal facing high concentrations of environmental peroxides to suppress defense against a temperature-insensitive threat when those defenses are detrimental towards defense towards hydrogen peroxide. This could occur, for example, because there is an energetic trade off when mounting multiple defense responses, or because specific defenses towards temperature-insensitive threats interfere with peroxide defense. As we noted in our response to reviewer #1’s point 2, our ability to predict survival to threats other than H2O2 (including temperature-independent threats) is limited not only by our lack of knowledge about the specific mechanisms that protect worms from those threats, but also by our inability to predict the extent to which defenses towards different threats operate independently, constructively, or destructively with those that provide hydrogen peroxide defense. We therefore think that those experiments would be better addressed in future studies.

Author Response:

Reviewer #1 (Public Review):

This manuscript investigates the gene regulatory mechanisms that are involved in the development and evolution of motor neurons, utilizing cross-species comparison of RNA-sequencing and ATAC-sequencing data from little skate, chick and mouse. The authors suggest that both conserved and divergent mechanisms contribute to motor neuron specification in each species. They also claim that more complex regulatory mechanisms have evolved in tetrapods to accommodate sophisticated motor behaviors. While this is strongly suggested by the authors' ATAC-seq data, some additional validation would be required to thoroughly support this claim.

Strengths of the manuscript:

1) The manuscript provides a valuable resource to the field by generating an assembly of the little skate genome, containing precise gene annotations that can now be utilized to perform gene expression and epigenetic analyses. The authors take advantage of this novel resource to identify novel gene expression programs and regulatory modules in little skate motor neurons.

2) Cross-species RNA-seq and ATAC-seq data comparisons are combined in a powerful approach to identify novel mechanisms that control motor neuron development and evolution.

Weaknesses:

1) It is surprising that the analysis of RNA-seq datasets between mouse, chick, and little skate only identified 5 genes that are common between the 3 species, especially given the authors' previous work identifying highly conserved molecular programs between little skate and mouse motor neurons, including core transcription factors (Isl1, Hb9, Lhx3), Hox genes and cholinergic transmission genes. This raises some questions about the robustness of the sequencing data and whether the genes identified represent the full transcriptome of these motor neurons.

To address reviewer #1’s questions, we have generated RNA sequencing data with mouse forelimb MNs and re-analyzed the RNA-seq data using only the homologous MN populations (Figure 3) among different species. As a result, many genes (1038 genes) are commonly expressed in MNs in different species, including many known MN marker genes. In the result section, we have added the following:

“The evolution of genetic programs in MNs was investigated unbiasedly by comparing highly expressed genes in pec-MNs (percentile expression > 70) of little skate with the ones from MNs of mouse and chick, two well-studied tetrapod species. In order to compare gene expression with homologous cell types from each species, we performed RNA sequencing on forelimb MNs of mouse embryos at embryonic day 13.5 (e13.5) and wing level MNs of chick embryos at Hamburger-Hamilton (HH) stage 26–27…”

We have also compared our re-analysis with previous results in Figure 2–figure supplement 1, shown above. Most of the fin MN genes (21/24) are highly expressed in pecMNs (percentile > 70), consistent with the previous in situ experiments. In the Results we have added the following:

“Although the total number of DEGs are different from the previous data (592 vs. 135 genes in pec-MN DEGs), which might be caused by different statistical analysis with different reference genome, previous RNA-seq data based on de novo assembly and annotation using zebrafish was mostly recapitulated in our DEG analysis based on our new skate genome (21 out of 24 previous fin MN marker genes have the expression level ranked above 70th percentile in Pec-MNs; Figure 2‒figure supplement 1).”

2) The authors suggest based on analysis of binding motifs in their ATAC-seq data that the greater number of putative binding sites in the mouse MNs allows for a higher complexity of regulation and specialization of putative motor pools. This could certainly be true in theory but needs to be further validated. The authors show FoxP1 as an example, which seems to be more heavily regulated in the mouse, but there is no evidence that FoxP1 expression profile is different between mouse and skate. It is suggested in Fig.5 that FoxP1 might be differentially regulated by SnaiI in mouse and skate but the expression of SnaiI in MNs in either species is not shown.

We have added further discussion and data about differential expression of Foxp1 in mouse and little skate in Figure 5–figure supplement 16 and have discussed as follows:

“Foxp1, the major limb/fin MN determinant appears to be differentially regulated in tetrapod and little skate. Although Foxp1 is expressed in and required for the specification of all limb MNs in tetrapods, Foxp1 is downregulated in Pea3 positive MN pools during maturation in mice (Catela et al., 2016; Dasen et al., 2008). In addition, preganglionic motor column neurons (PGC MNs) in the thoracic spinal cord of mouse and chick express half the level of Foxp1 expression than limb MNs. Although PGC neurons have not yet been identified in little skate, we tested the expression level of Foxp1 using a previously characterized tetrapod PGC marker, pSmad. We observed that Foxp1 is not expressed in MNs that express pSmad (Figure 5‒figure supplement 3). Since there is currently no known marker for PGC MNs in little skate, our conclusion should be taken with caution.”

As for Snai1, in the revision we performed a motif enrichment analysis with an unbiased gene list where Snai1 didn’t show up. However, when we performed an RNA in situ hybridization experiment for Snai1 (Figure 5–figure supplement 3), we found that Snai1 is expressed in MNs of both mouse and little skate, but not in chick, which has been shown previously (Cheung et al., 2005). In order to examine the function of Snai1 in the regulation of Foxp1 expression, we ectopically expressed Snai1 in chick spinal cord by performing in ovo electroporation. However, we did not detect any changes in Foxp1. Instead we observed an increase in the number of neurons and abnormal MN exits from the spinal cord, which is the reminiscent of a previous observation (Zander et al., 2014). Although we did not detect any changes in Foxp1 expression, we cannot rule out the possibility that Snai1 regulates Foxp1 in mouse and little skate, which may require a gene knock out experiment. Because binding sites of Snai1 were not enriched in the new gene sets that we analyzed in the revision, we have not further discussed the Snai1 in the text.

3) In their discussion section the authors state that they found both conserved and divergent molecular markers across multiple species but they do not validate the expression of novel markers in either category beyond RNA-seq, for example by in situ or antibody staining.

We have added RNA in situ hybridization results in Figure 3C and Figure 3–figure supplement 1 and 2. Most of the genes were expressed in tissues in accordance with the sequencing results (6 out of 9 common MN genes; 4 out of 6 mouse specific genes; 5 out of 7 skate specific genes). Specifcally, Uchl1, Slc5a7, Alcam, and Serinc1 are expressed in MNs of all three species; Coch, Ppp1rc, Ctxn1, and Clmp are expressed in MNs of mouse but not in MNs of other species; Eya1, Etv5, Dnmbp, and Spint1 are expressed in MNs of skate but not in MNs of other species. In the result section, we have summarized the results as follow:

“These results were validated by performing RNA in situ hybridization in tissue sections on a subset of species-specific genes …”

Author Response

Reviewer #1 (Public Review):

Switching between epithelial and mesenchymal populations is an important stage for cancer growth and metastasis but difficult to study as the cells in this transition are rare. In this study Xu et al investigate changes the splicing regulator environment and changes in specific splice events by monitoring colon cancer cell populations that have epithelial and mesenchymal properties (so are potentially in transition) compared their epithelial partners. Using these potentially transitioning cells should reveal new insights into the causative changes occurring during EMT, a key life threatening step in colon cancer progression, and other cancers too.

The authors were trying to establish if changes in the splicing environment occurred between epithelial and quasi-mesenchymal cells and to what extent this is important for colon cancer in establishing gene expression programs and cell behavior related to metastasis. The take home message is that these more "plastic" mesenchymal cells are expressing the mesenchymal transcription factor ZEB1 and reducing expression of the epithelial splicing factor ESRP1 (as well as some other RBPs). The FACS analysis showing that over-expression of ESRP1 alone can switch cell population ratios is very clear and indicates that reduction of this RBP plays a key role in making cells more metastatic. The lentiviral overexpression of CD44s and NUMB2/4 had very dramatic effects on increasing metastatic cellular properties. The clinical stratification analysis of splice isoforms and ZEB1/ESRP1 expression was very informative for understanding what is happening in actual tumors. The methods used and results from these studies are likely to have an impact on understanding the gene expression changes that take place during EMT.

Strengths: The authors have used cell lines that model switching cells between epithelial and quasimesenchymal, based on expression of the markers Epcam (epithelial cell adhesion molecule expressed in epithelial cells) and CD44. The study utilizes shRNA-mediated knockdown and lentiviral overexpression of

ESRP1 and splice isoforms, and monitors endogenous mRNA splice isoforms by RNAseq and qRTPCR, protein isoforms by western, cell surface expression of EpCAM and CD44 using FACS and metastatic potential using a mouse model, and patient gene expression data from TCGA.<br /> Weaknesses: Some of the data here might be novel for colon cancer, but the roles of these RNA binding proteins and ESRP1 target exons are better known in other cancers. Both CD44 and NUMB are known ESRP1 targets already in cells undergoing plasticity (e.g. PMID: 30692202). RBM47 is already known to be downregulated in EMT and quaking upregulated (PMID: 28680090; PMID: 27044866). There is also a lot of literature on ESRP1 expression in cancer and EMT. This should be better discussed.

Out of the 3 references mentioned, 2 are already discussed in the submitted manuscript, while the third (Rokavec et al.) has now been added to the Discussion. As specified above, we never claimed to be the first to report on these RBPs and downstream AS targets. Unfortunately, it is not clear how the reviewer wants us to improve on these aspects (“should be better discussed” is rather vague) but we have now tried to extend the discussion relative to these issues in the revised manuscript.

Author Response

Reviewer #1 (Public Review):

The authors sought to establish canine tissue-specific organoids for propagation, storage and potential use in biomedical and translational medicine.

Strengths - The project is ambitious in aim, seeking to raise 6 tissue-specific, stem cell-derived organoid lines.

Weaknesses -

1) While the manuscript refers to stem cell lines, no evidence of progressive organoid morphogenesis has been shown from undifferentiated single stem cells or stem cell clusters. This omission makes it difficult to distinguish true organoids from surviving pieces of parental tissue that the authors actually include within their cultures. The authors infer that high order tissue complexity can be generated within in short term 3D cultures. For example, their kidney organoids contained glomeruli, renal tubules and a Bowman's'capsule. These remarkable findings contrast with a previous study by Chen et al 2019 that showed kidney organoids had restricted morphogenic capacity, forming only simple epithelial dome-like structures (Chen et al 2019). Although the Chen study was cited, the major differences in study findings were not discussed. In the current study, no compelling evidence is provided for the integrated assembly of the glomerular microvascular capillary network, the glomerular epithelial capsule and complex tubular epithelial collecting ducts, during organoid growth.

Thank you, clarification was made regarding the differences between Chen et al. 2019 and our organoids in our revision of the manuscript (Lines 445-447). The sentence regarding glomeruli, renal tubules, and Bowman's capsules was modified to specifically state the morphological resemblance our organoids have to these structures. We further clarified in the text that we are not stating these structures are complete (glomerular microvascular capillary) (Line 236), as our data are too preliminary to support this statement. However, in future publications we are excited to complete more in-depth characterization and investigation, along with functional assessment. To aid in characterization, three immunohistochemistry (IHC) antibodies were added to Figure 2.

2) The potential of the organoids for freezing, storage and re-culture is unclear from the data presented.

We did not present data regarding the re-culturing of organoids from this manuscript. However, we are working on additional publications which have already thawed and regrown multiple cell lines from this manuscript leading us to believe it is possible for all lines cultured in this manuscript. Further investigation into long term expression changes after thawing is warranted in future investigations.

3) Organoid capacity for regenerative growth in xenograft models has not been tested.

We did not investigate this in the current manuscript, from our understanding, manuscripts which describe a new organoid model typically do not utilize xenografts to confirm the regenerative growth capacity. The use of organoids in xenograft models is an exciting avenue to explore in the future.

4) Figure 4 lacks appropriate positive and negative tissue controls.

Please see Figure 5-figure supplement 1 for all negative control images. The tissues of origin in Figure 4 (now Figure 5) were used as positive controls for the antibody.

5) Gene expression differences between tissues and organoids are inadequately explained.

Our apologies for the lack of clarity of our original manuscript. Differences of gene expression between tissues and organoids were compared in the revised Results section of each organ. To better describe the differences seen between tissues and organoids, information was added to the discussion elaborating on cell types present and missing from our samples (Lines 315-321).

6) Methodological detail is sparse. It is not clear how tissue biopsies are obtained, what size they are and how they are processed for organoid preparation.

Our apologies. Information regarding biopsies was added in Experimental Procedures specifically in the Tissue collection section (Lines 509-510, 519-535). Additional details on protocols for organoid preparation and culturing were added to the Experimental Procedures and are cited in Gabriel et al. 2022 (Lines 535-548).

7) The manuscript as a whole is poorly focussed and difficult to follow. The introduction is repetitive with only weak relevance to the main experiments.

We appreciate the reviewer’s concern. To better focus this manuscript, we re-ordered the introduction to be more linear and improve the focus regarding the main experiments. We hope our revisions will satisfy the reviewer.

Appraisal - The lack of morphogenesis and xenograft data undermines confidence that the authors have achieved their aims. The above concerns are also likely to hamper utility of the methods for the scientific community.

We appreciate the listed concerns. These novel organoid models are not limited to applications pertaining to xenografts. Our aim was to develop novel organoid lines that we believe can be of use to a variety of fields including pharmacology, virology, and basic research. The testing of these organoids in xenograft models is outside the scope of the current manuscript.

Reviewer #2 (Public Review):

Zydryski et al. develop a comprehensive toolbox of organ-specific canine organoids. Building on previous work on kidney, urinary bladder, and liver organoids, they now report on lung, endometrium, and pancreatic organoids; all six organoid lines are derived from two canines. The authors attempt to benchmark these organoids via histological, transcriptomic, and immunofluorescence characterization to their cognate organs. These efforts are a welcome development for the organoid field, broaden the scope of use to studies with canine models, and seek to establish robust standards. The organ specific RNAseq dataset is also likely to be useful to other researchers working with the canine model.

A key methodological advance would appear to be that the authors culture these organ-specific organoids using a common cell culture media. This is not the typical protocol in the organoid field; however, the authors do not provide enough information in the manuscript to evaluate if this is a good choice. Furthermore, it is likely that the authors were successful because they included additional tissue components in the co-culture for the organoids which might have provided the necessary tissue specific cues, but the methodological details to reproduce this and the technical evaluation of this approach are missing.

This is an excellent point and details were added to the methods section to better explain the embedding process in our revision of the manuscript (Lines 519-532). Your hypothesis about the tissue-specific cues is very intriguing and something we should explore in the future. Previous publications have isolated ECM from the native tissue (Giobbe et al. 2019), this may be a similar mechanism as you stated.

The authors also directly compare the transcriptional responses of the organoids with the organs, but this is a challenging enterprise given that the organoid models do not incorporate resident immune cells and typically are composed only of epithelial cells. This lack of an 'apples to apples' comparison might explain why in many cases the organoids and organs are highly divergent; however, it could also be that the common cell culture media did not lead to specific maturation of cell types.

We agree, this manuscript aimed to derive epithelial organoids, and we acknowledge the lack of all cell types present in the tissues. The comparison was meant to identify similarities (epithelial cells) and the current limitations of the organoid model. We added to the Discussion, specifically the Insights into organ-specific genes section to further clarify this point (Lines 315-321).

Reviewer #3 (Public Review):

Zydrski et al. describe the generation and characterization of multiple adult tissues from canines. While canine derived organoids could potentially be advantageous over murine and human organoids, the novelty of generation and characterization is limited, as organoid systems are now being rapidly genetically editing using CRISPR technologies and modeled within immunocompetent environments. Certain points limit my enthusiasm.

First, the authors do not support the use of serum (FBS) in their media and why they include the same growth and differentiation factors across all tissue types.

We added a sentence to the Discussion (Canine organoids as biomedical models) to further clarify the reasoning behind the inclusion of the same growth factors for all tissue types

“The use of the same media composition lends itself to future applications of co-culture or use in assembloid models where multiple organoid lines are combined and continued growth in a shared media is required”

As this media is based on canine intestinal organoid media, the FBS was included in case of potential applications require the co-culture of intestinal organoids.

Second, while bulk RNA sequencing data shows similarity per certain genes to the corresponding tissue, there is a lack of detailed characterization of what passage these organoids were harvested and how they change over time. Do they become more stem like and are they genetically stable?

The passage number of samples when they were harvested are listed in Supplemental file 1. The question of being genetically stable is an excellent point regarding organoids. We have not examined that yet in these canine organoids; however, we can leverage previous publications regarding organoids and how they are genomically stable over time regarding chromosome number and base pair changes, we added these citations into the introduction (Line 48). However, this current manuscript focuses on the derivation and initial characterization; future work will focus on the re-growth, genetic stability, and functional assessment of canine organoid lines.

Third, it would be important to demonstrate that these organoids can be genetically manipulated or be exposed to drugs and how they might be beneficial over murine and human organoids.

The genetic editing of twelve organoid lines is outside the scope of this paper and we plan to include this element in future publications. We believe that the organoids can be useful for veterinary medicine as well as being an important model or human disease as canines typically better represent humans better than mice (Lines 28-34, 77-95).

Fourth, the organoid complexity is not clear and cannot be ascertained from bulk RNA sequencing- for example, do kidney organoids recapitulate canonical markers at the protein level of proximal tubules, distal convoluted tubules, etc. Are different lung cells represented (AT1/AT2/club) and what is the composition of these cells? Why are these cells selected for?

Thank you. We agree that bulk RNA sequencing has its limitations when it comes to heterogenous cell populations. This was meant to give a first insight into whether the organoid lines resemble their tissue of origin, the addition of single cell RNA-sequencing in the future is worth investigating.

Fifth, as the authors note, methodically these canine organoids have been developed before from other tissues. For these reasons, my enthusiasm is diminished and unfortunately many of the necessary experiments for further consideration appear out of the scope of the study.

Three of these organoid lines have previously been published in canines. However, the growth and characterization of three novel organoid lines is included in this manuscript, while typically a manuscript focuses on one novel organoid line. Furthermore, unique to this study is the multi-organ comparisons of expression across both tissues and organoids from the same animal, with a biological replicate being a related individual which is unique to this study. In human and murine field, the organoid media must be adjusted to each individual organ the stem cells are isolated from. We show that our media composition, which is similar to that which previously supported hepatic and intestinal canine organoids, can now support organoids from six different tissues, bringing a novel approach to the field. To our knowledge, this is the most comprehensive comparison across tissue types of canine organoids. Additionally we have not seen any literature of the comparison of six different organoid lines from the same individual, with a related biological replicate in any other species.

Author Response

Reviewer #1 (Public Review):

This study examines whether the D2 receptor antagonist amisulpride and the mu-opioid receptor antagonist naltrexone bias model-based vs model-free behavior in a well-established two-step task of behavioral control. The authors find that amisulpride enhances model-based choices, which is further supported by computational modeling of the data, revealing an increase in the relative contribution of model-based control of behavior. Naltrexon on the other hand had no reliable effect on model-based behavior.

Overall, this is a very nice study with many strengths, including the task and data analysis. A particular strength of the design is the combination of a between-subject drug administration protocol with two within-subject (baseline vs. drug) sessions. This reduces between-subject variability in baseline model-based vs model-free behavior and enhances the power to detect drug effects.

The introduction could do a better job articulating the rationale for testing the effect of these two specific drugs. Currently, the rationale is that both transmitter systems targeted by these drugs are involved in drug addiction, which is characterized by an imbalance in model-based vs. habitual control of behavior. This appears somewhat indirect.

Blood draws were used to determine serum levels for amisulpride and naltrexone but these data are not included as covariates in the analysis.

We thank the reviewer for the high acclaim of our study, and for the constructive comments to improve it. We acknowledge that the introduction did not motivate the main research goal of the manuscript clearly enough. We have now extended this section and provided further insight into our reasoning behind the study design. Beyond the involvement of opioid and dopamine promoting drugs in addiction, there is abundant evidence from experimental studies showing comparable effects of manipulating receptors of both systems in model-free processes such as reinforcement, and habit formation. Based on this overlap one may predict that both neurotransmitter systems disrupt habit formation in a similar fashion, and that blocking their respective receptors will improve the ability to behave in a model-based manner. However, as we now elaborate in the manuscript, an argument against this could be that disrupting model-free processes might not be enough to promote model-based behaviour, as such behaviour relies heavily on cognitive control. It is therefore especially interesting to compare opioid antagonists, that do not enhance cognitive function, with a D2 antagonist at a dosage that has been shown to increase cognitive control as well as increase the desire to exert cognitive effort.

This is expressed in the following paragraphs of the Introduction (p.2 §3 and p.3 §1):

“Opiates, psychostimulants, and most other drugs of abuse increase the release of dopamine along the mesolimbic pathway (Chiara, 1999; Koob & Bloom, 1988), a circuit that plays a central role in reinforcement learning (Schultz, Dayan, & Montague, 1997). On top of this, the reinforcing properties of addictive drugs also depend on their ability to activate the μ opioid receptors (Becker, Grecksch, & Kraus, 2002; Benjamin, Grant, & Pohorecky, 1993; Le Merrer, Becker, Befort, & Kieffer, 2009). This suggests that both the dopamine and the opioid systems might be particularly relevant in model-free reinforcement learning processes that drive the formation of habitual behaviour. Studies in rodents show that activating receptors of both systems across the striatum increases cue-triggered wanting of rewards (Peciña & Berridge, 2013; Soares-Cunha et al., 2016). Conversely, inhibition of both D1-type and D2-type of dopamine receptors (referred to as D1 and D2 from here on) as well as opioid receptors reduces motivation to obtain or consume rewards (Laurent, Leung, Maidment, & Balleine, 2012; Peciña, 2008; Soares-Cunha et al., 2016). This data raises the hypothesis that the drift towards habitual control is enabled by dopamine and opioid receptors via a common neural pathway. Recent work in humans provides some evidence in this direction, whereby systemic administration of opioid and D2 dopamine receptor antagonists causes a comparable reduction of cue responsivity and reward impulsivity (Weber et al., 2016) and decreases the effort to obtain immediate primary rewards (Korb et al., 2020). This suggests that when allocating control between the model-based and model-free system, dopamine or opioid receptor antagonists might comparatively disrupt model-free behavioural strategies and increase model-based behaviour. Yet, no study in humans has directly investigated this. Furthermore, disrupting habit formation might not in itself lead to increased model-based control, without either increasing the perceived value of applying cognitive control or making it easier to do so.”

We also mention the implications of this direct comparison of the two compounds in the Discussion (p.8 §1):

“Our findings provide initial evidence for a divergent involvement of the dopamine and opioid neurotransmitter systems in the shift between habitual and goal-directed behaviour. The lack of effects of naltrexone on the model-based/model-free trade-off also provides some support for the notion that simply disrupting neurobiological systems that subserve habitual behaviour might not be enough to increase goal-directed behaviour in this task. An increase in the model-based/model-free weight following amisulpride administration advocates for dopamine playing a decisive role in flexibly applying cognitive control to facilitate model-based behavior and highlights the specific functional contribution of the D2 receptor subtype.”

Reviewer #3 (Public Review):

I think this is an interesting study on an important topic. I agree that there is not enough research to understand how the dopaminergic system interfaces with goal-directed planning, and I like the focus on specific types of dopamine receptors. It is interesting that they seem to find a specific effect on just the dopamine antagonist. I also appreciate the clarity with which the authors describe this field of research and their results. However, I also feel that there are several concerns with this paper, both in terms of framing and in terms of the experimental design and analysis. For completeness, I must note that I am not a dopamine expert.

I felt that the introduction of the paper did not sufficiently motivate the focus on the comparison between neurotransmitters systems, and (for the dopaminergic system) the distinction between D1/D2 receptors. Why is the mapping between stability/flexibility and D1/D2 receptors important? How does this relate to model-based control? Why do the authors predict that model-based control would increase when D2 receptors are blocked? If the hypothesis is about contrasting the contribution of D1 and D2 receptors to goal-directed control, why did the authors not use antagonists directly targeting these two systems?

In addition, the predictions that are more explicit, for example, that blocking D2 receptors increases MB control by stabilizing goal-relevant information, are fairly specific. However, the current version of the two-step task is not amenable to testing such a specific hypothesis, because it doesn't allow us to measure the specific components of planning (e.g., maintaining goals, the representation of the structure, prospective reasoning). Moreover, MB control in this version of the two-step task is marked by flexibility, because it requires the agent to be sensitive to switching starting states.

The predictions for the opioid system are also lacking. Why are the authors targeting this system? Why are they comparing the effects of the D2 antagonist with the opioid agonist? Why do the authors predict that amisulpride should have a stronger effect than naltrexone? In my opinion, these predictions were not sufficiently laid out, which made it difficult to appreciate the authors' motivation to run the study.

We thank the reviewer for their critical take on the manuscript and for clearly pointing out the weaknesses in argumentation. In particular, we appreciate the reviewer’s comment on the lack of clarity in describing why the comparison of dopamine and opioid antagonists’ effects on MB/MF behaviour might be particularly interesting and why we focused on D2 and not D1 receptors. We now extended the introduction section to clarify our rationale for comparing these two compounds (p.2-3). In short, apart from the fact that both systems are implicated in addiction, there is also abundant experimental evidence from human and non-human animal studies that the two systems are involved in processes related to forming habitual responses to primary and secondary rewards. This suggests that blocking receptors of either system might comparatively affect the MB/MF trade-off by impairing model-free processes. We therefore proceeded to compare opioid and dopamine antagonists.

As we note, using D1 antagonists would likely be detrimental to cognitive control related processes, and therefore more likely to decrease model-based performance. We therefore chose to compare opioid antagonists to D2 receptor antagonists. Another important reason for comparing the effects of opioid and D2 dopamine antagonists is the reasoning that it is not clear whether blocking model-free processes is in itself enough to promote model-based behaviour, without boosting cognitive control related processes. Given the recent evidence for D2 antagonists increasing cognitive effort (Westbrook et al., 2020) and the proposed role of prefrontal D2 receptors in destabilising prefrontal representations (according to the dual state theory of prefrontal dopamine function proposed by Durstewitz & Seamans, 2008)) we reasoned that D2 receptor blockade might also boost the ability (or willingness) to keep the mapping between spaceships and planets online while making choices.

We incorporated these arguments in the revised Introduction (p.2-3):

“Opiates, psychostimulants, and most other drugs of abuse increase the release of dopamine along the mesolimbic pathway (Chiara, 1999; Koob & Bloom, 1988), a circuit that plays a central role in reinforcement learning (Schultz et al., 1997). On top of this, the reinforcing properties of addictive drugs also depend on their ability to activate the μ opioid receptors (Becker et al., 2002; Benjamin et al., 1993; Le Merrer et al., 2009). This suggests that both the dopamine and the opioid systems might be particularly relevant in model-free reinforcement learning processes that drive the formation of habitual behaviour. Studies in rodents show that activating receptors of both systems across the striatum increases cue-triggered wanting of rewards (Peciña & Berridge, 2013; Soares-Cunha et al., 2016). Conversely, inhibition of both D1-type and D2-type of dopamine receptors (referred to as D1 and D2 from here on) as well as opioid receptors reduces motivation to obtain or consume rewards (Laurent et al., 2012; Peciña, 2008; Soares-Cunha et al., 2016). This data raises the hypothesis that the drift towards habitual control is enabled by dopamine and opioid receptors via a common neural pathway. Recent work in humans provides some evidence in this direction, whereby systemic administration of opioid and D2 dopamine receptor antagonists causes a comparable reduction of cue responsivity and reward impulsivity (Weber et al., 2016) and decreases the effort to obtain immediate primary rewards (Korb et al., 2020). This suggests that when allocating control between the model-based and model-free system, dopamine or opioid receptor antagonists might comparatively disrupt model-free behavioural strategies and increase model-based behaviour. Yet, no study in humans has directly investigated this. Furthermore, disrupting habit formation might not in itself lead to increased model-based control, without either increasing the perceived value of applying cognitive control or making it easier to do so. Crucially, there are important differences in how each of the two neurochemical systems relate to cognitive control that is pivotal for model-based behaviour. Across a wide range of studies using various dosing schemes, opioid receptor antagonists did not have an effect on tasks that require cognitive control, such as working memory (Del Campo, McMurray, Besser, & Grossman, 1992; File & Silverstone, 1981; Volavka, Dornbush, Mallya, & Cho, 1979), sustained attention(Zacny, Coalson, Lichtor, Yajnik, & Thapar, 1994), or mathematical problem-solving (Del Campo et al., 1992) (see (van Steenbergen, Eikemo, & Leknes, 2019) for a review). Dopaminergic circuits, on the other hand, play a central role in higher cognitive functions and goal-directed behaviour (Brozoski, Brown, Rosvold, & Goldman, 1979). In particular, D1 dopamine receptors in the prefrontal cortex enable maintenance of goal-relevant information and working memory(Goldman-Rakic, 1997; Sawaguchi & Goldman-Rakic, 1991; van Schouwenburg, Aarts, & Cools, 2010; Williams & Goldman-Rakic, 1995), while the D2 dopamine receptor activity disrupts prefrontal representations(Durstewitz & Seamans, 2008). In support of this, decreased working memory performance was observed after blocking prefrontal D1, but not prefrontal D2 receptors (Arnsten, 2011; Sawaguchi & Goldman-Rakic, 1991; Seamans & Yang, 2004). In humans, systemic administration of D2 antagonism increased the ability to maintain and manipulate working memory representations (Dodds et al., 2009; Frank & O’Reilly, 2006) and increased the value of applying cognitive effort (Westbrook et al., 2020). This data suggests that blocking D2 receptors, in contrast to blocking opioid receptors, could further facilitate model-based behaviour through enabling or encouraging flexible use of cognitive control.”

Another important point that the reviewer stresses is that the two-step task we use does not allow us to draw any conclusions through which mechanisms amisulpride increases model-based behaviour. Although we base our hypothesis that D2 might promote model-based behaviour (on top of disrupting habit formation) on previous work showing D2 blockade increasing cognitive effort and the ability to manipulate working memory representations, we completely agree that our setup does not give any definite answers about which of these cognitive processes mediated the increase in model-based weights. In the discussion we try to interpret our findings in the context of the dual-state hypothesis framework and within the framework of striatal control of adaptive behaviour (p.8 §3-4), whereby we centre our argumentation around dopaminergic circuits that subserve one or the other mechanism.

We agree with the reviewer that the task requires a high degree of flexible planning and that the dual-state theory might not be enough to account for our effects. We mention this in the Discussion (p. 8 §3):

“The effects of D2 antagonism on model-based/model-free behaviour in our study can be interpreted within this [dual-state] framework to result from increased ability to maintain prefrontal representation of the mapping between the spaceships and the planets online. However, this is difficult to reconcile with the fact that model-based behaviour in dynamic learning paradigms, such as the one used here, also requires flexible updating of action values.”

We also elaborate on the general limitations of drawing inference about the underlying cognitive/computational mechanisms in the Discussion (p. 14 §2):

“Importantly, it should also be acknowledged that the behavioural setup in our study does not allow us to draw definite conclusions about the mechanisms that mediate amisulpride’s effects on model-based or model-free behaviour. For example, it is not clear whether amisulpride increases the perceived benefit of applying cognitive control, or whether it increases the participant’s ability to do so through various possible complementary processes, such as goal maintenance or planning abilities. Future studies should further elucidate the mechanistic contributions of dopamine receptors to the distinct coding and utilisation of task relevant representations (Langdon, Sharpe, Schoenbaum, & Niv, 2018; Stalnaker et al., 2019).”

Related to this, I felt that the introduction was a bit too quiet on the genetic markers. Their discussion in the results was a bit surprising, and it wasn't quite clear why the authors decided to investigate these interaction effects.

We appreciate this comment as we were quite uncertain ourselves on how much weight to give to those data. Previous research had indeed shown profound variability in MB/MF behaviour across genotypes related to baseline dopamine function. The main purpose of the genetic analysis was to control for potential baseline differences and to explore the drug genotype interactions. However, including the serum data as a covariate in analyses, as suggested by the other reviewers, made most results relating to the genetic analysis disappear, even when using less conservative priors that likely understate the variance of posterior distributions of group effects. We have therefore opted to keep coverage of the genetic data to a minimum, but still report the results and make the data available online for future studies.

I found some of the core results confusing. Most importantly, why does amisulpride make people less like to stay after a reward when the first-stage state is the same? When first-stage states repeat, both an MB agent and an MF agent will be more likely to stay after a reward. To me, this kind of behavior doesn't seem particularly model-based. Why does this behavior occur under amisulpride? I was surprised that the authors did not really address it.

We agree that these results have been somewhat difficult to reconcile. However, adding amisulpride serum levels to our analyses now allow us to get a better understanding. It seems that across both serum groups model-based behaviour was increased, however, only in the high serum group did we additionally observe increased exploration. We also note that increased exploration was related to a reduced effect of previous points in the first same state trials, whereas the interaction term (effect of previous points in diff vs. same state trials) was more strongly associated with the model-based weight. In the manuscript this is described in the results section and in the discussion.

The following text is included in the Results (p.6):

“We first observed that the more model-based choices the participants made, the more money they earned (r = 0.65, 95% CI [0.53, 0.76]). This serves as a validity check of the task, which was designed to make cognitive control pay off (literally)45. We then looked at how the model parameters relate to the random slopes from the behavioural analysis of staying behaviour and found that the participant-level (random effect) slope for the effect of previous points on staying behaviour in different vs. same first state trials was most strongly related to ω (d = 0.493, P < 10e-3) and negatively related to the inverse temperature parameter η (d = -0.328, P < 10e-3), and the slope for trials with same first states was mostly related to η (d = 0.822, P < 10e-3), and less so to ω (d = 0.235, P < 10e-3).”

The following text is included in the Discussion (p.8 §2):

“Interestingly, amisulpride also increased choice stochasticity parametrised by the softmax inverse temperature parameter. In a paradigm with two choice options, it cannot be definitively determined whether this indicates higher decision-noise or increased exploration of alternative choices. We can however speculate that increased decision noise would lead to overall detrimental effects on learning in both trial types with same and different consecutive first stage states, which we do not observe in our data. The effect on the choice stochasticity parameter was only present in participants with a higher effective dose75, suggesting that the effect was more likely to be post-synaptic. Similarly, in the same effective dose group, we found some evidence that amisulpride reduces response stickiness indicating increased switching between actions. This is well in line with a prominent model of the cortico-striatal circuitry implicating post-synaptic D2 receptors in exploration/exploitation65 and supported by empirical data. In animal studies, activation of D2 receptors was shown to lead to choice perseverance and more deterministic behaviour, whereas D2 receptor inhibition increases the probability of performing competing actions and increases randomness in action selection76. In humans, a recent neurochemical imaging study showed that D2 receptor availability in the striatum correlated with choice uncertainty parameters across both reinforcement learning and active inference computational modelling frameworks77. Increased choice uncertainty was also observed in a social and non-social learning tasks in a study using 800 mg of sulpiride, a dose that is known to exert post-synaptic effects54,78. We note, however, that the evidence for the difference in exploration between the low and high serum groups was not robust (p=0.066). Furthermore, it has been suggested that increased striatal dopamine is also related to tendency for stochastic, undirected exploration79,80, arising due to overall uncertainty across available options79 or through increasing the opportunity cost of choosing the wrong option68,71. This suggests that the same biological signature that leads to increased cognitive effort expenditure also promotes choice exploration. In line with this, both prior studies that investigated the effect of increasing dopamine availability with L-DOPA on model-based/model-free behaviour observed increase choice exploration as well as increased model-based behaviour (although in one it was only present in individuals with a higher working memory capacity)55,58.”

With regards to the design, it is unfortunate that the order of drug administration is not counterbalanced. As far as I understand, model-based control is always measured without a drug in the first session, and then with the drug (or placebo) in the second. The change between sessions is then tested for all three conditions. Of course, it is possible that the increase in model-based control in the amisulpride condition is only driven by the drug. However, given the lack of counterbalancing, it's also possible that amisulpride increases model-based control only after the experience with the task. That is, if the authors had counterbalanced the drug effect, they may have found that amisulpride had a different effect if it was administered in the first session. That would have changed their interpretation quite a bit! As it stands, they are unable to verify their (admittedly simpler) hypothesis that there is only a main effect.

We thank the reviewer for this comment. Indeed, a full within-subject design would have been statistically more powerful and would have enabled us to exclude the possibility that amisulpride’s effect on model-based behaviour is indirect. We have now included the following paragraph in the discussion that aims to highlight the limitation of not counterbalancing the drug administration (p.10):

“One of the strengths of our design is a baseline measure, and the fact that the participants were all introduced to the task under no administration, thus avoiding potential effects of the treatment on task training. Although this design allowed to reduce between-subjects variability, we cannot completely exclude order effects. Although unlikely, it is possible that the effects of the treatment that we observe come indirectly from the effects of the two drugs on either skill transfer from the previous session, or simply on the effect of the drugs on the part of the experiment that preceded the task. For instance, participants under amisulpride could be less tired from other tasks and therefore more willing to exert effort in the task presented here. Speaking against this is the observation that we found no differences in mood between amisulpride and placebo regardless of low or high serum levels.”

Author Response

Reviewer #1 (Public Review):

This study presents a series of experiments that investigate maternal control over egg size in honey bees (Apis mellifera). Honey bees are social insects in which a single reproductive female (the queen) lays all the eggs in the colony. The first set of experiments presented here explore how queens change their egg size in response to changes in colony size. Specifically, they show that queens have relatively larger eggs in smaller colonies, and that egg size changes when queens are transplanted into colonies of a different size (i.e. confirming that egg size is a plastic trait in honey bee queens). The second set of experiments investigates candidate genes involved in egg size determination. Specifically, it shows that Rho1 plays a role in determining egg size in honey bee queens.

In principle, we agree with this summary, although we find the experimental demonstration that perceived colony size affects egg size (first set of experiments) and the overall proteomic comparison of ovaries that produce small and large eggs (second set of experiments that indicate the upregulation of metabolism, protein transport, cytoskeleton organization, and a few other processes in large egg-producing ovaries) also important.

A strength of the study is that it combines both manipulative field (apiary) experiments and molecular studies, and therefore attempts to consider broadly the mechanisms of plasticity in egg size. The link between these two types of dataset in the manuscript, however, is not strong. While the two parts are related, the molecular experiments do not follow from the conclusions of the field experiments but rather run in parallel (both using the same initial treatments of queens from large v small colonies).

We would welcome suggestions on how to further strengthen the integration between the field experiments and our molecular studies. We sought to explore the molecular basis of the observed plasticity in reproductive behavior and thus focused on samples from the first set of experiments for our proteome comparisons, realizing that every additional field experiment could have entail a similar molecular follow-up. We attempted to bring molecular studies and field experiments back together with the RNAi-mediated knock-down of Rho1 in queens that produce eggs in differently-sized colonies under realistic apicultural conditions. There may be better, additional opportunities for a closer integration of molecular and field experiments, but we could not conceive of them.

Another strength of the study is the focus on social cues for egg size control in a social insect. Particularly interesting is data showing that queens suddenly exposed to the cues of a larger colony (even where egg-laying opportunities did not actually increase) will decrease their egg size, in the same way as queens genuinely transplanted to larger colonies. That honey bee queens can control their egg sizes in response to cues in the colony is not unexpected, given that queens are known to vary egg size based on the cell type they are laying into (queen, drone or worker cell). Nevertheless, it is interesting to show that worker egg sizes over time are also mediated by social cues.

We thank the reviewer for this positive assessment and want to highlight that this experiment not only controls for egg laying opportunities, but also for potentially greater resource availability in larger colonies. These results are therefore important for the key argument that egg size is actively regulated by honey bee queens.

A weakness of the study is that the consequence of egg size on egg development and survival in honey bees is not made clear. The assumption is that larger egg size compensates for smaller colonies in some way. Do smaller eggs (i.e. those laid in large colonies) fare worse in smaller colonies than they do in large colonies? Showing that the variation in egg size is biologically relevant to fitness is an important piece of the puzzle.

We agree that the consequences of egg size variation are important to address beyond our previously published data set and the benefits demonstrated in other contexts by other authors. However, to comprehensively resolve the consequences requires considerable additional experiments that exceed the scope of our current study, which is primarily focused on the causes of the queens’ reproductive plasticity.

Also, the relationship between egg number and egg size in honey bees remains rather murky. Does egg size depend at least in part on daily egg laying rate (which is sure to be greater in larger colonies)? The study makes an effort to explore this by preventing queens from laying for two weeks and then comparing their egg size when they resume to those that did not have a pause in laying. Although egg size did not vary between the groups in this case, it is unclear whether the same effect would be seen if queens had simply been restricted from laying at such high rates (e.g. if available empty brood cells had been reduced rather than removed entirely).

We agree that the relation between egg number and egg size is complicated. We have added more data that show that egg laying rates can be higher in larger colonies than in smaller colonies. We also report now that the egg size is negatively correlated to egg number, although not in all instances, which partially supports (and partially contradicts) our previous findings (Amiri et al. 2020). We have modified the discussion of our results to account for the additional results and point out the limitation of the experiment with caged queens. It is important to realize though that the queens were caged on comb and not restricted in typical, small queen cages that are used for queen transport. It is not clear whether this treatment resulted in a downregulation of the reproductive efforts and/or the resorption of eggs.

Overall this study makes new contributions to our understanding of maternal control over egg size in honey bees. It provides stepping stones for further investigation of the molecular basis for egg size plasticity in insects.

We agree that we could not resolve everything in this study and that more investigations are needed.

Reviewer #2 (Public Review):

This paper builds on recent work showing that honeybee queens can change the size of the eggs they lay over the course of their life. Here the authors identified an environmental condition that reversibly causes queens to change their egg sizes: namely, being in a relatively small or large colony context. Recently published work demonstrated the existence of this egg size plasticity, but it was completely unknown what signaled to the queen. In a series of simple and elegant experiments they confirmed the existence of this egg size plasticity, and narrowed down the set of environmental inputs to the queen that could be responsible for signaling the change in the environment. They also began the work of identifying genes and proteins that might be involved in controlling egg size. They did a comparative proteomic analysis between small-egg-laying ovaries and large-egg-laying ovaries, and then selected one candidate gene (Rho1). They showed that it is expressed during oogenesis, and that when it is knocked down, eggs get smaller.

This is a good summary, although we think that it is fair to add that the expression of Rho1 is specific to the egg growth stage, and that we found an almost perfect correlation of Rho1 mRNA levels and egg size in two separate experiments (in addition the difference between large and small egg-producing ovaries at the protein level).

The experiments on honeybee colonies are well-designed, and they provide fairly strong evidence that the queens are reversibly changing egg size and that it is (at least some component of) their perception of colony size that is the signal. One minor but unavoidable weakness is that experiments on honeybees are necessarily done with small sample sizes. The authors were clear about this, however, and it was very effective that they showed all individual data points. Alongside the previous work on which this paper builds, I found their core results to be rather convincing and important.

We thank the reviewer for this positive evaluation.

I found the parts of the paper on oogenesis to be useful, but overall less informative in answering the questions that the authors set out for those sections. On balance, I think the best way to interpret the oogenesis results is as "suggestive and exploratory". For instance, the experiment aimed at understanding the relationship between egg-laying rate and egg size does not include a direct measurement of egg-laying rate, but instead puts queens in a place with no suitable oviposition sites. The proteomic analysis was fine, but since they were using whole ovaries, with tissue pooled across all stages of oogenesis including mature oocytes, I would be cautious in interpreting the results to mean that they had identified proteins involved in making larger eggs. These proteins might just as easily be the proteins that are put into larger eggs. In fact, for the one candidate gene that is examined, its transcripts seem as though they are predominantly in the oocyte cell itself rather than in the supporting cells that actually control the egg size (although it is hard to tell from the micrographs without a label for cell interfaces).

We have added data on the number of eggs produced in the first experiment, which actually show a negative correlation between egg size and egg number. In addition, we have cautioned our wording about the conclusions that can be drawn from the oviposition restriction experiment. Concerning the expression and role of Rho1, we apologize for the lack of a cell membrane marker. However, we share the reviewer’s interpretation that the mRNA is located in the oocyte. While we also agree that egg loading from the nurse cells is important, transport of vitellogenin from the follicle cells may also be quite significant for egg size (Wu et al. 2021 – doi:10.3389/fcell.2020.593613 and Fleig 1995 - doi:10.1016/0020-7322(95)98841-Z), a process that could be controlled by Rho1 in the documented location. We have added to the discussion to clarify this point.

On that note, with the caveat that the sample sizes are quite small, I agree that there is some evidence that Rho1 is involved in honeybee oogenesis. If this was the only gene they knocked down, and given that it results in a small size change with such a small sample size, it strikes me as a bit of a stretch to say that these results are evidence that Rho1 plays an important role in egg size determination. It is essential to know if this is a generic result of inhibiting cytoskeletal function or a specific function of Rho1. That is beyond the scope of this study, but until those experiments are done, it is hard to know how to interpret these results. For context, in Drosophila, there are lots and lots of genes such that if you knock them down, you get a smaller or differently shaped egg, including genes involved in planar polarity, cytoskeleton, basement membrane, protrusion/motility, septate junctions, intercellular signaling and their signal transduction components, muscle functions, insect hormones, vitellogenesis, etc. This is helpful, perhaps, for thinking about how to interpret the knockdown of just one gene.

We thank the reviewer for this perspective and have consequently cautioned our wording. The role of Rho1 in regulating the cytoskeletal function has been established in other organisms, but we do not have the tools to study the corresponding pathways and establish causality in honey bees. We have added to the discussion to alert the reader to the point that additional experiments are necessary.

Overall, I found the results to be technically sound, and there are several clever manipulations on honeybee colonies that will doubtless be repeated and elaborated in the future to great effect. The core result-that queens can change the size of their eggs quickly and reversibly, in response to some perceived signal-was honestly pretty astonishing to me, and it reveals that there are non-nutritive plastic mechanisms in insect oogenesis that we had no idea existed. I look forward to follow-up studies with interest.

We thank the reviewer for the overall evaluation and encouragement to continue our research.

Author Response

Reviewer #2 (Public Review):

In this manuscript, the authors performed single-cell RNA sequencing (scRNA-seq) analysis on bone marrow CD34+ cells from young and old healthy donors to understand the age-dependent cellular and molecular alterations during human hematopoiesis. Using a logistic regression classifier trained on young healthy donors, they identified cell-type composition changes in old donors, including an expansion of hematopoietic stem cells (HSCs) and a reduction of committed lymphoid and myeloid lineages. They also identified cell-type-specific molecular alterations between young and old donors and age-associated changes in differentiation trajectories and gene regulatory networks (GRNs). Furthermore, by comparing the single-cell atlas of normal hematopoiesis with that of myelodysplastic syndrome (MDS), they characterized cellular and molecular perturbations affecting normal hematopoiesis in MDS.

The present manuscript provides a valuable single-cell transcriptomic resource to understand normal hematopoiesis in humans and the age-dependent cellular and molecular alterations. However, their main claims are not well supported by the data presented. All results were based on computational predictions, not experimentally validated.

Major points:

1) The authors constructed a regularized logistic regression trained on young donors with manually annotated cell types and predicted cell type labels of cells from old and MDS samples. As the manual annotation of cell types was implicitly assumed as ground truth in this manuscript, I'm wondering whether the predicted cell types in old and MDS samples are consistent with the manual annotation. They should apply the same strategy used in young samples for manual annotation to old and MDS samples, and evaluate how accurate their classifier is.

We performed manual annotation for each MDS sample independently, and for the 3 healthy elderly donors integrated dataset. To do so, we performed unsupervised clustering with Seurat and annotated the clusters using the same set of canonical marker genes that we used for the young data. We then analyzed the correspondences between the annotated clusters and the predictions by GLMnet. Results are shown on Figure 1a. We observe that the biggest disagreements between methods occur between adjacent identities, such as HSC and LMPP, GMP and GMP with more prominent granulocytes profile, or MEP, early and late erythroid. When we explore these disagreements along the erythroid branch, we see that they particularly occur close to the border between subpopulations (Figure 1b). This is consistent with the continuous nature of the differentiation and the difficulty to establish boundaries between cell compartments. However, we observe that miss-labeling between different hematopoietic lineages is rare.

In addition, unsupervised clustering was not always able to directly separate the data in the expected subpopulations. We can see different clusters containing the same cell types (e.g. LMPP1, LMPP2), as well as individual clusters containing cells with different identities (e.g. pDC and monocyte progenitors). This is usually due to sources of variability different to cell identity present in the data Additional, supervised finetuning by local sub clustering and merging would be needed to correct for this. On the contrary, we believe that our GLMnet-based method focusses on gene expression related to identity, resulting in a classification that is better suited for our purpose.

Figure 1 Comparison between GLMnet predictions and manually annotated clusters A) Heatmaps showing percentages of cells in manually annotated clusters (columns) that have been assigned to each of the cell identities predicted by our GLMnet classification method (rows). The analysis was performed independently for the elderly integrated dataset and for every MDS sample. B) UMAP plots showing disagreements in classification between adjacent cell compartments in the erythroid branch. Cells from one erythroid cluster per patient are colored by the identity assigned by the GLMnet classifier. Cells in gray are not in the highlighted cluster, nor labeled as MEP, erythroid early or erythroid late by our classifier.

2) The cell-type composition changes in Figures 1 and 4 were descriptively presented without providing the statistical significance of the changes. In addition, the age-dependent cell-type composition changes should be validated by flow cytometry.

We thank the reviewer for the comment. Significance of the changes is included in Supplementary File 3. In addition, we included the percentage of several cell types we validated by flow cytometry, namely HSCs, GMPs and MEPs, in young and elderly healthy individuals in the manuscript, as Figure 1-figure supplement 3. Similarly to what we detected in our bioinformatic analyses, flow cytometry data demonstrated a significant increase in the percentage of HSCs, as well as an increasing trend in MEPs and a slight decrease in the percentage of GMPs in elderly individuals, corroborating our previous results.

3) In Figure 2, the authors used two different pseudo-time inference methods, STREAM, and Palantir. It is not clear why they used two different methods for trajectory inference. Do they provide the same differentiation trajectories? How robust are the results of trajectory inference algorithms? It seems to be inconsistent that the pseudotime inferred by STREAM was not used for downstream analysis and the new pseudotime was recalculated by using Palantir.

We thank the reviewer for the comment. The reason behind using two different methods to perform similar analyses, is that each of them provides specific outputs that can be used to perform a more robust and comprehensive analysis. STREAM allows to unravel the differentiation trajectories in a single cell dataset with an unsupervised approach. Also the visualization provided by STREAM (Figure 2C and 2D) allows for a simple interpretation of the results to the reader. On the other hand, Palantir provides a more robust analysis to dissect how gene expression dynamics interact and change with differentiation trajectories. For this reason, we decided to use this second method to investigate how specific genes were altered in the monocytic compartment.

As a resource article, the showcase of different methods can be valuable as it provides examples on how each tool can be used to obtain specific results, which can help any reader to decide which might be the best tool for their specific case.

Just to confirm that pseudotime results are similar, we perform a correlation analysis with the pseudotime values obtained from each method. We observed a correlation coefficient of 0.78 (p.val < 2.2e-16) confirming the similarity among both tools.

Figure 2. Correlation analysis of pseudotime values obtained with STREAM and PALANTIR.

4) In Figure 2D, some HSCs seem to be committed to the erythroid lineage. The authors should carefully examine whether these HSCs are genuinely HSCS, not early erythroid progenitors.

We thank the reviewer for the comment. We have performed a deep analysis regarding the classification of HSCs (See Figure 3). Our analyses reveal that none of the cells classified as HSCs express early erythroid progenitor markers. We have also used STREAM to show the expression of these markers along the obtained trajectory and observed that erythroid markers show expression in the erythroid trajectory but not in the HSC compartment (Figure 4).

Figure 3 Expression of marker genes in the HSC compartment. Dot plot depicting the normalized scaled expression of canonical marker genes by HSC of the 5 young and 3 elderly healthy donors. Marker genes are colored by the cell population they characterize. Dot color represents expression levels, and dot size represents the percentage of cells that express a gene.

Figure 4. Expression of erythroid markers in STREAM trajectories. Expression of GATA1 and HBB (erythroid markers) in the predicted differentiation trajectories.

5) It is not clear how the authors draw a conclusion from Figure 3D that the number of common targets between transcription factors is reduced. Some quantifications should be provided.

We thank the reviewer for the comment. We have updated the manuscript to better reflect our findings and emphasize that the predicted regulatory networks of HSCs in elderly donors is displayed as an independent network, compared to the young donors. (Page 6, line 36).

“Overall, we observed that the predicted regulatory network of elderly HSCs (Figure 3d) appeared as an independent network compared to the young GRN. This finding could result in the loss of co-regulatory mechanisms in the elderly donors.”

6) The constructed GRNs and related descriptions were based solely on the SCENIC analysis. By providing the results of an orthogonal prediction method for GRNs, the authors should evaluate how robust and consistent their predictions are.

We thank the reviewer for the comment regarding the method to build gene regulatory networks. As a resource article, our manuscript describes a complete workflow to perform different aspects of single cell analyses. These steps go from automated classification, trajectory inference and GRN prediction. All the selected algorithms have already been benchmarked and compared against other tools that perform similar analysis. SCENIC has already been benchmarked against other algorithms (11) and by others (12).

We do agree with the reviewer that these new predictions could provide strength to our findings, however we believe that these orthogonal predictions would better fit if our article was intended for the Research Article category instead of Tools and Resources.

7) The observed age-dependent cellular and molecular alterations in human hematopoiesis are interesting, but I'm wondering whether the observed alterations are driven by inflammatory microenvironment or intrinsic properties of a subpopulation of HSCs affected by clonal hematopoiesis (CH). To address this, the authors can perform genotyping of transcriptomes (GoT) on old healthy donors with CH. By comparing the transcriptomes of cells with and without CH mutations, we can evaluate the effects of CH on age-associated molecular alterations.

We thank the reviewer for the comment. Unfortunately, in order to perform GoT (genotyping of transcriptomes) on the healthy donors, requires modifying the standard 10x Genomics workflow to amplify the targeted locus and transcript of interest. This would require collecting new samples, optimizing the method and performing new analysis from scratch (from sequencing up to analysis). We believe this is not in the scope of the manuscript. On the other hand, we don’t have enough material to create new single cell libraries, this fact would require the addition of new donors and as a result, a complete new analysis to perform the integration.

Reviewer #3 (Public Review):

The authors have performed a transcriptional analysis of young/aged hematopoietic stem/progenitor cells which were obtained from normal individuals and those with MDS.

The authors generated an important and valuable dataset that will be of considerable benefit to the field. However, the data appear to be over-interpreted at times (for example, GSEA analysis does not have "functionality", as the authors claim). On the other hand, a comparison between normal-aged HSC and HSC from MDS patients appears to be under-explored in trying to understand how this disease (which is more common in the elderly) disrupts HSC function.

A more extensive cross-referencing of other normal HSPC/MDS HSCP datasets from aged humans would have been helpful to highlight the usefulness of the analytical tools that the authors have generated.

Major points

1) The authors detail methodology for identification of cell types from single-cell data - GLMnet. This portion of the text needs to be clarified as it is not immediately clear what it is or how it's being used. It also needs to be explained by what metric the classifier "performed better among progenitor cell types" and why this apparent advantage was sufficient to use it for the subsequent analysis. This is critical since interpretation of the data that follows depends on the validation of GLMnet as a reliable tool.

We thank the review for the comment. We have updated the corresponding section to better describe how GLMnet is used and that the reasoning on why we decided to use GLMnet as our cell type annotation method instead of other available tools such as Seurat, is based on the results of the benchmark described in Figure 1-figure supplement 1. We also described the main differences between our method and Seurat (See Answer to Review 1, Question # 4).

2) The finding of an increased number of erythroid progenitors and decreased number of myeloid cells in aged HPSC is surprising since aging is known to be associated with anemia and myeloid bias. Given that the initial validation of GLMnet is insufficiently described, this result raises concerns about the method. Along the same lines, the authors report that their tool detects a reduced frequency of monocyte progenitors. How does this finding correlate with the published data on aging humans? Is monocytopenia a feature of normal aging?

We thank the reviewer for this comment, as changes in the output of HSCs as a consequence of aging are of high interest. According to the literature, there is clear evidence of the loss of lymphoid progeny with age (13,14), which goes in agreement with our results. However, in the case of the myeloid compartment, the effects of aging are not as clear. Studies in mice have indeed observed that the loss of lymphoid cells is accompanied by increased myeloid output, starting at the level of GMPs (Rossi et al. 2005; Florian et al. 2012; Min et al. 2006). But studies on human individuals have not found changes in numbers of these myeloid progenitors (Kuranda et al. 2011; Pang et al. 2011). In addition, in the mentioned studies, myeloid production was measured exclusively by its white blood cells fraction. More recent studies have focused on the other myeloid compartments: megakaryocyte and erythroid cells. Results point towards the increase of platelet-biased HSC with age (Sanjuan-Pla et al. 2013; Grover et al. 2016) and a possible expansion of megakaryocytic and erythroid progenitor populations (Yamamoto et al. 2018; Poscablo et al. 2021; Rundberg Nilsson et al. 2016), which may represent a compensatory mechanism for the ineffective differentiation towards this lineage in elderly individuals. This goes in line with the accumulation of MEPs we see in our data. Finally, and in accordance with the reduced frequency of monocyte progenitors observed, it has been shown that with increasing age, there is a gradual decline in the monocyte count (15).

Regarding the concerns about our classification method raised by the reviewer, we have performed additional validations that we describe in answers to reviewer 1 comment #4 and reviewer 2 comment #1. To further confirm that the changes in cellular proportions we found are real, we applied two additional classification methods: Seurat transfer and Celltypist (16) to the elderly donors dataset. We obtained a similar expansion in MEPs, together with reduction of monocytic progenitors with the three methods (Figure 5).

Figure 5 Classification of HSPCs from elderly donors. Barplot showing proportions of every cell subpopulation per elderly donor, resulting from three classification methods: GLMnet-based classifier, Seurat transfer and Celltypist. For the three methods, cells with prediction scores < 0,5 were labeled as “not assigned”.

3) The use of terminology requires more clarity in order to better understand what kind of comparison has been performed, i.e. whether global transcriptional profiles are being compared, or those of specific subset populations. Also, the young/aged comparisons are often unclear, i.e. it's not evident whether the authors are referring to genes upregulated in aged HSC and downregulated in young HSC or vice versa. A more consistent data description would make the paper much easier to read.

We thank the reviewer for this comment. We have updated the manuscript to provide more clarity in the description of the different comparisons made in our analyses. Most changes are located in the Transcriptional profiling of human young and elderly hematopoietic progenitor systems sub-section within the Results.

4) The link between aging and MDS is not explored but could be an informative use of the data that the authors have generated. For example, anemia is a feature of both aging and MDS whereas neutropenia and thrombocytopenia only occur in MDS. Are there any specific pathways governing myeloid/platelet development that are only affected in MDS?

Thank you for raising this comment. We believe that discriminating events that take place during healthy aging from those associated to MDS will be helpful to understand this particular disease, as it is so closely related to age. This is why, when analyzing MDS, we have considered young and elderly donors as two separate sets of healthy controls, the eldery donors being the most suitable one for comparisons with MDS samples.

With regards to the comment on myeloid and platelet development, the GSEA analysis gives potentially useful information. MYC targets and oxidative phosphorylation are significantly enriched in the MEP compartment from MDS patients when compared to elderly donors, indicating that these progenitors may recover a more active profile with the disease. Hypoxia related genes, on the other hand, are more active in HSCs and MEPs from healthy elderly donors than in MDS. Hypoxia is known to be implicated in megakaryocyte and erythroid differentiation (17)

5) MDS is a very heterogeneous disorder and while the authors did specify that they were using samples from MDS with multilineage dysplasia, more clinical details (blood counts, cytogenetics, mutational status) are needed to be able to interpret the data.

We thank the reviewer for the comment. All the clinical details for each MDS patient are included in Supplementary File 5.

Author Response

Reviewer #1 (Public Review):

In this study, Sims et al. evaluate how system-level brain functional connectivity is associated with cognitive abilities in a sample of older adults aged > 85 years old. Because the study sample of 146 normal older adults has lived into advanced years of age, the novelty here is the opportunity to validate brain-behavioral associations in aging with a reduced concern of the potential influence of undetected incipient neuropsychological pathology. The participants afforded resting-state functional magnetic resonance imaging (rs-fMRI) data as well as behavioral neuropsychological test assessments of various cognitive abilities. Exploratory factor analysis was applied on the behavioral cognitive assessments to arrive at summary measures of participant ability in five cognitive domains including processing speed, executive functioning, episodic memory, working memory, and language. rsfMRI data were submitted to a graph-theoretic approach that derived underlying functional nodes in brain activity, the membership of these nodes in brain network systems, and indices characterizing the organizational properties of these brain networks. The study applies the classification of the various brain networks into a sensory/motor system of networks and an association system of network, with further sub-systems in the latter that includes the frontoparietal network (FPN), the default-mode network (DMN), the cingulo-opercular network (CON), and the dorsal (DA) and ventral (VA) attention networks. Amongst other graph metrics, the study focused on the extent to which networks in these brain systems were segregated (i.e., separable network communities as opposed to a more singular large community network). Evaluation of the brain network segregation indices and cognitive performance metrics showed that in general higher network functional segregation corresponds with higher cognitive performance ability. In particular, this association was seen between the general association system with overall cognition, and the FPN with overall cognition, and processing speed.

The results worthy of highlighting include the documentation of oldest-old individuals with detectable brain neural network segregation at the level of the association system and its FPN sub-system and the association of this brain functional state notably with general cognition and processing speed and less so with the other specific cognitive domains (such as memory). This finding suggests that (a) apparently better cognitive aging might stem from a specific level of neural network functional segregation, and (b) this linkage applies more specifically to the FPN and processing speed. These specific findings inform the broader conceptual perspective of how human brain aging that is normative vs. that which is pathological might be distinguished.

We appreciate this comment and we have added these points to the conclusion more explicitly.

To show the above result, this study defined functional networks that were driven more by the sample data as opposed to a pre-existing generic template. This approach involves a watershed algorithm to obtain functional connectivity boundary maps in which the boundary brain image voxels separate functionally related voxels from unrelated voxels by virtue of their functional covariance as measured in the immediate data. This is also a notable objective and data-driven approach towards defining functional brain regions-of-interest (ROIs), nodes, and networks that are age-appropriate and configured for a given dataset as opposed to using network definitions based on other datasets used as a generic template.

The sample size of 146 for this age group is generally sufficient.

For the analyses considering the significance of the effect of the brain network metrics on the cognitive variables, the usage of heirarchical regression to evaluate whether the additional variables (in the full model) significantly change the model fit relative to the reduced model with covariates-only (data collection site, cortical thickness), while a possible approach, might be problematic, particularly when the full model uses many more regressors than the reduced model. In general, adding more variables to regression models reduces the residual variance. As such, it is possible that adding more regressors in a full model and comparing that to a reduced model with much fewer regressors would yield significant changes in the R^2 fit index, even if the added regressors are not meaningfully modulating the dependent variable. This may not be an issue for the finding on the FPN segregation effect on overall cognition, but it may be important in interpreting the finding on the association system metrics on overall cognition.

Critically, we should note that the correlation effect sizes (justified by the 0.23 value based on the reported power analyses) were all rather small in size. The largest key brain-behavior correlation effect was 0.273 (between DMN segregation and Processing Speed). In the broader perspective, such effects sizes generally suggest that the contribution of this factor is minimal and one should be careful that the results should be understood in this context.

The recent, highly publicized paper from Marek and colleagues (citation below) offers some support for the assertion that these effect sizes are on the order that would be expected for ‘true’ signals in the brain. While the study reported here is not a “BWAS” as described in the Marek article (BWAS is a brain-wide association study, examining, without a priori hypotheses of brain network, all possible associations), and therefore our study does not fall prey to some of the multiple comparisons issues described in that paper, the general expected effect sizes based on that paper should be relevant here.

Marek and colleagues suggest that 1) effect sizes in the range of 0.273 are on the order of the larger brain-behavior relationships that can be expected to be replicable, and 2) samples that remove some drivers of individual variability are beneficial to the capability of a study to identify an effect. Relevant to the latter point, by reducing the variability in our sample due to age (our age range is tight) and early signs of neurological disease (these were screened out in our sample), this leaves a sample that is homogeneous along these variables, meaning that brain variability associated with cognitive performance can be more easily pulled from the data.

Our data have large variability on the behavior end, and large variability on the brain end, allowing better power for seeing effects between them.

Marek, S., Tervo-Clemmens, B., Calabro, F.J., Montez, D.F., Kay, B.P., Hatoum, A.S., Donohue, M.R., Foran, W., Miller, R.L., Hendrickson, T.J., et al. (2022). Reproducible brain-wide association studies require thousands of individuals. Nature 603, 654–660. 10.1038/s41586-022-04492-9.

Overall, the findings based on hierarchical regressions that evaluate the network segregation indices in accounting for cognition and the small correlation magnitudes are basically in line with the notion that more segregated neural networks in the oldest-old support better cognitive performance (particularly processing speed). However, the level of positive support for the notion based on these findings is somewhat moderate and requires further study.

The addition of a control analysis (sensorimotor network) in the newer version of the paper showed that these effects are not present in brain networks not thought to relate to cognition. We agree that further study of these questions is necessary for stronger claims to be made, but the current study advances the field by showing clearly that segregation of the association network and its components relates to behavior even in this oldest old cohort.

Reviewer #2 (Public Review):

The authors capitalised on the opportunity to obtain functional brain imaging data and cognitive performance from a group of oldest old with normative cognitive ability and no severe neurophysiological disorders, arguing that these individuals would be most qualified as having accomplished 'healthy ageing'. Combined with the derivation of a cohort-specific brain parcellation atlas, the authors demonstrated the importance of maintaining brain network segregation for normative cognition ability, especially processing speed, even at such late stage of life. In particular, segregation of the frontoparietal network (FPN) was found to be the key network property.

These results bolstered the findings from studies using younger old participants and are in agreement with the current understanding of the connectomme-cognition relationship. The inclusion of a modest sample size, power analysis, cohort-specific atlas, and a pretty comprehension neuropsychological assessment battery provides optimism that the observed importance of FPN segregation would be a robust and generalisable finding at least in future cross-sectional studies. The fact that FPN segregation is relatively more important to cognition than other associative networks also provides novel insight about the possible 'hierarchy' between age-related neural and cognitive changes, regardless of what mechanisms lead to such segregation at such an advanced age. it is also interesting that processing speed remains to be the 'hallmark' metric of age-related cognitive changes, indirectly speaking to its long assumption fundamental impact on overall cognition.

As laid out by the authors, if network differentiation is key to normative cognitive ability at old age, intervention and stimulation programs that could maintain or boost network segregation would have high translational value. With advent in mobile self-administrable devices that target behavioural and neural modifications, this potential would have increasing appeal.

However, I feel that a few things have prevented the manuscript to be a simple yet impactful submission

1) Interpretation and the major theme of discussion. While the authors attempted to discuss their findings with respect to both the compensatory and network dedifferentiation hypotheses, the results and their interpretation do not readily provide any resolution or reconciliation between the two, a common challenge in many ageing research. The authors did not further elaborate how the special cohort they had may provide further insights to this.

While the results certainly are in line with the dedifferentiation hypothesis, why 'this finding does not exclude the compensation hypothesis' (Discussion) was not elaborated enough. In particular, the authors seemed to suggest that maintained network specialisation may be in such a role, but the results and interpretations regarding network specialisation were not particularly focused on throughout the manuscript. In addition, both up regulation within a network and cross-network recruitment can both be potential compensatory strategies (Cabeza et al 2018, Rev Nat Neurosci). Without longitudinal data or other designs (e.g. task) it is quite difficult to evaluate the involvement of compensation. For instance, as rightly suggested by the authors, the two phenomena may not be mutually exclusive (e.g., maintenance of the FPN differentiation at such old age could be a result of 'compensation' that started when the participants were younger).

The reviewer makes some excellent points that we have taken to heart in this revision. We agree that the data as described do not directly address the compensation hypothesis, and therefore de-emphasized our descriptions of that hypothesis in service of a simpler, more impactful manuscript.

As described above in our response to the essential revisions “In the original submission, we noted relevant literature which describe both the dedifferentiation hypothesis and the compensation hypothesis of aging. Our original aim was to include more of a literature review of cognitive aging theories in the introduction and discussion, but that choice made it too confusing (and honestly left out much important literature). In responding to the reviews we realize that bypassing this cursory literature review here is preferable for the readability of the manuscript. Instead, we cite a literature review, and focus on the dedifferentiation hypothesis.

“The data we show here addresses the dedifferentiation hypothesis specifically since we are using the segregation metric- a reflection of dedifferentiation of network organization. The reviewers’ comments caused us to do a great deal of thinking on this topic, and we have a forthcoming review with our colleague Ian McDonough that covers this topic in more detail (McDonough, Nolin, and Visscher, 2022). We have substantially rewritten the relevant sections in the discussion (especially section 3.2) to be more clear for readers.”

As also described in our response to essential revisions 2c, we have added to the discussion regarding the utility of studying the oldest-old; this is in the second paragraph of the discussion, and reproduced above. Additionally, in the Introduction, we also briefly address the importance of this cohort. We state “Prior work has mostly been done in younger-old samples (largely 65-85 years old). Studying the younger-old can be confounded by including pre-symptomatic disease, since it is unknown which individuals may be experiencing undetectable, pre-clinical cognitive disorders and which will continue to be cognitively healthy for another decade. The cognitively unimpaired oldest-old have lived into late ages, and we can be more confident in determining their status as successful agers. A further benefit of studying these successful cognitive agers is that because of their advanced age and the normal aging and plasticity processes associated with it, there is greater variance in both their performance on neurocognitive tasks, and in brain connectivity measures than there is in younger cohorts (Christensen et al., 1994). This increased variance makes it easier to observe across-subject relationships of cognition and brain networks (Gratton, Nelson, & Gordon, 2022). We provide new insight into the relationship between the segregation of networks and cognition by investigating this relationship in an oldest old cohort of healthy individuals.”

2) Some further clarity about the data and statistical analyses would be desirable. First, since scan length determines the stability of functional connectivity, how long was the resting-state scan? Second, what is the purpose of using both hierarchical regression and partial correlation? While they do consider different variances in the dataset, they are quite similar and the decision looks quite redundant to me as not much further insights have been gained. [the main insight to including a regression is to be able to compare the different networks to each other.]

The resting-state fMRI scan is 8 minutes in length. This has been added to the text. After considering the redundancy the reviewer notes between hierarchical regression and correlations, we have simplified our statistical approach and only included correlations in the main body of the manuscript. We have put the regressions in the supplemental materials so if interested readers would like to be able to see those results, they are still available.

Author Response

Reviewer #2 (Public Review):

Zhukin et al., present the structure of the central scaffold component of the NuA4 complex. They hypothesise how the nucleosome interacting modules not present in the structure could be arranged, based on Alphafold modelling, and comparison of their structure to other complexes that use the same subunits. They show some interesting -albeit fairly preliminary - biochemistry on the binding of the flexible modules, suggesting a role for acetylation affecting H3K4me3 reading.

While the work builds upon previous structural studies on the Tra1 subunit in isolation and a previous 4.7A resolution structure from another group, there are clear differences and novel findings in this study. The data is presented beautifully and nicely annotated figures make following the many subunits and interactions therein simple. What could have been a very complex manuscript is easy to digest. Some of the figures could do with a couple of additional labels and detailed figure legends to make things a little clearer.

Overall, a nice study and a wonderfully detailed structure of a large multi-subunit assembly but we would recommend some further experimentation validation to bolster their findings.

Major comments

1) All 13 subunits of NuA4 are present by mass spec, however, based on the SDS-page gel (Fig1-1) components of the TINTIN sub-complex seem less than stoichiometric, with Eaf7 and Eaf3 certainly much weaker stained. This is particularly important with reference to Figure 3 and the discussion in the text which assumes the nucleosome interacting modules are all present equally, but too flexible to be observed in the structure.

Simple peptide numbers from mass spec cannot be used as a measure of protein abundance as this is sensitive to multiple confounding factors.

We did not identify the locations of individual modules (HAT, TINTIN and Yaf9) within the diffuse density, we merely indicate that this is a likely location for their presents based on the location of connections points and presence of crosslinks in previously published data. We did perform mass photometry analysis of the purified NuA4 sample to better determine the composition of the purified complex (Figure 1-1). We find that the major species peak is center at 1037 kDa, which is very close to the theoretical mass of 1043 kDa. There are a few other minor peaks but none of this would indicate a NuA4 complex lacking TINTIN (Eaf3,5,7) or any other distinct subcomplex.

2) A major novel biological finding and conclusion from the abstract concerns the binding to modified nucleosomes. However, this seemed somewhat preliminary, especially considering the discussion around the role of acetylation affecting binding to H3K4me3 nucleosomes based solely on the dCypher screen used.

The discussion on the role of HAT module binding preferential to acetylated and methylated tails concludes that the acetylation liberates the H3 tail from DNA interaction, making H3K4me3 more available for binding by the PHD domain. This is an interesting hypothesis but is stated as fact with very little evidence to make this assertion.

Whilst others have seen similar results (cited in the paper), no data is presented to disregard an alternative hypothesis that there is some additional acetyl-binding activity in the complex. Indeed, in one of the references they cite the authors do show a direct reading of acetylation as well as methylation.

TINTIN binding is subject to high background and a fairly minor effect. The biological relevance to these observations while intriguing needs to be proved further.

We have changed the language of this section to hopefully better leave open other possibilities. As for the TINTIN dCypher results, we do not try to draw too many conclusions, but the data indicates that there is very little (if any) interaction with the histones tails (at least for the modifications present in the panel). One thing we can say is that the TINTIN module does not seem to have any binding preference for H3K36me3 nucleosomes.

3) There is a large focus on the cross-linking mass spec study from another group and the previously published structure of the NuA4 complex. The authors are fairly aggressive in suggesting the other structure from Wang et al., is incorrect. It is very nice that their built structure shows a better interpretation of previous XL-MS data, but still many of the crosslinks are outside of the modelled density. One possibility that should be entertained is that the two studies are comparing different structures/states of NuA4. The authors of the Wang et al., paper indeed comment that Swc4 and Yaf9 are missing from their purified complex. It is of course possible that both structures are correct as they appear to be biochemically different, with the crosslinking in the Setiaputra paper better reflecting the complex presented here.

Response given above.

Author Response

Reviewer #1 (Public Review):

This manuscript seeks to identify the mechanism underlying priority effects in a plantmicrobe-pollinator model system and to explore its evolutionary and functional consequences. The manuscript first documents alternative community states in the wild: flowers tend to be strongly dominated by either bacteria or yeast but not both. Then lab experiments are used to show that bacteria lower the nectar pH, which inhibits yeast - thereby identifying a mechanism for the observed priority effect. The authors then perform an experimental evolution unfortunately experiment which shows that yeast can evolve tolerance to a lower pH. Finally, the authors show that low-pH nectar reduces pollinator consumption, suggesting a functional impact on the plant-pollinator system. Together, these multiple lines of evidence build a strong case that pH has far-reaching effects on the microbial community and beyond.

The paper is notable for the diverse approaches taken, including field observations, lab microbial competition and evolution experiments, genome resequencing of evolved strains, and field experiments with artificial flowers and nectar. This breadth can sometimes seem a bit overwhelming. The model system has been well developed by this group and is simple enough to dissect but also relevant and realistic. Whether the mechanism and interactions observed in this system can be extrapolated to other systems remains to be seen. The experimental design is generally sound. In terms of methods, the abundance of bacteria and yeast is measured using colony counts, and given that most microbes are uncultivable, it is important to show that these colony counts reflect true cell abundance in the nectar.

We have revised the text to address the relationship between cell counts and colony counts with nectar microbes. Specifically, we point out that our previous work (Peay et al. 2012) established a close correlation between CFUs and cell densities (r2 = 0.76) for six species of nectar yeasts isolated from D. aurantiacus nectar at Jasper Ridge, including M. reukaufii.

As for A. nectaris, we used a flow cytometric sorting technique to examine the relationship between cell density and CFU (figure supplement 1). This result should be viewed as preliminary given the low level of replication, but this relationship also appears to be linear, as shown below, indicating that colony counts likely reflect true cell abundance of this species in nectar.

It remains uncertain how closely CFU reflects total cell abundance of the entire bacterial and fungal community in nectar. However, a close association is possible and may be even likely given the data above, showing a close correlation between CFU and total cell count for several yeast species and A. nectaris, which are indicated by our data to be dominant species in nectar.

We have added the above points in the manuscript (lines 263-264, 938-932).

The genome resequencing to identify pH-driven mutations is, in my mind, the least connected and developed part of the manuscript, and could be removed to sharpen and shorten the manuscript.

We appreciate this perspective. However, given the disagreement between this perspective and reviewer 2’s, which asks for a more expanded section, we have decided to add a few additional lines (lines 628-637), briefly expanding on the genomic differences between strains evolved in bacteria-conditioned nectar and those evolved in low-pH nectar.

Overall, I think the authors achieve their aims of identifying a mechanism (pH) for the priority effect of early-colonizing bacteria on later-arriving yeast. The evolution and pollinator experiments show that pH has the potential for broader effects too. It is surprising that the authors do not discuss the inverse priority effect of early-arriving yeast on later-arriving bacteria, beyond a supplemental figure. Understandably this part of the story may warrant a separate manuscript.

We would like to point out that, in our original manuscript, we did discuss the inverse priority effects, referring to relevant findings that we previously reported (Tucker and Fukami 2014, Dhami et al. 2016 and 2018, Vannette and Fukami 2018). Specifically, we wrote that: “when yeast arrive first to nectar, they deplete nutrients such as amino acids and limit subsequent bacterial growth, thereby avoiding pH-driven suppression that would happen if bacteria were initially more abundant (Tucker and Fukami 2014; Vannette and Fukami 2018)” (lines 385-388). However, we now realize that this brief mention of the inverse priority effects was not sufficiently linked to our motivation for focusing mainly on the priority effects of bacteria on yeast in the present paper. Accordingly, we added the following sentences: “Since our previous papers sought to elucidate priority effects of early-arriving yeast, here we focus primarily on the other side of the priority effects, where initial dominance of bacteria inhibits yeast growth.” (lines 398-401).

I anticipate this paper will have a significant impact because it is a nice model for how one might identify and validate a mechanism for community-level interactions. I suspect it will be cited as a rare example of the mechanistic basis of priority effects, even across many systems (not just pollinator-microbe systems). It illustrates nicely a more general ecological phenomenon and is presented in a way that is accessible to a broader audience.

Thank you for this positive assessment.

Reviewer #2 (Public Review):

The manuscript "pH as an eco-evolutionary driver of priority effects" by Chappell et al illustrates how a single driver-microbial-induced pH change can affect multiple levels of species interactions including microbial community structure, microbial evolutionary change, and hummingbird nectar consumption (potentially influencing both microbial dispersal and plant reproduction). It is an elegant study with different interacting parts: from laboratory to field experiments addressing mechanism, condition, evolution, and functional consequences. It will likely be of interest to a wide audience and has implications for microbial, plant, and animal ecology and evolution.

This is a well-written manuscript, with generally clear and informative figures. It represents a large body and variety of work that is novel and relevant (all major strengths).

We appreciate this positive assessment.

Overall, the authors' claims and conclusions are justified by the data. There are a few things that could be addressed in more detail in the manuscript. The most important weakness in terms of lack of information/discussion is that it looks like there are just as many or more genomic differences between the bacterial-conditioned evolved strains and the low-pH evolved strains than there are between these and the normal nectar media evolved strains. I don't think this negates the main conclusion that pH is the primary driver of priority effects in this system, but it does open the question of what you are missing when you focus only on pH. I would like to see a discussion of the differences between bacteria-conditioned vs. low-pH evolved strains.

We agree with the reviewer and have included an expanded discussion in the revised manuscript [lines 628-637]. Specifically, to show overall genomic variation between treatments, we calculated genome-wide Fst comparing the various nectar conditions. We found that Fst was 0.0013, 0.0014, and 0.0015 for the low-pH vs. normal, low pH vs. bacteria-conditioned, and bacteria-conditioned vs. normal comparisons, respectively. The similarity between all treatments suggests that the differences between bacteria-conditioned and low pH are comparable to each treatment compared to normal. This result highlights that, although our phenotypic data suggest alterations to pH as the most important factor for this priority effect, it still may be one of many affecting the coevolutionary dynamics of wild yeast in the microbial communities they are part of. In the full community context in which these microbes grow in the field, multi-species interactions, environmental microclimates, etc. likely also play a role in rapid adaptation of these microbes which was not investigated in the current study.

Based on this overall picture, we have included additional discussion focusing on the effect of pH on evolution of stronger resistance to priority effects. We compared genomic differences between bacteria-conditioned and low-pH evolved strains, drawing the reader’s attention to specific differences in source data 14-15. Loci that varied between the low pH and bacteria-conditioned treatments occurred in genes associated with protein folding, amino acid biosynthesis, and metabolism.

Reviewer #3 (Public Review):

This work seeks to identify a common factor governing priority effects, including mechanism, condition, evolution, and functional consequences. It is suggested that environmental pH is the main factor that explains various aspects of priority effects across levels of biological organization. Building upon this well-studied nectar microbiome system, it is suggested that pH-mediated priority effects give rise to bacterial and yeast dominance as alternative community states. Furthermore, pH determines both the strengths and limits of priority effects through rapid evolution, with functional consequences for the host plant's reproduction. These data contribute to ongoing discussions of deterministic and stochastic drivers of community assembly processes.

Strengths:

Provides multiple lines of field and laboratory evidence to show that pH is the main factor shaping priority effects in the nectar microbiome. Field surveys characterize the distribution of microbial communities with flowers frequently dominated by either bacteria or yeast, suggesting that inhibitory priority effects explain these patterns. Microcosm experiments showed that A. nectaris (bacteria) showed negative inhibitory priority effects against M. reukaffi (yeast). Furthermore, high densities of bacteria were correlated with lower pH potentially due to bacteria-induced reduction in nectar pH. Experimental evolution showed that yeast evolved in low-pH and bacteria-conditioned treatments were less affected by priority effects as compared to ancestral yeast populations. This potentially explains the variation of bacteria-dominated flowers observed in the field, as yeast rapidly evolves resistance to bacterial priority effects. Genome sequencing further reveals that phenotypic changes in low-pH and bacteriaconditioned nectar treatments corresponded to genomic variation. Lastly, a field experiment showed that low nectar pH reduced flower visitation by hummingbirds. pH not only affected microbial priority effects but also has functional consequences for host plants.

We appreciate this positive assessment.

Weaknesses:

The conclusions of this paper are generally well-supported by the data, but some aspects of the experiments and analysis need to be clarified and expanded.

The authors imply that in their field surveys flowers were frequently dominated by bacteria or yeast, but rarely together. The authors argue that the distributional patterns of bacteria and yeast are therefore indicative of alternative states. In each of the 12 sites, 96 flowers were sampled for nectar microbes. However, it's unclear to what degree the spatial proximity of flowers within each of the sampled sites biased the observed distribution patterns. Furthermore, seasonal patterns may also influence microbial distribution patterns, especially in the case of co-dominated flowers. Temperature and moisture might influence the dominance patterns of bacteria and yeast.

We agree that these factors could potentially explain the presented results. Accordingly, we conducted spatial and seasonal analyses of the data, which we detail below and include in two new paragraphs in the manuscript [lines 290-309].

First, to determine whether spatial proximity influenced yeast and bacterial CFUs, we regressed the geographic distance between all possible pairs of plants to the difference in bacterial or fungal abundance between the paired plants. If plant location affected microbial abundance, one should see a positive relationship between distance and the difference in microbial abundance between a given pair of plants: a pair of plants that were more distantly located from each other should be, on average, more different in microbial abundance. Contrary to this expectation, we found no significant relationship between distance and the difference in bacterial colonization (A, p=0.07, R2=0.0003) and a small negative association between distance and the difference in fungal colonization (B, p<0.05, R2=0.004). Thus, there was no obvious overall spatial pattern in whether flowers were dominated by yeast or bacteria.

Next, to determine whether climatic factors or seasonality affected the colonization of bacteria and yeast per plant, we used a linear mixed model predicting the average bacteria and yeast density per plant from average annual temperature, temperature seasonality, and annual precipitation at each site, the date the site was sampled, and the site location and plant as nested random effects. We found that none of these variables were significantly associated with the density of bacteria and yeast in each plant.

To look at seasonality, we also re-ordered Fig 2C, which shows the abundance of bacteria- and yeast-dominated flowers at each site, so that the sites are now listed in order of sampling dates. In this re-ordered figure, there is no obvious trend in the number of flowers dominated by yeast throughout the period sampled (6.23 to 7/9), giving additional indication that seasonality was unlikely to affect the results.

Additionally, sampling date does not seem to strongly predict bacterial or fungal density within each flower when plotted.

These additional analyses, now included (figure supplements 2-4) and described (lines 290-309) in the manuscript, indicate that the observed microbial distribution patterns are unlikely to have been strongly influenced by spatial proximity, temperature, moisture, or seasonality, reinforcing the possibility that the distribution patterns instead indicate bacterial and yeast dominance as alternative stable states.

The authors exposed yeast to nectar treatments varying in pH levels. Using experimental evolution approaches, the authors determined that yeast grown in low pH nectar treatments were more resistant to priority effects by bacteria. The metric used to determine the bacteria's priority effect strength on yeast does not seem to take into account factors that limit growth, such as the environmental carrying capacity. In addition, yeast evolves in normal (pH =6) and low pH (3) nectar treatments, but it's unclear how resistance differs across a range of pH levels (ranging from low to high pH) and affects the cost of yeast resistance to bacteria priority effects. The cost of resistance may influence yeast life-history traits.

The strength of bacterial priority effects on yeast was calculated using the metric we previously published in Vannette and Fukami (2014): PE = log(BY/(-Y)) - log(YB/(Y-)), where BY and YB represent the final yeast density when early arrival (day 0 of the experiment) was by bacteria or yeast, followed by late arrival by yeast or bacteria (day 2), respectively, and -Y and Y- represent the final density of yeast in monoculture when they were introduced late or early, respectively. This metric does not incorporate carrying capacity. However, it does compare how each microbial species grows alone, relative to growth before or after a competitor. In this way, our metric compares environmental differences between treatments while also taking into account growth differences between strains.

Here we also present additional growth data to address the reviewer’s point about carrying capacity. Our experiments that compared ancestral and evolved yeast were conducted over the course of two days of growth. In preliminary monoculture growth experiments of each evolved strain, we found that yeast populations did reach carrying capacity over the course of the two-day experiment and population size declined or stayed constant after three and four days of growth.

However, we found no significant difference in monoculture growth between the ancestral stains and any of the evolved strains, as shown in Figure supplement 12B. This lack of significant difference in monoculture suggests that differences in intrinsic growth rate do not fully explain the priority effects results we present. Instead, differences in growth were specific to yeast’s response to early arrival by bacteria.

We also appreciate the reviewer’s comment about how yeast evolves resistance across a range of pH levels, as well as the effect of pH on yeast life-history traits. In fact, reviewer #2 pointed out an interesting trade-off in life history traits between growth and resistance to priority effects that we now include in the discussion (lines 535-551) as well as a figure in the manuscript (Figure 8).

Author Response

Reviewer #1 (Public Review):

This works makes an important contribution to the study of the cell cycle and the attempt to infer mechanism by studying correlations in division timing between single cells.

Given the importance of circadian rhythms to the ultimate conclusions of the study, I think it would be helpful to clarify the connection between possible oscillatory regulatory mechanisms and the formalism developed in e.g. Equation 3. The treatment appears to be a leading order expansion in stochastic fluctuations of the cell cycle regulators about the mean, but if an oscillatory process is involved, the fluctuations will be correlated in time and need not be small.

We thank the reviewer for the positive assessment of our work. We have introduced Section S7 in the Supplementary Information to address the connection between our theory and two existing models of circadian modulation of cell division. In the first model, the circadian clock drives the interdivision time, while in the second model, the clock drives cell size control. We find that, while both models satisfy the cousin inequality for comparable parameters, they differ in their interdivision time correlation patterns. The first model yields an alternator-oscillator mixed pattern, while the second gives an aperiodic-oscillator pattern.

The reviewer is right that our theory presents a leading-order expansion of cell cycle factor fluctuations. To overcome this limitation, we introduced the new Section S2 in the Supplementary Information, which shows how the correlation patterns are altered for moderately strong fluctuations. Interestingly, nonlinearity can be treated within our framework by introducing complexes of cell cycle factors. However, our model selection predicted that two cell cycle factors were enough to fit the present data without the need for complexes.

Reviewer #2 (Public Review):

This paper is of broad interest to scientists in the fields of cell growth, cell division, and cell-cycle control. Its main contribution is to provide a method to restrict the space of potential cell-cycle models using observed correlations in inter-division times of cells across their lineage tree. This method is validated on several data sets of bacterial and mammalian cells and is used to determine what additional measurements are required to distinguish the set of competing models consistent with a given correlation pattern.

The patterns of correlations in the division times of cells within their lineage tree contain information about the inheritable factors controlling cell cycles. In general, it is difficult to extract such information without a detailed model of cell cycle control. In this manuscript, the authors have provided a Bayesian inference framework to determine what classes of models are consistent with a given set of observations of division time correlations, and what additional observations are needed to distinguish between such models. This method is applied to data sets of division times for various types of bacterial and mammalian cells including cells known to exhibit circadian oscillations.

The manuscript is well-written, the analyses are thorough, and the authors have provided beautiful visualizations of how alternative models can be consistent with a finite set of observed correlations, and where and how extra measurements can distinguish between such models. Known models of growth rate correlations, cell-size regulation, and cell cycle control are analyzed within this framework in the Supplemental Information. A major advantage of the proposed method is that it provides a non-invasive framework to study the mechanism of cell-cycle control.

We thank the reviewer for the positive response to our manuscript.

Author Response

Reviewer #1 (Public Review):

In this manuscript the authors describe an approach for controlling cellular membrane potential using engineered gene circuits via ion channel expression. Specifically, the authors use microfluidics to track S. cerevisiae gene expression and plasma membrane potential (PMP) in single cells over time. They first establish a small engineered gene circuit capable of producing excitable gene expression dynamics through the combination of positive and negative feedback, tracking expression using GFP (Figure 1). Though not especially novel or complex, the data quality is high in Figure 1 and the results are convincing. Note that the circuit is excitable and not oscillatory; it is being driven periodically by a chemical inducer. I think the authors could have done a better job justifying the use of an excitable engineered gene circuit system, since you could get a similar result by just driving a promoter with the equivalent time course of inducer.

We restructured the manuscript by presenting the open-loop version of our synthetic circuit and demonstrate that closed loop system integrating feedbacks performs significantly better than its open-loop version (revised Figures 1 and 3). This open-loop system is based on Mar proteins that can synchronizes gene expression on extended spatiotemporal scales (PerezGarcia et al., Nat Comm, 2021). Other driven systems (i.e., TetR, AraC, LacI) can temporally synchronize gene expression in single bacteria cells to successive cycles of inducer. However, over time these bacterial systems build substantial delays in phases between cells, partially due to noise that ultimately led to desynchrony between individual cells even though they tend to follow the common inducer. This is clearly not the case in Mar-based systems (Perez-Garcia et al., Nat Comm, 2021) as eukaryotic cells synchronize to each other under guidance of common environmental stimuli with neglectable phase drift. Furthermore, in revised version we show that dual feedback strategy provides a robust solution to control ion channel expression and associated changes in PMP (see Conclusions lines 231-237).

The authors then use a similar approach to produce excitable expression of the bacterial ion channel KcsA, tracking membrane voltage using the voltage-sensitive dye ThT rather than GFP fluorescence (Figure 2). The experimental results in this figure are more novel as the authors are now using the expression of a heterologous ion channel to dynamically control plasma membrane potential. While fairly convincing, I think there are a few experimental controls that would make these results even more convincing. It is also unclear why the authors are now using power spectra to display observed frequencies compared to the much more intuitive histograms used in Figure 1.

Now we use violin plots with period distributions consistently in all figures to ease the comparison between scenarios.

Finally, the authors move on to use a similar excitable engineered gene circuit approach to produce inducible control of the K1 toxin which influences the native potassium channel TOK1 rather than the heterologous ion channel KcsA (Figure 3). I have a similar reaction to this figure as with Figure 2: the results are novel and interesting but would benefit from more experimental controls. Additionally, the image data shown in Figure 3b is very unclear and could be expanded and improved.

In revised version we have decided to remove K1 toxin data as we are aware that we cannot modulate K1 degradation rate due to its extracellular nature. Instead, we have decided to perform additional experiments in which we directly plugged our circuit to TOK1 native potassium channel to demonstrate that our feedback-integrating synthetic circuit is capable of controlling TOK1 dosage and associated PMP changes (revised Figure 3, and lines 209-220). We believe these new data make more direct connection between synthetic circuits phytohormones and native channel expression than presented earlier K1-based scenario.

Overall, in my opinion the claims in the abstract and title are a bit strong. I would deemphasize global coordination and "synchronous electrical signaling" since the authors are driving a global inducer. To make the claim of synchronous signaling I would want to see spatial data for cells near vs. far from K1 toxin producing cells in Figure 3 along with estimates of inducer/flow timescale vs. expression/diffusion of K1 toxin. As I read the manuscript, I see that most of the synchronicity comes from the fact that all cells are experiencing a global inducer concentration.

We agree with the Reviewer, synchronicity and global coordination comes from phytohormone sensing feedback circuit that is guided by cyclic environmental changes. We have revised definition of synchronous signaling as suggested, focusing on the macroscopic synchronization of ion channel expression achieved by external modulation, which is the key message coming from this work.

Reviewer #2 (Public Review):

The authors present a novel method to induce electrical signaling through an artificial chemical circuit in yeast which is an unconventional approach that could enable extremely interesting, future experiments. I appreciate that the authors created a computer model that mathematically predicts how the relationship between their two chemical stimulants interact with their two chosen receptors, IacR/MarR, could produce such effects. Their experimental validations clearly demonstrated control over phase that is directly related to the chemical stimulation. In addition, in the three scenarios in which they tested their circuit showed clear promise as the phase difference between spatially distant yeast communities was ~10%. Interestingly, indirect TOK1 expression through K1 toxin gives a nice example of inter-strain coupling, although the synchronization was weaker than in the other cases. Overall, the method is sound as a way to chemically stimulate yeast cultures to produce synchronous electrical activity. However, it is important to point out that this synchronicity is not produced by colony-colony communication (i.e., self-organized), but by a global chemical drive of the constructed gene-expression circuit.

The greatest limitation of the study lies in the presentation (not the science). There are two significant examples of this. First, the authors state this study 'provides a robust synthetic transcriptional toolbox' towards chemo-electrical coupling. In order to be a toolbox, more effort needs to be put into helping others use this approach. However little detail is given about methodological choices, circuit mechanisms in relation to the rest of the cell, nor how this method would be used outside of the demonstrated use case. Second, the authors stress that this method is 'non-invasive', but I fail to see how the presented methodology could be considered non-invasive, in in-vivo applications, as gene circuits are edited and a reliable way to chemically stimulate a large population of cells would be needed. It may be that I misunderstood their claim as the presentation of method and data were not done in a way that led to easy comprehension, but this needs to be addressed specifically and described.

We apologize Reviewer for a potential misunderstanding. By ‘non-invasive’ we meant that such systems would not need, for instance, the surgical installation of light components to control ion channel activity. Nonetheless, we have removed these confusing sentences from the revised manuscript.

The rational for using Mar-based system with feedback strategy data has been now presented in more structured and comprehensive way across the revised manuscript to demonstrate benefits from integrating feedback as well as potential of such systems for excitable dynamics with noise-filtering capability and faster responsiveness. We also show how system can be coupled to native potassium channels, opening ways to integrate synthetic circuit into other organisms.

In terms of classifying the synchronicity, while phase difference among communities was the key indicator of synchronization, there were little data exploring other aspects of coupled waveforms, nor a discussion into potential drawbacks. For example, phase may be aligned while other properties such as amplitude and typical wave-shape measures may differ. As this is presented as a method meant for adoption in other labs, a more rigorous analytical approach was expected.

In the revised manuscript, we have analyzed synchronicity using several different approaches:

(1) we calculate cumulative autocorrelations of response between communities.

(2) to complement autocorrelation analysis, we developed a quantitative metric of ‘synchrony index’ defined as 1 - R where R is the ratio of differences in subsequent ThT peak positions amongst cell communities (phase) to expected period. This metrics describes how well synchronized are fungi colonies with each other under guidance of the common environmental signal.

(3) we analyzed amplitudes and peak widths for all presented scenarios and we conclude that while periods and peak widths are robust across communities there is noticeable variation in amplitudes (i.e. Figure 3E).

We therefore believe that this multistep quantitative approach is rigorous in identifying oscillatory signal characteristics.

Reviewer #3 (Public Review):

We are enthusiastic about this paper. It demonstrates controlled expression of ion channels, which itself is impressive. Going a step further, the authors show that through their control over ion channel expression, they can dynamically manipulate membrane potential in yeast. This chemical to electrophysiological conversion opens up new opportunities for synthetic biology, for example development of synthetic signaling systems or biological electrochemical interfaces. We believe that control of ion channel expression and hence membrane potential through external stimuli can be emphasized more strongly in the report. The experimental time-lapse data were also high quality. We have two major critiques on the paper, which we will discuss below.

First, we do not believe the analyses used supports the authors' claims that chemical or electrical signals are propagating from cell-to-cell. The text makes this claim indirectly and directly. For example, in lines 139-141, the authors describe the observed membrane potential dynamics as "indicative of the effective communication of electrical messages within the populations". There are similar remarks in lines 144 and 154-156. The claim of electrical communication is further established by Figure 2 supplement 3, which is a spatial signal propagation model. As far as we can tell, this model describes a system different from the one implemented in the paper.

Second, it is not clear why the excitable dynamics of the circuit are so important or if the circuit constructed does in fact exhibit excitable dynamics. Certainly, the mathematical model has excitable dynamics. However, not enough data demonstrates that the biological implementation is in an excitable regime. For example, where in the parameter space of Figure 1 supplement 1 does the biological circuit lie? If the circuit has excitable dynamics, then the authors should observe something like Figure 1 supplement 1B in response to a nonoscillating input. Do they observe that? Do they observe a refractory period? Even if the circuit as constructed is not excitable, we don't think that's a major problem because it is not central to what we believe is the most important part of this work - controlling ion channel expression and hence membrane potential with external chemical stimuli.

We thank Reviewer for encouraging comments and positive evaluation of our work.

Author Response

Reviewer 1# (Public Review):

Purkinje cells (PCs) in the cerebellum extend axonal collaterals along the PC layer and within the molecular layer. Previous anatomical studies have shown the existence of these tracts and recently, the existence of functional synapses from PCs to PCs, molecular layer interneurons (MLIs), and other cell types was demonstrated by Witter et al., (Neuron, 2016) using optogenetics. In this manuscript, Halverson et al., first characterize the PC to MLI synapse properties in the slice using optogenetics and dual patch recordings. They then use computer simulations to predict the role of these connections in eyelid conditioning and test these predictions using in vivo recordings in rabbits. Authors claim that PCs fire before their target MLIs and that their activity is anticorrelated. They further suggest that the special class of MLIs receiving inhibitory input from PCs might serve to synchronize PCs during eyelid conditioning.

Major comments:

1) The manuscript is quite long with 9 main figure panels and 6 supplementary figures. The flow of the results is not smooth. While the first 4 figures are nicely done in terms of their results and organization, the same cannot be said about the rest of the figures.

To address this concern, we have revised the Results section extensively. We believe that it is now much more accessible and better integrated.

In fact, it would make sense to split the manuscript in two, one describing the synaptic properties and circuit mapping of the PC-PC-MLI circuit and the other describing their role in eyelid conditioning. As it stands, this manuscript is a tough read and difficult to get through.

We acknowledge that our results, which were done in two different labs and employed a variety of different techniques, could have been split into two (or even more) separate papers. However, we believe that there is high value to our readers in providing a comprehensive study that integrates many different types of analyses to attack the same fundamental question. That is why we chose to organize the content in the way that we did and that is why we still prefer to keep the entire story together. However, we do agree with the reviewer’s point that the previous version was unwieldy and too challenging to understand. Therefore, we have invested a lot of effort to improve the readability of the revised version.

Further, the authors have not connected the initial slice physiology with the later in vivo work to argue for their presence in the same paper. For example, the quantal content measurement, the short-term plasticity, the mobilization rate measurement, etc do not figure in the latter half of the manuscript at all. I strongly suggest carving figures 1-4 out into a separate manuscript.

The slice work motivated the computational simulation and the in vivo recordings of MLI activity. While it is true that it is hard to correlate every aspect of the slice work (e.g. quantal content measurements, etc.) with the in vivo recordings, and vice versa, there are elements of each that have informed the other. As a result, consistent properties of the PC-to-PC-MLI circuit emerged. We have highlighted the cross-connections more in the revised manuscript, including the following passages:

1) “Only a subset of MLIs (8.7%) showed clear inverse correlation with eyelid PCs and they were within approximately 120 µm of eyelid PCs. These connectivity rates and distances are comparable to our observations in cerebellar slices, where we found that approximately 5-6% of MLIs receive PC feedback inhibition (Figure 1b) that extends over 200 m or less (Figure 3).” (p.13, para 1)

2) “The pattern of cross-correlation between connected PCs and PC-MLIs was qualitatively similar to that observed in slices…” (p.14, para 2).

3) “This correspondence strengthens the conclusion that putative PC-MLI identified in vivo are equivalent to the PC-MLI identified in slices” (p. 15, para 1).

4) “The need for relatively large changes in PC activity in vivo highlights the importance of the frequency-independent synaptic synaptic transmission at the PC-to-PC-MLI synapse illustrated in Figure 2.” (p. 16, para. 1)

We have more closely harmonized the style of all figures, to subliminally emphasize the close connection between the slice and in vivo results.

Above we have addressed the suggestion to split the paper into two. Instead of breaking up the paper, we worked hard to better integrate the two parts and make them easier to read as a whole.

2) Authors conclude that eyelid PCs and eyelid PC-MLIs are inversely correlated and that PCs precede PC-MLIs during CRs and therefore could drive their activity. Both of these points are insufficiently justified by their analysis. First, it is not clear how eyelid PCs are identified – I’m assuming this is based on negative correlation with CRs just like positively correlated MLIs are assigned as eyelid PC-MLIs.

We apologize for failing to mention that eyelid PCs were identified by the presence of US -evoked (eyelid stimulation) complex spikes. This criterion is completely independent of the responses of PCs during expression of eyelid CRs and also provides an in vivo tool for identifying the “eyelid” region of the cerebellar cortex, which should also be where eyelid PC-MLIs are located. To address this omission, we have now describe the method used to identify eyelid PCs in the Methods section (p. 31, para. 2) and the Results section (p. 11, para. 2).

If this is how PCs and PC-MLIs are identified, then the inverse correlation between the two cell types results from this definition itself. And, their activity pattern during CRs, illustrated in many figure panels is hardly surprising.

Yes of course this would be circular logic, but it is not at all what we did! Again, we apologize for the confusion.

Second, to show that PCs fire ahead of PC-MLIs, the authors calculate the difference in fractional change in spike rate before and after the start of the CR (PC-MLI). Their reasoning is that if the bulk of firing rate change happened before the start of CR for PCs, but at the start or later for PC-MLIs, then this value will be positive, else it will be negative. The distribution of these values was shifted to the positive side leading them to conclude that PCs fire ahead of PC-MLIs. However, this is a huge logical jump. The sign of (PC-MLI) is dependent on the depth of modulation in each cell type as well and does not necessarily indicate relative timing. In any case, such caveats have not been ruled out in their analysis. This analysis to establish timing is unconvincing. Would it not be better to look at the timing of the spike modulation start directly rather than the round-about method they are using?

We agree with the reviewer that PC and PC-MLI activities undergo complex time-dependent changes, particularly during CRs, which makes it challenging to have a single parameter that uniquely represents the differences in timing between the activity of the two cell types. In our revised manuscript, we have addressed this issue by creating a new section that is entirely devoted to analysis of the temporal relationship between PC and PC-MLI activity (pp. 14-17). In brief, here are the main lines of evidence that PCs fire prior to PC-MLIs, both in baseline conditions and during conditioned eyelid responses.

We have provided 3 types of evidence that PCs fire prior to putative PC-MLIs during baseline activity:

1) A spike-triggered average of PC and putative PC-MLI activity during baseline firing showed a modest decrease in PC-MLI firing rate in response to a PC action potential (Figure 8a; see also Figure 8-figure supplement 1a and 1b).

2) A pause in PC activity caused a very substantial rise in activity in putative PC-MLIs (Figure 8c; see also Figure 8-figure supplement 1c).

3) A burst of PC activity caused a decline in putative PC-MLI activity (Figure 8d; see also Figure 8-figure supplement 1d).

We have an additional 3 lines of evidence showing that PCs fire prior to putative PC-MLIs during CRs:

1) Simultaneous recordings of the time course in changes in PC and putative PC-MLI activity during CRs indicate that PC activity usually declined prior to the activity of putative PC-MLIs. This is clearly visible in the examples shown in Figure 9c, as well as the averaged data shown in Figure 9-figure supplement 1a.

2) We measured the delay between the time at which PC activity reached 50% of its minimum during the CS and the time at which the activity of putative PC-MLIs reached 50% of its maximum during single trials. Whenever CRs were observed, PCs reached their half-maximal response before putative PC-MLIs did (Figure 9─figure supplement 1b).

3) We also measured the collective timing of changes in the activity of putative PC-MLIs and eyelid PCs during conditioning across all of our paired recordings. This was done by calculating a ratio representing the magnitude of changes in activity prior to CR onset, normalized to the peak amplitude of the change during the entire interval. The distribution of differences in the timing of changes in PC and PC-MLI activity has a mean that is greater than zero (Figure 10), indicating that eyelid PCs decreased their activity before putative PC-MLIs increased their activity in a majority of cases.

We hope that these improvements have adequately addressed the reviewer’s concern.

3) Many figure panels make the same point and appear redundant. For example, that PCs and PC-MLIs are inversely correlated with each other in vivo during CRs is shown in Figure 7, figure 8a, S2, S4, and S5. Of course, in each case, the data are sorted differently (according to ISI, CR initiation, cumulative distributions, etc.,) but surely, the point regarding inverse relationship can be conveyed more concisely?

As mentioned in response to the reviewer’s previous comment, we have made significant changes to this part of the manuscript, including creating a section that addresses the temporal relationship between PC and PC-MLI activity. This has involved removing some of the analyses listed by the reviewer, adding some new analysis and relegating some previous figures to the supplementary materials. We believe that these changes allow the manuscript to efficiently clarify the relationship between PC and PC-MLI activity and highlight the value of each type of analysis that is included.

4) Several details are missing in the methods section even though parts of it may have been published before. For instance, how are CRs calculated in the simulation? Methods state that 'The averaged and smoothed activity of the eight deep nucleus neurons was used to represent the output of the simulation and the predicted "eyelid response" of the simulation'. It is not clear what the nature of this transform is and if any calibration factors were used. How comparable are the simulated CRs in kinetics and amplitude to experimental CRs?

In response to this comment, we have revised the Methods section to include much more detail about the simulation methods, including two schematic diagrams. The methods employed for the experimental work in the paper are already described in detail.

The simulation can produce simulated CRs (smoothed histogram of nucleus activity) with kinematic variables that are comparable to experimental CRs. A detailed account of how this is accomplished is described in Medina and Mauk (2000) and are briefly summarized on p. 39, para.2: “The averaged and smoothed activity of the eight deep nucleus neurons was used to represent the output of the simulation and the predicted “eyelid response” of the simulation.” Although this approach is not intended to simulate the precise kinematics of an eyeblink, comparison of Figures 11a (simulation) and 6a (rabbit) show that there is a reasonable concordance. The real value of the simulation is in predicting the relative changes in eyelid responses that occur during conditioning.

Author Response

Reviewer #1 (Public Review):

Figures 2 through 6. There is no description of the relationship between the findings and the anatomical location of the electrodes (other than distal versus local). Perhaps the non-uniform distribution of electrodes makes these analyses more complicated and such questions might have minimal if any statistical power. But how should we think about the claims in Figures 2-6 in relationship to the hippocampus, amygdala, entorhinal cortex, and parahippocampal gyrus? As one example question out of many, is Figure 2C revealing results for local pairs in all medial temporal lobe areas or any one area in particular? I won't spell out every single anatomical question. But essentially every figure is associated with an anatomical question that is not described in the results.

To address the reviewer’s point we now report the distribution of spike-LFP pairs across anatomical regions for each Figure 2-6. The results split by anatomical regions are reported in Figure 2 – figure supplement 7, Figure 3 – figure supplement 7, Figure 4 – figure supplement 1, Figure 5 – figure supplement 2, and Figure 6 – figure supplement 3. We also calculated a non-parametric Kruskal-Wallis Test to statistically examine the effect of anatomical regions on the results shown in each figure. Generally, these new results show that the effects are similar across regions, apart from two exceptions (i.e. Figure 4 – supplement 1; and Figure 5 – supplement 2). However, we would like to stress that these results should be taken with a huge grain of salt because the electrodes were not evenly distributed across regions (i.e. ~75% of observations pertain to the hippocampus), and patients as the reviewer correctly points out. This leads to sometimes very low numbers of observations per region and it is difficult to disentangle whether any apparent differences are driven by regional differences, or differences between patients. Detailed results are reported below.

Manuscript lines 207-212: “In the above analysis all MTL regions were pooled together to allow for sufficient statistical power. Results separated by anatomical region are reported in Figure 2 – figure supplement 7 for the interested reader. However, these results should be interpreted with caution because electrodes were not evenly distributed across regions and patients making it difficult to disentangle whether any apparent differences are driven by actual anatomical differences, or idiosyncratic differences between patients.”

Manuscript lines 255-258: “Finally, we report the distal spike-LFP results separated by anatomical region in Figure 3 – figure supplement 7, which did not reveal any apparent differences in the memory related modulation of theta spike-LFP coupling between regions.”

Manuscript lines 264-266: “PSI results separated by anatomical regions are reported in Figure 4 – figure supplement 1, which revealed that the PSI results were mostly driven by within regional coupling.”

Manuscript lines 399-303: “We also analyzed whether the memory-dependent effects of cross-frequency coupling differ between anatomical regions (see Figure 5 – figure supplement 2). This analysis revealed that the results were mostly driven by the hippocampus, however we urge caution in interpreting this effect due to the large sampling imbalance across regions.”

Manuscript lines 343-346: “As for the above analysis we also investigated any apparent differences in co-firing between anatomical regions. These results are reported in Figure 6 – figure supplement 3 and show that the earlier co-firing for hits compared to misses was approximately equivalent across regions.”

Figure 1

1A. I assume that image positions are randomized during a cued recall?

Yes, that was the case. We now added that information in the methods section.

Manuscript lines 526: “Image positions on the screen were randomized for each trial.”

What was the correlation between subjects' indication of how many images they thought they remembered and their actual performance?

We did not log how many images the patients thought they remembered. Specifically, if the patients answered that they remembered at least one image, then they were shown the selection screen where they could select the appropriate images. Therefore, we cannot perform this analysis. We report this now in the methods section. However, albeit interesting, the results of such an analysis would not affect the main conclusions of our manuscript.

Manuscript lines 523-524: “The experimental script did not log how many images the patient indicated that they thought to remember.”

1B. Chance is shown for hits but not misses. I assume that hits are defined as both images correct and misses as either 0 or 1 image correct. Then a chance for misses is 1-chance for hits = 5/6. It would be nice to mark this in the figure.

Done as suggested (see Figure 1).

The authors report that both incorrect was 11.9%. By chance, both incorrect should be the same as both correct, hence also 1/6 probability, hence the probability of both incorrect seems quite close to chance levels, right?

Yes, that is correct, however, across sessions the proportion of full misses (i.e. both incorrect) was significantly below chance (t(39)=-1.9214; p<0.05). Nevertheless, the proportion of fully forgotten trials appears to be higher than expected purely by chance. This is likely driven by a tendency of participants to either fully remember an episode, or completely forget it, as demonstrated previously in behavioural work (Joensen et al., 2020; JEP Gen.). We report this now in the manuscript.

Manuscript lines 132-136: “Across sessions the proportion of full misses (i.e. both incorrect) was significantly below chance (t39=-1.92; p<0.05). However, the proportion of fully forgotten trials appears to be higher than expected purely by chance. This is likely driven by a tendency of participants to either fully remember an episode, or completely forget it, as demonstrated previously in behavioral work (25).”

1C. How does the number of electrodes relate to the number of units recorded in each area?

The distribution of neurons per region is shown in the new Figure 1D (see above). It approximately matches the distribution of electrodes per region, except for the Amygdala where slightly more neurons where recorded. This is because of one patient (P08) who had two electrodes in the left and right Amygdala and who contributed at lot of sessions (i.e. 9 sessions, comparing to an average of 4.44 per patient).

Line 152. The authors state that neural firing during encoding was not modulated by memory for the time window of interest. This is slightly surprising given that other studies have shown a correlation between firing rates and memory performance (see Zheng et al Nature Neuroscience 2022 for a recent example). The task here is different from those in other studies, but is there any speculation as to potential differences? What makes firing rates during encoding correlate with subsequent memory in one task and not in another? And why is the interval from 2-3 seconds more interesting than the intervals after 3 seconds where the authors do report changes in firing rates associated with subsequent performance? Is there any reason to think that the interval from 2-3 seconds is where memories are encoded as opposed to the interval after 3 seconds?

Zheng et al. used a movie-based memory paradigm where they manipulated transitions between scenes to identify event cells and boundary cells. They show that boundary cells, which made up 7.24% of all recorded MTL cells, but not event cells (6.2% of all MTL cells) modulate their firing rate around an event depending on later memory. There are quite a few differences between Zheng et al’s study and our study that need to be considered. Most importantly, we did not perform a complex movie-based memory paradigm as in Zheng et al. and therefore cannot identify boundary cells, which would be expected to show the memory dependent firing rate modulation. This alone could contribute to the fact that no significant differences in firing rates in the first second following stimulus onset were observed. Such an absence of a difference of neural firing depending on later memory is not unprecedented. In their seminal paper, Rutishauser et al. (2010; Nature) report no significant differences in firing rates (0-1 seconds after stimulus onset, which is similar to our 2-3 seconds time window) between later remembered or later forgotten images. This finding is also in line to Jutras & Buffalo (2009; J Neurosci) who also show no significant difference in firing rates of hippocampal neurons during encoding of remembered and forgotten images.

The 2-3 seconds time interval, which corresponds to 0-1 seconds after the onset of the two associate images, is special because it marks the earliest time point where memory formation can start, therefore allowing us to investigate these very early neural processes that set the stage for later memory-forming processes. While speculative, these early processes likely capture the initial sweep of information transfer into the MTL memory system which arguably is reflected in the timing of spikes relative to LFPs. It is conceivable that these initial network dynamics reflect attentional processes, which act as a gate keeper to the hippocampus (Moscovitch, 2008; Can J Exp Psychol) and thereby set the stage for later memory forming processes. This interpretation would be consistent with studies in macaques showing that attention increases spike-LFP coupling, whilst not affecting firing rates (Fries et al., 2004; Science). We modified the discussion section to address this issue.

Manuscript lines 468-474: “Interestingly, these early modulations of neural synchronization by memory encoding were observed in the absence of modulations of firing rates, which is consistent with previous results in humans (16) and macaques (12), but contrasts with (43). Studies in macaques showed that attention increases spike-LFP coupling whilst not affecting firing rates (44). It is therefore conceivable that these initial network dynamics reflect attentional processes, which act as a gate keeper to the hippocampus and thereby set the stage for later memory forming processes (45).”

Lines 154-157 and relationship to the subsequent analyses. These lines mention in passing differences in power in low-frequency bands and high-frequency bands. To what extent are subsequent results (especially Figures 3 and 4) related to this observation? That is, are the changes in spike-field coherence, correlated with, or perhaps even dictated by, the changes in power in the corresponding frequency bands?

To address this question we repeated the analysis that we performed for SFC for Power in those channels whose LFP was locally coupled to spikes in gamma, and distally coupled to spikes in theta. Furthermore, we correlated the difference in peak frequency between hits and misses between Power and SFC. If power would dictate the effects seen in SFC then we would expect similar effects of memory in power as in SFC, that is an increase of peak frequency for hits compared to misses for gamma and theta. Furthermore, we would expect to find a correlation between the peak frequency differences in power and SFC. None of these scenarios were confirmed by the data. These results are now reported in Figure 2 – figure supplement 5 for gamma, and Figure 3 – figure supplement 5 for theta.

Manuscript lines 195-199: “We also tested whether a similar shift in peak gamma frequency as observed for spike-LFP coupling is present in LFP power, and whether memory-related differences in peak gamma spike-LFP are correlated with differences in peak gamma power (Figure 2 – figure supplement 5). Both analyses showed no effects, suggesting that the effects in spike-LFP coupling were not coupled to, or driven by similar changes in LFP power.”

Manuscript lines 248-253: “As for gamma, we also tested whether a similar shift in peak theta frequency is present in LFP power, and whether there is a correlation between the memory-related differences in peak theta spike-LFP and peak theta power (Figure 3 – figure supplement 5). Both analyses showed no effects, suggesting that the effects in spike-LFP coupling were not coupled to, or driven by similar changes in LFP power.”

Do local interactions include spike-field coherence measurements from the same microwire (i.e., spikes and LFPs from the same microwire)?

Yes, they do. Out of the 53 local spike-SFC couplings found for the gamma frequency range, 11 (20.75%) were from pairs where the spikes and LFPs were measured on the same microwire. We assume that the reviewer is asking this question because of a concern that spike interpolation may introduce artifacts which may influence the spectrograms and consequently the spike-LFP coupling measures. This was also pointed out by Reviewer #2. To address this concern, we split the data based on whether the spike and LFP providing channels were the same or different. The results show that (i) the spectrogram of SFC is highly similar between the two datasets, with a prominent gamma peak present in both and no significant differences between the two; (ii) restricting the analysis to those data where the LFP and spike providing channels are different replicated the main finding of faster gamma peak frequencies for hits compared to misses; and (iii) limiting the SFC analysis further to only ‘silent’ channels, i.e. channels where no SUA/MUA activity was present at all also replicated the main finding of faster gamma peak frequencies for hits compared to misses.

These analyses suggest that the SFC results were not driven by spike interpolation artefacts.

Manuscript lines 199-203: “To rule out concerns about possible artifacts introduced by spike interpolation we repeated the above analysis for spike-LFP pairs where the spike and LFP providing channels are the same or different, and for ‘silent’ LFP channels (i.e. channels were no SUA/MUA activity was detected (see Figure 2 – figure supplement 6). “

60 Hz. It has always troubled me deeply when results peak at 60 Hz. This is seen in multiple places in the manuscript; e.g., Figures 2B, 2E. What are the odds that engineers choosing the frequency for AC currents would choose the exact same frequency that evolution dictated for interactions of brain signals? This is certainly not the only study that reports interesting observations peaking at 60 Hz. One strong line of argument to suggest that this is not line noise is the difference between conditions. For example, in Figure 2B, there is a difference between local and distal interactions. It is hard for me to imagine why line noise would reveal any such difference. Still ...

The frequency for AC currents in Europe is 50 Hz, not 60 Hz as in the US. Therefore, our SFC effects are well outside the range of the notch.

Figure 6. I was very excited about Figure 6, which is one of the most novel aspects of this study. In addition to the anatomical questions about this figure noted above, I would like to know more. What is the width of the Gaussian envelope?

The width of the Gaussian Window used in the original results was 25ms. We chose this time window because in our view it represents a good balance between integrating over a long-enough time window and thus allowing for some jitter in neural firing between pairs of neurons, whilst still being temporally specific. Finding the right balance here is not trivial because a too short time window underestimates co-firing, and a too long time window may not provide the temporal specificity necessary to detect co-firing lags (Cohen & Kohn, 2011; Nat Neurosci). To test whether this choice critically affected our results, we repeated the analysis for different window sizes, i.e. 15, 35, and 45 ms. The results show that the pattern of results did not change, with hits showing earlier peaks in co-firing compared to misses. Critically, the difference in co-firing peaks was significant for all window sizes, except for the shortest one which presumably is due to the increase in noise because of the smaller time window over which spikes are integrated. These issues are now mentioned in the methods section, and the results for the different window sizes are reported in Figure 6 – figure supplement 4.

Manuscript lines 346-347: “The co-firing analyses were replicated with different smoothing parameters (see Figure 6 – figure supplement 4).”

Manuscript lines 894-898: “We chose this time window because it should represent a good balance between integrating over a long-enough time window and thus allowing for some jitter in neural firing between pairs of neurons, whilst still being temporally specific (57). To test whether this choice critically affected our results, we repeated the analysis for different window sizes, i.e. 15, 35, and 45 ms (see Figure 6 – figure supplement 4).”

Are these units on the same or different microwires?

All units used for the analysis shown in Figure 6 come from different microwires. This was naturally the case because the putative up-stream neuron was distally coupled to the theta LFP, and the putative down-stream neuron was locally coupled to gamma at this same theta LFP electrode. This information is listed in Figure 6 – source data 1 which lists the locations and electrode IDs for all neuron pairs shown in figure 6.

How do the spike latencies reported here depend on the firing rates of the two units?

To address this question we first tested whether firing rates (averaged across the putative up-stream and down-stream neurons) differ between hits and misses. If they do, this would be suggestive of a dependency of the spike latency differences between hits and misses on firing rates. No such difference was observed (p>0.3). Second, we correlated the differences between hits and misses in Co-firing peak latencies with the differences in firing rates. Again, no significant correlation was observed (R=-0.06; p>0.7), suggesting that firing rates had no influence on the observed differences in co-firing latencies. These control analyses are now reported in the main text.

Manuscript lines 347-350: “No significant differences in firing rates between hits and misses were found (p>0.3), and on correlations between firing rates and the co-firing latencies were obtained (R=-0.06; p>0.7), suggesting that firing rates had no influence on the observed co-firing differences between hits and misses.”

What do these results look like for other pairs that are not putative upstream/downstream pairs?

As we reported in the original manuscript in lines 352-355 we did not find a memory dependent effect on co-firing latencies if we select neuron pairs solely on the basis of distal theta SFC. Within this analysis the distally theta coupled neuron would be the up-stream neuron and the neuron recorded at the site where the theta LFP is coupled would be the down-stream neuron. This null-result suggests that in order for the memory dependent difference in co-firing lags to emerge, the down-stream neurons need to be coupled to a local gamma rhythm in order for the memory effect on co-firing latencies to emerge. However, within this previous analysis there is still a notion of up-stream and down-stream neurons because neuron pairs were selected based on distal theta phase coupling. We therefore repeated this analysis for all pairs of neurons in a completely unconstrained fashion such that all possible pairs of neurons that were recorded from different electrodes were entered into the co-firing analysis. This analysis also revealed no difference in co-firing lags, neither for positive lags nor for negative lags. Instead, what this analysis showed is tendency for hits to show a higher occurrence of simultaneous or near simultaneous firing, which is in line with Hebbian learning. These results are now reported in Figure 6 – figure supplement 1.

Manuscript lines 333-335: “In addition, a completely unconstrained co-firing analysis where all pairs possible pairings of units were considered also showed no systematic difference in co-firing lags between hits and misses (Figure 6 – figure supplement 1).”

Reviewer #2 (Public Review):

Roux et al. investigated the temporal relationship between spike field coherence (SFC) of locally and distally coupled units in the hippocampus of epilepsy patients to successful and unsuccessful memory encoding and retrieval. They show that SFC to faster theta and gamma oscillations accompany hits (successful memory encoding and retrieval) and that the timing of the SFC between local and distal units for hits comports well with synaptic plasticity rules. The task and data analyses appear to be rigorously done.

Strengths: The manuscript extends previous work in the human medial temporal lobe which has shown that greater SFC accompanies improved memory strength. The cross-regional analyses are interesting and necessary to invoke plasticity mechanisms. They deploy a number of contemporary analyses to disentangle the question they are addressing. Furthermore, their analyses address limitations or confound that can arise from various sources like sample size, firing rates, and signal processing issues.

Weaknesses:

Methodological:

The SFC coherence measures are dependent in part on extracting LFPs derived from the same or potentially other electrodes that are contaminated by spikes, as well as multiunit activity. In the methods, they cite a spike removal approach. Firstly, the incomplete removal or substitution of a signal with a signal that has a semblance to what might have been there if no spike was present can introduce broadband signal time-locked to the spike and create spurious SFC. Can the authors confirm that such an artifact is not present in their analyses? Secondly, how did they deal with the removal of the multiunit activity? It would be suspected that the removal of such activity in light of refractory period violation might be more difficult than well-isolated units, and introduce artifacts and broadband power, again which would spuriously elevate SFC. Conversely, the lack of removal of multiunit activity would seem to for a surety introduce significant broadband power. One way around this might be that since it is uncommon to have units on all 8 of the BF microwires, to exclude the microwire(s) with the units when extracting the LFP to avoid the need to perform spike removal.

The reviewer raises a valid concern which we address as follows. Firstly, an artefact introduced into SFC by linear interpolation would be a problem for those local SFCs where the spike providing channel and the LFP providing channel are identical. Out of the 53 local spike-SFC couplings found for the gamma frequency range, only 11 (20.75%) were from pairs where the spikes and LFPs come from the identical microwire. It is unlikely that this minority of data would have driven the results. Furthermore, it is unlikely that the interpolation would introduce a frequency shift of SFC that is memory dependent, because the interpolation is more likely to cause a general increase in broadband SFC (as opposed to having a frequency band specific effect). However, to address this concern, we split the data based on whether the spike and LFP providing channels were the same or different. The results show that (i) the spectrogram of SFC is highly similar between the two datasets, with a prominent gamma peak present in both and no significant differences between the two; (ii) restricting the analysis to those data where the LFP and spike providing channels are different replicated the main finding of faster gamma peak frequencies for hits compared to misses.

Secondly, we followed the reviewer’s suggestion and repeated the SFC analysis for ‘silent’ microwires, i.e. microwires where no single or multi-units were detected. This analysis replicated the same memory effects as observed in the analysis with all microwires. Specifically, we found an increase in the local gamma peak SFC frequency for hits compared to misses, as well as an increase in distal theta peak SFC frequency for hits compared to misses. These results are reported in the main manuscript and in Figure 2 – figure supplement 6 for gamma, and figure 3 – figure supplement 6 for theta.

Manuscript lines 199-203: “To rule out concerns about possible artifacts introduced by spike interpolation we repeated the above analysis for spike-LFP pairs where the spike and LFP providing channels are the same or different, and for ‘silent’ LFP channels (i.e. channels were no SUA/MUA activity was detected (see Figure 2 – figure supplement 6).”

Manuscript lines 253-255: “We also repeated the above analysis for spike-LFP pairs by only using ‘silent’ LFP channels (i.e. channels were no SUA/MUA activity was detected (see Figure 3 – figure supplement 6) to address possible concerns about artefacts introduced by spike interpolation.”

In a number of analyses the spike train is convolved with a Gaussian in places with a window length of 250ms and in others 25ms. It is suspected that windows of varying lengths would induce "oscillations" of different frequencies, and would thus generate results biased towards the window length used. Can the authors justify their choices where these values are used, and/or provide some sensitivity analyses to show that the results are somewhat independent of the window length of the Gaussian used to convolve with the times series.

The different choices in window length for the Gaussian convolution reflect the different needs of the two analyses where these convolutions were applied. In one analysis we wanted to get a smooth estimate of spike densities that we can average across trials, similar to a peri-stimulus spike histogram. For this analysis we used a window length of 250 ms which we found appropriate to yield a good balance between retaining smooth time courses whilst still being temporally sensitive. Importantly, for the statistical analysis of the firing rates, spike densities were averaged in much larger time windows than 250 ms (i.e. 1 – 2 seconds) therefore our choice of window length for spike densities would not have any bearing on the averaged firing rate analysis.

In the other analysis, which is more central for our manuscript, we used a cross-correlation between spike trains to estimate co-firing lags in the range of milliseconds. Therefore, this analysis necessitated a much higher temporal precision. We used a Gaussian Window with a width of 25ms because it represents a good balance between integrating over a long-enough time window and thus allowing for some jitter in neural firing between pairs of neurons, whilst still being temporally specific. Finding the right balance here is not trivial because a too short time window would be prone to noise and underestimates co-firing, whereas a too long time window may not provide the temporal specificity necessary to detect co-firing lags (Cohen and Kohn, 2013; Nat Neurosci). To test whether this choice critically affected our results, we repeated the analysis for different window sizes, i.e. 15, 35, and 45 ms. The results show that the basic pattern of results did not change, with hits showing earlier peaks in co-firing compared to misses. Critically, the difference in co-firing peaks was significant for all window sizes, except for the shortest one which is likely due to the increase in noise because of the smaller time window over which spikes are integrated. These issues are now mentioned in the methods section, and the results for the different window sizes are reported in Figure 6 – figure supplement 4.

Manuscript lines 346-347: “The co-firing analyses were replicated with different smoothing parameters (see Figure 6 – figure supplement 4).”

Manuscript lines 894-898: “We chose this time window because it should represent a good balance between integrating over a long-enough time window and thus allowing for some jitter in neural firing between pairs of neurons, whilst still being temporally specific (57). To test whether this choice critically affected our results, we repeated the analysis for different window sizes, i.e. 15, 35, and 45 ms (see Figure 6 – figure supplement 4).”

Conceptual:

The co-firing analyses are very interesting and novel. In table S1 are listed locally and distally coupled neurons. There are some pairs for example where the distally coupled neuron is in EC and the downstream one in the hippo, and then there is a pair that is the opposite of this (dist: hippo, local EC). There appear to be a number of such "reversal", despite the delay between these two regions one would assume them to be similar in sign and magnitude given the units are in the same two regions. It seems surprising that in two identical regions of the hippo the flow of information or "causality", could be reversed, when/if one assumes information flows through the system from EC to hippo. This seems unusual and hard to reconcile given what is known about how information flows through the MTL system.

The reviewer is correct that the spike co-firing analysis suggests a bi-directional flow of information between the hippocampus and surrounding MTL regions (e.g. entorhinal cortex; see Figure 6 – figure supplement 3). However, this bi-directional flow of information is not incompatible with neuroanatomy and the memory literature. The entorhinal cortex serves as an interface between the hippocampus and the neocortex with superficial layers providing input into the hippocampus (via the perforant pathway), and the deeper layers receiving output from the hippocampus (van Strien et al., 2009; Nat Rev Neurosci). Therefore, on a purely anatomical basis we can expect to see a bi-directional flow of information between the hippocampus and the entorhinal cortex, albeit in different layers. Importantly, reversals as shown in our Figure 6 – source data 1 involved different microwires and therefore different neurons (i.e. the entorhinal unit in row 1 was recorded from microwire 3, whereas the entorhinal unit in row 2 was recorded from microwire 8). It is conceivable that these two neurons correspond to different layers of the entorhinal cortex and therefore reflect input vs. output paths. Moreover, studies in humans demonstrated that successful encoding of memories depends not only on the input from the entorhinal cortex into the hippocampus, but also on the output of the hippocampal system into the entorhinal cortex, and indeed on the dynamic recurrent interaction between these input and output paths (Maass et al. 2014; Nat Comms; Koster et al., 2018; Neuron). Our bi-directional couplings between hippocampal and surrounding MTL regions (such as the EC) are in line with these findings. We have added a discussion of this issue in the discussion section.

Manuscript lines 447-452: “Notably, the neural co-firing analysis indicates a bidirectional flow of information between the hippocampus and surrounding MTL areas, such as the entorhinal cortex (see Figure 6 – figure supplement 3; Figure 6 – source data 1). This result parallels other studies in humans showing that successful encoding of memories depends not only on the input from surrounding MTL areas into the hippocampus, but also on the output of the hippocampal system into those areas, and indeed on the dynamic recurrent interaction between these input and output paths (43, 44).”

Author Response

Reviewer #3 (Public Review):

Canetta et al have characterized the developmental regulation of PV neurons in PFC. The experiments have been carefully conducted and even though this is an area of broad scientific interest, there are several issues that require consideration.

1) The dosing regime of the CNO that has been employed will not provide persistent inhibition. Inhibition will operate on a 16 hr on/ 8 hr off cycle. Under such circumstances, it will be very difficult to rule out interspersed inhibition-related artifacts.

Our approach of twice daily injections of CNO is consistent with that of other publications that have used similar chemogenetic approaches to chronically alter activity1-3. However, the reviewer is correct that our twice daily CNO injection protocol may only intermittently inhibit PV cells, and it is possible that persistent inhibition might result in even stronger behavioral and circuit effects. However, it is also possible that more continuous CNO administration could lead to hM4DGi desensitization. Given these caveats, we respectfully submit that repeating all the experiments under conditions that would allow constant chronic dispensation of CNO (such as implantation of minipumps) is an excellent future experiment but currently outside the scope of this manuscript.

2) The second major issue with the dosing regime is that it is long (35 days). Realizing that the development of PFC circuitry is complex but at P90, the animals will have been dosed for more than a third of their lives. How can the authors rule out compensatory changes that do not have anything to do with critical periods?

In future studies we hope to refine the timing of the developmental window mediating these long-term effects by comparing inhibiting during shorter developmental periods. Our current studies demonstrate that a 35-day window of inhibition during development, but not during adulthood, leads to long-lasting effects on behavior and prefrontal network function. If the developmental manipulation is more impactful because the manipulation represents a longer proportion of the animals’ lifetime, we would expect that the effect should wane as the animal gets older. However, for the rescue experiments that take place at P120 and P130, rather than P90, we still find that the Dev Inhibition animals are impaired, suggesting that it is not the proportion of the animals’ lifetime that has been inhibited, but the timing during which this inhibition occurs, that matters most.

3) To this point, in the discussion first para line 8 - please change "transient" to something more suitable to reflect the duration of treatment.

We have replaced transient with reversible.

Author Response

Reviewer #1 (Public Review):

This is a well performed study to demonstrate the antiviral function and viral antagonism of the dynein activating adapter NINL. The results are clearly presented to support the conclusions.

This reviewer has only one minor suggestion to improve the manuscript.

Add a discussion (1) why the folds of reduction among VSV, SinV and CVB3 were different in the NINL KO cells and (2) why the folds of reduction of VSV in the NINL KO A549 and U-2 OS cells.

Thank you for this suggestion. We have amended the results section to include additional information about these observations and possible explanations for these results.

Reviewer #2 (Public Review):

This manuscript is of interest to readers for host-viral co-evolution. This study has identified a novel human-virus interaction point NINL-viral 3C protease, where NINL is actively evolving upon the selection pressure against viral infect and viral 3Cpro cleavage. This study demonstrates that the viral 3Cpros-mediated cleavage of host NINL disrupts its adaptor function in dynein motor-mediated cargo transportation to the centrosome, and this disruption is both host- and virus-specific. In addition, this paper indicates the role of NINL in the IFN signaling pathway. Data shown in this manuscript support the major claims.

In this paper, the authors have identified a novel host-viral interaction, where viral 3C proteases (3Cpro) cleave at specific sites on a host activating adaptor of dynein intracellular transportation machinery, ninein-like protein (NINL or NLP in short) and inhibit its role in the antiviral innate immune response.

The authors firstly found that, unlike other activating adaptors of dynein intracellular transportation machinery, NINL (or NLP) is rapidly evolving. Thus, the authors hypothesized that this rapid evolution of NINL was caused by its interaction with viral infection. The authors found that viruses replicated higher in NINL knock-out (KO) cells than in wild-type (WT) cells and the replication level was not attenuated upon IFNa treatment in NINL KO cells, unlike in WT cells. Next, the authors investigated the role of NINL in type I IFN-mediated immune response and found that the induction of Janus kinase/signal transducer and activation of transcription (JAK/STAT) genes were attenuated in NINL KO cells upon IFNa treatment. The author further showed that the reduction of replication IFNa sensitive Vaccinia virus mutant upon IFNa treatment was decreased in NINL KO A549 cells compared to WT cells. The authors further showed that the virus antagonized NINL function by cleaving it with viral 3Cpro at its specific cleavage sites. NINL-peroxisome ligation-based cargo trafficking visualization assay showed that the redistribution of immobile membrane-bound peroxisome was disrupted by cleavage of NINL or viral infection.

This paper has revealed a novel host-virus interaction, and an antiviral function of a rapidly evolving activating adaptor of dynein intracellular transportation machinery, NINL. The major conclusions of this paper are well supported by data, but several aspects can be improved.

1) It would be necessary to include a couple of other pathways involved in innate immune response besides JAK/STAT pathway.

We are very interested in this question as well. Our RNAseq data (Supplementary file 4 and Figure 3 – Figure supplement 4) suggest that there are several transcriptional changes that result from NINL KO. Our goal in this manuscript was to focus on IFN signaling in order to understand this specific effect of NINL KO since it might have wide-ranging consequences on viral replication. While we agree that broadening our studies to other signaling pathways, including other pathways involved in innate immune response, is a good idea, we feel that those experiments would take longer than two months to perform and therefore fall outside of the scope of this paper.

2) The in-cell cleavages of NINL by viral 3Cpros were well demonstrated and supported by data of high quality. A direct biochemical demonstration of the cleavage is needed with purified proteins.

We agree with the reviewer that a direct biochemical cleavage assay would further demonstrate that viral 3Cpros cleave NINL specifically. However, our attempts to purify full-length NINL have been unsuccessful due to solubility issues (see example gel below), which is not surprising given that NINL is a >150 kDa human protein that has multiple surfaces that bind to other human proteins. As such, we focused our efforts on in-cell cleavage assays using specificity controls for cleavage. Specifically, we used catalytically inactive CVB3 3Cpro to show a dependence on protease catalytic activity and a variety of NINL constructs in which the glutamine in the P1 position is replaced by an arginine to show site specificity of cleavage. Notably, the cleavage sites in NINL that we mapped using this mutagenesis were predicted bioinformatically from known sites of 3Cpro cleavage in viral polyproteins, further indicating that cleavage is 3Cpro-dependent. We believe these results thus demonstrate that cleavage of NINL is dependent on viral protease activity and occurs in a sequence-specific manner. In light of the difficulty of purifying full-length NINL that would make biochemical experiments very challenging and likely take longer than two months to perform, we believe that our in cell data should be sufficient to demonstrate activity-dependent site-specific cleavage of NINL by viral 3Cpros.

Sypro stained SDS-PAGE gel showing supernatant (S) and insoluble pellet (P) fractions across multiple purifications with altered buffer conditions.

3) The author used different cell types in different assays. Explain the rationale with a sentence for each assay.

Throughout this work, we choose to use a variety of cell lines for specific purposes. A549 cells were chosen as our main cell line as they are widely used in virology, are susceptible to the viruses we used, are responsive to interferon, and express both NINL and our control NIN at moderate levels. In the case of our virology and ISG expression data, we performed the same experiments with NINL KOs in other cell lines confirm that the phenotypes we observed in A549 cells could be attributed to the absence of NINL rather than off-target CRISPR perturbations or cell-line specific effects. All cleavage experiments were performed in HEK293T for their ease of transfection and protein expression. The inducible peroxisome trafficking assays were performed in U-2 OS cells as their morphology is ideal for observing the spatial organization of peroxisomes via confocal microscopy, and based on the fact that we had recapitulated the virology results and ISG expression results in those cells. At the suggestion of the reviewer, we have amended the text to include rationales where appropriate.

4) While cell-based assays well support the conclusions in this paper, further demonstration in vivo would be helpful to provide an implication on the pathogenicity impact of NINL.

We agree. However, we believe that examining the impact of the loss of or antagonism of NINL on the pathogenesis of infectious diseases in an in vivo model is outside the scope of this study.

In summary, this manuscript contributes to a novel antiviral target. In addition, it is important to understand the host-virus co-evolution. The use of the evolution signatures to identify the "conflict point" between host and virus is novel.

Author Response

Reviewer #3 (Public Review):

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

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

1) The context of how different structures of the Platynereis larval body, by changing their position, move the body needs much more introduction than the short paragraph given at the end of the Introduction.

-My understanding is that the larval body is segmented, and contraction of the segments can cause a certain type crawling or swimming: does it? Do the longitudinal muscles, for example, insert at segment boundaries, and alternating contraction left-right cause some sort of "wiggling" or peristalsis?

Longitudinal muscles do not insert only at segment boundaries, but have desmosomal connections along the entire length of the cell. Individual longitudinal muscle cells can span up to 3 segments. However the cells are staggered in such a way that all longitudinal muscle cells with somas in one segment can collectively cover up to 4 segments. Longitudinal muscles are involved in turning when swimming (Randel et al., 2014). The undulatory trunk movements and parapodial walking movements are due to the contraction of oblique and parapodial muscles. The longitudinal muscles provide support during crawling (via desmosomal links) but it is unlikely that these muscles contract segmentally. Disentangling the distinct contributions of 53 types of muscles during crawling will require further studies.

-In addition, there are segmental processes (parapodia, neuropodia), and embedded in them are long chitinous hairs (Chaetae, Acicula). Do certain types of the muscles described in the study insert at the base of the parapodia/neuropodia (coming from different angles), such that contraction would move the entire process, including the chaetae/acicula embedded in their tips?

Yes, acicular muscles insert at the proximal base of the acicula, and by moving the acicula they move the entire noto-/neuropodia. We have presented the anatomy of all acicular and chaetal muscles types in the figures and videos.

-Or is it that only these chaetae/acicula move, by means of muscles inserting at their base (the latter is clearly part of the story)? Or does both happen at the same time: parapodium moves relative to the trunk, and chaeta/acicula moves relative to the parapodium? How would these movements lead to different kind of behaviors?

-Diagrams should be provided that shed light on these issues.

We have extended Video2 to show individual muscles and their relation to the aciculae in one of the parapodia. We also clarified this in the text:

“Several acicular muscles attach on one end to the proximal base of the aciculae and on the other end to the paratrochs and epidermal cells. Oblique muscles attach to the basal lamina, epidermal and midline cells at their proximal end, run along the anterior edge of parapodia and attach to epidermal and chaetal follicle cells at their distal tips. Both of these muscle groups are involved in moving the entire parapodium. Acicular muscles move the proximal tips of the aciculae, while oblique muscles move the parapodium by moving the tissue around the chaetae and the aciculae. All acicular movements also correspond to parapodial movements. Chaetae are embedded in the parapodium and therefore move with it, but the chaetal sac muscles can also independently retract the chaetae into the parapodium or protract them and make them fan out.”

2) The main problem I have with the analysis is the way a muscle cell is treated, namely as a "one dimensional" node, rather than a vector.

-In the current state of the analysis, the authors have mapped all desmosomes of a given muscle cell to its attached "target" cell. But how is that helpful? The principal way a muscle cell acts is by contracting, thereby pulling the cells it attaches to at its two end closer together. As the authors state (p.4) "...desmosomes..are enriched at the ends of muscle cells indicating that these adhesive structures transmit force upon muscle-cell contraction."

At the level of the current analysis our data reveal which cells may be moved by the contractions of the individual muscle cells. The reviewer is right that treating a muscle as a vector (or set of vectors) would be a more accurate description, which would potentially also open up the possibility of computational modelling. We have provided such a vectorised dataset in the revised version, where each muscle-cell skeleton is subdivided into short linear segments (Figure2–source–data 2). This dataset may be useful to approach the problem with a three dimensional approach, which is beyond the scope of the current analysis. We also included an additional video (Video 7) showing examples of muscles and their partners where the cells and the desmosomes connecting them are highlighted. This reveals that the desmosomes connecting two cells are often at the very end of the muscle cell.

-for that reason, the desmosomes at the muscle tips have to be treated as (2) special sets. Aside from these tip desmosomes there are other desmosomes (inbetween muscles, for example), but they (I would presume) have a very different function; maybe to coordinate muscle fiber contraction? Augment the force caused by contraction?

Desmosomes between muscles only occur between muscles of different types, not for homotypic connections. There are other types of junctions (adhaerens-like junctions) that connect individual cells of a muscle bundle together (not analysed here). We clarified this in the text.

• As far as I understand for (all of) the desmosome connectome plots, there is no differentiation made between desmosome subsets located at different positions within the muscle fiber. I therefore don't see how the plots are helpful to shed light on how the multiplicity of muscles represented in the graphs cause specific types of neurons.

We would like to point out that the cells and structures that muscles connect to via desmosomes are very likely the parts of the body that will move during the contraction of the muscle or will provide structural support (e.g. basal lamina) for the muscle cell to contract. This is most evident in the parapodial complex. The majority of muscles in the body connect to the aciuclar folliclecells and the aciculae are the most actively moving parts in the body during crawling (see Video 4). In any case, since we provide all skeleton reconstructions and the xyz coordinates of all desmosomes, the data could be further analysed following these suggestions by the reviewer.

• As it stands these plots "merely" help to classify muscles, based on their position and what cell type they target: but that (certainly useful) map could have probably also be achieved by light microscopic analysis.

This has never been achieved by light microscopy analysis in the hundreds of papers on invertebrate muscle anatomy (e.g. by phalloidin staining). For an LM analysis, it would not be sufficient to label the muscle fibres, but one would also need to label the desmosomes and a multitude of non-muscle cell types including the extent of their cytoplasm. This is technically very challenging (we would nevertheless be happy to hear specific suggestions for markers etc. from the Reviewer). Currently, only EM provides the required depth of structural information and resolution. This is why we believe that our dataset and analysis is unique, despite over a century of research in invertebrate anatomy.

3) Section "Local connectivity and modular structure of the desmosomal connectome" p.4-7" undertakes an analysis of the structure of the desmosome network, comparing it with other networks.

-What is the rationale here? How do the conclusions help to understand how the spatial pattern of muscles and their contraction move the body?

We hope that our analysis may also be of interest to the community of network scientists and we believe that the reconstruction of a quite large and novel type of biological network warrants a more quantitative network analysis, using the standard methods and measures of network science – as we presented e.g. in Figure 4 – even if these mathematical analyses may not directly reveal how muscles move the body. We hope that some readers with an interest in quantitative analyses will also appreciate the broader picture here.

-Isn't, on the one hand (given that position of the desmosome was apparently not considered), the finding that desmosome networks stand out (from random networks) by their high level of connectivity ("with all cells only connecting to cells in their immediate neighbourhood forming local cliques") completely expected?

We disagree that the result was completely expected. Even if this was the case, we think it is quite different to say that a result is expected or to thoroughly quantify certain parameters and mathematically characterise key properties of the desmosomal graph (as we have done). These network analyses help to conceptualise our findings and to think about the muscle system in more global, whole-body terms.

-On the other hand, does this reflect the reality, given that (many?) muscle cells are quite long, connecting for example the anterior border of a segment with the posterior border.

Indeed, a quantitative analysis helped us to identify cases where the reality deviated somewhat from what was completely expected, and we thank the reviewer for these comments. As we explain in the revised version, some longitudinal muscles show an unexpected position in the force-field layout of the graph, due to their long-range connections. We have added extra clarifications to the text: “To analyse how closely the force-field-based layout of the desmosomal connectome reflects anatomy, we coloured the nodes in the graph based on body regions (Figure 5). In the force-field layout, nodes are segregated by body side and body segment. Exceptions include the dorsolateral longitudinal muscles (MUSlongD) in segment-0. These cells connect to dorsal epidermal cells that also form desmosomes with segment-1 and segment-2 MUSlongD cells. These connections pull the MUSlongD_sg0 cells down to segment-2 in the force-field layout (Figure 5D).”

1. In the section "Acicular movements and the unit muscle contractions that drive them" the authors record movement of the acicula and correlate it with activity (Ca imaging) of specific muscle types. This study gives insightful data, and could be extended to all movements of the larva.

-The fact that a certain muscle is active when the acicula moves in a certain direction can be explained (in part) by the "connectivity": as shown in Fig.7L, the muscle inserts at a acicular follicle cell on the one side, and to an epithelial (epidermal?) cell and the basal lamina on the other side. But how meaningful is a description at this "cell type level" of resolution? The direction of acicula deflection depends on where (relative to the acicula base) the epithelial cell (or point in the basal lamina) is located. This information is not given in the part of the connectome network shown in Fig.7L, or any of the other graphs.

This information is indeed not shown in the graphs, where each cell is treated as a node. However, we provide this information in the detailed anatomical figures in Figure 6 – figure supplement 1-3 and Video 7, where the individual acicular and oblique muscle types are visualised. In principle, one could subdivide aciculae into e.g. proximal and distal halves and derive a more detailed network. We have not done this but since all the EM, anatomical rendering and connectivity data are available in our public CATMAID server (https://catmaid.jekelylab.ex.ac.uk/), we hope that the interested readers will be able to further analyse the data.

We renamed ‘epithelial’ cells to ‘epidermal’ cells.

Author Response

Reviewer #1 (Public Review):

Using a large neonatal dataset from the developmental Human Connectome project, Li and colleagues find that cortical morphological measurements including cortical thickness are affected by postnatal experience whereas cortical myelination and overall functional connectivity of ventral cortex developed significantly were not influenced by postnatal time. The authors suggest that early postnatal experience and time spent inside the womb differentially shape the structural and functional development of the visual cortex.

The use of large data set is a major strength of this study, furthermore an attempt to examine both structural and functional measures, and connectivity analysis and separating these analyses based on the pre-and full-term infants is impressive and strengthens the claims made in the paper. While I find this work theoretically well-motivated and the use of the large dHCP dataset very exciting, there are some concerns, that need to be addressed.

There is a bit of confusion if the authors really compared the structural-functional measures in the final analysis. If the authors wish to make claims about the relationship, then there must be a compelling analysis detailing these findings.

Thanks for the suggestions. We have added analysis to directly investigate the relationship between the development of homotopic connection and corresponding structural measurements in the area V1 (Page 13 Line 5-16):

“The above results revealed that structural and functional properties of the ventral visual cortex both developed with PMA, but were differently influenced by the in-utero and external environment (Table 1). We further investigated the relationship between structural and functional development based on area V1, which showed a strong developmental effect in both structural and functional analyses. Mediation analysis was employed to see whether the development (GA or PT) of the homotopic connection between bilateral V1 was mediated by the structural properties (CT or CM). We found that the PT had a significant direct effect on the homotopic function that was not mediated by CT or CM (Fig 6a-b). In contrast, the direct effect of GA on the homotopic connection was not significant but the indirect effect of GA through CM on the connection was significant (Fig 6c-d).”

There is also a bit of confusion in the terminology used in the study regarding ages; the gestational age, premenstrual age, and postnatal time. I think clarifying and simplifying it down to GA and postnatal time will help the reader and avoid confusion.

Thank you for the suggestion. We have made extensive revision regarding the terminology throughout the paper and simplified it down to GA and PT. Please see the response to the 1st major concern in the Essential Revisions (for the authors) section above.

*Reviewer #2 (Public Review):

The authors utilize the publicly available dHCP dataset to ask an interesting question: how does postnatal experience and prenatal maturation influence the development of the visual system. The authors report that experience and prenatal maturation differentially contribute to different aspects of development. Namely, the authors quantify cortical thickness, myelination, and lateral symmetry of function as three different metrics of development. The homotopy and preterm infant analyses are strengths that, on their own, could have justified reporting. However, I have concerns about the analytic approaches that were used and the conclusions that were drawn. Below I list my major concerns with the manuscript.

PMA vs. GA vs. PT

The authors seek to understand the contribution of experience and prenatal development, yet I am unsure why the authors focused on the variables they did. There are three variables of interest used throughout this study: Gestational age at birth (GA), postnatal time (PT), and postmenstrual age at the time of scan (PMA). The last metric, PMA, is straightforwardly related to GA and PT since PMA = GA + PT. In most (but not all) of the manuscript, the authors use PMA and PT, with GA used without justification in some cases but not in others.

It is unclear why PMA is used at all: PMA is necessarily related to PT and GA, making these variables non-independent. Indeed, the authors show that PMA and PT are highly correlated. The authors even say that "the contribution of postnatal experience to the development was not clarified because PMA reflects both prenatal endogenous effect and postnatal experience." So, why not use GA at birth instead of PMA? Clearly, GA is appropriate in some cases (e.g., Figure S4 or in some of the ANOVA applications), and to me, it seems to isolate the effect the authors care about (i.e., duration of prenatal development). Perhaps there is some theoretical justification for using PMA, but if so, I am unaware.

That said, I expect that replacing all analyses involving PMA with GA will substantially change the results. I do not see this as a bad thing as I think it will make the conclusions stronger. As is, I am left unsure about what the key takeaways of this paper are.

We appreciate the suggestions, and we have replaced the related analyses involving PMA with GA in the manuscript. Please see the Response to the 1st major concern in the Essential Revisions (for the authors) section above for more detail.

Using GA instead of PMA will have several benefits: 1) It will be much simpler to think of these two variables since they contrast the duration of fetal maturation and time postnatally. 2) This will help the partial correlation analyses performed since the variance between the variables is more independent. It will also mean that the negative relationships observed between PT and cortical thickness when controlling for PMA (e.g., Figure 2h) might disappear (reversed signs for partial correlations are common when two covariates are correlated). 3) this will allow the authors to replace Figure 1a with a more informative plot. Namely, they could use a scatter of GA and PT, giving insight into the descriptive statistics of both dimensions.

We have revised the manuscript throughoutly following the reviewer’s suggestion. However, we thought it would be necessary to show the overall development of CT and CM across the general age (PMA) in Figure 1. Therefore, we didn’t replace the figure 1a but added a scatter figure between GA and PT in Figure 2-figure supplement 1 and added descriptive statistics of them in the manuscripts: “The mean GA of the neonates was 39.93 weeks (SD = 1.26) and the mean PT was 1.21 weeks (SD = 1.25), the correlation between them was not significant (r = - 0.08, p > 0.1; Figure 2-figure supplement 1).” Moreover, the negative relationships between PT and CT when controlling for PMA disappeared in the revised results as the reviewer’s predicted.

I suspect that one motivation for the use of PMA over GA is for the analysis in Figure 6. In this analysis, the authors pick a group of term infants with a PMA equal to the preterm infants. Since PMA is the same, the only difference between the groups (according to the authors) is the amount of postnatal experience. However, this is not the only difference between the groups since they also vary in GA (and now PT and GA are negatively correlated almost perfectly). I don't know how to interpret this analysis since both the amount of prenatal maturation and postnatal experience vary between the groups.

We appreciate the reviewer’s opinion that both GA and PT were different between preterm and term-born neonates. Then any of the differences between the two groups might came from the combined effect of GA and PT in our results, and unfortunately, we might not able to separate them in this analysis. However, the preceding results indicated that the CT was significantly influenced by PT and GA while CM was significantly influenced by GA, which So we discuss the preterm and term-born comparison in the context of these findings (Page 19 Line 26-29 and Page 20 Line 1-5): “We found CT in the ventral cortex was generally lower in the term-born than preterm-born infants, while the CM showed the opposite trend in the two groups. Since the preterm babies have longer PT but shorter GA compared to full-term infants at the same PMA, this result supported the above analysis that CT was preferably influenced by PT while CM was largely dependent on GA during the neonatal period”. Furthermore, we added a description in the limitation section to stress the caveat (Page 20 Line17-19): “Meantime, both GA and PT were different between preterm and term-born neonates. Then any of the differences between the two groups might came from the combined effect of GA and PT, and unfortunately, we were not able to separate them in this study.”

Justification of conclusions and statistical considerations

I had concerns about some of the statistical tests and conclusions that the authors made. I refer to some of these in other sections (e.g., the homotopy analyses), but I raise several here.

I am not sure what evidence the authors are using to make this claim: "we found that the cortical myelination and overall functional connectivity of ventral cortex developed significantly with the PMA but was not directly influenced by postnatal time." Postnatal time is significantly correlated with cortical myelination, as shown in Figures 2g, 2h, 3b, 3c, and postnatal time is significantly correlated with functional connectivity, as shown in Figures 4h, 5c, 5d, and 5e. Hence, this general claim that "the development of CT was considerably modulated by the postnatal experience while the CM was heavily influenced by prenatal duration" doesn't seem to be supported: both myelination and thickness are affected by postnatal experience and prenatal duration (as measured by PMA). A similar sentiment is expressed in the abstract. Perhaps the authors suggest different patterns in the strength of change for PMA vs. PT across these metrics, but if so, then statistical tests need to support that conclusion, and the claims need to reflect that sentiment.

Interestingly, Figure S4 presents a compelling ANOVA that does support this conclusion. Still, this result is relegated to the supplement, and it also uses GA, rather than PMA, making it hard to reconcile with the other claims made in the main text. Moreover, it uses ANOVAs, which dichotomizes a continuous variable. Here and elsewhere in the manuscript (e.g., Figures 3d, 3e), the authors split the infants into quartiles and compare them with ANOVAs. Their use for visualization is helpful, but it is unclear what the statistical motivation for this is rather than treating these as continuous variables like is possible with linear mixed-effects models. Moreover, it is unclear why the authors excluded half the data from the study (i.e., quartiles 2 and 3) in this ANOVA when all four quartiles could be used as factors.

We appreciate the reviewer’s comments. We have clarified our results and conclusion in the revised manuscript based on the new analyses that replaced PMA with PT and GA (See the response to the 1st major concern in the Essential Revisions). The previous claims have been changed as following:” the postnatal time could modulate the cortical thickness in ventral visual cortex and the functional circuit between bilateral primary visual cortices. But the cortical myelination, particularly that of the high-order visual cortex, developed without significant influence of postnatal time in such early period” (Page 2, Lines 8-12). This claims could be supported by the results in figure 2. Moreover, to support the claims about the comparison of the influence between GA and PT on structural development, we replaced the ANOVA analysis with a linear mixed-effect model as the reviewer mentioned.

1) To compare the influence of GA against PT on the structural development in the whole ventral visual cortex (Page 7 Line 15-19), “We applied a linear mixed-effect model to test whether the CT (or CM) of the whole ventral cortex were differently influenced by the GA vs. PT, and found that the GA had a significantly stronger effect on the CM than PT (interaction between GA and PT, p < 0.05) but no significant difference was found of the effect on the CT between the ages (p > 0.6).”

2) To compare the influence of GA against PT on the structural development in the area V1 and VOTC, we applied a similar linear mixed-effect model analysis for the two ROIs (Page 8 Line 17-18 and Page 9 Line 1-4): “Moreover, we applied a linear mixed-effect model to test the developmental influence of GA vs. PT on the cortical structure , and the results showed that the CT in two ROIs showed non-significantly different influences from GA against PT (p > 0.3), but CM showed at least marginally significant results in both two ROIs (V1: p < 0.01 and VOTC: p < 0.09).”

It is unclear what the evidence is to support the following claim: "Both CT and CM show higher correlation with PMA in the posterior than anterior region, and higher correlation in the medial than lateral part within the anatomical mask (Figure 2a and Figure S2b-c [sic])" From Figure 2 or Figure S2, I don't see a gradient. From Figure S3, there might be a trend in some plots, but it is hard to interpret since it is non-monotonic. More generally, is there a statistical test to support this claim?

We added a correlation analysis between the diction (x: lateral to medial; y: posterior to anterior) and measurements (CT and CM) in the ventral visual cortex, and the resulting coefficient was all significant (r = 0.7/-0.8 for CT along x/y axis, and r = 0.91/-0.83 for CM along x/y axis; p < 0.001). See Figure 1-figure supplement 2. However, the consideration provided by the reviewer still exists that such significance was driven by part of the areas and the gradient was non-monotonic. Therefore, we replaced the original claim with the following sentence (Page 6 Line 3-8): “In addition, we found distinct spatial variation along ventral cortex, e.g. posterior-anterior and medial-lateral directions (Figure 1-figure supplement 2a-b). Generally, both CT and CM showed higher correlation with PMA in the posterior than anterior region (r = -0.8 and -0.83; p < 0.001), and higher correlation in the medial than lateral part within the ventral visual cortex (r = 0.7 and 0.91; p < 0.001; Figure 1-figure supplement 2c-d).”.

"and the interaction [sic] was more prominent in CM (simple effect: t = 10.98, p < 10-9) that in than CT (t = 2.07, p < 0.05)." Does 'more prominent' mean it is 'significantly stronger'? If not, then the authors should adjust this claim

The claim ‘more prominent’ did express ‘significantly stronger’ since we found that the interaction between CM and CT along PMA or PT was significant in the ANOVA analysis. This analysis has been removed because we thought that the comparison between two structural measurements is not very relevant to the conclusion of the paper. We now applied a linear mixed-effect model to compare the influence of GA against PT on specific structural development. So this result and claim have been removed from the new manuscript.

Are the authors Fisher Z transforming their correlations? In numerous places, correlation values seem to be added together or used as the input to other correlation analyses. It is unclear from the methods whether the authors are transforming their correlation values to make that use appropriate.

We are sorry for the confusion. All the statistical analyses involving correlation coefficients were Fisher-Z transformed. We have added a clear description in the manuscripts involving the Fisher-Z transformation (Page 25 Line 16-18).

Homotopy analyses

The homotopy section is a strength of the paper, but I have doubts about the approach taken to analyze this data and some of the conclusions drawn. I don't expect any of my suggestions to change the takeaway of this section, but I do think they are essential criticisms to address.

I do not think that the non-homotopic control condition is appropriate. In Arcaro & Livingstone (2017), the authors had 3 categories for this analysis: homotopic pairs (e.g., left V1 vs. right V1), adjacent pairs (e.g., left V1 vs. right V2), and distal pairs (e.g., left V1 vs. right PHA1). In the homotopy analysis performed by Li and colleagues, they compare homotopic pairs with all other pairs. I don't think that is generous to the test since non-homotopic pairs include adjacent pairs that should be similar and distal pairs that shouldn't be similar. This may explain why some non-homotopic distribution overlaps with the homotopic distribution in Figure 4c.

Thanks for these suggestions. In the revised manuscript, we reanalyzed the data by dividing the connections into three groups for each subject. See Page 26 Line 24-29: “For each subject, Pearson correlations were carried out on the ROI-averaged time series within and across the left and right ventral cortex. The resulting connections were divided into three groups, namely the homotopic connection (the connection between two paired areas in two hemispheres. e.g. right and left V1), adjacent connection (e.g., right V1 and left V2 since V1 and V2 are adjacent) and distant connections (two areas that were not the paired or adjacent)”.

Regardless of this decision, I think the authors should reconsider their statistical test. I think the authors are using a between samples t-test to compare the 34 homotopic pairs with the hundreds of non-homotopic pairs. This is statistically inappropriate since the items are not independent (i.e., left V1 vs. right V1 is not independent of left V1 vs. right V2, which is also not independent of left V3 vs. right V2). This means the actual degrees of freedom are much lower than what is used. Moreover, I am unsure how the authors do this analysis across participants since this test can be done within participants. The authors should clarify what they did for this analysis and justify its appropriateness.

Thank you for the suggestion. In the previous manuscript, we first averaged the connection matrix across subjects and then calculated the homotopic (or non-homotopic) connections between areas, and therefore, statistical analysis could not be performed. In the revised paper, we calculated the three groups of connections for each subject before the average. We applied a non-parameter statistical analysis (Wilcoxon signed-rank) to address the issue of the independent comparison among the connections, and found the homotopic connections were significantly stronger than the adjacent or distant connections.

See (Page 26 Line 29 and Page 27 Line 1-3): “Independent-sample T-test was used to test whether the homotopic correlation was significantly greater than zero across subjects. To compare the correlation among the three types of connections, we applied a non-parameter statistical analysis (Wilcoxon signed-rank) across subjects”.

The results showed that (Page 9 Line 17-21) “the homotopic connections in all ROIs of ventral cortex were significant (mean r = 0.13– 0.43, t > 12.87, s < 10-9; Fig 4a-b), and were significantly higher than adjacent connections (0.29 ± 0.12 vs. 0.19 ± 0.10, Wilcoxon signed rank test on the Fisher-Z transformed r value: z = 16.32, p < 10-9) and distal connections (0.04 ± 0.06, z = 16.32, p < 10-9; Fig. 4c)”.

Could the authors speculate on why the correlations in homotopic regions are so much lower than what Arcaro and Livingstone (2017) found. I can think of a few possibilities: higher motion in infants, less rfMRI data per participant, different sleep/wake states, and different parcellation strategies. Regarding the last explanation, I think this is a real possibility: the bilateral correlation may be reduced if the Glasser atlas combines functionally heterogeneous patches of the cortex. Hence, the authors should consider this and other possible explanations.

Thank you for the suggestion. The neonates included in this study were all under natural sleep during the scan, so sleep/wake states would not be one of the causes. We added some possible reasons for this difference following the related results (Page 19 Line 9-13): “However, the present homotopic connections in the human neonates were lower than those in neonate macaca mulattas (Arcaro and Livingstone, 2017). This difference might relate to the higher motion in human infants, less r-fMRI data in the present study, coarser parcellation in the visual cortex used in this work, and the developmental difference between primates and humans in the neonatal period.”

The authors assume that the homotopic analyses mean that there are lateral connections between hemispheres (e.g., "Furthermore, the connections among the ventral visual cortex have developed during this early stage. Specifically, the homotopic connections between bilateral V1 and between bilateral VOTC both increased with GA, indicating an increased degree of functional distinction"). While this might be true, it doesn't need to be. Functional connectivity can be observed between regions that lack anatomical connectivity. Instead, two regions could both be driven by another region. In this case, the thalamus might drive symmetrical activity in the visual cortex.

We agree with the reviewer’s view that the development of functional connectivity might be driven by other regions like thalamus. So we added this interpretation in the discussion section (Page 19 Line 23-25): “It is worth noting that the increased homotopic connection can be direct or indirect, e.g., the effect might be driven external regions with enhanced connection to both of the areas (e.g. thalamus)”.

Miscellaneous

I am not sure what the motivation of this line is: "Moreover, those studies did not fully control the visual experience in the first few weeks of the subjects, thus cannot give a clear conclusion whether the innate functional connectivity is unrelated to postnatal visual experience." Arcaro, Schade, Vincent, Ponce, & Livingstone (2017) did control the visual experience of subjects. Moreover, the research here doesn't control infant experience in the way this sentence implies: it implies an experiment manipulation (i.e., fully control) rather than a statistical control that is done here. Consider rephrasing

We have rephrased this sentence in the introduction section (Page 5 Line 2-5): “Moreover, the human infants participating in a previous study (Kamps et al., 2020) were around one month old (mean age: 27 d; range from 6 to 57 d), who might already acquire some visual experience, and thus this study could not exclude postnatal visual experience on the innate functional connectivity”.

I am not sure why this claim is made: "Area V1 was selected because this region is the most basic region for visual processing and probably is the most experience-dependent area during early development". Is there evidence supporting this claim? Plasticity is found throughout the visual cortex, and I think which region is most plastic depends on the definition of plasticity. For instance, most people have the same tuning properties to gabor gratings (e.g., a cardinality bias), but there is enormous variability in face tuning across cultures.

We have removed this claim in the manuscript.

The abstract says 783 infants were included in this study, but far fewer are actually used. The authors should report the 407 number in the abstract if any number at all.

We have revised the number accordingly.

Any comparisons of preterms and terms ought to be given the caveat that the preterm environment can be very different than the term environment: whereas a term infant goes home and sees friends and family without restriction, the preterm environment can be heavily regulated if they are in a NICU. Authors should either provide details about the environments of the preterms in their study, or they should consider how differences in the richness of visual experience - regardless of quantity - may affect visual development.

We agree with the reviewer’s concern, and added a paragraph in the limitation section to stress the caveat (Page 20 Line 12-16): “One limitation of this study is the comparison between preterm and term-born infants did not consider the different visual experience in these infants. The preterm-born neonates may experience very different environment than those of the term-born, e.g. the preterm environment can be heavily regulated if they were in a NICU, but we didn’t have detailed information about the postnatal environment to control for it.”

Reviewer #3 (Public Review):

The authors use a large neonatal dataset to examine how development may occur differently based on whether on not the neonate spent that time in gestation or out of the womb accruing potentially accruing visual experience. In this manner, the authors hope to tease apart those aspects of development that are biologically programmed versus those that occur in response to experience within the visual cortex. They show structurally that cortical thickness is affected by postnatal experience while cortical myelination is not, and functionally they find regional differentiation present between visual areas at birth and that their connectivity changes with development and postnatal experience. The conclusions seem well supported by the data and analyses and provide some insight into which aspects of brain structure at birth are sculpted more by postnatal experience and which are more determined by endogenous developmental timelines.

The analyses are based on a large sample of infants, and the authors were careful to statistically separate which aspects of an infant's age, gestational or postnatal, are driving brain development, providing a deeper picture of infant brain development than previous publications. Overall, the findings seem well supported by the data as the analyses are relatively straightforward.

Visualization of the data and findings could be improved, as a few figures are difficult to interpret without having to read the methods.

We have extensively revised the figures in the manuscript to improve the readability. See updated Figures 2-7.

The acronyms regarding gestation, postnatal, and post-menstrual time are a little distracting. Please consider explicitly writing "gestational time" etc when referring to these numbers to improve readability.

We have replaced the analyses involving PMA with gestational age (GA) or postnatal time (PT) in the revised manuscript to simplify the terminology. Please see the Response to the 1st major concern in the Essential Revisions (for the authors) section above. We believe this change makes the paper easier to follow even with the abbreviations.

Because the cortical ribbon of infants is so thin at birth, there seems to be a possibility that partial-volume effects could be more prevalent in less-developed infants and impact myelin metrics. If not modeled or estimated, it should at least be discussed.

In fact, the cortical thickness of the neonatal brain is not thinner than that of the adult. Particularly, the average cortical thickness of infants aged 0-5 months is around 2-2.5 mm (Wang et al., 2019), which is similar to adults (Fjell et al., 2015). Therefore, the partial-volume effect for cortical gray matter is not a special concern for infants.

Nevertheless, we agree that the partial-volume effects might have different influences on infants of different ages. We added this consideration in the limitation section (Page 20 Line 20-24). “Another concern was about the partial-volume effect on the cortical measurements. The changing thickness of cortical ribbon during development may changes the degree of partial-volume effect, and thus may affect the cortical myelination measurement and may contribute to the myelination difference observed between preterm and term-born groups.”

Structural and functional development could be more formally compared using quantitative models if the authors want those points more strongly related; the two are only qualitatively discussed at present.

We have added a formal analysis to investigate the relationship between structural and functional development. Please see the Response to the 1st concern of Reviewer 1 (public review).

Author Response

Reviewer #1 (Public Review):

The previous study as the authors stated showed a weaker expression of DMP1 in skeletal muscle. The authors provide a clear justification that sarcopenia-like phenotype was unlikely caused by DMP1-cre expression in muscle cells given there is no change of muscle cell numbers. It would be helpful to provide some quantification data of muscle cells to further preclude this possibility.

To define how osteocyte partial ablation was achieved, we performed the quantification of empty lacunae ratio of DTAhet mice at 13 weeks. About 80% empty lacunae was observed in DTAhet mice at 13 weeks which increased about 20% compared to 4 weeks (Line 127-131, Figure 1 – figure supplement 1B), indicating diphtheria toxin (DT) has an accumulative effect with age in DTAhet mice. We speculated that when DT accumulated to a threshold, osteocytes were ablated.

The underlying molecular mechanism is not shown in the current study, but it might be worthwhile to provide some more-depth discussions and hypotheses concerning how osteocytes could influence cell lineage commitment in bone marrow.

We thank the reviewer’s suggestion, and we now have updated this in the Discussion in the revised version (Lines 424-433).

Reviewer #3 (Public Review):

The finding that osteocyte reduction induced senescence in osteoprogenitors and myeloid lineage cells is intriguing. However, further validation of cellular senescence in bone/bone marrow is lacking. Additional approaches, such as immunostaining of key senescence markers in bone tissue sections, are needed to validate the phenotype.

According to the reviewer’s suggestion, we performed the senescence associated 𝛽galactosidase (SA-𝛽Gal) staining of frozen sections of WT and DTAhet mice femur (Figure 6 - figure supplement 1D). Accordingly, the details were given in Response to Essential Revision 2.

It is interesting that partial osteocyte ablation alters mesenchymal lineage commitment, i.e. increased adipogenesis and impaired osteogenesis. The authors should perform further analysis of their scRNA-Seq data and conduct trajectory analysis to confirm the phenomenon. Additional functional assays of bone marrow mesenchymal stem/progenitor cells, such as CFU-F and tri-lineage differentiation assays, are needed to claim the lineage commitment change of the cells.

As we used total bone marrow cells to perform scRNA-seq, the number of MSCs was not enough to perform further re-clustering and trajectory analysis. We performed GO enrichment analysis of MSC cluster which revealed that downregulated genes after osteocyte ablation were enriched in ossification and biomineral tissue development (Figure 6 - figure supplement 1E), which was consistent with the finding of impaired osteoblast differentiation (Figure 4H-J). Further, as reviewer suggested, we performed qPCR to verify related gene changes during tri-lineage differentiation. We found that the mRNA level of osteogenic markers including Alp, Ocn, Runx2 was decreased (Figure 4J), indicating the impaired osteogenesis after osteocyte ablation. Meanwhile, the mRNA level of adipogenic markers including Adipoq, Fabp4, Ppap𝛾 and Cebpa was significantly increased (Figure 6 - figure supplement 1F), indicating the promoted adipogenesis and altered MSC commitment. Besides, the mRNA level of cartilage anabolism related genes (Col1a2, Acan, Sox9 and Prg4) and catabolism related genes (Mmp3, Mmp13, Adamts1 and Adamts5) was not significantly changed (Figure 6 - figure supplement 1G), indicating that chondrogenesis was not altered after osteocyte ablation. And we now have updated this in the revised version (Lines 324-333) and trilineage differentiation methods and information of primers have been updated in Material and methods (Lines 579-590, Lines 623-637).

The mechanism why osteocyte reduction causes cellular senescence of the surrounding cells is an interesting question. It would be helpful if the authors provide evidence or give an explanation on this point. Does the phenotype recapitulate age-associated bone impairment? The laboratories of Sundeep Khosla (Mayo Clinic) and Maria Almeida (University of Arkansas for Medical Sciences) reported that osteocytes are a major cell type in bone that become senescent during aging. Although most of osteocytes were eliminated in the mouse model used in this study, were the rest osteocytes undergoing cellular senescence?

We thank the reviewer’s suggestion, and we now have updated this in the Discussion in the revised version (Line 424-433). The details were given in Response to Essential Revision 4.

We thought that the phenotypes after osteocyte ablation were similar with the ageassociated bone impairment, and to certain degree this phenotype recapitulated the ageassociated bone impairment, which further indicated the important role of osteocytes in maintaining the bone homeostasis during aging. We performed the SA-𝛽Gal staining of frozen sections of WT and DTAhet mice femur, in which we observed SA-𝛽Gal+ cell in the cortical bone region of DTAhet mice (Figure 6 - figure supplement 1D). As cortical bone mainly contains osteocyte and matrix, we inferred that the rest osteocytes may also underwent cellular senescence.

Author Response

Reviewer #3 (Public Review):

In invertebrates, learning-dependent plasticity was reported to take place predominantly in presynaptic neurons. In Drosophila appetitive olfactory learning, cholinergic synapses between presynaptic Kenyon cells and postsynaptic MBONs undergo behaviourally relevant associative plasticity, and it was shown to reside largely in Kenyon cell output sites. This study provided several lines of evidence for postsynaptic plasticity in MBONs. The authors nicely showed the requirement of Kenyon cell output during training, strongly suggesting that behaviourally relevant associative plasticity also resides downstream of Kenyon cell output. This is further supported by impaired appetitive memory by downregulating nAChR subunits (a2, a5) and scaffold protein Dlg in specific MBONs. Live imaging experiments demonstrated that the learning-dependent depression in M4-MBON was reduced upon knocking down the a2 nAChR subunit. Using in-vivo FRAP experiments, the authors showed recovery rates of nAChR-a2::GFP were altered by the co-application of olfactory stimulation and DA. All these lines of evidence point to the significance of nAChR subunits in MBONs for postsynaptic plasticity.

On the technical side, this study achieved a very high standard, such as the measurement of lowexpressed receptor mobility by in-vivo FRAP. The authors conducted a wide array of experiments for collecting data supporting postsynaptic mechanisms. The downside of this multitude is somewhat compromised coherence. To give an example, the authors duplicated many behaviour and imaging experiments in different MBONs for non-associative learning (Fig. 7 and 8), which is primarily out of the scope of this paper (cf. title).

We thank the reviewer for their positive assessment and constructive criticism. We have thought a lot about removing data on non-associative learning (Fig. 7 and 8.), however feel that they do add important experiments that are not feasible to address for the other MBONs due to technical constraints (complexity of training protocols and localization of imaging area). We also decided, as reviewer 1 was happy with these experiments, that it is important to show that the receptor plasticity is not confined to associative appetitive memory but also is important for other postsynaptic memory storage mechanisms. As a response to this reviewer, we have adjusted the title to:

Postsynaptic plasticity of cholinergic synapses underlies the induction and expression of appetitive and familiarity memories in Drosophila

We also now include familiarity learning in the abstract. Moreover, we now expanded our explanation on to why we conduct these additional experiments and now state:

line 436ff: ‘Our data so far suggest that regulation of α2 subunits downstream of α5 are involved in postsynaptic plasticity mechanisms underlying appetitive, but not aversive memory storage. Besides associative memories, non-associative memories, such as familiarity learning, a form of habituation, are also stored at the level of Drosophila MBs. We next asked whether postsynaptic plasticity expressed through α5 and α2 subunit interplay, was exclusive to appetitive memory storage, or would represent a more generalizable mechanism that could underlie other forms of learning represented in the MBs. We turned to the α’3 compartment at the tip of the vertical MB lobe that has previously been shown to mediate odor familiarity learning. This form of learning allows the animal to adapt its behavioral responses to new odors and permits for assaying direct odor-related plasticity at the level of a higher order integration center. Importantly, this compartment follows different plasticity rules, because the odor serves as both the conditioned (activating KCs) and unconditioned stimulus (activating corresponding dopaminergic neurons)15. While allowing us to test whether the so far uncovered principles could also be relevant in a different context, it also provides a less complex test bed to further investigate whether α5 functions upstream of α2 dynamics.’

We also would like to emphasize that - if the reviewer feels that keeping these data / this information as part of our manuscript would prevent publication - we are prepared to remove these data from the manuscript, and submit these data in their own right (potentially as a research advance subsequently).

Author Response

Reviewer #1 (Public Review):

1. Probably the shortest review I've ever written! Most birds today can lift the upper beak independently of the brain case. This is made possible by a series of mobile joints and bending zones in the skull. To investigate the evolution of this phenomenon, the authors successfully CT-scanned the thoroughly squished skull of the Early Cretaceous stem-bird Yuanchuavis. The detailed description and illustration of the shapes and positions of the skull bones leave no doubt about the conclusion that the toothed snout was unable to move independently of the brain case. They also show, however, that the loss of a few extensions from specific skull bones would have made mobility possible. This plugs a major gap in our understanding of the evolution of mobility within the skull in birds (and by extension elsewhere, notably in the similarly diverse lizards & snakes).

Yes, we are delighted that this work will further advance our understandings about the avian skull evolution.

Reviewer #2 (Public Review):

1. Wang et al. present a detailed description and analysis of the previously reported cranial remains of enantiornithine bird Yuanchuavis. The authors use X-ray CT scan data to reconstruct the cranial elements and retro-deform the facial and palatal skeleton. The authors also use principle component analysis with geometric morphometrics data to investigate where Yuanchuavis falls in palatine phylomorphospace. The authors use these data to make inferences about the kinetics of the Yuanchuavis skull as well as the evolution of cranial kinesis across birds. Generally, I find the authors' direct interpretation of their anatomical and PCA data to be convincing and compelling. The anatomical description is thorough and accurate. The methods used for the geometrics morphometrics and PC analyses are appropriate. I find compelling the authors' interpretations that Yuanchuavis largely retained the ancestral non-avialan akinetic skull.

One of the greatest strengths of this paper are the extremely attractive figures. In particular, I find figure 4 to be exceptionally useful - this is easily the most effective illustration I have yet seen of avian cranial kinesis and the shifts in cranial morphology that underlie its evolution. I applaud whoever designed this figure. My one major concern with this paper's methodology is that the palatine used for Ichthyornis is incorrect. Torres et al. (2021) published the correct palatines, which were very different from those incorrectly (but understandably) identified in Field et al. (2018) and used here. I strongly urge the authors to rerun their GMM analysis with corrected data

We thank the reviewer for supporting this study. As for the palatine of Ichthyornis, we have used the palatine reconstruction in Torres et al. (2021) and reperformed the GMM analyses. This certainly changes the GMM result, and the main conclusion has not been strongly influenced. We are grateful for this comment.

Author Response

Reviewer 2 (Public Review):

1) The hypothesis that the genes responsible for the Mendelian traits are also the causal genes for the cognate complex traits does not seem to hold, given the prior work and the data shown in the study. For example, if this hypothesis is true, it is unexplained why the candidate genes were not even enriched in the GWAS regions for height and breast cancer.

Following the removal of a data artifact from our breast cancer analysis and the inclusion of Backman et al.’s larger list of genes implicated in height, every phenotype in our analysis displays enrichment in proximity to GWAS peaks. Enrichment is present not only in genes selected based on cognate Mendelian phenotypes, but also on those from Backman et al., which examined the same complex trait phenotypes that were used for GWAS. In that work, the enrichment GWAS signal near of genes selected on coding variants was as high as 59.3-fold.

Our use of Mendelian-trait-causing genes is not dependent on GWAS. Short of large-scale experimental work, we do not know any better way to confirm the genes’ broad relevance to GWAS phenotypes than their enrichment near peaks. This enrichment has been persuasively demonstrated by previous research. Freund et al. (2019) tested the enrichment of 20 Mendelian disorder gene sets against 62 complex phenotypes. Though there was no statistically significant overlap of phenotypically non-matched Mendelian genes and GWAS peaks (2% matched), the overlap of matched Mendelian genes and GWAS peaks was significant (54% matched).

We have included additional evidence and references for this relationship in Supp. Note 1.

2) The only evidence supporting their hypothesis appears to be the enrichment of the candidate genes in the GWAS regions for seven out of the nine traits. However, significant enrichment of the candidate genes in the GWAS regions does not necessarily mean that a large proportion of the candidate genes are the causal genes responsible for the GWAS signals. Analogously, we cannot use the strong enrichment of eQTLs in GWAS regions as evidence to claim that a large proportion of the GWAS signals are driven by eQTLs.

Our gene sets were selected by considering two criteria: whether they are relevant to each complex trait, and whether they are biologically interpretable.

The genes identified in Backman et al. have a strong case for relevance. They are evaluated for association, not with cognate Mendelian phenotypes, but with the exact same complex traits used for GWAS.

Our genes, selected based on cognate Mendelian traits, are less obviously relevant, but have advantages for interpretation. Many have well-understood biological roles and are part of pathways that have been studied in great detail. Because most of these genes can cause dramatic phenotypic changes with one variant, the direction of effect is easier to understand than genes identified through burden testing. In fact, loss-of-function coding variants that cause autosomal dominant traits can be thought of as large-effect, context-independent eQTLs—they cause phenotypic change by decreasing gene expression roughly 50% across cell types, developmental stages, etc.

Ideal genes for our analysis would combine the advantages of both sets. They would have individual coding variants that could be tied to complex traits using exome sequences. However, natural selection creates tradeoffs between variant frequencies and variant effect sizes. Large-effect variants (such as those responsible for Mendelian traits) are generally too rare to be detected in population sequencing. Coding variants that reach frequencies detectable in databases such as UK Biobank typically have smaller effect sizes, requiring them to be aggregated in order to implicate genes.

We believe that our original gene set is plausible both because of its collective enrichment in GWAS signal and because each gene is individually known to cause cognate phenotypes. Enrichment is not proof, but can serve as strong evidence when backed up by known biology. Though selection precludes a perfect gene set, the enrichment in both our Mendelian gene set and the set from Backman et al. addresses each criterion—interpretability and relevance—individually, and, taken together, provides an argument for the relevance of genes selected based on coding variants.

3) Considering the large numbers of GWAS signals, we would expect a substantial number of genes in the GWAS regions by chance. It would be interesting to quantify the number of genes in the GWAS regions if the 143 genes are randomly selected. Correcting the observed number of genes for that expected by chance (e.g., subtracting the observed number by that expected by chance), the proportion of the candidate genes in the GWAS regions would be small.

The proportion of the candidate genes whose eQTL signals were colocalized with the GWAS signals or in close physical proximity with the fine-mapped GWAS hits was small. However, I would not be surprised if they are significantly enriched, compared with that expected by chance (e.g., quantified by repeated sampling of the 143 genes at random).

Taking random sets of genes, or the entire set of non-putatively-causative genes shows that, given the size of our gene set, we would expect 43 randomly selected genes to fall within 1 Mb of a peak (95% confidence interval: 31.5-54.5). Instead, we find 147 peak-adjacent genes. When looking closer to genes, the enrichment increases. At a distance of 100 kb, we find 104 putatively causative genes, but the null model predicts only 11 (95% CI 4.5-17.0), a roughly ten-fold difference.

Enrichment remains significant even when using a more conservative null. It may be that genes like ours, with importance to phenotype, are more likely than random genes to fall near GWAS peaks, even if their phenotype does not correspond to the GWAS phenotype. In this case, we might see enrichment even in the absence of a relationship between our Mendelian and complex traits. To account for this, we also tested significance by testing genes sets against different phenotypes (e.g. testing our LDL genes with a UC GWAS, and our height genes with a T2D GWAS). The results of this permutation are visible in Supp. Fig. 1, and further confirm the enrichment.

Finally, non-expression based analysis found that Mendelian genes had large enrichments in heritability. As in our study, they included Mendelian genes for diabetes and LDL—the Mendelian diabetes genes were enriched 65-fold for common-variant heritability and the Mendelian LDL genes were enriched 212-fold (Weiner et al. 2022).

Though it is true that the number of colocalizations and TWAS hits likely represents a statistically significant enrichment over all genes, we feel that this does not affect the conclusions of the paper. The model that noncoding variants identified by GWAS act as eQTLs certainly has some truth—colocalization and TWAS studies have found, in total, many associations. But the model’s success has not lived up to its expectations. This has been suggested, albeit inconclusively, by the failure of most GWAS peaks to colocalize. By evaluating, not the portion of loci that can be tied to a gene, but the portion of already-implicated genes that can be tied to a locus, we believe the model’s deficiencies are both more clear and more puzzling.

4) It is unclear how the authors selected the breast cancer genes. If the genes were selected based on tumor somatic mutations, it is a problem because there is no evidence supporting that somatic mutation target genes are also cancer germline risk genes.

Genes for breast cancer were selected using the MutPanning method (Dietlein et al. 2020), which takes somatic mutations found in tumors, and evaluates them in the context of known mutation patterns. The relationship between somatic and germline variants in cancer is little studied. We believe it is meaningful that, as explained in our response to overall comment 2ii, we do now find an enrichment of our breast cancer genes near GWAS peaks. Though these genes are very unlikely to be a perfect set, the conclusions of our paper remain true with or without the inclusion of this phenotype.

5) The authors observed no enrichment of the candidate genes in height and breast cancer GWAS regions. In this case, should these traits and the corresponding genes be removed from the subsequent analyses?

The reviewers’ notes about enrichment—and its absence in height and BC—prompted us to review our analysis of it. The enrichment for five of our phenotypes remained significant, and the lack of enrichment for breast cancer genes proved artifactual. After accounting for the artifact, the enrichment of breast cancer genes displays the same pattern as most other phenotypes, displaying highly significant enrichment as compared to the genomic background and a permutation analysis. Supplementary figure 1 has been updated to reflect this change, and to add the enrichments found in Backman et al.

Because our original analysis of height has nominal, but not corrected, significance for enrichment, the problem may be one of power. The set of height genes identified by Backman et al. is larger than our original set and displays a significant enrichment in proximity to GWAS signal. This enrichment is also present when the two gene sets are combined, as shown in the updated Supp. Figure 1.

Reviewer 3 (Public Review):

1) The positive results are substantially reduced when restricting the analyses to a set of selected tissues of relevance to the trait. Isn't it implicated that the selection of relevant tissues in this study is not comprehensive, and further, tissue specificity is common in mediating genetic effects by gene expression? First, it seems some apparently relevant tissues are not selected (Table 2), such as bone for height (Finucane et al. 2015 NG). One approach to assess the relevant tissues for the predefined set of putatively causative genes is to see if these genes are enriched in the differentially expressed gene sets for those tissues. Second, among 84 putatively causative genes overlapped with GWAS signals, they identified 39 genes by TWAS, 11 genes by fine mapping with linear distance to chromatin modification features, and 41 genes by fine mapping with ChromHMM enhancer annotations, but these numbers reduced substantially to 9, 5 and 27 when restricting the same analysis to the selected tissues for each trait. If genes function only in the relevant tissues, I think using bulk expression data would lose power but is unlikely to give false positives. Thus, it is possible that for the traits analysed, not all relevant tissues are selected so that only a fraction of genes identified in bulk expression analysis can be replicated in the tissue-specific analysis. This appears to me a notable piece of evidence to support the hypothesis of biological context that the authors tend to have reservations in discussion.

Testing for colocalizations or TWAS hits in all tissues may increase power for several reasons. First, it is possible that some GTEx tissues have unrealized relevance to our phenotypes. Secondly, in the event that a tissue is not present in GTEx, we may still detect relevant eQTLs in a tissue that is not itself involved in the trait, but which has similar patterns of expression. Finally, some tissues may be correct, but underpowered due to their small sample size. In this case, we may better detect the colocalization in tissues that are “irrelevant,” but are well-powered and have correlated expression.

However, this creates problems of interpretation. Say we find, for example, a colocalization of an APOE eQTL with an LDL GWAS peak in skin tissue. Does this mean that skin tissue contributes to LDL levels? Is it simply because skin tissue has more samples than liver? Are we uncovering a strange, unexpected pleiotropy?

We believe we can achieve both objectives—power and interpretability—with our use of MASH (Urbut et al. 2019) as described in response 3 of the first section. Briefly, MASH is a Bayesian tool that we use to update the estimates of eQTLs in GTEx data. Each tissue is adjusted to incorporate signals detected in other tissues with similar expression. This mitigates the danger of ignoring the correct tissue, and increases the power of tissues with small sample sizes. Its benefit is demonstrated by the substantial increase in the number of expression-GWAS colocalizations identified by coloc—however, the number of genes identified that fall within our putatively causative gene sets remains strikingly small.

2) How much do both LD differences between GWAS and eQTL samples and the presence of allelic heterogeneity contribute to the observed low colocalization rate? One of their main findings is the low colocalization between trait-associated variants and eQTL in non-coding regions, which accounts for only 7% of the putatively causative genes. In discussion, the authors believe that this finding cannot be explained by lack of statistical power and is directly supported by a Bayesian analysis which reported high posterior probabilities of distinct signals for GWAS and eQTL. I agree that power is probably not a big issue. However, my concern is that given the large difference in sample size between GWAS and GTEx datasets, any small differences in LD between the two samples might cause a statistical separation of the signals even when trait phenotype and gene expression truly share a causal variant. Moreover, the presence of more than one causal variant with allelic heterogeneity in the locus may also play a part in the failure of colocalization. Consider two causal variants for the complex trait, one regulating the target gene and the other regulating another gene in co-expression. Potentially, the presence of the second causal variant would diminish the colocalization probability at the target gene.

The ability of our statistical tools to actually find colocalizations is a critical one in this project. Small sample size increases the variance of the LD matrix, but is one of only many factors that influence power, which include LD differences between study populations and eQTL effect sizes.

Though we restricted both GWAS and GTEx samples to subjects with European ancestry and used PCs as covariates, reviewers are correct that there are likely to be LD differences between samples, due to both slight variations in populations and the smaller sample sizes of GTEx. Analysis of colocalization tools in cases of mismatched LD have shown that decreases in power are small. Chun et al. (2017) tested JLIM in simulated conditions of modest population mismatch, using CEU haplotypes to create the GWAS, and haplotypes from all non-Finnish Europeans for eQTL associations. They then attempted to distinguish shared vs. distinct causative variants for GWAS and eQTL, finding no decrease in sensitivity or specificity (Supp. Fig. 6 of Chun et al. 2017).

The case in which two genes are co-regulated by nearby variants, both causative for the GWAS trait, creates a condition of allelic heterogeneity for the GWAS trait (as opposed to the expression trait). Chun et al. evaluated JLIM’s loss of power as a result of AH, and found that the power loss is small, except in cases in which the two variants have equal effects (Supp. Fig. 10). Testing cases in which the AH occurs for the expression trait returned a similar result (Supp. Fig. 9).

Hukku et al. (2021) performed similar analyses on coloc, eCAVIAR, and fastENLOC. Allelic heterogeneity was found to damage the power of coloc (by about a factor of 2). Testing on different pairs of populations, they conclude that extreme LD mismatches (e.g. Finnish vs. Yoruban samples) can lead to substantial power loss, but moderate LD mismatches (e.g. Finnish vs. British samples) do not. Though a factor of two is substantial, it would not change the qualitative conclusions of this paper. Overall, given the variety of methods we employ (including those, such as JLIM, more robust to AH), we are confident that they have, when taken together, been shown to be robust to the concerns raised.

Finally, TWAS should, by design, be less vulnerable to LD differences and allelic heterogeneity. This can result in false positives, when genes with correlated expression are identified together, despite only one being causative. It can also result in non-causative genes being prioritized over causative ones, however, generally both genes will be identified (Wainberg et al. 2019).

3) Perhaps the authors can perform some simulations to quantify the influence of tissue-specific expression effects, LD differences between eQTL and well-powered GWAS, and allelic heterogeneity, as discussed above, on their analyses. I understand that the authors may not be willing to do as it would involve a lot of work. But I'd like to see at least some discussion on how these questions can be better addressed in the future research.

These are nuanced technical questions, and to address them by simulation in our paper would, as noted, involve a lot of work. We have summarized previous work that evaluated the effects of LD differences and AH in our response to essential revision 4. We discuss our concerns about the possibility of an overly broad tissue search in essential revisions 3 and 5, and our decision to address this question using MASH in essential revision 3.

4) It looks quite striking that only 6% of the putatively causative genes are identified by TWAS with the correct effect direction. But I think this number is slightly misleading as one may interpret it as only 6% of the functionally relevant genes are regulated by trait-associated variants. In fact, 46% of the genes are detected by TWAS but only 11% are confirmed in their selected tissues, among which about half (5/9) have correct effect direction. First, the result could be limited by the selection of relevant tissues, as discussed above. Second, the fact that half of the genes do not show correct effect direction may reflect a nonlinear relationship between expression and trait, or the presence of cell-type heterogeneity within a tissue. These may not necessarily overturn the assumption that these genes are regulated by trait-associated variants in the causal tissues or cell types.

In our initial submission, we had been reluctant to expand the list of tissues for two reasons. First, increasing from the small number of tissues with known biological relevance to all tissues (or all non-brain tissues) increases the multiple-testing correction burden. Second, and, in our eyes, more important, colocalizations in tissues without clear biological relevance are not biologically interprable. Such hits can be results of complicated genetic architecture (e.g. shared eQTLs), power differences in tissues with correlated expression, or biology not directly related to the trait in question.

That said, the tissue data we have access to are incomplete, and we are without question missing some relevant tissues. Additionally, some relevant tissues have lower sample sizes, and thus lower power, than tissues that are not relevant but may still share eQTLs. To overcome these problems, we applied Multivariate Adaptive Shrinkage (MASH), a Bayesian method that detects correlations between different (in this case tissues) and uses them to produce posterior estimates of summary statistics in each tissue (Urbut et al. 2019). Unlike meta-analysis, which produces one result, the effect size estimates for each tissue are distinct, though informed by one another.

Using MASH has a pronounced effect on colocalization results. The number of non-putatively causative genes colocalizing increases from 389 to 489, while the number of putatively causative genes in our Mendelian set is unchanged, remaining at 2. The number of genes from the Backman et al. set increases from 2 to 5. Though this is a proportionally large increase, it still represents a small fraction of genes. We have updated our paper to use these results—which should be less dependent on the tissues we selected—but the message has not changed.

5) While they highlight the roles of alternative regulatory mechanisms, few testable hypotheses are put forward for the field, which is somewhat disappointing but understandable given how little we know about the human genome at the mechanistic level.

We have added a set of models that may explain the “missing heritability” to Table 4 in the discussion. Though we do not propose experiments, we have included citations for research relevant to confirming or disproving these models.

Author Response

Reviewer #1 (Public Review):

In this manuscript, Kowalczyk and colleagues report on identifying coding and non-coding genetic determinants of hairlessness in mammals using an approach they developed called RER-converge. The approach has previously been employed to examine several different traits in previous publications from this group. The authors determine that hairlessness is associated with relaxed evolutionary constraint at genetic loci and identify both coding genes and non-coding sequencing associated with this phenotype. Several known-hair-associated and novel genes and microRNAs are observed.

This is a strong manuscript with interesting results. It is remarkable how robust this method is. There are a few places where I was not fully convinced of the choice to highlight a gene as "significant" however.

In Figure 4 and the associated text and figure legend the claim is made that non-coding regions exhibit accelerated evolution of matrix and dermal papilla elements. However, the enrichment, even prior to multiple testing correction is not significant. Should this be reported on?

We agree that some of the results that we displayed had borderline significance and we have clarified this in the text so the reader is aware. Our rationale for highlighting tissue annotations from borderline-significant enrichment results from noncoding analyses (matrix p=0.078 adj.p=0.18, dermal sheath p=0.059 adj.p=0.16, dermal papilla p=0.049 adj.p=0.16) is because we believe that these are an honest depiction of the trends we see in this scan (with alpha=0.2 for adjusted p-values), particularly when supported by effect sizes as reported through AUC. We prefer to set a generous threshold to avoid missing any meaningful results rather than setting a more stringent threshold. A more generous alpha is also more forgiving of the noise related to identifying noncoding regions and assigning them to genes.

Related to the above, Table 1 includes just one 'significant gene,' with the remainder of the genes highlighted because they have a Bayes Factor ratio >5. Should a gene with a BF HvM be highlighted as a gene "whose evolutionary rates are significantly associated with the hairless phenotype?" Perhaps I am incorrect, but the hypothesis that is being tested by this approach seems distinct from "is the gene associated with hair loss."

Similar to pathway enrichment analyses, we also used generous significance thresholds for gene-specific results to show our top, most significant results from protein-coding analyses. Significance of noncoding enrichment was not a criterion for inclusion/exclusion of genes in Table 1. Generally, some genes with significant convergent evolutionary rate shifts in protein coding sequence also have significant enrichment of convergent rate shifts in nearby noncoding regions (like PTPRM in Table 1), but many do not, which is also shown in Figure 6A. We have clarified column titles in Table 1 by adding (Gene) or (Noncoding) to indicate which sequences the values refer to.

Bayes factors (BF) are a complementary Bayesian approach to analyze statistical associations that we use here to supplement information we get from our more traditional Kruskal-Wallis test. BF are easy to interpret because they directly describe the amount of support for our alternative hypothesis rather than indirectly describing support as p-values do. For example, in Table 1, the hypothesis that the evolution of FGF11 is associated with the evolution of mammalian hairlessness has 6,354.7 times more support than the null hypothesis that phenotype and gene evolution are not related. These large values are interpreted as supporting the alternative, which is equivalent to what we want to be able to interpret from p-values (i.e. low p-values allow us to reject the null and implicitly support the alternative).

BF values in Table 1 are calculated using evolutionary rates in protein coding sequence and so are not expected to match values in the “Noncoding” columns. “BF Hairless” is directly related to the “Statistic” and “p-adj” columns, which is why the “BF Hairless” values are all quite large, indicating a large amount of support for an association between gene and phenotype evolution.

The hypotheses that are tested with the “Statistic” and “p-adj” columns and the “BF HvM” column are colloquially the same: they both test to determine if the evolutionary rate of the gene is different in hairy mammals compared to hairless mammals. Only the details are different. The traditional statistics test for an association without accounting for marine mammals as a potential confounder. The BF tests check for a significant association that is driven more strongly by hairlessness than by marine habitat.

Slightly more description of the Bayes factor calculation would be beneficial to the supplement. e.g. is the R package BayesFactor package being used here... or something else?

We agree that a clearer description of Bayes factors is appropriate and have modified the methods description as follows:

“In addition to calculating element-specific association statistics, Bayes factors were calculated for each gene using the marine and hairless phenotypes using the BayesFactor R package (Morey & Rouder, 2021). These values were calculated to disentangle the two phenotypes, which are heavily confounded since nearly all marine mammals in the genome alignment used for this work are hairless. Briefly, Bayes factors are a Bayesian approach complementary to more standard statistical tests. Instead of returning statistics and p-values, Bayes factors directly quantify the amount of support for an alternative hypothesis. For example, a Bayes factor value of 5 for a particular statistical test would indicate 5 times more support for the alternative hypothesis than the null hypothesis. Bayes factors can also be used to compare different alternative hypotheses by calculating the ratio of two Bayes factors. When considering the hairless phenotype, we use Bayes factors to quantify the support for a linear model predicting phenotype using evolutionary rate information from each gene, with a higher Bayes factor indicating greater support. We perform this calculation for two alternative hypotheses: 1) a gene shows different evolutionary rates in hairless versus hairy species, and 2) a gene shows different evolutionary rates in marine species versus non-marine species. The ratio of Bayes factors between the hairless and marine phenotypes quantifies the level of support of one phenotype over the other and thus can be used to tease apart intricacies of the two heavily-confounded phenotype. When the Bayes factor for the hairless phenotype is much larger than the Bayes factor for the marine phenotype, that indicates stronger support for signal driven by hairlessness.”

Why are the qq-plot distributions of non-coding elements so distinct compared to coding? Some comment on this would be appreciated in the main text, even if briefly.

We have added the following text as a tentative speculation about why noncoding elements seem to show more signal than coding signal:

“Interestingly, noncoding regions appeared to show even stronger deviation from uniformity than coding regions, perhaps because regulatory changes more strongly underlie the convergent evolution of hairlessness.”

Reviewer #3 (Public Review):

The authors present a phylogenetic analysis of evolutionary rates as they correlate with independently derived "hairlessness" across mammals. This is a very good paper, well written and very carefully analyzed. This paper makes a number of interesting biological insights, including the identification of protein coding as well as noncoding regions that appear to evolve in correlated fashion with hairlessness.

I have several recommendations:

1) The main assumption behind this experiment is that species "use" the same genes to accomplish hairlessness. Only then would one predict correlated rate shifts along hairless lineages. If, on the other hand, each hairless species used a unique gene to accomplish hairlessness, then one might only see a rate shift on that species' lineage. Therefore, a complementary approach might be to i) define all genes with known involvement in hair morphology (i.e., genes in the categories listed in Fig. 1C). ii) test how many of those genes show a significant rate shift in at least one hairless lineage. iii) test whether hair genes are more likely to show at least one rate shift compared to genomic background. This complementary analysis would relax the assumption that all hairless species show similar rate shifts compared to haired species.

Our analyses detect convergently evolving genomic elements associated with hairlessness for two reasons. First, species-specific analyses may detect genomic changes associated with any unique phenotypes in a particular species and it is difficult to distinguish which of those genomic changes are associated with hairlessness. Second, we are seeking genomic elements associated with hair growth in all mammals and species-specific adaptations will not be shared across all mammals.

Nevertheless, we conducted a complementary analysis to test for rate shifts specific to each hairless species compared to all of the non-hairless species. We then tested for enrichment of hair follicle genes among genes with significant rate shifts in different numbers of hairless species. For example, among all genes with significant rate shifts in at least one hairless species, is there an enrichment of hair follicle genes? Then, among all genes with significant rate shifts in at least two hairless species, is there an enrichment of hair follicle genes? Et cetera until we test for enrichment only in genes with rates shifts in all ten hairless species. As expected, the signal of enrichment gets stronger as more species share the rate shift (the “convergent signal”). This happens because the genes with shared rate shifts are more hair-specific than the genes with unshared rate shifts.

We also performed another analysis to test for enrichment of hair follicle genes among genes with significant rate shifts per hairless species. For example, in orca, are the genes with significant rate shifts enriched for hair follicle genes? To complement this analysis, we also repeated the procedure for non-hairless species for comparison. Only two of the ten hairless species show species-specific hair follicle enrichments, which indicates that most of the hairless species alone are insufficient to detect hair signal at all. Even among the two species with significant enrichment, there are thousands of total genes identified, many of which are likely related to other unique characteristics of those species other than hairlessness, and it is impossible to distinguish the hair-related genes from the other genes without additional information.

All of these results are reported in the manuscript in the text and figures shown below:

Species-Specific Analyses

In addition to conducting convergent evolution analyses to identify genetic elements evolving at different rates across all hairless species, we also conducted complementary analyses to detect elements evolving at different rates in individual hairless species to demonstrate the importance of convergent evolution in our analyses. Indeed, the strength of enrichment for hair follicle-related genes among top hits steadily increases as more hairless species share rate shifts in those genes, an indicator of the power of the convergent signal (Figure 2). Further, analyses on single species alone only show enrichment for hair follicle-related genes among top hits in two hairless species out of ten – armadillo and pig (Figure 2 Supplement 1). Together, these results demonstrate the importance of testing for convergent evolutionary rate shifts across all hairless mammals to best detect hair-related elements.

Also of important note is that every individual hairless species has thousands of genes with significant rate shifts in that species (Supp. File 10). It is impossible to tell which of those rate shifts is associated with hairlessness specifically because the species have many unique phenotypes other than hairlessness that could be responsible for rate shifts in their respective genes. Convergent analyses allow for more concrete identification of hair-related elements by weeding out rate shifts that are not shared across species with the convergent hairless phenotype.

2) It would be interesting to break up noncoding into additional strata. For example, one might predict that rate shifts in predicted transcription factor binding sites would have a larger functional impact than rate shifts in noncoding regions with no function. Or... that rate shifts in highly conserved noncoding regions vs. less conserved noncoding regions.

We have performed extensive analyses to investigate the roles of TFBSs in the convergent evolution of hairlessness and found little enrichment of specific TFBS in our top noncoding regions from RERconverge. Perhaps because the noncoding regions are highly conserved, they contain many potential locations for TF binding and so it may be more reasonable to consider their full stretch of sequence as functional than it would be if they were less conserved.

We have calculated conservation scores for noncoding regions and found no global association between RERconverge results and sequence conservation score.

3) Why is aardvark considered a haired species? Aardvarks have as much (or as little) hair as pigs.

Body hair is a difficult phenotype to categorize in mammals because all mammals do have hair. In order to create a binary distinction between hairy and hairless mammals, we needed to make a choice about where to draw that line. We were particularly concerned about the impact of assigning some of the hairier mammals, like pig, armadillo, and human, as hairless, so we performed the drop-out tests shown in Figure 4 to demonstrate that removing individual hairless species from our analyses does not change the overall signal. Indeed, removing pig impacts detection of genes in the two hair-related pathways shown less than removing clearly hairless species like killer whale or dolphin. We believe that these results are sufficient to demonstrate that subtle differences in phenotyping decisions will not substantially change the findings stated in our manuscript.

4) The primary goal of the paper is to identify coding/noncoding regions that show shifts in evolutionary that are correlated on hairless vs. haired lineages. I was left wondering... when these correlations are found, how often is it due to the same mutations hitting the regions vs. mutations randomly hitting the same regions. If the former, this would suggest some limited way that species can achieve "hairlessness".

In general, we do not expect amino acid convergence (for genes) or nucleotide convergence (for noncoding regions) to drive much of the signal we detect using RERconverge. For species separated by millions of years of evolutionary time, it is highly unlikely that a change in a single amino acid (or nucleotide) would drive exactly the same phenotypic change for a highly complex phenotype like hairlessness. However, we argue that there do appear to be some limited ways that species become hairless, albeit at the scale of evolutionary rates across a length of sequence rather than individual bases.

Related to this point is the distinction between positively selected regions compared to regions under reduced constraint, which we would expect to accumulate mutations randomly.

For genes, we believe that accelerated evolution of specific genomic regions in hairless species is caused by an accumulation of random mutations, not positive selection or specific targeted mutations. As stated in the manuscript, we performed branch-site tests for positive selection on our top genes, all KRTs, and all KRTAPs, and we found little indication that quickly evolving genes are undergoing positive selection specific to hairless species. This conclusion is also consistent with the hypothesis that genes under relaxation of evolutionary constraint will have rate shifts that are easier to detect over long periods of evolutionary time compared to genes under more subtle and short-lived periods of positive selection in association with the establishment of a new phenotype.

For noncoding regions, it is much more difficult to distinguish positive selection from relaxation of evolutionary constraint because it is difficult to establish an estimate of neutral evolution for those sequences. Models of positive selection in regulatory sequence is a current area of emerging research in the field and are not yet reliable enough to make the distinction between positive selection and accumulation of random mutations.

Author Response

Reviewer #1 (Public Review):

The authors took advantage of an existing protein-trap resource in zebrafish to identify genes important for normal pacemaker function in adults. They generated a collection of lines with mutation in genes that expressed at reasonably high levels in the heart and assess their ECG. They identified 3 candidates with increased incidence of sinus arrest and focused on validation of dnajb6b. The dnjb6b mutant fish display other defects including enhanced response to atropine and carbacol and bradycardia. They show that dnajb6b is expressed in a subset of cells in the sinus node in zebrafish. In mouse sinus node, DNAJB6 expressing cells have low expression of TBX3 and its target HCN4. In addition, Dnajb6b+/- mice also display similar phenotypes. Analysis of pacemaker function in ex vivo mouse hearts by high-resolution fluorescent optical mapping of action potentials revealed that the number of leading pacemakers in Dnajb6b+/- hearts is decreased in the sinus node, with a concomitant increase in the auxiliary pacemakers. RNAseq analysis of the right atrial tissues detected expression changes in ion channels and genes involved in Ca2+ handling and Wnt signaling. Overall, the results support the conclusion that DNAJB6 is important for proper sinus node function, thus adding it to the short list of sick sinus syndrome genes. However, the manuscript has several weaknesses.

Weakness:

The manuscript does not address the mechanism by which decreased DNAJB6B causes sick sinus syndrome. For example, it is unknown if DNAJB6B functions cell autonomously or non-cell autonomously in the sinus node. The RNAseq analysis identified changes in ion channels in the right atrial tissues of 1-year old mice, cellular electrophysiology of the sinus node cells was not assessed.

The main goal of this research is to prove the feasibility of discovering novel SSS genes in adults via a forward genetic approach in zebrafish. Thus, the major hallmark would be to prove causality and specificity of the candidate genes identified from this screen, such as Dnajb6. Comprehensive mechanistic study would be a focus for future studies.

Nevertheless, we carried out the following experiments to address the mechanisms. Based on these data, a new section was added to the discussion section (Lines 424-465).

(1) In mice, we did more antibody immunostaining and confirmed a negative correlation in terms of expression intensity between the Dnajb6 and Tbx3 proteins. We further detected a significantly increased Tbx3 immunostaining signal in the SAN tissues of Dnajb6 heterozygous mice compared to WT controls (new Figure 3D-F).

(2) In zebrafish, we compared expression patterns of the sqET33-mi59B conduction system reporter line between the GBT411/dnajb6b heterozygous and homozygous mutants. We found the atrio-ventricular canal (AVC) signal became diffused in GBT411/dnajb6b homozygous adult hearts. In addition, the ring-like structure usually seen in the SAN region of WT controls and in the GBT411/dnajb6 heterozygous was largely lost in 3 out of 9 GBT411/dnajb6b homozygous adult hearts examined (new Figure 2).

Together with the ectopic pacemaker activity detected in the Dnajb6 heterozygous mice (new Figure 5A and 5B), we speculate that Dnajb6 might act as a suppressor of Tbx3 transcription factor in defining cell fate specification into SAN pacemaker myocytes. Since Tbx3 was reported to suppress chamber myocardial differentiation (Mommersteeg et al., Circ Res. 2007;100(3):354-62), upregulation of Tbx3 may thus contribute to enhanced atrial ectopic activity in Dnajb6 heterozygous mice.

Furthermore, TBX3 has been recently identified as a component of the Wnt/β-catenin-dependent transcriptional complex (Zimmerli et al., eLife. 2020;9:e58123), which is significantly affected in Dnajb6 heterozygous mice (see new Figure 7B-C). This further supports a possible role of TBX3 in both SAN and atrial remodeling.

(3) Finally, in collaboration with Drs. Grandi, Morotti, and Ni from University of California Davis, we utilized a population-based computational modeling approach to determine the cellular/ionic mechanisms that could underlie the ex vivo observed SSS phenotype in the Dnajb6 heterozygous mice (new Figure 6). We used our previously published model of the mouse SAN myocyte (Morotti et al. Int J Mol Sci. 2021; 22(11):5645) and enhanced it with addition of both sympathetic and parasympathetic stimulations to model the effects of isoproterenol- and carbachol-induced changes in pacemaker activity (i.e., firing rate), respectively. We generated a population of 10,000 mouse SAN myocyte models by random modification of selected model parameters describing maximum ion channel conductances and ion transport rates from the baseline model and assessed isoproterenol- and carbachol-induced effects on each model variant. We then separated this population of models in two subpopulations representing the WT and Dnajb6+/- mice phenotypes: namely, we extracted the model variants that recapitulate changes observed in Dnajb6+/- vs. WT mice, including a reduced firing rate at baseline, an increased response to isoproterenol, and a decreased response to carbachol administration (new Figure 6). This filtering process resulted in n=438 models that correspond to the Dnajb6+/- mice phenotype and n=6,995 models that correspond to the WT phenotype. We analyzed the parameter value differences in these two subgroups to revealed several crucial parameters that are significantly correlated with the observed electrophysiological changes. The analysis revealed a significant decrease in the maximal conductances of the fast (Nav1.5) sodium current, the L-type Ca2+ current (ICa,L), the transient outward, sustained, and acetylcholine-activated K+ currents, the background Na+ and Ca2+ currents, as well as the ryanodine receptor maximal release flux of the Dnajb6+/- vs. WT model variants. We also found a significant increase in the Na+/Ca2+ exchanger (NCX) maximal transport rate, and conductances of the T-type Ca2+ current and the slowly-activating delayed rectifier K+ current. These new studies provide some novel mechanistic insights into the observed SSS phenotype in Dnajb6+/- mice. Importantly, these new in silico experiments add another conceptual level to the phenotype-based screening approach introduced in the current study to identify new genetic factors associated with SAN dysfunction. Direct testing of these mechanisms would require a substantial amount of single SAN cell patch clamp and confocal microscopy experiments which are out of scope of the current manuscript and will be pursued in a follow-up study.

The manuscript does not address why the zebrafish homozygous mutants are adult viable while the mouse homozygotes are embryonic lethal. The insertion of the GBT411 disrupt dnajb6b(L) but not dnajb6b(S), while the mouse mutation deletes the entire gene. Does this difference partially explain the difference?

Indeed, the difference between zebrafish and mouse can be partially explained by the fact that only the long isoform of dnajb6b gene, dnajb6b(L), was disrupted in the GBT411 mutant, while both the long-Dnajb6(L) and short-Dnajb6(S) isoforms of Dnajb6 gene was largely deleted in the Dnajb6 knockout mice. However, we think the main reason is probably that functional redundancy in zebrafish but not mouse: zebrafish has two dnajb6 homologues, dnajb6b and dnajb6a, while mouse has only one Dnajb6 homologue. We added these points to the paper (Lines 377-379).

Reviewer #2 (Public Review):

In this manuscript, the authors expand upon previous work describing development of a protein trap library made with the gene-break transposon. This library was screened to identify lines displaying gene trap expression in the heart (zebrafish insertional cardiac mutant collection). A pilot screen of these lines using adult ECG phenotypes identifies dnajb6b as a new gene important for cardiac rhythm. Using the GBT/dnajb6b zebrafish line, Ding et al. find a proportion of aged homozygous mutant fish (1.5-2 years) present sinus arrest episodes and reduced heart rate. Treating GBT411/dnajb6b mutant adults with compounds revealed aberrant responses to autonomic stimuli, and sinus arrest episodes were induced following verapamil exposure, providing evidence that GBT411/dnajb6b as an arrhythmia mutant. This conclusion could be better supported by presenting specific ECG parameters to characterize the conduction defect more thoroughly. The authors then report that Dnajb6+/- adult mice recapitulate some of the phenotypes observed in zebrafish, including sinus arrest and AV blocks, as well as impaired (although different) responses to autonomic stimuli. The authors describe that these are features of sick sinus syndrome in the absence of cardiomyopathy phenotypes in either the zebrafish or mouse lines. However, overall cardiac morphology is not well described for either the GBT411/dnajb6b or Dnajb6+/- models.

We carried out more experiments to examine left ventricular (LV) structure in Dnajb6 heterozygous mice at 1 year of age, using H&E staining, Masson’s trichrome staining, and transmission electron microscopy (TEM) analysis. We now show clearly that there are no significant myocardium structural changes in the LV as well as atrial and SAN tissues of Dnajb6 heterozygous mice (new Supplemental Figures 3 and 5), when the SSS phenotype was already noticeable. However, in the GBT411/dnajb6b heterozygous mutant at ~2 years of age, we detected severe sarcomere structural abnormality in 1 out of 3 fish hearts examined (see Response-only Figure 1). In addition, in a previous publication (Ding et al., Circ Res, 2013:112(40:606-17), we reported evident cardiac remodeling phenotypes in the GBT411/dnajb6b homozygous fish at 12 months of age.

Together, we have obtained more experimental evidence to strengthen the claim that arrhythmia is not due to cardiomyopathy/structural remodeling in the Dnajb6+/- mice. However, the evidence from fish remains weak. Therefore, we removed the claim that “when structural remodeling/cardiac dysfunction have not yet occurred” in fish and modified our statement in mice accordingly (Lines 372-377, 385-386).

To further support a role for Dnajb6 in sinoatrial node dysfunction, the authors performed optical mapping of action potentials from isolated mouse atrial tissue. These data reveal that Dnajb6+/- cultures exhibit ectopic pacemakers outside of the sinoatrial node, including within the atrial wall and inter-atrial septum. These data also show prolongation of SAN recovery time at baseline and following autonomic stimulation, further suggesting SAN dysfunction. RNA-sequencing experiments of DNAjb6+/- adult right atrial tissue showed differentially expressed genes encoding Ca2+ handling related proteins, ion channels, and WNT pathway related proteins. As these genes are involved in the cardiac conduction system, the authors suggest these pathways as molecular mechanisms underlying SSS phenotypes in Dnajb6 models.

Sick sinus syndrome is a relatively rare arrhythmia most commonly found in older populations. Therefore, it has been challenging to establish clinically relevant models and there is a limited understanding of mechanisms of SSS pathogenesis. One particular strength of this manuscript is the ECG phenotype-based forward screen of the gene-breaking transposon (GBT)-based gene trap library in aged animals. This pilot study provides proof-of-concept that this screening approach is well suited to identify regulators of cardiac function in adults and genes linked to adult diseases like SSS.

Thank you very much for recognizing the major strength of our manuscript!

Author Response

Reviewer #1 (Public Review):

In this manuscript, the authors investigated the role of a long noncoding RNA VPS9D1-AS1(VPS) in colorectal cancer (CRC). They found that a high level of VPS was negatively associated with T cell infiltration in CRC patients; in cell line-derived xenograft models or a conditional knock-in mouse model, VPS overexpression enhanced tumor growth and suppressed the infiltration of CD8+ T cells, which was reverted by VPS antisense oligonucleotide (ASO) treatment. They also investigated the molecular mechanisms underlying VPS function and revealed a VPS/TGF-β/ISG signaling cascade in tumor cells and crosstalk between tumors and T cells depending on IFNAR1 level.

The authors had performed extensive analyses on the functions of VPS using patient samples, CRC cell lines, xenograft tumors, and drug-induced tumors, and the data were of relatively good quality; they targeted VPS overexpression in cell line-derived xenografts or mouse tumors by ASO treatment as potential therapeutics, although the overexpression level may not be physiologically relevant. The authors also made great efforts to explore the mechanisms in vitro and proposed a very interesting model of ribosomes/VPS/TGF-β/ISG signaling axis in tumor cells and opposing regulation on IFNAR1 in tumor and T cells; however, the mechanistic model was tested in vitro, not in cell line-derived xenografts or mouse tumors used in the study, which undermined the authors' claims.

Thanks for these positive comments from reviewer #1, and we attached great importance to critical comments about in vivo data.

Reviewer #2 (Public Review):

In this paper, Yang et al. seek to show the importance of the lncRNA VPS9D1-AS1 in the biology and pathology of colorectal cancer (CRC). Starting with the analysis of patient data, and proceeding to cellular and animal cancer models.

Specifically, the authors report higher VPS9D1-AS1 levels in tumor tissues in two independent cohorts of CRC patients. There was a positive association between VPS9D1-AS1 levels and molecules involved in TGFb signaling, yet a negative association between VPS9D1-AS1 levels and levels of tumor-infiltrating CD8+ T cells (and a negative correlation of these levels of tumor-infiltrating CD8+ T cells and protein expression of molecules involved in TGFb signaling). Cell line studies revealed a positive feedback loop between VPS9D1-AS1 and TGFb signaling molecules, with a cell-intrinsic, pro-proliferative, and pro-survival effect of VPS9D1-AS1 on CRC cancer cells. VPS9D1-AS1 also controls the expression of several genes in the IFN pathway, in particular the ISGs IFI27 and OAS1. In addition, IFI27 and OAS1 expression are controlled by TGFb, TGFBR1, and SMAD1, and the promoter of OAS1 is targeted by SMAD4 (but also TGFb), which binds to it. VPS9D1-AS1 expression in tumor cells promotes PD1 expression and negatively affects IFNAR1 on T cells to reduce their effector functions. In vivo, MC38 CRC cells overexpressing VPS9D1-AS1 show increased tumor growth in mice, and animals with transgenic VPS9D1-AS1 expression in the intestine develop larger CRC lesions upon AOM/DSS treatment. Finally, in vivo targeting of VPS9D1-AS1 using anti-sense oligo reduced tumor size. The data indicate a series of intricate molecular and cellular interactions and suggest that VPS9D1-AS1 can help with patient stratification, improving prognostic prediction and allowing for personalized treatment.

Taken together, there is a multitude of datasets and several complementary experiments using patient-derived samples, genetically engineered cell lines, and mouse models. Definitely, the paper includes many avenues of inquiry that cover the broad field of cancer molecular biology, biochemistry, and pathogenesis. However, this broad approach renders the paper difficult to follow at times and also leads to numerous typographical and interpretive (but, largely, not methodological), mistakes. In addition, the quality of some of the figures needs to be improved before they can be properly evaluated.

In methodology, the authors are largely successful, and I would not recommend major changes to the work, other than to recommend a "focusing" of the manuscript objectives, or a paring of the data to better convey the desired story.

The experiments presented herein, particularly those that test the efficacy of the lncRNA as cancer therapeutics are important for the field, and should be of high import to other cancer biologists.

We thank you very much for your constructive comments. We had replied all your concerns

Reviewer #3 (Public Review):

The authors have accomplished large amounts of work to prove the role of VPS9D1-AS1 in promoting immune escape from cytotoxic T cells, and the mechanistic exploration is valid enough to support the conclusions, as well as the translational significance of this target through in vivo experiments. However, the logicality of the diagram requires improvement, and several revisions are warranted.

We thank for reviewer’s positive comments. We revised our manuscript according to your suggestions

Author Response

Reviewer #2 (Public Review):

This study has investigated the pathway for degradation of the inner nuclear membrane protein SUN2. Based on earlier studies that had searched for bTrCP substrates and interactors, the authors postulated that SUN2 might be a target of this ligase. They found two potential bTrCP recognition sites and showed that the second of these, Site2, is important for SUN2 turnover. A phospho-mimetic mutant is turned over faster, and a phospho-resistant mutant is turned over slower. The degradation is slowed by inhibitors of Neddylation (and therefore, of Cullin Ring Ligases), inhibitors of p97, and proteasome inhibitors. Using a genetic screen, they find bTrCP, components of the Cullin ring ligase, p97, the proteasome, and subunits of CK2. They use inhibitors to show that CK2 is needed for maximal SUN2 degradation, and a phosphatase called CTDNEP1 antagonises CK2-mediated SUN2 degradation. Using a non-degraded variant of SUN2, the authors show that its overexpression can influence nuclear morphology and various nuclear functions. In sum, the authors outline a pathway for regulated degradation of the inner nuclear membrane SUN2.

The study is generally sound in its logic, well written, and appropriately interpreted for the most part. The data are of high quality. The findings are new and will provide a foundation to now examine how LINC complex abundance is regulated. I have a number of suggestions for improvement, listed in order of importance. Only the first two should require any experimental work, and the second item is potentially optional depending on the authors' response. The remaining items can be handled with adjustments to the manuscript.

1) It is surprising that nowhere in the paper is an experiment directly and rigorously establishing that bTrCP is required for SUN2 degradation. I realise this is quite plausible from the shown experiments, but it seems to be a rather glaring oversight (apologies if I have missed it somewhere). At present, the current evidence for its role is the similarity of Site2 to a bTrCP recognition motif, the physical interaction of SUN2 with bTrCP, and the modulation of this interaction by mutants intended to mimic or eliminate phosphorylation. The inhibitor experiment is not strong evidence because it inhibits all CRLs. I would therefore recommend, at the least, to present an experiment knocking down or out bTrCP2 (i.e., FBXW11, which nicely showed up in the genetic screen). This simple experiment could be included in the validation experiments in Fig. S4b. It would be worth also including FBXW1A for comparison, and if needed, the double-knockdown. This seems essential to complete the study.

We thank the reviewer for suggestion. These experiments have now been included in the manuscript. For further information, see response to Editor’s point 1.

2) The experiments with TBCA are not complemented with knockdown experiments of CK2 subunits. I realise CK2 is essential, but cells can evidently tolerate acute knockdown sufficiently well to do experiments given that this came up in the CRISPR screen. I would think such knockdown experiments would strengthen the argument and mitigate any concern about the off-target effects of TBCA. Kinase inhibitors are often only partially specific, so arguments about the involvement of any kinase are stronger if inhibitor studies are complemented with genetic perturbations.

We thank the reviewer for suggestion. We have now included knockdown experiments with independent sgRNAs that validate our conclusion on the role of CK2 in SUN2 degradation (Figure Supplement 4C). In addition, we would like to point out that besides the essential CK2 regulatory subunit CSNK2B, our genome widescreen also identified the catalytic subunit CSNK2A2 (non-essential as it is redundant with CSNK2A1) (see Figure S3A). Considering that our library contains 4 sgRNAs per gene, this makes a total of 8 sgRNAs targeting subunits of CK2. Importantly no other kinase was identified in our screen. Moreover, TBCA is well established as a specific CK2 inhibitor. Altogether, these various observations make us quite confident that CK2 is the prime kinase controlling SUN2 stability.

3) Lines 173-183: MLN4924 is used interchangeably with inhibition of SCFbTrCP. But MLN4924 is an NAE inhibitor that indirectly inhibits all CRLs. It seems premature to invoke SCFbTrCP as being involved because the experiments have not yet established a role for this specific CRL (see point 1 above). Instead, the conclusion should be that the data indicate a role for one or more CRLs. At this point in the narrative, the only evidence that bTrCP is involved is the sequence similarity of site1 and site2 to canonical bTrCP recognition sites. However, this is not enough evidence as no experiments knocking down or knocking out bTrCP, or experiments showing a physical interaction, have been presented yet. That comes in the subsequent section.

We thank the reviewer for pointing this out. The text was modified and additional data on the depletion of βTrCP has been included in the revised manuscript to support our conclusions.

4) Line 195 - At this point in the narrative, there is no evidence that SUN2 is ubiquitinated by SCFbTrCP. This needs to be rephrased. I would think one can conclude at this point that SUN2 is degraded by a pathway that relies on a CRL, p97, and the proteasome. The degradation is controlled by Site2, potentially by phosphorylation (again, this has not really been established at this point in the story, even if it seems plausible based on the mutagenesis).

The sentence has been modified.

5) I think the discussion needs to include some thoughts on what the authors believe happens to the rest of the SUN2 trimer or more broadly, the LINC complexes. In other words, what is the consequence of degrading a single protein of a much larger complex? In this vein, the model shows monomeric SUN2. Is it worth showing that it is part of a trimer and part of the LINC complexes? Regardless of how the authors depict the model, discussing this issue seems worthwhile.

We thank the reviewer for the suggestion. We observe that the turnover rates of endogenous SUN2 is affected by the exogenous expression of SUN2 and primarily its derivatives Site 2A and Site 2D (Figure 1E). The effects are likely due to the assembly of trimers containing both endogenous and exogenous SUN2. This observation also suggests that degradation of one of the subunits in the trimer leads to the degradation of the other two. However, in the current manuscript we do not directly test or analyse these models or look at SUN2 complexes.

6) Lines 225-226 - again, MLN4924 is not an inhibitor of SCFbTrCP, but rather a CRL inhibitor. The evidence for bTrCP being the key ligase is still missing at this point in the narrative.

We now present evidence in an earlier figure that βTrCP is the F-Box involved in SUN2 degradation. In this context, the sentence appears correct.

7) Fig. 5G is not especially convincing - to my eye, the effect on endogenous SUN2 is very similar to the effect on the transgene SUN2-site2A mutant, but simply a fainter exposure. Can the authors provide some numbers to allay this concern? It might well be that there is little difference between the behaviour of the endogenous and exogenous SUN2 in this experiment because they engage in heterotrimeric complexes. Also, why is the transgenic SUN2 not detected on the SUN2 blot? Would it not be evident at ~100 kD?

We have consistently seen that SUN2 Site 2A is refractory to CTDNEP1 regulation. The blot has been replaced to better convey this result.

The transgenic SUN2 is not detected in this blot because while the same cell lines were used for this experiment, to visualise the endogenous SUN2, doxycycline were not added to these cells. Thus, two sets of lysates were collected, one for cells that were treated with Doxycycline (transgene) and one without Doxycycline (endogenous). This is explained in the figure legend.

8) In panel 1E, the heterologously expressed SUN2 protein has two bands, with the upper band being more readily degraded than the lower band in some cases. Is the upper band the phosphorylated product? Might be worth a comment if anything is known about what the two bands represent.

We believe that the two bands do not correspond to different phosphorylated SUN2 forms. This is based on the analysis of SUN2 by SDS-PAGE in presence of Phostag reagent and the fact that two bands are seen both also for non-phosphorylatable and phospho-mimetic SUN2 derivatives. The appearance of two bands has been observed for other ERAD substrates characterized in our lab (for example Weijer et al. 2020) and appears to depend on the lysis conditions (see for example Figure 2 and 3).

9) Worth mentioning in the main text that FBXW11 is bTRCP2. Also, it is worth noting whether bTRCP1 (FBXW1A) was a hit on the screen or not.

Thanks for the suggestion. We have now included this information.

Reviewer #3 (Public Review):

The manuscript by Krshnan et al. reports a cellular mechanism akin to the endoplasmic reticulum-associated degradation (ERAD) that degrades SUN2, a nuclear inner membrane protein. The authors previously identified the Asi ubiquitin ligase complex that mediates the degradation of inner nuclear membrane proteins in budding yeast. In this manuscript, they identified the SCF β TrCP, and SCF as another ligase that regulates the ubiquitination and degradation of SUN2 in mammalian cells. The key findings include the identification of a substrate recognition motif that appears to undergo casein kinase (CK) dependent phosphorylation. Mutagenesis studies show that mutants defective in phosphorylation are stabilized while a phosphor-mimetic mutant is more unstable. They further show that the degradation of SUN2 requires the AAA ATPase p97, which allows them to draw the analogy between SUN2 degradation and Vpu-induced degradation of CD4, which occurs on the ER membrane via the ERAD pathway. Lastly, they show that the stability of endogenous SUN2 is regulated by a phosphatase and that over-expression of a non-degradable SUN2 variant disrupts nuclear envelope morphology, cell cycle kinetics, and DNA repair efficiency. Overall, the study dissects another example of inner nuclear envelope protein turnover and the involvement of a pair of kinase and phosphatase in this regulation. The data are of extremely high quality and the manuscript is clearly written. That being said, the following questions should be addressed to improve the robustness of the conclusions and to avoid potential misinterpretation of the data.

1) Since SUN2 is normally incorporated into a SUN2-SYNE2-KASH2 LINC heterohexamer complex, the authors should be cautious with the use of over-expressed SUN2 in this study. Over-expressed SUN2 is expected to stay mostly as unassembled molecules and thus is likely degraded by a protein quality control mechanism that targets unassembled proteins. Consistent with this possibility, CK2 has been implicated in the regulated turnover of aggregation-prone proteins (Watabe, M. et al., JCS 2011). This mechanism would be potentially distinct from the one proposed for endogenous SUN2 degradation.

We thank reviewers for the suggestion to provide further genetic evidence of the involvement of βTrCP1 and 2 F-box proteins in the degradation of SUN2. We now show that maximum stabilization of endogenous (Figure Supplement S4D) and transgenic (Figure S2) SUN2 is observed upon simultaneous depletion of βTrCP1 and βTrCP2 indicating that these F-Box proteins are redundant. Depletion of βTrCP1 alone did not impact SUN2 levels while depletion of βTrCP2 increased SUN2 steady state levels, with the effect being more pronounced for overexpressed SUN2. Depletion of other F-Box proteins did not affect SUN2 levels indicating that the effect observed for βTrCP1 is specific (Figure S2B). These results are in line with the results of our genome wide screen (Figure 4 and S3) and the literature. The differences in the effects of βTrCP1 and βTrCP2 depletion likely result from the relative abundance of the two F-Box proteins in the HEK cells used in this study.

2) Certain conclusions appear to be an overstatement. This is particularly the case for the title, which implies that SUN2 is a protein that undergoes regulated turnover (under certain physiological conditions). Given that CK2 is a constitutive kinase and that the authors have not identified the conditions under which the activity of CTDNEP1 is regulated, it is premature to make such a conclusion.

We disagree with the reviewer in this point. We present clear evidence that the turnover rate of SUN2 (both overexpressed and endogenous) is regulated by opposing kinase/phosphatase activities. This per se implies a mode of regulation. Similar kinase/phosphatase balances regulate a plethora of physiologic processes (from cell cycle progression to DNA repair) and the term “regulation” is commonly used in these contexts. We agree with the reviewer that upstream events controlling SUN2 remain elusive however, we do present evidence the balance of CK2 and CTDNEP1 activities regulate SUN2 degradation.

3) Likewise, the demonstration of the impact of SUN2 accumulation on different cellular pathways mainly relies on the over-expression of a non-degradable SUN2 mutant. Whether similar defects could be seen when the degradation of endogenous SUN2 is blocked remains an open question.

It would be great to gene edit the SUN2 locus to introduce the desired mutations. But as pointed out this is not trivial, in particular considering that the desired mutations would need to be introduced in both chromosomal copies.

Author Response

Reviewer #1 (Public Review):

“Overall this is an interesting study of the function of ATP6AP2 in the osteoblastic lineage. This gene is unstudied in the osteoblast, despite its known role in WNT signaling. In this study, the authors first show that loss of this gene in mature osteoblasts results in a strong cortical bone phenotype, with reduced osteocyte numbers and disorganized collagen. This phenotype is not present at birth but progressively worsens as the animals reach weaning age. In the compact bone, they show that loss of ATP6AP2 results in osteocytes largely devoid of dendritic processes. Loss of this gene starting at the osteocyte stage results in a milder phenotype. They then show that the osteocytes presenting have reduced MMP14 and that partial restoration of MM14 attenuates the severity of the cortical phenotype.”

Strengths

“This study uses cutting-edge microscopy to thoroughly characterize how and where the loss of ATP6AP2 in either the mature osteoblast or the osteocyte results in disorganized bone. Innovative proteomics techniques are used to identify cell surface proteins, including MMP14 that may mediate this phenotype. Two cre-drivers are used to determine when in the osteoblast-osteocyte lineage this gene has the maximum effect. Lastly, in vivo lentivirus replacement is used to test if the replacement of MMP14 can rescue the phenotype. This latter experiment solidifies the importance of MMP14 as a major player in the downstream sequela of ATP6AP2 action.”

Weaknesses

“Unfortunately, all of the histology is conducted on demineralized bone, and counts of osteoblasts and osteoclasts on the bone surface are not presented. This reduces the ability to interpret all downstream work. As such, the extent of the mineralization defects is difficult to interpret. Much of this paper is focused on the osteocyte, which is curious as the phenotype of the mature osteoblasts ATP6AP2 knockout mice is so much more severe than that of the osteocyte ATP6AP2 knockout mice. While it is clear how MMP14 was identified as being deficient in the mature osteoblasts ATP6AP2 knockout cells, it is not obvious how this gene became the sole focus of the remainder of this paper. This phenotype progresses as the mice become ambulatory and therefore weight bearing on their limbs. This could partially explain the presentation of the mouse phenotype, but this is not discussed.”

Good suggestions! We have performed the suggested experiments on mineralized bone sections, and quantified both osteoblasts and osteoclasts on the bone surface.

The results, shown in Fig. 1A-B above, demonstrated increased osteoclast numbers in both trabecular and endocortical bone surfaces in ATP6AP2 mutant mice, which were accompanied with elevated bone resorption (see Fig. 1C, measured by serum levels of PYD). However, upon bisphosphonates (alendronate) treatment, an inhibitor of osteoclastic activity, the trabecular bone mass was restored, but little effect on the cortical bone phenotype in the mutant mice (Fig. 2A-F). These results thus suggest an osteoclast activity independent cortical bone phenotype in the mutant mice.

We thus further investigated the cortical bone phenotypes and the osteoclast independent underlying mechanisms. Whereases no significant change in the number of osteoblasts was detected in the metaphysis region of femur in ATP6AP2Ocn-cre mice (see Fig. 6A-B below), we did detect mineralization deficit in the mutant mice by both in vivo and in vitro experiments (see Fig. 5 and Fig. 7). These results suggest that the increased cortical woven bone in the mutant mice is likely due to an impairment in the replacement of woven bone with the mineralized cortical bone matrix.

Additionally, the expression levels of ATP6AP2 in osteocytes appeared to be similar to that in osteoblasts and BMSCs (Fig. 8). The phenotype of osteocyte ATP6AP2 knockout mice (ATP6AP2DMP-Cre) appeared to be weaker than that of the BMSCs/osteoblasts ATP6AP2 knockout mice (ATP6AP2OCN-Cre) led us to speculate that ATP6AP2 in Ocn-Cre+ osteoblastic cells (e.g., immature osteocytes) may play a more critical role than that in DMP1-Cre+ mature osteocytes in regulating cortical bone matrix remodeling and osteocyte development.

These results and points, in line with our model, will be included into a revised manuscript.

Reviewer #3 (Public Review):

“In this work, the authors have assessed the bone phenotype of a mouse with targeted ablation of the vacuolar ATPase accessory protein ATP6AP2 in the osteoblast lineage. They observe a clear increase in cortical thickness, but the cortex is highly porous and contains remnant cartilage as well as extensive woven bone. They then follow this by suggesting that one cause of this phenotype may be a change in the surface expression of the protein MMP14, a matrix metalloproteinase, known to be involved in bone matrix degradation, at least in osteoclasts. They provide evidence that this protein may also regulate matrix degradation surrounding osteocytes and an increase in this protein in osteocytes lacking ATP6AP2 may be a cause of the initial phenotype described.”

While the phenotype described is very dramatic, the interpretation that it reflects a defect in osteoblast to osteocyte transition is questioned by this reviewer. The phenotype appears to be an osteopetrosis, including a lack of remodelling of the cortex. Cartilage and woven bone are not replaced effectively by lamellar bone. The bone contains ample osteocytes, but they are the osteocytes typical of woven bone, with rounded cell bodies, disordered organisation, low sclerostin expression, and short dendritic processes. The defect in the ATP6AP2 mice is a lack of cortical remodelling during cortical consolidation (for review see PMID: 34196732). Cartilage and woven bone remnants, which are normally remodelled as cortical bone matures, remain in the cortex until adulthood. It is not clear whether this results from reduced or increased remodelling of the cortex, but it is not because the osteoblasts cannot form osteocytes.

Some of the data is very challenging to interpret because of low sample numbers (n=4 for much of the analysis), and lack of detail as to the sex of the animals. Regions used for imaging, histomorphometry, and dynamic histomorphometry all need to be defined throughout the work. Since the cortex differs dramatically by site, and by distance from the growth plate (due to the different stages of maturation) this is critical. Some methods are not defined, although they could be of great use to the field (e.g. the method for assessing bone degradation by MMP14).

Good suggestions! We will describe the results more precisely in a revised manuscript.

Author Response

Reviewer #1 (Public Review):

In this work, Bentley et al. describe the development and use of a novel microfluidic platform to study motility of green algae. By confining algae to circular corrals of various diameters (and with a height that renders the system quasi-two-dimensional), the authors gather extremely long time series of the swimming trajectories under various degrees of lateral confinement, in the presence of several different kinds of perturbations.

The data is presented in a number of ways, most importantly by means of transitions between the three characteristic states of motion for these algae. This allows contact to be made with ideas from nonequilibrium dynamical systems by examining the transition probabilities between those states and identifying nonequilibrium characteristics of the fluxes between them.

Overall the work is extremely impressive in terms of the data acquisition and careful time series analysis. The work falls short though in not following through on the many interesting observations that can be deduced from the data to come to precise conclusions about the biology and physics. For example, we see in Figs. 2 and 3 the effects of confinement on the trajectories, leading to clearly chiral motion at the strongest confinement. I would have expected the next step of the analysis to be a study of this problem in the context of, say, a Fokker-Planck equation for the probability distribution function for orientations, complete with boundary conditions that encode the scattering laws that we know from prior work by Kantsler et al. and others. Similar comments can be made about the other observations, which are followed up with any clear mechanistic analysis or comparison with theory.

The example above suggests that this paper, in its current form, is more akin to a "Methodology" paper than one that discovers new phenomena and explains them.

We thank the reviewer for their summary of our work and for these pertinent comments. As discussed above, we performed new experiments and modelling to successfully answer the main question of why chiral circling appears in the smallest traps (highest confinement), and also why the chirality depends on light. As demonstrated in the prior Cammann et al PNAS (and to some extent also the Ostapenko et al, PRL) study, encoding the scattering laws measured from Kantsler et al for a basic swimmer produces chiral motion (circular movement). An analytical treatment in terms of FP equations already appears in these prior studies. However, the novelty here is why this circular movement should remain chiral in the time-averaged sense.

In the revisions, we restrict ourselves to a conceptual-level where we show how a small internal asymmetry at the cellular level suffices to produce macroscopic chirality and how this depends on the size of the trap. Our new explanation reaches a precise conclusion about how a fundamentally biological phenomenon (slightly asymmetric flagellar beating in a phototactic swimmer), leads to a confinementinduced physical phenomenon (a preferred sense of circular swimming).

In a separate follow-up study, we will extend our model to incorporate more realistic parameters from the current dataset (e.g. time-dependent speed, stochastic reorientations, shock-responses, softness of the potential etc) to understand more subtle aspects of the high-resolution data we acquired.

Reviewer #2 (Public Review):

The authors use microfluidic devices to follow single swimmers for long periods, measuring their movement in detail and allowing detailed statistics at a level that has never been possible before and machine learning.

Its strength is the extraordinary detail and the doors opened by the quality of the resultant data. As such it makes a substantial contribution to a narrow field and adds slightly more subtly to an important field of full mathematically accessible descriptions of migration phenotypes.

Its weakness is that these tools are not yet used for any particularly enlightening tests. The directed probability fluxes are interesting, but not surprising. The strength of this paper is in the method, the analysis, and the ability to generate rigorous datasets.

We thank the reviewer for highlighting the quality and detail of our datasets, and we agree with the criticisms raised. We hope these weaknesses are now rectified in the revision. There is clearly scope to do much more now that we have access to this data, and we demonstrate this in the revision with our new model/interpretation.

We highlight three innovations of our work that may not have been made clear in the previous draft.

1. We have suggested a new paradigm for analysis of microbial motility and behaviour. The extraction of state transition probabilities from single-cell trajectories reveals exactly how motility changes at the subcellular level, which is much more informative than whether an organism speeds up or slows down on average. This tells us how does a given individual modulates the balance of possible behavioural states in response to their environment and also over time. These concepts apply not just microbes but to any behaving organism.

2. The emphasis on keeping track of the ‘arrow of time’ in the analysis of movement trajectories in important, can again be applied to any organism. As discussed above, while circling behaviour or symmetry breaking in confinement may not be surprising itself (though that does not prevent the flurry of experimental and theoretical papers on this topic), we argue that the emergence of chirality in the timeaveraged trajectory is surprising and does requires more subtle treatment. We now suggest this is down to a very small amount of internal symmetry breaking – it is interesting that such a small amount of symmetry-breaking at the sub-cellular scale can manifest as robust symmetry-breaking at the macroscopic scale.

This kind of insight has broad implications for understanding how (even simple) organisms can dramatically alter how they interact with their physical environment by effecting even minute internal adjustments. This could also motivate the design of novel biomimetic artificial devices or microswimmers.

1. Our approach of fusing droplets to investigate rapid motility responses to chemicals has plenty of potential for drug screening and also for investigating cellular transduction pathways (e.g. functional assays of mutants). We demonstrate its operation here as proof of concept on one species for one chemical only, but there are clear advantages over traditional approaches involving setting up chemical gradients or similar, where it is impossible to get a handle on instantaneous cell reactions nor individual-level responses.

Author Response

Reviewer #1 (Public Review):

This paper follows several innovative articles from the authors exploring the molecular mechanisms of insulin and IGF1 receptors activation by their ligands using cryo-electron microscopy. Here the authors explore the role of an alpha helical C-terminal segment (called the alpha-CT motif) of a disordered disulfide-linked insert domain in the FnIII-2 module of the insulin and IGF1 receptors (at the end of the alpha subunit), in the mechanism of ligand binding, negative cooperativity and receptor activation.

Biochemical data gathered over several decades have suggested that insulin and IGF1 use two separate binding sites, site 1 and site 2, to bind to two distinct domains (sites 1 and 2, and 1'and 2') on each protomer of the homodimeric receptors, disposed in an antiparallel symmetry. This disposition was corroborated by the early x-ray crystallographic studies of the unliganded insulin receptor ectodomain (apo-receptor). A subsequent somewhat surprising finding was that the insulin receptor site 1 is in fact a composite, made of the beta surface of the L1 module of one protomer, and of the alpha-CT motif of the other protomer which binds perpendicularly to the L1 surface (a "tandem binding element"), with insulin binding more to the alpha-CT motif than to L1.

Previous work from the authors showed that the subsaturated insulin receptor has an asymmetric configuration while the receptor saturated with 4 insulins has a symmetric T-shaped configuration. In contrast, the IGF1R shows only one IGF1 bound to an asymmetric configuration, indicating according to the authors a stronger negative cooperativity. This is attributed to a more rigid and elongated conformation of the alpha-CT motives that restricts the structural flexibility of the alternate binding site.

To test this hypothesis, the authors determined the cryo-EM structure of IGF1 bound to IGF1R with a mutated alpha-CT motif elongated by four glycine residues. Strikingly, a portion of these constructs adopt a T-shaped symmetric structure.

Conversely, they show that the cryo-EM structure of insulin bound to an insulin receptor with non-covalently bound alpha-CTs insert domains by mutation of the cysteines to serine adopts asymmetric conformations even at saturated insulin concentrations. They conclude that the alpha-CTs in disulfide-linked insert domains of the insulin receptor play an important role in the structural transition from asymmetric to symmetric during the insulin-induced insulin receptor activation.

All in all, this is a very interesting and well-designed study that represents an advance in the knowledge of the insulin/IGF1 receptor systems, although the details of the structural interpretations deserve some discussion.

This is very clear and succinct summary of our work. We thank Dr. Pierre De Meyts for the positive assessment of our manuscript, and we greatly appreciate the constructive comments which we have addressed.

Reviewer #2 (Public Review):

Li et al build upon recent observations that the alphaCT peptide is a key element in the IGF-1R and IR that regulates negative cooperativity and receptor activation. The use of IGF-1R and IR mutants builds upon previous observations with these mutants by Li et al (IGF-1R) and Weis et al. (IR).

Here they determined the structures of the IGF-1R mutant, IGF-1R-P673G4, which has a 4 glycine motif inserted at residue P673 at 4Å resolution. By introducing structural flexibility in the alphaCT the IGF-1R is able to bind 2 IGFs and adopts a symmetric T conformation, which is in contrast to the single IGF bound WT IGF-1R that adopts an asymmetric conformation. The ability to bind two IGFs is taken as a sign that negative cooperativity has been affected and confirms the importance of the alphaCT in constraining the IGF-1R into the asymmetric conformation. The increased flexibility of the alphaCT linkage between the two receptor monomers results in reduced ability of IGF-I to activate the IGF-1R and Erk leading to reduced IGF-1R internalisation. This is consistent with previous reports that effective Erk signalling is dependent on endosomal signalling. A second mutant, IGF-1R -3CS, was also used where a cysteine triplet in the alphaCT is mutated to serine to perturb the disulfide bonding between the alphaCTs of the two monomers. IGF-1R activation and signalling by this mutant was also reduced.

In addition, the structure of an equivalent insulin receptor mutant, IR-3CS, was determined with complexes formed with excess insulin. Again, the increased flexibility of the alphaCT altered the structural rearrangement upon ligand binding. Three conformations with 4 insulins bound were detected, two unique asymmetric (4.5 Å and 4.9 Å) and one symmetric (3.7 Å), whereas WT IR:insulin complex predominantly forms a symmetric T 4 insulin bound structure. This suggests the IR alphaCT is important in stablising the active T structure. In contrast to the IGF-1R -3CS, the IR-3CS has the same affinity for insulin as WT IR, is more potently activated (pY1150/Y1151) by insulin but has a reduced signalling response. This demonstrates the role of alphaCT in the activation.

Whilst the symmetric IR-3CS:insulin complex structure is compared with the WT IR: insulin complex, no comparisons were made between the asymmetric conformations described here and those previously reported. Is the ligand binding in the asymmetric conformations different to the asymmetric binding seen when WT IR:insulin complexes were generated at low insulin concentrations? It would be interesting to see these overlaid. How do these asymmetric conformations relate to the existing asymmetric conformations reported by Nielsen (10.1016/j.jmb.2022.167458) and Xiong (DOI 10.1038/s41589-022-00981-0)?

Thank you for the good suggestions. We have now prepared a new Figure 4-supplement 2 that compares the structure of asymmetric IR-3CS/insulin with that of asymmetric IR bound with subsaturated insulin previously published by us and others. All asymmetric structures of IR bound with subsaturated insulin have similar structural features, i.e., in one half of the complex, one insulin bound at site-1 also contacts site-2 from adjacent protomer, or vice versa. However, in the asymmetric structure of IR-3CS/insulin, two insulins were bound at the hybrid site in the middle of the IR-3CS/insulin complex. To accommodate the binding of two insulins, the L1/αCT together with bound site-1 insulin move outward as compared to the asymmetric structure of IR bound with subsaturated insulin. This is the major structural difference between these asymmetric structures. We have discussed this in the revised manuscript.

What is the distance between the FnIII-3 domains of the IR:insulin asymmetric conformations and in the symmetric structure? Does this correlate to activity as is seen for the IGF-1R-P673G4? It would be good to comment on this, particularly as there is an interesting disconnect between the receptor activation and downstream signalling activity. Why is there greater pY1150/Y1151 activation than for the WT IR and how can the lower downstream signalling activity be explained?

We thank Dr. Briony Forbes for raising this point. Asymmetric IR-3CS/insulin, asymmetric IR/insulin and symmetric IR/insulin have similar distances between their membrane-proximal regions (approximately 30 – 35 Å). This indicates that the distances between the membrane-proximal regions within these complexes are all short enough to allow the intracellular kinase to undergo efficient autophosphorylation, in contrast to IGF1R.

As indicated by Dr. Forbes, our cellular functional assays showed that the IR-3CS has higher levels of autophosphorylation, but lower levels of downstream signaling activity and a defect in endocytosis. Although the distances between the membrane-proximal regions are similar, the relative positions and orientations between the two membrane proximal regions are significantly different between asymmetric IR-3CS and symmetric IR. Given the fact that the FnIII-3 domain is connected to the transmembrane domain by a short linker containing four residues, we speculate that the structural differences in the extracellular domains may lead to both differential dimeric assembly of transmembrane and intracellular domains, as well as the stable interaction between the intracellular IR domains and downstream adaptors and effectors. This could in part explain why IR-3CS can still undergo robust autophosphorylation but its downstream signaling becomes defective. Similar hypothesis has been proposed in the EGF and TGF-α induced activation of EGFR (PMID: 34846302). The endocytosis defects of IR-3CS might be the result of reduced IR signaling, but it is tempting to speculate that less endocytosis of IR-3CS may cause defective downstream signaling. The structure of transmembrane and intracellular domains in the context of the entire full-length/insulin complex needs to be further investigated. We have included new analysis and expanded the discussion.

It would be good to reword the opening statement that "IGF1 only has one type of ligand binding site (site-1)" to acknowledge that two binding sites on IGF-I have been detected through analysis of competition binding studies which are fitted to a two-site sequential model and detect both high affinity and low affinity binding (Kiselyov). Site directed mutagenesis studies of both IGFs have detected two binding surfaces analogous to insulin's site 1 and site 2 (Gaugin et al and Alvino et al). Furthermore, binding assays with mini-IGF-1R (L1, CR, L2 fused to alphaCT, ie site 1 only) clearly demonstrated that IGF-II site 2 residues do contribute to overall binding affinity (Alvino et al). Perhaps we are yet to capture site 2 of IGF-1R as it is not in the same location as IR site 2? It would be good to comment on this.

Point accepted. Gauguin L. et al. demonstrated that alanine mutagenesis in IGF1, including E9A, D12A, F16A, D53A, L54A, and E58A, markedly reduced IGF1R-binding affinity. With the exception of IGF1 E9 (Site-1b of IGF1R, the same position in IGF2, E12), none of IGF1 D12, F16, D53, L54, and E58 are involved in the binding to site-1, suggesting that IGF1 has an additional site that maximizes the binding to the receptor. Despite saturated IGF1 levels, however, our previous and current structural studies did not reveal the putative site-2 of IGF1R-IGF1 binding. We speculate that IGF1 binds to site-2 transiently, which might be important for IGF1-induced activation of IGF1R. We have revised the manuscript and expanded the discussion.

Reviewer #3 (Public Review):

Li et al. present cryo-EM structures of the insulin receptor (IR) and insulin-like growth factor-1 receptor (IGF1R), exploring the functional roles of the disulfide-linked alphaCT regions in ligand binding and receptor activation.

Cryo-EM structures of mutants of IGF1R and IR designed to increase the flexibility between disulfide-linked alphaCT regions revealed conformational states that were distinct from those of the wild-type (WT) receptors. Mutant (P673G4) IGF1R displayed conformations in which two IGF1 molecules were bound, rather than the 1:1 ligand:receptor state observed previously for WT IGF1R. Mutant (3CS) IR displayed asymmetric conformations with four insulin molecules bound, as well as the symmetric T conformation with four insulin molecules bound observed previously for WT IR. In each case, the mutant receptor was shown in cells to be poorly activated by its respective ligand.

This study demonstrates the importance of the disulfide-coupled alphaCT regions in the IR and IGF1R for ligand binding and receptor activation. What is not resolved in this study is whether differences in the alphaCT regions of these two highly related receptors contribute to their disparate active states - asymmetric for IGF1R (and 1:1 IGF1:IGF1R) vs. symmetric (T) for IR (and 4:1 insulin:IR).

We thank Dr. Stevan Hubbard for the positive assessment of our manuscript, and we greatly appreciate the constructive comments which we have addressed.

Author Response

Reviewer #1 (Public Review):

In one of the most creative eDNA studies I have had the pleasure to review, the authors have taken advantage of an existing program several decades old to address whether insect declines are indeed occurring - an active area of discussion and debate within ecology. Here, they extracted arthropod environmental DNA (eDNA) from pulverized leaf samples collected from different tree species across different habitats. Their aim was to assess the arthropod community composition within the canopies of these trees during the time of collection to assess whether arthropod richness, diversity, and biomass were declining. By utilizing these leaf samples, the greatest shortcoming of assessing arthropod declines - the lack of historical data to compare to - was overcome, and strong timeseries evidence can now be used to inform the discussion. Through their use of eDNA metabarcoding, they were able to determine that richness was not declining, but there was evidence of beta diversity loss due to biotic homogenization occurring across different habitats. Furthermore, their application of qPCR to assess changes in eDNA copy number temporally and associate those changes with changes to arthropod biomass provided support to the argument that arthropod biomass is indeed declining. Taken together, these data add substantial weight to the current discussion regarding how arthropods are being affected in the Anthropocene.

Thank you very much for the positive assessment of our work.

I find the conclusions of the paper to be sound and mostly defensible, though there are some issues to take note of that may undermine these findings.

Firstly, I saw no explanation of the requisite controls for such an experiment. An experiment of this scale should have detailed explanations of the field/equipment controls, extraction controls, and PCR controls to ensure there are no contamination issues that would otherwise undermine the entirety of the study. At one point in the manuscript the presence of controls is mentioned just once, so I surmise they must exist. Trusting such results needs to be taken with caution until such evidence is clearly outlined. Furthermore, the plate layout which includes these controls would help assess the extent of tag-jumping, should the plate plan proposed in Taberlet et al., 2018 be adopted.

Second, without the presence of adequate controls, filtering schemes would be unable to determine whether there were contaminants and also be unable to remove them. This would also prevent samples from being filtered out should there be excessive levels of contamination present. Without such information, it makes it difficult to fully trust the data as presented.

Finally, there is insufficient detail regarding the decontamination procedures of equipment used to prepare the samples (e.g., the cryomil). Without clear explanations of the steps the authors took to ensure samples were handled and prepared correctly, there is yet more concern that there may be unseen problems with the dataset.

We are well aware of the potential issues and consequences of contamination in our work. However, we are also confident that our field and laboratory procedures adequately rule out these issues. We agree with the reviewer that we should expand more on our reasoning. Hence, we have now significantly expanded the Methods section outlining controls and sample purity, particularly under “Tree samples of the German Environmental Specimen Bank – Standardized time series samples stored at ultra-low temperatures” (lines 303-304), “Test for DNA carryover in the cryomill” (lines 448-464) and “Statistical analysis” (lines 570-575).

We ran negative control extractions as well as negative control PCRs with all samples. These controls were sequenced along with all samples and used to explore the effect of experimental contamination. With the exception of a few reads of abundant taxa, these controls were mostly clean. We report this in more detail now in the Methods under “Sequence analysis” (lines 570-575). This suggests that our data are free of experimental contamination or tag jumping issues.

We have also expanded on the avoidance of contamination in our field sampling protocols. The ESB has been set up for monitoring even the tiniest trace amounts of chemicals. Carryover between samples would render the samples useless. Hence, highly clean and standardized protocols are implemented. All samples are only collected with sterilized equipment under sterile conditions. Each piece of equipment is thoroughly decontaminated before sampling.

The cryomill is another potential source of cross-contamination. The mill is disassembled after each sample and thoroughly cleaned. Milled samples have already been tested for chemical carryover, and none was found. We have now added an additional analysis to rule out DNA carryover. We received the milling schedule of samples for the past years. Assuming samples get contaminated by carryover between milling runs, two consecutive samples should show signatures of this carryover. We tested this for singletaxon carryover as well as community-wide beta diversity, but did not find any signal of contamination. This gives us confidence that our samples are very pure. The results of this test are now reported in the manuscript (Suppl. Fig 12 & Suppl. Table 3).

Reviewer #2 (Public Review):

Krehenwinkel et al. investigated the long-term temporal dynamics of arthropod communities using environmental DNA (eDNA) remained in archived leave samples. The authors first developed a method to recover arthropod eDNA from archived leave samples and carefully tested whether the developed method could reasonably reveal the dynamics of arthropod communities where the leave samples originated. Then, using the eDNA method, the authors analyzed 30-year-long well-archived tree leaf samples in Germany and reconstructed the long-term temporal dynamics of arthropod communities associated with the tree species. The reconstructed time series includes several thousand arthropod species belonging to 23 orders, and the authors found interesting patterns in the time series. Contrary to some previous studies, the authors did not find widespread temporal α-diversity (OTU richness and haplotype diversity) declines. Instead, β-diversity among study sites gradually decreased, suggesting that the arthropod communities are more spatially homogenized in recent years. Overall, the authors suggested that the temporal dynamics of arthropod communities may be complex and involve changes in α- and β-diversity and demonstrated the usefulness of their unique eDNA-based approach.

Strengths:

The authors' idea that using eDNA remained in archived leave samples is unique and potentially applicable to other systems. For example, different types of specimens archived in museums may be utilized for reconstructing long-term community dynamics of other organisms, which would be beneficial for understanding and predicting ecosystem dynamics.

A great strength of this work is that the authors very carefully tested their method. For example, the authors tested the effects of powdered leaves input weights, sampling methods, storing methods, PCR primers, and days from last precipitation to sampling on the eDNA metabarcoding results. The results showed that the tested variables did not significantly impact the eDNA metabarcoding results, which convinced me that the proposed method reasonably recovers arthropod eDNA from the archived leaf samples. Furthermore, the authors developed a method that can separately quantify 18S DNA copy numbers of arthropods and plants, which enables the estimations of relative arthropod eDNA copy numbers. While most eDNA studies provide relative abundance only, the DNA copy numbers measured in this study provide valuable information on arthropod community dynamics.

Overall, the authors' idea is excellent, and I believe that the developed eDNA methodology reasonably reconstructed the long-term temporal dynamics of the target organisms, which are major strengths of this study.

Thank you very much for the positive assessment of our work.

Weaknesses:

Although this work has major strengths in the eDNA experimental part, there are concerns in DNA sequence processing and statistical analyses.

Statistical methods to analyze the temporal trend are too simplistic. The methods used in the study did not consider possible autocorrelation and other structures that the eDNA time series might have. It is well known that the applications of simple linear models to time series with autocorrelation structure incorrectly detect a "significant" temporal trend. For example, a linear model can often detect a significant trend even in a random walk time series.

We have now reanalyzed our data controlling for autocorrelation and for non-linear changes of abundance and recover no change to our results. We have added this information to the manuscript under “Statistical analysis” (lines 629-644).

Also, there are some issues regarding the DNA sequence analysis and the subsequent use of the results. For example, read abundance was used in the statistical model, but the read abundance cannot be a proxy for species abundance/biomass. Because the total 18S DNA copy numbers of arthropods were quantified in the study, multiplying the sequence-based relative abundance by the total 18S DNA copy numbers may produce a better proxy of the abundance of arthropods, and the use of such a better proxy would be more appropriate here. In addition, a coverage-based rarefaction enables a more rigorous comparison of diversity (OTU diversity or haplotype diversity) than the readbased rarefaction does.

We did not use read abundance as a proxy for abundance, but used our qPCR approach to measure relative copy number of arthropods. While there are biases to this (see our explanations above), the assay proved very reliable and robust. We thus believe it should indeed provide a rough estimate of biomass. As biomass is very commonly discussed in insect decline (in fact the first study on insect decline entirely relies on biomass; Hallmann et al. 2017), we feel it is important go include a proxy for this as well. However, we also discuss the alternative option that a turnover of diversity is affecting the measured biomass. A pattern of abundance loss for common species has been described in other works on insect decline.

We liked the reviewer’s suggestion to use copy number information to perform abundance-informed rarefaction. We have done this now and added an additional analysis rarefying by copy number/biomass. A parallel analysis using this newly rarefied table was done for the total diversity as well as single species abundance change. Details can be found in the Methods and Results section of the manuscript. However, the result essentially remains the same. Even abundance-informed rarefaction does not lead to a pattern of loss of species richness over time (see “Statistical analysis”).

The overall results are supporting a scenario of no overall loss of species richness over time, but a loss of abundance for common species. And we indeed see the pattern of declining abundance for once-common species in our data, for example the loss of the Green Silver-Line moth, once a very common species in beech canopy (Suppl. Fig. 10). We have added details on this to the Discussion (lines 254-260).

These points may significantly impact the conclusions of this work.

Reviewer #3 (Public Review):

The aim of Weber and colleagues' study was to generate arthropod environmental DNA extracted from a unique 30-year time series of deep-frozen leaf material sampled at 24 German sites, that represent four different land use types. Using this dataset, they explore how the arthropod community has changed through time in these sites, using both conventional metabarcoding to reconstruct the OTUs present, and a new qPCR assay developed to estimate the overall arthropod diversity on the collected material. Overall their results show that while no clear changes in alpha diversity are found, the βdiversity dropped significantly over time in many sites, most notable in the beech forests. Overall I believe their data supports these findings, and thus their conclusion that diversity is becoming homogenized through time is valid.

Thank you for the positive assessment.

While overall I do not doubt the general findings, I have a number of comments. Firstly while I agree this is a very nice study on a unique dataset - other temporal datasets of insects that were used for eDNA studies do exist, and perhaps it would be relevant to put the findings into context (or even the study design) of other work that has been done on such datasets. One example that jumps to my mind is Thomsen et al. 2015 https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/1365-2656.12452 but I am sure there are others.

We have expanded the introduction and discussion on this citing this among other studies now (lines 71-72, 276-278).

From a technical point of view, the conclusions of course rely on several assumptions, including (1) that the biomass assay is effective and (2) that the reconstructed levels of OTU diversity are accurate,

With regards to biomass although it is stated in the manuscript that "Relative eDNA copy number should be a predictor for relative biomass ", this is in fact only true if one assumes a number of things, e.g. there is a similar copy number of 18s rDNA per species, similar numbers of mtDNA per cell, a similar number of cells per individual species etc. In this regard, on the positive side, it is gratifying to see that the authors perform a validation assay on 7 mock controls, and these seem to indicate the assay works well. Given how critical this is, I recommend discussing the details of this a bit more, and why the authors are convinced the assay is effective in the main text so that the reader is able to fully decide if they are in agreement. However perhaps on the negative side, I am concerned about the strategy taken to perform the qPCR may have not been ideal. Specifically, the assay is based on nested PCR, where the authors first perform a 15cycle amplification, this product is purified, then put into a subsequent qPCR. Given how both PCR is notorious for introducing amplification biases in general (especially when performed on low levels of DNA), and the fact that nested PCRs are notoriously contamination prone - this approach seems to be asking for trouble. This raises the question - why not just do the qPCR directly on the extracts (one can still dilute the plant DNA 100x prior to qPCR if needed). Further, given the qPCRs were run in triplicate I think the full data (Ct values) for this should be released (as opposed to just stating in the paper that the average values were used). In this way, the readers will be able to judge how replicable the assay was - something I think is critical given how noisy the patterns in Fig S10 seem to be.

We agree with this point, and this is why we do not want to overstate the decline in copy number. This is an additional source of data next to genetic and species diversity. We have added to our discussion of turnover as another potential driver of copy number change (lines 257-260). We have also added text addressing the robustness of the mock community assay (lines 138-141).

However, we are confident of the reliability and robustness of our qPCR assay for the detection of relative arthropod copy number. We performed several validations and optimizations before using the assay. We have added additional details to the manuscript on this (see “Detection of relative arthropod DNA copy number using quantitative PCR”, lines 548-556). We got the idea for the nested qPCR from a study (Tran et al.) showing its high accuracy and reproducibility. We show that our assay has a very high replicability using triplicates of each qPCR, which we will now include in the supplementary data on Dryad. The SD of Ct values is very low (~ 0.1 on average). NTC were run with all qPCRs to rule out contamination as an issue in the experiments. We also find a very high efficiency of the assay. At dilutions far outside the observed copy number in our actual leaf data, we still find the assay to be accurate. We found very comparable abundance changes across our highly taxonomically diverse mock communities. This also suggests that abundance changes are a more likely explanation than simple turnover for the observed drop in copy number. A biomass loss for common species is well in line with recent reports on insect decline. We can also rely on several other mock community studies (Krehenwinkel et al. 2017 & 2019) where we used read abundance of 18S and found it to be a relatively good predictor of relative biomass.

The pattern in Fig. S10 is not really noisy. It just reflects typical population fluctuations for arthropods. Most arthropod taxa undergo very pronounced temporal abundance fluctuations between years.

Next, with regards to the observation that the results reveal an overall decrease in arthropod biomass over time: The authors suggest one alternate to their theory, that the dropping DNA copy number may reflect taxonomic turnover of species with different eDNA shedding rates. Could there be another potential explanation - simply be that leaves are getting denser/larger? Can this be ruled out in some way, e.g. via data on leaf mass through time for these trees? (From this dataset or indeed any other place).

This is a very good point. However, we can rule out this hypothesis, as the ESB performs intensive biometric data analysis. The average leaf weight and water content have not significantly changed in our sites. We have addressed this in the Methods section (see ”Tree samples of the German Environmental Specimen Bank – Standardized time series samples stored at ultra-low temperatures”, lines 308-311).

With regards to estimates of OTU/zOTU diversity. The authors state in the manuscript that zOTUs represent individual haplotypes, thus genetic variation within species. This is only true if they do not represent PCR and/or sequencing errors. Perhaps therefore they would be able to elaborate (for the non-computational/eDNA specialist reader) on why their sequence processing methods rule out this possibility? One very good bit of evidence would be that identical haplotypes for the individual species are found in the replicate PCRs. Or even between different extractions at single locations/timepoints.

We have repeated the analysis of genetic variation with much more stringent filtering criteria (see “Statistical analysis”, lines 611-615). Among other filtering steps, this also includes the use of only those zOTUs that occur in both technical replicates, as suggested by the reviewer. Another reason to make us believe we are dealing with true haplotypic variation here is that haplotypes show geographic variation. E.g., some haplotypes are more abundant in some sites than in others. NUMTS would consistently show a simple correlation in their abundance with the most abundant true haplotype.

With regards to the bigger picture, one thing I found very interesting from a technical point of view is that the authors explored how modifying the mass of plant material used in the extraction affects the overall results, and basically find that using more than 200mg provides no real advantage. In this regard, I draw the authors and readers attention to an excellent paper by Mata et al. (https://onlinelibrary.wiley.com/doi/full/10.1111/mec.14779) - where these authors compare the effect of increasing the amount of bat faeces used in a bat diet metabarcoding study, on the OTUs generated. Essentially Mata and colleagues report that as the amount of faeces increases, the rare taxa (e.g. those found at a low level in a single faeces) get lost - they are simply diluted out by the common taxa (e.g those in all faeces). In contrast, increasing biological replicates (in their case more individual faecal samples) increased diversity. I think these results are relevant in the context of the experiment described in this new manuscript, as they seem to show similar results - there is no benefit of considerably increasing the amount of leaf tissue used. And if so, this seems to point to a general principal of relevance to the design of metabarcoding studies, thus of likely wide interest.

Thank you for this interesting study, which we were not aware of before. The cryomilling is an extremely efficient approach to equally disperse even traces of chemicals in a sample. This has been established for trace chemicals early during the operation of the ESB, but also seems to hold true for eDNA in the samples. We have recently done more replication experiments from different ESB samples (different terrestrial and marine samples for different taxonomic groups) and find that replication of extraction does not provide much more benefit than replication of PCR. Even after 2 replicates, diversity approaches saturation. This can be seen in the plot below, which shows recovered eDNA diversity for different ESB samples and different taxonomic groups from 1-4 replicates. A single extract of a small volume contains DNA from nearly all taxa in the community. Rare taxa can be enriched with more PCR replicates.

Author Response

Reviewer #1 (Public Review):

Previous studies have linked several lifestyle-related factors, such as body mass index and smoking, alcohol use with accelerated biological aging measured using epigenetic clocks, however, most of them focused on single lifestyle factors based on cross-sectional data from older adults. The current study has a couple of major strengths: it has a decent sample size, lifestyle was measured longitudinally during puberty and adolescence, it looked at the effect of multiple lifestyle measures collectively, it looked at multiple epigenetic clocks, and due to the data from twins, it could examine the contribution of genetic and environmental influences to the outcomes. I have a couple of comments that are mainly aimed at improving the clarity of the methods (e.g. how was multiple testing correction done, how did the association model account for the clustering of twin data, how many samples were measured on 450k vs EPIC and were raw or pre-QC'd data supplied to the online epigenetic age calculator), and interpretation of findings (why were 2 measures of Dunedin PACE of aging used, how much are results driven by BMI versus the other lifestyle factors, and the discussion on shared genetic influences should be more nuanced; it includes both pleiotropic effects and causal effects among lifestyle and biological ageing).

Thank you for the encouraging comments and important suggestions.

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.

The twin dataset is large and longitudinal, making this an excellent resource to assess the research questions.

The analyses are advanced including LCA capitalizing on the strength of these data.

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.

Thank you for the encouraging comments and important notes.

We have now acknowledged that the standard deviations of epigenetic age estimates were high (lines 409-418). Due to the narrow age range of this study, the correlations between chronological age and epigenetic age estimates were weak. We aimed to overcome these weaknesses and calculated the epigenetic age estimates using recently developed principal component (PC)-based clocks, which are shown to improve the reliability and validity of epigenetic clocks (Higgins-Chen et al., 2022). In our data, the standard deviations of epigenetic age estimates were similar or even higher compared with those obtained with the original clocks, but the correlations between epigenetic age acceleration measures assessed with different clocks were consistently higher when PC-based epigenetic clocks were used. Importantly, the observed associations with the adolescent lifestyle behavior patterns did not substantially change.

Moreover, we have now more carefully reported and interpreted the results obtained using different epigenetic aging measures and acknowledged their heterogeneity (lines 459-467).

Figure 4 presents the genetic and environmental influences on biological aging shared with adolescent lifestyle and biological aging. There are also unique genetic and environmental influences on biological aging not shown in the figure. Therefore, the unexplained variation in biological aging was not that large. Most of the total variation in biological aging was explained by the genetic factors unique to biological aging. We have now clarified the description of the estimation of genetic and environmental influences (lines 283-300) and the presentation of the results (lines 437-449).

References:

Higgins-Chen, A. T., Thrush, K. L., Wang, Y., Minteer, C. J., Kuo, P.-L., Wang, M., Niimi, P., Sturm, G., Lin, J., Moore, A. Z., Bandinelli, S., Vinkers, C. H., Vermetten, E., Rutten, B. P. F., Geuze, E., Okhuijsen-Pfeifer, C., van der Horst, M. Z., Schreiter, S., Gutwinski, S., … Levine, M. E. (2022). A computational solution for bolstering reliability of epigenetic clocks: implications for clinical trials and longitudinal tracking. Nature Aging, 2(7), 644–661. https://doi.org/10.1038/s43587-022-00248-2

Author Response

Reviewer #1 (Public Review):

This report describes evidence that the main driving force for stimulation of glycolysis in cultured DGC neurons by electrical activity comes from influx of Na+ including Na+ exchanging into the cell for Ca2+. The findings are presented very clearly and the authors' interpretations seem reasonable. This is important and impactful because it identifies the major energy demand in excited neurons that stimulates glycolysis to supply more ATP.

Strengths are the highly rigorous use of fluorescent probes to directly monitor the concentrations of NADH/NAD+, Ca2+ and Na+. The strategies directly test the roles of Na+ and Ca2+.

A weakness is an ambiguity about the effects of ouabain to inhibit the Na+/K+ ATPase directly and the absence of biochemical controls to validate the interpretation of the ouabain experiment.

We appreciate the reviewer's comments about the work. While we can not rule out non-specific effects of ouabain at the concentrations needed to block Na+/K+ ATPase in these experiments, we do think that we can rely on the prior biochemical work characterizing the multiple components of ouabain binding in fresh mouse brain tissue, which is a close match to the acute mouse brain slice tissue used here.

Reviewer #2 (Public Review):

This study seeks to determine how neuronal glycolysis is coupled to electrical activity. Previous studies had found that glycolytic enzymes cluster within nerve terminals (in C. elegans) during activity. Furthermore, the glucose transporter GLUT4 is recruited to synaptic surface during activity. The authors previously showed that Ca2+ does not stimulate glycolysis in active neurons. Here, the authors show that the cytosolic Na+, not Ca2+, and the activity of the Na+/K+ pump drive glycolysis. However, it is important to note that in this study, glycolysis was examined in the soma, not nerve terminals, where some of the previous studies were conducted. A few other caveats in the interpretation of the findings are listed below:

1) The NADH/NAD+ ratio is used throughout as the only measurement reflecting glycolytic flux.

In this and previous work, we have validated that increased cytosolic NADH production (whose major sources are related to glycolysis), rather than altered NADH reoxidation, produces the changes in NADH/NAD+ ratio.

2) It has been hypothesized that the close association of glycolytic enzymes with ion transporters (such as the Na+/K+ pump) is meant to provide localized ATP to power these pumps. How does bulk glycolysis (monitored with NADH/NAD+ ratio) relate to localized/compartmentalized glycolysis?

Even if glycolysis is indeed localized to the plasma membrane (an interesting and difficult-to-address hypothesis), we believe that because the mitochondrial shuttles are the main pathway for NADH re-oxidation, and most mitochondria are not localized to the plasma membrane, changes in glycolytic NADH production are likely to be reflected in changes of the bulk cytosolic NADH/NAD+.

3) Related to point 2, most of the Peredox measurements in the paper have been made at baseline, in the absence of electrical activity. Therefore, it is not clear how the findings relate to activity-driven glycolysis.

The ion exchange experiments and even the faster Ca2+ puff experiments can mimic but indeed cannot match the speed of activity-driven changes in ion concentrations. Unfortunately, it is impossible to induce normal electrical activity in neurons in the absence of extracellular Na+. We believe that the complete inability of Ca2+ elevation alone (without Na+-Ca2+ exchange) to stimulate glycolysis, combined with the substantial Ca2+ contribution to activity-driven glycolysis, makes a good argument that Ca2+ entering during activity is likely to stimulate glycolysis via Na+ entry and the Na+/K+ ATPase.

4) The finding that inhibition of SERCA during stimulation actually elevates cytosolic NADH level argues against Na+ being the only ion that regulates glycolysis.

The ability of SERCA inhibition to produce a small increase in activity-driven glycolysis is consistent with the simple argument that reduced SERCA-driven uptake of Ca2+ into ER results in additional Ca+ removal via Na+/Ca2+ exchange (which can then affect glycolysis via Na+ levels).

5) The finding that "SBFI ΔF/F transients were longer in duration than the RCaMP LT transient" does not necessarily mean that Na+ elevation lasts longer than Ca2+ in the cell. This could be an artefact of the SBFI on/off rate relative to RCaMP. In fact, prolonged elevation of cytosolic Na+ would make neurons refractive to depolarization in AP trains.

The rates of Na+ binding and unbinding to SBFI are likely to occur on the microsecond timescale (based on the known properties of crown ether molecules), much faster than the observed transient duration of approximately one minute. Prolonged elevation of cytosolic Na+ alone (to the levels seen here) should not cause neurons to be refractory to firing; refractoriness typically occurs in the setting of prolonged depolarization and consequent inactivation of NaV channels.

Reviewer #3 (Public Review):

Meyer et al have studied the mechanisms of glycolysis activation in the hippocampus during neuronal activity. The study is logically laid out, uses sophisticated fluorescence lifetime imaging technology and smart experimental designs. The support for intracellular [Na+] vs [Ca2+] rise driving glycolysis is strong. The evidence for the direct involvement of the Na+/K+ pump is based only on pharmacology using ouabain but the Na+/K+ pump is admittedly not an easy subject for specific perturbations. I still think that the Authors should strengthen the support for the pathway.

We are happy that the reviewer feels that the evidence for Na+ rather than Ca2+ as the effector of glycolysis is strong. The tools for investigating the role of the Na+/K+ pump (NKA) are indeed limited to pharmacology, because (as the reviewer says) there are not many other options. The requirement for Na+ elevation (which stimulates NKA activity) to trigger glycolysis and the ability of ouabain, a specific NKA inhibitor, to prevent this seem like strong implication of NKA in the mechanism of glycolysis activation. Genetic manipulation of the NKA may be unable to change the level of pump activity, because of compensation by altered expression of other subunits (PMID 17234593); it also is unclear how any chronic manipulation would shed light on the role of NKA in triggering glycolysis. But perhaps future studies of knock-in mice in which the α1 isoform of NKA has made more sensitive to ouabain (PMIDs 15485817; 34129092) might allow the identification of the NKA as the target of ouabain in this situation to be made even more secure.

Also, there is a long list of publications on the connection between the Na+/K+ pump and glycolysis. It might be useful to highlight the role of the NCX- Na+/K+ pump coupling in the activation of glycolysis in the title.

Author Response

Reviewer #1 (Public Review):

Dotov et al. took joint drumming as a model of human collective dynamics. They tested interpersonal synchronization across progressively larger groups composed of 1, 2, 4 and 8 individuals. They conducted several analyses, generally showing that the stability of group coordination increases with group numerosity. They also propose a model that nicely mirrors some of the results.

The manuscript is very clear and very well written. The introduction covers a lot of relevant literature, including animal models that are very relevant in this field but often ignored by human studies. The methods cover a wide range of distinct analyses, including modelling, giving a comprehensive overview of the data. There are a few small technical differences across the experiments conducted with small vs. large groups, but I think this is to some extent unavoidable (yet, future studies might attempt to improve this). Furthermore, the currently adopted model accounts well for behaviors where all individuals produce a similar output and therefore are "equally important". However, it might be interesting to test to what extent this can be generalized to situations where each individual produces a distinct sound (as in a small orchestra) and therefore might selectively adapt to (more clearly) distinguishable individuals.

We agree that this is important. We discuss this in a new section (4.1) at the end of the discussion. We suggest that heterogeneity makes it possible for other modes of organization to compete with the attractive tendency towards the global average. We also point out that factors such as individual skill, task difficulty, delays, and selective attention enable such heterogeneity in the ensemble.

Similarly, it would be interesting to test to what extent the current results (and model) can be generalized to interactions that more strongly rely on predictive behavior (as there is not much to predict here given that all participants have to drum at a stable, non-changing tempo).

We can only speculate that the present results are less relevant to interactions that rely strongly on predicitive behavior, as behaviour in our simple task could be modeled well by our hybrid single oscillator Kuromoto model. We inserted the idea that the presence of a group rhythm can diminish the demands for individuals to predict each other’s notes, the end of paragraph 1, page 27.

An important implication of this study is that some well-known behaviors typically studied in dyadic interaction might be less prominent when group numerosity increases. I am specifically referring to "speeding up" (also termed "joint rushing") and "tap-by-tap error correction" (Wolf et al., 2019 and Konvalinka et al., 2010, also cited in the manuscript, are two recent examples). I am not sure whether this depends on how the data is analyzed (e.g. averaging the behavior of multiple drummers), yet this might be an important take-home message.

Thank you for the suggestion. We edited to emphasize that the relevant part of the analysis of the drumming data was performed at the individual level and using the same methods as typically done in dyadic tapping (first sentences of Section 2.7.2). Speeding up was the only variable where we used group-averages. For consistency, and to avoid confusion, in the present version we re-did the stats (the changed statistical parameters are highlighted) and figures using the individual data points and we did not observe major changes.

I am confident that this study will have a significant impact on the field, bringing more researchers close to the study of large groups, and generally bridging the gap between human and animal studies of collective behavior.

Reviewer #2 (Public Review):

In this manuscript Dotov et al. study how individuals in a group adjust their rhythms and maintain synchrony while drumming. The authors recognize correctly that most investigation of rhythm interaction examines pairs (dyads) rather than larger groups despite the ubiquity of group situations and interactions in human as well as non-human animals. Their study is both empirical, using human drummers, and modeling, evaluating how well variations of the Kuramoto coupled-oscillator describe timing of grouped drummers. Based on temporal analyses of drumming in groups of different sizes, it is concluded that this coupled oscillator model provides a 'good fit' to the data and that each individual in a group responds to the collective stimulus generated by all neighbors, the 'mean field'.

I have concerns about 1) the overall analysis and testing in the study and about 2) specific aspects of the model and how it relates to human cognition. Because the study is largely empirical, it would be most critical for the authors to propose two - or more - alternative hypotheses for achieving and maintaining synchrony in a group. Ideally, these alternatives would have different predictions, which could be tested by appropriate analyses of drummer timing. For example, in non-human animals, where the problem of rhythm interaction in groups has been examined more thoroughly than in humans, many acoustic species organize their timing by attending largely to a few nearby neighbors and ignoring the rest. Such 'selective attention' is known to occur in species where dyads (and triads) keep time with a Kuramoto oscillator, but the overall timing of the group does not arise from individual responses to the mean field. Can this alternative be evaluated in the drumming data ? Would this alternative fit the drumming data as well as, or better than , the mean field, 'wisdom of the crowd' model ?

These are very important points. The present paper is restricted to a simple task where participants are instructed to synchronize with each other. However, we now more explicitly acknowledge the limitations of our study and include a new section, “Beyond the group average” at the end of the Discussion that is dedicated to this issue and discussed other organizing tendencies that are particularly relevant in larger and more diverse ensembles. In the context of the present task, the relative difference between local and global interactions was likely negligible because of the small differences in timing, from 4 to 16 ms, between the closest and most distant pairs.

It will be interesting in future studies to introduce acoustic heterogeneity by varying the timbre of the instruments, for example. In the present study, the instruments had the same timbre with narrowly varying fundamental frequencies (117-129 Hz in the duets/quartets and 249-284 Hz in the octets), a situation that encourages integration of all the acoustic information. We do point out that the present approach needs to be expanded to be able to account for competitive pressure and selective attention.

The well-known Vicsek model (discussed briefly in paragraph 2, page 15), related to the Kuramoto under certain assumptions, can account for a variety of dynamic behaviors in flocking animals. The ability for selective attention in the form of a heterogeneous coupling matrix, combined with the existence of competitive pressure in the form of negative coupling terms can result in spontaneous formation of clusters and spatiotemporal patterns of movement. This is consistent with prior research in chorusing animals (insects and anurans). Large musical ensembles also involve groupings of instruments such as separate sections that change their relative loudness across time. Typically these are not spontaneous but composed and conducted, yet they may satisfy the same constraints.

We also pointed out that we see these as complementary organizing principles. Even in the Vicsek model, there is a notion of a ‘local order parameter’ whereby individuals are coupled to a group average within a narrow interaction radius. The relative importance of other organization tendencies depends on the layout of the acoustic environment and the competitive and collaborative aspects of the task. Hence, parameters such as delay and individual heterogeneity could act as symmetry breaking terms that enable different stabilities from the basic global group synchrony.

A second concern arises from relying on a hybrid, continuous - pulsed version of the Kuramoto coupled oscillator. If the human drummers in the test could only hear but not see their neighbors, this hybrid model would seem appropriate: Each drummer only receives sensory input at the exact moment when a neighbor's drumstick strikes the drum. But the drummers see as well as hear their neighbors, and they may be receiving a considerable amount of information on their neighbors' rhythms throughout the drum cycle. Can this potential problem be addressed? In general, more attention should be paid to the cognitive aspects of the experiment: What exactly do the individual drummers perceive, and how might they perceive the 'mean field' ?

This is all very relevant. We instructed participants to focus on X’s in the centers of their drums and not look at their peers (edited to mention that in at the end of Section 2.4, page 9). Additionally, the pattern of results for tempo change, cross-correlations, and variability in the dyadic condition was consistent with previous studies that involved purely auditory tapping tasks (emphasized in the begging of paragraph 2, page 26). The best way to address this limitation would be to repeat the study and block the visual contact among participants, as well as include a condition emphasizing visual contact.

It is beyond the scope of the present paper to make model-based predictions of effects of coupling and information availability, but this should be done in future work. For the present paper, we now include a simulation involving continuous coupling (end of section 2.9.2, page 16) and Supplementary Figure 8A) which fails to reproduce the results for variability, results that are well captured by the hybrid continuous-pulsed model we developed, see the Supplementary Materials.

Reviewer #3 (Public Review):

The contribution provides approaches to understanding group behaviour using drumming as a case of collective dynamics. The experimental design is interestingly complemented with the novel application of several methods established in different disciplines. The key strengths of the contribution seem to be concentrated in 1) the combination of theoretical and methodological elements brought from the application of methods from neurosciences and psychology and 2) the methodological diversity and creative debate brought to the study of musical performance, including here the object of study, which looks at group drumming as a cultural trait in many societies.

Even though the experimental design and object of study do not represent an original approach, the proposed procedures and the analytical approaches shed light on elements poorly addressed in music studies. The performers' relationships, feedbacks, differences between solo and ensemble performance and interpersonal organization convey novel ideas to the field and most probably new insights to the methodological part.

It must be mentioned that the authors accepted the challenge of leaving the nauseatic no-frills dyadic tests and tapping experiments in the direction of more culturally comprehensive (and complex) setups. This represents a very important strength of the paper and greatly improves the communication with performers and music studies, which have been affected by the poor impact of predictable non-musical experimental tasks (that can easily generate statistical significant measurements). More specifically, the originality of the experiment-analysis approach provided a novel framework to observe how the axis from individual to collective unfolds in interaction patterns. In special, the emergence of mutual prediction in large groups is quite interesting, although similar results might be found elsewhere.

Thank you for these comments.

On another side, important issues regarding the literature review, experimental design and assumptions should be addressed.

I miss an important part of the literature that reports similar experiments under the thematic framework of musical expressivity/expression, groove, microtiming and timing studies. From the participatory discrepancies proposed in 1980's Keil (1987) to the work of Benadon et al (2018), Guy Madison, colleagues and others, this literature presents formidable studies that could help understand how timing and interactions are structured and conceptualized in the music studies and by musicians and experts. (I declare that I have no recent collaborations with the authors I mentioned throughout the text and that I don't feel comfortable suggesting my own contributions to the field). This is important because there are important ontological concerns in applying methods from sciences to cultural performances.

Thank you for the suggestions. We included a brief discussion in the newly added “Beyond the group average” section at the end of the Discussion, specifically the first paragraph, pages 27-8. We think that expressive timing naturally fits in continuation with the other reviewers’ concerns about how much the idea of the group average generalizes to real musical situations. By design and instruction, we stripped individual expression from the present task. Specific cultural contexts and performance styles may want to escape or at least expressively tackle this constraint of our task, and we believe that now that we have established the mean field as one factor affecting group behaviour, further studies can take on the challenge of developing models that make predictions in more complex situations closer to real musical interactions – and testing those models empirically.

One ontological issue that different cultural phenomena differ from, for example, animal behaviour. For example, the authors consider timing and synchrony in a way that does not comply with cultural concepts: p.4 "Here we consider a musical task in which timing consistency and synchrony is crucial". A large part of the literature mentioned above and evidence found in ethnographic literature indicate that the ability to modulate timing and synchrony-asynchrony elements are part of explicit cultural processes of meaning formation (see, for example, Lucas, Glaura and Clayton, Martin and Leante, Laura (2011) 'Inter-group entrainment in Afro-Brazilian Congado ritual.', Empirical musicology review., 6 (2). pp. 75-102.). Without these idiosyncrasies, what you listen to can't be considered a musical task in context and lacks basic expressivity elements that represent musical meaning on different levels (see, for example, the Swanwick's work about layers/levels of musical discourse formation).

Indeed, this is an important issue. We often use cultural phenomena merely as a motivation but do not dive in the relevant details. Here, in addition to the previous discussion, we now reiterate that the tendency towards the group average is one organizing tendency but there are additional ones, enabled by individual heterogeneity and context. For example, marching bands and chanting crowds probably impose different constraints than individual artistic expression by skillful musicians.

Such plain ideas about the ontology of musical activities (e.g. that musical practice is oriented by precision or synchrony) generate superficial constructs such as precision priority, dance synchrony, imaginary internal oscillators, strict predictive motor planning that are not present in cultural reports, excepting some cultures of classical European music based on notation and shaped by industrial models. The lack of proper cultural framing of the drumming task might also have induced the authors to instruct the participants to minimize "temporal variability" (musical timing) and maintain the rate of the stimulus (musical tempo), even though these limiting tasks mostly take part of musical training in some societies (examples of social drumming in non-western societies barely represent isochronous tempo or timing in any linguistic or conceptual way). The authors should examine how this instruction impacts the validity of results that describe the variability since it was affected by imposed conditions and might have limited the observed behaviour. The reporting of the results in the graphs must also allow the diagnosis of the effect of timing in such small time frame windows of action.

We agree totally. We made changes and tried to be more specific about the cultural framing, delineating contexts where the present ideas are more relevant and where they are less relevant, or at least incomplete (the bottom of page 3, and pages 27-8).

Author Response

Reviewer #1 (Public Review):

Mitotic spindles are macromolecular machines that accurately segregate duplicate chromosomes between two daughter cells during cell division. To perform this task, spindles exert forces that are orchestrated in space and time. On the other hand, non-functioning spindles can generate chromosome segregation errors, which are present in cancers, miscarriages, and Down syndrome. Therefore, understanding spindle mechanics is a big biological challenge. In this elegant study, the authors explore the mechanical properties of the mitotic spindle. They combine a variety of experimental biophysical approaches, including microneedle manipulation and quantitative imaging, with theoretical modeling. By systematically exploring the shape of kinetochore fibers that are not manipulated, they find the force and moments that exist in the native spindles. Analyzing previously published data obtained by microneedle manipulations, where kinetochore fibers were mechanically perturbed, the authors observe a dramatic change in the shape of the kinetochore fibers. Comparing this observation and theoretical predictions, they discover a lateral anchorage near the chromosome. Taken together, this paper nicely demonstrates existence of lateral anchorage near chromosomes, offering exciting ideas about the balance of forces of the entire mitotic spindle.

We appreciate the reviewer’s enthusiasm about the work and their thoughtful questions and suggestions to improve the manuscript.

Major points:

(1) In order to describe the shape of unmanipulated kinetochore fibers, the authors use a simple physical model in which they describe these fibers as a single elastic rod. They find that the observed shape is a consequence of compressive forces, or a combination of bending moments and perpendicular forces. However, it is well known that kinetochores are under the tension. For this reason, the plus end of kinetochore fibers should be under tension rather than under compression. In order to describe forces that shape unmanipulated kinetochore fibers, the authors should revise the model by setting the tensile force at the plus end of the kinetochore fiber.

We thank the reviewer for their comment on this important point .

(2) The authors compare the shapes of inner and outer kinetochore fibers. By using the model, they find that the forces and moments are similar for both, the inner and outer kinetochore fibers, whereas the difference arises because these fibers have a different length. In classical beam theory, we distinguish between buckling (caused by a compressive force) and bending (caused by a bending moment). In the case of buckling, which is caused by a same critical force, different curvatures can be obtained, whereas in the case of bending the curvature is proportional to the bending moment. Based on the data presented by the authors, it seems that their model operates in the buckling regime. It would be important to elaborate on this more systematically. Also, one should warn the reader that in the case of bending, the inner and outer kinetochore fibers will be characterized by different bending moments.

We thank the reviewer for raising this nuanced point on the shape generation mechanisms in inner and outer k-fibers. We believe that the mechanisms that the reviewer suggested are valid ways to generate varying k-fiber deflections in the scenario where the k-fiber end-to-end length is held fixed. However, we argue that the natural variability in the lengths of inner vs. outer k-fibers is alone sufficient to give rise to diverse k-fiber shapes without requiring the end-forces to change.

We added a new Appendix section 1.4 (pages S4-S5 in the revised appendix) in our revised submission where we provide the details of our argument. We demonstrate analytically that when only a moment at the pole is present and held at a fixed value, then the normalized maximum deflection scales linearly with the k-fiber’s end-to-end length (Appendix 1 – figure 3a,b). And in the case where both a moment at the pole and an axial force are present and held at fixed values, the dependence on k-fiber length is stronger (faster than linear), thereby allowing for a wide range of k-fiber deflections created with identical end-forces (Appendix 1 – figure 3c,d).

Reviewer #2 (Public Review):

Suresh and co-workers apply classical beam bending theory to analyze shapes of the microtubule bundles that push and pull on mitotic chromosomes and drive chromosome separation in dividing cells. The bundles attach at one end to chromosomes via specialized protein assemblies called kinetochores, and at the other end they are associated with spindle poles. The shapes of these k-fiber bundles are analyzed in unperturbed control cells and in cells where the bundles have been forcibly deformed using microneedles. From their analysis, the authors infer the extent and nature of mechanical anchorage at each end of the bundles, finding that anchorage is more extensive and more restrictive at the kinetochore-attached ends compared to the pole-proximal ends. Anchorage at the pole-proximal ends is apparently limited to the bundle tips, allowing some swiveling of the bundles around the poles. In contrast, the kinetochore-attached ends appear to have "lateral anchorage", i.e. force-bearing connections to the sides of the bundles, that extend several micrometers away from the kinetochores. This lateral anchorage resists swiveling of the bundles around their kinetochore-attached ends.

A major strength of this study is its high degree of novelty. The microneedle data on which the analyses are based have been published previously, but are entirely unique - based on classic, groundbreaking experiments performed nearly half a century ago on cells from grasshoppers and mantids, and now being done only in the Dumont lab, in mammalian cells for the first time, and with the benefit of modern fluorescence and molecular perturbation techniques. Such a unique and interesting dataset certainly deserves careful analytical scrutiny, which is the focus of this new paper.

The application here of classical beam theory to analyze k-fiber shapes is also clever, apparently well done, and well described. The unique approach provides a direct way to assess the extent to which k-fiber bundles are mechanically linked to surrounding material, including to non-k-fiber microtubules and potentially to neighboring k-fibers. The main conclusion that lateral anchorage of the k-fibers in the local vicinity (within a few micrometers) of kinetochores is needed to explain the shapes that the k-fibers adopt during manipulations seems well justified by the data and analyses - particularly by the negative curvatures measured near the kinetochore-attached ends, and the tendency for the orientations of the kinetochore-proximal portions to be maintained even 1 to 3 micrometers away from the kinetochore-attached ends. The assumptions of the analysis also seem mostly reasonable and are clearly explained. Under these assumptions, the analysis shows convincingly that forces and moments applied only at kinetochore-attached ends would be insufficient to explain the observed shapes.

We appreciate the reviewer’s enthusiasm about the work and their thoughtful questions and suggestions to improve the manuscript.

Author Response

Reviewer #1 (Public Review):

This paper primarily assessed the host/phage interactions for bacteria in the order of Cornyebacteriales to identify novel host factors necessary for phage infection, in regards to genes responsible for bacterial envelope assembly. Bacteria in this order, such as Mycobacterium tuberculosis and Corynebacterium diphtheriae have unique, complex envelopes composed of peptidoglycan, arabinogalactan, and mycolic acids. This barrier is a potent protector against the therapeutic effects of antibiotics. Phages can be used to discover novel aspects of this bacterial envelope assembly because they engage with cell surface receptors. To uncover new factors, the researchers challenged a high-density transposon library of Corynebacterium glutamicum (called Cglu in the paper) with phages, Cog, and CL31. Results by transposon sequencing identified loci that were interrupted, leading to phage resistance. This study implicated the importance of Cglu genes, ppgS, cgp_0658, cgp_0391, and cgp_0393. They also identified a new gene called cgp_0396 necessary for arabinogalactan modification and recognized a conserved host factor called Ahfa (Cpg_0475) that plays a crucial role in Cglu mycolic acid synthesis. Ultimately, this work implicated the importance of mycomembrane porins, arabinogalactan, and mycolic acid synthesis pathways in the assembly of the Cornyebacteriales envelope.

Strengths of the research:

• Language choice: A major strength of the paper is that this could easily be given to an undergraduate student with introductory knowledge of biology and they would still be able to get the gist of this paper. The language is written in a clear, concise fashion with explanations of terms not everyone would immediately know unless they worked in the field specifically.

• These figures are generally explained in a direct manner, clearly stating the major conclusions the reader should get after carefully analyzing the presented data

We thank the reviewer for the enthusiasm for our work and our description of it.

How the research could be strengthened:

• It could be worthwhile to describe some of your results mathematically. For example, the differences you see in your phage infections relating to the differences in logs, etc. Bar graphs also should be described in mathematical terms, when "something is lower compared to the WT," how much is lower, etc?

To keep the text streamlined, we refrained from adding descriptions of the results mathematically in the text. The reader can refer to the figures to get the magnitudes of any changes observed.

• There were no p values relating to the statistical significance of any of the data presented, which should be changed for the final manuscript implicating the importance of this work.

We added the p-values as requested.

• Figure 8 was not entirely supported by the data, especially Figure 8A which either could be improved with better images that support the author's claims, etc.

We do not understand why the reviewer believes that Figure 8A does not support our conclusions. The mutant cells do not label with the 6-TMR-Tre dye whereas the WT control does. The dye labels mycolic acid such that our conclusion that AhfA is involved in mycolic acid synthesis is valid. In any case, we have included an additional supplementary source data file of the uncropped image of the 6-TMR-Tre treated cells to show a larger number of mutant cells that fail to stain, further supporting our conclusion.

Reviewer #2 (Public Review):

In this manuscript, McKitterick and Bernhardt use genetic approaches to investigate genes in Corynebacterium glutamicum that are required for efficient phage infection. They make use of a high-density transposon library that was generated in the Bernhardt lab recently. They challenged the library with two phages, CL31 and Cog. Importantly, they elegantly adapted the phages to the laboratory strain MB001 before. The MB001 strain is ideal for genetic experiments since all prophage elements were removed in this strain. The evolved phages are likely a very useful tool for further investigations aiming to understand host/virus interactions in this model. The phage-infected libraries were plated and the collected colonies were sequenced. Genes involved in efficient phage infection had multiple transposon insertions. Using this method the authors identified specific genes required for infection with Cog and CL31. The Cog phage needs apparently the porin proteins in the mycolic acid membrane for efficient infection and the authors speculate that the porins may act as auxiliary receptors for phage adsorption. Furthermore, genes involved in putative arabinogalactan modification were found to be important. Mutants in these genes did not abolish phage adsorption and thus play a role in viral genome injection. For phage CL31 the authors show that in particular genes involved in mycolic acid synthesis are essential. The genes identified include one coding for a protein involved in protein mycoloylation. A candidate for such a lipidation is the porin protein complex PorAH. The trehalose-6-phosphate synthase OtsA was also identified as important for phage infection. Also strictly required for the establishment of the myco membrane, otsA deletions are viable in C. glutamicum. As part of their analysis, they also identified an unknown factor in mycolic acid synthesis in C. glutamicum. Analysis of a spontaneous resistant mutant to CL31 revealed a mutation in cg_0475 (renamed ahfA). Deletion of ahfA drastically reduced mycolic acid production. This was proven by thin layer chromatography and fluorescent staining. Interestingly, deletion of ahfA also results in a cell morphology defect, indicating the importance of a correct mycolic acid layer for cell shape.

In summary, the authors provide an excellent paper that is clearly written and experiments are conducted nicely.

We thank the reviewer for their kind words and enthusiasm for the work.

Reviewer #3 (Public Review):

In their manuscript, McKitterick and Bernhardt perform a screen to determine host factors, such as receptors, which are important for bacterial viruses (phages) to infect Corynebacterium glutamicum., an organism that shares the unique membrane of mycobacteria (mycomembrane), with M. tuberculosis. To do so, they challenge a previously described Tn-seq library with a high MOI of 2 phages - Cgl and Cog. The surviving strains are those in which genes important for phage infection (such as receptors) are disrupted. The authors' screen is successful, and the authors identify and validate several factors important for the infection of each phage, providing the first such screen in Corynebacterium. Moreover, the authors perform a suppressor screen to identify additional factors and experimentally follow up several genes of interest. Finally, the authors use the newly determined host specificity of te phages to implicate new genes in mycolic acid synthesis. As a whole, this is a strong work that paves the way to a deeper understanding of Corynebacterial and (by extension) Mycobacterial phages and should be of broad interest.

Below, we suggest additional analyses, context, and elaboration that will help the ms. elaboration to fully realize its impact.

Major points:

1. Although the authors' experimental design is fundamentally sound, I am worried about the possibility of "jackpotting" in shaping their results, particularly in the uninfected control experiment. If the authors' Tn-seq library is ~200,000 strains, and they don't plate at least 10-100x times that many colonies then any given strain (regardless of its phenotype) may or may not be represented in the output of the experiment, causing false phenotypes to be ascribed to genes based on chance. This is particularly a problem for the uninfected control, where the authors choose to dilute the culture 1000fold to mimic the number of colonies that survive infection. They may be better served by plating the whole culture on the plates, to ensure adequate representation of the library. Part of the reason for this concern is that an overwhelming majority of statistically significant hits (something like 80-90%) appear to confer susceptibility rather than resistance (source data Fig 2) - something the authors' experimental design should not be able to measure. The lack of accurate representation of distributions of strains in the starting culture also calls into question the quantitative differences they present in the results

We thank the reviewer for their thorough analysis of our experimental design. The Tn-Seq experiments were repeated with the uninfected controls plated at a density that maintains the representation of the original library. The overall results are largely unchanged because we maintain our focus on hits that become greatly enriched following phage infection not those that become depleted. The vast majority of these hits were validated for their involvement by constructing mutant strains, indicating the robustness of the current and previous analyses. With respect to the depletion of insertion mutants, we mentioned in the original submission that they are unlikely to be biologically meaningful.

a. L138. Where the authors describe their initial experimental design it would be helpful to add more details. What is the size of the Tn library? What is the coverage in their experiment? Approximately how many colonies are recovered on the plates after phage infection and in the uninfected control?

This information has been added (Fig. 2 table supplement 1).

b. it is important to know how the number of colonies on the plates compares to the number of reads in the experiment. In the analysis of most HT screens, one implicitly assumes that each read corresponds to 1 cell, hence each read can be treated as statistically independent. This assumption is critical to the statistical methods used to analyze this data. By scraping a plate of colonies (which may be required for efficient phage infection), the authors potentially violate this assumption (since the number of cells → number of colonies, which are the actual statistically independent entities in the experiment). Does this assumption hold (or approximately hold) for the screen? If not, a different statistical method should be used to determine p-values.

We respectfully disagree with the reviewer on this point. In our view, a slurry of colonies from a plate is no different than a culture. Both contain a mixture of cells containing an array of different transposon mutants each represented multiple times in the population due to replication of the original mutant. We do not think there is any meaningful difference to the analysis whether this replication occurs in liquid or on a plate. In both cases, a read corresponds to a single cell/molecule of purified genomic DNA from the population.

1. The authors' Tn-seq methodology is different from previously published HT-phage screens (e.g. Mutalik et al., 2020 and Rousset et al., 2018). Based on my knowledge of classical phage biology, I agree that plating the infected cells has advantages. However, the rationale will not be clear for most people performing such experiments. Please explain the rationale for the experimental protocol.

Although the authors in the Mutalik et al paper did do competition experiments in liquid over several infection cycles, they also made use of a solid platebased assay in which they adsorbed their phages to the library cells for 15 minutes before plating. These plates were incubated overnight and resistant colonies were scraped, pelleted, and DNA prepped in a similar manner to the approach we took.

We prefer plating over liquid growth because colony formation is an easy way to ensure that the mutant population has undergone numerous rounds of doubling under a given condition before the analysis is performed.

a. Why did the authors plate the cultures after initial phage absorption instead of remaining in liquid?

We were concerned that some potential receptor-related mutants would be less fit and would therefore be lost in a competition experiment. As such, plating after phage adsorption would decrease the competition between phage survivors. Furthermore, we thought that plating would additionally ensure that the bacteria that are sequenced are true survivors and not just reflect remnant DNA in the culture.

b. How reproducible are the authors' Tn-seq results? The SRA ascension shows multiple replicates but this is not described in the manuscript nor reflected in the supplementary data. Given the potential for bottleneck and jackpotting effects in this assay, some measure of reproducibility is important for interpreting the results (see point 1).

We performed completely new Tn-seq experiments for each phage in duplicate. The hit lists remained largely unchanged from our initial analysis and those that were investigated further were enriched for insertions in both new data sets. Thus, the results are highly reproducible.

c. L587 "Significant hits with fewer than 10 insertions on each strand were manually removed." Why did the authors choose this criterion? Almost all of the genes they removed have very asymmetric distributions (e.g. in the Cog experiment, cgp3051 has 47853 fwd reads and 6 rev reads. Asymmetric distribution of insertions suggests that overexpression of downstream genes has an important (positive or negative) effect. This is a worthwhile pursuit, and many automated analysis pipelines can disambiguate these effects, including those developed in the Walker Lab (e.g. doi: 10.1038/s41589018-0041-4). These genes shouldn't be thrown away when they are arguably some of the most informative hits!

We have updated the criteria we used for selecting the most impactful insertion enrichments. Our concern in this report was to investigate mutants that affect phage infection when inactivated. We will pursue genes that affect phage infection when overexpressed (as indicated by asymmetric insertion orientation distributions) in a follow-on study. We think such a study would best be carried out with a different transposon containing a strong outward facing promoter.

1. There is a somewhat extensive phylogeny of M. smegmatis phages (phagesdb.org). Are the phages that the authors work on related to any of these phages? If so, what cluster do they map to? What is the host range of other phages in that cluster? If not, may be worthwhile to mention that these are quite distinct from other studied phages.

We agree that the phylogenetic history of corynephages is quite interesting. Very few phages that infect Cglu have been isolated and sequenced, let alone studied. Neither Cog nor CL31 share significant nucleotide identity with other sequenced phages, thus they do not have assigned clusters at the moment.

1. Given that cgp_0475 was a strong hit in the Tn-seq, why was it not identified in the previous chemical genomics experiments from the lab (https://doi.org/10.7554/eLife.54761) ?

We appreciate the reviewer’s interest in previous work from the lab. In the prior phenotypic analysis, cgp_0475 was identified as having severe fitness defects across many conditions. However, it was not possible to correlate its phenotype with other genes involved in mycolic acid synthesis like pks and fadD2 because they were found to be so sick in the phenotypic outgrowth that they were classified as essential.

1. Is there any relationship between the growth-rate of the mutants and their phage susceptibility? This can be analyzed using the authors' previous studies of this library.

While some of the phage resistant mutants are associated with poor fitness (namely those involved in mycolic acid synthesis), not all were associated with decreased growth. For example, there were minimal fitness defects associated with deletions of either porAH or the genes involved GalN decoration. However, loss of these genes greatly inhibited the ability of Cog to infect.

Author Response

Reviewer #1 (Public Review):

Main concerns:

1) Validation of the MCS reporters is not shown. This is particularly important for pCLIP and GoPo, which have not been reported before. Fluorescence complementation between two proteins that normally localize to different organelles is far from demonstrating the existence of a MCS between those organelles. It would be important to demonstrate using marker proteins and ideally electron microscopy/CLEM the existence of the mentioned MCS and the suitability of the fluorescent reporter.

We thank the reviewer for pointing this out and have now added supplementary characterization of the pCLIP and GoPo contact sites. The pCLIP has been previously described by us (Shai et al. 2018 Nat Commun 9, 1761. doi:10.1038/s41467-018-03957-8) and so we have only added one new figure (Figure 2 S1A) which shows the co-localization of the contact site reporter with a LD marker (MDH) and a cell periphery marker (TRITC-ConA). For the GoPo, since this is the first demonstration of a reporter for this contact site, we have rigorously characterized it by looking at the frequency of co-localization between peroxisomes and the Golgi in the absence of the reporter (Figure 1 S1B), the co-localization of the contact site reporter with a peroxisome marker (CFP-SKL) and a Golgi marker (Sec7-mCherry) (Figure 1 S1C), and by identifying a condition where this contact site is increased (Figure 1 S1D).

Since all supported their function as bone-fide reporters and since performing electron microscopy experiments on these reporters was not possible for us at this time and has not been the standard in the field for other reporters, we hope that this is satisfactory.

2) As pointed out above, the identification of a phenotype in ergosterol distribution for Ypr097W/Lec1 is very interesting. However, it is unclear how this observation relates with the localization of Lec1 to LDs, which is observed only upon over-expression.

We would like to clarify that at endogenous levels Lec1 also localizes to LDs. However, this localization is less pronounced. To clarify this in the text and show this experimentally we have now added an example of the endogenous GFP-tagged protein with the LD marker Faa4-mCherry (Figure 3 S1B), and added a section in the text.

Instead, further characterization of Ypr097w phenotype (via mutagenesis, modulation of ergosterol biosynthetic pathway, test ability to bind ergosterol, etc) in ergosterol distribution would be a plus.

To further characterize the Lec1 phenotype, we looked at changes in ADHpr-GFP-Lec1 localization in cells treated with 40 µg/ml of fluconazole for 3h (Figure 5 S2B-C). Fluconazole is a known inhibitor of Erg11 and treatment with this drug strongly reduces the overall levels of cellular ergosterol, which can be clearly observed by the loss of binding of mCherry-D4H to the plasma membrane (cytosolic signal) (Figure 5 S2B lower right panels). After 3h of treatment with fluconazole, there is a small increase in the number of cells with bud/ bud neck localization for GFP-Lec1. The GFP-Lec1 signal in these cells generally appears brighter than in untreated cells, suggesting that loss of ergosterol potentiates Lec1 accumulation at the bud / bud neck. This result suggests that Lec1 cellular localization is affected by the levels of ergosterol. However, since treatment with high concentration of fluconazole leads to growth arrest (Zhang et al. 2010. PLOS Pathogens 6:e1000939. doi:10.1371/journal.ppat.1000939), it is also possible that this signal increase is the result of Lec1 accumulation at the bud due to a stalling in budding. We now discuss this in the text.

We have also extensively mutagenized Lec1 as requested in an attempt to find a mutant that is still localized to LDs and stable yet not causing sterol redistribution. However, despite great efforts this has proven to be challenging (See below in detailed response to this request from reviewer #2).

Author Response

Reviewer #1 (Public Review):

The authors have employed a variety of techniques (single-molecule fluorescence kinetic and steady state measurements, cryo-EM structure determination, and in vivo measurements of protein synthesis and cell proliferation) to investigate the mechanism of action of two molecule products: Didemnin B and Ternatin-4. Both molecules have previously shown to target eEF1A and have potential as cancer therapeutics. In addition, the structure of Didemnin B, bound to eEIF1A and to an elongation complex, have previously been solved.

The authors here show that both compounds disrupt the dynamic accommodation of tRNA driven by eEF1A and its activation by the GTPase activation center of the ribosomal large subunit, relying on previous assignment of the FRET intensities observed in pre-steady state single-molecule fluorescence experiments in which peptide-tRNA and incoming aminoacylated tRNA are labeled with donor and acceptor dyes, respectively. They further show that this inhibition is dose dependent for both compounds and sensitive to the A399V eEF1A mutant, which creates a steric clash with didemnin B in its usual binding site. Subsequent analysis of steady-state single-molecule FRET experiments shows that didemnin B more strongly inhibits transitions between the intermediate (0.45) FRET state and the high (0.8) FRET states (though the authors choose to focus only on the effect of transitions from 0.45 to 0.8) previously assigned to the GTPase activated and fully accommodated conformations of the ternary complex, respectively. Further single-molecule experiments provide initial evidence that Didemnin B remains more stably bound to elongation complexes than does Ternatin-4.

The authors then turn to cryo-EM structures of each compound bound to elongation complexes purified either from lysate or assembled from purified components. The structure of the Ternatin-4 complex shows additional density in the same binding cleft observed for Didemnin B in a prior structure reported elsewhere, with which the Didemnin B structures reported here also agree. This binding location provides structural evidence for both compounds effects on ternary complex dynamics, as well as their previously described effects on tRNA accommodation and elongation. Further comparison of the Didemnin B and Ternatin-4 structures reveals decreased electron density in the Ternatin-4 structure for elements of eEF1A (switch loops 1 and 2 and helix alpha2), as compared to the Didemnin B structures. The authors interpret this as evidence for greater mobility of these elements, which might explain the more modest restriction of A-site tRNA dynamics they observe in the presence of Ternatin-4 (as opposed to Didemnin B). Certainly this decreased density (which might be more convincingly demonstrated using difference maps of the two structures) is consistent with that interpretation. That said, it is certainly not a smoking gun.

We have worked to soften the language pertaining to this point and have updated Fig. 4 to more accurately highlight the observed differences between the didemnin and ternatin-4 structures.

Finally, the authors turn to in vivo measurements of protein synthesis and effects on cellular proliferation or survival in the presence of both compounds. Consistent with their single-molecule experiments, they observe more severe and durable inhibition of protein synthesis in the presence of Didemnin B, whereas Ternatin-4 exhibits more modest effects that are more rapidly restored upon removal of the drug in solution. Interestingly, Ternatin-4 appears to elicit similar, and perhaps more rapid, effects on cellular survival, increasing apoptosis more rapidly than Didemnin B, though these effects (like those on protein synthesis rates) are once again more sensitive to removal of the drug. The authors describe these results as evidence that Didemnin-B "irreversibly inhibits" protein synthesis in cells. I find this assertion strange, given that the authors have previously measured a dissociation rate for this molecule from elongation complexes and they have not performed measurements to ensure that activity is not simply restored at timescales longer than their initial measurements. That said, I concede that this might be a semantic distinction if the vast majority of cells perish prior to dissociation of the drug. In either case, I would suggest the authors apply a somewhat more nuanced interpretation of these results lest they be misunderstood.

We thank Reviewer 1 for bringing this point to our attention. We have changed the title of this section to “Protein synthesis inhibition by ternatin-4, but not didemnin, can be reversed in cells,” and have softened the interpretation.

Overall, this is a rigorous and well reasoned study that employs multiple complementary techniques to investigate the mechanism of action of compounds of potential therapeutic interest. In places, the higher order interpretation of the experimental data leaks into the results section (as opposed to being fully explored in the discussion) and is at times somewhat aggressive. Nonetheless, the results presented here illuminate important questions at the intersection of translational mechanism, cell proliferation, and cancer.

We are grateful to Reviewer 1 for their assessment of this work as rigorous and well-reasoned. We have made significant updates to the text and figures, and we hope they find that we have addressed all concerns.

Reviewer #2 (Public Review):

The manuscript of Juette et al presents a combined structural and dynamic view of how a class of inhibitors (Didemins) block human ribosomal elongation. Prior work had shown that these cyclic peptide drugs bind to eEF1A in the ternary complex on the ribosome, between the Domains I and III of the factor, blocking the dissociation of the elongation factor from elongator tRNA and ribosome during decoding. Here the authors use beautiful single-molecule and structural approaches to probe the mechanisms of two related drugs-Didemnin B and Ternatin-4. Their results expand on prior observations of drug mechanism, and provide clarity for the similarities and differences on how the two drugs work both in vitro and in vivo. Using single-molecule tRNA-tRNA FRET, the authors show that the drugs (at saturating concentrations) block progression of the tRNA from a mid-FRET (GTPase activating) state to the fully accommodated (high FRET) state; they observe slightly more transitions to high FRET in the presence Ternatin-4 than Didemnin B (more below on this). These results are consistent with the idea that the drugs trap the ternary complex on the ribosome after GTP hydrolysis. Using the fraction of ribosomes that lead to accommodation, the authors performed a titration to determine the apparent Ki for the drugs (which were similar in the range of 5-10nM). They also performed clever washout experiments (always in presence of cycloheximide to block further conformational dynamics once a tRNA accommodates). These experiments probed the drug dissociation rate and showed marked differences between Didemnin-B (slow rate) and Ternatin-4 (faster rate). The authors then recapitulate the prior structural work (at lower resolution in RRL) using a reconstituted system. Their results show a similar structure as that solved previously, but with more disordered loops in the presence of ternatin-4, although the resolution here is moderate (3.2 and 3.8Å for the two drug complexes). Finally, the authors perform in vivo analyses of drug action on protein synthesis using clickable amino acid incorporation. They show that the two drugs block protein synthesis in a dose dependent manner, and that the effect of ternatin-4 can be reversed by washout of the drug, whereas that of didemnin-2 is poorly reversed, explaining differences in drug action despite the similar binding site.

Overall, this is a rigorous and well performed study probing the mechanisms of drug action in human translation elongation. The combination of dynamics measurements and structure are particularly novel, and will complement ongoing investigations (and publications) by the Blanchard lab on human elongation in general.

We thank Reviewer 2 for their assessment of this work as rigorous and novel.

Reviewer #3 (Public Review):

In this article, Juette et al employed single-molecule FRET, cryo-EM, and Hpg incorporation (in cell translation assays) to compare the mechanisms by which Didemnin B and Ternatin-4 inhibit translation elongation. They found that, while binding to the same pocket of eEF1A and blocking accommodation after GTP hydrolysis, Didemnin B had an irreversible effect on protein synthesis, but Ternatin-4, while still a potent inhibitor, allowed more flexibility in complexes (increased disorder of regions in cryo-EM structures) that allowed increased sampling of on-pathway accommodated states (observed by smFRET), and reversibility of effects on protein synthesis in cultured cells (by Hpg incorporation). This is a straightforward study and the conclusions are well-supported by the data using appropriate techniques. The work will be of impact to the ribosome field, which may use these drugs in other mechanistic studies, and researchers wanting to employ the drugs to combat cancer and other diseases.

We are thankful to Reviewer 3 for their assessment of this work as well-supported and impactful.

Author Response

Reviewer #2 (Public Review):

I have only one concern with the study. I am not fully convinced that the disruption of behavioral updating is specifically due to NA signaling within OFC. In the first two studies, they observed non-specific anatomical effect likely due to the ablation of fibers of passage through OFC. The DREADD experiment is claimed to allay this concern. However, the DCZ was injected systemically. This means that any collaterals of LC NA neurons outside OFC will also be suppressed. While the lack of effect with the mPFC projection is interesting, this does not preclude an effect mediated in other target regions. Overall, I believe that none of the experiments truly demonstrate a specific effect of NA in OFC. A few experimental options that could be considered are injection of DCZ directly in OFC, optogenetic inhibition of fibers in OFC, or pharmacological disruption of NA signaling in OFC.

The other options are to measure the effect of the toxin ablations from experiments 1 and 2 not just in mPFC but in other regions. If the non-specific effect is truly only in mPFC outside of OFC, that would lead to more confidence that mPFC projection is the only other viable pathway mediating the effect.

As requested, we have quantified the effect of toxin ablations in neighbouring cortical regions known to be involved in the goal directed behavior, namely the insular cortex (IC, e.g., Balleine & Dickinson, 2000; Parkes & Balleine, 2013) the medial orbitofrontal cortex (MO, e.g., Bradfield et al., 2015; Gourley et al., 2016) and secondary motor cortex (M2, Gremel et al., 2016). Briefly, we found that injection of the saporin toxin in the VO and LO (Experiment 1) led to a significant decrease in NA fiber density in all examined regions. Injection of 6-OHDA also produced significant loss of NA fibres in MO and M2 but not insular cortex. These results are presented in Suppl. Figures 1 and 3 (pages 28 and 30) and the statistics are reported in the main text (page 6 and page 11)

We have also added the following to our discussion on the reason for the off-target depletions that we observed and acknowledged the potential role of collateral LC neurons:

Page 21, line starting 374: “The use of the saporin toxin led to a dramatic decrease of NA fiber density in all analysed cortical areas (Suppl Fig 1). This may be due to diffusion of the toxin from the injection site, the existence of collateral LC neurons and/or fibers passing through the ventral portion of the OFC but targeting other cortical areas (Cerpa et al 2019). However, injection of 6OHDA led to much less offsite NA depletion suggesting that a large part of the previous observation is toxin-specific. Indeed, no significant loss of NA fibers was visible in the insular cortex, which has been previously implicated in goal-directed behaviour (Balleine & Dickinson, 2000; Parkes et al., 2013; 2015; 2017). We did nevertheless observe an offsite depletion in more proximal prefrontal areas (prelimbic and medial orbitofrontal cortices) albeit a more modest depletion that what was observed using the saporin toxin. Several studies have described the projection pattern of LC cells. These studies, using various techniques, indicate that LC cells mainly target a single region, and that only a small proportion of LC neurons collateralize to minor targets (Plummer et al., 2020, Kebschull et al 2016, Uematsu et al 2017, Chandler et al 2014). Therefore, even if the OFC noradrenergic innervation is presumably specific (Chandler et al 2013), we cannot rule out a possible collateralization of some neurons toward neighbouring prefrontal areas (PL and MO). We have previously discussed that the posterior ventral portion of the OFC is an entry point for LC fibers en passant, which ultimately target other prefrontal areas (Cerpa et al 2019).

To achieve a greater anatomical selectivity we used a CAV-2 vector carrying the noradrenergic promoter PRS to target either the LC:A32 or the LC:OFC pathways (Hayat et al., 2020; Hirschberg et al., 2017). It has been shown that the CAV-2 vector can infect axons-of-passage, however the vector does not spread more than 200 µm from the injection site (Schwarz et al 2015). Therefore, when targeting the OFC we injected anteriorly to the level where the highest density of fibers of passage is expected (Cerpa et al 2019) in order to minimize infection of such fibers and restrict inhibition to our pathway of interest.

Overall, the current behavioural results are in line with our previous work showing that the ability to associate new outcomes to previously acquired actions is impaired following chemogenetic inhibition of the VO and LO (Parkes et al., 2018) or disconnection of the VO and LO from the submedius thalamic nucleus (Fresno et al 2019). These results point to a necessary role of the ventral and lateral parts of the OFC and its noradrenergic innervation for updating A-O associations. However, it is worth mentioning that different subregions of the OFC, both along the medio-lateral and antero-posterior axes of OFC, display clear functional heterogeneities (Dalton et al 2016, Izquierdo 2017, Panayi & Killcross, 2018, Bradfield et al 2018, Barreiros et al 2021). Therefore, while we have previously focused on the anatomical heterogeneity of the noradrenergic innervation in these prefrontal subregions (Cerpa et al 2019), a thorough characterization of its functional role in each of these subregions still needs to be addressed.”

One last concern is that the lack of the effect due to disruption of the mPFC projection is not guaranteed to not be from experimental issues. If the authors have some evidence that the mPFC projection disruption produced some other behavioral effect, that would make the lack of effect in this case more convincing.

Unfortunately, we do not provide evidence in the current paper that disrupting the LC:mPFC (now termed LC:A32 in the current study, based on the recommendation of reviewer 1) projection produces some other behavioural effect. However, in an on-going series of experiments, using the same tools as the current study, we found that inhibiting the LC:A32, but not LC:OFC, pathway impairs Pavlovian contingency degradation as shown in the figure below. We therefore believe that the failure of LC:mPFC pathway inhibition to effect outcome identity reversal in the present study is not due to experimental issues. Please note that in the figure below mPFC is referred to as area 32 (A32), as requested by reviewer 1.

Figure 1. A) Experimental timeline for the Pavlovian contingency degradation procedure. Prior to behavioural training, rats were injected with CAV2-PRS-hM4D-mCherry into either the vlOFC or area 32 (A32). Number of food port entries during the non-degraded CS and degraded CS for rats injected with vehicle and rats injected with DCZ during degradation training (B, D) and the test in extinction (C, E). Inhibition of the LC:vlOFC had no effect on Pavlovian contingency degradation, whereas inhibition of LC:A32 during degradation training rendered rats insensitive to the change in the causal relationship between the CS and the US.

Reviewer #3 (Public Review):

I would be curious about the authors' thoughts regarding the recent Duan ... Robbins Neuron paper (https://pubmed.ncbi.nlm.nih.gov/34171290/), in which marmosets displayed paradoxical responses to VLO inactivation and stimulation in contingency degradation tasks. Are there ways to reconcile these reports?

We previously argued that the updating processes underlying changes in causal contingency versus outcome identity may be supported by different prefrontal regions (Cerpa et al., 2021, Behav Neurosci). Unfortunately, the tasks used in the current study do not allow us to test if our rats are sensitive to changes in the action-outcome contingency. In fact, the effect of inactivation (or overactivation) of the ventral and lateral regions of OFC on an instrumental contingency degradation task similar to that used in Duan et al (2022) has not yet been examined in rats.

Indeed, while it is stated in Duan et al (2022) that rats with lesions of lateral OFC are insensitive to contingency degradation, none of the citations provided support this conclusion (Balleine & Dickinson, 1998; Corbit & Balleine, 2003; Ostlund and Balleine, 2007; Yin et al., 2005). Balleine and Dickinson (1998) assessed the effect of prelimbic and insular cortex lesions (insular anteroposterior coordinate +1.2), with only the former affecting instrumental contingency degradation. Ostlund and Balleine (2007) assessed the effect of orbitofrontal lesions on Pavlovian contingency degradation (degradation of the S-O contingency) not instrumental contingency degradation. Finally, Corbit and Balleine (2003) and Yin et al (2005) assessed the effect of prelimbic and dorsomedial striatum lesions, respectively. Nevertheless, there are some reports on the effect of chemogenetic inhibition of VO/LO on degradation in a nose-poke response task but the results are conflicting (e.g., Whyte et al., 2019; Zimmerman et al., 2017; 2018). It would be very interesting to study the impact of both inactivation and overactivation of VO and LO in rats to compare with the results found in marmosets, using comparable tasks.

We have added the following to our discussion, which cites Duan et al (2022) and the need to better understand the role of VO and LO in contingency degradation.

Page 24, line starting 450: “However, it is not yet clear if the NA-OFC system is also involved in detecting the causal relationship between an action and its outcome (see Cerpa et al., 2021 for a discussion). Some have reported impaired adaptation to contingency changes following inhibition of VO and LO or BDNF-knockdown in these regions (Whyte et al., 2019; Zimmerman et al., 2017), while another study shows that inhibition of VO/LO leaves sensitivity to degradation intact, at least during an initial test (Zimmerman et al., 2018). Interestingly, a recent paper in marmosets demonstrates that inactivation of anterior OFC (area 11) improves instrumental contingency degradation, whereas overactivation impairs degradation (Duan et al., 2022). The potential role of the rodent ventral and lateral regions of OFC, and the NA innervation of OFC, in adapting to degradation of instrumental contingencies requires further investigation.”

Author Response

Reviewer #2 (Public Review):

The role of cMAF in the formation of iNKT10 is only suggested ny the transcriptional signatures analyzed here. There is no direct evidence that cMAF is indeed needed to generate iNKT10. This should be investigated.

We thank the reviewer for their comments on the link between IL-10, NKT10 cells, and cMAF. We agree that our study provides evidence that cMAF is a promising candidate regulator of IL-10 production by iNKT cells, and we attempted to address this using gene-specific knockout mice. Since mice lacking expression of cMAF exhibit post-natal mortality and severe developmental defects13⁠⁠ we attempted to breed Maffl/flCd4-cre mice, which have previously been used to study the role of cMAF in T cell function14⁠⁠. However, we were not able to successfully breed enough of these mice to assay whether or not cMAF is required for the production of IL-10 by iNKT cells. Therefore, our study can only suggest that cMAF is a promising candidate regulator of NKT10 cells based on our transcriptomic data and flow cytomery data showing that production of IL-10 is associated with expression of cMAF. However, we present further correlative or indirect evidence to this effect. It has previously been demonstrated that restimulation of activated iNKT cells at 72 hours post-⍺GalCer results in increased production of IL-10 compared to the stimulation of iNKT cells at steady state15⁠⁠. We found that the frequency of splenic cMAF+ iNKT cells was greatly increased at 72 hours post-⍺GalCer compared to steady state (Figure S3B, Figure S3D) and this increase in expression of cMAF correlated with increased production of IL-10 (Figure S3E-S3F). Therefore, we believe that cMAF is a promising candidate for future work examining the functional landscape of NKT10 cells and we anticipate that our study will be a useful transcriptomic reference for such studies.

The Kronenberg group recently published a similar analysis, using RNAseq and ATACseq. Although I don't believe the cMAF signature was highlighted at the time, one could argue that this previously published study dampens the originality of this manuscript. Although this study (Murray et al.) is clearly acknowledged, the similarities and differences in both the methodology and findings should be clearly discussed.

As the reviewer stated, the excellent study by Murray et al. (2021) did not identify or highlight a population of cMAF+ iNKT cells expressing a regulatory gene signature, as presented in our study, and as the reviewer mentions, we cite and discuss the Murray et al. study in our manuscript. We believe that both studies together provide a comprehensive transcriptomic analysis of iNKT cells after activation, and that ours provides unique insight not found in Murray et al. Our study uses scRNA-Seq rather than bulk RNA-Seq or bulk ATAC-Seq methods, enabling us to study transcriptomic characteristics of activation among heterogeneous iNKT cell subsets without needing to sort pre-identified iNKT cell populations or subsets. It is the use of unbiased scRNA-Seq that allowed us to identify cMAF+ iNKT cells, since this population has not been previously described in the literature. Notably, we also sequenced the largest number of iNKT cells to date, 48,813 cells, to the best of our knowledge, which provides deeper insight. We also performed transcriptomic characterization of activated iNKT cells at different stages of activation to those characterized by Murray et al. Importantly, we profiled the phenotype of iNKT cells at 4 hours post-⍺GalCer and 72 hours post-⍺GalCer, when iNKT cells engage in a rapid cytokine production or undergo proliferation and expansion. This revealed several novel transcriptional insights including rapid metabolic gene reprogramming that occurs twice during this activation timeline. By contrast, Murray et al. focused on analysis of iNKT cells at steady state and 6 days post-⍺GalCer. Finally, we performed transcriptional characterization of adipose iNKT cells in our study, which are known to represent an unusual regulatory population of iNKT cells at steady state16⁠⁠, whereas the study by Murray et al. (2021) did not study adipose iNKT cells. Therefore, we propose that our study complements the excellent work performed by Murray et al. (2021) but provides novel insight in terms of focus, discovery, and scope.

The authors should clearly describe the genes that were used to define iNKT1/2/17 identity in their study. This is important in order to track that identity over time following activation, at it is well known that the expression of some of the markers typically used change following activation. This would bring clarity to the manuscript.

We agree with the reviewer and we had originally removed two clarifying figures for iNKT cell subset identification due to space, but now we have included these two clarifying supplemental figures (Figure S4, Figure S5) to illustrate how we identified NKT1, NKT2, and NKT17 cell subsets in our scRNA-Seq data. We have also added further details to the Methods section (please see the “Downstream scRNA-Seq data analysis” section) and we have changed the title of the activated iNKT cell data in Figure 2A-2D and Figure 4C from “4 hours post-⍺GalCer” to “Activated (4 hours post-⍺GalCer)” to reflect our subset identification protocol as accurately as possible (please see below).

Steady state and activated splenic NKT1, NKT2 and NKT17 cell subset identification was performed as follows: We identified spacial separation and graph-based clustering of five main populations of cells at steady state (Figure S4A). We then used the expression of the published marker genes Tbx21, Zbtb16, Rorc and Mki67 to identify NKT1, NKT2, NKT17 and Cycling cells (Figure S4B). We identified spacial separation and graph-based clustering of three main populations at 4 hours post-αGalCer (Figure S4D). However, we found that Tbx21 and Zbtb16 expression was increased across multiple clusters and did not effectively demarcate NKT1 and NKT2 cells at the RNA level (Figure 2B), and so we instead used the flagship cytokines Ifng, Il4, Il13 and Il17a and Il17f to demarcate NKT1, NKT2 and NKT17 cells. We then combined the identified NKT1, NKT2 and NKT17 cell populations from steady state and 4 hours post-⍺GalCer together (i.e. cells from the same subset at the two different time points were combined together) and performed reclustering of the cells within each subset (Figure S5). It has previously been shown that there can be differences in the activation kinetic of different splenic iNKT cell subsets, for example NKT1 versus NKT2 cells, which may be in part due to physical localization, for example in the red pulp versus the white pulp of the spleen (see Lee et al. 2015)17⁠. We observed a similar phenotype for NKT2 cells in our data, whereby a proportion of NKT2 cells at 4 hour post-⍺GalCer clustered with NKT2 cells from mice that received no ⍺GalCer (Figure S5A). To prevent differences in activation kinetic from biasing our analysis of transcriptional signatures of iNKT cell subset activation, we performed low-level graph-based reclustering within each iNKT cell subset to accurately segregate activated and steady state iNKT cells (Figure S5B). We validated our reclustering using the expression of activation markers and flagship cytokines (Figure S5C-S5D). Finally, these reclustered subset data were recombined and renormalized to generate the final analysis as shown in Figure 2 of the manuscript.

Author Response

Reviewer #1 (Public Review):

Champer et al. evaluate two homing drives that have been developed in the Anopheles mosquito. Variants of one of these (zpg) are possibly being further investigated for an eventual release. Work with the other has seemingly been discontinued because of unintended fitness costs. The authors argue that this second drive may be in fact better if the experimental results are interpreted more favourably. An important point if true, but somewhat separate from the findings in the paper. To a large extent, this point could be made without any of the results in the paper. However, the authors do show through modelling that this difference may in fact be relevant.

This careful justification of the model parameters increases its relevance to the evaluation of those specific gene drives. The zpg drive will likely be extensively investigated and the specific relevance of this work is a valuable contribution. While a range of parameters is tested for each expression pattern, there are no step-by-step investigations of how the drive outcomes are effect by changes to the underlying DNA-repair/deposition/fitness parameters. So while a reader may learn one drive is better than the other, the ability to get a deeper understanding of the underlying relationship is limited. This means this work has a more limited scope and relies on the relevance of the chosen parameters. In that regard, there may be room for improvement. The chosen parameters for zpg and nos may not be completely fair in regards to the target site and I believe this needs to be addressed.

The second aspect of this paper is the comparison between the commonly used panmictic modelling approach and spacial models. This also somewhat relies on the drive parameters being chosen well, as a more comprehensive evaluation of the spacial approach has been done in prior work by this group. However, showing that these particular extremely efficient drives may still struggle when additional spacial factors are considered is useful and relevant. That a second Anopheles-specific spacial model further reduces the drive performance is a relevant finding. This is helped by a specific analysis of the effect of changes to the migration rates and the low-density growth rate. This spacial modelling also has relevant findings for the homing X-shredder design.

In our previous study (Champer, Kim, et al, 2021 in Molecular Ecology) that we reference, we varied some of these drive performance parameters, which may address some of the reviewer’s concerns. We view this study as building off that one, but with a more specific focus (mosquitoes and existing drives). We also now discuss how using parameters for a different target site may have affected our results (see below - nos may actually have been shortchanged since zpg performs better at dsx than at nudel).

Reviewer #2 (Public Review):

Champer and colleagues present forward simulations of several gene drive systems that have been designed to suppress the malaria mosquito, Anopheles gambiae. These gene drives have all been validated in laboratory cage experiments but have not yet progressed to field trials. The authors are particularly concerned with the phenomenon of "chasing," in which local success of the drive will lead to continuous cycles of recolonization by wildtype mosquitoes, preventing complete suppression of the population. In addition to their spatially-explicit model, they additionally present results from a model in which the parameters are tuned to the ecology of the mosquito.

Though there are a few additions that would improve the manuscript, the authors achieved these aims and their conclusions are supported by their modeling, which appears to be technically sound and well executed.

Strengths:

The work represents a useful, model-based comparison of the various Anopheles gene drives that have had success in laboratory conditions. With the incorporation of spatial dynamics, the authors are thereby able to focus on the problem of chasing, or a fluctuating equilibrium state that is impossible to study in laboratory colonies. Through a comparative framework, the authors additionally provide key information on the differences in predicted success between the various gene drive systems. For these reasons, the work will be a useful addition to the other published forecasts of gene drive success. Given the importance of the topic to diverse stakeholders that vary in their familiarity with gene drives and ecological modeling, I was glad to see the authors summarize their findings cogently and accessibly.

We thank the reviewer for these kind comments.

Weaknesses:

The main area in which the manuscript could be strengthened is the description of the Anopheles-specific model. Based in part on the differences observed between their discrete generation model and Anopheles-specific model, the authors correctly note that "the outcome of a drive release could be very sensitive to the precise ecological characteristics of the targeted population." It was sometimes unclear which model parameter choices were informed by literature and how much confidence was had in each. A more explicit summary or perhaps a table of parameters, references, and estimates with confidence ranges informed by the authors' knowledge of literature would strengthen this section.

We now have added Table S1 showing all model parameters. The parameters themselves often require more text for justification than just references, but we have improved our methods section throughout to increase the visibility and clarity of these sections. Note that mosquito ecology is a fairly understudied field, resulting in widely varying parameter ranges throughout different studies, so it is difficult to provide confidence ranges for our parameters, other than that they are designed to fall within estimates from different studies (see “Anopheles-specific spatial model” methods subsection).

In the framing of the work, the authors imply that their modeling study "suggest(s) an alternative interpretation of [the] performance [of homing gene drives]" from recent studies (e.g., Simoni et al. 2020 and Kyrou et al. 2018). I am not certain this framing is justified, given the original authors' circumspection in correctly noting their drives had success in the cage experiments, without claiming they would be successful in the wild. I would prefer this study be presented as building on those previous studies and extending their work.

In crafting this sentence, we had in mind very specific technical interpretations of drive performance mechanisms (paternal deposition vs. more somatic fitness cost, existing of somatic fitness cost in nos males) rather than general performance in any given environment. To more clearly convey our intended meaning here, we have adjusted our wording. The sentence now reads: “Here, we analyze data associated with each of these gene drives and consider both the original and alternative interpretations of these drives’ characteristics and performance parameters.”

Likely impact:

This work will be of interest to research scientists whose interests range from transgenic mosquitoes to ecological modelers to post-release assessment. The authors correctly note that additional refinement of the ecological parameters will increase the utility of the model, but the framework as it stands will be an important contribution to the literature. Given the timeliness of this topic, the subject is of interest to other stakeholders in the regulatory or policymaking realm, as well as governmental and funding agencies deciding between gene drive systems.

Reviewer #3 (Public Review):

This is a computational modeling study to evaluate the merits (likely success) of different 'suppression' gene drive systems. Gene drives offer a possible simple and low-effort means of suppressing or even extinguishing pest populations. Using CRISPR technology, several gene drive systems have been developed in the last decade for key mosquito vector species. As no gene drive has been approved for release in the wild, efforts to evaluate their likely success are limited to cage trials and modeling, the latter as done here. In contrast to some modeling studies, the effort here is to develop and analyze models that match the gene drive and mosquito biology closely. The models are thus parameterized with values representative of what is known about mosquito biology and of the various gene drive constructs that have been developed for lab studies.

In these models, gene drive success or failure in population suppression largely depends on (i) how well the drive spreads throughout the population, and (ii) whether the population persists because of a type of ongoing spatial 'group selection' in which local pockets invaded by the drive die out and are then repopulated by migrants lacking the drive. Formal evolution of functional resistance is not allowed. The numerical results show striking differences in suppression success with different gene drive constructions, and these differences are likely to be of use when designing drives for actual releases.

The basic group selection outcome that allows population persistence amid a suppression gene drive has been shown before, as cited in the ms. The novelty provided by the present study is to tie the models to the biology of known gene drive constructions. Given the high specificity of the models, the audience for this work is likely to be somewhat narrow, confined to those involved in gene drive design. The work is nonetheless significant in view of the strong potential of gene drives in global public health efforts.

The software used to generate the trials is freely available from one of the authors for anyone wishing to repeat the simulations. There is an extensive supplement of results referenced (but not otherwise included) in the main text.

We thank the reviewer for these comments, and we note that an analysis of functional resistance was performed in our previous study. Because the results of this study are likely to be fully applicable to our new results, we did not repeat it with our mosquito model and with our parameterized drives. However, it is certainly an important topic, and we explicitly mention in the discussion how the possibility of chasing requires further consideration in the acceptable rate of functional resistance allele formation. The text reads, “We did not consider the possibility of.... functional resistance, which can evolve more readily during lengthy chases11.”

Author Response

Reviewer #1 (Public Review):

Overall, it is an interesting work exploring stochastic and deterministic aspects of embryonic cell division in plants. The power of the authors' approach lies in the quantitative analysis of 3D cell geometries combined with quantitative computer modelling.

I am a bit confused about how authors relate stochasticity as an emergent property of a deterministic process. Typically, stochasticity is the low-level process resulting in variation of subcellular components those also related to the positioning of the cell division plane. Perhaps a more elaborated and clearer connection between stochasticity at the subcellular level and phenotypic variability should be provided.

Actually, our interpretation does not directly relate variability in division patterns to a deterministic selection of division plane orientation. A key intermediate between these two scales is the variability in cell geometry. Based on our results, we propose that the selection of division plane orientation would obey a deterministic principle based on geometrical constraints. Variability in cell division patterns would ensue from the expression of this deterministic rule in the variable context of cell geometries. Variability in cell geometry would itself result from noise in the precise positioning of the division plane along the optimal, deterministic orientation. We have added a new summary Figure 10 to illustrate and clarify this interpretation.

I have a number of specific questions/concerns that I would like the authors to address as listed below:

1) Major variability of cell shapes is observed in the apical domain as opposed to the basal domain. What would be an underlying principle to asymmetric shaping of the apical-basal domain? The authors describe beautifully the observations but give relatively little discussion on this matter, leaving the reader guessing.

We added a new paragraph (before the last) in the Discussion on this point. The origin for a larger variability in the apical than in the basal domain can be found in part in the different cell shapes present in the two domains at stage 16C. Along the path tetrahedron -> triangular prism -> cuboid, the apical domain indeed appears farther from the final absorbing state represented by the cuboid shape, hence more time will be spent in the intermediate shapes in this domain. In addition, the tetrahedral and triangular prism shapes are closer to rotational symmetry and thus represent a larger source of variability in division plane orientation. Lastly, apical and basal cells have distinct environments, the basal cells being constrained between the suspensor, on one side, and apical cells, on the other. We can only speculate about the functional significance, if any, of the larger variability in the apical domain. For example, we can relate it to the future morphological transition that will characterize the apical domain with the emergence of the cotyledons at the heart stage. A variable pattern of cell walls could be required to establish a specific mechanical pattern at the tissue scale to favor this shape transition.

2) Authors used graph theory to explain variability in cell division for the same topological feature. In light of quite a discrepancy between predictions and observations (i.e., Figure 4C) question arises of how this prediction could be affected by undergoing cell expansions as this element I believe is neglected in their graph theory approach?

It is indeed the case that cell edge lengths are ignored in the graph-theoretical approach described in Section 2.4. In this part of the paper, our objective was to objectively test if non-topological factors were implicated in the determination of division orientations. The rationale was thus to compare observed patterns to predictions obtained using topological information only. The strong discrepancy between observations and predictions (Figure 4C) confirmed that topology alone was not sufficient to predict observed patterns. The integration of cell geometry (including edge lengths) into the predictions is considered in the next sections (Sections 2.5 and 2.6).

We believe the modifications we made in Section 2.4 to answer Reviewer 3’s comment on this part (see below) should make this point clearer.

3) Tetrahedron shape repeats in only 4% of embryos at the 16-cell stage. What could be the criticality of this shape for the entire embryo patterning?

At the 16C stage, there are four domains (apical/basal x inner/outer) represented each by exactly one cell shape. The triangular prismatic shape is observed in two domains (outer apical and inner basal). The two other cell shapes, cuboid and tetrahedron, are specific to the two other domains, respectively outer basal and inner apical. Hence, the tetrahedron shape represents 25%, not 4% of cell shapes at the 16-cell stage, a proportion large enough to potentially impinge on embryo patterning.

Our graph analysis shows that, under a topologically random regime of cell division, the tetrahedron shape should progressively vanish because, due to 4-way junction avoidance, a tetrahedron cannot divide into two tetrahedra and because divisions of triangular prisms and cuboids generate a minor proportion of tetrahedra only in comparison with the other cell shapes (Figure 4C). In addition, our analysis also shows that the triangular prismatic shape is a necessary intermediate to transit from a tetrahedron to a cuboid shape.

Altogether, the presence of the tetrahedron shape in the inner apical domain at 16-cell stage could be responsible for the large variability in cell shape subsequently observed in this domain. (See also our answer to Comment 1 of the Reviewer).

4) It is not clear to me whether stochastic cell division modelling takes into account the mechanical influence of adjacent cells? In any case, authors should discuss how this could potentially affect their analysis.

Indeed, our stochastic cell division model only takes into account the geometry of the mother cell, ignoring the possible influence of the environment of the cell within the tissue (through mechanical signals or other, such as hormonal signals) - except of course for indirect effects that the environment could exert on cell shape. The possible mechanical influence of the cell environment was already discussed in the original version of our manuscript (last paragraph of the Discussion). In particular, we mentioned that the specific localization deep inside the embryo of the inner basal cells could mechanistically influence the positioning of the division plane, thus explaining the strong discrepancy between observations and predictions in this domain. This negative result illustrates how the cell-autonomous model can be useful in pointing to possible environmental influence (or alternative geometrical rules yet to be identified) and in suggesting future directions of investigation.

To remove any ambiguity, we have made more explicit the cell-autonomous nature of the model (Section 2.5 and Material and Methods).

5) Authors should perform model parameter sensitivity analysis (i.e., position of surfaces) to confirm the convergence and robustness of their approach.

In the first version, we already reported in Supplementary Figures S8 to S15, for each domain and each orientation of division, the distribution plots (surface area x distance to cell center) of simulations performed in different cells. In each of these figures, the cells shared the same shape (cuboid, triangular prism or tetrahedron) but differed in their exact geometry. As can be seen from these graphs, similar point distributions were obtained for different cells and, more importantly, the simulations matching best with observed patterns shared the same relative localization within these distributions (except of course when the geometrical rule was not valid, as in the basal inner domain). Therefore, these results already provide a sensitivity analysis to shape fluctuations. To make this point more explicit, we now show in a new Supplementary Figure S12 the 3D shapes associated with the first set of graphs (basal outer domain; Figure S11) and added reference to this figure in Section 2.5.

In addition to biological variability, possible minor errors and uncertainties at the image processing and segmentation step may also affect mother cell geometry. To illustrate the robustness of our approach to this potential source of geometrical variability, we have added a new Supplementary Figure S10 showing the distribution plots of simulations performed within a raw mother cell mask and within its mask following filtering using a mathematical morphological opening with radius of 1, 2, or 3 voxels. Morphological opening is an image processing operation that smoothes binary objects, removing extrusions having a radius smaller than the prescribed radius. The obtained results show the robustness of our results to such alterations of mother cell geometry. In the four conditions (R=0,1,2,3), the simulated patterns matching best with the observed pattern are located at the same bottom left position of the plot, corresponding to the geometrical rule.

Lastly, we have also added a new Supplementary Figure S9 to illustrate the reproducibility of simulations results obtained within a given mother cell. In this figure, we show the distribution plots for two independent sets of 1000 simulations each. The graphs show similar distributions, with identical locations at the bottom left of the distribution of the simulations that best matched with the observed division plane.

Concerning the convergence of the 3D cell division computer model, we have added in Section 4.3 (Material and Methods: Computer modeling of cell divisions) the justification about the number of Monte Carlo cycles. We have added a new Supplementary Figure S7 illustrating the convergence of the algorithm over different independent runs. We also corrected a typo on the number of Monte Carlo cycles (which was 500 instead of 5000 as initially written).

Reviewer #2 (Public Review):

This is an interesting manuscript aiming at identifying minimal rules that account for cell divisions in early Arabidopsis embryos. This research has two main strengths. The authors consider cell division in 3 dimensions, whereas most other studies on the orientation of cell divisions are restricted to 2 dimensions. Based on their observations, the authors proposed that the previously proposed probabilistic rule for cell division can be replaced by a deterministic rule, with sources of stochasticity coming from irregularities/imperfections in cell geometry. The manuscript is overall well-written. I nevertheless have a few concerns.

1) What is the effect of embryo fixation on cell geometry? Could the irregularities be an artefact due to fixation? How robust are the conclusions to numerical perturbations of the position of cell surfaces?

We used the fixation and staining protocol developed by one of us (JCP) (Truernit et al 2008). Yoshida et al. Developmental Cell (2014) used this same protocol, which they validated by comparison with live imaging data. The fixation and the following treatment could have an impact on cell geometry. For this reason, we have selected among a thousand embryo acquisitions, the embryos that are not or very few damaged with this treatment. The robustness of our results and conclusions to variability and alterations of cell geometries was also questioned by Reviewer #1. Please see above our in-depth answer to this point.

2) Section 2.7 on attractor patterns is essentially descriptive and the conclusions seem to be based on qualitative observation of a few cases. Can the authors support them with quantitative measures? Or with simulations?

We have completed this section with quantitative data when it was missing. In the apical outer domain, we had 135 observations at G6, which had been reached from G4 according to one or the other of the two main pathways shown in Figure 9A. These two possibilities accounted for 40% and 42% of observations, respectively.

The other attractors shown in Figure 9 are rare cases (Fig. 9B: 1 case over 173; Fig. 9C: less than 9 cases over 309). The case shown in Figure 9B was previously documented in Scheres et al 1995 (cited in the manuscript).

The lower frequency in the basal domain of alternative sequences leading to a same attractor pattern is consistent with the lower variability in this domain. However, the conclusion is the same as in the apical domain where the distribution between alternative sequences is more balanced: different sequences of division over several generations can lead to similar cell patterns.

Author Response

Reviewer #2 (Public Review):

Fibular hemimelia (FH) is a rare genetic disorder with unknown mechanisms. In this study, the authors generated Axin1 conditional knockout (cKO) mice by depleting Axin1 gene specifically in Prx-1 expressing mesenchymal cells and demonstrated that Axin1 cKO mice developed FH phenotype with various severities. FH phenotype in Axin1 cKO mice can be rescued by either β-catenin or BMP inhibition if the inhibition was applied to the pregnant mother at E9.5 to E12.5. For mechanistic study, the authors showed elevated expression of BMP signaling molecules in limb tissue of Axin1 cKO mice and Axin1 regulated the degradation of pSmad5 in mesenchymal cells.

The study has many strengths. 1) The study was performed with high rigor. Utilization of various cre lines to conditional KO Anix1 in different cell types to formally demonstrate the expression of Anix1 in Prx-1-expressing mesenchymal cells, but not in Sox9-, Col2-, and Osx-expressing cells is required for normal fibular development. 2) Treatment of Axin1 cKO mice with β-catenin and BMP inhibitor at different time points to demonstrate that inhibitors should be given during the early embryonic development, a very important point for considering the translational potential of the study. 3) Detailed in vitro experiments were performed to investigate the molecular mechanisms of Axin 1 on Smad 5 stability. 4) Both β-canenin and BMP signaling pathways are important, including skeletal development, this study used Axin1 cKO mice to integrate these two pathways together, which is a important and new contribution.

Weaknesses of the study have been described below:

1) Authors need to report/describe findings/pheotypes in bones other than fibula in Axin1 cKO mice (4-8-week-old) first, and then focus on fibular development. From the X-ray data shown in Figure 1D-E, it appears that Axin1 cKO mice have high bone mass or osteopetrosis. Thus, histology of bones (femur, tibia, knee joint) other than fibula should be provided.

We have performed histology in femur, tibia and knee joint in Axin1 KO mice as the reviewer suggested.

2) Fig. 2 described Axin1/2 dKO mice. I suggest to remove Figure 2 or move it to supplemental data. Including Axin1/2 dKO mice in the main text makes the story complicated and difficult to explain because the most of figures in this manuscript were on Axin 1, such as rescue experiments and molecular mechanistic study. Further, various severity of FH in Axin1 cKO mice are closer to human FH cases (various severity) than Axin1/2 dKO mice that have a completed loss of fibula. The title is also on Axin1. If Axin1/2 dKO mouse data are included in the main text, authors need to provide molecular explanation why Axin1/2 dKO mice have more severe phenotypes.

To make the entire story more straightfoward, we have removed the Axin1/2 double KO data (Fig. 2) as the reviewer suggested.

3) Please include a paragraph in the discussion regarding the limitation of the study. Is there any human report that FH patients have mutation in Axin 1 and its related downstream signal proteins such as β-catenin and BMP? Can FH being directed before birth and to treat pregnant mother? Do authors plan to use unbiased approaches such as RNAseq or proteomics to discover new gene/proteins that are regulated by Axin1 in mesenchymal cells?

We have added a paragraph to discuss the limitation in the discussion section as the reviewer suggested. We have collaborated with Dr. Qinglin Kang and collected 9 samples from patients with FH disease and identified a mutation of β-catenin gene, encoding a potential phosphorylation site, which may lead to upregulation of β-catenin protein levels. In the future, we will investigate if the mutation of β-catenin affects its function in mesenchymal cells. We are currently planning to perform the RNA-Seq and proteomics experiments to identify novel downstream target gene(s) of Axin1.