eLife assessment

In this study, the authors (1) build a detailed model of the process whereby visual information is stored in neuronal activity from the moment that a stimulus is presented until some time later when such information needs to be recalled, and (2) test numerous mechanistic assumptions by fitting the model to rich psychophysical experiments performed by human participants. The evidence supporting the claims is convincing, utilizing detailed model comparison to evaluate potential mechanisms. Overall, the results represent a valuable advance in our understanding of how sensory representations are encoded in and then recalled from, working memory.

Reviewer #1 (Public Review):

Summary:<br /> This study investigates how the neural representation of a stimulus transitions from that evoked by the presence of the stimulus (sensory) to one that exists only as a memory trace once the stimulus disappears (mnemonic). In simple terms, it explores the transition from so-called "iconic memory" (akin to residual sensory-driven neural activity) to working memory proper (self-sustained activity). The authors build a computational model for this transition and test it against data from two new psychophysical experiments plus two datasets from prior experiments.

Strengths and weaknesses:<br /> I really liked this work. It considers a fairly complex process but builds a mechanistically comprehensive scheme that is intuitive and testable. This is a hefty paper; the full model built by the authors has a lot of moving parts. But these are all carefully justified, and in fact, many of them are specifically tested by fitting customized variants of the model to the experimental data (which are rich enough to distinguish all of these variants, not only quantitatively but also qualitatively). Said differently, both the assumptions used to build the model and the conclusions drawn after comparison with the experimental data are well justified. In the end, although it takes some effort to put the whole scheme together, I think the reader learns a lot about memory mechanisms. The Discussion is rich, as beyond working memory per se, the work relates to numerous issues (e.g., perception, attention, neural dynamics, population coding). Importantly, although part of the value of the study lies in the way it integrates many prior results into a cohesive framework, it also makes an important novel point: that iconic and working memory are not qualitatively different things, but rather just different extreme manifestations of the one, continuous process whereby perceptual information is stored (as a pattern of neural activation) and made accessible to other cognitive functions. In this conceptualization, working memory corresponds to a readout of activity a significant time (typically > 1 s) after stimulus offset, whereas iconic memory is consistent with a readout from the same neural population but immediately or very shortly after stimulus offset. This account not only is parsimonious but also provides a specific hypothesis (or a set of hypotheses) that can be tested further.

I did not find any major weaknesses. The paper does require some time and effort in order to appreciate all that it contains, but this is inevitable, as it aims to (1) build a compact but mechanistically detailed account of a process that is somewhat complex, and (2) test key predictions through psychophysical experiments that must be sufficiently rich. In the end, I found the effort quite rewarding.

Reviewer #2 (Public Review):

Summary:<br /> Previous work has shown subjects can use a form of short-term sensory memory, known as 'iconic memory', to accurately remember stimuli over short periods of time (several hundred milliseconds). Working memory maintains representations for longer periods of time but is strictly limited in its capacity. While it has long been assumed that sensory information acts as the input to working memory, a process model of this transfer has been missing. To address this, Tomic and Bays present the Dynamic Neural Resource (DyNR) model. The DyNR model captures the dynamics of processing sensory stimuli, transferring that representation into working memory, the diffusion of representations within working memory, and the selection of memory for report.

The DyNR model captures many of the effects observed in behavior. Most importantly, psychophysical experiments found the greater the delay between stimulus presentation and the selection of an item from working memory, the greater the error. This effect also depended on working memory load. In the model, these effects are captured by the exponential decay of sensory representations (i.e., iconic memory) following the offset of the stimulus. Once the selection cue is presented, residual information in iconic memory can be integrated into working memory, improving the strength of the representation and reducing error. This selection process takes time, and is slower for larger memory loads, explaining the observation that memory seems to decay instantly. The authors compare the DyNR model to several variants, demonstrating the importance of the persistence of sensory information in iconic memory, normalization of representations with increasing memory load, and diffusion of memories over time.

Strengths:<br /> The manuscript provides a convincing argument for the interaction of iconic memory and working memory, as captured by the DyNR model. The analyses are cutting-edge and the results are well captured by the DyNR model. In particular, a strength of the manuscript is the comparison of the DyNR model to several alternative variants.

The results provide a process model for interactions between iconic memory and working memory. This will be of interest to neuroscientists and psychologists studying working memory. I see this work as providing a foundation for understanding behavior in continuous working memory tasks, from either a mechanistic, neuroscience, perspective or as a high-water mark for comparison to other psychological process models.

Finally, the manuscript is well written. The DyNR model is nicely described and an intuition for the dynamics of the model is clearly shown in Figures 2 and 4.

Weaknesses:<br /> Despite its strengths, the paper does have some (relatively minor) weaknesses. In particular, the authors could consider the role of sensory processing, and its limitations, and variability in selecting an item from working memory as other factors affecting memory accuracy.

Reviewer #3 (Public Review):

Summary:<br /> The authors set out to formally contrast several theoretical models of working memory, being particularly interested in comparing the models regarding their ability to explain cueing effects at short cue durations. These benefits are traditionally attributed to the existence of a high capacity, rapidly decaying sensory storage which can be directly read out following short latency retro-cues. Based on the model fits, the authors alternatively suggest that cue-benefits arise from a freeing of working memory resources, which at short cue latencies can be utilized to encode additional sensory information into VWM.

A dynamic neural population model consisting of separate sensory and VWM populations was used to explain temporal VWM fidelity of human behavioral data collected during several working memory tasks. VWM fidelity was probed at several timepoints during encoding, while sensory information was available, and maintenance when sensory information was no longer available. Furthermore, set size and exposure durations were manipulated to disentangle contributions of sensory and visual working memory.

Overall, the model explained human memory fidelity well, accounting for set size, exposure time, retention time, error distributions, and swap errors. Crucially the model suggests that recall at short delays is due to post-cue integration of sensory information into VWM as opposed to direct readout from sensory memory. The authors formally address several alternative theories, demonstrating that models with reduced sensory persistence, direct readout from sensory memory, no set-size dependent delays in cue processing, and constant accumulation rate provide significantly worse fits to the data.

I congratulate the authors for this rigorous scientific work. I have only very few remarks that I hope the authors can clarify.

Reviewer #2 (Public Review):

The contribution of glial cells to the pathogenesis of amyotrophic lateral sclerosis (ALS) is of substantial interest and the investigators have contributed significantly to this emerging field via prior publications. In the present study, authors use a SOD1G93A mouse model to elucidate the role of astrocyte ephrinB2 signaling in ALS disease progression. Erythropoietin-producing human hepatocellular receptors (Ephs) and the Eph receptor-interacting proteins (ephrins) signaling is an important mediators of signaling between neurons and non-neuronal cells in the nervous system. Recent evidence suggests that dysregulated Eph-ephrin signaling in the mature CNS is a feature of neurodegenerative diseases. In the ALS model, upregulated Eph4A expression in motor neurons has been linked to disease pathogenesis. In the present study, authors extend previous findings to a new class of ephrinB2 ligands. Urban et al. hypothesize that upregulated ephrinB2 signaling contributes to disease pathogenesis in ALS mice. The authors successfully test this hypothesis and their results generally support their conclusion.

Major strengths of this work include a robust study design, a well-defined translational model, and complementary biochemical and experimental methods such that correlated findings are followed up by interventional studies. Authors show that ephrinB2 ligand expression is progressively upregulated in the ventral horn of the cervical and lumbar spinal cord through pre-symptomatic to end stages of the disease. This novel association was also observed in lumbar spinal cord samples from post-mortem samples of human ALS donors with a SOD1 mutation. Further, they use a lentiviral approach to drive knock-down of ephrinB2 in the central cervical region of SOD1G93A mice at the pre-symptomatic stage. Interestingly, in spite of using a non-specific promoter, authors note that the lentiviral expression was preferentially driven in astrocytes.

Since respiratory compromise is a leading cause of morbidity in the ALS population, the authors proceed to characterize the impact of ephrinB2 knockdown on diaphragm muscle output. In mice approaching the end stage of the disease, electrophysiological recordings from the diaphragm muscle show that animals in the knock-down group exhibited a ~60% larger amplitude. This functional preservation of diaphragm function was also accompanied with the preservation of diaphragm neuromuscular innervation. However, it must be noted that this cervical ephrinB2 knockdown approach had no impact on disease onset, disease duration, or animal survival. Furthermore, there was no impact of ephrinB2 knockdown on forelimb or hindlimb function. This is an expected result, given the fairly focal approach of ephrinB2 knockdown in C3-C5 spinal segments.

The major limitation of the study is the conclusion that the preservation of diaphragm output following ephrinB2 knockdown in SOD1 mice is mediated primarily (if not entirely) by astrocytes. The authors present convincing evidence that a reduction in ephrinB2 is observed in local astrocytes (~56% transduction) following the intraspinal injection of the lentivirus. However, the proportion of cell types assessed for transduction with the lentivirus in the spinal cord was limited to neurons, astrocytes, and oligodendrocyte lineage cells. Microglia comprise a large proportion of the glial population in the spinal grey matter and have been shown to associate closely with respiratory motor pools. This cell type, amongst the many other that comprise the ventral gray matter, have not been investigated in this study. Nonetheless, there is convincing evidence to suggest astrocytes play a significant role, as compared to oligodendrocytes in promoting ALS pathogenesis.

In summary, this study by Urban et al. provides a valuable framework for Eph-Ephrin signaling mechanisms imposing pathological changes in an ALS mouse model. The role of glial cells in ALS pathology is a very exciting and upcoming field of investigation. The current study proposes a novel astrocyte-mediated mechanism for the propagation of disease that may eventually help to identify potential therapeutic targets.

eLife assessment

This is a valuable study of Eph-Ephrin signaling mechanisms generating pathological changes in amyotropic lateral sclerosis. There are exciting findings bearing on the role of glial cells in this pathology. The study emerges with solid evidence for a novel astrocyte-mediated mechanism for disease propagation. It may help identify potential therapeutic targets.

Reviewer #1 (Public Review):

In the manuscript by Urban et al., the authors attempt to further delineate the role with which non-neuronal CNS cells play in the development of ALS. Towards this goal, the transmembrane signaling molecule ephrinB2 was studied. It was found that there is an increased expression of ephrinB2 in astrocytes within the cervical ventral horn of the spinal cord in a rodent model of ALS. Moreover, reduction of ephrinB2 reduced motoneuron loss and prevented respiratory dysfunction at the NMJ. Further driving the importance of ephrinB2 is an increased expression in the spinal cords of human ALS individuals. Collectively, these findings present compelling evidence implicating ephrinB2 as a contributing factor towards the development of ALS.

Author Response

The following is the authors’ response to the original reviews.

We would like to thank the editors and reviewers for their overall positive assessment of this work. We have carefully revised the manuscript and implemented near all reviewers’ public and confidential recommendations. We believe these modifications have strengthened the manuscript and hope it will further convince the editors and reviewers.

We below provide a point-by-point response to the reviewers’ comments.

Reviewer #1 (Public Review):

To further understand the plasticity of vestibular compensation, Schenberg et al. sought to characterize the response of the vestibular system to short-term and partial impairment using gaze stabilization behaviors. A transient ototoxic protocol affected type I hair cells and produced gain changes in the vestibulo-ocular reflex and optokinetic response. Interestingly, decreases in vestibular function occurred in coordination with an increase in ocular reflex gain at frequencies where vestibular information is more highly weighted over visual. Moreover, computational approaches revealed unexpected detriment from low reproducibility on combined gaze responses. These results inform the current understanding of visual-vestibular integration especially in the face of dysfunction.

Strengths

The manuscript takes advantage of VOR measurements which can be activated by targeted organs, are used in many species including clinically, and indicate additional adverse effects of vestibular dysfunction. The authors use a variety of experimental procedures and analysis methods to verify results and consider individual performance effects on the population data. The conclusions are well-justified by current data and supported by previous research and theories of visuo-vestibular function and plasticity.

The authors thank reviewer 1 for emphasizing these positive aspects of the work.

Weaknesses

The manuscript describes the methodology as inducing reversible changes (lines 44, 67,) but the data shows a reversible effect only in hair cell histology (Fig 3A-B) not in function as demonstrated by the persistent aVOR gain reduction in week 12 (Fig 1C) and increase of OKR gain in weeks 6-12 (Fig 4C/D).

Rodents exposed to IDPN in the drinking water show from complete to null reversibility of the function loss depending on the IDPN concentration and duration of exposure, and the relationship between exposure and effect varies as a function of species, strain and sex of the exposed animals (Llorens and Rodríguez-Farré, Neurotoxicol. Teratol., 1997; Seoane et al., J. Comp. Neurol. 2001; Sedó-Cabezón et al., Dis. Model. Mech., 2015; Greguske et al., Arch. Toxicol., 2019). In addition, there is individual variability. The concentration of IDPN and time of exposure used in this study were selected to result in a loss followed by complete reversion but, as noted by the referee, the reversion was complete on Hair cells, while the gaze stabilizing reflexes show differential degrees of recovery depending on the functional tests (complete recovery on OCR; partial on aVOR and OKR). These make the IDPN subchronic protocol an interesting methodology to study the long term consequences of partial/reversible inner ear impairment. To be more accurate in the description of the reversibility, we have now introduced the following changes:

Lines 43: Subchronic exposure to IDPN in drinking water at low doses allowed for progressive ototoxicity, leading to a partial and largely reversible loss of function.

Lines 67-68: We demonstrate that despite the significant recovery in their vestibulo-ocular reflexes, the visuo-vestibular integration remains notably impaired in some IDPN-treated mice

Lines 578: Previous experiments (Greguske et al., 2021) had demonstrated that at these concentrations, ototoxic lesions produced by IDPN are largely reversible.

Reviewer 1: The manuscript begins with the mention of fluctuating vestibular function clinically, but does not connect this to any specific pathologies nor does it relate its conclusions back to this motivation.

Thank you. We have now added a conclusion (lines 525-552) to discuss the results in a clinical perspective.

Reviewer 1: The conclusions of frequency-specific changes in OKR would be stronger if frequency-specific aVOR effects were demonstrated similar to Figure 4D.

We have presented the frequency-selective effects in Figure 1 supplement and related text; changes observed in aVOR are mostly (see below) comparable for all frequencies >0.2Hz. However, we have edited the text to better highlight when the IDPN differentially affect aVOR tested at different frequencies (see lines 97-99).

Reviewer #2 (Public Review):

This is a very nice study showing how partial loss of vestibular function leads to long term alterations in behavioural responses of mice. Specifically, the authors show that VOR involving both canal and otolith afferents are strongly attenuated following treatment and partially recover. The main result is that loss of VOR is partially "compensated" by increased OKR in treated animals. Finally, the authors show that treatment primarily affects type I hair cells as opposed to type II. Overall, these results have potentially important implications for our understanding of how the VOR Is generated using input from both type I and type II hair cells. As detailed below however, more controls as well as analyses are needed.

The authors thank reviewer 2 for positive evaluation regarding the potential implication of the work.

Major points:

Reviewer 2: The authors analyze both canal and otolith contributions to the VOR which is great. There is however an asymmetry in the way that the results are presented in Figure 1. Please correct this and show time series of fixations for control and at W6 and W12. Moreover, the authors are plotting table and eye position traces in Fig. 1B but, based on the methods, gains are computed based on velocity. So please show eye velocity traces instead. Also, what was the goodness of fit of the model to the trace at W6? If lower than 0.5 then I think that it is misleading to show such a trace since there does not seem to be a significant VOR.

Figure 1 was modified as suggested. Panel B now shows velocity traces, and goodness of fit is reported in figure legend. Panel E now shows raw OCR traces at W0, W6, W12.

Reviewer 2: This is important to show that the loss is partial as opposed to total. It seems to me that the treatment was not effective at all for aVOR for at least some animals. What happens if these are not included in the analysis?

The reviewer is correct, there is indeed variability in the alteration observed during the treatment, as previously described and expected from previous experiments (Llorens and Rodríguez-Farré, Neurotoxicol. Teratol., 1997; Seoane et al., J. Comp. Neurol. 2001; SedóCabezón et al., Dis. Model. Mech., 2015; Greguske et al., Arch. Toxicol., 2019). It was actually one of the goal of the study to compare hair cell loss and VOR responses in individuals. The individual aVOR gain and phase responses during the IDPN treatment are all presented in Figure 1 supplement. aVOR was reduced in all animals, although 2/21 only showed a decrease of less than 10% of their initial gain at W6. If these were excluded from the analysis, the statistical differences between the 2 groups would be reinforced.

Reviewer 2: Figure 2A shows a parallel time course for gains of aVOR and OCR at the population level. Is this also seen at the individual level?

Yes, this is seen in individuals. This result is presented in Figure 2C and 2D which illustrate the similar effect of IDPN on aVOR and OCR responses at week 6 and week 12 at the individual level (each symbol represents an individual mouse). The plotted delta gain of both aVOR and OCR represents the relative loss of vestibular function for each individual mouse at W6 and W12, respectively.

Reviewer 2: Figure 3: please show individual datapoints in all conditions.

Figure 3 was modified to show individual datapoints in all conditions (see Figure 3 A2, A3, C2 and C3).

Reviewer 2: Figure 4: The authors show both gain and phase for OKR. Why not show gain and phase for aVOR and OCR in Figure 1. I realize that phase is shown in sup Figures but it is important to show in main figures. The authors show a significant increase in phase lead for aVOR but no further mention is made of this in the discussion. Moreover, how are the authors dealing with the fact that, as gain gets smaller, the error on the phase will increase. Specifically, what happens when the grey datapoints are not included?

As pointed by the reviewer, we have included all aVOR phase results in Figure 1 supplement and also stated it in the main text (lines 100-102). There is however no phase calculated for the OCR which is a static test, as better illustrated in new Figure 1E. Error in phase calculations increases as gain gets smaller. To take this into account, the phase corresponding to the grey points (VAF<0,5; corresponding to Gains<0.10) were not included in the statistical analysis of the aVOR phase. This point is now made clearer in methods lines 639-640.

Reviewer 2: Discussion: As mentioned above, the authors should discuss the mechanisms and implications of the observed phase lead following treatment. Moreover, recent literature showing that VN neurons that make the primary contribution to the VOR (i.e., PVP neurons) tend to show more regular resting discharges than other classes (i.e., EH cells), and that such regularity is needed for the VOR should be discussed (Mackrous et al. 2020 eLife). Specifically, how are type I and type II hair cells related to discharge regularity by central neurons in VN?

We have now added discussion regarding mechanisms and implications of the phase changes in lines 363-371. The authors thank reviewer 1 for pointing at the Mackrous et al. 2020 eLife paper which is now included in the updated discussion. The relations between type I and type II and discharge regularity in afferents and central VN is further discussed 442-449.

Below we provide answers to specific recommendations for the authors.

Reviewer #1 (Recommendations For The Authors):

Reviewer 1: Were hair cells counted for the whole organ? what was the control for epithelial size differences?

The effect of the treatment on hair cells was estimated by counting numbers of cells in square area of the central and peripheral parts of the sensory epithelia. The text has been modified to better describe the method, lines 748-751.

Reviewer 1: The title of the article leads readers to expect more emphasis on hair cell changes, while the content of the manuscript is more functional and encompassing the visual and vestibular systems.

We have retained the original tittle.

Reviewer 1: Please provide acronym definitions before they are used. Examples: HC (line 63), W6 etc (line 82-83)

Done as suggested on lines 63, 82 and 107.

Reviewer 1: Please describe the ages of animals used in the study.

The animals used in the study were 6 weeks old at the beginning of the protocol and 20 weeks old at the end. The text has been modified accordingly, line 564.

Reviewer 1: Consider changing "until" to "through" when describing time ranges (initially line 88), as the following time mentioned is included in the statement. E.g., line 216-217 sounds as if gain was insignificantly different at W12.

Done as suggested, lines 88 and 219.

Reviewer 1: Line 162: lower case for "immunostaining".

Done, line 164.

Reviewer 1: Consider regrouping or renumbering panels of Figure 3 for more clarity.

Panels in Figure 3 were regrouped as suggested, with first the canal-related data in panels A-B followed by the utricule-related data in panels C-D.

Reviewer 1: Lines 222-223: reword as gain increased not frequency.

Thank you, the text has been reworded, line 224-225.

Reviewer 1: It is unclear if the two subgroups revealed in CGR analysis (line 288) are relevant to the two groups described in VOR responses (line 137-138). Please clarify if these are the same mice or distinct clusters.

The two subgroups found in the CGR analysis differ from the clusters revealed by the decrease of the aVOR gain; the text has been modified lines 300-301.

Reviewer 1: Consider adding that irregular afferents + calyces are relevant specifically to type I HCs (lines 411-426).

The text has been modified to clarify the contacts between the two types of vestibular afferents and hair cells, lines 431-435.

Reviewer 1: Line 434: clarify which "scheme" given context before and after this phrase

In order to clarify this part of the discussion, the text has been modified and this term no longer appears.

Reviewer 1: Please indicate the time gap from surgery to treatment.

The time gap from the surgery to treatment, at least 72h, has been updated in the methods, lines 575.

Reviewer 1: Line 619-620: It is unclear if VOR and OKR sessions were randomized in order or if the authors have considered training or adaptive effects from the initial test session.

VOR and OKR sessions were performed on different days to limit cross effects, lines 639-640.

Reviewer 1: Line 688: typo-change ifG to IgG.

modified, line 744.

Reviewer 1: Line 692-693: were hair cells counted for the whole organ? what was the control for epithelial size differences?

The effect of the treatment on hair cells was estimated by counting numbers of cells in square area of the central and peripheral parts of the sensory epithelia. The text has been modified to better explain the method, lines 748-751.

Reviewer 1: Change decimal indicator to be consistent (commas used in lines 732, 759, 776, Figure 6C),

Thank you; modified as suggested.

Reviewer 1: Line 763: "stimulation at 0.5Hz &10{degree sign}/s" is unclear.

The text has been modified, line 817.

Reviewer 1: Line 765: bold (E)

The police format has been updated, line 820.

Reviewer 1: Line 770-771: A) insert OKR to be "mean delta aVOR and delta OKR gain", B) plot is OKR as a function of VOR.

Thank you, done as suggested. The text has been modified, line 824. Reviewer 1: Describe Figure 6 delta at initial mention (line 784 instead of 786) Authors: thank you, done as suggested, line 839.

Reviewer 1: It is unclear why the tables are included if never mentioned in the text.

The tables are now mentioned, lines 90 and 218.

Reviewer 1: Figure 1: is the observed gain for Sham group expected value rather than closer to 1?

Yes, as the value reported on Figure 1 is a mean of the values obtained during aVOR test in the dark at frequencies in range 0.2-1Hz (see also Figure 1 Supplement).

Reviewer 1: Figure 2: A) give enough space to see error bars at W2. Consider making test data more easily distinguishable. B) is OCR mean or specific stimulation? C/D) move 1Hz label from title to x-axis label as it does not describe OCR test. Figure 5: C) consider making color specific to frequency for better distinction on C+D as symbols previously indicated individual data. D) 1Hz specific to OKR? move to axis label instead of title

The Figures 2 and 5 have been modified according to reviewer 1 suggestions.

Reviewer 1: Figure 6: A/B) what time point are these, W12?

Those points correspond to W6 and W12, the text has been updated to specify the time points, lines 834 and 835.

Reviewer #2 (Recommendations For The Authors):

The authors should perform additional analyses that will help strengthen their results.

We are unsure about the exact implementation of this recommendation. However, we have strengthened our results by following all reviewers’ specific recommendations.

eLife assessment

This paper provides a fundamental expansion of vestibular compensation into transient and partial dysfunction, as well as insights into the adaptation of visual reflexes in this process. The conclusions are convincingly supported with paired histological and behavioral measurements, which are additionally modeled for further interpretation. This work would be of interest to neuroscientists working in multisensory integration and recovery mechanisms.

Reviewer #1 (Public Review):

To further understand the plasticity of vestibular compensation, Schenberg et al. sought to characterize the response of the vestibular system to short-term and partial impairment using gaze stabilization behaviors. A transient ototoxic protocol affected type I hair cells and produced gain changes in the vestibulo-ocular reflex and optokinetic response. Interestingly, decreases in vestibular function occurred in coordination with an increase in ocular reflex gain at frequencies where vestibular information is more highly weighted over visual. Moreover, computational approaches revealed unexpected detriment from low reproducibility on combined gaze responses. These results inform the current understanding of visual-vestibular integration especially in the face of dysfunction.

Strengths<br /> The manuscript takes advantage of VOR measurements that can be activated by targeted organs, are used in many species including clinically, and indicate additional adverse effects of vestibular dysfunction.

The authors use a variety of experimental procedures and analysis methods to verify results and consider individual performance effects on the population data.

The conclusions are well-justified by current data and supported by previous research and theories of visuo-vestibular function and plasticity.

Reviewer #2 (Public Review):

This is a very nice study showing how partial loss of vestibular function leads to long term alterations in behavioural responses of mice. Specifically, the authors show that VOR involving both canal and otolith afferents are strongly attenuated following treatment and partially recover. The main result is that loss of VOR is partially "compensated" by increased OKR in treated animals. Finally, the authors show that treatment primarily affects type I hair cells as opposed to type II hair cells. Overall, these results have important implications for our understanding of how the VOR Is generated using input from both type I and type II hair cells.

The major strength of the study lies in the use of partial inactivation of hair cells to look at the effects on behaviors such as VOR and OKR. Some weaknesses stem from the fact that the effects of inactivation are highly variable across specimens and that there is no recovery of behavioral function.

eLife assessment

This is a fundamental work that significantly advances our understanding of the role of mossy cells in the dentate gyrus in Fragile X Syndrome. Carefully designed experiments provide evidence that changes in their excitability occur due to up-regulation of Kv7 currents. While the evidence supporting the authors' conclusions is solid, some of their claims did not consider other potential factors and explanations. The work will be of interest to neuroscientists working on unveiling the mechanisms of Fragile X pathology.

Reviewer #1 (Public Review):

Summary:<br /> In this work, the authors provide evidence to show that an increase in Kv7 channels in hilar mossy cells of Fmr1 knock out mice results in a marked decrease in their excitability. The reduction in excitatory drive onto local hilar interneurons produces an increased excitation/inhibition ratio in granule cells. Inhibiting Kv7 channels can help normalize the excitatory drive in this circuit, suggesting that they may represent a viable target for targeted therapeutics for fragile-x syndrome.

Strengths:<br /> The work is supported by a compelling and thorough set of electrophysiological studies. The authors do an excellent job of analysing their data and present a very complete data set.

Weaknesses:<br /> There are no significant weaknesses in the experimental work, however the complexity of the data presentation and the lack of a schematic showing the organizational framework of this circuit make the data less accessible to non-experts in the field. I highly encourage a graphical abstract and network diagram to help individuals understand the implications of this work.

The work is important as it identifies a unique regional and cell-specific abnormality in Fmr1 KO mice, showing how the loss of one gene can result in region-specific changes in brain circuits.

Reviewer #3 (Public Review):

The paper by Deng, Kumar, Cavalli, Klyachko describes that, unlike in other cell types, loss of Fmr1 decreases the excitability of hippocampal mossy cells due to up-regulation of Kv7 currents. They also show evidence that while muting mossy cells appears to be a compensatory mechanism, it contributes to the higher activity of the dentate gyrus, because the removal of mossy cell output alleviates the inhibition of dentate principal cells. This may be important for the patho-mechanism in Fragile X syndrome caused by the loss of Fmr1.

These experiments were carefully designed, and the results are presented ‎in a very logical, insightful, and self-explanatory way. Therefore, this paper represents strong evidence for the claims of the authors. In the current state of the manuscript, there are only a few points that need additional explanation.

One of the results, which is shown in the supplementary dataset, does not fit the main conclusions. Changes in the mEPSC frequency suggest that in addition to the proposed network effects, there are additional changes in the synaptic machinery or synapse number that are independent of the actual activity of the neurons. Since the differences of the mEPSC and sEPSC frequencies are similar and because only the latter can signal network effects, while the former is typically interpreted as a presynaptic change, it cannot be claimed that sEPSC frequency changes are due to the hypo-excitability of mossy cells.

An apparent technical issue may imply a second weak point in the interpretation of the results. Because the IPSCs in the PP stimulation experiments (Fig 8) start within a few milliseconds, it is unlikely that its first ‎components originate from the PP-GC-MC-IN feedforward inhibitory circuit. The involvement of this circuit and MCs in the Kv7-dependent excitability changes is the main implication of the results of this paper. But this feedforward inhibition requires three consecutive synaptic steps and EPSP-AP couplings, each of them lasting for at least 1ms + 2-5ms. Therefore, the inhibition via the PP-GC-MC-IN circuit can be only seen from 10-20ms after PP stimulation. The earlier components of the cPSCs should originate from other circuit elements that are not related to the rest of the paper. Therefore, more isolated measurements on the cPSC recordings are needed ‎which consider only the later phase of the IPSCs. This can be either a measurement of the decay phase or a pharmacological manipulation that selectively enhances/inhibits a specific component of the proposed circuit.

I suggest refraining from the conclusions saying "‎MCs provide at least ~51% of the excitatory drive onto interneurons in WT and ~41% in KO mice", because too many factors (eg. IN cell types, slice condition, synaptic reliability) are not accounted for in these actual numbers, and these values are not necessary for the general observation of the paper.

There are additional minor issues about the presentation of the results.

Author Response

Reviewer #1 (Public Review):

Assessment:

The manuscript titled 'Rab7 dependent regulation of goblet cell protein CLCA1 modulates gastrointestinal 1 homeostasis' by Gaur et al discusses the role of Rab7 in the development of ulcerative colitis by regulating the lysosomal degradation of Clca1, a mucin protease. The manuscript presents interesting data and provides a potential molecular mechanism for the pathological alterations observed in ulcerative colitis. Gaur et al demonstrate that Rab7 levels are lowered in UC and CD. However, a similar analysis of Rab7 levels in ulcerative colitis (UC) and Crohn's disease (CD) patient samples was conducted recently (Du et al, Dev Cell, 2020) which showed that Rab7 levels are found to be elevated under these conditions. While Gaur et al have briefly mentioned Du et al's paper in passing in the discussion, they need to discuss these contradictory results in their paper and clarify these differences. Additionally, Du et al are not included in the list of references.

Strengths:

The manuscript used a multi-pronged approach and compares patient samples, mouse models of DSS, and protocols that allow differentiation of goblet cells. They also use a nanogel-based delivery system for siRNAs, which is ideal for the knockdown of specific genes in the gut.

Weaknesses:

Du et al, Dev Cell 2020 (https://doi.org/10.1016/j.devcel.2020.03.002) have previously shown that Rab7 levels are elevated in a similar set of colonic samples (age group, number etc) from UC and CD patients. Gaur et al have not discussed this paper or its findings in detail, which directly contradicts their results. Clarification regarding this should be provided.

We thank and appreciate the reviewer for bringing this point.

The results shown by Du et al, Dev Cell, 2020 depict elevated expression of Rab7 in UC and CD patients compared to controls. In first occurrence, these results appear contradictory, but there may be a few possible explanations for this.

Firstly, Rab7 expression levels may fluctuate in the tissue depending on the degree of the gut inflammation. This can be concluded from our observations in DSS-mice dynamics model and the human patient samples with mild and moderate UC. Furthermore, Du et al provide no information of the severity of the condition among the patients employed in the study. Our motive, in the current work, was to emphasise this aspect. This point was mentioned in the discussion section of the manuscript. However, in view of the reviewer’s concern, we now intend to add a detailed comment on this in the main text of the revised version of the manuscript.

Secondly, the control biopsies in our investigation were acquired from non-IBD patients, and not what was done by Du et al., wherein biopsies from the normal para-carcinoma region of the colorectal cancer patients was used. One can not overlook the fact that physiological and molecular changes are apparent even in non-inflamed regions in the gut of an IBD or CRC patient. It is possible that the observed discrepancy arises due to the differences in the sample type used for comparing the Rab7 expression.

Finally, the main sub-tissue region showing a decrease in Rab7 expression in UC samples, appeared to be the Goblet cells which was not covered by Du et al.

Keeping these points in mind we do not think that there is a contradiction in our findings with that of Du et al., 2020. In the revised submission some of these explanations will be incorporated. Include Du et al in the reference list and add the comment in main text.

This was an oversight from our side. We have actually mentioned Du et al., 2020 in the discussion (line number 338) but somehow the reference was missing in the main list. We will ensure that the reference is included in the revised version and that their findings are included both in main text and in the discussion.

Reviewer #2 (Public Review):

Summary:

In this work, the authors report a role for the well-studied GTPase Rab7 in gut homeostasis. The study combines cell culture experiments with mouse models and human ulcerative colitis patient tissues to propose a model where, Rab7 by delivering a key mucous component CLCA1 to lysosomes, regulates its secretion in the goblet cells. This is important for the maintenance of mucous permeability and gut microbiota composition. In the absence of Rab7, CLCA1 protein levels are higher in tissues as well as the mucus layer, corroborating with the anticorrelation of Rab7 (reduced) and CLCA1 (increased) from ulcerative colitis patients. The authors conclude that Rab7 maintains CLCA1 level by controlling its lysosomal degradation, thereby playing a vital role in mucous composition, colon integrity, and gut homeostasis.

Strengths:

The biggest strength of this manuscript is the combination of cell culture, mouse model, and human tissues. The experiments are largely well done and in most cases, the results support their conclusions. The authors go to substantial lengths to find a link, such as alteration in microbiota, or mucus proteomics.

Weaknesses:

There are also some weaknesses that need to be addressed. The association of Rab7 with UC in both mice and humans is clear, however, claims on the underlying mechanisms are less clear. Does Rab7 regulate specifically CLCA1 delivery to lysosomes, or is it an outcome of a generic trafficking defect? CLCA1 is a secretory protein, how does it get routed to lysosomes, i.e. through Golgi-derived vesicles, or by endocytosis of mucous components? Mechanistic details on how CLCA1 is routed to lysosomes will add substantial value.

We thank the reviewer for the insightful comment. We would like to bring forth the following explanation for each these concerns:

(a) Our immunofluorescence imaging experiments revealed co-localization of Rab7 protein with CLCA1 and the lysosomes (Fig 7I). In addition, the absence of Rab7 affects the transport of CLCA1 to lysosomes (Fig 7J). This demonstrates that Rab7 may be involved in regulation of CLCA1 transport (presumably along with other cargo), to lysosomes selectively. However, we do recognise that the point raised by the reviewer about possible effect of a generic trafficking defect is valid. (b) As mentioned in the manuscript, the trafficking of CLCA1 protein or CLCA1-containing vesicles within the goblet cell is unknown, with no information on the proteins involved in its mobility. The switching of CLCA1 containing vesicles from the secretory route to lysosomes needs extensive investigation involving overall trafficking of the protein. Taken together, the complete answer to both these important questions will need a series of experiments and those may be interesting avenues for future research.

(a) Why does the level of Rab7 fluctuate during DSS treatment (Fig 1B)? (b) Does the reduction seen in Rab7 levels (by WB) also reflect in reduced Rab7 endosome numbers?

This is a very thoughtful point from the reviewer. We detected a distinct pattern of Rab7 expression fluctuation in intestinal epithelial cells after DSS-dynamics treatment in mice. Perhaps, these changes are the result of complex cellular signalling in response to the DSS treatment. Rab7, being a fundamental protein involved in protein sorting pathway, is expected to undergo alteration based on cells requirement. Presently there are no reports suggesting the regulatory mechanisms that govern Rab7 levels in the gut. (b) We observed reduction in Rab7 expression both at RNA and protein levels. To confirm whether this alteration will lead to reduced Rab7 positive endosome numbers may require detailed investigations.

Are other late endosomal (and lysosomal) populations also reduced upon DSS treatment and UC? Is there a general defect in lysosomal function?

There are no direct evidences showing reduction in the late endosomal and lysosomal population during gut inflammation, but few studies link lysosomal dysfunction with risk for colitis (doi: 10.1016/j.immuni.2016.05.007).

The evidence for lysosomal delivery of CLCA1 (Fig 7 I, J) is weak. Although used sometimes in combination with antibodies, lysotracker red is not well compatible with permeabilization and immunofluorescence staining. The authors can substantiate this result further using lysosomal antibodies such as Lamp1 and Lamp2. For Fig 7J, it will be good to see a reduction in Rab7 levels upon KD in the same cell.

We used Lysotracker red in live cells followed by fixation. So, permeabilization issues were resolved. Lamp1, as suggested by the reviewer, is definitely a better marker for lysosomes in immunofluorescence studies, but is also shown to mark late endosomes (doi: 10.1083/jcb.132.4.565). As Rab7 protein also marks the late endosomes, using Lamp1 may leave the ambiguity of CLCA1 in Rab7 positive late endosomes versus lysosomes. Nevertheless, we will be carrying out this experiment and the data will be shared in revised version of the work.

In this connection, Fig S3D is somewhat confusing. While it is clear that the pattern of Muc2 in WT and Rab7-/- cells are different, how this corroborates with the in vivo data on alterations in mucus layer permeability -- as claimed -- is not clear.

The data in Fig. S3D suggest the involvement of Rab7 in packaging of Muc2. The whole idea for doing this experiment was to support our observation in the Rab7KD-mice model where mucus layer was seen to be loose and more permeable in Rab7 deficient mice.

Overall, the work shows a role for a well-studied GTPase, Rab7, in gut homeostasis. This is an important finding and could provide scope and testable hypotheses for future studies aimed at understanding in detail the mechanisms involved.

We thank the reviewer for this comment.

eLife assessment

The study provides important insights into the mechanisms underlying ulcerative colitis, a chronic and debilitating gastrointestinal condition. The article provides solid evidence of the role of the vesicular trafficking protein Rab7 in regulating the colonic mucus system and its implications in ulcerative colitis.

Reviewer #1 (Public Review):

Assessment:

The manuscript titled 'Rab7 dependent regulation of goblet cell protein CLCA1 modulates gastrointestinal 1 homeostasis' by Gaur et al discusses the role of Rab7 in the development of ulcerative colitis by regulating the lysosomal degradation of Clca1, a mucin protease. The manuscript presents interesting data and provides a potential molecular mechanism for the pathological alterations observed in ulcerative colitis. Gaur et al demonstrate that Rab7 levels are lowered in UC and CD. However, a similar analysis of Rab7 levels in ulcerative colitis (UC) and Crohn's disease (CD) patient samples was conducted recently (Du et al, Dev Cell, 2020) which showed that Rab7 levels are found to be elevated under these conditions. While Gaur et al have briefly mentioned Du et al's paper in passing in the discussion, they need to discuss these contradictory results in their paper and clarify these differences. Additionally, Du et al are not included in the list of references.

Strengths:

The manuscript used a multi-pronged approach and compares patient samples, mouse models of DSS, and protocols that allow differentiation of goblet cells. They also use a nanogel-based delivery system for siRNAs, which is ideal for the knockdown of specific genes in the gut.

Weaknesses:

Du et al, Dev Cell 2020 (https://doi.org/10.1016/j.devcel.2020.03.002) have previously shown that Rab7 levels are elevated in a similar set of colonic samples (age group, number etc) from UC and CD patients. Gaur et al have not discussed this paper or its findings in detail, which directly contradicts their results. Clarification regarding this should be provided.

Reviewer #2 (Public Review):

Summary:

In this work, the authors report a role for the well-studied GTPase Rab7 in gut homeostasis. The study combines cell culture experiments with mouse models and human ulcerative colitis patient tissues to propose a model where, Rab7 by delivering a key mucous component CLCA1 to lysosomes, regulates its secretion in the goblet cells. This is important for the maintenance of mucous permeability and gut microbiota composition. In the absence of Rab7, CLCA1 protein levels are higher in tissues as well as the mucus layer, corroborating with the anti-correlation of Rab7 (reduced) and CLCA1 (increased) from ulcerative colitis patients. The authors conclude that Rab7 maintains CLCA1 level by controlling its lysosomal degradation, thereby playing a vital role in mucous composition, colon integrity, and gut homeostasis.

Strengths:

The biggest strength of this manuscript is the combination of cell culture, mouse model, and human tissues. The experiments are largely well done and in most cases, the results support their conclusions. The authors go to substantial lengths to find a link, such as alteration in microbiota, or mucus proteomics.

Weaknesses:

There are also some weaknesses that need to be addressed. The association of Rab7 with UC in both mice and humans is clear, however, claims on the underlying mechanisms are less clear. Does Rab7 regulate specifically CLCA1 delivery to lysosomes, or is it an outcome of a generic trafficking defect? CLCA1 is a secretory protein, how does it get routed to lysosomes, i.e. through Golgi-derived vesicles, or by endocytosis of mucous components? Mechanistic details on how CLCA1 is routed to lysosomes will add substantial value.

Why does the level of Rab7 fluctuate during DSS treatment (Fig 1B)? Does the reduction seen in Rab7 levels (by WB) also reflect in reduced Rab7 endosome numbers? Are other late endosomal (and lysosomal) populations also reduced upon DSS treatment and UC? Is there a general defect in lysosomal function?

The evidence for lysosomal delivery of CLCA1 (Fig 7 I, J) is weak. Although used sometimes in combination with antibodies, lysotracker red is not well compatible with permeabilization and immunofluorescence staining. The authors can substantiate this result further using lysosomal antibodies such as Lamp1 and Lamp2. For Fig 7J, it will be good to see a reduction in Rab7 levels upon KD in the same cell. In this connection, Fig S3D is somewhat confusing. While it is clear that the pattern of Muc2 in WT and Rab7-/- cells are different, how this corroborates with the in vivo data on alterations in mucus layer permeability -- as claimed -- is not clear.

Overall, the work shows a role for a well-studied GTPase, Rab7, in gut homeostasis. This is an important finding and could provide scope and testable hypotheses for future studies aimed at understanding in detail the mechanisms involved.

eLife assessment

This study presents useful findings from a large sample of participants from the UK Biobank on the relationship between menopause (including status, type, and age of onset), cognition, neuroanatomical measures derived from magnetic resonance imaging, and Alzheimer's disease. The strength of evidence is incomplete, and the study would benefit from clearer methodological descriptions, more careful consideration of potential confounds, and better theoretical integration with prior work in the field. This paper will be of interest to people working in the fields of cognitive neuroscience, endocrinology, and dementia.

Reviewer #1 (Public Review):

Summary:<br /> Costantino et al report on data from thousands of participants from the UK Biobank whereby they assessed relationships between menopausal status, menopause type (surgical or natural), and age at menopause with cognition, neuroanatomical measures derived from magnetic resonance imaging and Alzheimer's disease (AD) risk.

Strengths:<br /> This is a really important field of research. Alzheimer's disease is a leading cause of death in women and better understanding whether hormonal and brain changes associated with the menopause transition are contributing to this risk is a crucial research question. Access to such a large database, with cognitive assessment alongside structural MRI data, is a strength of this study. The authors report a positive association between earlier age of menopause as well as surgical menopause and a higher risk of developing AD. The authors also report associations between age at natural menopause and performances on various cognitive tests. Positive associations were found between the age of menopause and fluid intelligence, numeric memory, and pair matching.

Weaknesses:<br /> The manuscript would benefit from further clarification about the sample and descriptions of analyses. At the moment, it is difficult to determine whether the conclusions align with the results. In terms of the method, this is a cross-sectional analysis, with different subgroups selected depending on the research question and model. Some further clarification on the full sample and the participants selected for each analysis would be helpful. Some clarification on how menopause status and AD diagnosis were determined would be helpful. The results and discussion refer to menopause having an impact on specific cognitive tasks - the domains that these tasks assess would be worthy of some discussion.

Reviewer #2 (Public Review):

Summary:

Using in vitro and in vivo approaches, the authors first demonstrate that BEST4 inhibits intestinal tumor cell growth and reduces their metastatic potential, possibly via downstream regulation of TWIST1.

They further show that HES4 positively upregulates BEST4 expression, with direct interaction with BEST4 promoter region and protein. The authors further expand on this with results showing that negative regulation of TWIST1 by HES4 requires BEST4 protein, with BEST4 required for TWIST1 association with HES4. Reduction of BEST1 expression was shown in CRC tumor samples, with correlation of BEST4 mRNA levels with different clinicopathological factors such as sex, tumor stage, and lymph node metastasis, suggesting a tumor-suppressive role of BEST4 for intestinal cancer.

Strengths:

• Good quality western blot data.<br /> • Multiple approaches were used to validate the findings.<br /> • Logical experimental progression for readability.<br /> • Human patient data / In vivo murine model / Multiple cell lines were used, which supports translatability / reproducibility of findings.

Weaknesses:

• Interpretation of figures and data (unsubstantiated conclusions).<br /> • Figure quality.<br /> • Figure legends lack information.<br /> • Lacking/shallow discussion.<br /> • Requires more information for reproducibility regarding materials and methods.

eLife assessment

The findings of this valuable manuscript advance our understanding of the significance of Bestrophin isoform 4 (BEST4) in suppressing colorectal cancer (CRC) progression. The authors used appropriate and validated methodology, such as the knockout of BEST4 using CRISPR/Cas9 in CRC cells, to provide a solid foundation for elucidating the potential link between BEST4 and CRC progression.

Reviewer #1 (Public Review):

Summary:

In this study, the authors describe the participation of the Hes4-BEST4-Twist axis in controlling the process of epithelial-mesenchymal transition (EMT) and the advancement of colorectal cancers (CRC). They assert that this axis diminishes the EMT capabilities of CRC cells through a variety of molecular mechanisms. Additionally, they propose that reduced BEST4 expression within tumor cells might serve as an indicator of an adverse prognosis for individuals with CRC.

Strengths:

• Exploring the correlation between the Hes4-BEST4-Twist axis, EMT, and the advancement of CRC is a novel perspective and gives readers a fresh standpoint.<br /> • The whole transcriptome sequence analysis (Figure 5) showing low expression of BEST4 in CRC samples will be of broad interest to cancer specialists as well as cell biologists although further corroborative data is essential to strengthen these findings (See Weaknesses).

Weaknesses:

• The authors employed three kinds of CRC cell lines, but not untransformed cells such as intestinal epithelial organoids which are commonly used in recent research.<br /> • The authors use three different human CRC cell lines with a lack of consistency in the selection of them. Please clarify 1) how these lines are different from each other, 2) why they pick up one or two of them for each experiment. To be more convincing, at least two lines should be employed for each in vitro experiment.<br /> • The authors demonstrated associations between BEST4 and cell proliferation/viability as well as migration/invasion, utilizing CRC cell lines, but it should be noted that these findings do not indicate a tumor-suppressive role of BEST4 as mentioned in line 120. Furthermore, while the authors propose that "BEST4 functions as a tumor suppressor in CRC" in line 50, there seems no supporting data to suggest BEST4 as a tumor suppressor gene.<br /> • The HES4-BEST4-Twist1 axis likely plays a significant role in CRC progression via EMT but not CRC initiation. Some sentences could lead to a misunderstanding that the axis is important for CRC initiation.<br /> • The authors mostly focus on the relationship of the HES4-BEST4-Twist1 axis with EMT, but their claims sometimes appear to deviate from this focus.<br /> • Some experiments do not appear to have a direct relevance to their claims. For example, the analysis using the xenograft model in Figure 2E-J is not optimal for analyzing EMT. The authors should analyze metastatic or invasive properties of the transplanted tumors if they intend to provide some supporting evidence for their claims.<br /> • In Figure 4H, ZO-1 and E-cad expression looks unchanged in the BEST4 KD.<br /> • The in vivo and in vitro data supporting the whole transcriptome sequence analysis (Figure 5) is mostly insufficient. Including the following experiments will substantiate their claims: 1) BEST4 and HES4 immunostaining of human surgical tissue samples, 2) qPCR data of HES4, Twist1, Vimentin, etc. as shown in Figure 5C, 5D.<br /> • Some statements are inconsistent probably due to grammatical errors. (For example, some High/low may be reversed in lines 234-244.)

Author Response

The following is the authors’ response to the original reviews.

eLife assessment

This study and associated data is compelling, novel, important, and well-carried out. The study demonstrates a novel finding that different chemotherapeutic agents can induce nucleolar stress, which manifests with varying cellular and molecular characteristics. The study also proposes a mechanism for how a novel type of nucleolar stress driven by CDK inhibitors may be regulated. The study sheds light on the importance of nucleolar stress in defining the on-target and offtarget effects of chemotherapy in normal and cancer cells.

We are thankful to the reviewers and the editor for their feedback and thorough assessment of our work. Our responses to the comments and suggestions are below.

Reviewer #1 (Public Review):

The study titled "Distinct states of nucleolar stress induced by anti-cancer drugs" by Potapova and colleagues demonstrates that different chemotherapeutic agents can induce nucleolar stress, which manifests with varying cellular and molecular characteristics. The study also proposes a mechanism for how a novel type of nucleolar stress driven by CDK inhibitors may be regulated. As a reviewer, I appreciate the unbiased screening approach and I am enthusiastic about the novel insights into cell biology and the implications for cancer research and treatment. The study has several significant strengths: i) it highlights the understudied role of nucleolar stress in the on- and off-target effects of chemotherapy; ii) it defines novel molecular and cellular characteristics of the different types of nucleolar stress phenotypes; iii) it proposes novel modes of action for well-known drugs. However, there are several important points that should be addressed:

• The rationale behind choosing RPE cells for the screen is unclear. It might be more informative to use cancer cells to study the effects of chemotherapeutic agents. Alternatively, were RPE cells selected to evaluate the side effects of these agents on normal cells? Clarifying these points in the introduction and discussion would guide the reader.

RPE1, a non-cancer-derived cell line, was chosen for this study to evaluate the effects of anticancer drugs on normal nucleolar function, with the underlying premise that nucleolar stress in normal cells can contribute to non-specific toxicity. This clarification is added to the introduction. Another factor that played in selecting a normal cell line for the drug screen and subsequent experiments was the spectrum of known and unknown genetic and metabolic alterations present in various cancer cell lines. These variables are often unique to a particular cancer cell line and may or may not impact nucleolar proteome and function. Therefore, the nucleolar stress response can be influenced by the spectrum of alterations inherent to each cancer. Our primary focus was to determine the impact of these drugs under normal conditions.

That said, the selected hits of main drug classes were validated in a panel of cell lines that included two other hTERT lines (BJ5TA and CHON-002) and two cancer lines (DLD1 and HCT116). In cancer cells starting nucleolar normality scores were lower than in hTERT cells, suggesting that genetic and metabolic changes in these cells may indeed affect nucleolar morphology. Nonetheless, all drugs from a panel of selected hits from different target classes validated in both cancer cell lines (Fig. 2F).

• Figure 2F indicates that DLD1 and HCT116 cells are less sensitive to nucleolar changes induced by several inhibitors, including CDK inhibitors. It would be crucial to correlate these differences with cell viability. Are these differences due to cell-type sensitivity or variations in intracellular drug levels? Assessing cell viability and intracellular drug concentration for the same drugs and cells would provide valuable insights.

One of the reasons for the reduced magnitude of the effects of selected drugs in DLD1 and HCT116 cells is their lower baseline normality scores compared to hTERT cells (now shown in Sup. Fig. 1B-C). Other potential factors include proteomic and metabolic shifts and alterations in signaling pathways that control ribosome production. The less-likely possibility of variations in intracellular drug levels cannot be excluded, but measuring this for every compound in every cell line was not feasible in this study. These limitations are now noted in the results section.

Regarding the point about viability - our initial screen output, in addition to normality scores, included cell count (cumulative count of cells in all imaged fields), which serves as a proxy for viability. By this measure, all hit compounds in our screen were cytostatic or cytotoxic in RPE1 cells (Fig. 2C). The impact of these drugs on the viability of cancer cells that can have various degrees of addiction to ribosome biogenesis merits a separate study of a large cancer cell line panel.

• Have the authors interpreted nucleolar stress as the primary cause of cell death induced by these drugs? When cells treated with CDK inhibitors exhibit the dissociated nucleoli phenotype, is this effect reversible? Is this phenotype indicative of cell death commitment? Conducting a washout experiment to measure the recovery of nucleolar function and cell viability would address these questions.

Whether nucleolar toxicity is the primary cause of cytotoxicity for a given chemotherapy drug is an incisive and thought-provoking question. Our screen did not discern whether the cytotoxic effects of our hits were due to inhibition of their intended targets, their impact on the nucleolus, or a combined effect. This point is now mentioned in the results section. Regarding the reversibility of the nucleolar disassembly phenotype seen in CDK inhibitors –in the case of flavopiridol, which is a reversible CDK inhibitor, we demonstrated that nucleoli re-assembled within 4-6 hours after the drug was washed out. An example of this is shown in Sup. Figure 3 and in Video 5. For these experiments, cells were pretreated with the drug for 5 hours, not long enough to cause cell death.

• The correlation between the loss of Treacle phosphorylation and nucleolar stress upon CDK inhibition is intriguing. However, it remains unclear how these two events are related. Would Treacle knockdown yield the same nucleolar phenotype as CDK inhibition? Moreover, would point mutations that abolish Treacle phosphorylation prevent its interaction with Pol-I? Experiments addressing these questions would enhance our understanding of the correlation/causation between Treacle phosphorylation and the effects of CDK inhibition on nucleolar stress.

We agree that the Treacle finding is interesting and warrants further investigation. In our attempts to knock down Treacle with siRNA, its protein levels were reduced by no more than 50%, which was not sufficient to cause a strong nucleolar stress response. Therefore, these data were not incorporated into the manuscript. However, in our view, Treacle is unlikely to be the only nucleolar CDK substrate whose dephosphorylation is causing the “bare scaffold” phenotype caused by the transcriptional CDK inhibitors. Our phospho-proteomics studies identified multiple nucleolar CDK substrates with established roles in the formation of the nucleolus. For instance, the granular component protein Ki-67 was also dephosphorylated on multiple sites and dispersed throughout the nucleus (shown in Sup. Fig 4). Given that CDKs typically phosphorylate many substrates that can have multiple phosphorylation sites, identifying a sole protein or phosphorylation site responsible for nucleolar disassembly may be an unattainable target.

Overall, this study is significant and novel as it sheds light on the importance of nucleolar stress in defining the on-target and off-target effects of chemotherapy in normal and cancer cells.

Thank you, we appreciate the positive and constructive assessment of our study.

Reviewer #2 (Public Review):

This is an interesting study with high-quality imaging and quantitative data. The authors devise a robust quantitative parameter that is easily applicable to any experimental system. The drug screen data can potentially be helpful to the wider community studying nucleolar architecture and the effects of chemotherapy drugs. Additionally, the authors find Treacle phosphorylation as a potential link between CDK9 inhibition, rDNA transcription, and nucleolar stress. Therefore I think this would be of broad interest to researchers studying transcription, CDKs, nucleolus, and chemotherapy drug mechanisms. However, the study has several weaknesses in its current form as outlined below.

1) Overall the study seems to suffer from a lack of focus. At first, it feels like a descriptive study aimed at characterizing the effect of chemotherapy drugs on the nucleolar state. But then the authors dive into the mechanism of CDK inhibition and then suddenly switch to studying biophysical properties of nucleolus using NPM1. Figure 6 does not enhance the story in any way; on the contrary, the findings from Fig. 6 are inconclusive and therefore could lead to some confusion.

This study was specifically designed to examine a broad range of chemotherapy drugs. The newly created nucleolar normality score enabled us to measure nucleolar stress precisely and in high throughput. Our primary objective was to find drugs that disrupt the normal nucleolar morphology and then study in-depth the most interesting and novel hits. We have made revisions to emphasize that these are the primary focal points of the manuscript.

As context, we were motivated to explore the biophysical properties of the nucleolus because they are thought to underlie its formation and function, which also suggested a potential predictive value for modeling nucleolar responses to drug treatments. For this, we edited the RPE1 cell line by endogenously tagging NPM1, a granular component protein that behaves in line with the phase-separation paradigm in vitro and when over-expressed. We fully expected to confirm that its behavior in vivo would be consistent with LLPS, but instead found that even in an untreated scenario, the dynamics of endogenous NPM1 could not be fully explained by the phase separation theory (Fig. 6 A-C). Our message is that accurately predicting drug responses using the nucleolar normality score as a readout, based on our current understanding of the biophysical forces governing nucleolar assembly, is unworkable. For instance, normality scores decrease and NPM1 dynamics increase radically when CDKs are inhibited, without changes in NPM1 concentration or concentrations of other protein components (Fig.6 E-H). These observations are important because they highlight our gaps in understanding the relative contribution of phase separation versus active assembly in nucleolar formation. We believe that these observations are worth sharing with the scientific community.

2) The justification for pursuing CDK inhibitors is not clear. Some of the top hits in the screen were mTOR, PI3K, HSP90, Topoisomerases, but the authors fail to properly justify why they chose CDKi over other inhibitors.

We decided to focus on CDK inhibitors for several reasons. First, their effects were completely new and unexpected, suggesting the existence of an unknown mechanism regulating nucleolar structure and function. In addition, CDK inhibitors caused a very strong and distinct nucleolar stress phenotype with the lowest normality scores that merited its own term, the “bare scaffold” phenotype. One more reason for pursuing CDK-inhibiting drugs was their high rate of failure in clinics because of the intense and hard-to-explain toxicity. We suspect that this toxicity may be due at least in part to their profound effect on nucleolar organization and ribosome production throughout the body. We stated this rationale more explicitly in the manuscript.

3) In addition to poor justification, it seems like a very superficial attempt at deciphering the mechanism of CDK9imediated nucleolar stress. I think the most interesting part of the study is the link between CDK9, Pol I transcription, and nucleolar stress. But the data presented is not entirely convincing. There are several important controls missing as detailed below.

We agree with the reviewer that follow-up studies of CDK9, Pol I, and nucleolar stress connection are important long-term goals. However, the primary objective of this study was to ascertain the scope of anticancer agents that can cause nucleolar stress and the establishment of nucleolar stress categories. This is an important advance and could serve as the foundation for a standalone in-depth study or multiple studies. We have included the complete screen, proteomics, and phospho-proteomics results (Sup. Tables 1, 2, and 3), which will enable other investigators to mine the screen information based on their specific interests. Furthermore, we have made multiple text revisions to clarify rationale and interpretation, and incorporated additional data that strengthen the manuscript.

4) The authors did not test if inhibition of CDK7 and/or CDK12 also induces nucleolar stress. CDK7 and CDK12 are also major kinases of RNAPII CTD, just like CDK9. Importantly, there are well-established inhibitors against both these kinases. It is not clear from the text whether these inhibitors were included in the screen library.

Our anticancer compound library contained CDK7 inhibitor THZ1⦁2HCL, and it was a hit at both 1 and 10 uM concentrations (Sup. Table 1). However, its nucleolar stress phenotype was morphologically distinct from CDK9 inhibitors, resembling the stress caps phenotype instead of the bare scaffold phenotype. We did not pursue CDK7 because of its two hard-to-separate functions: in addition to its role as an RNAPII CTD kinase, it also acts as a CDK-activating kinase (CAK) by promoting the associations of multiple CDKs with their cyclin partners. This dual role of CDK7 makes the interpretation of THZ1-induced nucleolar stress phenotype difficult because it could be attributed to either or both of these functions. Moreover, it was reported to cause DNA damage, which may explain why it causes stress caps. An image depicting nucleolar stress phenotype caused by THZ1⦁2HCL is provided in Author response image 1.

Author response image 1.

Control and THZ1 - treated RPE1 cells, images from screen plates.

We are not aware of specific inhibitors of CDK12, as they also reportedly inhibit CDK13. None of the CDK12/CDK13 inhibitors were present in our library, therefore we can neither confirm nor exclude the possible involvement of these kinases in regulating nucleolar structure. Many other existing CDK inhibitors were absent from our library. Our work highlights the importance of assessing their potential to induce nucleolar stress and offers an approach for this assessment.

5) In Figure 4E, the authors show that Pol I is reduced in nucleolus/on rDNA. The authors should include an orthogonal method like chromatin fractionation and/or ChIP

We acknowledge the reviewer’s request for additional validation of reduced occupancy of rDNA by Pol I.<br /> Nucleolar chromatin fractionation in cells treated with CDK inhibitors is unlikely to work due to nearly complete nucleolar disassembly. Chromatin immunoprecipitation would require finding and validating a suitable ChIP-grade antibody. Moreover, the evaluation of repetitive regions by ChIP is non-trivial and error-prone. To help address this request and further confirm the POLR1A immunofluorescence results in 4E, we included additional immunofluorescence data obtained with a different POLR1A antibody (Sup. Fig. 3D), and the results were similar.

6) In Fig. 5D, in vitro kinase lacks important controls. The authors should include S to A mutants of Treacle S1299A/S1301A to demonstrate that CDK9 phosphorylates these two residues specifically.

7) To support their model, the authors should test if overexpression of Treacle mutants S1299A/S1301A can partially phenocopy the nucleolar stress seen upon CDK9 inhibition. This would considerably strengthen the author's claim that reduced Treacle phosphorylation leads to Pol I disassociation from rDNA and consequently leads to nucleolar stress.

8) Additionally, it would be interesting if S1299D/S1301D mutants could partially rescue CDK9 inhibition.

Points (6-8):

We reiterate that transcriptional CDKs target multiple nucleolar proteins, and the observed phenotype might be due to the combined effects of de-phosphorylation of multiple substrates. We concur that deconstructing the role of Treacle phosphorylation sites is very interesting and warrants further in-depth studies. The phospho-proteomics enrichment method, while an effective first-pass strategy, might not capture 100% of the phosphorylated sites. Treacle is a phospho-protein with an abundance of serine and threonine residues. It could potentially have been selectively dephosphorylated on more sites than were detected by this method. Therefore, the suggested mutations may not be the exclusive contributors responsible for the functional phenotype. Additionally, overexpressing Treacle impairs the viability of RPE1 cells, complicating the interpretation of experiments involving overexpression of both wild-type and mutant proteins. A conceivable strategy would involve generating phosphomimetic and non-phosphorylatable mutants by gene editing, studying their interactions by biochemical approaches, and determining their impact on nucleolar function, but this may take years of additional work. We hope that our work will inspire further studies that explore Treacle phosphorylation and other functions of transcriptional CDKs in nucleolar formation.

Thank you for the thoughtful review and suggestions.

Reviewer #2 (Recommendations For The Authors):

1) The manuscript could be re-organized to focus on 'CDK9-Treacle-Pol I-nucleolar stress' as the central part of the story.

While we acknowledge this suggestion, it's important to emphasize that the primary focus of this manuscript is on the identification of anticancer drugs that induce nucleolar stress and the establishment of nucleolar stress categories.

2) Include a "no ATP" control in the in vitro kinase assay and indicate molecular sizes.

We provided an additional kinase assay (Sup. Fig. 4B) that includes no ATP control lanes and a fragment of a Coomassie blue stained gel showing molecular weight markers. No ATP control assays (lanes 4 and 5) were blank as expected. Molecular weight markers were added to all other kinase assays based on the known sizes of isolated Pol II holoenzyme subunits Rbp1 (191 kDa) and Rbp2 (138 kDa).

3) For in vitro phosphorylation, please provide an explanation for using CDK9/cyclin K instead of Cyclin T1 which is the predominant cyclin for CDK9

Recombinant CDK9/cyclin K complex was used for in vitro kinase assays for a technical reason: CDK9/cyclin T obtained from the same vendor appeared to be low quality, as it showed only minimal activity toward our positive control, the isolated Pol II complex. The kinase assays using recombinant CDK9/cyclin T in parallel with CDK9/cyclin K are now presented it Sup. Fig. 4B. The first two assays in this experiment contained Pol II as a substrate, and it is evident that Pol II was phosphorylated much stronger by CDK9/cyclin K than CDK9/cyclin T (comparing lane 1 vs lane 2). Therefore, the lack of detectable Treacle phosphorylation by CDK9/Cyclin T (lane 7), in contrast to strong phosphorylation by CDK9/cyclin K (lane 6), was likely attributable to poor reagent quality rather than physiological differences. We can conclude that CDK9/cyclin K reliably phosphorylates Treacle in vitro, but CDK9/cyclin T kinase assays were inconclusive.

eLife assessment

This study and associated data is compelling, novel, important, and well-carried out. The study demonstrates a novel finding that different chemotherapeutic agents can induce nucleolar stress, which manifests with varying cellular and molecular characteristics. The study also proposes a mechanism for how a novel type of nucleolar stress driven by CDK inhibitors may be regulated. The study sheds light on the importance of nucleolar stress in defining the on-target and off-target effects of chemotherapy in normal and cancer cells.

Joint Public Review:

Summary:

This is an interesting study with high quality imaging and quantitative data. The authors devise a robust quantitative parameter that is easy applicable to any experimental system. The drug screen data can potentially be helpful to the wider community studying nucleolar architecture and effects of chemotherapy drugs. Additionally, the authors find Treacle phosphorylation as a potential link between CDK9 inhibition, rDNA transcription and nucleolar stress. Therefore I think this would be of broad interest to researchers studying transcription, CDKs, nucleolus and chemotherapy drug mechanisms.

Revised manuscript:

While most of my concerns related were addressed, a PolI ChIP on rDNA would be an important experiment to establish the relevance of some of the conclusions of the paper using well established protocols with validated antibodies for PolI ChIP. Furthermore, additional S to A mutants of Treacle S1299A/S1301A is an important control which could have provided evidence if indeed S1299/S1301 were the only sites being phosphorylated by CDK9. To support their model, the authors should test if overexpression of Treacle mutants S1299A/S1301A can partially phenocopy the nucleolar stress seen upon CDK9 inhibition. This would considerably strengthen the author's claim that reduced Treacle phosphorylation leads to Pol I disassociation from rDNA and consequently leads to nucleolar stress. If not, it would have strengthened the authors' argument that Treacle could have multiple sites targeted by CDK9 and that mutating any one or two may not be sufficient to cause disassociation from PolI.

Overall, I believe the primary conclusions regarding the impact of various chemotherapy drugs on nucleolar state are solid and valuable to the broader scientific community. However, the mechanistic exploration of CDK9i is not sufficiently developed, and the authors have not adequately addressed the feedback provided in the original manuscript.

eLife assessment

This manuscript describes valuable findings on the expression pattern of orexin receptors in the midbrain and how manipulating this system influences several behaviors, such as context-induced locomotor activity and exploration. The overall strength of evidence - which includes anatomical, viral manipulation studies, and brain imaging - is solid and broadly supports claims in the paper, however, there are several areas in which the conclusions are only partially supported by the data provided. These results have implications for understanding the neural underpinnings of reward and will be of interest to neuroscientists and cognitive scientists with an interest in the neurobiology of reward.

Reviewer #1 (Public Review):

Summary:<br /> In this manuscript, the role of orexin receptors in dopamine neurons is studied. Considering the importance of both orexin and dopamine signalling in the brain, with critical roles in arousal and drug seeking, this study is important to understand the anatomical and functional interaction between these two neuromodulators. This work suggests that such interaction is direct and occurs at the level of SN and VTA, via the expression of OX1R-type orexin receptors by dopaminergic neurons.

Strengths:<br /> The use of a transgenic line that lacks OX1R in dopamine-transporter-expressing neurons is a strong approach to dissecting the direct role of orexin in modulating dopamine signalling in the brain. The battery of behavioural assays to study this line provides a valuable source of information for researchers interested in the role of orexin-A in animal physiology.

Weaknesses:<br /> The choice of methods to demonstrate the role of orexin in the activation of dopamine neurons is not justified and the quantification methods are not described with enough detail. The representation of results can be dramatically improved and the data can be statistically analysed with more appropriate methods.

Reviewer #2 (Public Review):

Summary:<br /> This manuscript examines the expression of orexin receptors in the midbrain - with a focus on dopamine neurons - and uses several fairly sophisticated manipulation techniques to explore the role of this peptide neurotransmitter in reward-related behaviors. Specifically, in situ hybridization is used to show that dopamine neurons predominantly express the orexin receptor 1 subtype and then go on to delete this receptor in dopamine neurons using a transgenic strategy. Ex vivo calcium imaging of midbrain neurons is used to show that in the absence of this receptor orexin is no longer able to excite dopamine neurons of the substantia nigra.

The authors proceed to use this same model to study the effect of orexin receptor 1 deletion on a series of behavioral tests, namely, novelty-induced locomotion and exploration, anxiety-related behavior, preference for sweet solutions, cocaine-induced conditioned place preference, and energy metabolism. Of these, the most consistent effects are seen in the tests of novelty-induced locomotion and exploration in which the mice with orexin 1 receptor deletion are observed to show greater levels of exploration, relative to wild-type, when placed in a novel environment, an effect that is augmented after icv administration of orexin.

In the final part of the paper, the authors use PET imaging to compare brain-wide activity patterns in the mutant mice compared to wildtype. They find differences in several areas both under control conditions (i.e., after injection of saline) as well as after injection of orexin. They focus on changes in the dorsal bed nucleus of stria terminalis (dBNST) and the lateral paragigantocellular nucleus (LPGi) and perform analysis of the dopaminergic projections to these areas. They provide anatomical evidence that these regions are innervated by dopamine fibers from the midbrain, are activated by orexin in control, but not mutant mice, and that dopamine receptors are present. Thus, they argue these anatomical data support the hypothesis that behavioral effects of orexin receptor 1 deletion in dopamine neurons are due to changes in dopamine signaling in these areas.

Strengths:<br /> Understanding how orexin interacts with the dopamine system is an important question and this paper contains several novel findings along these lines. Specifically:<br /> (1) The distribution of orexin receptor subtypes in VTA and SN is explored thoroughly.<br /> (2) Use of the genetic model that knocks out a specific orexin receptor subtype from only dopamine neurons is a useful model and helps to narrow down the behavioral significance of this interaction.<br /> (3) PET studies showing how central administration of orexin evokes dopamine release across the brain is intriguing, especially since two key areas are pursued - BNST and LPGi - where the dopamine projection is not as well described/understood.

Weaknesses:<br /> The role of the orexin-dopamine interaction is not explored in enough detail. The manuscript presents several related findings, but the combination of anatomy and manipulation studies does not quite tell a cogent story. Ideally, one would like to see the authors focus on a specific behavioral parameter and show that one of their final target areas (dBNST or LPGi) was responsible or at least correlated with this behavioral readout. In addition, some more discussion on what the results tell us about orexin signaling to dopamine neurons under normal physiological conditions would be very useful. For example, what is the relevance of the orexin-dopamine interaction blunting novelty-induced locomotion under wildtype conditions?

In some places in the Results, insufficient explanation and reporting is provided. For example, when reporting the behavioral effects of the Ox1 deletion in two bottle preference, it is stated that "[mutant] mice showed significant changes..." without stating the direction in which preference was affected.

The cocaine CPP results are difficult to interpret because it is unclear whether any of the control mice developed a CPP preference. Therefore, it is difficult to conclude that the knockout animals were unaffected by drug reward learning. Similarly, the sucrose/sucralose preference scores are also difficult to interpret because no test of preference vs. water is performed (although the data appear to show that there is a preference at least at higher concentrations, it has not been tested).

eLife assessment

In this important study, the authors integrated genetic and genomic datasets from humans and mice to unveil shared networks and pathways associated with coronary artery disease. Their compelling analysis led to the identification of new regulatory genes and pathways in vascular tissues and in the liver, allowing for a more in-depth understanding of the pathogenesis of coronary artery disease.

Reviewer #1 (Public Review):

This manuscript represents an elegant bioinformatics approach to addressing causal pathways in vascular and liver tissue related to atherosclerosis/coronary artery disease, including those shared by humans and mice and those that are specific to only one of these species. The authors constructed co-expression networks using bulk transcriptome data from human (aorta, coronary) and mouse (aorta) vascular and liver tissue. They mapped human CAD GWAS data onto these modules, mapped GWAS SNPs to putatively causal genes, identified pathways and modules enriched in CAD GWAS hits, assessed those shared between vascular and liver tissues and between humans and mice, determined key driver genes in CAD-associated supersets, and used mouse single-cell transcriptome data to infer the roles of specific vascular and liver cell types. The overall approach used by the authors is rigorous and provides new insights into potentially causal pathways in vascular tissue and liver involved in atherosclerosis/CAD that are shared between humans and mice as well as those that are species-specific. This approach could be applied to a variety of other common complex conditions.

Reviewer #2 (Public Review):

Summary:<br /> Mouse models are widely used to determine key molecular mechanisms of atherosclerosis, the underlying pathology that leads to coronary artery disease. The authors use various systems biology approaches, namely co-expression and Bayesian Network analysis, as well as key driver analysis, to identify co-regulated genes and pathways involved in human and mouse atherosclerosis in artery and liver tissues. They identify species-specific and tissue-specific pathways enriched for the genetic association signals obtained in genome-wide association studies of human and mouse cohorts.

Strengths:<br /> The manuscript is well executed with appropriate analysis methods. It also provides a compelling series of results regarding mouse and human atherosclerosis.

Author Response

The following is the authors’ response to the original reviews.

Reviewer #1 (Public Review):

The manuscript by Muthana et al. describes the effect of injection of an antibody specific for human CTLA4 conjugated to a cytotoxic molecule (Ipi-DM1) in knock-in mice expressing human CTLA4. The authors show that Ipi-DM1 administration causes a partial decrease (about 50% in absolute number) of mature B cells in blood and bone marrow 9-14 days after the beginning of treatment. Ipi-DM1 also results in a partial decrease in Foxp3+ Tregs (about 40% in absolute number) and a slight increase in activation of conventional T cells (Tconvs) in the blood at D9. Tconv depletion, CTLA4-Ig or anti-TNF mAb partially prevents the effect of ipi-DM1 on B cells. This work is interesting but has the following major limitations:

1) This work could have been of more interest if the Ipi-DM1 molecule would be used in the clinic. As this is not the case, the intimate mechanism of the effect of this molecule in mice is of reduced interest.

The goal of the current study is to use Ipi-DM1 ADC as probe to study mechanism of B cell loss observed in Treg-deficient host.

2) The fact that a partial deletion of Tregs is associated with activation of Tconvs and a decrease in B cells has been published several times and is therefore not new. According to the authors, their work would be the first to show that activation of Tconvs would lead to B cell depletion. However, this is shown in an indirect way and the mechanisms are not really elucidated. Indeed, this work shows a correlation between an increase in Tconv activation and a decrease in the number of B cells in the blood. The experiments to try to show a causal link are of 2 types: deletion of T cells (Fig 4) and blocking T cell activation with CTLA4-Ig (Fig 5) (neutralization of TNF addresses another question). Neither of these 2 experiments is totally convincing. Indeed, the absence of B cell depletion when T cells are deleted can be explained by other mechanisms than the preservation of B cell destruction by activated T cells. The phenomenon could be explained by B cell recirculation to lymphoid tissues or an effect of massive T cell death for example. The experiment shown in Fig. 5 with Belatacept is more convincing because this time the effect is targeted to activated T cells only. However, the prevention of B cell ablation is only partial. Again, since only blood is analyzed, other mechanisms could explain the B cell loss, such as their recirculation in lymphoid tissues.

While the concept of treg depletion leads to activation of Tconv cells and reduced B cells has been previously published, B cell loss was explained on basis of defective B cell lymphopoiesis due to low production of stroma cell-derived IL-7 or destruction of stromal cells by effector T cells. Our new data established that loss of B cells in the context of Treg depletion was not due to defects in the number of pre-/pro-B cells. Rather it is the death of mature B cells in the bone marrow.

To address the reviewer’s concern that the B cell loss was merely caused by a change in circulating pattern, we performed a new study on the effect of the ADC on B cells in bone marrow. Our new data reveal loss of mature bone marrow B cells, and that such loss is associated with increased apoptosis of mature B cells. Therefore, the loss of B cells in the peripheral blood is not due to a changed circulation. Furthermore, our data show that B cell progenitor, Pre-B, cells are not changed. Therefore, B cell lymphopoiesis is not the reason for B cell loss in our model system.

3) It is disappointing that only the blood (and sometimes the bone marrow) was studied in this work. The interest of doing experiments in mice is to have access to many tissues such as the spleen, lymph nodes, colon, lung, and liver. To conclude that there is B cell deletion without showing lymphoid organs (where the majority of B cells reside) is insufficient. As discussed above, the drop in B cells in the blood could be due to their recirculation in lymphoid organs. In addition, there is no measurement of functional B cells activity. Do mice treated with Ipi-DM1 have a decreased ability to develop an antibody response following immunization?

We have analyzed lymph nodes and spleen at the same time points. Unfortunately, Treg depletion was no longer observed at these time points. As expected, we did not see a clear depletion of B cells (Figure 1-figure supplement 6). In regards to functional B cell activity, we observed an increase of plasma immunoglobulins especially IgE which are now shown in Figure 3-figure supplement 1.

4) Although it is difficult to study in vivo, there is not a single evidence of increased B cell death after injection of Ipi-DM1.

Figure 2 & Figure 2-supplement 1 provides B cell death comparisons between IpiDM1 and hIgGFc group for bone marrow, blood, spleen, and lymph nodes. Statistically significant increase in B cell death is observed in mature B cells in bone marrow.

5) In most of the experiments, B cells are quantified with the B220 marker alone, but this marker, in some cases, can be expressed by other cells. It would have been preferable to use a marker more specific to B cells such as CD19 for example.

We have added data to support the death of mature B cells using other markers.

Minor points.

1) It should be indicated whether human CTLA4 binds normally to mouse CD80 CD86. We do not know if knock-in mice with human CTLA4 have a fully functional immune system.

We have indicated this point as suggested and cited our previous work line 226-227 (ref 23 & 24)

2) The manuscript is too long. Some of the data in the figures should be moved to supplemental figures. This is the case, for example, for some trivial stainings (Fig 1F, Fig 4B, 4F, Fig 5A, D, F, G). The figure legends and the Materials and Methods section are far too long. On the other hand, Fig 5-Fig Sup 1 could go into the main figures.

The figure legends, materials, and methods may be too long, but our intention is to provide as much info as possible for others who may be interested in our model system.

3) The anti-CTLA4 ADC reagent should be better explained and defined in the text.

The anti-CTLA-4 ADC reagent synthesis described in materials/methods under “Antibody-drug conjugate preparation.”

Reviewer #2 (Public Review):

Despite the fact that CTLA-4 is a critical molecule for inhibiting the immune response, surprisingly individuals with heterozygous CTLA-4 mutations exhibit immunodeficiency, presenting with antibody deficiency secondary to B cell loss. Why the loss of a molecule that regulates T cell activation should lead to B cell loss has remained unclear. In this study, Muthana and colleagues use an anti-CTLA-4 antibody drug conjugate (aCTLA-4 ADC) to delete cells expressing high levels of CTLA-4, and show that this leads to a reduction in B cells. The aCTLA-4 ADC is found to delete a subset of Tregs, leading to hyperactivation of T cells that is associated with B cell depletion. Using blocking antibodies, the authors implicate TNFa in the observed B cell loss.

The reciprocal regulation of T and B cell homeostasis is an important research area. While it has been shown that Treg defects are associated with B cell loss, the mechanisms at play are incompletely understood. CTLA-4 is not normally expressed in B cells so an indirect mechanism of action is assumed. The authors show that the decrease in Treg following aCTLA-4 ADC treatment is associated with activation of T cells, and that B cell loss is blunted if T cells are depleted. A role for both CD4 and CD8 T cells is identified by selective CD4/CD8 depletion. T cells appear to require CD28 costimulation in order to mediate B cell loss, since the response is partially inhibited in the presence of the costimulation blockade drug belatacept (CTLA-4-Ig). Finally, experiments using the anti-TNFa antibody adalimumab suggest a potential role for TNFa in the depletion of B cells.

While the manuscript makes a useful contribution, a number of questions remain. Perhaps most important is the extent to which this model mimics the natural situation in individuals with CTLA-4 mutations (or following CTLA-4-based clinical interventions). aCTLA-4 ADC treatment permits acute deletion of Treg expressing high levels of CTLA-4, whereas in patients the Treg population remains but is specifically impaired in CTLA-4 function. Secondly, although the requirement for T cells to mediate B cell loss is convincingly demonstrated, the incomplete reversal by TNFa blockade suggests additional unidentified factors contribute to this effect. Finally, although the manuscript favours peripheral killing of mature B cells over alterations to B cell lymphopoiesis, one concern is that this may simply reflect the model employed: the shortterm (6 day) treatment used here may be too acute to alter B cell development, but this may nevertheless be a feature of prolonged immune dysregulation in humans.

We appreciate reviewer’s comments and the difference between short-term depletion and permanent inactivation of Treg by genetic mutation is discussed. We would note that apart from mutation, dynamic Treg perturbation does occur under autoimmune conditions. Therefore, our data have significant implications for T-B cell interactions.

TNF-alpha is implicated in B cell loss as evidenced by the partial rescue with Anti-TNF treatment. We did not try to exclude the possibility that other mechanisms are involved.

Our data shows loss of circulating B cell in peripheral blood and mature bone marrow B cells. B cell progenitor, Pre-B, cells are not changed due Ipi-DM1 induced treg impairment, therefore B cell lymphopoiesis is not the reason for B cell loss in our model system. Evidence of increased cell death is only observed in mature B cells (Figure 2).

1) Following aCTLA-4 ADC treatment, it is surprising how subtle the deletion of Treg is (from ~8% to ~7%, Fig 1G), compared to the marked deletion of CTLA-4-expressing CHO cells. Is this a feature of in vivo versus in vitro treatment? If Treg are treated in vitro is deletion more efficient? How does the expression level of CTLA-4 in the CHO cells compare with the Treg in these assays?

We appreciate reviewer’s comments. The anti-CTLA-4 ADC targets CTLA-4 on cell surface. On average about 5% of Tregs express surface CTLA-4 at given moment while human CTLA-4 expressing CHO cell line stains > 90%. Nevertheless, Treg cell number in peripheral blood is reduced by >40%. Additionally, we have included bone marrow data, which shows a greater percentage of Treg depletion (Figure 1J).

2) The decrease in CTLA-4 seen after ipi-DM1 is complicated by the fact that the control DM1 conjugate (IgG1-DM1) appears to significantly increase CTLA-4 expression (Fig 1 supplement 2). It would be useful to clarify when hIgGFc is used versus hIgGFc-DM1 given the additional complexity introduced here (comparisons lacking a payload differ in more than one variable, while the hIgGFc-DM1 is clearly not inert).

We appreciate reviewer’s comments. We agree that the hIgGFc-DM1 control slightly increased CTLA-4 level; nevertheless, it did not alter B cells, T cells or their proliferation capacity when compared to hIgGFc. Our point here is that B cell depletion is not mediated by DM1 payload off target release (new-version Figure 1-Figure supplement 4, old version Figure 1-figure supplement 2). As for the clarification comment when hIgGFc is used versus hIgGFcDM1 is used, the information is clarified in the figure legend. Comparisons are made between (hIgGFc VS Ipi-DM1) or (hIgGFc VS hIgGFc-DM1).

3) T cell-derived IFNg is another potential contender for influencing B cell homeostasis - have you considered testing whether this also contributes in your model?

We appreciate reviewer’s suggestion. IFN was reported to induce apoptosis and cell arrest in Pre- B cells, however these studies are invitro studies Garvey et.al Immunology. 1994 Mar; 81(3): 381–388; Grawunder et.al Eur. J. Immunol. 23, 544–551. Since we did not observe any effect on Pre-B cells, we have not followed the literature to investigate the role of IFNy in B cell loss in our model.

Reviewer #3 (Public Review):

The co-suppressive molecule CTLA-4 has a critical role in the maintenance of peripheral tolerance, primarily by Treg mediated control of the co-stimulatory molecules CD80 and CD86. As stated by the authors, previous studies have found a variety of effects of anti-CTLA-4 antibody treatment or genetic loss of CTLA-4 on B-cells. These include increased B-cell activation and antibody production, autoantibody production, impairment of B-cell production in the bone marrow and loss of peripheral B-cells. In this article Muthana et al use a CTLA-4 humanized mouse model and examine the effects of drug conjugated CTLA-4 on the immune system. They observe a transient loss of B-cells in the blood of the treated mice. They then use a range of immune interventions such as T-cell depletion and blocking antibodies to demonstrate that this effect is dependent on T-cell activation.

Since anti-CTLA-4 immunotherapy is in active clinical use exploration of its effects are welcome, this is helped by the use of a humanized CTLA-4 system which should be considered a strength of the paper. However, currently, the central premise of this paper, that B-cells are depleted, seems underexplored. Direct evidence of T-cell killing of B-cells is never presented, rather it is inferred from the reduced numbers of B-cells in the blood. The status of B-cells in sites that contain a large proportion of B-cells such as the spleen and lymph nodes is not examined. Additionally, no examination of B-cell antibody production is performed.

We appreciate reviewer’s comments. To address the reviewer’s concerns we performed additional experiments to evaluate the impact on B cells in other organs, as detailed in our responses to specific questions.

1) Examination of B-cell apoptosis/cell death and T-cell mediated cytotoxicity is needed. The authors repeatedly refer to auto destructive T-cells without ever demonstrating their presence or any direct evidence that B-cells are dying. This is particularly important in the context of the blood since an alternative hypothesis would be a change in B cell trafficking and infiltration into tissues.

We appreciate reviewer’s comments. To address the reviewer’s concern that B cell loss in blood might be caused by a change in B cell trafficking pattern. We performed new study on the effect of the ADC on B cells in bone marrow. Our new data reveal loss of mature bone marrow B cells, and that such loss is associated with increased apoptosis of mature B cells (Figure 2). Therefore, the loss of B cells in the peripheral blood is not due to B cell trafficking and infiltration into tissues.

2) The authors demonstrate that B-cells are mostly reduced in blood at around days 10 to 15, I believe it is critical to determine if this is also reflected in the lymphoid organs such as the spleen and lymph nodes.

We appreciate reviewer’s comments. We have analyzed lymph node and spleen at the same time points. Unfortunately, Treg depletion was no longer observed at these time points. As expected, we did not see a clear depletion of B cells (Figure 1-figure supplement 6).

3) Related to the above point do the authors see evidence of Splenomegaly or lymphadenopathy?

We appreciate reviewer’s comment. Evidence of splenomegaly and lymphadenopathy is presented in Figure 3-figure supplement 2.

4) Minimal examination of the status of the B-cells or antibody production is performed. Previous reports would suggest that plasma cell induction and antibody responses may be expected. Do serum antibody levels change in this system?

We appreciate reviewer’s comment. Increases of plasma immunoglobulins especially IgE are now shown in Figure 3-figure supplement 1.

5) Its unclear how the authors interpret their experiment with anti-TNFa (figure 6). Are they suggesting that TNFa itself depletes B-cells or that it is part of the inflammatory milieu that contributes to wider T-cell activation and, in turn, B-cell depletion?

We have discussed these possibilities in the revised manuscript.

Reviewer #1 (Public Review):

The manuscript by Muthana et al. describes the effect of injection of an antibody specific for human CTLA4 conjugated to a cytotoxic molecule (Ipi-DM1) in knock-in mice expressing human CTLA4. The authors show that Ipi-DM1 administration causes a partial decrease (about 50% in absolute number) of mature B cells in blood and bone marrow 9-14 days after the beginning of treatment. B cell progenitors and pre-B cells in the BM are not affected. Ipi-DM1 also results in a partial decrease in Foxp3+ Tregs (about 40% in absolute number) and a slight increase in activation of conventional T cells (Tconvs) in the blood, spleen, BM and LNs at D9 as well as increased plasma immunoglobulins especially IgE. Tconv depletion, CTLA4-Ig or anti-TNF mAb partially prevents the effect of ipi-DM1 on B cells. This effect of Ipi-DM1 on the reduction B cells and Tregs at D9 is not observed in the spleen and lymph nodes (maybe not the good timing to see it), and there is even an increase in the number of Treg and the frequency and number of B cells in lymph nodes. This work is interesting but has the following major limitations:

1- This work could have been of more interest if the Ipi-DM1 molecule would be used in the clinic. As this is not the case, the intimate mechanism of the effect of this molecule in mice is of reduced interest.

2- The fact that a partial deletion of Tregs is associated with activation of Tconvs and a decrease in B cells is not new. According to the authors, their work would be the first to show that activation of Tconvs would lead to B cell death. However, this is shown in an indirect way and the mechanisms are not really elucidated. The experiments to try to show a causal link are of 2 types: deletion of T cells (Fig 5) and blocking T cell activation with CTLA4-Ig (Fig 6). These 2 experiments are not fully convincing. The absence of B cell depletion in the blood when T cells are deleted can be explained by other mechanisms, such as B cell recirculation to lymphoid tissues or an effect of massive T cell death for example. The experiment with CTLA4-Ig is more convincing because the effect is targeted to activated T cells only. However, the prevention of B cell ablation is only partial. Since only blood is analyzed, other mechanisms could explain the B cell loss, such as their recirculation in lymphoid tissues.

3- The authors propose that the drop in B cell numbers in the blood in mice treated with Ipi-DM1 results from reduced mature B cells in the bone marrow. However, B cells are continuously recirculating between the blood and secondary lymphoid tissues. The drop of blood B cells could be well explained by an increased recirculation to lymphoid organs. The increased numbers of B cells in lymph nodes support this latter hypothesis.

4- The new Figure 2 suggests direct evidence of apoptosis of mature B cells in the BM of treated mice using a PI/annexin V staining assay. This is an important point to support the point of the manuscript. However, using the same assay, the level of B cell apoptosis is of 80% in lymph nodes and 50% in the spleen in control mice (see new Figure 2-figure supplement 1), which is way too high and questions the reliability of this assay. It is likely that B cells enter apoptosis only in vitro due to some artefactual stress.

Reviewer #2 (Public Review):

Despite the fact that CTLA-4 is a critical molecule for inhibiting the immune response, surprisingly individuals with heterozygous CTLA-4 mutations exhibit immunodeficiency, presenting with antibody deficiency secondary to B cell loss. Why the loss of a molecule that regulates T cell activation should lead to B cell loss has remained unclear. In this study, Muthana and colleagues use an anti-CTLA-4 antibody drug conjugate (aCTLA-4 ADC) to delete cells expressing high levels of CTLA-4, and show that this leads to a reduction in B cells. The aCTLA-4 ADC is found to delete a subset of Tregs, leading to hyperactivation of T cells that is associated with B cell depletion. Using blocking antibodies, the authors implicate TNFa in the observed B cell loss.

The reciprocal regulation of T and B cell homeostasis is an important research area. While it has been shown that Treg defects are associated with B cell loss, the mechanisms at play are incompletely understood. CTLA-4 is not normally expressed in B cells so an indirect mechanism of action is assumed. The authors show that the decrease in Treg following aCTLA-4 ADC treatment is associated with activation of T cells, and that B cell loss is blunted if T cells are depleted. A role for both CD4 and CD8 T cells is identified by selective CD4/CD8 depletion. T cells appear to require CD28 costimulation in order to mediate B cell loss, since the response is partially inhibited in the presence of the costimulation blockade drug belatacept (CTLA-4-Ig). Finally, experiments using the anti-TNFa antibody adalimumab suggest a potential role for TNFa in the depletion of B cells.

While the manuscript makes a useful contribution, a number of limitations remain. Perhaps most important is the extent to which this model mimics the natural situation in individuals with CTLA-4 mutations (or following CTLA-4-based clinical interventions). aCTLA-4 ADC treatment permits acute deletion of Treg expressing high levels of CTLA-4, whereas in patients the Treg population remains but is specifically impaired in CTLA-4 function. Secondly, although the requirement for T cells to mediate B cell loss is convincingly demonstrated, the incomplete reversal by TNFa blockade suggests additional unidentified factors contribute to this effect. Finally, although the manuscript favours peripheral killing of mature B cells over alterations to B cell lymphopoiesis, one concern is that this may simply reflect the model employed: the short-term (6 day) treatment used here may be too acute to alter B cell development, but this may nevertheless be a feature of prolonged immune dysregulation in humans.

Reviewer #3 (Public Review):

The co-suppressive molecule CTLA-4 has a critical role in the maintenance of peripheral tolerance, primarily by Treg mediated control of the co-stimulatory molecules CD80 and CD86. As stated by the authors, previous studies have found a variety of effects of anti-CTLA-4 antibody treatment or genetic loss of CTLA-4 on B-cells. These include increased B-cell activation and antibody production, autoantibody production, impairment of B-cell production in the bone marrow and loss of peripheral B-cells. In this article Muthana et al use a CTLA-4 humanized mouse model and examine the effects of drug conjugated CTLA-4 on the immune system. They observe a transient loss of B-cells in the blood of the treated mice. They then use a range of immune interventions such as T-cell depletion and blocking antibodies to demonstrate that this effect is dependent on T-cell activation.

Since anti-CTLA-4 immunotherapy is in active clinical use exploration of its effects are welcome, this is helped by the use of a humanized CTLA-4 system which should be considered a strength of the paper. However, currently the central premise of this paper, that B-cells are depleted seems underexplored. Direct evidence of T-cell killing of B-cells is never presented, rather it is inferred from the reduced numbers of B-cells in the blood and increased apoptosis in the bone marrow. It is not made clear if B-cell numbers in the bone marrow are reduced.

Upon examining lymphoid organs it seems that the spleen is relatively unchanged while the lymph nodes have a large increase in B-cells alongside increased serum antibody levels. The paper does underline the importance of looking at the differences of multiple immune compartments and interesting phenomenon are described in each compartment. Simultaneous inhibition of B-cell lymphopoiesis and blood trafficking with strong activation and antibody production of lymphoid resident (presumably germinal center) B-cells appears to be occurring. However the current overall interpretation that B-cells are broadly depleted is perhaps too simplistic and largely ignores the lymphoid organs and serum antibodies.

eLife assessment

This study reports that neural activity in the auditory cortex (field L) of singing male songbirds can be modulated by social context. These potentially important findings indicate that the presence of a female conspecific alters the response of auditory cortical neurons to the male bird's own song and to perturbations of auditory feedback that the bird has been trained to expect. While they extend recent work showing that the activity of dopaminergic neurons in songbirds is also affected by an audience, the evidence presented is incomplete since it is unclear how much of the apparent modulation of cortical neurons may be due to other factors, such as changes in the recorded neurons or their properties over time, which will require additional analyses to work out.

Reviewer #1 (Public Review):

Summary:<br /> This study examines the context-dependent modulation of auditory cortical neurons in response to expected sensory input, either self-generated sounds or expected perturbations of self-generated sounds. Specifically, using songbirds, the authors ask whether social context (the presence of a female conspecific) affects 1) the response of auditory cortical neurons to the bird's own song when he is singing; and 2) the response of neurons to perturbations of auditory feedback that the bird has been trained to expect.

Strengths:<br /> First, the authors report that across the population, the responses of the neurons does not differ when a male bird sings alone or if he sings to a female. A fraction of auditory cortical neurons, however, do show significant differences in the firing rate, precision, and/or degree of burst firing when males sing alone vs. when they sing to females. This finding is broadly consistent with the literature showing that sensory neurons (visual, auditory, somatosensory, etc.) can be rapidly reconfigured into different "information processing modes" depending on behavioral state (e.g, quiescence vs vigilance).

For the perturbation experiments, the authors trained birds to expect distorted auditory feedback during a particular syllable. They found that some neurons showed greater responses during perturbation when a female was present (compared to when males were alone) while other neurons had smaller responses during perturbation when a female was present. In addition, the response of a small number of auditory cortical neurons were not affected by behavioral state. These results contrast with their prior report that the responses of midbrain dopaminergic neurons that project to the basal ganglia are "uniformly reduced" in the presence of a female, raising a question of how an evaluation signal is transformed in the circuit from the primary sensory region to the midbrain.

Weaknesses:<br /> While the experiments and analysis are solid, the finding that social context can alter responses of auditory cortical neurons in a multitude of ways (increase, decrease or no change) raises several questions that can be examined with additional analysis. For example, do context-dependent differences in auditory responses derive from context-dependent differences in the songs? Are context-dependent differences present in all classes of neurons and throughout the auditory system?

The observed heterogeneity in the firing properties of auditory cortical neurons, both in response to self-generated sounds and during perturbations of auditory feedback, raises the question of which neurons are sensitive to social context (which likely can be addressed by the authors in a revision). The authors should provide additional details about the recordings:

a) What are the locations of the recording sites?<br /> Prior work has shown that there is an organized map of spectrotemporal features of sounds in the auditory cortex of songbirds; spectral tuning widths change along the medial-lateral axis and temporal tuning widths differ between the input and output layers of Field L. Were the recordings primarily in Field L2 (thalamo-recipient region), L1 or L3? Were some recordings lateral to Field L in secondary auditory regions? Were the neurons that showed context-dependent changes in firing properties localized or distributed throughout Field L (i.e., were the context-dependent differences in neural responses truly brain-wide)? At a minimum, the authors should include a schematic showing the different regions of Field L and a summary of the location of the recording sites. Images of the processed tissue with electrolytic lesions would also be helpful.

b) Was the context-dependent modulation limited to a particular class of neurons (distinguished by spike waveform shape, spontaneous firing rate, or other feature)?

While the authors attribute differences in the responses of single auditory cortical neurons to the presence of a female, other potential explanations for the observed differences should be examined (and potentially ruled out):

a) Prior work has shown that songs of zebra finches differ slightly when males sing alone compared to when they sing to females: songs are faster; pitch is less variable; and the number of introductory elements is greater when males sing to females. Do some of the observed social context-dependent differences in the responses of auditory neurons reflect differences in the songs in the two conditions? This idea is supported in part by a prior study in juvenile zebra finches (Keller & Hahnloser, 2009) showing that ~20% of the neurons they recorded in Field L and a secondary auditory region (CLM) showed anticipatory activity even before the onset of a song bout, suggesting a source of premotor (or at least non-auditory drive) to neurons in the auditory cortex. Did the authors of this study also find premotor activity in Field L, and if so, did it differ between the two social contexts? Might differences in Field L responses reflect motor/song differences?

b) For the perturbation experiments, the authors report heterogeneous responses to playback, with some neurons firing more and other firing less when a female is present compared to when the male is alone. Keller and Hahnloser (2009) found that in juvenile birds, responses of Field L to perturbations of auditory feedback were sensitive to sound amplitude; perturbation responses increased with relative perturbation amplitude. This raises a question of whether perturbation amplitude is different when a male is alone and when a female is present (i.e., the male may move towards the female when she is present and if the speaker is close to the female, the perturbation may be louder than when the male is alone; alternatively, the male be more active when he is alone so the loudness of the perturbation may be more variable across song bouts). It would be useful to know if (and how much) perturbation amplitude varied depending on the location inside the cage as well as whether the sound pressure level of the underlying song was higher (e.g., Lombard effect). Addition of details of the experimental setup/procedure would help to allay concerns that the amplitude of the white noise varied significantly depending on behavioral context.

Finally, I am still trying to make sense of the differences in the context-dependent modulation of responses of auditory cortical neurons vs. midbrain dopaminergic neurons. Given the heterogeneity of responses in Field L, both to self-generated sounds and to expected perturbations during singing, how are the signals decoded downstream of Field L? At the population level, neither the mean firing rate nor the timing of firing of Field L neurons changed with courtship. Similarly, across the population, the responses to perturbations of auditory feedback were not affected by courtship state (error signal attenuated in 11 neurons, increased in 22 neurons and not affected in 10 neurons). Yet, the courtship state "uniformly" reduces the response of midbrain dopaminergic neurons to auditory perturbation. It would be helpful if the authors could include a model and/or more discussion of how this change may arise.

Reviewer #2 (Public Review):

Summary:

In the manuscript 'Auditory cortical error signals retune during songbird courtship', Jones and Goldberg study auditory cortex in male zebra finches. They explore song-related responses in two different contexts, when the male is either alone or in the presence of a female. Social-context related responses are hypothesized based on previous results on downstream VTA neurons where such modulation is found. They play jamming stimuli through a loudspeaker to probe sensitivity of song-related neural responses to these external stimuli. They find a heterogeneity of responses, in line with auditory cortical neurons computing the social modulation of responses found in VTA.

Strengths:

In general, the work is interesting and sheds light onto auditory processing and self-perception mechanisms in songbirds.

Weaknesses:

Stability of responses has not been studied: some neurons seem to have responses that slowly drift in time, which could lead to observed differences between alone and with-female conditions. Also, possible motor confounds and sound-of-audience confounds should be addressed. The language is often imprecise.

Stability and Reversal: It is a bit unfortunate that stability of effects seemingly has not been studied by reversing experimental conditions. The work would be much stronger if authors could show that audience-dependent tuning is robust in individual cells. Did they record from some neurons during reversal back to the alone condition? Ideally, the responses should be identical before and after recording with an audience. This would control for possible non-stationarities in their neuron recordings/spike-sorting/circadian trends. If authors do not have such data, it would be worth wile to even just try to divide the dataset for each neuron and condition (either the audience or isolate condition) into two parts to verify that the response is the same in either part (provided sufficient song renditions are recorded). See also my comment below about Fig. 2A.

Motor responses: Does DAF playback change song? If so, especially if it applies only in one of the two conditions (audience/no audience), then the observed response differences could be motor-related rather than auditory responses. Analyses of song spectrograms right after DAF would presumably provide the answer.

Similarly, motif-aligned spiking activity was time warped to the median duration of undirected or directed motifs. Could the shorter motifs during directed song (as has been reported in other studies) lead to alignment differences that would account for the different error responses in alone/wfemale conditions? In other words, could increased error responses be due to the fixed 100 ms analysis window of the audience condition that extends into a song region beyond the 100 ms region of the no-audience condition where there is increased firing? And vice versa for observed decreases in error responses, i.e. is there a firing pause just after the offset of the 100 ms window in the no-audience condition that causes audience dependence of responses? A simple compensation of song tempo differences by shortening/stretching the analysis window in one of the two conditions would allow to test for this.

Audience versus sound of audience: In the first sentence of the discussion authors write: we discovered that auditory representations of an animal's own vocalizations change with an audience. Is it truly the audience that causes the difference in error responses or is it the sounds the audience makes? To control for that would be to play back stimuli that simulate a non-silent audience through a loudspeaker to see whether error responses depend on the soundscape created by a typical audience (either present or absent). Authors probably do not have such data and to record it would go beyond the scope of this study, but it would be important to discuss this possibility or perform some analysis in that vein.

Reviewer #3 (Public Review):

Summary:

In this study, Jones et al. examine how neural activity in a primary auditory area (field L) of singing male songbirds is modulated by the presence or absence of an audience (a female conspecific). Prior work has demonstrated that the presence of an audience attenuates the responses of dopaminergic neurons to distortions of auditory feedback (DAF). Here the authors report that even in a region that is primarily considered sensory, responses to DAF are also modulated by the audience, although in a heterogeneous manner that does not readily explain previously observed attenuation. These findings address an interesting question and will potentially be important in adding to an understanding of how non-sensory factors can alter response properties of neurons even in primary sensory regions in a context dependent fashion. However, to be fully persuasive, additional analyses will be required to address how much of the apparent modulation by audience may be explained by other factors such as changes in recorded neurons or their properties over time.

Full Public Review:

In this study, Jones et al. examine how neural activity in a primary auditory area (field L) of singing male songbirds is modulated by the presence or absence of an audience (a female conspecific). They test whether activity in Field L differs between conditions in which the male is singing to a female (directed song) or alone (undirected song) and whether response to distortions of auditory feedback (DAF) differ between these conditions. Previous work has shown that in other parts of the songbird brain, sensory-motor activity can differ between directed and undirected song, and that responses to DAF are attenuated when males sing directed song versus undirected song. These prior results raise the interesting question of the extent to which such modulations of activity by the presence of an audience are already present in primary sensory areas such as Field L. This possibility is also motivated by prior work that has shown that Field L activity is not exclusively explained by auditory input, but can also be modulated by the bird's state - whether it is singing or not.

Against this background, the questions asked here are of interest for two inter-related reasons:

1) the authors address whether the presence of an audience (a female conspecific) alters activity in a primary auditory area during singing. Primary auditory areas such as Field L, and analogous mammalian thalamo-recipient cortical regions such as A1, are often thought of as responding very specifically to the features of sensory stimuli, but are also understood to be modulated by a variety of factors including the attentional and behavioral state of the animal. For audition, such modulation includes whether or not animals are vocalizing and listening to themselves or listening to playback of their own vocalizations. Cited works from Keller (2009) as well as Eliades and Wang (2008) have indicated that the act of vocalizing can modulate auditory responses to self-generated feedback in primary auditory areas relative to those arising from playback of the same sounds. Here, the question is whether responses to self-generated feedback differ between conditions of singing alone versus singing to a female audience. A demonstration that the presence of an audience matters to responses in Field L would add to a general understanding of how it is that non-auditory factors can modulate sensory responses.

2) the authors address the possible source of an audience-dependent modulation of responses to feedback perturbation in the VTA previously reported by Goldberg and colleagues (2023). In the VTA, responses to perturbations during singing are consistently attenuated when males are singing to females versus when they are singing alone, but the underlying mechanisms of this modulation are unknown. Here, the authors test the possibility that such modulation by an audience is already present at the level of Field L. The previously reported attenuation in VTA is quite striking and reflects a nice example of how neural processing can differ with varying behavioral priorities. Understanding whether this modulation of responses to DAF arises already in primary auditory areas would further a mechanistic understanding of an intriguing example of state-dependent modulation of sensory processing and behavior, and lend broad insight into related phenomena.

The authors report 1) that activity in Field L differs between directed and undirected singing at many individual recording sites, but that these changes are heterogeneous, with both increases and decreases in activity, so that there is no consistent change across the population and 2) that the responses to DAF differ between directed and undirected song, but that there is no consistent attenuation of response (as observed in the VTA) and instead heterogeneous increases and decreases in response to DAF so that there is no net change at the population level.

These findings, if firmly established, are important and of general interest. While they do not readily explain the source of the audience-dependent attenuation of auditory responses to DAF in the VTA, the demonstration of audience-dependent modulation of self-generated feedback and its disruption in a primary auditory area is an exciting result that would provide an opportunity for further investigation of how changes in social context influence brain and behavior. The manuscript is generally well written, although the presentation is terse. My main reservations about the current manuscript relate to aspects of experimental design and analysis that need to be clarified and addressed before these conclusions will be fully persuasive. There are also some places where further discussion of the findings and their relationship to prior studies would be helpful.

1. A central concern relates to whether the main reported effects associated with differences in singing directed versus undirected song reflect only those changes in conditions, versus contributions from changes in unit isolation or response properties over time. The authors record undirected song in a block in the morning and only after collecting at least 40 renditions do they later record responses during directed song over a series of repeated exposures to a female. Therefore, differences between data collected during undirected song and directed song also reflect differences between data collected initially during the morning versus later. It is unclear from methods whether any of these recordings during undirected and directed conditions are interleaved, but if this is not the case, then it is crucial to ask how stable were neural recordings with respect to unit isolation, and potential changes to response properties, over the duration of the experiments. This would be less of a concern if the results mirrored those observed in the VTA, where attenuation of responses was observed across the entire population during directed versus undirected conditions - it is hard to explain a phenomenon that is consistently observed across the population as arising from a change in which neurons and spikes are contributing to responses, or other forms of non-stationarity. However, because there are no significant differences reported at the population level in the current study, it is important to address the possibility that observed differences between conditions reflect some form of noise or drift in recorded units, rather than being entirely due to directed versus undirected singing. I have elaborated in more detail below on this concern, including places where the data seems to suggest some non-stationarity of responses, and have some suggestions for ways in which this concern might be addressed.

2. A second concern, related to this first one, has to do with the categorical definition of 'error neurons'. The authors note in their text that it could be problematic to apply categorical definitions to continuous distributions, and yet that seems to be what they then do. The authors have a metric of error sensitivity that they apply to each neuron's response to DAF in both undirected and directed conditions (the error score). They show that there is a continuous distribution of error scores (Figure 2 - figure supplement 1) across the population, with no bimodality that would be suggestive of distinct error sensitive and error-insensitive neurons. One nice feature of their analysis is that they also show the distribution of error scores computed in an analogous fashion for a period of neural activity in the song prior to DAF. This control data set makes it persuasive that there is a significant response to DAF, but also shows that there can be a broad range of error scores even when no DAF has been played, and that this range of 'noise' responses to DAF overlaps substantially with the actual responses to DAF. Despite the continuum of error scores, the authors define a subset of neurons as error responsive only if their responses to DAF exceed a specific threshold (2.5 standard deviations). One of the main conclusions of the paper is based on finding a subset of 22 neurons that exhibited error responses (by this definition) only during singing to a female and 11 neurons that exhibited error responses only when singing alone. These neurons are described as 'retuned' because they have error responses in only one condition.

The problem here is that for some, if not many, of the neurons that are categorically defined as being responsive to DAF in only one condition (directed versus undirected) there is almost certainly not a significant difference in the actual responses to DAF between conditions. This is apparent in the relevant data figure (figure 2 - figure supplement 1) and is a consequence of using a threshold to split a continuous distribution into groups defined as error responsive or not. For example, several neurons in this plot that have almost identical scores in the directed and undirected condition are counted as examples of retuning because the error scores are just a bit over 2.5 in the directed condition and just a bit under 2.5 in the undirected condition.

That this kind of categorical approach may be problematic is apparent in the control data in the plot. Despite the absence of any perturbation, there are error responsive neurons present in these data that are considered selective for directed versus undirected singing - this is an expected consequence of using a threshold on dispersed or noisy biological data. Shifting to a more stringent threshold of three standard deviations, as the authors do, does not help with this problem, as that still treats as categorically different responses that fall on either side of a line, even if only by a tiny amount. I suggest that the authors devise a measure for each neuron to test whether the responses to DAF are significantly different under the two conditions (directed versus undirected). As noted above, this measure should take into account some assessment of the stationarity of responses, as well as the distribution of responses (which, in some of the examples does not seem to be Gaussian around a mean response level, but rather highly variable across trials).

3. There are several places where further discussion of the previous literature and how the current results relate to that literature would be helpful. This includes:

3a. Some discussion of what is already known about the auditory tuning of field L, and the extent to which responses associated with distortion of feedback may reflect the frequency tuning of field L neurons versus something that might be construed as more specifically as detecting an error in perceived feedback. For example, Field L neurons have previously been characterized as having relatively simple spectro-temporal receptive fields, often with a single frequency band that is excitatory and nearby frequency bands that are inhibitory. It would be beyond the scope of this paper to directly assess the extent to which both song responses and responses to DAF are well predicted by simple STRFs that might be measured for the recorded neurons, or computed from activity during a range of vocalizations, but perhaps worth discussing whether a neuron with such frequency tuning would potentially exhibit 'error responses' of the sort described here, simply because the DAF stimulus happens to fall into the excitatory or inhibitory regions of the neuron's receptive field. While it is OK to use the term 'error responsive' in the current study, it would be good to make clear that changes in firing associated with playing DAF should be expected even for neurons that have simple auditory receptive fields (i.e. with center surround tuning to specific frequencies in a tonotopic map, as has been described for Field L) without necessarily indicating that these neurons are specifically registering any deviation or 'error' between expected feedback and experienced feedback. In this respect, there are multiple subdivisions of Field L with different tuning properties. Please specify further what criteria were used to determine recording locations and how these correspond with previously defined subdivisions.

3b. It would also be useful to discuss further previous work on differences in auditory tuning or responses between conditions when subjects are vocalizing, versus when vocalizations are played back (as in Keller, Eliades) and whether the results in the current study are similar or different. For example, this prior work has indicated that efference copy or other signals that precede vocalizations can reach and influence activity in auditory areas - with the most compelling evidence for this being the modulation of activity prior to the onset of vocalizations. Was this also observed in the current study, and to what extent might this kind of mechanism contribute to the processing of feedback distortions? With respect to this kind of efference signal, or other possibilities, can the authors provide some discussion or speculation about possible mechanisms that might be differentially engaged between conditions of singing directed versus undirected song?

3c. The previous study on DAF responses in VTA indicates enhanced responses to female calls during directed song. To what extent did the current study control for any vocalizations or other sounds produced by females during the directed singing, and could this have contributed to differences in Field L activity between conditions? This question is motivated partly by the highly variable responses in raster plots even within one condition - might some of this reflect motifs during which transient noises are produced from female calling or other movements by the male or female?

More regarding stability of recordings:

The data presented in Figure 1D illustrate some of my concerns about the stationarity of recordings. In the directed condition there are no spikes at all following the first handful of motif renditions. Were the directed and undirected recordings interleaved here? If not, could the recorded neuron simply have been lost, changed in amplitude of recorded spikes so that it was no longer counted, or reduced its responsiveness over the course of the recordings? Because the recordings of undirected and directed singing are described as occurring sequentially, it seems likely that this type of change in recorded signal could contribute to changes in measured responses over time, independently of effects due to directed versus undirected singing.

A minor issue of this example is that the raw example trace with male alone does not seem to have a corresponding set of points in the roster plot. For panel E, I also cannot find rasters that correspond to the example recordings shown at top.

Figure 2A also shows a neuron that looks like it has non-stationarity; for the alone condition without altered feedback, the main peak has no spikes for the bottom half of the rasters. For the directed condition, much of the difference between control and distorted feedback conditions seems to come from a few trials towards the bottom of the raster plot that show more and earlier firing than most other rasters.

Other more subtle examples are suggested in the figures, such as Figure 1F where responses in the alone condition seem to increase over the course of recordings. A related issue apparent in some of the raster plots is that the firing rate distributions within a given condition sometimes appear to be very non-gaussian, with some motifs during which there is a lot of activity, or apparent bursting, and others in which there is little activity. In addition to the examples above, this includes<br /> responses in Fig 1E and Fig 2F. Does anything distinguish these cases or trails? Where differences between conditions are driven by firing differences that are present on only a subset of trials, such as in Fig 2A, there is some deviation from the normal criteria for use of T-tests/Z-scores. Please consider this point and discuss any caveats and/or apply other tests (Monte Carlo? Non-parametric?) as appropriate.

These potential issues of non-stationarily, and non-Gaussian firing rate distributions in each condition, make it complicated to think about what differences in activity reflect changes from undirected to directed conditions versus these other factors.

Approaches to addressing this issue could include more specifically indicating examples in which recordings from the alone condition and directed condition are interleaved and exhibit reversible (between conditions) changes in the pattern of responses (both without DAF in comparing alone versus directed, and with DAF demonstrating differences in DAF influences between conditions). Some good interleaved examples of this sort would be very helpful to illustrate the robustness of differences between conditions. More generally, the methods and or raster plots should include some further explanation of the time periods over which recordings were made in the alone versus directed conditions, and the extent to which they are interleaved or not.

Another approach that could be used if there are not many instances of inter-leaved recordings is to try to document the stationarily or stability of unit isolation and/or responses over time. It would be most helpful when applied to recordings from a given singing condition (i.e. alone or directed) that are interleaved, but even in cases where this is not possible perhaps one could assess the stability of waveforms and unit isolation across time. For example in Figure 2 - Supplementary figure 2, the left-hand and middle examples appear to have quite good unit isolation, and might be the sorts of cases where measures of unit isolation and waveform stability could be used to argue that a gain or loss of spikes due to drift in recordings or changes to SNR and spike detection are not contributing to changes in firing patterns over time (and across conditions).

It potentially would also be informative to present the prevalence of the main effects reported in the study as a function of some measures of unit isolation, SNR, and recording stability. It would be reassuring to see that significant differences between conditions are equally or more prevalent under the conditions of greatest unit isolation and recording stability than in cases with worse SNR or stability.

One other way that the authors might be able to address my main concern would be to look at the stability of firing patterns within conditions, where differences across trials most directly indicate the potential contributions of technical or biological changes in neural activity over time that are not related to the experimental conditions.

To further address some of these issues, it would be helpful to have additional explanations in this paper (rather than by reference to Goldberg and Fee, 2010) of the criteria that were used for counting spikes, and assessing stability of recordings. All I found about this in the Goldberg and Fee, 2010 reference was that "Spikes were sorted off-line using custom Matlab software" Does this require human inspection and judgment? Is there a simple threshold, or waveform measurement used for detecting spikes from single units? Are some sort of signal to noise measures, or ISI violations used to score how well units are isolated?

For the specific examples shown in figures, it would be useful to indicate by small tick marks or otherwise which spikes were counted as single units. For example in figure 2 column B, for the condition with female, did only the 1-3 largest spikes get counted, or also the spikes of medium height?

Page 11: "Many channels on the probes recorded multi-unit activity, which were taken note of but not analyzed in this study."

What were the criteria for this? For several of the examples in the figures there are spikes of varying amplitudes and as mentioned above it would be helpful to clarify how the spikes were sorted into single units in such cases.

Categorical scores:

Page 13: "Neurons with error responses greater than 2.5 in only one condition (undirected versus directed) were considered to have retuned; neurons with error scores greater than 2.5 in both conditions were considered not to have retuned."

This definition results in cases where responses of 2.45 vs 2.55 are described as 'retuned', even if these responses are not significantly different. The figure (Figure 2 - figure supplement 1) indicates that multiple neurons that were scored as retuning had responses that fall very near the threshold in this way.

Page 13, "Our results did not fundamentally change with ... a more stringent threshold of 3..."

The stringency is not issue here, rather the categorical threshold. Retuning would be more persuasively demonstrated if the authors could provide a test of whether or not the responses for individual neurons differ significantly between conditions appropriately taking into account multiple comparisons, stability of recordings, non-Gaussian firing rate distributions across motif renditions, etc. and use this metric to report effects, rather than setting a categorical threshold.

eLife assessment

This important study discovered DBT as a novel gene implicated in the resistance to MG132-mediated cytotoxicity and potentially also in the pathogenesis of ALS and FTD, two fatal neurodegenerative diseases. The authors provided solid evidence to support a mechanism by which loss of DBT suppresses MG132-mediated toxicity. While activation of autophagy is shown to be associated with DBT knockdown, it remains unclear if this is the underlying mechanism driving improved survival.

Reviewer #1 (Public Review):

Summary:<br /> Through an unbiased genomewide KO screen, the authors identified loss of DBT to suppress MG132-mediated death of cultured RPE cells. Further analyses suggested that DBT reduces ubiquitinated proteins by promoting autophagy. Mechanistic studies indicated that DBT loss promotes autophagy via AMPK and its downstream ULK and mTOR signaling. Furthermore, loss of DBT suppresses polyglutamine- or TDP-43-mediated cytotoxicity and/or neurodegeneration in fly models. Finally, the authors showed that DBT proteins are increased in ALS patient tissues, compared to non-neurological controls.

Strengths:<br /> The idea is novel, the evidence is mostly convincing, and the data are clean. The findings have implications for human diseases.

Weaknesses:<br /> More experiments are needed to establish the connections between DBT and autophagy. The mechanistic studies are somewhat biased, and it's unclear whether the same mechanism (i.e., AMPK-->mTOR) can be applied to TDP-43-mediated neurodegeneration. Also, some data interpretation has to be more accurate.

Reviewer #2 (Public Review):

Summary:<br /> Hwang, Ran-Der et al utilized a CRISPR-Cas9 knockout in human retinal pigment epithelium (RPE1) cells to evaluate for suppressors of toxicity by the proteasome inhibitor MG132 and identified that knockout of dihydrolipoamide branched chain transacylase E2 (DBT) suppressed cell death. They show that DBT knockout in RPE1 cells does not alter proteasome or autophagy function at baseline. However, with MG132 treatment, they show a reduction in ubiquitinated proteins but with no change in proteasome function. Instead, they show that DBT knockout cells treated with MG132 have improved autophagy flux compared to wildtype cells treated with MG132. They show that MG132 treatment decreases ATP/ADP ratios to a greater extent in DBT knockout cells, and in accordance causes activation of AMPK. They then show downstream altered autophagy signaling in DBT knockout cells treated with MG132 compared to wild-type cells treated with MG132. Then they express the ALS mutant TDP43 M337 or expanded polyglutamine repeats to model Huntington's disease and show that knockdown of DBT improves cell survival in RPE1 cells with improved autophagic flux. They also utilize a Drosophila model and show that utilizing either a RNAi or CRISPR-Cas9 knockout of DBT improves eye pigment in TDP43M337V and polyglutamine repeat-expressing transgenic flies. Finally, they show evidence for increased DBT in postmortem spinal cord tissue from patients with ALS via both immunoblotting and immunofluorescence.

Strengths:<br /> This is a mechanistic and well-designed paper that identifies DBT as a novel regulator of proteotoxicity via activating autophagy in the setting of proteasome inhibition. Major strengths include careful delineation of a mechanistic pathway to define how DBT is protective. These conclusions are largely justified, but additional experiments and information would be useful to clarify and extend these conclusions.

Weaknesses:<br /> The large majority of the experiments are evaluating suppression of drug (MG132) toxicity in an in vitro epithelial cell line, so the generalizability to disease is unclear. Indeed, MG132 itself has been shown to modulate autophagy, and off-target effects of MG132 are not addressed. While this paper is strengthened by the inclusion of mouse-induced motor neurons, Drosophila models, and postmortem tissue, the putative mechanisms are minimally evaluated in these models.

Also, this effect is only seen with MG132 treatment, at a dose that causes markedly impaired cell survival. In this setting, it is certainly plausible that changes in autophagy could be the result of differences in cell survival, as opposed to an underlying mechanism for cell survival. Additional controls would be useful to increase confidence that DBT knockdown is protective via modulation of autophagy.

While the authors report increased DBT in postmortem ALS tissue as suggestive that DBT may modulate proteotoxicity in neurodegeneration, this point would be better supported with the evaluation of overexpression of DBT in their model.

eLife assessment

This fundamental study provides an unprecedented understanding of the roles of different combinations of NaV channel isoforms in nociceptors' excitability, with relevance for the design of better strategies targeting NaV channels to treat pain. Although the experimental combination of electrophysiological, modeling, imaging, molecular biology, and behavioral data is convincing and supports the major claims of the work, some conclusions need to be strengthened by further evidence or discussion. The work may be of broad interest to scientists working on pain, drug development, neuronal excitability, and ion channels.

Reviewer #1 (Public Review):

Summary:<br /> In this work, Xie, Prescott, and colleagues have reevaluated the role of Nav1.7 in nociceptive sensory neuron excitability. They find that nociceptors can make use of different sodium channel subtypes to reach equivalent excitability. The existence of this degeneracy is critical to understanding neuronal physiology under normal and pathological conditions and could explain why Nav subtype-selective drugs have failed in clinical trials. More concretely, nociceptor repetitive spiking relies on Nav1.8 at DIV0 (and probably under normal conditions in vivo), but on Nav1.7 and Nav1.3 at DIV4-7 (and after inflammation in vivo).

The conclusions of this paper are mostly well supported by data, and these findings should be of broad interest to scientists working on pain, drug development, neuronal excitability, and ion channels.

Strengths:<br /> The authors have employed elegant electrophysiology experiments (including specific pharmacology and dynamic clamp) and computational simulations to study the excitability of a subpopulation of DRGs that would very likely match with nociceptors (they take advantage of using transgenic mice to detect Nav1.8-expressing neurons). They make a strong point showing the degeneracy that occurs at the ion channel expression level in nociceptors, adding this new data to previous observations in other neuronal types. They also demonstrate that the different Nav subtypes functionally overlap and are able to interchange their "typical" roles in action potential generation. As Xie, Prescott, and colleagues argue, the functional implications of the degenerate character of nociceptive sensory neuron excitability need to be seriously taken into account regarding drug development and clinical trials with Nav subtype-selective inhibitors.

Weaknesses:<br /> The next comments are minor criticisms, as the major conclusions of the paper are well substantiated. Most of the results presented in the article have been obtained from experiments with DRG neuron cultures, and surely there is a greater degree of complexity and heterogeneity about the degeneracy of nociceptors excitability in the "in vivo" condition. Indeed, the authors show in Figures 7 and 8 data that support their hypothesis and an increased Nav1.7's influence on nociceptor excitability after inflammation, but also a higher variability in the nociceptors spiking responses. On the other hand, DRG neurons targeted in this study (YFP (+) after crossing with Nav1.8-Cre mice) are >90% nociceptors, but not all nociceptors express Nav1.8 in vivo. As shown by Li et al., 2016 ("Somatosensory neuron types identified by high-coverage single-cell RNA-sequencing and functional heterogeneity"), there is a high heterogeneity of neuron subtypes within sensory neurons. Therefore, some caution should be taken when translating the results obtained with the DRG neuron cultures to the more complex "in vivo" panorama.

Although the authors have focused their attention on Nav channels, it should be noted that degeneracy concerning other ion channels (such as potassium ion channels) could also impact the nociceptor excitability. The action potential AHP in Figure 1, panel A is very different comparing the DIV0 (blue) and DIV4-7 examples. Indeed, the conductance density values for the AHP current are higher at DIV0 than at DIV7 in the computational model (supplementary table 5). The role of other ion channels in order to obtain equivalent excitability should not be underestimated.

Reviewer #2 (Public Review):

Summary:<br /> The authors have noted in preliminary work that tetrodotoxin (TTX), which inhibits NaV1.7 and several other TTX-sensitive sodium channels, has differential effects on nociceptors, dramatically reducing their excitability under certain conditions but not under others. Partly because of this coincidental observation, the aim of the present work was to re-examine or characterize the role of NaV1.7 in nociceptor excitability and its effects on drug efficacy. The manuscript demonstrates that a NaV1.7-selective inhibitor produces analgesia only when nociceptor excitability is based on NaV1.7. More generally and comprehensively, the results show that nociceptors can achieve equivalent excitability through changes in differential NaV inactivation and NaV expression of different NaV subtypes (NaV 1.3/1.7 and 1.8). This can cause widespread changes in the role of a particular subtype over time. The degenerate nature of nociceptor excitability shows functional implications that make the assignment of pathological changes to a particular NaV subtype difficult or even impossible.

Thus, the analgesic efficacy of NaV1.7- or NaV1.8-selective agents depends essentially on which NaV subtype controls excitability at a given time point. These results explain, at least in part, the poor clinical outcomes with the use of subtype-selective NaV inhibitors and therefore have major implications for the future development of Nav-selective analgesics.

Strengths:<br /> The above results are clearly and impressively supported by the experiments and data shown. All methods are described in detail, presumably allow good reproducibility, and were suitable to address the corresponding question. The only exception is the description of the computer model, which should be described in more detail.

The results showing that nociceptors can achieve equivalent excitability through changes in differential NaV inactivation and expression of different NaV subtypes are of great importance in the fields of basic and clinical pain research and sodium channel physiology and pharmacology, but also for a broad readership and community. The degenerate nature of nociceptor excitability, which is clearly shown and well supported by data has large functional implications. The results are of great importance because they may explain, at least in part, the poor clinical outcomes with the use of subtype-selective NaV inhibitors and therefore have major implications for the future development of Nav-selective analgesics.

In summary, the authors achieved their overall aim to enlighten the role of NaV1.7 in nociceptor excitability and the effects on drug efficacy. The data support the conclusions, although the clinical implications could be highlighted in a more detailed manner.

Weaknesses:<br /> As mentioned before, the results that nociceptors can achieve equivalent excitability through changes in differential NaV inactivation and NaV expression of different NaV subtypes are impressive. However, there is some "gap" between the DRG culture experiments and acutely dissociated DRGs from mice after CFA injection. In the extensive experiments with cultured DRG neurons, different time points after dissociation were compared. Although it would have been difficult for functional testing to examine additional time points (besides DIV0 and DIV4-7), at least mRNA and protein levels should have been determined at additional time points (DIV) to examine the time course or whether gene expression (mRNA) or membrane expression (protein) changes slowly and gradually or rapidly and more abruptly. It would also be interesting to clarify whether the changes that occur in culture (DIV0 vs. DIV4-7) are accompanied by (pro-)inflammatory changes in gene and protein expression, such as those known for nociceptors after CFA injection. This would better link the following data demonstrating that in acutely dissociated nociceptors after CFA injection, the inflammation-induced increase in NaV1.7 membrane expression enhances the effect of (or more neurons respond to) the NaV1.7 inhibitor PF-71, whereas fewer CFA neurons respond to the NaV1.8 inhibitor PF-24.

The results shown explain, at least in part, the poor clinical outcomes with the use of subtype-selective NaV inhibitors and therefore have important implications for the future development of Nav-selective analgesics. However, this point, which is also evident from the title of the manuscript, is discussed only superficially with respect to clinical outcomes. In particular, the promising role of NaV1.7, which plays a role in nociceptor hyperexcitability but not in "normal" neurons, should be discussed in light of clinical results and not just covered with a citation of a review. Which clinical results of NaV1.7-selective drugs can now be better explained and how?

Another point directly related to the previous one, which should at least be discussed, is that all the data are from rodents, or in this case from mice, and this should explain the clinical data in humans. Even if "impediment to translation" is briefly mentioned in a slightly different context, one could (as mentioned above) discuss in more detail which human clinical data support the existence of "equivalent excitability through different sodium channels" also in humans.

Although speculative, it would be interesting for readers to know whether a treatment regimen based on "time since injury" with NaV1.7 and NaV1.8 inhibitors might offer benefits. Based on the data, could one hypothesize that NaV1.7 inhibitors are more likely to benefit (albeit in the short term) in patients with neuropathic pain with better patient selection (e.g., defined interval between injury and treatment)?

Reviewer #3 (Public Review):

Summary:<br /> In this study, the authors used patch-clamp to characterize the implication of various voltage-gated Na+ channels in the firing properties of mouse nociceptive sensory neurons. They report that depending on the culture conditions NaV1.3, NaV1.7, and NaV1.8 have distinct contributions to action potential firing and that similar firing patterns can result from distinct relative roles of these channels. The findings may be relevant for the design of better strategies targeting NaV channels to treat pain.

Strengths:<br /> The paper addresses the important issue of understanding, from an interesting perspective, the lack of success of therapeutic strategies targeting NaV channels in the context of pain. Specifically, the authors test the hypothesis that different NaV channels contribute in a plastic manner to action potential firing, which may be the reason why it is difficult to target pain by inhibiting these channels. The experiments seem to have been properly performed and most conclusions are justified. The paper is concisely written and easy to follow.

Weaknesses:<br /> 1) The most critical issue I find in the manuscript is the claim that different combinations of NaV channels result in equivalent excitability. For example, in the Abstract it is stated that: "...we show that nociceptors can achieve equivalent excitability using different combinations of NaV1.3, NaV1.7, and NaV1.8". The gating properties of these channels are not identical, and therefore their contributions to excitability should not be the same. I think that the culprit of this issue is that the authors reach their conclusion from the comparison of the (average) firing rate determined over 1 s current stimulation in distinct conditions. However, this is not the only parameter that determines how sensory neurons convey information. For instance, the time dependence of the instantaneous frequency, the actual firing pattern, may be important too. Moreover, the use of 1 s of current stimulation might not be sufficient to characterize the firing pattern if one wants to obtain conclusions that could translate to clinical settings (i.e., sustained pain). A neuron in which NaV1.7 is the main contributor is expected to have a damping firing pattern due to cumulative channel inactivation, whereas another depending mainly on NaV1.8 is expected to display more sustained firing. This is actually seen in the results of the modelling.

2) In Fig. 1, is 100 nM TTX sufficient to inhibit all TTX-sensitive NaV currents? More common in literature values to fully inhibit these currents are between 300 to 500 nM. The currents shown as TTX-sensitive in Fig. 1D look very strange (not like the ones at Baseline DIV4-7). It seems that 100 nM TTX was not enough, leading to an underestimation of the amplitude of the TTX-sensitive currents.

3) Page 8, the authors conclude that "Inflammation caused nociceptors to become much more variable in their reliance of specific NaV subtypes". However, how did the authors ensure that all neurons tested were affected by the CFA model? It could be that the heterogeneity in neuron properties results from distinct levels of effects of CFA.

eLife assessment

This useful study summarises the effect of optical stimulation of the A13 region on locomotion in healthy mice and experimental Parkinsonism and could potentially be of interest to basic and clinical neuroscientists. Behavioural analyses and evidence for pro-locomotor effects of stimulation are solid. However, anatomical analyses are incomplete and do not yield mechanistic insights due to various issues with specificity, sample size, statistical analysis, and data presentation in the present form of the study.

Reviewer #1 (Public Review):

Summary:<br /> This study aimed to investigate the effects of optically stimulating the A13 region in healthy mice and a unilateral 6-OHDA mouse model of Parkinson's disease (PD). The primary objectives were to assess changes in locomotion, motor behaviors, and the neural connectome. For this, the authors examined the dopaminergic loss induced by 6-OHDA lesioning. They found a significant loss of tyrosine hydroxylase (TH+) neurons in the substantia nigra pars compacta (SNc) while the dopaminergic cells in the A13 region were largely preserved. Then, they optically stimulated the A13 region using a viral vector to deliver the channelrhodopsine (CamKII promoter). In both sham and PD model mice, optogenetic stimulation of the A13 region induced pro-locomotor effects, including increased locomotion, more locomotion bouts, longer durations of locomotion, and higher movement speeds. Additionally, PD model mice exhibited increased ipsilesional turning during A13 region photoactivation. Lastly, the authors used whole-brain imaging to explore changes in the A13 region's connectome after 6-OHDA lesions. These alterations involved a complex rewiring of neural circuits, impacting both afferent and efferent projections. In summary, this study unveiled the pro-locomotor effects of A13 region photoactivation in both healthy and PD model mice. The study also indicates the preservation of A13 dopaminergic cells and the anatomical changes in neural circuitry following PD-like lesions that represent the anatomical substrate for a parallel motor pathway.

Strengths:<br /> These findings hold significant relevance for the field of motor control, providing valuable insights into the organization of the motor system in mammals. Additionally, they offer potential avenues for addressing motor deficits in Parkinson's disease (PD). The study fills a crucial knowledge gap, underscoring its importance, and the results bolster its clinical relevance and overall strength.

The authors adeptly set the stage for their research by framing the central questions in the introduction, and they provide thoughtful interpretations of the data in the discussion section. The results section, while straightforward, effectively supports the study's primary conclusion - the pro-locomotor effects of A13 region stimulation, both in normal motor control and in the 6-OHDA model of brain damage.

Weaknesses:<br /> 1) Anatomical investigation. I have a major concern regarding the anatomical investigation of plastic changes in the A13 connectome (Figures 4 and 5). While the methodology employed to assess the connectome is technically advanced and powerful, the results lack mechanistic insight at the cell or circuit level into the pro-locomotor effects of A13 region stimulation in both physiological and pathological conditions. This concern is exacerbated by a textual description of results that doesn't pinpoint precise brain areas or subareas but instead references large brain portions like the cortical plate, making it challenging to discern the implications for A13 stimulation. Lastly, the study is generally well-written with a smooth and straightforward style, but the connectome section presents challenges in readability and comprehension. The presentation of results, particularly the correlation matrices and correlation strength, doesn't facilitate biological understanding. It would be beneficial to explore specific pathways responsible for driving the locomotor effects of A13 stimulation, including examining the strength of connections to well-known locomotor-associated regions like the Pedunculopontine nucleus, Cuneiformis nucleus, LPGi, and others in the diencephalon, midbrain, pons, and medulla. Additionally, identifying the primary inputs to A13 associated with motor function would enhance the study's clarity and relevance.

The study raises intriguing questions about compensatory mechanisms in Parkinson's disease and a new perspective on the preservation of dopaminergic cells in A13, despite the SNc degeneration, and the plastic changes to input/output matrices. To gain inspiration for a more straightforward reanalysis and discussion of the results, I recommend the authors refer to the paper titled "Specific populations of basal ganglia output neurons target distinct brain stem areas while collateralizing throughout the diencephalon from the David Kleinfeld laboratory." This could guide the authors in investigating motor pathways across different brain regions.

2) Description of locomotor performance. Figure 3 provides valuable data on the locomotor effects of A13 region photoactivation in both control and 6-OHDA mice. However, a more detailed analysis of the changes in locomotion during stimulation would enhance our understanding of the pro-locomotor effects, especially in the context of 6-OHDA lesions. For example, it would be informative to explore whether the probability of locomotion changes during stimulation in the control and 6-OHDA groups. Investigating reaction time, speed, total distance, and even kinematic aspects during stimulation could reveal how A13 is influencing locomotion, particularly after 6-OHDA lesions. The laboratory of Whelan has a deep knowledge of locomotion and the neural circuits driving it so these features may be instructive to infer insights on the neural circuits driving movement. On the same line, examining features like the frequency or power of stimulation related to walking patterns may help elucidate whether A13 is engaging with the Mesencephalic Locomotor Region (MLR) to drive the pro-locomotor effects. These insights would provide a more comprehensive understanding of the mechanisms underlying A13-mediated locomotor changes in both healthy and pathological conditions.

Reviewer #3 (Public Review):

Kim, Lognon et al. present an important finding on pro-locomotor effects of optogenetic activation of the A13 region, which they identify as a dopamine-containing area of the medial zona incerta that undergoes profound remodeling in terms of afferent and efferent connectivity after administration of 6-OHDA to the MFB. The authors claim to address a model of PD-related gait dysfunction, a contentious problem that can be difficult to treat with dopaminergic medication or DBS in conventional targets. They make use of an impressive array of technologies to gain insight into the role of A13 remodeling in the 6-OHDA model of PD. The evidence provided is solid and the paper is well written, but there are several general issues that reduce the value of the paper in its current form, and a number of specific, more minor ones. Also, some suggestions, that may improve the paper compared to its recent form, come to mind.

The most fundamental issue that needs to be addressed is the relation of the structural to the behavioral findings. It would be very interesting to see whether the structural heterogeneity in afferent/effects projections induced by 6-OHDA is related to the degree of symptom severity and motor improvement during A13 stimulation.

The authors provide extensive interrogation of large-scale changes in the organization of the A13 region afferent and efferent distributions. It remains unclear how many animals were included to produce Fig 4 and 5. Fig S5 suggests that only 3 animals were used, is that correct? Please provide details about the heterogeneity between animals. Please provide a table detailing how many animals were used for which experiment. Were the same animals used for several experiments?

While the authors provide evidence that photoactivation of the A13 is sufficient in driving locomotion in the OFT, this pro-locomotor effect seems to be independent of 6-OHDA-induced pathophysiology. Only in the pole test do they find that there seems to be a difference between Sham vs 6-OHDA concerning the effects of photoactivation of the A13. Because of these behavioral findings, optogenic activation of A13 may represent a gain of function rather than disease-specific rescue. This needs to be highlighted more explicitly in the title, abstract, and conclusion.

The authors claim that A13 may be a possible target for DBS to treat gait dysfunction. However, the experimental evidence provided (in particular the lack of disease-specific changes in the OFT) seems insufficient to draw such conclusions. It needs to be highlighted that optogenetic activation does not necessarily have the same effects as DBS (see the recent review from Neumann et al. in Brain: https://pubmed.ncbi.nlm.nih.gov/37450573/). This is important because ZI-DBS so far had very mixed clinical effects. The authors should provide plausible reasons for these discrepancies. Is cell-specificity, which only optogenetic interventions can achieve, necessary? Can new forms of cyclic burst DBS achieve similar specificity (Spix et al, Science 2021)? Please comment.

In a recent study, Jeon et al (Topographic connectivity and cellular profiling reveal detailed input pathways and functionally distinct cell types in the subthalamic nucleus, 2022, Cell Reports) provided evidence on the topographically graded organization of STN afferents and McElvain et al. (Specific populations of basal ganglia output neurons target distinct brain stem areas while collateralizing throughout the diencephalon, 2021, Neuron) have shown similar topographical resolution for SNr efferents. Can a similar topographical organization of efferents and afferents be derived for the A13/ ZI in total?

In conclusion, this is an interesting study that can be improved by taking into consideration the points mentioned above.

eLife assessment

This study provides valuable insights into how researchers can use perceptual metamers to formally explore the limits of visual representations at different processing stages. While the study is overall convincing in terms of approach and results, issues were identified with respect to novelty, sample size, incomplete psychophysical methodology, and better motivation of the models tested.

Reviewer #1 (Public Review):

This is an interesting study of the nature of representations across the visual field. The question of how peripheral vision differs from foveal vision is a fascinating and important one. The majority of our visual field is extra-foveal yet our sensory and perceptual capabilities decline in pronounced and well-documented ways away from the fovea. Part of the decline is thought to be due to spatial averaging ('pooling') of features. Here, the authors contrast two models of such feature pooling with human judgments of image content. They use much larger visual stimuli than in most previous studies, and some sophisticated image synthesis methods to tease apart the prediction of the distinct models.

More importantly, in so doing, the researchers thoroughly explore the general approach of probing visual representations through metamers-stimuli that are physically distinct but perceptually indistinguishable. The work is embedded within a rigorous and general mathematical framework for expressing equivalence classes of images and how visual representations influence these. They describe how image-computable models can be used to make predictions about metamers, which can then be compared to make inferences about the underlying sensory representations. The main merit of the work lies in providing a formal framework for reasoning about metamers and their implications, for comparing models of sensory processing in terms of the metamers that they predict, and for mapping such models onto physiology. Importantly, they also consider the limits of what can be inferred about sensory processing from metamers derived from different models.

Overall, the work is of a very high standard and represents a significant advance over our current understanding of perceptual representations of image structure at different locations across the visual field. The authors do a good job of capturing the limits of their approach and I particularly appreciated the detailed and thoughtful Discussion section and the suggestion to extend the metamer-based approach described in the MS with observer models. The work will have an impact on researchers studying many different aspects of visual function including texture perception, crowding, natural image statistics, and the physiology of low- and mid-level vision.

The main weaknesses of the original submission relate to the writing. A clearer motivation could have been provided for the specific models that they consider, and the text could have been written in a more didactic and easy-to-follow manner. The authors could also have been more explicit about the assumptions that they make.

Reviewer #2 (Public Review):

Summary<br /> This paper expands on the literature on spatial metamers, evaluating different aspects of spatial metamers including the effect of different models and initialization conditions, as well as the relationship between metamers of the human visual system and metamers for a model. The authors conduct psychophysics experiments testing variations of metamer synthesis parameters including type of target image, scaling factor, and initialization parameters, and also compare two different metamer models (luminance vs energy). An additional contribution is doing this for a field of view larger than has been explored previously.

General Comments<br /> Overall, this paper addresses some important outstanding questions regarding comparing original to synthesized images in metamer experiments and begins to explore the effect of noise vs image seed on the resulting syntheses. While the paper tests some model classes that could be better motivated, and the results are not particularly groundbreaking, the contributions are convincing and undoubtedly important to the field. The paper includes an interesting Voronoi-like schematic of how to think about perceptual metamers, which I found helpful, but for which I do have some questions and suggestions. I also have some major concerns regarding incomplete psychophysical methodology including lack of eye-tracking, results inferred from a single subject, and a huge number of trials. I have only minor typographical criticisms and suggestions to improve clarity. The authors also use very good data reproducibility practices.

Specific Comments

Experimental Setup<br /> Firstly, the experiments do not appear to utilize an eye tracker to monitor fixation. Without eye tracking or another manipulation to ensure fixation, we cannot ensure the subjects were fixating the center of the image, and viewing the metamer as intended. While the short stimulus time (200ms) can help minimize eye movements, this does not guarantee that subjects began the trial with correct fixation, especially in such a long experiment. While Covid-19 did at one point limit in-person eye-tracked experiments, the paper reports no such restrictions that would have made the addition of eye-tracking impossible. While such a large-scale experiment may be difficult to repeat with the addition of eye tracking, the paper would be greatly improved with, at a minimum, an explanation as to why eye tracking was not included.

Secondly, many of the comparisons later in the paper (Figures 9,10) are made from a single subject. N=1 is not typically accepted as sufficient to draw conclusions in such a psychophysics experiment. Again, if there were restrictions limiting this it should be discussed. Also (P11) Is subject sub-00 is this an author? Other expert? A naive subject? The subject's expertise in viewing metamers will likely affect their performance.

Finally, the number of trials per subject is quite large. 13,000 over 9 sessions is much larger than most human experiments in this area. The reason for this should be justified.

Model<br /> For the main experiment, the authors compare the results of two models: a 'luminance model' that spatially pools mean luminance values, and an 'energy model' that spatially pools energy calculated from a multi-scale pyramid decomposition. They show that these models create metamers that result in different thresholds for human performance, and therefore different critical scaling parameters, with the basic luminance pooling model producing a scaling factor 1/4 that of the energy model. While this is certain to be true, due to the luminance model being so much simpler, the motivation for the simple luminance-based model as a comparison is unclear.

The authors claim that this luminance model captures the response of retinal ganglion cells, often modeled as a center-surround operation (Rodieck, 1964). I am unclear in what aspect(s) the authors claim these center-surround neurons mimic a simple mean luminance, especially in the context of evidence supporting a much more complex role of RGCs in vision (Atick & Redlich, 1992). Why do the authors not compare the energy model to a model that captures center-surround responses instead? Do the authors mean to claim that the luminance model captures only the pooling aspects of an RGC model? This is particularly confusing as Figures 6 and 9 show the luminance and energy models for original vs synth aligning with the scaling of Midget and Parasol RGCs, respectively. These claims should be more clearly stated, and citations included to motivate this. Similarly, with the energy model, the physiological evidence is very loosely connected to the model discussed.

Prior Work:<br /> While the explorations in this paper clearly have value, it does not present any particularly groundbreaking results, and those reported are consistent with previous literature. The explorations around critical eccentricity measurement have been done for texture models (Figure 11) in multiple papers (Freeman 2011, Wallis, 2019, Balas 2009). In particular, Freeman 20111 demonstrated that simpler models, representing measurements presumed to occur earlier in visual processing need smaller pooling regions to achieve metamerism. This work's measurements for the simpler models tested here are consistent with those results, though the model details are different. In addition, Brown, 2023 (which is miscited) also used an extended field of view (though not as large as in this work). Both Brown 2023, and Wallis 2019 performed an exploration of the effect of the target image. Also, much of the more recent previous work uses color images, while the author's exploration is only done for greyscale.

Discussion of Prior Work:<br /> The prior work on testing metamerism between original vs. synthesized and synthesized vs. synthesized images is presented in a misleading way. Wallis et al.'s prior work on this should not be a minor remark in the post-experiment discussion. Rather, it was surely a motivation for the experiment. The text should make this clear; a discussion of Wallis et al. should appear at the start of that section. The authors similarly cite much of the most relevant literature in this area as a minor remark at the end of the introduction (P3L72).

White Noise:<br /> The authors make an analogy to the inability of humans to distinguish samples of white noise. It is unclear however that human difficulty distinguishing samples of white noise is a perceptual issue- It could instead perhaps be due to cognitive/memory limitations. If one concentrates on an individual patch one can usually tell apart two samples. Support for these difficulties emerging from perceptual limitations, or a discussion of the possibility of these limitations being more cognitive should be discussed, or a different analogy employed.

Relatedly, in Figure 14, the authors do not explain why the white noise seeds would be more likely to produce syntheses that end up in different human equivalence classes.

It would be nice to see the effect of pink noise seeds, which mirror the power spectrum of natural images, but do not contain the same structure as natural images - this may address the artifacts noted in Figure 9b.

Finally, the authors note high-frequency artifacts in Figure 4 & P5L135, that remain after syntheses from the luminance model. They hypothesize that this is due to a lack of constraints on frequencies above that defined by the pooling region size. Could these be addressed with a white noise image seed that is pre-blurred with a low pass filter removing the frequencies above the spatial frequency constrained at the given eccentricity?

Schematic of metamerism:<br /> Figures 1,2,12, and 13 show a visual schematic of the state space of images, and their relationship to both model and human metamers. This is depicted as a Voronoi diagram, with individual images near the center of each shape, and other images that fall at different locations within the same cell producing the same human visual system response. I felt this conceptualization was helpful. However, implicitly it seems to make a distinction between metamerism and JND (just noticeable difference). I felt this would be better made explicit. In the case of JND, neighboring points, despite having different visual system responses, might not be distinguishable to a human observer.

In these diagrams and throughout the paper, the phrase 'visual stimulus' rather than 'image' would improve clarity, because the location of the stimulus in relation to the fovea matters whereas the image can be interpreted as the pixels displayed on the computer.

Other<br /> The authors show good reproducibility practices with links to relevant code, datasets, and figures.

eLife assessment

This study provides valuable findings about pre-saccadic foveal prediction and the extent to which it is influenced by the visibility of the saccade target relative to its background. The results and research methodology are technically solid, but there are questions about the interpretation of data which if addressed, would benefit our understanding of this phenomenon. This work should be of broad interest to visual neuroscientists, as well as those interested in understanding perception in the context of eye movements and in modeling visually guided actions.

Reviewer #1 Public Review:

Summary:<br /> This study examines to what extent this phenomenon varies based on the visibility of the saccade target. Visibility is defined as the contrast level of the target with respect to the noise background, and it is related to the signal-to-noise ratio of the target. A more visible target facilitates the oculomotor behavior planning and execution, however, as speculated by the authors, it can also benefit foveal prediction even if the foveal stimulus visibility is maintained constant. Remarkably, the authors show that presenting a highly visible saccade target is beneficial for foveal vision as the detection of stimuli with an orientation similar to that of the saccade target is improved, the lower the saccade target visibility, the less prominent the effect.

Strengths:<br /> The results are convincing and the research methodology is technically sound.

Weaknesses:<br /> Discussion on how this phenomenon may unfold in natural viewing conditions when the foveal and saccade target stimuli are complex and are constituted by different visual properties is lacking. Some speculations regarding feedforward vs feedback neural processing involved in the phenomenon and the speed of the feedforward signal in relation to the visibility of the target, are not well justified and not clearly supported by the data.

Reviewer #2 Public Review:

Summary:<br /> In this manuscript, the authors ran a dual task. Subjects monitored a peripheral location for a target onset (to generate a saccade to), and they also monitored a foveal location for a foveal probe. The foveal probe could be congruent or incongruent with the orientation of the peripheral target. In this study, the authors manipulated the conspicuity of the peripheral target, and they saw changes in performance in the foveal task. However, the changes were somewhat counterintuitive.

Strengths:<br /> The authors use solid analysis methods and careful experimental design.

Weaknesses:<br /> I have some issues with the interpretation of the results, as explained below. In general, I feel that a lot of effects are being explained by attention and target-probe onset asynchrony etc, but this seems to be against the idea put forth by the authors of "foveal prediction for visual continuity across saccades". Why would foveal prediction be so dependent on such other processes? This needs to be better clarified and justified.

Specifics:<br /> The explanation of decreased hit rates with increased peripheral target opacity is not convincing. The authors suggest that higher contrast stimuli in the periphery attract attention. But, then, why are the foveal results occurring earlier (as per the later descriptions in the manuscript)? And, more importantly, why would foveal prediction need to be weaker with stronger pre-saccadic attention to the periphery? What is the function of foveal prediction? What of the other interpretation that could be invoked in general for this type of task used by the authors: that the dual task is challenging and that subjects somehow misattribute what they saw in the peripheral task when planning the saccade. i.e. foveal hit rates are misperceptions of the peripheral target. When the peripheral target is easier to see, then the foveal hit rate drops.

The analyses of Fig. 3C appear to be overly convoluted. They also imply an acknowledgment by the authors that target-probe temporal difference matters. Doesn't this already negate the idea that the foveal effects are associated with the saccade generation process itself? If the effect is related to target onset, how is it interpreted as related to a foveal prediction that is associated with the saccade itself? Also, the oscillatory nature of the effect in Fig. 3C for 59% and 90% opacity is quite confusing and not addressed. The authors simply state that enhancement occurs earlier before the saccade for higher contrasts. But, this is not entirely true. The enhancement emerges then disappears and then emerges again leading up to the saccade. Why would foveal prediction do that?

The interpretation of Fig. 4 is also confusing. Doesn't the longer latency already account for the lapse in attention, such that visual continuity can proceed normally now that the saccade is actually eventually made? In all results, it seems that the effects are all related to the dual nature of the task and/or attention, rather than to the act of making the saccade itself. Why should visual continuity (when a saccade is actually made, whether with short or long latency) have different "fidelity"? And, isn't this disruptive to the whole idea of visual continuity in the first place?

Small question: is it just me or does the data in general seem to be too excessively smoothed?

eLife assessment

In this study, the authors develop a useful strategy for fluorophore-tagging endogenous proteins in human induced pluripotent stem cells (iPSCs) using a split mNeonGreen approach. Experimentally, the methods are solid, and the data presented support the author's conclusions. Overall, these methodologies should be useful to a wide audience of cell biologists who want to study protein localization and dynamics at endogenous levels in iPSCs.

Reviewer #1 (Public Review):

Summary:<br /> In this manuscript, the authors have applied an asymmetric split mNeonGreen2 (mNG2) system to human iPSCs. Integrating a constitutively expressed long fragment of mNG2 at the AAVS1 locus, allows other proteins to be tagged through the use of available ssODN donors. This removes the need to generate long AAV donors for tagging, thus greatly facilitating high-throughput tagging efforts. The authors then demonstrate the feasibility of the method by successfully tagging 9 markers expressed in iPSC at various, and one expressed upon endoderm differentiation. Several additional differentiation markers were also successfully tagged but not subsequently tested for expression/visibility. As one might expect for high-throughput tagging, a few proteins, while successfully tagged at the genomic level, failed to be visible. Finally, to demonstrate the utility of the tagged cells, the authors isolated clones with genes relevant to cytokinesis tagged, and together with an AI to enhance signal-to-noise ratios, monitored their localization over cell division.

Strengths:<br /> Characterization of the mNG2 tagged parental iPSC line was well and carefully done including validation of a single integration, the presence of markers for continued pluripotency, selected off-target analysis, and G-banding-based structural rearrangement detection.

The ability to tag proteins with simple ssODNs in iPSC capable of multi-lineage differentiation will undoubtedly be useful for localization tracking and reporter line generation.

Validation of clone genotypes was carefully performed and highlights the continued need for caution with regard to editing outcomes.

Weaknesses:<br /> IF and flow cytometry figures lack quantification and information on replication. How consistent is the brightness and localization of the markers? How representative are the specific images? Stability is mentioned in the text but data on the stability of expression/brightness is not shown.

The localization of markers, while consistent with expectations, is not validated by a second technique such as antibody staining, and in many cases not even with Hoechst to show nuclear vs cytoplasmic.

For the multi-germ layer differentiation validation, NCAM is also expressed by ectoderm, so isn't a good solo marker for mesoderm as it was used. Indeed, the kit used for the differentiation suggests Brachyury combined with either NCAM or CXCR4, not NCAM alone.

Only a single female parental line has been generated and characterized. It would have been useful to have several lines and both male and female to allow sex differences to be explored.

The AI-based signal-to-noise enhancement needs more details and testing. Such models can introduce strong assumptions and thus artefacts into the resolved data. Was the model trained on all markers or were multiple models trained on a single marker each? For example, if trained to enhance a single marker (or co-localized group of markers), it could introduce artefacts where it forces signal localization to those areas even for others. What happens if you feed in images with scrambled pixel locations, does it still say the structures are where the training data says they should be? What about markers with different localization from the training set? If you feed those in, does it force them to the location expected by the training data or does it retain their differential true localization and simply enhance the signal?

Reviewer #2 (Public Review):

Summary:<br /> The authors have generated human iPSC cells constitutively expressing the mNG21-10 and tested them by endogenous tagging multiple genes with mNG211 (several tagged iPS cell lines clones were isolated). With this tool, they have explored several weakly expressed cytokinesis genes and gained insights into how cytokinesis occurs.

Strengths:<br /> Human iPSC cells are used.

Weaknesses:<br /> i) The manuscript is extremely incremental, no improvements are present in the split-fluorescent (split-FP) protein variant used nor in the approach for endogenous tagging with split-FPs (both of them are already very well established and used in literature as well as in different cell types).

ii) The fluorescence intensity of the split mNeonGreen appears rather low, for example in Figure 2C the H2BC11, ANLN, SOX2, and TUBB3 signals are very noisy (differences between the structures observed are almost absent). For low-expression targets, this is an important limitation. This is also stated by the authors but image restoration could not be the best solution since a lot of biologically relevant information will be lost anyway.

iii) There is no comparison with other existing split-FP variants, methods, or imaging and it is unclear what the advantages of the system are.

Reviewer #3 (Public Review):

The authors report on the engineering of an induced Pluripotent Stem Cell (iPSC) line that harbours a single copy of a split mNeonGreen, mNG2(1-10). This cell line is subsequently used to take endogenous protein with a smaller part of mNeonGreen, mNG2(11), enabling the complementation of mNG into a fluorescent protein that is then used to visualize the protein. The parental cell is validated and used to construct several iPSC lines with endogenously tagged proteins. These are used to visualize and quantify endogenous protein localisation during mitosis.

I see the advantage of tagging endogenous loci with small fragments, but the complementation strategy has disadvantages that deserve some attention. One potential issue is the level of the mNG2(1-10). Is it clear that the current level is saturating? Based on the data in Figure S3, the expression levels and fluorescence intensity levels show a similar dose-dependency which is reassuring, but not definitive proof that all the mNG2(11)-tagged protein is detected.

Do the authors see a difference in fluorescence intensity for homo- and heterozygous cell lines that have the same protein tagged with mNG2(11)? One would expect two-fold differences, or not?

Related to this, would it be favourable to have a homozygous line for expressing mNG2(1-10)?

The complementation seems to work well for the proteins that are tested. Would this also work for secreted (or other organelle-resident) proteins, for which the mNG2(11) tag is localised in a membrane-enclosed compartment?

The authors present a technological advance and it would be great if others could benefit from this as well by having access to the cell lines.

eLife assessment

This study describes a method to track MHC class II binding peptides on dendritic cell (DC) surfaces using a tetracystein tag and a thiol-reactive dye, which can then be investigated in vitro and in vivo. This is a valuable study for the impact on immunology and potentially other areas where the detection of cell-associated peptides is required. The methods are convincing based on the use of MHC class I/II deficient mice that have significantly reduced signal, but the non-zero background is detected, and it is not clear that this is lower than if the peptides were directly labelled with fluorophores.

Reviewer #1 (Public Review):

Summary:<br /> The authors develop a method to fluorescently tag peptides loaded onto dendritic cells using a two-step method with a tetracystein motif modified peptide and labelling step done on the surface of live DC using a dye with high affinity for the added motif. The results are convincing in demonstrating in vitro and in vivo T cell activation and efficient label transfer to specific T cells in vivo. The label transfer technique will be useful to identify T cells that have recognised a DC presenting a specific peptide antigen to allow the isolation of the T cell and cloning of its TCR subunits, for example. It may also be useful as a general assay for in vitro or in vivo T-DC communication that can allow the detection of genetic or chemical modulators.

Strengths:<br /> The study includes both in vitro and in vivo analysis including flow cytometry and two-photon laser scanning microscopy. The results are convincing and the level of T cell labelling with the fluorescent pMHC is surprisingly robust and suggests that the approach is potentially revealing something about fundamental mechanisms beyond the state of the art.

Weaknesses:<br /> The method is demonstrated only at high pMHC density and it is not clear if it can operate at at lower peptide doses where T cells normally operate. However, this doesn't limit the utility of the method for applications where the peptide of interest is known. It's not clear to me how it could be used to de-orphan known TCR and this should be explained if they want to claim this as an application. Previous methods based on biotin-streptavidin and phycoerythrin had single pMHC sensitivity, but there were limitations to the PE-based probe so the use of organic dyes could offer advantages.

Reviewer #2 (Public Review):

Summary:<br /> The authors here develop a novel Ovalbumin model peptide that can be labeled with a site-specific FlAsH dye to track agonist peptides both in vitro and in vivo. The utility of this tool could allow better tracking of activated polyclonal T cells particularly in novel systems. The authors have provided solid evidence that peptides are functional, capable of activating OTII T cells, and that these peptides can undergo trogocytosis by cognate T cells only.

Strengths:<br /> -An array of in vitro and in vivo studies are used to assess peptide functionality.<br /> -Nice use of cutting-edge intravital imaging.<br /> -Internal controls such as non-cogate T cells to improve the robustness of the results (such as Fig 5A-D).<br /> -One of the strengths is the direct labeling of the peptide and the potential utility in other systems.

Weaknesses:<br /> 1. What is the background signal from FlAsH?<br /> The baselines for Figure 1 flow plots are all quite different. Hard to follow. What does the background signal look like without FLASH (how much fluorescence shift is unlabeled cells to No antigen+FLASH?). How much of the FlAsH in cells is actually conjugated to the peptide? In Figure 2E, it doesn't look like it's very specific to pMHC complexes. Maybe you could double-stain with Ab for MHCII. Figure 4e suggests there is no background without MHCII but I'm not fully convinced. Potentially some MassSpec for FLASH-containing peptides.

2. On the flip side, how much of the variant peptides are getting conjugated in cells? I'd like to see some quantification (HPLC or MassSpec). If it's ~10% of peptides that get labeled, this could explain the low shifts in fluorescence and the similar T cell activation to native peptides if FlasH has any deleterious effects on TCR recognition. But if it's a high rate of labeling, then it adds confidence to this system.

3. Conceptually, what is the value of labeling peptides after loading with DCs? Why not preconjugate peptides with dye, before loading, so you have a cleaner, potentially higher fluorescence signal? If there is a potential utility, I do not see it being well exploited in this paper. There are some hints in the discussion of additional use cases, but it was not clear exactly how they would work. One mention was that the dye could be added in real-time in vivo to label complexes, but I believe this was not done here. Is that feasible to show?

4. Figure 5D-F the imaging data isn't fully convincing. For example, in 5F and 2G, the speeds for T cells with no Ag should be much higher (10-15micron/min or 0.16-0.25micron/sec). The fact that yours are much lower speeds suggests technical or biological issues, that might need to be acknowledged or use other readouts like the flow cytometry.

eLife assessment

This valuable work by Rivera et al. probes to understand how the regulation of oligodendrocyte progenitor cell (OPC) remyelination and function contributes to the treatment of multiple sclerosis. The authors provide incomplete evidence for the platelets to mediate OPC differentiation and remyelination. Both reviewers have raised important questions. This work will be of broad interest to biologists in general.

Reviewer #1 (Public Review):

Summary:<br /> The authors have studied the effects of platelets in OPC biology and remyelination. For this, they used mutant mice with lower levels of platelets as a demyelinating/remyelinating scenario, as well as in a model with large numbers of circulating platelets.

Strengths:<br /> -The work is very focused, with defined objectives.<br /> -The work is properly done.

Weaknesses:<br /> -There is no clear effect on a single cell type and/or mechanism involved.

Reviewer #2 (Public Review):

Summary:<br /> This paper examined whether circulating platelets regulate oligodendrocyte progenitor cell (OPC) differentiation for the link with multiple sclerosis (MS). They identified that the interaction with platelets enhances OPC differentiation although persistent contact inhibits the process in the long-term. The mouse model with increased platelet levels in the blood reduced mature oligodendrocytes, while how platelets might regulate OPC differentiation is not clear yet.

Strengths:<br /> The use of both partial platelet depletion and thrombocytosis mouse models gives in vivo evidence. The presentation of platelet accumulation in a time-course manner is rigorous. The in vitro co-culture model tested the role of platelets in OPC differentiation, which was supportive of in vivo observations.

Weaknesses:<br /> How platelets regulate OPC differentiation is not clear. What the significance of platelets is in MS progression is not clear.

eLife assessment

This study presents a valuable finding on the mechanism of glucocorticoid-induced osteonecrosis of the femoral head. The data were collected and analyzed using solid and validated methodology and can be used as a starting point for functional studies of the development of glucocorticoid-induced osteonecrosis. This paper would be of interest to cell biologists and biophysicists working on the potential pharmacological treatments for glucocorticoid-induced osteonecrosis.

Reviewer #1 (Public Review):

Summary:<br /> The manuscript by Xia et al. investigated the mechanisms underlying Glucocorticoid-induced osteonecrosis of the femoral head (GONFH). The authors observed that abnormal osteogenesis and adipogenesis are associated with decreased β-catenin in the necrotic femoral head of GONFH patients, and that the inhibition of β-catenin signalling leads to abnormal osteogenesis and adipogenesis in GONFH rats. Of interest, the deletion of β-catenin in Col2-expressing cells rather than in osx-expressing cells leads to a GONFH-like phenotype in the femoral head of mice.

Strengths:<br /> A strength of the study is that it sets up a Col2-expressing cell-specific β-catenin knockout mouse model that mimics the full spectrum of osteonecrosis phenotype of GONFH. This is interesting and provides new insights into the understanding of GONFH. Overall, the data are solid and support their conclusions.

Reviewer #2 (Public Review):

Summary:<br /> In this manuscript, the authors reported a study to uncover that β-catenin inhibition disrupting the homeostasis of osteogenic/adipogenic differentiation contributes to the development of Glucocorticoid-induced osteonecrosis of the femoral head (GONFH). In this study, they first observed abnormal osteogenesis and adipogenesis associated with decreased β-catenin in the necrotic femoral head of GONFH patients, but the exact pathological mechanisms of GONFH remain unknown. They then performed in vivo and in vitro studies to further reveal that glucocorticoid exposure disrupted osteogenic/adipogenic differentiation bone marrow stromal cells (BMSCs) by inhibiting β-catenin signaling in glucocorticoid-induced GONFH rats, and specific deletion of β-catenin in Col2+ cells shifted BMSCs commitment from osteoblasts to adipocytes, leading to a full spectrum of disease phenotype of GONFH in adult mice.

Strengths:<br /> This innovative study provides strong evidence supporting that β-catenin inhibition disrupts the homeostasis of osteogenic/adipogenic differentiation that contributes to the development of GONFH. This study also identifies an ideal genetically modified mouse model of GONFH. Overall, the experiment is logically designed, the figures are clear, and the data generated from humans and animals is abundant supporting their conclusions.

Weaknesses:<br /> There is a lack of discussion to explain how the Wnt agonist 1 works. There are several types of Wnt ligands. It is not clear if this agonist only targets Wnt1 or other Wnts as well. Also, why Wnt agonist 1 couldn't rescue the GONFH-like phenotype in β-cateninCol2ER mice needs to be discussed.

Reviewer #3 (Public Review):

Summary:<br /> In this manuscript, the authors are trying to delineate the mechanism underlying the osteonecrosis of the femoral head.

Strengths:<br /> The authors provided compelling in vivo and in vitro data to demonstrate Col2+ cells and Osx+ cells were differentially expressed in the femoral head. Moreover, inducible knockout of β-catenin in Col2+ cells but not Osx+ cells lead to a GONFH-like phenotype including fat accumulation, subchondral bone destruction, and femoral head collapse, indicating that imbalance of osteogenic/adipogenic differentiation of Col2+ cells plays an important role in GONFH pathogenesis. Therefore, this manuscript provided mechanistic insights into osteonecrosis as well as potential therapeutic targets for disease treatment.

Weaknesses:<br /> However, additional in-depth discussion regarding the phenotype observed in mice is highly encouraged.

eLife assessment

The authors have shown a valuable tumor suppressive function of the non-core regions of RAG1/2 recombinases, by using a set of animal models. The work is solid and the conclusions are supported by their data. Some areas of mechanistic work can be improved.

Reviewer #1 (Public Review):

Summary:<br /> In this report, Yu et al ascribe potential tumor suppressive functions to the non-core regions of RAG1/2 recombinases. Using a well-established BCR-ABL oncogene-driven system, the authors model the development of B cell acute lymphoblastic leukemia in mice and found that RAG mutants lacking non-core regions show accelerated leukemogenesis. They further report that the loss of non-core regions of RAG1/2 increases genomic instability, possibly caused by increased off-target recombination of aberrant RAG-induced breaks. The authors conclude that the non-core regions of RAG1 in particular not only increase the fidelity of VDJ recombination, but may also influence the recombination "range" of off-target joints, and that in the absence of the non-core regions, mutant RAG1/2 (termed cRAGs) catalyze high levels of off-target recombination leading to the development of aggressive leukemia.

Strengths:<br /> The authors used a genetically defined oncogene-driven model to study the effect of RAG non-core regions on leukemogenesis. The animal studies were well performed and generally included a good number of mice. Therefore, the finding that cRAG expression led to the development of more aggressive BCR-ABL+ leukemia compared to fRAG is solid.

Weaknesses:<br /> In general, I find the mechanistic explanation offered by the authors to explain how the non-core regions of RAG1/2 suppress leukemogenesis to be less convincing. My main concern is that cRAG1 and cRAG2 are overexpressed relative to fRAG1/2. This raises the possibility that the observed increased aggressiveness of cRAG tumors compared to fRAG tumors could be solely due to cRAG1/2 overexpression, rather than any intrinsic differences in the activity of cRAG1/2 vs fRAG1/2; and indeed, the authors allude to this possibility in Fig S8, where it was shown that elevated expression of RAG (i.e. fRAG) correlated with decreased survival in pediatric ALL. Although it doesn't mean the authors' assertions are incorrect, this potential caveat should nevertheless be discussed.

Some of the conclusions drawn were not supported by the data.<br /> 1. I'm not sure that the authors can conclude based on μHC expression that there is a loss of pre-BCR checkpoint in cRAG tumors. In fact, Fig. 2B showed that the differences are not statistically significant overall, and more importantly, μHC expression should be detectable in small pre-B cells (CD43-). This is also corroborated by the authors' analysis of VDJ rearrangements, showing that it has occurred at the H chain locus in cRAG cells.

2. The authors found a high degree of polyclonal VDJ rearrangements in fRAG tumor cells but a much more limited oligoclonal VDJ repertoire in cRAG tumors. They concluded that this explains why cRAG tumors are more aggressive because BCR-ABL induced leukemia requires secondary oncogenic hits, resulting in the outgrowth of a few dominant clones (Page 19, lines 381-398). I'm not sure this is necessarily a causal relationship since we don't know if the oligoclonality of cRAG tumors is due to selection based on oncogenic potential or if it may actually reflect a more restricted usage of different VDJ gene segments during rearrangement.

3. What constitutes a cancer gene can be highly context- and tissue-dependent. Given that there is no additional information on how any putative cancer gene was disrupted (e.g., truncation of regulatory or coding regions), it is not possible to infer whether increased off-target cRAG activity really directly contributed to the increased aggressiveness of leukemia.

4. Fig. 6A, it seems that it is really the first four nucleotide (CACA) that determines fRAG binding and the first three (CAC) that determine cRAG binding, as opposed to five for fRAG and four for cRAG, as the author wrote (page 24, lines 493-497).

5. Fig S3B, I don't really see why "significant variations in NHEJ" would necessarily equate "aberrant expression of DNA repair pathways in cRAG leukemic cells". This is purely speculative. Since it has been reported previously that alt-EJ/MMEJ can join off target RAG breaks, do the authors detect high levels of microhomology usage at break points in cRAG tumors?

6. Fig. S7, CDKN2B inhibits CDK4/6 activation by cyclin D, but I don't think it has been shown to regulate CDK6 mRNA expression. The increase in CDK6 mRNA likely just reflects a more proliferative tumor but may have nothing to do with CDKN2B deletion in cRAG1 tumors.

Insufficient details in some figures. For instance, Fig. 1A, please include statistics in the plot showing a comparison of fRAG vs cRAG1, fRAG vs cRAG2, cRAG1 vs cRAG2. As of now, there's a single p-value (0.0425) stated in the main text and the legend but why is there only one p-value when fRAG is compared to cRAG1 or cRAG2? Similarly, the authors wrote "median survival days 11-26, 10-16, 11-21 days, P < 0.0023-0.0299, Fig. S2B." However, it is difficult for me to figure out what are the numbers referring to. For instance, is 11-26 referring to median survival of fRAG inoculated with three different concentrations of GFP+ leukemic cells or is 11-26 referring to median survival of fRAG, cRAG1, cRAG2 inoculated with 10^5 cells? It would be much clearer if the authors can provide the numbers for each pair-wise comparison, if not in the main text, then at least in the figure legend. In Fig. 5A-B, do the plots depict SVs in cRAG tumors or both cRAG and fRAG cells? Also in Fig. 5, why did 24 SVs give rise to 42 breakpoints, and not 48? Doesn't it take 2 breaks to accomplish rearrangement? In Fig. 6B-C, it is not clear how the recombination sizes were calculated. In the examples shown in Fig. 4, only cRAG1 tumors show intra-chromosomal joins (chr 12), while fRAG and cRAG2 tumors show exclusively inter-chromosomal joins.

Insufficient details on certain reagents/methods. For instance, are the cRAG1/2 mice of the same genetic background as fRAG mice (C57BL/6 WT)? On Page 23, line 481, what is a cancer gene? How are they defined? In Fig. 3C, are the FACS plots gated on intact cells? Since apoptotic cells show high levels of gH2AX, I'm surprised that the fraction of gH2AX+ cells is so much lower in fRAG tumors compared to cRAG tumors. The in vitro VDJ assay shown in Fig 3B is not described in the Method section (although it is described in Fig S5b). Fig. 5A-B, do the plots depict SVs in cRAG tumors or both cRAG and fRAG cells?

Reviewer #2 (Public Review):

Summary: In the manuscript, the authors summarized and introduced the correlation between the non-core regions of RAG1 and RAG2 in BCR-ABL1+acute B lymphoblastic leukemia and off-target recombination which has certain innovative and clinical significance.

eLife assessment

The authors studied key characteristics of MYC-driven cancers: dysregulated pre-mRNA splicing and altered metabolism. The reviewers agree that this is an interesting study and that the findings are important. Overall the data was considered solid although the paper would benefit from revisions. The manuscript has the potential to be of broad interest to cancer biologists due to its therapeutic implications.

Reviewer #1 (Public Review):

Summary of Author's Objectives:

The authors aimed to explore JMJD6's role in MYC-driven neuroblastoma, particularly in the interplay between pre-mRNA splicing and cancer metabolism, and to investigate the potential for targeting this pathway.

Strengths:

1. The study employs a diverse range of experimental techniques, including molecular biology assays, next-generation sequencing, interactome profiling, and metabolic analysis. Moreover, the authors specifically focused on gained chromosome 17q in neuroblastoma, in combination with analyzing cancer dependency genes screened with Crispr/Cas9 library, analyzing the association of gene expression with prognosis of neuroblastoma patients with large clinical cohort. This comprehensive approach strengthens the credibility of the findings. The identification of the link between JMJD6-mediated pre-mRNA splicing and metabolic reprogramming in MYC-driven cancer cells is innovative.

2. The authors effectively integrate data from multiple sources, such as gene expression analysis, RNA splicing analysis, JMJD6 interactome assay, and metabolic profiling. This holistic approach provides a more complete understanding of JMJD6's role.

3. The identification of JMJD6 as a potential therapeutic target and its correlation with the response to indisulam have significant clinical implications, addressing an unmet need in cancer treatment.

Weaknesses:

1. The manuscript contains complex technical details and terminology that may pose challenges for readers without a deep background in molecular biology and cancer research. Providing simplified explanations or additional context would enhance accessibility.

2. It would be beneficial to explore whether treatment with JMJD6 inhibitors, both in vitro and in vivo, can effectively target the enhanced pre-mRNA splicing of metabolic genes in MYC-driven cancer cells.

Appraisal of Achievement and Conclusion Support:

The authors have effectively met their objectives by offering valuable insights into JMJD6's role in MYC-driven neuroblastoma. The results robustly underpin their conclusions about JMJD6's contribution to metabolic reprogramming through alternative splicing and its connection to the therapeutic response to indisulam.

Likely Impact on the Field and Utility of Methods/Data:

The study's findings have the potential to significantly impact the field of cancer research by identifying JMJD6 as a promising therapeutic target for MYC-driven cancers. The methods and data presented in the manuscript offer valuable resources to the research community for further investigations into cancer metabolism and splicing regulation.

Additional Context for Interpretation:

Understanding the complex interplay between cancer metabolism and splicing regulation is crucial for developing effective cancer treatments. This study sheds light on a previously poorly understood aspect of MYC-driven cancers and opens new avenues for targeted therapies. However, the transition from preclinical findings to clinical applications may face challenges, which should be considered in future research and clinical trials.

Reviewer #2 (Public Review):

Summary:

Jablonowski and colleagues studied key characteristics of MYC-driven cancers: dysregulated pre-mRNA splicing and altered metabolism. This is an important field of study as it remains largely unclear as to how these processes are coordinated in response to malignant transformation and how they are exploitable for future treatments. In the present study, the authors attempt to show that Jumonji Domain Containing 6, Arginine Demethylase And Lysine Hydroxylase (JMJD6) plays a central role in connecting pre-mRNA splicing and metabolism in MYC-driven neuroblastoma. JMJD6 collaborates with the MYC protein in driving cellular transformation by physically interacting with RNA-binding proteins involved in pre-mRNA splicing and protein regulation. In cell line experiments, JMJD6 affected the alternative splicing of two forms of glutaminase (GLS), an essential enzyme in the glutaminolysis process within the central carbon metabolism of neuroblastoma cells. Additionally, the study provides in vitro (and in silico) evidence for JMJD6 being associated with the anti-proliferation effects of a compound called indisulam, which degrades the splicing factor RBM39, known to interact with JMJD6.

Overall, the findings presented by Jabolonowski et al. begin to illuminate a cancer-promoting metabolic, and potentially, a protein synthesis suppression program that may be linked to alternative pre-mRNA splicing through the action of JMJD6 - downstream of MYC. This discovery can provide further evidence for considering JMJD6 as a potential therapeutic target for the treatment of MYC-driven cancers.

Strengths:

Alternative Splicing Induced by JMJD6 Knockdown: the study presents evidence for the role of JMJD6 in alternative splicing in neuroblastoma cells. Specifically, the RNA immunoprecipitation experiments demonstrated a significant shift from the GAC to the KGA GLS isoform upon JMJD6 knockdown. Moreover, a significant correlation between JMJD6 levels and GAC/KGA isoform expression was identified in two distinct neuroblastoma cohorts. This suggests a causative link between JMJD6 activity and isoform prevalence.

Physical Interaction of JMJD6 in Neuroblastoma Cells: The paper provides preliminary insight into the physical interactome of JMJD6 in neuroblastoma cells. This offers a potential mechanistic avenue for the observed effects on metabolism and protein synthesis and could be exploited for a deeper investigation into the exact nature, and implications of neuroblastoma-specific JMJD6 protein-protein interactions.

Weaknesses:

There are several areas that would benefit from improvements with regard to the current data supporting the claims of the paper (i.e., the conclusion presented in Figure 8).

Neuroblastoma Modelling Strategy: The study heavily relies on cell lines without incorporating patient-derived cells/biomaterials. Using databases to fill gaps in the experimental design can only fortify the observations to a certain extent. A critical oversight is the absence of non-cancerous control cells in many figures, and the rationale for selecting specific cell lines for assays/approaches remains somewhat unclear. A foundational control for such experiments should involve the non-transformed neural crest cell line, which the authors have readily available. Are the observed splicing and metabolic effects of JMJD6 specific to neuroblastoma? Is there a neuroblastoma-specific JMJD6 interactome? Is MYC function essential?

In Vivo Modelling: The inclusion of a genetic mouse model combined with an inducible JMJD6 knockdown, would enhance the study by allowing examination of JMJD6's role during both tumor initiation and growth in vivo. For instance, the TH-MYCN mice overexpressing MYCN in neural crest cells, could be a promising choice.

Dependence on Colony Formation Assay: The study leans on 2D and semi-quantitative colony formation assays to assess malignant growth. To validate the link between the mechanistic insights discussed (e.g., reduced protein synthesis) and JMJD6-mediated malignant growth as a potential therapeutic target, evidence from in vivo or representative 3D models would be crucial.

Data Presentation and Rigor: The presented data is predominantly qualitative and necessitates quantification. For instance, Western blots should be quantified. The RNAseq, metabolism, and pull-down data should be transparently and numerically presented. The figure legends seem elusive and their lack of transparency (often with regards to biological repeats, error bars, cell line used etc.) is concerning. Adequate citation and identification of all data sources, including online resources, are imperative. The manuscript would also benefit from a more rigorous depiction and quantification of RNA interference of both stable and transient knockdowns with quantitative validation at mRNA and protein levels.

Novelty Concerns: The emphasis on JMJD6 as a novel neuroblastoma target is contingent on the new mechanistic revelations about the JMJD6-centered link between splicing, metabolism, and protein synthesis. Given that JMJD6 has been previously linked to neuroblastoma biology, the rationale (particularly in Figure 1) for concentrating on JMJD6 may stem more from bias rather than data-driven reasoning.

Depth of Mechanistic Investigation: Current evidence lacks depth in key areas such as JMJD6-RNA binding. A more thorough approach would involve pinpointing specific JMJD6 binding sites on endogenous RNAs using techniques such as cross-linking and immunoprecipitation, paired with complementary proximity-based methodologies. Regarding the presented metabolism data, diving deeper into metabolic flux via isotope labeling experiments could shed light on dynamic processes like TCA and glutaminolysis. As it stands, the 'pathway cartoon' in Figure 6d appears overly qualitative.

eLife assessment

This paper proposes a valuable new method for the assessment of the mean kurtosis for diffusional kurtosis imaging by utilizing a recently introduced sub-diffusion model. The evidence supporting the claims that this technique is robust and accurate in brain imaging is incomplete. The work could be of interest in the research and clinical arena.

Reviewer #1 (Public Review):

Summary:<br /> This study introduces an innovative method for assessing the mean kurtosis, utilizing the mathematical foundation of the sub-diffusion framework. In particular, a new fitting technique that incorporates two different diffusion times is proposed to estimate the parameters of the sub-diffusion model. The evaluation of this technique, which generates kurtosis maps based on the sub-diffusion framework, is conducted through simulations and the examination of data obtained from human subjects.

Strengths:<br /> The utilization of the sub-diffusion model for tissue characterization is a significant conceptual advancement for the field of diffusion MRI. This study adeptly harnesses this approach for an accurate estimation of the parameters of the widely employed diffusion model, DKI, leveraging their established analytical interconnection as evidenced in prior research. Notably, this approach not only proposes a robust, fast, and accurate technique for DKI parameter estimation but also underscores the viability of deploying the sub-diffusion model for tissue characterization, substantiated by both simulated and human subject analyses. The paper is very-well written; well-organized; and coherent. The simulation study included different aspects of water diffusion as captured by diffusion-weighted MRI such as varying diffusion times and different b-value subpopulations, resulting in a comprehensive and thorough discussion.

Weaknesses:<br /> The primary objective of this study is to demonstrate a robust approach for estimating DKI parameters by directly calculating them using the parameters of the sub-diffusion model. This premise, however, relies on the assumption that the sub-diffusion model effectively characterizes the diffusion MRI signal and that its parameters are both robust and accurate. Throughout the manuscript, the term "ground truth kurtosis K" is frequently used to denote the "true K" value in the context of the simulation study. Nonetheless, given that the data is simulated using the new sub-diffusion model - an approximation of the DKI-based signal expression- this value cannot truly be considered the "ground truth K". The simulation study highlights the robustness and accuracy of D* and K*, but it inherently operates under the assumption that the observed data is in the form of the sub-diffusion model.

Reviewer #2 (Public Review):

Summary: The authors present a technique for fitting diffusion magnetic resonance images (dMRI) to a sub-diffusion model of the diffusion process within brain imaging. The authors suggest that their technique provides robust and accurate calculation of diffusional kurtosis imaging parameters from which high quality images can be calculated from short dMRI data acquisitions at two diffusion times.

Strengths: If the authors can show that the dMRI signal in brain tissue follows a sub-diffusion model decay curve then their technique for accurately and robustly calculating diffusional kurtosis parameters from multiple diffusion times would be of benefit for tissue microstructural imaging in research and clinical arenas.

Weaknesses: The applied sub-diffusion model has two parameters that are invariant to diffusion time, D_β and β which are used to calculate the diffusional kurtosis measures of a diffusion time dependent D* and a diffusion time invariant K*. However, the authors do not demonstrate that the D_β, β and K* parameters are invariant to diffusion time in brain tissue. The authors' results visually show that there is time dependence of the K* measure (in Figure 6) that is more apparent in white matter with K* values being higher for diffusion times of ∆=49 ms than ∆ = 19 ms. The diffusion time dependence of K* indicates there is also diffusion time dependence of β. Furthermore, Figure 7 shows that there is a tissue specific root mean squared error in model fitting over the two diffusion times which indicates greater deviation from the model fit in white matter than grey matter. To show that the sub-diffusion model is robust and accurate (and consequently that K* is robust and accurate) the authors would have to demonstrate that there is no diffusion time-dependence in both D_β and β in application to brain imaging data for each diffusion time separately. Simulated data should not be used to demonstrate the robustness and accuracy of the sub-diffusion model or to determine optimization of dMRI acquisition parameters without first demonstrating that D_β and β are invariant to diffusion time. This is because simulated signals calculated by using the sub-diffusion charateristic equation of dMRI signal decay will necessarily have diffusion time invariant D_β and β parameters.

Without further information demonstrating diffusion time invariance of D_β, β and K* it is not possible to determine whether the authors have achieved their aims or that their results support their conclusions.

Author Response

The following is the authors’ response to the original reviews.

Thank you for reviewing and assessing our paper. Reviewer2 had only posive comments. Reviewer 1 also had posive comments but included a list of suggesons. The revised version includes text edits to address the suggesons.

Reviewer 1:

… First, it is unclear whether the experiments and analyses were set up to be able to rule out more specific candidate funcons of the ZI.

The list of possible funcons performed by the ZI is broad. Nevertheless, our study considers a rather long list of neural processes related to the behaviors listed below.

Second, many important details of the experiments and their results are hard to decipher given the current descripons and presentaons of the data.

The procedures used in the present study have all been used and described in our previous studies (cited). We used the same descripons and presentaons as in the prior studies. We have gone over the Methods and figures to ensure that all details required to understand the experiments are provided, but we also added further details following the suggesons noted below.

The paper could be significantly strengthened by including more details from each experiment, stronger jusficaons for the limited behaviors and experimental analyses performed, and, finally, a broader analysis of how the recorded acvity in the ZI relates to behavioral parameters.

The paper studied several behaviors including: 1) spontaneous movement of head-fixed mice on a spherical treadmill, 2) tacle (whisker, and body parts) and auditory (tones and white noise) smuli applied to head fixed mice, 3) spontaneous movement iniaon, change, and turns in freely moving mice, 4) auditory tone (frequency and SPL) mapping in freely behaving mice, 5) auditory-evoked orienng head movements (responses) in the context of several behavioral tasks, 6) signaled acve avoidance responses and escapes (AA1), 7) unsignaled/signaled passive avoidance responses (AA2ITI/AA3-CS2), 8) sensory discriminaon (AA3), 9) CS-US interval ming discriminaon (AA4), and 10) USevoked unsignaled escape responses.

In freely moving experiments, the behavior is connuously tracked and decomposed into translaonal and rotaonal movement components. Discrete responses are also evaluated (e.g., acve avoids, escapes, passive avoids, errors, intertrial crossings, latencies, etc.). These behavioral procedures evaluate many neural processes, including decision making (Go/NoGo in AA1-3), response control/inhibion (unsignaled and signaled passive avoidance in AA2/3), and smulus discriminaon (AA3). The applied smuli, discrete responses, and tracked movement are always related to the recorded ZI acvity using a variety of techniques (e.g., cross-correlaons, PSTHs, event-triggered me extracons, etc.), which relate the discrete and me-series parameters to the neural acvity. We do not think all this qualifies as, “limited behaviors”.

(1) Anatomical specificaon: The ZI contains many disnct subdivisions--each with its own topographically organized inputs/outputs and putave funcons. The current manuscript doesn't reference these known divisions or their behavioral disncons, and one cannot tell exactly which poron(s) of the ZI was included in the current study. Moreover, the elongated structure of the ZI makes it very difficult to specifically or completely infect virally. The data could be beter interpreted if the paper included basic informaon on the locaons of recordings, the extent of the AAV spread in the ZI in each viral experiment, and what fracon of infected neurons were inside versus outside ZI.

Our experiments employed Vgat-Cre mice to target ZI neurons. In this line, GABAergic neurons from the enre ZI express Cre, including the dorsal and ventral subdivisions (see (Vong et al., 2011; Hormigo et al., 2020)). Consequently, AAV injecons in Vgat-Cre mice produce restricted expression in the ZI that can fully delineate the nucleus as shown in the papers referenced above (including ours). There is nil expression in structures above or below ZI because they do not express Cre in these mice (e.g., thalamus and subthalamic nucleus), which allows for selecve targeng of ZI. Our optogenec manipulaons and photometry recordings were not aimed at specific ZI subdivisions. We targeted the area of ZI indicated by the stereotaxic coordinates (see Methods), which are aimed at the center of the structure to maximize success in recording/manipulang neurons within ZI. While all the animals included in the study expressed opsins and GCaMP within ZI that in many animals fully delineated the nucleus, there was normal variability in the locaon of opcal fibers, but we did not detect any differences in the results related to these variaons.

Fiber photometry and optogenecs experiments are performed with rather large diameter opcal probes, which record/manipulate relavely large areas of the targeted structure. This is useful because our goal was to idenfy funconal roles of the enre ZI, which could then be parsed. In the present study, we did not perform experiments to target specific ZI populaons (e.g., retrograde Cre expression from target areas), which may have revealed differences atributed to their projecon sites. However, in the last experiment, we selecvely excited ZI fibers targeng three different areas (midbrain tegmentum, superior colliculus, and posterior thalamus), which revealed clear differences on movement. Thus, future experiments should explore these different populaons (e.g., using retrograde/anterograde expression systems), which may be in different subdivisions.

We have enhanced the Methods secon to clarify these points, including the addion of these references.

(2) Electrophysiological recording on the treadmill: The authors are commended for this technically very difficult experiment. The authors do not specify, however, how they knew when they were recording in ZI rather than surrounding structures, parcularly given that recording site lesions were only performed during the last recording session. A map of the locaons of the different classes of units would be valuable data to relate to the literature.

We have added details about this procedure in the Methods secon. These recordings are performed based on coordinates, and categorizing neurons as belonging to ZI is obviously an esmate based on the final histological verificaon. Nevertheless, the marking lesions revealed that the electrodes were on target, which likely resulted from the care taken during the surgical procedure to define reference points used later during the recording sessions (see Methods). Regarding a map of the unit locaons, we performed several analyses that did not reveal clear differences based on site. For example, we compared depth vs cell class, “There was no difference in recording depth between the four classes of neurons (ANOVA F(3,337)= 1.06 p=0.3676)”. Future experiments that employ addional methods (labelling, opto-tagging, etc.) would be more appropriate to address mapping quesons. Finally, as we state in the paper, “However, these recordings do not target GABAergic neurons and may sample some neurons in the tissue surrounding the zona incerta. Therefore, we used calcium imaging fiber photometry to target GABAergic neurons in the zona incerta”.

(3) The raonale of the analysis of acvity with respect to “movement peak”: It is unclear why the authors did not assess how ZI acvity correlates with a broad set of movement parameters, but rather grouped heterogeneous behavioral epochs to analyze firing with respect to “movement peaks”.

The reviewer is referring to movement peaks on the spherical treadmill. On the treadmill, we used the forward locomotor movement of the animal because this is the main acvity of the mice on the treadmill. We considered “all peaks” (or movements) and “>4 sec peaks”, which select for movement onsets. Compared to the treadmill, in freely movement condions during various behavioral tasks, there is a richer behavioral repertoire, which was analyzed in more detail (i.e., translaonal, and rotaonal components during spontaneous ongoing movement and movement onsets, movement related to various behaviors such as orienng, acve and passive avoidance, escape, sensory smulaon, discriminaon, etc.). Thus, we focused on a broader set of movement parameters in the Cre-defined ZI cells of freely behaving mice.

(4) The display of mean categorical data in various figures is interesng, however, the reader cannot gather a very detailed view of ZI firing responses or potenal heterogeneity with so litle informaon about their distribuons.

The PCA performs the heterogeneity classificaon in an unbiased manner, which we feel is a thoughul approach. The firing rates and correlaons with movement for each category of neurons are detailed in the results. Furthermore, the sensory responses for these neurons are also detailed. Together, we think this provides a detailed view of the units we recorded in awake/head-fixed mice. As already stated, further study would benefit from an addional level of cell site verificaon.

(5) Somatosensory firing responses in ZI: It is unclear why the authors chose the specific smuli used in the study. How oen did they evoke reflexive motor responses? What was the latency of sensory-evoked responses in ZI acvity and the latency of the reflexive movement?

These are broad quesons, and we assume that the reviewer is asking about somatosensory evoked responses on the spherical treadmill. We used air-puffs applied to the whiskers and on the back (le vs right) because the whiskers represent an important sensory representaon for mice, and the back is a part of the body (trunk), which we oen use to movate the animals to move forward on the treadmill. Regarding the latency of the somatosensory evoked responses, in this case, we did not correct them based on the me it takes the air-puff to travel to the whiskers or body part, and therefore we did not provide latencies. Moreover, air-puffs are not a very good method to quanfy whisker-evoked latencies, which are beter measured using other methods (whisker deflecons of single/mulple whiskers using piezo-devices or other mechanical devices, as we and others have done in many studies). We are not sure what the reviewer means by “reflexive behavior”; we did not measure any reflexive behavior under these condions. We have gone over the Methods and Results to ensure that sufficient details are provided about these experiments.

(6) It would be valuable to see example traces in Figure 3 to get a beter sense of the me course and contexts under which Ca signals in ZI tracks movement. What is the typical latency? What is the typical range of magnitudes of responses? Does the Ca signal track both fast and slow movements? How are the authors sure that there are no movement arfacts contribung to the calcium imaging? It seems there is more informaon in the dataset that could be valuable.

As is well known, fiber photometry calcium imaging is a slow populaon signal. We do not think it would be valuable to get into ming issues beyond what is already detailed in the study (i.e., magnitudes measured as areas or peaks, and ming as me-to-peaks). Regarding “movement arfacts”, these signals are absent (flat) in animals that do not express GCAMP. We agree that there must be addional valuable informaon in our datasets (as in most me-series). However, the current paper is already rather extensive. We will connue to peruse our datasets and report addional findings in new papers.

(7) Figure 4: The raonale for quanfying the F/Fo responses over a 6-second window, rather than with respect to discrete movement parameters, is not well explained. What types of movement are binned in this approach and might this broad binning hinder the ability to detect more specific relaonships between acvity and movement?

Figure 4 is focused on characterizing the relaonship between turns (ipsiversive and contraversive) during movement and ZI acvity. We tested different binning windows to find differences, including the 6 sec window in figure 4 for populaon measures (-3 to 3 sec around the turns). This binning approach is effecve at revealing differences where they exist (e.g., superior colliculus) as shown in our previous studies (e.g. (Zhou et al., 2023)). Moreover, the turns in the different direcons can be considered discrete responses at their peak, and the ming of the related acvaons (e.g., me to peaks), which we evaluated, are rather sensive and would have revealed differences, but we did not find them.

(8) Separaon of sensory and motor responses in Figure 5: The current data do not adequately differenate whether the responses are sensory or motor given the high correlaon of the sensory inputs driving motor responses. Because isoflurane can diminish auditory responses early in the auditory pathway, this reviewer is not convinced the isoflurane experiments are interpretable.

The reviewer is referring to Fig. 5C,D. Indeed, the point of this experiment was to show that it is difficult to differenate whether neural responses are sensory or motor in awake and freely moving condions. As we stated in the Results secon, “Although arousal and movement were not dissected in the present experiment (this would likely require paralyzing and ventilating the animal), the results indicate that activation of zona incerta neurons by sensory stimulation is primarily associated with states when sensory-evoked movement is also present”. This is followed in the Discussion by, “…as already noted, the suppression of sensory responses may be due to changes in arousal (Castro-Alamancos, 2004; Lee and Dan, 2012) and not caused by the abolishment of the movements per se”.

(9) Given the broad duraon of the mean avoidance response (Fig. 6 C, botom), it would be useful to know to what extent this plot reflects a prolonged behavior or is the result of averaging different animals/trials with different latencies. Given that the shapes of the F/Fo responses in ZI appear similar across avoids and escapes (Fig. 6D), despite their apparent different speeds and movement duraons (Fig 6C), it would be valuable to know how the ming of the F/Fo relates to movement on a trial-by-trial basis.

The duraon of the avoidance response cannot be ascertained from CS onset (panel 6C botom) and avoids are not wide but rather sharp. We have now made this clearer when Fig. 6C is first menoned (“note that since avoids occur at different latencies after CS onset they are best measured from their occurrence as in Fig. 6D”). Like other related condioned and uncondioned responses, avoids and escapes are similar, varying in the noted parameters. Regarding ming, as already menoned above, we think that the characteriscs of the populaon calcium signal make it unsuitable for further ming consideraons than what we included, parcularly for movements occurring at the fast speeds of avoids and escapes.

(10) Lesion quanficaon: One cannot tell what rostral-caudal extent of ZI was lesioned and quanfied in this experiment. It would be easier to interpret if also ploted for each animal, so the reader can tell how reliable the method is. The mean ablaon would be beter shown as a normalized fracon of cells. Although the authors claim the lesions have litle impact on behavior, it appears the incompleteness of the lesions could warrant a more conservave interpretaon.

The lesion experiment was a complement to the optogenecs inacvaon experiments we performed in our preceding ZI paper and in the present paper. Thus, the finding that the lesions had litle impact on behavior is supporve of the optogenecs findings. Regarding cell counts, we did not select any parts of the ZI to quanfy the number of neurons in either control or lesion mice. We considered the full rostrocaudal extent in our measurements. We are not sure what “fracon” the reviewer is suggesng, considering that these counts are from two different groups of mice (control vs lesion). Note that the red-marked neurons, as shown in Fig. 8A, reveal healthy non-Vgat-Cre neurons outside ZI that mark the extent of the AAV diffusion, which as shown spanned the full extent of the ZI in the coronal plane (and in other planes as the AAV spreads in all direcons).

(11) Optogenecs: the locaon of infected neurons is poorly described, including the rostral-caudal extent and the fracon of neurons inside and outside of ZI. Moreover, it is unclear how strongly the optogenec manipulaons in this study are expected to affect neuronal acvity in ZI.

We discussed the first point in (1) above. Regarding, how optogenec manipulaons are expected to affect neuronal acvity in ZI and its targets, we have conducted extensive electrophysiological recordings in slices and in vivo to detail the effects of our manipulaons on GABAergic neurons (e.g. (Hormigo et al., 2016; Hormigo et al., 2019; Hormigo et al., 2021a; Hormigo et al., 2021b), including ZI neurons (Hormigo et al., 2020). In fact, we never use an opsin we have not validated ourselves using electrophysiology. Moreover, our experiments employ a spectrum of optogenec light paterns (including trains/cont at different powers) that trate the optogenec effects within each session/animal. As shown in fig. 11 and 12, these paterns produce different behavioral effects related to the different levels of neural firing they induce. For ChR2-expressing neurons in ZI, firing is frequency dependent and maximal during Cont blue light (at the same power). For Arch-expressing neurons only Cont is used, and inhibion is a funcon of the green light power. When blue light is applied in ZI fibers targeng different areas, this relaonship changes. Blue light trains (1-ms pulses) at 40-66 Hz become the most effecve means of inducing sustained postsynapc inhibion compared to Cont or low frequencies.

References

Castro-Alamancos MA (2004) Dynamics of sensory thalamocorcal synapc networks during informaon processing states. Progress in Neurobiology 74:213-247.

Hormigo S, Vega-Flores G, Castro-Alamancos MA (2016) Basal Ganglia Output Controls Acve Avoidance Behavior. J Neurosci 36:10274-10284.

Hormigo S, Zhou J, Castro-Alamancos MA (2020) Zona Incerta GABAergic Output Controls a Signaled Locomotor Acon in the Midbrain Tegmentum. eNeuro 7.

Hormigo S, Zhou J, Castro-Alamancos MA (2021a) Bidireconal control of orienng behavior by the substana nigra pars reculata: disnct significance of head and whisker movements. eNeuro. Hormigo S, Vega-Flores G, Rovira V, Castro-Alamancos MA (2019) Circuits That Mediate Expression of Signaled Acve Avoidance Converge in the Pedunculoponne Tegmentum. J Neurosci 39:45764594.

Hormigo S, Zhou J, Chabbert D, Shanmugasundaram B, Castro-Alamancos MA (2021b) Basal Ganglia Output Has a Permissive Non-Driving Role in a Signaled Locomotor Acon Mediated by the Midbrain. J Neurosci 41:1529-1552.

Lee SH, Dan Y (2012) Neuromodulaon of brain states. Neuron 76:209-222.

Vong L, Ye C, Yang Z, Choi B, Chua S, Jr., Lowell BB (2011) Lepn acon on GABAergic neurons prevents obesity and reduces inhibitory tone to POMC neurons. Neuron 71:142-154.

Zhou J, Hormigo S, Busel N, Castro-Alamancos MA (2023) The Orienng Reflex Reveals Behavioral States Set by Demanding Contexts: Role of the Superior Colliculus. J Neurosci 43:1778-1796.

eLife assessment

This important study uses a range of technical approaches to investigate the responses of zona incerta neurons to movement and sensory stimuli. The majority of neurons exhibited movement related activity but only a small proportion were modulated by whisker deflections. The major conclusion of the study is that the zona incerta distributes a general motor signal. The evidence supporting this claim is solid, although the study would be improved by greater transparency and discussion of experimental methods and histological verification of recording sites, viral spread, and which territories of the zona incerta were investigated. The work will be of interest to behavioral and physiological neuroscientists.

Joint Public Review:

The manuscript presents compelling evidence for the role of the zona incerta area of the brain in regulating movement and sensory stimuli in mice. The study uses appropriate and validated methodology in line with the current state-of-the-art, including optogenetic manipulation and recording of single-unit activity. The authors' claims and conclusions are well-supported by their data, which includes a comprehensive review of previous research on the zona incerta. Overall, the manuscript provides solid evidence for the role of the zona incerta in regulating movement and sensory processing.

Major strengths and weaknesses of the methods and results.<br /> The zona incerta have many integrative functions that link sensory stimuli with motor responses to guide behavior.<br /> The study explored the activation of zona incerta GABAergic neurons during cued avoidance tasks and found that these neurons activate during goal-directed avoidance movement. Optogenetic manipulation of these neurons affected movement speed and performance during active avoidance tasks.<br /> The findings suggest that the zona incerta area of the brain plays a significant role in regulating movement and responding to salient auditory tones in association with movement in mice. The evidence presented is fundamental and provides a comprehensive review of previous research on the zona incerta and its involvement in various behaviors and sensory processing.

The article is very well written, with a correct hypothesis and a cutting-edge methodology to achieve the expected objectives. Moreover, they use statistical rigorous approaches in the analysis of the results. Also, analyzes are performed using scripts that automate all aspects of data analysis, ensuring their objectivity. The results are very novel, and provides solid evidence for the role of the zona incerta in regulating movement and sensory processing.

Author Response

The following is the authors’ response to the original reviews.

We thank the editor and the reviewers for their very useful and constructive comments. We went through the list and gladly received all their suggestions. The reviewers mostly pointed to minor revisions in the text, and we acted on all of those. The one suggestion that required major work was the one raised in point 13, about the processing pipeline being unconvincingly scattered between different tools (R → Python → Matlab). I agree that this was a major annoyance, and I am happy to say we have solved it integrating everything in a recent version of the ethoscopy software (available on biorxiv with DOI https://www.biorxiv.org/content/10.1101/2022.11.28.517675v2 and in press with Bioinformatics Advances). End users will now be able to perform coccinella analysis using ethoscopy only, thus relying on nothing else but Python as their data analysis tool. This revised version of the manuscript now includes two Jupyter Notebooks as supplementary material with a “pre-cooked” sample recipe of how to do that. This should really simplify adoption and provides more details on the pipeline used for phenotyping.

Please find below a point-by-point description of how we incorporated all the reviewers’ excellent suggestions.

Recommendations for the authors: please note that you control which, if any, revisions, to undertake

1) Line 38: "collecting data simultaneously from a large number of individuals with no or limited human intervention" is a bit misleading, as the entire condition the individuals are put in are highly modified by humans and most times "unnatural". I understand the point that once the animals are placed in these environments, then recording takes place without intervention, but it would be nice to rephrase this so that it reflects more accurately what is happening.

We have now rephrased this into the following (L39):

Collecting data simultaneously from a large number of individuals, which can remain undisturbed throughout recording.

2) Line 63: please add a reference to the Ethoscopes so that readers can easily find it.

Done.

2b) And also add how much they cost and the time needed to build them, as this will allow readers to better compare the proposed system against other commercially available ones.

This information is available on the ethoscope manual website (http://lab.gilest.ro/ethoscope). The price of one ethoscope, provided all necessary tools are available, is around ~£75 and the building time very much depends on the skillset of the builder and whether they are building their first ethoscope or subsequent ones. In our experience, building and adopting ethoscopes for the first time is not any more time-expensive than building a (e.g.) deeplabcut setup for the first time. We have added this information to L81

Ethoscopes are open source and can be manufactured by a skilled end-user at a cost of about £75 per machine, mostly building on two off-the-shelf component: a Raspberry Pi microcomputer and a Raspberry Pi NoIR camera overlooking a bespoke 3D printed arena hosting freely moving flies.

3) Line 88: The authors describe that in the current setting, their system is capable of an acquisition rate of 2.2 frames per second (FPS). Would reducing the resolution of the PiCamera allow for higher FPS? I raise this point because the authors state that max velocity over a ten second window is a good feature for classifying behaviors. However, if animals move much faster than the current acquisition rate, they could, for instance, be in position X, move about and be close to the initial position when the next data point is acquired, leading to a measured low max velocity, when in fact the opposite happened. I think it would be good to add a statement addressing this (either data from the literature showing that the low FPS does not compromise data acquisition, or a test where increasing greatly FPS leads to the same results).

We have previously performed a comparison of data analysed using videos captured at different FPSs, which is published in Quentin Geissman’s doctoral Thesis (2018, DOI: https://doi.org/10.25560/69514 ) in chapter 2, section 2.8.3, figure 2.9 ). We have now added this work as one of the references at L95 (reference 19).

4) Still on the low FPS, would a Raspberry Pi 4 help with the sampling rate? Given that they are more powerful than the RPi3 used in the paper?

It would, but it would be a minor increase, leading from 2.2 to probably 3-5 FPS. A significantly higher number of FPSs would be best achieved by lowering the camera’s resolution, as the reviewer’s suggested, or by operating offline. I think the interesting point being implied by the reviewers is that, for Drosophila, the current limits of resolution are more than sufficient. For other animals, perhaps moving more abruptly, they may not. The reviewer is right that we should add a line of caveat about this. We now do so in the discussion, lines 215-224.

Coccinella is a reductionist tool, not meant to replace the behavioural categorization that other tools can offer but to complement it. It relies on raspberry PIs as main acquisition devices, with associated advantages and limitations. Ethoscopes are inexpensive and versatile but have limitations in terms of computing power and acquisition rates. Their online acquisition speed is fast enough to successfully capture the motor activity of different species of Drosophilae28, but may not be sufficient for other animals moving more swiftly, such as zebrafish larvae. Moreover, coccinella cannot apply labels to behaviour (“courting”, “lounging”, “sipping”, “jumping” etc.) but it can successfully identify large behavioural phenotypes and generate unbiased hypothesis on how behaviour – and a nervous system at large – can be influenced by chemicals, genetics, artificial manipulations in general.

5) Along the same line of thought, would using a simple webcam (with similar specs to the PiCamera - ELP has cameras that operate on infrared and are quite affordable too) connected to a more powerful computer lead to higher FPS? - The reason for the question about using a simple webcam is that this would make your system more flexible (especially useful in the current shortage of RPi boards on the market) lowering the barrier for others to use it, increasing the chances for adoption.

Completely bypassing ethoscopes would require the users to setup their own tracking solution, with a final result that may or may not match what we describe here. If a greater temporal resolution is necessary, the easiest way to achieve more FPSs would be to either decrease camera resolution or use the Pis to take videos offline and then process those videos at a later stage. The combination of these two would give FPS acquisition of 60 fps at 720p, which is the maximum the camera can achieve. We now made this clear at lines 83-92.

The temporal and spatial resolution of the collected images depends on the working modality the user chooses. When operating in offline mode, ethoscopes are capable to acquire 720p videos at 60 fps, which is a convenient option with fast moving animals. In this study, we instead opted for the default ethoscope working settings, providing online tracking and realtime parametric extraction, meaning that images are analysed by each raspberry Pi at the very moment they were acquired (Figure 1b). This latter modality limits the temporal resolution of information being processed (one frame every 444 ms ± 127 ms, equivalent to 2.2 fps on a Raspberry Pi3 at a resolution of 1280x960 pixels with each animal being constricted in an ellipse measuring 25.8 ± 1.4 x 9.85 ±1.4 pixels - Figure 1a) but provides the most affordable and high-throughput solution, dispensing the researcher from organising video storage or asynchronous video processing for animals tracking.

6) One last point about decreasing use barrier and increasing adoption: Would it be possible to use DeepLabCut (DLC) to simply annotate each animal (instead of each body part) and feed the extracted data into your current analysis with coccinella? This way different labs that already have pipelines in place that use DLC would have a much easier time in testing and eventually switching to coccinella? I understand that extracting simple maximal velocity this way would be an overkill, but the trade-off would again be a lowering of the adoption barrier.

It would certainly be possible to calculate velocity from the whole animal pose measurement and then use this with HCTSA or Catch22, thus mimicking the coccinella pipeline, but it would be definitely overkilled, as the reviewers correctly points out. Given that we are trying to make an argument about high-throughput data acquisition I would rather not suggest this option in the manuscript.

7) Line 96: The authors state that once data is collected, it is put through a computational frameworkthat uses 7700 tests described in the literature so that meaningful discriminative features are found. I think it would be interesting to expand a bit on the explanation of how this framework deals multiple comparison/multiple testing issues.

We always use the full set of features on aggregate to train a classifier (e.g., TS_Classify in HCTSA) and that means no correction is necessary because the trained classifier only ever makes a single prediction (only one test is performed), so as long as it is done correctly (e.g., proper separation of training and test sets, etc.) then multiple hypothesis correction is not appropriate. This has been confirmed with the HCTSA/Catch22 author (Dr Ben Fulcher, personal communication). We have added a clarifying sentence about this to the methods (L315-318)

8) It would be nice to have a couple of lines explaining the choice of compounds used for testing and also why in some tests, 17 compounds were used, while in others 40, and then 12? I understand how much work it must be in terms of experiment preparation and data collection for these many flies and compounds, but these changes in the compounds used for testing without a more detailed explanation is suboptimal.

This is another good point. We have now added this information to the methods, in a section renamed “choice, handling and preparation of drugs” L280-285, which now reads like this:

The initial preliminary analysis was conducted using a group of 12 compounds “proof of principle” compounds and a solvent control. These compounds were initially used to compare both the video method and ethoscope method. After testing these initial compounds, it was found that the ethoscope methodology was more successful, and then the compound list was expanded to 17 (including the control) only using the ethoscope method. As a final test, we included additional compounds for a single concentration, bringing up the total to 40 (including control), also for the ethoscope method.

9) Line 119 states: "A similar drop in accuracy was observed using a smaller panel of 12 treatments (Supplementary Figure 2a)". It is actually Supplementary Figure 1c.

Thank you for noticing that! Now corrected. The Supplementary figures have also been renamed to obey eLife’s expected nomenclature (both Figure 1 – Figure supplements)

10) In some places the language seems a little outlandish and should either be removed or appropriately qualified. a- Lines 56-59 pose three questions that are either rhetorical or ill-posed. For example, "...minimal amount of information...behavior" implies there is a singular response but the response depends on many details such as to what degree do the authors want to "classify behavior".

Yes, those were meant as rhetorical questions indeed, but we prefer to keep them in, because we are hoping to generate this type of thoughts with the readers. These are concepts that may not be so obvious to someone who is just looking to apply an existing tool and may spring some reflection about what kind of data do they really want/need to acquire.

b) Some of the criticisms leveled at the state-of-the-art methods are probably unwarranted because the goals of the different approaches are different. The current method does not yield the type of rich information that DeepLabCut yields. So, depending on the application DeepLabCut may be the method of choice. The authors of the current manuscript should more clearly state that.

In the introduction and discussion we do try to stress that coccinella is not meant to replace tools like DLC. We have now added more emphasis to this concept, for instance to L212:

[tools like deeplabcut] are ideal – and irreplaceable – to identify behavioural patterns and study fine motor control but may be undue for many other uses.

And L215:

Coccinella is a reductionist tool not meant to replace the behavioural categorization that other tools can offer but to complement it

11) The application to sleep data appears suddenly in the manuscript. The authors should attempt to make with text change a smoother transition from drug screen to investigation into sleep.

I agree with this observation. We have now tried to add a couple of sentences to contextualise this experiment and hopefully make the connection appear more natural. Ultimately, this is a proof-ofprinciple example anyway so hopefully the reader will take it for what it is (L169).

Finally, to push the system to its limit, we asked coccinella to find qualitative differences not in pharmacologically induced changes in activity, but in a type of spontaneous behaviour mostly characterised by lack of movement: sleep. In particular, we wondered whether coccinella could provide biological insights comparing conditions of sleep rebound observed after different regimes of sleep deprivation. Drosophila melanogaster is known to show a strong, conserved homeostatic regulation of sleep that forces flies to recover at least in part lost sleep, for instance after a night of forceful sleep deprivation.

11b) Additionally, the beginning section of sleep experiments talks about sleep depth yet the conclusion drawn from sleep rebound says more about the validity of the current 5 min definition of sleep than about sleep depth. If this conclusion was misunderstood, it should be clarified. If it was not, the beginning text of the sleep section should be tailored to better fit the conclusion.

I am afraid we did not a good job at explaining a critical aspect here: the data fed to coccinella are the “raw” activity data, in which we are not making any assumption on the state of the animal. In other words, we do not use the 5-minutes at this or any other point to classify sleep and wakening. Nevertheless, coccinella picks the 300 seconds threshold as the critical one for discerning the two groups. This is interesting because it provides a full agnostic confirmation of the five minutes rule in D. melanogaster. We recognise this was not necessarily obvious from the text and now added a clarification at L189-201:

However, analysis of those same animals during rebound after sleep deprivation showed a clear clustering, segregating the samples in two subsets with separation around the 300 seconds inactivity trigger (Figure 3d). This result is important for two reasons: on one hand, it provides, for the third time, strong evidence that the system is not simply overfitting data of nought biological significance, given that it could not perform any better than a random classifier on the baseline control. On the other hand, coccinella could find biologically relevant differences on rebound data after different regimes of sleep deprivation. Interestingly enough, the 300 seconds threshold that coccinella independently identified has a deep intrinsic significance for the field, for it is considered to be the threshold beyond which flies lose arousal response to external stimuli, defining a “sleep quantum” (i.e.: the minimum amount of time required for transforming inactivity bouts into sleep bouts23,24,28). Coccinella’s analysis ran agnostic of the arbitrary 5-minutes threshold and yet identified the same value as the one able to segregate the two clusters, thus providing an independent confirmation of the fiveminutes rule in D. melanogaster.

12) Line 227: (standard food) - please add a link to a protocol or a detailed description on what is "standard food". This way others can precisely replicate what you are using. This is not my field, but I have the impression that food content/composition for these animals makes big changes in behaviour?

Yes, good point. We have now added the actual recipe to the methods L240:

Fly lines were maintained on a 12-hour light: 12-hour dark (LD) cycle and raised on polenta and yeast-based fly media (agar 96 g, polenta 240 g, fructose 960 g and Brewer’s yeast 1,200 g in 12 litres of water).

13) Data acquisition and processing: please add links to the code used.

Both the code and the raw data used to generate all the figures have been uploaded on Zenodo and available through their repository. Zenodo has a limit of 50GB per uploaded dataset so we had to split everything into two files, with two DOIs, given in the methods (L356, section “code and availability” - DOIs: 10.5281/zenodo.7335575 and 10.5281/zenodo.7393689). We have now also created a landing page for the entire project at http://lab.gilest.ro/coccinella and linked that landing page in the introduction (L64).

13b) Also your pipeline seems to use three different programming languages/environments... Any chance this could be reduced? Maybe there are R packages that can convert csv to matlab compatible formats, so you can avoid the Python step? (nothing against using the current pipeline per se, I am just thinking that for usability and adoption by other labs, the smaller amount of languages, the better?

This is a very important suggestion that highlights a clear limitation of the pipeline. I am happy to say that we worked on this and solved the problem integrating the Python version of Catch22 into the ethoscopy software. This means the two now integrate, and the entire analysis can be run within the Python ecosystem. HCTSA does not have a Python package unfortunately but we still streamlined the process so that one only has to go from Python to Matlab without passing through R. To be honest, Catch22 is the evolution of HCTSA and performs really well so I think that is what most users will want to use. We provide two supplementary notebooks to guide the reader through the process. One explains how to go from ethoscope data to an HCTSA compatible mat file. The other explains how ethoscope data integrate with Catch22 and provides many more examples than the ones found in the paper figures.

14) There are two sections named "References" (which are different from each other) on the manuscript I received and also on BioRxiv. Should one of them be a supplementary reference? Please correct it. I spent a bit of time trying to figure out why cited references in the paper had nothing to do with what was being described...

The second list of references actually applied only to the list of compounds in the supplementary table 1. When generating a collated PDF this appeared at the end of the document and created confusion. We have now amended the heading of that list in the following way, to read more appropriately:

eLife assessment

This study presents an important open-source resource for high-throughput behavioral screening. The protocols employ inexpensive, off the shelf hardware, and allow real-time analysis of hundreds of behaving flies. Although these protocols were developed using Drosophila melanogaster, they could easily be applied to other models. The evidence in support of the conclusions is solid and the revisions carried out by the authors go a long way towards providing the user with an integrated system that is also more user-friendly.

Joint Public Review:

In the current paper, Jones et al. describe a new framework, named "coccinella", for real-time high-throughput behavioral analysis aimed at reducing the cost of analyzing behavior. In the setup used here each fly is confined to a small circular arena and able to walk around on an agar bed spiked with nutrients or pharmacological agents. The new framework, built on the researchers' previously developed platform Ethoscope, relies on relatively low-cost Raspberry Pi video cameras to acquire images at ~0.5 Hz and pull out, in real time, the maximal velocity (parameter extraction) during 10 second windows from each video. Thus, the program produces a text file, and not voluminous videos requiring storage facilities for large amounts of video data, a prohibitive step in many behavioral analyses. The maximal velocity time-series is then fed to an algorithm called Highly Comparative Time-Series Classification (HCTSA)(which itself is based on a large number of feature extraction algorithms) developed by other researchers. HCTSA identifies statistically salient features in the time-series which are then passed on to a type of linear classifier algorithm called support vector machines (SVM). In cases where such analyses are sufficient for characterizing the behaviors of interest this system performs as well as other state-of-the-art systems used in behavioral analysis (e.g., DeepLabCut)

In a pharmacobehavior paradigm testing different chemicals, the authors show that coccinella can identify specific compounds as effectively as other more time-consuming and resource-consuming systems.

The new paradigm should be of interest to researchers involved in drug screens, and more generally, in high-throughput analysis focused on gross locomotor defects in fruit flies such as identification of sleep phenotypes. By extracting/saving only the maximal velocity from video clips, the method is fast. However, the rapidity of the platform comes at a cost--loss of information on subtle but important behavioral alterations. When seeking subtle modifications in animal behavior, solutions like DeepLabCut, which are admittedly slower but far superior in terms of the level of details they yield, would be more appropriate.

The manuscript reads well, and it is scientifically solid. The comments listed below were directed to the original submission and were satisfactorily addressed in the revised version.

1- The fact that Coccinella runs on Ethoscopes, an open source hardware platform described by the same group, is very useful because the relevant publication describes Ethoscope in detail. However, the current version of the paper does not offer details or alternatives for users that would like to test the framework, but do not have an Ethoscope. Would it be possible to overcome this barrier and have coccinella run with any video data (and, thus, potentially be used to analyze data obtained from other animal models)?

2- Readers who want background on the analytical approaches that the platform relies on following maximal velocity extraction, will have to consult the original publications. In particular, the current manuscript does not provide much explanation on Highly Comparative Time-Series Classification (HCTSA) or SVM; this may be reasonable because the methods were developed earlier by others. While some readers may find that the lack of details increases the manuscript's readability, others may be left wanting to see more discussion on these not-so-trivial approaches. In addition, it is worth noting that the same authors that published the HCTSA method, also described a shorter version named catch22, that runs faster with a similar output. Thus, explaining in more detail how HCTSA operates, considering is a relatively new method, will make the method more convincing.

eLife assessment

This paper presents an important contribution to the field of hippocampal registration by introducing a novel surface-based approach that utilizes the topological and morphological features of the hippocampus for anatomical registration across individuals, rather than volumetric-based methods commonly used in the literature. The study provides compelling evidence for the efficacy of this approach using histological samples from three different datasets and offers validation of the method through comparison with traditional volumetric registration. This is significant work given the large number of studies that examine hippocampal shape, thickness, and function in large cohorts, providing strong support for the use of hippocampal unfolding methods in future studies.

Author Response

The following is the authors’ response to the original reviews.

Thank you for reviewing our manuscript. We do find that the reviews are constructive and meaningful. Accordingly, we incorporated most suggestions into our revision. We provided a point-by-point responses to the reviews below.

Reviewer #1 (Public Review):

The evolution of dioecy in angiosperms has significant implications for plant reproductive efficiency, adaptation, evolutionary potential, and resilience to environmental changes. Dioecy allows for the specialization and division of labor between male and female plants, where each sex can focus on specific aspects of reproduction and allocate resources accordingly. This division of labor creates an opportunity for sexual selection to act and can drive the evolution of sexual dimorphism.

In the present study, the authors investigate sex-biased gene expression patterns in juvenile and mature dioecious flowers to gain insights into the molecular basis of sexual dimorphism. They find that a large proportion of the plant transcriptome is differentially regulated between males and females with the number of sex-biased genes in floral buds being approximately 15 times higher than in mature flowers. The functional analysis of sex-biased genes reveals that chemical defense pathways against herbivores are up-regulated in the female buds along with genes involved in the acquisition of resources such as carbon for fruit and seed production, whereas male buds are enriched in genes related to signaling, inflorescence development and senescence of male flowers. Furthermore, the authors implement sophisticated maximum likelihood methods to understand the forces driving the evolution of sexbiased genes. They highlight the influence of positive and relaxed purifying selection on the evolution of male-biased genes, which show significantly higher rates of nonsynonymous to synonymous substitutions than female or unbiased genes. This is the first report (to my knowledge) highlighting the occurrence of this pattern in plants. Overall, this study provides important insights into the genetic basis of sexual dimorphism and the evolution of reproductive genes in Cucurbitaceae.

Thank you for your positive comments. Greatly appreciated.

There are, however, parts of the manuscript that are not clearly described or could be otherwise improved.

• The number of denovo-assembled unigenes seems large and I would like to know how it compares to the number of genes in other Cucurbitaceae species. The presence of alternatively assembled isoforms or assembly artifacts may be still high in the final assembly and inflate the numbers of identified sex-biased genes.

The majority of unigenes were annotated by homologs in species of Cucurbitaceae (63%), including Momordica charantia (16.3%), Cucumis melo (11.9%), Cucurbita pepo (11.9%), Cucurbita moschata (11.5%), Cucurbita maxima (10.1%) and other species of Cucurbitaceae (Fig. S1C). We admit that in the final assembly, transcripts may be still overestimated due to the unavoidable presence of isoforms, although we have tried our best to filter it by several strategies of clustering methods. Additionally, we assessed the transcripts using BUSCOv5.4.5 and embryophyta_odb10 database with 1,614 plant orthologs assessment. Some 95.0% of these orthologs were covered by the unigenes, in which 1447 (89.7%) BUSCO genes were “Complete BUSCOs”, 85 (5.3%) were “Fragmented BUSCOs”, and only 82 (5.0%) were “Missing BUSCOs” (Table S2). Overall, our assessment suggested that we have generated high-quality reference transcriptomes in the absence of a reference genome. Subsequently, we revised the manuscript (lines 175-181).

• It is interesting that the majority of sex-biased genes are present in the floral buds but not in the mature flowers. I think this pattern could be explored in more detail, by investigating the expression of male and female sex-biased genes throughout the flower development in the opposite sex. It is also not clear how the expression of the sex-biased genes found in the buds changes when buds and mature flowers are compared within each sex.

Thank you for your advice for further understanding of this interesting pattern. In the near future, we would like to study these issues through more development stages of flowers in each sex, probably with the aid of single-cell techniques and a reference genome. We have revised the manuscript to reflect these in Results, in the section "Tissue-biased/stage-biased gene expression" (lines 202216).

• The statistical analysis of evolutionary rates between male-biased, female-biased, and unbiased genes is performed on samples with very different numbers of observations, therefore, a permutation test seems more appropriate here.

Thank you for your suggestion. However, all comparisons between sex-biased and unbiased genes were tested using Wilcoxon rank sum test in R software, which is more commonly used. Additionally, we tested some datasets, which were consistent with Wilcoxon rank sum test.

• The impact of pleiotropy on the evolutionary rates of male-biased genes is speculative since only two tissue samples (buds and mature flowers) are used. More tissue types need to be included to draw any meaningful conclusions here.

Thank you for your advice for further understanding of the impact of pleitropy. In the near future, we would like make further investigations through more development stages of flowers and new technologies in each sex to consolidate the conclusion.

Reviewer #2 (Public Review):

Summary:

This study uses transcriptome sequence from a dioecious plant to compare evolutionary rates between genes with male- and female-biased expression and distinguish between relaxed selection and positive selection as causes for more rapid evolution. These questions have been explored in animals and algae, but few studies have investigated this in dioecious angiosperms, and none have so far identified faster rates of evolution in male-biased genes (though see Hough et al. 2014 https://doi.org/10.1073/pnas.1319227111).

Strengths:

The methods are appropriate to the questions asked. Both the sample size and the depth of sequencing are sufficient, and the methods used to estimate evolutionary rates and the strength of selection are appropriate. The data presented are consistent with faster evolution of genes with male-biased expression, due to both positive and relaxed selection.

This is a useful contribution to understanding the effect of sex-biased expression in genetic evolution in plants. It demonstrates the range of variation in evolutionary rates and selective mechanisms, and provides further context to connect these patterns to potential explanatory factors in plant diversity such as the age of sex chromosomes and the developmental trajectories of male and female flowers.

Weaknesses:

The presence of sex chromosomes is a potential confounding factor, since there are different evolutionary expectations for X-linked, Y-linked, and autosomal genes. Attempting to distinguish transcripts on the sex chromosomes from autosomal transcripts could provide additional insight into the relative contributions of positive and relaxed selection.

Thank you for your meanful suggestions. We agree that the identification of chromosome origins for transcripts would greatly improve the insights of selection, and we will investigate these issues, probably with a reference genome in the near future.

Reviewer #3 (Public Review):

The potential for sexual selection and the extent of sexual dimorphism in gene expression have been studied in great detail in animals, but hardly examined in plants so far. In this context, the study by Zhao, Zhou et al. al represents a welcome addition to the literature.

Relative to the previous studies in Angiosperms, the dataset is interesting in that it focuses on reproductive rather than somatic tissues (which makes sense to investigate sexual selection), and includes more than a single developmental stage (buds + mature flowers).

The main limitation of the study is the very low number of samples analyzed, with only three replicate individuals per sex (i.e. the whole study is built on six individuals only). This provides low power to detect differential expression. Along the same line, only three species were used to evaluate the rates of non-synonymous to synonymous substitutions, which also represents a very limited dataset, in particular when trying to fit parameter-rich models such as those implemented here.

A third limitation relates to the absence of a reference genome for the species, making the use of a de novo transcriptome assembly necessary, which is likely to lead to a large number of incorrectly assembled transcripts. Of course, the production of a reference transcriptome in this non-model species is already a useful resource, but this point should at least be acknowledged somewhere in the manuscript.

Each of these shortcomings is relatively important, and together they strongly limit the scope of the conclusions that can be made, and they should at least be acknowledged more prominently. The study is valuable in spite of these limitations and the topic remains grossly understudied, so I think the study will be of interest to researchers in the field, and hopefully inspire further, more comprehensive analyses.

We acknowledged that our sample size was relatively small. We will investigate these issues at the population level, probably with a reference genome in the near future. We acknowledged in the revised manuscript that there may be some incorrectly assembled transcripts. We assessed the transcripts using BUSCOv5.4.5 and the latest embryophyta_odb10 database with 1,614 plant orthologs assessment. As mentioned, 95.0% of these orthologs were covered by the unigenes, which of 1447 (89.7%) BUSCO genes were “Complete BUSCOs”, 85 (5.3%) were “Fragmented BUSCOs”, and only 82 (5.0%) were “Missing BUSCOs” (Table S2). In short, the quality of transcriptome was high in the absence of a reference genome.

Reviewer #1 (Recommendations For The Authors):

My main criticism of this manuscript is that it refers to gene names and orthogroups throughout the text, however, the assembled transcripts are not accessible. The reference trascriptome, orthology data, and alignments used for evolutionary analysis should be made available through a public repository to support reproducibility and efficient use of produced resources in this study.

We have uploaded these datasets in Researchgate (https://www.researchgate.net/publication/373194650_Trichosanthes_pilosa_datasets Positive_selection_and_relaxed_purifying_selection_contribute_to_rapid_evolution of_male-biased_genes_in_a_dioecious_flowering_plant).

Comments to the authors:

1) I have an issue with the tissue-biased gene expression analysis. Looking at Fig.3, it seems to me there are 3,204 male-biased genes that are expressed at the same level in male buds and mature flowers (same for 5,011 female-biased genes in female buds and flowers), however, only a handful of genes show sex bias between mature male and female flowers. Taking the male-biased genes as an example, if the 3,204 M1BGs experience the same expression levels in mature male flowers and are no longer male-biased when mature male vs female flowers are compared, why there are not found as female tissue biased (F2TGs)? I may be wrong, but one scenario would be that the M1BGs increase their expression in female flowers and become unbiased. However, that increase in expression (low expression in the female buds → higher expression in the female flowers) should classify them as female tissue-biased genes (F2TGs). Can you please clarify how are the M1BGs and F1BGs expressed in the flowers of the opposite sex?

As to Fig. 3A, 3,204 male-biased genes expressed in male floral buds are part of all male-biased genes (3204+286+724=4214), as shown in Fig.2A. However, only 233 male-biased genes (88+1+144=233, Fig.2B and Fig.3B) expressed in male mature flowers. So, they are not expressed at the same level between male floral buds and mature flowers. Only 288 genes are sex-biased (M1BGs), as well as tissue/stage-biased (M1TGs) in male floral buds. M1BGs (4,214 male-biased genes) and F1BGs (5,096 female-biased genes) are 0 overlaps, except for 44,326 unbiasedgenes shown in Fig.2A. That is, F1BGs (5,096 female-biased genes) are low expression or no expression in M1BGs (4,214 male-biased genes). The expression levels of some genes have been shown in Table S14.

2) Paragraph (407-416) describes the analysis of duplicated genes under relaxed selection but there is no mention of this in the results.

In fact, these results have been shown in Table S13. It is not necessary for us to describe them in detail in the results.

3) How did the authors conclude that the identified functions in male flowers make them more adapted to biotic and abiotic environments (line 347-350)? In the paragraph above (line 338-342) the authors describe that female buds are better equipped against herbivores, which are a biotic factor?

Following your concerns, we have revised the manuscript as follows: For line 338-342, we revised the text as “Indeed, functional enrichment analysis in chemical pathways such as terpenoid backbone and diterpenoid biosynthesis indicated that relative to male floral buds, female floral buds had more expressed genes that were equipped to defend against herbivorous insects and pathogens, except for growth and development (Vaughan et al., 2013; Ren et al., 2022) (Fig. S7A and Table S11).” For line 347-350, we revised text as “We also found that male-biased genes with high evolutionary rates in male buds were associated with functions to abiotic stresses and immune responses (Tables S12 and S13), which suggest that male floral buds through rapidly evolving genes are adapted to mountain climate and the environment in Southwest China compared to female floral buds through high gene expression.”

4) Line 417-418: decreasing codon usage bias is linked to decreasing synonymous substitution rates, should this be the opposite?

No. Codon usage bias was positively related to synonymous substitution rates. That is, stronger codon usage bias may be related to higher synonymous substitution rates (Parvathy et al., 2022).

5) Figures and Tables are not standalone and are missing details in the legends. - Fig.2C, which genes are plotted on the heatmap and what is the color scale corresponding to?

• All Supplementary figures are missing the descriptions of individual panels (A, B, C,etc.) in the legends. In addition, please add the numbers of observations under boxplots.

• Supplementary Fig.5 and 6: Panel B is not a Venn diagram, I suggest removing it from the figures.

• Supplementary Fig.7: Should be 'sex-biased genes'. What is the x-axis on the plot?

• Supplementary Fig.8: Please add the description of the abbreviations in the legend. - Supplementary Tables S4, S5, S6: Please add information about the foreground and background branches.

• Supplementary Table S6, S7, S8, S9, S10: Please add more details about the column headers (what is Model-A, background ω 2a, Unconstrained_1.p, K, which was the foreground branch etc.).

• Supplementary Table S11: Please add gene IDs for each KEGG category.

We have revised/fixed these issues following your concerns and suggetions.

Minor comments:

Line 28: 'algae' in place of 'algas'

Line 53-56: Please provide more recent references.

Line65: 'most' instead of 'almost'

Line 86-87: It is not clear from the sentence if the sex-biased expression was detected in flowers compared to leaves, or were the sex-biased genes detected between male and female leaves? Please clarify.

Line 107-108: positive selection is referred to as adaptive evolution, please choose one or the other.

Line 109: 'force' instead of 'forces'

Line 110: 'algae' instead of 'alga'

Line 132: '..mainly distributed from Southwest,' the country is missing.

Line 202: 'protein sequence evolution'?

Line 232: what does the 'number of evolutionary rates' refers to?

Line 253: please provide a reference for the RELAX model.

Line 274: 'relaxed selective male-biased genes' should be 'male-biased genes under relaxed purifying selection'?

Line 318: Please add a sentence explaining why the Cucurbitaceae family is a great model to study the evolution of sexual systems.

Line 321: 'genes' instead of 'gene'.

Line 366: male-biased genes experience 'higher' or 'more rapid' evolutionary rates. line 377: in the present study and in the case of Ectocarpus alga, positive selection plays an important role in male-biased genes evolution, but does not account for the majority of evolutionary change. Therefore, I would not call it a 'primary' force.

Line 477: missing reference for DESeq2 package.

Line 480: 'used'.

Line 498: 'coding sequences'.

Line516: 'to' instead of 'by'.

Line 553: 'the' is repeated twice.

Sorry for the typos and grammatical issues. We have revised them accordingly.

Reviewer #2 (Recommendations For The Authors):

There are two areas for improvement, one empirical and one theoretical.

Empirically, the analyses could be expanded by an attempt to distinguish between genes on the autosomes and the sex chromosomes. Genotypic patterns can be used to provisionally assign transcripts to XY or XX-like behavior when all males are heterozygous and all females are homozygous (fixed X-Y SNPs) and when all females are heterozygous and males are homozygous (lost or silenced Y genes). Comparing such genes to autosomal genes with sex-biased expression would sharpen the results because there are different expectations for the efficacy of selection on sex chromosomes. See this paper (Hough et al. 2014; https://www.pnas.org/doi/abs/10.1073/pnas.1319227111), which should be cited and does in fact identify faster substitution rates in Y-linked genes (and note that pollenexpressed genes, at least, are concentrated on the sex chromosome in this system: https://academic.oup.com/evlett/article/2/4/368/6697528, https://royalsocietypublishing.org/doi/10.1098/rstb.2021.0226).

We have cited Hough et al. 2014 and noticed that several species have been observed to exhibit rapid evolutionary rates of sequences on sex chromosomes compared to autosomes, which has been related to the evolutionary theories of fast-X or fast-Z (lines 482-484).

On the theoretical side, this study is making a very specific intervention, namely identifying more rapid evolutionary rates in genes with male-biased than femalebiased expression in a dioecious plant. The writing in the introduction and the discussion needs to be improved to differentiate between this comparison and similar comparisons, e.g. sex-biased expression in other dioecious plants (76-81), between Xlinked and Y-linked genes (Hough et al. 2014), sex chromosomes and autosome (several studies already cited), gametophytic and sporophytic tissue, and male and female reproductive tissue in hermaphroditic plants. Setting out this distinction early in the introduction will make the specific goals and novelty of this work clearer.

Thank you for your constructive suggestions. We have revised the relevant part of the Introduction accordingly (lines 74-107).

Specific comments by line:

Sorry for the typos or wording issues. We have revised them.

26 - driven not driving

28 - check house style (algae vs algas)

28-29 - consider clarifying the antecedent of "them" (evolutionary forces, not algas) 35 - maybe, but don't the signalling genes involved in stress responses function in many capacities, not just stress? Also, there's evidence that reproductive recognition machinery in plants may ultimately derive from immune function (e.g. https://doi.org/10.1111/j.1469-8137.2008.02403.x), so the GO category "biotic stress" may be too vague

39 - maybe clarify that "for the first time" refers to male rather than female, since there have been other studies in dioecious plants

66-68 - asserting that something is "essential" after describing how rare it is doesn't quite follow, since diecious plants - especially with sex chromosomes - are basically an exception. I agree that understanding the evolution of dioecious plants is important, but this isn't the most compelling way to make that case - perhaps try something else.

137ff - this sentence can be consolidated and streamlined

142 - "floral tissue" rather than "flowers tissue," here and elsewhere

144 - divergence (singular)

235 - "evidence for the contributions of" = "evidences" is unidiomatic 250 - efficiency or efficacy?

300 - why is "inositol" capitalized here and elsewhere?

300ff - are these typical patterns in male tissue in other species?

308 - is that interesting? It seems like exactly what I'd expect. Perhaps start with the unsurprising but reassuring observation (anther and pollen development genes are indeed expressed in male buds) before moving on to the more surprising findings.

319 - remove "the"

321 - genes (plural)

330 - replace "these differences" with "the differences" 336 - perhaps recap proportions / percents here?

340 - unnecessary comma after diterpenoid

341 - this seems like a big leap from the evidence, especially in the absence of supporting information about the chemical defenses of these species and how they differ by sex. Don't terpenoids have a diverse array of functions, not just defense? Here's a review: https://link.springer.com/chapter/10.1007/10_2014_295

We have revised the text as “Indeed, functional enrichment analysis in chemical pathways such as terpenoid backbone and diterpenoid biosynthesis indicated that relative to male floral buds, female floral buds had more expressed genes that were equipped to defend against herbivorous insects and pathogens, except for growth and development (Vaughan et al., 2013; Ren et al., 2022) (Fig. S7A and Table S11)” (lines 373-378).

349 - as mentioned in line 35, this is a big speculative leap. The discussion is the place for speculation, but consider other explanations too. How does the development of flowers work? Are male flowers suppressing or resorbing female primordial organs? Do male flowers in fact senesce faster? perhaps spell out the logic in more detail.

We have revised the text as “In addition, the enrichment in regulation of autophagy pathways could be associated with gamete development and the senescence of male floral buds (Table S14) (Liu and Bassham, 2012; Li et al., 2020; Zhou et al., 2021). In fact, it was observed that male flowers senesced faster (Wu et al., 2011). We also found that homologous genes of two male-biased genes in floral buds (Table S14) that control the raceme inflorescence development (Teo et al., 2014) were highly expressed compared to female floral buds. Taken together, these results indicate that expression changes in sex-biased genes, rather than sex-specific genes play different roles in sexual dimorphic traits in physiology and morphology (Dawson and Geber, 1999).” (lines 390-402).

351 - senescence of, not senescence for

363 - but Hough et al. 2014 did show rapid evolution of Y-linked genes, and those are by definition sex biased ...

391 - perhaps reiterate here that while some sex-BIASED genes did, sex-SPECIFIC genes did not, to avoid confusion

We also revised them accordingly.

Reviewer #3 (Recommendations For The Authors):

1- lines 56-57 : « have facilitated » : this wording confounds correlation with causation. Consider rephrasing as « is associated with »

2- lines 58-60 : vague wording : what are these variations ? e.g. which tissues and stages are generally enriched?

3- line 63 : this sentence is a bit misleading: consider changing it to « Most dioecious plants possess homomorphic sex-chromosomes » [and explain what homomorphic means in this context].

4- line 68 : a reference is missing here. Also perhaps, allude to the fact that sexual selection in plants has long been considered a contentious issue (e.g. https://doi.org/10.1016/j.cub.2010.12.035)

5- lines 72-76 : beyond simply describing the pattern, say what evolutionary processes are revealed by these observations.

6- line 92 : remind the reader what these 5 studies are.

7- line 94-95 : explain why the comparison of vegetative vs vegetative and vegetative vs reproductive tissues is a problem.

The published studies only compared gene expression in vegetative versus vegetative tissues and vegetative versus reproductive tissues. Because it limited our understanding of sexual selection at different floral development stages. Revised accordingly (lines 103-104). We are very interested in flower development stage for sex-biased genes. The datasets could investigate sexual selection using two developmental stage (buds + mature flowers).

8- line 100 « Evolutionary dynamic analyses » : this wording is vague

9- line 110 : brown algae are NOT plants

10- line 137-140 or in M&M : needs to describe somewhere how the male flowers differ from the female flowers and vice-versa: are the whole morphological structures related to female (male) reproduction entirely missing, or is their development arrested later on and they are still present but simply not producing gametes? This has consequences for the interpretation of the genes they express.

We have revised the typos or wording issues accordingly. However, because the sampled floral buds were equal or less than 3 mm in size, we did not observe much morphological structural difference. Indeed, the male and female flowers at antheses were markedly different in this dioecious plant as shown in Fig. 1. Additionally, we found that dioecy is the ancestral state of Trichosanthes, and transitions to monoecy (Guo et al., 2020) based on our analysis (not shown in this study), which suggest that in the early stages of flower development, female floral buds may tend to masculinize, and vice versa (Fig. 2C).

11- line 152 : it is important to be very transparent on the sample sizes here: « from three females and three males of the dioecious ... »

12- line 153 : along the same line, explain here why a de novo transcriptome had to be generated here: « In the absence of an assembled reference genome for this nonmodel species, we de novo assembled ... »

13- line 164-165 : « we have generated high-quality reference trancriptomes » : I am not entirely convinced of the quality of the transcriptome obtained without a reference genome, so I suggest simply removing this subjective sentence.

Our assessment suggested that we have generated high-quality reference transcriptomes in the absence of a reference genome, which will be the next step of our work.

14- line 169 : briefly explain the criteria used to call differentially expressed genes. Given the threshold (log-fold change >=1.3 if I read the figure correctly, but the M&M says >=1), explain how it was chosen.

Sorry, you may have misunderstood the X, Y coordinates. The value of y coordinate represents -log10(FDR), and the value of x coordinate represents log2 (Fold Change).

15- line 174 : Not clear to me how Fig2C is « revealing strong sexual dimorphism », since genes cluster neither by sex nor by tissue. This should be explained more clearly.

16- line 174-177 : The fact that more ex-biased genes were identified in early buds than in mature flowers is an interesting observation that could be given more prominence in the manuscript, but it is not really explained. Could it reflect the fact that more genes are expressed in early buds because meiotic processes happen early in flower development? Also, the genes involved in male and female organ cell fate determination might also be expected to be expressed early, with mostly organ growth genes being expressed in the mature flower.

17- line 181 : a wrap-up sentence might be useful here to drive the point home that sex-bias is more prevalent in buds than mature flowers.

18- line 184 : « tissue-biased » : a more appropriate wording here would be « stagebiased », no ? These are indeed the same tissues but at different developmental stages.

19- line 183-195 : this section could benefit from setting clear expectations in a hypothesis testing framework laying out the reasons to expect a different bias between stages and between sexes. How does that fit with the level of morphological divergence between sexes (relates to my point 10 above).

20- line 197-204. A number of essential pieces of information are missing here: how many species, how divergent, say that one other is dioecious, and precise their relative phylogenetic placement (which is important to understand the models used below). Maybe adding a phylogeny of these species in Figure 4 could be useful. Also, briefly explain the « two-ratio » and « free-ratio » models here.

21- line 196 and following: In these analyses, I could not understand the rationale for keeping buds vs mature flowers as separate analyses throughout. Why not combine both and use the full set of genes showing sex-bias in any tissue? This would increase the power and make the presentation of the results a lot more straightforward.

As you pointed earlier (in the public review, paragraphy 2), “the dataset is interesting in that it focuses on reproductive rather than somatic tissues (which makes sense to investigate sexual selection), and includes more than a single developmental stage (buds + mature flowers)”, we totally agree with your points and were very interested in floral development stages for sex-biased genes.

22- line 216 : say explicitly that the reason for not detecting a significant difference in spite of a relatively large effect size is probably related to the low number of genes, conferring low statistical power to detect a difference. An important feature also not highlighted here is that the trend (though not significant) is in the opposite direction than in the buds, and that both the 2-ratio and the free-ratio models agree on these inverted trends. This could be the basis for an interesting comparison.

Thank you for your suggestions.

23- line 220 : needs to explain more clearly how this « free-ratio » differs from the « two-ratio » model.

24- line 232-234 : I don't see why this is necessary. Why not combine both? See also my comment 21 above.

25- line 237 : the «A-model » was not defined before.

26- line 237 : « male-biased » is missing after 343.

27- line 253-258 : briefly explain what these other models are based on and how they are not redundant and instead complement the previous analyses and each other. 28- line 266-268 : the use of a more precise set of codons for male-biased genes than the others (if I understood correctly) could also be interpreted as another sign of stronger selective constraint, no?

Codon usage bias is influenced by many factors, such as levels of gene expression. Highly expressed genes have a stronger codon usage bias and could be encoded by optimal codons for more efficient translation (Frumkin et al., 2018; Parvathy et al., 2022).

29- line 269-291 : removing genes on a post-hoc basis seems statistically suspicious to me. I don't think your analysis has enough power to hand-pick specific categories of genes, and it is not clear what this brings here. I suggest simply removing these analyses and paragraphs.

30- line 325 : say whether this patterns parallels / or not those in animals.

31- line 335 : yes, these biological pieces of information are important and should be given in the introduction already.

32- the discussion should focus at some point on the observation that more femalebiased genes are found in general, but that male-biased genes seem to be under greater selection. How do you reconcile these two apparently contradictory observations?

We found that male-biased genes with high evolutionary rates in male floral buds were associated with functions to abiotic stresses and immune responses (Tables S12 and S13), which suggests that male floral buds through rapidly evolving genes are adapted to mountain climate and the environment in Southwest China compared to female floral buds through high gene expression (lines 387-390).

33- line 355 : not clear how this follows from the previous sentences.

34- line 356-358 : vagiue. not clear what the message of this sentence is.

35- line 378-383 : say that these conclusions rely on the quality of gene annotation in this non-model species, which is probably pretty low (just like any other non-model species).

36- line 403 : this conclusion seems far-fetched. Explain how exactly you reached this conclusion.

37- line 406-416: these speculations on the role of paralogs seem unnecessary, in particular since the de novo transcriptome onto which all analyses are based cannot distinguish orthologs from paralogs.

38- line 417-424. The discussion should not contain new results.

39- line 510 : why were genes with dN/dS >2 discarded here? This might strongly bias the comparison, no? This needs to be properly justified.

40- lines 516-523 : references to the models are missing.

41- line 527: « omega = 1.5 » : why/how was this arbitrary threshold chosen?

42- Fig 2 : write out « buds » and « mature flowers » on top of the graphs

43- Fig 4 : add a phylogeny of the species showing the branch being compared. Also, add the number of genes in each category on each plot.

Thanks, we revised/fixed these issues accordingly.

Reviewer #1 (Public Review):

The evolution of dioecy in angiosperms has significant implications for plant reproductive efficiency, adaptation, evolutionary potential, and resilience to environmental changes. Dioecy allows for the specialization and division of labor between male and female plants, where each sex can focus on specific aspects of reproduction and allocate resources accordingly. This division of labor creates an opportunity for sexual selection to act and can drive the evolution of sexual dimorphism.

In the present study, the authors investigate sex-biased gene expression patterns in juvenile and mature dioecious flowers to gain insights into the molecular basis of sexual dimorphism. They find that a large proportion of the plant transcriptome is differentially regulated between males and females with the number of sex-biased genes in floral buds being approximately 15 times higher than in mature flowers. The functional analysis of sex-biased genes reveals that chemical defense pathways against herbivores are up-regulated in the female buds along with genes involved in the acquisition of resources such as carbon for fruit and seed production, whereas male buds are enriched in genes related to signaling, inflorescence development and senescence of male flowers. Furthermore, the authors implement sophisticated maximum likelihood methods to understand the forces driving the evolution of sex-biased genes. They highlight the influence of positive and relaxed purifying selection on the evolution of male-biased genes, which show significantly higher rates of non-synonymous to synonymous substitutions than female or unbiased genes. This is the first report (to my knowledge) highlighting the occurrence of this pattern in plants. Overall, this study provides important insights into the genetic basis of sexual dimorphism and the evolution of reproductive genes in Cucurbitaceae.

eLife assessment

This valuable paper examines gene expression differences between male and female individuals over the course of flower development in the dioecious angiosperm Trichosantes pilosa. The authors show that male-biased genes evolve faster than female-biased and unbiased genes. This is frequently observed in animals, but this is the first report of such a pattern in plants. In spite of the limited sample size, the evidence is mostly solid and the methods appropriate for a non-model organism. The resources produced will be used by researchers working in the Cucurbitaceae, and the results obtained advance our understanding of the mechanisms of plant sexual reproduction and its evolutionary implications: as such they will broadly appeal to evolutionary biologists and plant biologists.

Reviewer #2 (Public Review):

Summary:

This study uses transcriptome sequence from a dioecious plant to compare evolutionary rates between genes with male- and female-biased expression and distinguish between relaxed selection and positive selection as causes for more rapid evolution. These questions have been explored in animals and algae, but few studies have investigated this in dioecious angiosperms, and none have so far identified faster rates of evolution in male-biased genes (though see Hough et al. 2014 https://doi.org/10.1073/pnas.1319227111).

Strengths:

The methods are appropriate to the questions asked. Both the sample size and the depth of sequencing are sufficient, and the methods used to estimate evolutionary rates and the strength of selection are appropriate. The data presented are consistent with faster evolution of genes with male-biased expression, due to both positive and relaxed selection.

This is a useful contribution to understanding the effect of sex-biased expression in genetic evolution in plants. It demonstrates the range of variation in evolutionary rates and selective mechanisms, and provides further context to connect these patterns to potential explanatory factors in plant diversity such as the age of sex chromosomes and the developmental trajectories of male and female flowers.

Weaknesses:

The presence of sex chromosomes is a potential confounding factor, since there are different evolutionary expectations for X-linked, Y-linked, and autosomal genes. Attempting to distinguish transcripts on the sex chromosomes from autosomal transcripts could provide additional insight into the relative contributions of positive and relaxed selection.

Reviewer #3 (Public Review):

The potential for sexual selection and the extent of sexual dimorphism in gene expression have been studied in great detail in animals, but hardly examined in plants so far. In this context, the study by Zhao, Zhou et al. al represents a welcome addition to the literature.

Relative to the previous studies in Angiosperms, the dataset is interesting in that it focuses on reproductive rather than somatic tissues (which makes sense to investigate sexual selection), and includes more than a single developmental stage (buds + mature flowers).

Some aspects of the presentation have been improved in this new version of the manuscript. Specifically:

- the link between sex-biased and tissue-biased genes is now slightly clearer,<br /> - the limitation related to the de novo assembled transcriptome is now formally acknowledged,<br /> - the interpretation of functional categories of the genes identified is more precise,<br /> - the legends of supplementary figures have been improved<br /> - a large number of typos have been fixed.

However, overall the analyses are largely unchanged and the manuscript did not mature much in response to this first round of reviews. As I detail below, many of the relevant and constructive suggestions by the previous reviewers were not taken into account in this revision. For instance:

- Reviewer 2 made precise suggestions for trying to take into account the potential confonding factor of sex-chromosomes. This suggestion was not followed.<br /> - Reviewer 1 & 3 indicated that results were mentioned in the discussion section without having been described before. This was not fixed in this new version.<br /> - Reviewer 1 asked for a comparison between the number of de novo assembled unigenes in this transcriptome and the number of genes in other Cucurbitaceae species. I could not see this comparison reported.<br /> - Reviewer 1 pointed out that permutation tests were more appropriate, but no change was made to the manuscript.<br /> - Reviewer 3 pointed out the small sample size (both for the RNA-seq and the phylogenetic analysis), but again this limitation is not acknowledged very clearly.<br /> - Reviewer 1 & 3 pointed out that Fig 3 was hard to understand and asked for clarifications that I did not see in the text and the figure in unchanged.<br /> - Reviewer 3 suggested to combine all genes with sex-bias expression when evaluating the evolutionary rate, in addition to the analyses already done. This suggestion was not followed.<br /> - Reviewer 3 pointed out that hand-picking specific categories of genes was not statistically valid, and in fact not necessary in the present context. This was not changed.<br /> - Reviewer 1 asked for all data to be public, but I could not find in the manuscript where the link to the data on ResearchGate was provided.<br /> - Reviewers 1 & 3 pointed out that since only two tissues were compared, the claims on pleiotropy should have been toned down, but no change was made to the text.<br /> - Reviewer 1 asked for a clarification on which genes are plotted on the heatmap of Fig3C and an explanation of the color scale. No change was made.<br /> - Reviewer 1 asked for panel B in Fig 5 and 6 to be removed. They are still there. They asked for abbreviations to be explained in the legend of Fig S8. This was not done. They asked for details about coluln headers. Such detailed were not added. They asked for more recent references on line 53-56 : this was not done.

eLife assessment

This manuscript aims to identify the pacemaker cells in the lymphatic collecting vessels - the cells that initiate the autonomous action potentials and contractions needed to drive lymphatic pumping. Through the exemplary use of existing approaches (genetic deletions and cytosolic calcium detection in multiple cell types), the authors convincingly determine that lymphatic muscle cells are the origin of the action potential that triggers lymphatic contraction. This fundamental discovery establishes a new standard for the field of lymphatic physiology.

Reviewer #1 (Public Review):

Summary:

This manuscript explores the multiple cell types present in the wall of murine-collecting lymphatic vessels with the goal of identifying cells that initiate the autonomous action potentials and contractions needed to drive lymphatic pumping. Through the use of genetic models to delete individual genes or detect cytosolic calcium in specific cell types, the authors convincingly determine that lymphatic muscle cells are the origin of the action potential that triggers lymphatic contraction.

Strengths:

The experiments are rigorously performed, the data justify the conclusions, and the limitations of the study are appropriately discussed.

There is a need to identify therapeutic targets to improve lymphatic contraction and this work helps identify lymphatic muscle cells as potential cellular targets for intervention.

Weaknesses:

My only major comment would be that the manuscript provides a lot of rich information describing the cellular components of the muscular lymphatic vessel wall and that these data are not well represented by the title. The title (while currently accurate) could be tweaked to better represent all that is in this manuscript. Maybe something like "Characterization/Interrogation of the cellular components of murine collecting lymphatic vessels reveals that lymphatic muscle cells are the innate pacemaker cells regulating lymphatic contractions" or "Discovery/Confirmation of lymphatic muscle cells as innate pacemaker cells of lymphatic contraction through characterization of the cellular components of murine collecting lymphatic vessels". Potentially a cartoon summary figure of the components that make up the collecting lymphatic vessel wall could also be included. In my opinion, these changes will make this manuscript of more interest to a broader group of scientists. I have a few additional comments for consideration to improve the clarity and enhance the discussion of this work.

Reviewer #2 (Public Review):

Summary:

This is a well-written manuscript describing studies directed at identifying the cell type responsible for pacemaking in murine-collecting lymphatics. Using state-of-the-art approaches, the authors identified a number of different cell types in the wall of these lymphatics and then using targeted expression of Channel Rhodopsin and GCaMP, the authors convincingly demonstrate that only activation of lymphatic muscle cells produces coordinated lymphatic contraction and that only lymphatic muscle cells display pressure-dependent Ca2+ transients as would be expected of a pacemaker in these lymphatics.

Strengths:

The use of a targeted expression of channel rhodopsin and GCaMP to test the hypothesis that lymphatic muscle cells serve as the pacemakers in musing lymphatic collecting vessels.

Weaknesses:

The only significant weakness was the lack of quantitative analysis of most of the imaging data shown in Figures 1-11. In particular, the colonization analysis should be extended to show cells not expected to demonstrate colocalization as a negative control for the colocalization analysis that the authors present.

Reviewer #3 (Public Review):

Summary:

Zawieja et al. aimed to identify the pacemaker cells in the lymphatic collecting vessels. Authors have used various Cre-based expression systems and optogentic tools to identify these cells. Their findings suggest these cells are lymphatic muscle cells that drive the pacemaker activity in the lymphatic collecting vessels.

Strengths:

The authors have used multiple approaches to test their hypothesis. Some findings are presented as qualitative images, while some quantitative measurements are provided.

Weaknesses:

- More quantitative measurements.<br /> - Possible mechanisms associated with the pacemaker activity.<br /> - Membrane potential measurements.

Author Response

The following is the authors’ response to the original reviews.

We thank the reviewers and editors for their thoughtful assessment and critiques. As detailed below in the point-by-point replies, we have modified the text and figures to clarify points of ambiguity and to document statistical significance in places where we had inadvertently neglected to do so. The manuscript is clearer and more rigorous as a result of the review process.

Reviewer #1 (Public Review):

This study addresses the fundamental question of how the nucleotide, associated with the beta-subunit of the tubulin dimer, dictates the tubulin-tubulin interaction strength in the microtubule polymer. This problem has been a topic of debate in the field for over a decade, and it is essential for understanding microtubule dynamics.

McCormick and colleagues focus their attention on two hypotheses, which they call the "self-acting" model and the "interface-acting" model. Both models have been previously discussed in the literature and they are related to the specific way, in which the GTP hydrolysis in the beta-tubulin subunit exerts an effect on the microtubule lattice. The authors argue that the two considered models can be discriminated based on a quantitative analysis of the sensitivity of the growth rates at the plus- and minus-ends of microtubules to the concentration of GDP-tubulins in mixed nucleotide (GDP/GMPCPP) experiments. By combing computational simulations and in vitro observations, they conclude that the tubulin-tubulin interaction strength is determined by the interfacial nucleotide.

The major strength of the paper is a systematic and thorough consideration of GDP as a modulator of microtubule dynamics, which brings novel insights about the structure of the stabilizing cap on the growing microtubule end.

I think that the study is interesting and valuable for the field, but it could be improved by addressing the following critical points and suggestions. They concern (1) the statistical significance of the main experimental finding about the distinct sensitivity of the plus- and minus-ends of microtubules to the GTP-tubulin concentration in solution, and (2) the validity of the formulation of the "self-acting" model with an emphasis solely on the longitudinal bonds.

We thank the reviewer for the comment about statistical significance, and we regret our oversight to have not included that analysis in the original manuscript. We have now included an analysis of statistical significance for the main experimental results supporting the interface-acting model (Fig. 2C and the replotting of those data against a different abscissa in Fig. 3C,D), and more broadly we have ensured that all figure legends contain information about the number of measurements and whether error bars indicate SD or SEM.

The reviewers comment about the sole emphasis on longitudinal bonds helped us realize that a change to Fig. 1 (where we illustrate the two models) would improve clarity. We had originally chosen to illustrate Figure 1 using ‘pure’ longitudinal interactions (with no lateral contacts), and this may be what triggered the reviewer’s comment. We have now revised the figure to show ‘corner’ (longitudinal + lateral) interactions. There are two main reasons for this decision. First, the corner interactions are more long-lived and therefore more important for the phenomena under study. Second, because illustrating corner interactions provides a better basis for us to discuss what is a subtle aspect of our model – that the ‘GDP penalty’ affecting longitudinal or lateral interactions in a corner site is completely equivalent. Thus, our model is not quite as narrow/exclusive as the reviewer suggested. We appreciate having had the chance to clarify this.

Reviewer #2 (Public Review):

McCormick, Cleary et al., explore the question of how the nucleotide state of the tubulin heterodimer affects the interaction between adjacent tubulins.

(1) The setup of the authors' model, which attributes the dynamic properties of the growing microtubule only to the differences in interface binding affinities, is unrealistic. They excluded the influence of the nucleotide-dependent global conformational changes even in the 'Self-Acting Nucleodide' model (Fig. 1A). As the authors have found earlier, tubulin in its unassembled state may be curved irrespective of the species of the bound nucleotide (Rice et al., 2008, doi: 10.1073/pnas.0801155105), but at the growing end of microtubules, the situation could be different. Considering the recently published papers from other laboratories, it may be more appropriate to include the nucleotide-dependent change in the tubulin conformation in the Self-Acting Nucleotide model.

We understand the reviewer’s perspective, which may be summarized as: “We know conformational changes are happening and that they affect tubulin:tubulin interactions, so why isn’t your model trying to account for that?” In text added to the revised manuscript, we address this critique in the following ways. First, there is not a consensus in the field about how to parameterize the different conformations of tubulin and how they influence tubulin:tubulin interactions. Second, any attempt to explicitly account for different conformations of tubulin would substantially increase the number of adjustable model parameters, which in turn makes the fitting to growth rates more complicated. Third, compared to traditional ‘dynamics’ assays that use GTP, using mixtures of GMPCPP and GDP simplifies the biochemistry by eliminating GTPase. This results in a more uniform composition of nucleotide state in the body of the microtubule polymer, which diminishes the importance of explicitly modeling nucleotide-influenced changes in conformation. Fourth, it seems likely that different conformations of tubulin will modulate both longitudinal interactions (as tubulin becomes straighter the longitudinal contact area grows larger) and lateral interactions (as tubulin becomes straighter, the lateral contact areas on α- and β-tubulin come into better alignment). Our model treats longitudinal and corner (defined as longitudinal + lateral) interactions as independent, so in principle it could be implicitly capturing some of these conformational effects. By refining the strengths of the longitudinal and corner interactions independently, the model effectively allows the strength of longitudinal contacts to be different for pure longitudinal and corner interactions, which might implicitly capture some variations in longitudinal contacts for different tubulin conformations. Our model treats ‘bucket’-type sites (one longitudinal and two lateral interactions) as simply having an additional lateral interaction of equal strength as the first, but because bucket sites have such a high affinity, they rarely dissociate and this small oversimplification is unlikely to have a substantial effect. We have introduced text in several places (bottom of p. 7 and elsewhere) to cover these points.

(2) The result that the minus end is insensitive to GDP (Fig. 2) was previously published in a paper by Tanaka-Takiguchi et al. (doi: 10.1006/jmbi.1998.1877). The exact experimental condition was different from the one used in Fig. 2, but the essential point of the finding is the same. The authors should cite the preceding work, and discuss the similarities and differences, as compared to their own results.

Thank you for reminding us of this paper! We agree that it is an ‘on target’ citation, and have cited and discussed it in the revised manuscript (last paragraph of Introduction, third paragraph of Discussion).

Reviewer #1 (Recommendations For The Authors):

1) In my opinion, the way in which the authors have depicted their "self-acting" model in Fig. 1 and in Supplementary Figure 1, makes the model look intuitively implausible. The drawings seem to imply that at the plus-end the GTP hydrolysis in the beta-tubulin subunit somehow allosterically affects the alpha-tubulin subunit of the same dimer to weaken its longitudinal bond with adjacent tubulin dimer. Conversely, at the minus end, the same reaction now affects the very same beta-tubulin subunit, and modulates its longitudinal interaction with the next dimer.

However, a more realistic formulation of the "self-acting" model would be that the exchangeable nucleotide affects the lateral bonds, formed by the same beta-tubulin with its lateral neighbors. Although the experimental data in this regard are controversial, at least some supporting evidence for this idea comes from structural arguments, e.g. [Manka, S.W., Moores, C.A. Nat Struct Mol Biol 25, 607-615 (2018).] This "lateral selfacting", but not the "longitudinal self-acting" hypothesis, seems more natural, and it was the one previously implemented in the seminal paper by [Vanburen et al, 2002 Proceedings of the National Academy of Sciences 99.9 (2002): 6035-6040.] and later by other some other models as well.

This point has been addressed above, in part by modifying the cartoon in Fig. 1.

2) To better clarify, which exact models are considered in this manuscript, it would be helpful if the authors provided a detailed table with all simulation parameters, including, k_off_loner, k_off_bucket and k_off_corner, for both nucleotide states, in both the selfacting and the interface-acting models.

Thank you for the suggestion. We have added tables that show all simulation parameters, as well as the corresponding calculated on- and off-rates for each interaction.

3) I am not sure that using some 'arbitrarily chosen' parameters is very helpful in Chapter 1 of Results. In fact, the results, obtained with an unconstrained set of parameters may be misleading or provide ambiguous answers. In other words, how reliable are the conclusions, based on the arbitrary parameter set? For example, could the dependences of the microtubule growth rate on the GDP-tubulin content be more or less pronounced with a different set of arbitrarily chosen parameters, compared to the graphs in Fig. 1BC?

This is a fair criticism. In response, we have added three new sets of simulations that each test different choices of the biochemical parameters used in Figure 1. With respect to the original parameters, we tested a weaker and stronger choice for the longitudinal interaction (KDlong, a 100-fold range), the corner interaction (KDcorner, a 25-fold range), and the GDP weakening factor (a 100-fold range). The predicted supersensitivity of plus-end growth rates to GDP in the self-acting vs interface-acting mechanisms is robust across the range of different choices for the above parameters (Figure 1 Supplements 1 and 2). Parameters for these new simulations are shown in Tables 3 and 4.

4) It took me some time to comprehend why the minus-end growth rate is assumed to be dependent only on the concentration of the GMPCPP-tubulin (in section 2 of Results). It would be great if the authors simply plotted the simulated dependence of the growth rate on the GMPCPP-tubulin concentration in the case when no GDP-tubulin was added. As I understand, that curve should almost exactly match the dependence observed in Fig 1B, correct? Otherwise, it does not seem obvious, why GDP-tubulin does not impede the minus-end growth. Again, is this conclusion model- and parameterdependent? This question is related to point 3 above.

The minus-end growth rates decrease in proportion to the concentration of GMPCPPtubulin. We have added a note on minus-end growth rates in the Figure 1 legend.

5) I was not quite convinced by the evidence for distinct sensitivities of the plus- and minus-end growth rates to GDP-tubulin concentration (Figure 2C and Fig 3C, D). These are the key experimental measurements in the paper. Therefore, I suggest that the authors try to strengthen this point by additional measurements to increase statistics. Or at least, please, explain the data points, the error bars, and provide some information on the number of independent measurements and the statistical significance between the curves. Maybe, they could be directly compared after normalizing by the "all GMPCPP growth rate"? How was the "1.5-fold" ratio obtained in Fig 2C? Does that number refer only to a certain GDP-tubulin concentration or does that value somehow characterize the whole range of the concentrations measured?

This has been addressed above.

Reviewer #2 (Recommendations For The Authors):

These look identical to above and were addressed there.

(1) The setup of the authors' model, which attributes the dynamic properties of the growing microtubule only to the differences in interface binding affinities, is unrealistic. They excluded the influence of the nucleotide-dependent global conformational changes even in the 'Self-Acting Nucleodide' model (Fig. 1A). As the authors have found earlier, tubulin in its unassembled state may be curved irrespective of the species of the bound nucleotide (Rice et al., 2008, doi: 10.1073/pnas.0801155105), but at the growing end of microtubules, the situation could be different. Considering the recently published papers from other laboratories, it may be more appropriate to include the nucleotide-dependent change in the tubulin conformation in the Self-Acting Nucleotide model.

(2) The result that the minus end is insensitive to GDP (Fig. 2) was previously published in a paper by Tanaka-Takiguchi et al. (doi: 10.1006/jmbi.1998.1877). The exact experimental condition was different from the one used in Fig. 2, but the essential point of the finding is the same. The authors should cite the preceding work, and discuss the similarities and differences, as compared to their own results.

Reviewer #1 (Public Review):

This study addresses the fundamental question of how the nucleotide, associated with the beta-subunit of the tubulin dimer, dictates the tubulin-tubulin interaction strength in the microtubule polymer. This problem has been a topic of debate in the field for over a decade, and it is essential for understanding microtubule dynamics.

McCormick and colleagues focus their attention on two hypotheses, which they call the "self-acting" model and the "interface-acting" model. Both models have been previously discussed in the literature and they are related to the specific way, in which the GTP hydrolysis in the beta-tubulin subunit exerts an effect on the microtubule lattice. The authors argue that the two considered models can be discriminated based on a quantitative analysis of the sensitivity of the growth rates at the plus- and minus-ends of microtubules to the concentration of GDP-tubulins in mixed nucleotide (GDP/GMPCPP) experiments. By combing computational simulations and in vitro observations, they conclude that the tubulin-tubulin interaction strength is determined by the interfacial nucleotide.

The major strength of the paper is a systematic and thorough consideration of GDP as a modulator of microtubule dynamics, which brings novel insights about the structure of the stabilizing cap on the growing microtubule end.

eLife assessment

This important study combines in vitro experiments with simulations to identify the mechanisms governing modulation of microtubule dynamics by GTP hydrolysis. The authors introduce a convincing new approach by using a mixed GDP/GMPCPP lattice and varying GDP concentration to reveal that the nucleotide at the interface of two tubulin dimers determines the strength of the interaction between two dimers. Overall, the findings will be of interest to biophysicists and cell biologists, especially in the field of microtubule biology.

Reviewer #2 (Public Review):

In their manuscript, McCormick, Cleary et al., explore the question of how the nucleotide state of the tubulin heterodimer affects the interaction between adjacent tubulins. They use a solid combination of biochemical reconstitution assays and modeling to reveal that the nucleotide at the interface of two tubulin dimers determines the strength of the interaction between two dimers. Overall, the findings will be valuable to the field of microtubule biology.

Author Response

The following is the authors’ response to the original reviews.

Response to Public Reviews

Reviewer #1:

We thank this reviewer for their comments on our paper. We have adjusted the methods secon to ensure it is clear, including an updated descripon of the stascs and in some cases updated stascal methods to ensure uniformity in analyses across datasets. The discussion has been modified so that the message regarding our results is set appropriately in the literature.

Reviewer #2:

We are grateful to this reviewer for their insight. We have modified the text of the discussion to address the points of this reviewer, including providing a greater focus on the significance of our results without overgeneralizing. We have addionally reframed our argument regarding the detecon of pescides by Bombus terrestris to more carefully consider conflicng results from other papers.

Response to Recommendaons For The Authors

Response to Reviewer #1

We thank this reviewer for their thoughul comments and ideas. We have made several changes to the text of the manuscript to improve the clarity of our wring, and we are grateful to the reviewer for raising several important points that we had not sufficiently discussed in the paper previously. We feel the paper has been improved with the inclusion of a more thorough discussion and clarified methods. Please see below our responses to the points they raised.

A few general thoughts that I had when reading your manuscript: I assume you have only tested the acve pescide ingredients, but not the formula generally applied in the field. Given that these formulas contain addional compounds but the acve ingredients, might it not be possible that they could be perceived by bees?

For this study, we were interested specifically with the taste of acve pescide compounds, although we agree it could be interesng to explore the taste of other formula compounds, it was not within the scope of this paper to test these.

Is there an alternave to quinine as a negave control? As you state, quinine is generally used in studies and likely oen in concentraons which are beyond what can be seen in e.g. floral nectar, which might explain its aversive effect. I would like to see it tested in natural concentraons and ideally in combinaon with other potenally toxic plant secondary metabolites (PSMs).

The purpose of including quinine in our study was to provide an in-depth characterizaon of “biter” taste responses using the sensilla on bumblebee labial palps and galea (i.e., through the atenuaon of GRN firing). This was not to show how bumblebees may interact with plants containing quinine in the field, or other PSMs. It would indeed be interesng to explore other plant secondary metabolites, however this was beyond the scope of our paper.

L177-187 AND 233-238 Could you, please, provide a photo or schemac drawing to illustrate your assay? I have a very hard me picturing the actual set-up.

We have provided a labeled diagram of the bumblebee mouthparts and sensillum types (Fig 1A), as well as an image of the bumblebee feeding from a capillary in the behavioural assay (Fig 1G). Further details about the feeding assay (including a JoVe video) can be found with the Ma 2016 paper that we cite throughout our methods secon.

L219 Why did you choose 5 sec here?

This feeding bout duraon was selected based on the criteria defined in Ma et al 2016. We have added a citaon to that sentence.

L221-224 How precisely was the behavior scored? Just length of bouts or also repeated short contacts? Please, specify.

We used the first bout duraon and the cumulave bout duraon in our analyses. A sentence has been added to specify this.

L231/233 Please, provide some brief details here, as many readers may find it annoying to find and read another study for methods' details.

We have added three sentences in the methods to further explain the electrophysiological method.

L238-245 See also my general methods comment: concentraons used for pescides and quinine differ quite substanally, which may explain their different effects on the bees' percepon. Are the concentraons used for quinine realisc? If not that is totally fine for a negave control, but it would be interesng to see a comparison of effects conducted for similar concentraons.

The concentraons used of quinine were selected so that they would elicit a known “biter response” – these concentraons are not meant to be field-realisc, and our data (and others, e.g., Tiedeken et al 2014) show that lower concentraons of quinine are not detected by bumblebees.

L277-301 I assume that this is a standard stascal approach to analyze electrophysiological data. However, I am really struggling with fully understanding what you did here. It might be good to add some addional explanaon/detail, e.g. on why you constructed firing rate histograms or how you derived slopes (aren't smulus and bin categorical variables?).

Firing rate histograms are indeed very commonly used for visualizing neuron spikes over me. We have changed the text somewhat in an effort to make it more clear. Likewise, the “slopes” were derived from the LMEs, and in this case “bin” is a connuous me variable – any me variable will involve some binning depending on the resoluon used but should not be considered categorical.

L291-295 As you were averaging and normalizing your data, could you, please, provide some informaon on variaon within animals?

We have now included informaon on the coefficient of variaon for spike rates across sensilla for a given animal/smulus (CV range, median, and IQR).

L295 I assume t-SNE represent a mulvariate approach for ordinaon, correct? Can you explain why you chose to use this approach? Did you use Euclidean Distance?

Yes, t-SNE is a mulvariate technique for dimensionality reducon. It is parcularly well-suited for the visualizaon of high-dimensional datasets, as it can reveal the underlying structure of the data by embedding it in a lower-dimensional space (e.g., 2D) while preserving the local structure of the data as much as possible. We used t-SNE because it has been shown to be effecve in visualizing clusters of similar data points in high-dimensional data. Euclidean distance was used as the distance metric for the t-SNE embedding. Euclidean distance is the default distance metric for most implementaons of t-SNE and is appropriate for connuous data like the spike counts in this study. We have adjusted the methods to clarify this.

L304 Why did you not always use LMEs?

We have adjusted the text to show that we used LME for all comparisons, and the stascs have been updated accordingly in the results secon. None of the outcomes changed with the implementaon of LME for all tests.

L306 Would it not make sense to also include the interacon between smulus and concentraon in your models?

We have now included a sentence to explain that the interacon term was removed due to it being nonsignificant, and the models without the interacon term having beter model fit (determined by having lower AIC and BIC values).

Results:<br /> L337, 339 and more: I would prefer to see actual p-values, not just "p > 0.05".

This has been adjusted on L337 and 339. As far as we are aware, there are no other instances where exact p-values were not given (except when p < 0.0001).

Discussion:<br /> L470 This is true, but the bees' behavior changed significantly, indicang that they may respond to this small change in firing paterns already?

It is true that the bees’ behaviour changed significantly with 0.1mM QUI, but this was not the case with the pescides. Bees drank less overall of 0.1mM QUI than OSR because of the rapid posngesve effects of this compound. It’s important that the duraon of the first bout was not affected (i.e., they didn’t directly avoid it by taste upon first encountering it, as they do with 1mM QUI), but just that they drank less of the 0.1mM QUI over 2 minutes. Post-ingesve effects may occur as quickly as 30s aer inial consumpon. For the pescides, the small changes in GRN firing were not associated with any effects on consumpon or our other measures of feeding behaviour, and we suggest this results from a lack of rapid negave posngesve consequences. We now include further discussion of these important points.

L481 But they consumed significantly less of the 0.1 mM QUI!?

This is true, but they did not reject it (i.e., not drink it at all), and there were no changes in the amount of me the bees spent in contact with the 0.1mM QUI soluon compared to OSR. We have adjusted the text for clarificaon.

L504/505 AND 520/521 AND 533-536 I see your point, but I am wondering whether the bees may need some me but are generally able to learn the taste of pescides, which may explain why e.g. Arce et al. only saw an effect over me. For example, learning to 'focus' on the pescide taste may require some internal feedback (bees not feeling well) or larvae feedback. If I understood right, you tested workers only, which might be less sensive than queens or larvae. I think these aspects should be discussed.

In our study, we invesgated the mechanism of taste detecon of pescides. We agree that bees likely use posngesve mechanisms to learn to associate the locaon (or another cue) of a food source with posive or negave posngesve cues. ‘Focus’ is a higher-order process that involves increased atenon to sensory smuli but does not affect sensaon at the level of the receptor. We show that bees are unable to taste pescides using the gustatory receptors on their mouthparts, so post-ingesve learning would not be able to associate the pescides with any taste cue. Indeed, there may be caste-specific differences with foraging queens, however a discussion of this would be beyond the scope of our paper.

I also recommend broadening the scope of your discussion. For example, you only focus on nectar, while the story might be different for pollen, which is also contaminated with pescides but represents a different chemical matrix with potenally different taste properes. Also, unlike nectar, pollen is collected with tarsae and hence on contact with other bee body parts.<br /> I would also like to see a discussion of your study's implicaons for other bee species and other potenally toxic compounds (e.g. PSMs).

We do not include any data in this paper regarding tarsal or antennal taste or other potenally toxic compounds. In this paper we present one mechanism of biter taste percepon (i.e., of quinine) and specifically show that the buff-tailed bumblebee is unable to taste certain pescides using their mouthparts. To avoid overgeneralizing, we have not included discussions about other species or compounds, which may or may not share similaries with our study.

Response to Reviewer #2

We thank this reviewer for their comments. We have adjusted the text to avoid overgeneralizaons with our conclusions, and atempted to soen language in the discussion that may have been construed as combave towards the Arce et al (2018) paper. We hope this reviewer finds these adjustments to be in line with their expectaons.

1) In two parts of the manuscript, the authors made broad predicons and conclusions beyond what the evidence in the paper can support and wrote "Future studies will be necessary to confirm this." (Lines 508-509) and " Future studies that survey a greater variety of compounds will be necessary to confirm this." (563-564). Instead of making conclusions based on what experimental data in future studies may support, I would ask the authors instead to make conclusions that their current study can support based on experimental evidence in this paper.

We have removed these predicons that extend beyond the scope of the paper.

2) Line 315 "GRNs encode differences in sugar soluon composion". The logic of the tle is wrong.

This has been fixed.

3) Since this study is only performed in one bumblebee species, then I would suggest that the tle be more specific e.g., "Mouthparts of the bumblebee Bombus terrestris exhibit poor acuity for the detecon of pescides in nectar".

We have made this change.

Reviewer #1 (Public Review):

Summary:<br /> Parkinson and colleagues address an interesting and important question, i.e. whether the bumblebee Bombus terrestris can receive field-realistic concentrations of different pesticides in a sugar solution mimicking nectar. The study directly follows up on a previous study conducted by the same team (Kessler et al. 2015, Nature), which was partly questioned by another more recent study (Arce et al. 2018, Proc R. Soc. B). The authors apply a combination of electrophysiological measurements and behavioral feeding tests to answer this question. Their results strongly suggest that B. terrestris workers are not able to perceive field-realistic doses of pesticides in a sugar solution. They additionally show that B. terrestris can physiologically differentiate between solutions varying in sugar composition.

Strengths:<br /> Sophisticated methodology, combination of approaches, clear and precise language. The stats questions have been addressed to my satisfaction. In terms of interpretation, however, several suggestions and comments were provided from an ecological perspective, which was deemed important, while the authors have expressed their intent to concentrate on the electrophysiological mechanism. Given that this study was motivated by conflicting results from earlier research, which were frequently employed to discuss the authors' findings, I still find that the discussion needs to be expanded in order to encompass a wider context.

Reviewer #2 (Public Review):

Summary:<br /> This manuscript is part of the Wright lab's ongoing studies that investigate whether the bumblebee B. terrestris can detect the presence of pesticides when feeding. Previously, they showed that B. terrestris cannot detect neonicotinoids and would prefer food containing neonicotinoids (Kessler et al. 2015). However, in that paper, they showed that B. terrestris cannot taste neonicotinoids but did not provide evidence on why B. terrestris prefer food containing neonicotinoids. In the current paper, the authors continue to suggest that B. terrestris cannot taste neonicotinoids as well as another insecticide, sulfoxaflor, based on additional behavioral experiments and electrophysiological experiments focusing on specific GRNs. While the data from these experiments continue to suggest that B. terrestris cannot taste these insecticides using their mouthparts, whether B. terrestris can actually perceive these insecticides, and why this species prefers food containing these compounds remains unknown.

Strengths:<br /> The authors provided additional evidence that B. terrestris cannot taste neonicotinoids with their mouthparts. The authors have addressed my concerns regarding overgeneralization and that parts of the manuscript were written in a way that sounded combative with studies from other groups that had come to slightly different conclusions from their previous paper.

Author Response

The following is the authors’ response to the original reviews.

We thank the reviewers for recognizing the importance of our work and for their insightful suggestions. A point-by-point response to their comments is listed underneath each reviewer’s section.

Reviewer #1 (Recommendations For The Authors):

Major comments

1) Have the authors optimized the expression level of dCas9? I cannot find this information in this paper or in their 2021 paper. It is important to avoid the toxicity phenomenon that occurs when using guide RNAs that share specific five base seed sequences (referred to as 'bad seeds').

Cui L., Vigouroux A., Rousset F., Varet H., Khanna V., Bikard D. A CRISPRi screen in E. coli reveals sequence-specific toxicity of dCas9. Nat. Commun. 2018; 9:1912.

Rostain W., Grebert T., Vyhovskyi D., Thiel Pizarro P., Tshinsele-Van Bellingen G., Cui1 L., Bikard D. Cas9 off-target binding to the promoter of bacterial genes leads to silencing and toxicity. Nucleic Acids Research, 2023, gkad170.

2) One guide per gene is highly unusual given that different guides block the RNA polymerase with different efficiency. This was even shown by the Machner lab in the Legionella context in Figure 1c of Ellis et al. 2021 for sidM and vipD. Typically, genes need three guides minimum to ensure that the gene of interest is knocked down fully unless it is not possible as the gene is too small and/or it is difficult to find an NGG sequence. The authors have used one guide per effector, how can they be sure that each gene is knocked down? The Machner lab themselves in Figure 3c of Ellis et al. 2021 shows not all genes targeted using multiplex CRISPRi are equally efficiently knocked down. Please justify why only one guide per gene was chosen and add controls to validate the results. The authors themselves state that the strategy of one guide may be problematic. Lines 315-316 it reads... A possible explanation was the incomplete knockdown of a seemingly important process.

3) Given what the Machner lab observed about spacer location in Ellis et al. 2021 would it not make more sense to take one set of redundant effectors and make multiplex randomized CRISPRi with them in different locations? For Figure 1 at least.

4) Following infection, it seems that the bacteria were not plated onto antibiotic media, so it is not known how well the plasmid harboring guides is kept through infection.

Specific comments

A) The first results paragraph describes the set-up of 10-plex synthesized CRISPR arrays, where 10 effector encoding genes of specific gene families are selected. The rationale of the choice of these genes is not given. Please explain.

B) Please also add some biological data on what these genes code for, and what is their known or predicted function. It is not very informative and exciting to have tables of lpg numbers without any knowledge of what these genes code for and why they were selected, at least some.

C) Figure 1 A Why are only some of the MC arrays shown? Please, at least include in supplementary the others. Again one array in detail would be more informative, showing true knockdown of all genes by qPCR and ideally by western blot.

D) I am not convinced that the gene silencing efficiency qPCR comparison is done in the correct way. In my opinion, each of the genes to be knocked down should be tested against the expression of a control gene e.g. rpoS and then these results should be compared and not the results of empty plasmid or CRISPR array containing plasmid directly. L. pneumophila are very sensitive to growth conditions and inoculum, thus the two strains might not be completely at the same growth stage when being compared which can impact the results.

E) Figure 1 B As stated in general comment number 4, the authors do not appear to plate onto antibiotic so we don't know how well the plasmid harboring the guides is kept through infection. The sustained presence of the guide is particularly important for CRISPRi.

F) The authors found only a few growth phenotypes and mainly this was due to single genes and not combinations of genes. This might again be due to the fact that only one guide per gene was used. How do the authors know that all genes targeted were indeed knocked down?

G) Line 119 Alternatively, the genes were not 100% all knocked down, escaping the knockdown effect expected. Could authors take three genes with three guides each and look at impact instead of only one?

H) The authors then develop the randomized multiplexed arrays and chose to test 44 TME encoding genes. Line 141 Justify why these effectors were chosen in the text.

I) Unfortunately, the method is not clearly described, and many parts are complicated and the text needs to be re-read several times to be understood (lines 150 - 166). Please re-write to better explain to the reader. In line 156 the authors point to a supplementary note 1. This information should be in the main text.

J) What is the copy number of the CRISPR plasmid? Please add in the Material and Method section also the origin of this plasmid.

Figure 2

K) In the paper (line 154-160) and the extra notes, it states that authors attempt to size select CRISPR arrays. However, this is not apparent in Figure 2 schematic. Or are the authors stating that plasmids only containing one guide were selected out? However, line 312 would suggest not. Please clarify

L) A limiting factor in making multiplex guide CRISPR, as the authors are trying to establish in this study, is cloning of multiple guides. In the pre-determined CRISPR arrays in this study, the guides were synthesized. For the randomized multiplex CRISPR in this study, the authors adapt a Golden Gate cloning method to generate multiple sgRNAs in the Cas9 vector. A similar protocol was established in the below paper. Please add this reference.

Zuckermann, M.; Hlevnjak, M.; Yazdanparast, H.; Zapatka, M.; Jones, D.T.W.; Lichter, P.; Gronych, J. A novel cloning strategy for one-step assembly of multiplex CRISPR vectors. Sci. Rep. 2018

M) As the authors note, Zuckermann et al. similarly note that plex of 3 or 4 is most common and above 5 is rare. This thus appears to still be the limiting step of multiplex CRISPR technology. Please discuss

Figure 4

N) The idea of multiplexed CRISPRi seq to address the biological phenomenon of redundancy is an interesting one, however, I am missing the in-depth functional characterization and discussion of at least one of the redundant functions discovered. Please add.

Figure5/6

O) As noted above, the strength of the experiments is undermined by how CRISPRi is set up. Having an average multiplex of 2 or three and again only using one guide per gene weakens the study and the results obtained. Furthermore, as stated in general comment number 4, the authors do not appear to plate onto antibiotic so again, we don't know how well the plasmid harboring the guides is kept through infection. The sustained presence of the guide is particularly important for CRISPRi. Please add a validation that the guides are all present.

Response to Reviewer #1

We are grateful to the reviewers for their insightful comments and suggestions on how to further improve the manuscript.

Regarding the issue of ‘bad seed sequences’ (comment #1), we had previously evaluated the expression level of dcas9 (plotted in Figure 1b of the 2021 Communications Biol paper) and tuned our induction conditions accordingly (40 ng/mL as described in the Methods). Since all strains used in this study express dcas9 from the chromosome, not a plasmid, this eliminates the possibility of fluctuations in expression levels due to variabilities in plasmid copy numbers.

In the rare event that toxicity by any given guide occurs, we would expect that guide to already be underrepresented or missing in the input pool following 24+ hours of CRISPRi induction during axenic growth. Our data, now discussed in the manuscript (Lines 211-216 and Figure S2), revealed that this was not the case and that all guide-encoding spacers were present in roughly equal amounts (median of >5000 occurrences). As with any knockdown study, the creation of true chromosome deletions was performed throughout as to alleviate the issue of false positives.

Regarding comments #2, #3, and specific comments made under point F, G, and O, on the topic of using single vs. multiple guides, we agree that there are circumstances under which using more than one guide per target may be advantageous, for example when attempting to delete a gene from mammalian cells using conventional CRISPR engineering. In the study described here, this is not the case. In fact, we did create a second array library with alternative guides targeting the same group of genes at locations other than the “optimal location” identified in our 2021 paper and found that these “sub-optimal” guides were inefficient for identifying critical effectors as described in Supplemental Note S1 under the heading “Sub-optimal annealing sites” (Lines 919+). These data suggest that adding sub-optimal guides to the arrays of optimal guides might ‘poison’ the arrays and limit rather than enhance their ability to identify gene combinations.

Regarding comment #2, #3, and specific comments made under point C, F, and G, on the topic of confirming efficient gene knockdown for the identification of critical genes, we remind Reviewer 1 that we did confirm knockdown of 60 of the target genes of the 10-plex screen to be at least 2-fold, with an average fold repression of one order of magnitude or more (Figure 1A). While knockdown of every gene in every 10-plex construct would be an unprecedented ask of any published CRISPR screen, we believe that these 60 genes provide a large enough sampling of all guides to elucidate the range of knockdown to be expected by our CRISPRi platform. As with other knockdown technologies, such as RNAi, there is no expectation of accomplishing complete knockdown for any given target. Hence, the data in Figure 1A suggest that the lack of identifying critical genes using pre-determined 10-plex arrays was not due to a lack of knockdown efficiency, but rather the difficulty to accurately predict redundancy within a cohort of uncharacterized genes, accentuating the need for array randomization with MuRCiS.

On the topic of antibiotic use for plasmid selection (comments #4, E and O), we would like to clarify that the CRISPR plasmids were selected by thymidine prototrophy, not antibiotic resistance, and we apologize for not making this clearer. The laboratory strain Lp02 is a thymidine auxotroph (thyA-) L. pneumophila variant, and plasmid retention is routinely achieved by including the thymidine biosynthesis gene (thyA) on the plasmid backbone. Only with a plasmid bearing the thyA gene can L. pneumophila grow on CYE (thymidine-) plates. Our use of vectors bearing thyA and plating on CYE plates is described in the Methods section. Further, in Figure 7 of our 2021 paper, we show that CRISPR plasmids are efficiently retained by Lp02 for the duration of a 48-hour infection, resulting in efficient multi-gene knockdown even at the end of the intracellular growth experiment.

Regarding comments A and B, on publishing the biological data used to classify genes in gene families for 10-plex silencing, we do not consider it critical to provide additional information beyond the broad classification (e.g. kinases, phosphatases, etc) described in Table S1. Structural predictions constantly change due to continuously evolving databases. Our initial analyses were made in 2015 using HHPRED Hidden-Markov models and, in all likelihood, those predictions have been refined since then. Moreover, with the recent advent of Alphafold, anyone interested in learning more about select effectors from our list is advised to simply access the most recent functional predictions directly on the Alphafold webpage (https://alphafold.ebi.ac.uk/). We clarify how predictions were made on Lines 97-101.

Regarding specific comment D, on our method for qPCR normalization and comparison, we point Reviewer 1 to the Methods section (Lines 460+) where we describe that data obtained from each CRISPRi strain were in fact normalized to the levels of rpsL prior to comparing them to the normalized data from the strain with the empty control plasmid. This normalization to rpsL, a gene encoding a ribosomal protein, also corrects for growth differences between samples.

Regarding specific comment H, the justification for studying 44 transmembrane effector-encoding genes was driven by the fact that activities mediated by transmembrane proteins are difficult (though not impossible) to be replaced by cytosolic proteins, for example the transport of metabolites across the LCV membrane. And since transmembrane regions can be predicted with high confidence, we decided to probe this group of TMEs for synthetic lethality with the randomized CRISPRi approach as proof-of-concept. To make this clearer, we have added more detail to the text (Lines 151-155).

Regarding specific comment I, we have further simplified the description of the cloning technique to increase clarity (Lines 156+). The information listed under Supplemental Note S1, though informative, is not critical for the overall understanding of this highly technical section, and since the reviewer already considered this section to be difficult to follow, we would prefer to not further complicate the text by including these non-essential details.

Regarding the origin of the CRISPRi plasmid (specific comment J), we point Reviewer 1 to the reference (Hammer BK and Swanson MS (Mol Microbiol 1999)) listed in Table S10: Strains and Plasmids Used in this Study.

Regarding specific comment K and O, on the clarity of depicting the CRISPR array size selection process, we have updated the Figure 2 schematic. Reviewer 1 is correct in that despite our best efforts to exclude short CRISPR arrays, inevitably some 1-plex arrays remained in our input vector pool. Still, the average length of all arrays used in our pilot study exceeded three crRNA-encoding spacers. Further, having a population of 1- or 2-plex arrays in our libraries did allow us to pin-point the most critical effectors of a larger arrays within the same MuRCiS experiment (Table S5 and Table S7), a strength of MuRCiS as described in the discussion (Lines 378+).

Regarding specific comment L, we appreciate Reviewer 1’s suggestion of an additional reference and we have added it to the manuscript as reference #23 (Line 71). While this reference does use a Golden Gate strategy to build a multiplex array, that array was not randomized but had a predefined order. Hence, our assembly method is unique due to its randomization.

Regarding specific comment M, on array length cloning limitations, we agree with the conclusion of Zuckermann in Figure 1d of their article that longer inserts are generally harder to get into vector backbones. The challenge of cloning longer inserts is a common phenomenon of general biology and is not unique to cloning CRISPR arrays. We have altered the wording in our manuscript to better describe the intrinsic competition between short and long inserts during cloning (Lines 162-164).

Regarding specific comment N, we second Reviewer 1’s desire to learn more about the critical effector pairs discovered here. With that said, the goal of this manuscript is to report the development of a novel MuRCiS pipeline to identify these critical pairs. Biochemical and molecular investigations of the encoded protein pairs are on-going and will be the topic of a future manuscript.

Reviewer #2 (Recommendations For The Authors):

Specific points

1) The effector repertoire of L. pneumophila seems to have evolved in response to the plethora of potential protozoan hosts (PMID: 31988381). To further assess evolutionary aspects of the vast L. pneumophila effector arsenal, it would be interesting to test the single and double effector mutant strains (Fig. 5FG, Fig. 6EF) for growth in protozoa other than A. castellanii.

2) Most CRISPR arrays comprising genes encoding functionally similar proteins or encoding evolutionarily conserved proteins did not substantially affect intracellular growth of L. pneumophila (Fig. 1B). This rather surprising result should be further discussed.

3) l. 118/119: "Similar results ..., where none of the MC arrays ..." This statement should be phrased more precisely, since some CRISPR arrays did indeed have an effect on intracellular growth of L. pneumophila in U937 macrophages, while none affected intracellular growth in A. castellanii (Fig. 1B).

4) Typos:

• l. 852: ... (arbitrarily set to -100).

• l. 862: ... Legionella-containing vacuole (LCV).

• l. 895: ... and so we would recommend ...

Regarding point 1, we thank Reviewer 2 for the suggestion of testing effector mutants in different hosts. While the primary purpose of the current manuscript was to optimize the MuRCiS platform, future studies using this technology to investigate specific biological questions related to Legionella infection would certainly benefit from including more than one amoebaean species.

Regarding point 2, we agree that the lack of substantial growth defects seems surprising. Yet only two of the seven core effectors (found in all Legionella sp.), lpg2300 and mavN, individually attenuated Legionella intracellular growth when deleted (Burstein 2016 Nat Genetics; Isaac et al., 2015 PNAS). Thus, we hypothesize that the functions many effectors fulfil are of such importance for intracellular survival that that redundancy reaches beyond the boundary of conservation or like-function. We have added a statement emphasizing this at the end of the Figure 1 results section (Line 122-125).

Regarding points 3 and 4, we thank Reviewer 2 for catching these errors and have corrected where needed in the text.

-l. 852 (now Line 874): … (arbitrarily set to -100,000) is correct for Figure 6E.

eLife assessment

This important study uses CRISPRi to silence multiple effectors in the pathogen, Legionella pneumophila. It provides a technique that will allow researchers to address functional redundancy amongst effectors, a problem that has persisted even after decades of study. The methodology used is convincing, and further improvement (such as using multiple guides per gene) can lead to the identification of novel virulence factors.

Reviewer #1 (Public Review):

The article "A randomized multiplex CRISPRi-Seq approach for the identification of critical combinations of genes" describes the development of a multiplex randomized CRISPRi screening method that they named MurCiS and applied it to study redundancy of L. pneumophila virulence factors. The authors used a L. pneumophila strain carrying dCas9 on the chromosome that they had constructed for a CRISPRi screen they had published recently and here combined it with self-assembly randomized multiplex CRISPR arrays that they developed. The strains carrying the dCas9 and the different CRISPRi arrays were used to infect U937 or Acanthamoeba castellanii cells and the intracellular growth phenotypes were recorded as readout. This allowed the authors to identify certain gene combinations that when knocked down induced a growth defect in either or both cells tested but not when they were knocked down alone. A particular gene combination caught their attention, as the genes lpg2888 and lpg3000 were inducing a growth defect only when both were knocked down in U937 cells but in A. castellanii cells lpg3000 alone was sufficient to cause a growth defect.

The concept of using CRISPRi to look at functional redundancy in effectors is a very useful one to the Legionella field and where biological redundancy limits studies. It has the potential to uncover virulence effectors of importance that have not been described before.

Comments on revised version: In this revised version the authors have answered our concerns satisfactorily except the point related to the use of only one guide per gene.

Reviewer #2 (Public Review):

The study by Ellis et al. documents the development of a CRISPR interference (CRISPRi) screen aiming at identifying virulence-critical genes of Legionella pneumophila, the facultative intracellular bacterium causing Legionnaires' disease. L. pneumophila employs the Dot/Icm type IV secretion system to translocate more than 300 different "effector proteins" into host cells. Many effector proteins appear to have redundant functions, and therefore, depleting several of them is required to observe a strong intracellular replication phenotype. In the current study, Ellis et al. develop a "multiplex, randomized CRISPRi sequencing" (MuRCiS) approach to silence several effector genes simultaneously and randomly, thereby possibly causing synthetic lethality for L. pneumophila upon infection of host cells.

The MuRCiS approach comprises the ligation of different CRISPR spacers flanked by repeats in presence of "dead end" oligonucleotide pairs capping a random array of building blocks to be inserted into a library vector. Thus, spacer arrays with an average of 3.3 spacers per array were obtained. As a proof-of-concept, spacer arrays targeting 44 transmembrane effector-encoding L. pneumophila genes were employed to screen for intracellular growth defects in macrophages and amoeba. Consequently, novel pairs of synergistically functioning effector genes were identified by comparative next-generation sequencing of the input and output pools of spacer arrays.

A major strength of this well-written and straightforward study is the construction and use of random and multiplexed CRISPRi arrays, allowing an unbiased and comprehensive screen for multiple genes affecting the intracellular growth of L. pneumophila. The ingenious approach established by Ellis et al. will be useful for further genetic analysis of L. pneumophila infection and might also be adopted for other pathogens employing a large set of (functionally redundant) virulence factors.<br /> The reviewer's suggestion to test the single and double L. pneumophila effector mutant strains for growth in protozoa other than A. castellanii was considered beyond the scope of the current manuscript describing the optimization of the MuRCiS platform. The authors have satisfactorily addressed the minor points raised previously.

Author Response

The following is the authors’ response to the previous reviews

Comments from reviewer 1:

Comment 1. Regarding SBSMMA, the authors may complement their discussion by mentioning recent work (PMID: 35738428) where SBSMMA was used to exemplify a potential fragment-based design approach for developing allosteric effectors for kinases.

Thank you for the suggestion, we have added a short summary of the work where SBSMMA is used as a basis for developing small molecules to target kinases using fragment-based design approach

eLife assessment

One of the most promising strategies in development of drugs targeting kinases is provided by using allosteric control that allows specific regulation and study of kinase function without directly targeting the active site. This important paper reviews convincingly the current repertoire of tools for regulating the activity of protein kinases with the ultimate goal of developing novel approaches in treating diseases associated with signal dysregulation.

Joint Public Review:

This concise review provides a clear and instructive picture of the state-of-the-art understanding of protein kinases' activity and sets of approaches and tools to analyse and regulate it.

Three major parts of the work include: methods to map allosteric communications, tools to control allostery, and allosteric regulation of protein kinases. The work provides an important and timely view of the current status of our understanding of the function of protein kinases and state-of-the-art methods to study its allosteric regulation and to develop allosteric approaches to control it.

Authorr Response

The following is the authors’ response to the original reviews.

Reviewer #1 (Public Review):

The objective of this investigation was to determine whether experimental pain could induce alterations in cortical inhibitory/facilitatory activity observed in TMS-evoked potentials (TEPs). Previous TMS investigations of pain perception had focused on motor evoked potentials (MEPs), which reflect a combination of cortical, spinal, and peripheral activity, as well as restricting the focus to M1. The main strength of this investigation is the combined use of TMS and EEG in the context of experimental pain. More specifically, Experiment 1 investigated whether acute pain altered cortical excitability, reflected in the modulation of TEPs. The main outcome of this study is that relative to non-painful warm stimuli, painful thermal stimuli led to an increase on the amplitude of the TEP N45, with a larger increase associated with higher pain ratings. Because it has been argued that a significant portion of TEPs could reflect auditory potentials elicited by the sound (click) of the TMS, Experiment 2 constituted a control study that aimed to disentangle the cortical response related to TMS and auditory activity. Finally, Experiment 3 aimed to disentangle the cortical response to TMS and reafferent feedback from muscular activity elicited by suprathreshold TMS applied over M1. The fact that the authors accompanied their main experiment with two control experiments strengthens the conclusion that the N45 TEP peak could be implicated in the perception of painful stimuli.

Perhaps, the addition of a highly salient but non-painful stimulus (i.e. from another modality) would have further ruled out that the effects on the N45 are not predominantly related to intensity/saliency of the stimulus rather than to pain per se.

We thank the reviewer for their comment on the possibility of whether stimulus intensity influences the N45 as opposed to pain per se. We agree that the ideal experiment would have included multiple levels of stimulation. We would argue, however, that that in Experiment 1, despite the same level of stimulus intensity for all participants (46 degrees), individual differences in pain ratings were associated with the change in the N45 amplitude, suggesting that the results cannot be explained by stimulus intensity, but rather by pain intensity.

Reviewer #2 (Public Review):

The authors have used transcranial magnetic stimulation (TMS) and motor evoked potentials (MEPs) and TMS-electroencephalography (EEG) evoked potentials (TEPs) to determine how experimental heat pain could induce alterations in these metrics.
In Experiment 1 (n = 29), multiple sustained thermal stimuli were administered over the forearm, with the first, second, and third block of stimuli consisting of warm but non-painful (pre-pain block), painful heat (pain block) and warm but non-painful (post-pain block) temperatures respectively. Painful stimuli led to an increase in the amplitude of the fronto-central N45, with a larger increase associated with higher pain ratings. Experiments 2 and 3 studied the correlation between the increase in the N45 in pain and the effects of a sham stimulation protocol/higher stimulation intensity. They found that the centro-frontal N45 TEP was decreased in acute pain. The study comes from a very strong group in the pain fields with long experience in psychophysics, experimental pain, neuromodulation, and EEG in pain. They are among the first to report on changes in cortical excitability as measured by TMS-EEG over M1. While their results are in line with reductions seen in motor-evoked responses during pain and effort was made to address possible confounding factors (study 2 and 3), there are some points that need attention. In my view the most important are:

1) The method used to calculate the rest motor threshold, which is likely to have overestimated its true value : calculating highly abnormal RMT may lead to suprathreshold stimulations in all instances (Experiment 3) and may lead to somatosensory "contamination" due to re-afferent loops in both "supra" and "infra" (aka. less supra) conditions.

The method used to assess motor threshold was the TMS motor threshold Assessment Tool (MTAT) which estimates motor threshold using maximum likelihood parametric estimation by sequential testing (Awiszus et al., 2003; Awiszus and Borckardt, 2011). This was developed as a quicker alternative for calculating motor threshold compared to the traditional Rossini-Rothwell method which involves determining the lowest intensity that evokes at least 5/10 MEPs of at least 50 microvolts. The method has been shown to achieve the same accuracy of determining motor threshold as the traditional Rossini-Rothwell method, but with fewer pulses (Qi et al., 2011; Silbert et al., 2013).

We have now made this clearer in the manuscript:

“The RMT was determined using the TMS motor thresholding assessment tool, which estimates the TMS intensity required to induce an MEP of 50 microvolts with a 50% probability using maximum likelihood parametric estimation by sequential testing (Awiszus, 2003; Awiszus & Borckardt, 2011). This method has been shown to achieve the accuracy of methods such as the Rossini-Rothwell method (Rossini et al., 1994; Rothwell et al., 1999) but with fewer pulses (Qi, Wu, & Schweighofer, 2011; Silbert, Patterson, Pevcic, Windnagel, & Thickbroom, 2013). The test stimulus intensity was set at 110% RMT to concurrently measure MEPs and TEPs during pre-pain, pain and post-pain blocks.”

Therefore, the high RMTs in our study cannot be explained by the threshold assessment method. Instead, they are likely explained by aspects of the experimental setup that increased the distance between the TMS coil and the scalp, including the layer of foam placed over the coil, the EEG cap and the fact that the electrodes we used had a relatively thick profile. This has been explained in the paper:

“We note that the relatively high RMTs are likely due to aspects of the experimental setup that increased the distance between the TMS coil and the scalp, including the layer of foam placed over the coil, the EEG cap and relatively thick electrodes (6mm)”

Awiszus, F. (2003). TMS and threshold hunting. In Supplements to Clinical neurophysiology (Vol. 56, pp. 13-23). Elsevier.

Qi, F., Wu, A. D., & Schweighofer, N. (2011). Fast estimation of transcranial magnetic stimulation motor threshold. Brain stimulation, 4(1), 50-57.

Silbert, B. I., Patterson, H. I., Pevcic, D. D., Windnagel, K. A., & Thickbroom, G. W. (2013). A comparison of relative-frequency and threshold-hunting methods to determine stimulus intensity in transcranial magnetic stimulation. Clinical Neurophysiology, 124(4), 708-712.

2) The low number of pulses used for TEPs (close to ⅓ of the usual and recommended)

We agree that increasing the number of pulses can increase the signal to noise ratio. During piloting, participants were unable to tolerate the painful stimulus for long periods of time and we were required to minimize the number of pulses per condition.

We note that there is no set advised number of trials in TMS-EEG research. According to the recommendations paper, the number of trials should be based on the outcome measure e.g., TEP peaks vs. frequency domain measures vs. other measures and based on previous studies investigating test-retest reliability (Hernandez-Pavon et al., 2023). The choice of 66 pulses per condition was based on the study by Kerwin et al., (2018) showing that optimal concordance between TEP peaks can be found with 60-100 TMS pulses delivered in the same run (as in the present study). The concordance was particularly higher for the N40 peak at prefrontal electrodes, which was the key peak and electrode cluster in our study. We have made this clearer:

“Current recommendations (Hernandez-Pavon et al., 2023) suggest basing the number of TMS trials per condition on the key outcome measure (e.g., TEP peaks vs. frequency measures) and based on previous test-retest reliability studies. In our study the number of trials was based on a test-retest reliability study by (Kerwin, Keller, Wu, Narayan, & Etkin, 2018) which showed that 60 TMS pulses (delivered in the same run) was sufficient to obtain reliable TEP peaks (i.e., sufficient within-individual concordance between the resultant TEP peaks of each trial).”

Further supporting the reliability of the TEP data in our experiment, we note that the scalp topographies of the TEPs for active TMS at various timepoints (Figures 5, 7 and 9) were similar across all three experiments, especially at 45 ms post-TMS (frontal negative activity, parietal-occipital positive activity).

In addition to this, the interclass correlation coefficient (Two-way fixed, single measure) for the N45 to active suprathreshold TMS across timepoints for each experiment was 0.90 for Experiment 1 (across pre-pain, pain, post-pain time points), 0.74 for Experiment 2 (across pre-pain and pain conditions), and 0.95 for Experiment 3 (across pre-pain conditions). This suggests that even with the fluctuations in the N45 induced by pain, the N45 for each participant was stable across time, further supporting the reliability of our data. These ICCs are now reported in the supplementary material (subheading: Test-retest reliability of N45 Peaks).

Hernandez-Pavon, J. C., Veniero, D., Bergmann, T. O., Belardinelli, P., Bortoletto, M., Casarotto, S., ... & Ilmoniemi, R. J. (2023). TMS combined with EEG: Recommendations and open issues for data collection and analysis. Brain Stimulatio, 16(3), 567-593

Kerwin, L. J., Keller, C. J., Wu, W., Narayan, M., & Etkin, A. (2018). Test-retest reliability of transcranial magnetic stimulation EEG evoked potentials. Brain stimulation, 11(3), 536-544.

Lack of measures to mask auditory noise.

In TMS-EEG research, various masking methods have been proposed to suppress the somatosensory and auditory artefacts resulting from TMS pulses, such as white noise played through headphones to mask the click sound (Ilmoniemi and Kičić, 2010), and a thin layer of foam placed between the TMS coil and EEG cap to minimize the scalp sensation (Massimini et al., 2005). However, recent studies have shown that even when these methods are used, sensory contamination of TEPs is still present, as shown by studies that show commonalities in the signal between active and sensory sham conditions that mimic the auditory/somatosensory aspects of real TMS (Biabani et al., 2019; Conde et al., 2019; Rocchi et al., 2021). This has led many authors (Biabani et al., 2019; Conde et al., 2019) to recommend the use of sham conditions to control for sensory contamination. To separate the direct cortical response to TMS from sensory evoked activity, Experiment 2 included a sham TMS condition that mimicked the auditory/somatosensory aspects of active TMS to determine whether any alterations in the TEP peaks in response to pain were due to changes in sensory evoked activity associated with TMS, as opposed to changes in cortical excitability. Therefore, the lack of auditory masking does not impact the main conclusions of the paper.

We have made this clearer:

“… masking methods have been used to suppress these sensory inputs, (Ilmoniemi and Kičić, 2010; Massimini et al., 2005). However recent studies have shown that even when these methods are used, sensory contamination of TEPs is still present, as shown by commonalities in the signal between active and sensory sham conditions that mimic the auditory/somatosensory aspects of real TMS (Biabani et al., 2019; Conde et al., 2019; Rocchi et al., 2021). This has led many leading authors (Biabani et al., 2019; Conde et al., 2019) to recommend the use of sham conditions to control for sensory contamination.”

Ilmoniemi, R. J., & Kičić, D. (2010). Methodology for combined TMS and EEG. Brain topography, 22, 233-248.

Massimini, M., Ferrarelli, F., Huber, R., Esser, S. K., Singh, H., & Tononi, G. (2005). Breakdown of cortical effective connectivity during sleep. Science, 309(5744), 2228-2232.

Biabani, M., Fornito, A., Mutanen, T. P., Morrow, J., & Rogasch, N. C. (2019). Characterizing and minimizing the contribution of sensory inputs to TMS-evoked potentials. Brain stimulation, 12(6), 1537-1552.

Conde, V., Tomasevic, L., Akopian, I., Stanek, K., Saturnino, G. B., Thielscher, A., ... & Siebner, H. R. (2019). The non-transcranial TMS-evoked potential is an inherent source of ambiguity in TMS-EEG studies. Neuroimage, 185, 300-312.

Rocchi, L., Di Santo, A., Brown, K., Ibáñez, J., Casula, E., Rawji, V., ... & Rothwell, J. (2021). Disentangling EEG responses to TMS due to cortical and peripheral activations. Brain stimulation, 14(1), 4-18.

3) A supra-stimulus heat stimulus not based on individual HPT, that oscillates during the experiment and that lead to large variations in pain intensity across participants is unfortunate.

The choice of whether to calibrate or fix stimulus intensity is a contentious question in experimental pain research. A recent discussion by Adamczyk et al., (2022) explores the pros and cons of each approach and recommends situations where one method may be preferred over the other. That paper suggests that the choice of the methodology is related to the research question – when the main outcome of the research is objective (neurophysiological measures) and researchers are interested in the variability in pain ratings, the fixed approach is preferrable. Given we explored the relationship between MEP/N45 modulation by pain and pain intensity, this question is better explored by using the same stimulus intensity for all participants, as opposed to calibrating the intensity to achieve a similar level of pain across participants.

We have made this clearer:

“Given we were interested in the individual relationship between pain and excitability changes, the fixed temperature of 46ºC ensured larger variability in pain ratings as opposed to calibrating the temperature of the thermode for each participant (Adamczyk et al., 2022).”.

Adamczyk, W. M., Szikszay, T. M., Nahman-Averbuch, H., Skalski, J., Nastaj, J., Gouverneur, P., & Luedtke, K. (2022). To calibrate or not to calibrate? A methodological dilemma in experimental pain research. The Journal of Pain, 23(11), 1823-1832.

So is the lack of report on measures taken to correct for a fortuitous significance (multiple comparison correction) in such a huge number of serial paired tests.

Note that we used a Bayesian approach for all analyses as opposed to the traditional frequentist approach. In contrast to the frequentist approach, the Bayesian approach does not require corrections for multiple comparisons (Gelman et al., 2000) given that they provide a ratio representing the strength of evidence for the null vs. alternative hypotheses as opposed to accepting or rejecting the null hypothesis based on p-values. As such, throughout the paper, we frame our interpretations and conclusions based on the strength of evidence (e.g. anecdotal/weak, moderate, strong, very strong) as opposed to referring to the significance of the effects.

Gelman A, Tuerlinckx F. (2000). Type S error rates for classical and Bayesian single and multiple comparison procedures. Computational statistics, 15(3):373-90.

Reviewer #3 (Public Review):

The present study aims to investigate whether pain influences cortical excitability. To this end, heat pain stimuli are applied to healthy human participants. Simultaneously, TMS pulses are applied to M1 and TMS-evoked potentials (TEPs) and pain ratings are assessed after each TMS pulse. TEPs are used as measures of cortical excitability. The results show that TEP amplitudes at 45 msec (N45) after TMS pulses are higher during painful stimulation than during non-painful warm stimulation. Control experiments indicate that auditory, somatosensory, or proprioceptive effects cannot explain this effect. Considering that the N45 might reflect GABAergic activity, the results suggest that pain changes GABAergic activity. The authors conclude that TEP indices of GABAergic transmission might be useful as biomarkers of pain sensitivity.

Pain-induced cortical excitability changes is an interesting, timely, and potentially clinically relevant topic. The paradigm and the analysis are sound, the results are mostly convincing, and the interpretation is adequate. The following clarifications and revisions might help to improve the manuscript further.

1) Non-painful control condition. In this condition, stimuli are applied at warmth detection threshold. At this intensity, by definition, some stimuli are not perceived as different from the baseline. Thus, this condition might not be perfectly suited to control for the effects of painful vs. non-painful stimulation. This potential confound should be critically discussed.

In Experiment 3, we also collected warmth ratings to confirm whether the pre-pain stimuli were perceived as different from baseline. This detail has been added to them methods:

“In addition to the pain rating in between TMS pulses, we collected a second rating for warmth of the thermal stimulus (0 = neutral, 10 = very warm) to confirm that the participants felt some difference in sensation relative to baseline during the pre-pain block. This data is presented in the supplementary material”.

We did not include these data in the initial submission but have now included it in the supplemental material. These data showed warmth ratings were close to 2/10 on average. This confirms that the non-painful control condition produced some level of non-painful sensation.

2) MEP differences between conditions. The results do not show differences in MEP amplitudes between conditions (BF 1.015). The analysis nevertheless relates MEP differences between conditions to pain ratings. It would be more appropriate to state that in this study, pain did not affect MEP and to remove the correlation analysis and its interpretation from the manuscript.

The interindividual relationship between changes in MEP amplitude and individual pain rating is statistically independent from the overall group level effect of pain on MEP amplitude. Therefore, conclusions for the individual and group level effects can be made independently.

It is also important to note that in the pain literature, there is now increasing emphasis placed on investigating the individual level relationship between changes in cortical excitability and pain as opposed to the group level effect (Seminowicz et al., 2019; Summers et al., 2019). As such, it is important to make these results readily available for the scientific community.

We have made this clearer:

‘As there is now increasing emphasis placed on investigating the individual level relationship between changes in cortical excitability and pain and not only the group level effect, (Chowdhury et al., 2022; Seminowicz et al., 2018; Seminowicz, Thapa, & Schabrun, 2019; Summers et al., 2019) we also investigated the correlations between pain ratings and changes in MEP (and TEP) amplitude”

Chowdhury, N. S., Chang, W. J., Millard, S. K., Skippen, P., Bilska, K., Seminowicz, D. A., & Schabrun, S. M. (2022). The Effect of Acute and Sustained Pain on Corticomotor Excitability: A Systematic Review and Meta-Analysis of Group and Individual Level Data. The Journal of Pain, 23(10), 1680-1696.

Summers, S. J., Chipchase, L. S., Hirata, R., Graven-Nielsen, T., Cavaleri, R., & Schabrun, S. M. (2019). Motor adaptation varies between individuals in the transition to sustained pain. Pain, 160(9), 2115-2125.

Seminowicz, D. A., Thapa, T., & Schabrun, S. M. (2019). Corticomotor depression is associated with higher pain severity in the transition to sustained pain: a longitudinal exploratory study of individual differences. The Journal of Pain, 20(12), 1498-1506.

3) Confounds by pain ratings. The ISI between TMS pulses is 4 sec and includes verbal pain ratings. Considering this relatively short ISI, would it be possible that verbal pain ratings confound the TEP? Moreover, could the pain ratings confound TEP differences between conditions, e.g., by providing earlier ratings when the stimulus is painful? This should be carefully considered, and the authors might perform control analyses.

It is unlikely that the verbal ratings contaminated the TEP response as the subsequent TMS pulse was not delivered until the verbal rating was complete and given that each participant was cued by the experimenter to provide the pain rating after each pulse (rather than the participant giving the rating at any time). As such, it would not be possible for participants to provide earlier ratings to more painful stimuli.

We have made this clearer:

"To avoid contamination of TEPs by verbal ratings, the subsequent TMS pulse was not delivered until the verbal rating was complete, and the participant was cued by the experimenter to provide the pain rating after each pulse.”

4) Confounds by time effects. Non-painful and painful conditions were performed in a fixed order. Potential confounds by time effects should be carefully considered.

Previous research suggests that pain alters neural excitability even after pain has subsided. In a recent meta-analysis (Chowdhury et al., 2022) we found effect sizes of 0.55-0.9 for MEP reductions 0-30 minutes after pain had resolved. As such, we avoided intermixing pain and warm blocks given subsequent warm blocks would not serve as a valid baseline, as each subsequent warm block would have residual effects from the previous pain blocks.

Chowdhury, N. S., Chang, W. J., Millard, S. K., Skippen, P., Bilska, K., Seminowicz, D. A., & Schabrun, S. M. (2022). The Effect of Acute and Sustained Pain on Corticomotor Excitability: A Systematic Review and Meta-Analysis of Group and Individual Level Data. The Journal of Pain, 23(10), 1680-1696.

At the same time, given there was no conclusive evidence for a difference in N45 amplitude between pre-pain and post-pain conditions of Experiment 1 (Supplementary Figure 1), it is unlikely that the effect of pain was an artefact of time i.e., the explanation that successive thermal stimuli applied to the skin results an increase in the N45, regardless of whether the stimuli are painful or not. We will make this point in our next revision.

We have discussed this issue:

“Lastly, future research should consider replicating our experiment using intermixed pain and no pain blocks, as opposed to fixed pre-pain and pain blocks, to control for order effects i.e., the explanation that successive thermal stimuli applied to the skin results an increase in the N45 peak, regardless of whether the stimuli are painful or not. However, we note that there was no conclusive evidence for a difference in N45 peak amplitude between pre-pain and post-pain conditions of Experiment 1 (Supplementary Figure 1), suggesting it is unlikely that the observed effects were an artefact of time.”

5) Data availability. The authors should state how they make the data openly available.

We have uploaded the MEP, TEP and pain data on the Open science framework https://osf.io/k3psu/

Reviewer #1 (Recommendations For The Authors):

I think the study is quite solid and I only have very minor recommendations for the authors:

• Introduction, p. 3: "Functional magnetic resonance imaging has helped us understand where in the brain pain is processed". This is an overstatement. fMRI provides us with potential biomarkers (e.g. "the pain signature"), but the specificity of these responses for pain is debated and we still do not know where in the brain pain is processed.

We have amended to:

“functional magnetic resonance imaging has assisted in the localization of brain structures implicated in pain processing”

• Introduction, p. 5: "neural baseline" should be "neutral baseline"?

We thank the reviewer for identifying this – this has now been amended.

Reviewer #2 (Recommendations For The Authors):

INTRODUCTION

The introduction mentions how important extra-motor areas can be explored by TMS-EEG, then the effects of DLPFC rTMS on TEPs ... but you do not explore the DLPFC... Perhaps the introduction should be reframed.

The current work explores cortical excitability throughout the brain (as shown in our cluster-based permutation and source localization analyses), so our investigations are in line with the introductions statement about the importance of studying non-motor areas.

The reference to DLPFC rTMS was to highlight current existing research that has applied TMS-EEG to understand pain. It was not used as a methodological rationale to investigate the DLPFC in the present study. To make the research gap clearer, we state:

“While these studies assist us in understanding whether TEPs might mediate rTMS-induced pain reductions, no study has investigated whether TEPs are altered in direct response to pain”

Lignes 63-65 the term "TMS" is used to refer to motor corticospinal excitability measures, in contrast to TMS-EEG measures of TEPs. Then the authors come back to TMS-EEG and then again back to MEPs. This is rather confusing: TMS means TMS... the concept of MEP/ motor corticospinal excitability measures is not intuitive when using the term "TMS". I suggest using motor corticospinal excitability measures when referring to MEP/MEP-based measures of cortical excitability...) and M1TMS-EEG-evoked potentials (usually abbreviated to TEPs) to refer to TMS-EEG responses as measured here.

Throughout the manuscript, we now use the term TEPs when referring to TMS-EEG measures, and MEPs when referring to TMS-EMG measure. The use of TEPs vs. MEPs will make it easier for readers to follow which measures we are referring to.

Line 83: "As such, the precise origin of the pain mechanism cannot be localized." Please rephrase, the sentence conveys the idea that it is indeed possible to localize the origin of a pain mechanism with a different approach, and we know this is not currently possible, irrespective of the methodological setup.

We have replaced this with:

“This makes it unclear as to whether pain processes occur at the cortical, spinal or peripheral level.”

How can one predetermine the temperature that will be perceived as painful by someone else, and not base it on individual HPT? This is against principles of psychophysics. Please comment. Attesting all participants had HPT below 46 is important, but then being stimulated at 46C when our HPT is 45C is different from when our HPT is 39C. Please explain why the pain intensity was not standardised based on individual HPT.

Please refer to our response to the public review related to the issue

Line 38: "if we had used an alternative design with blocks of warm stimuli intermixed with blocks of painful stimuli, the warm stimuli blocks would not serve as a valid non-painful baseline". I do not understand why it is not possible to have a pain-free baseline, followed by a pain/warm sequence.

In our study, we had the choice of either intermixing blocks or to use a fixed sequence. Previous research suggests that pain alters neural excitability even after pain has subsided. In a recent meta-analysis (Chowdhury et al., 2022) we found effect sizes of 0.55-0.9 for MEP reductions 0-30 minutes after pain had resolved. As such, we avoided intermixing pain and warm blocks given subsequent warm blocks would not serve as a valid baseline, as each subsequent warm block would have residual effects from the previous pain blocks.

We have updated the manuscript to be clearer about why we used a fixed sequence:

“The pre-pain/pain/post-pain design has been commonly used in the TMS-MEP pain literature, as many studies have demonstrated strong changes in corticomotor excitability that persist beyond the painful period. Indeed, in a systematic review, we showed effect sizes of 0.55-0.9 for MEP reductions 0-30 minutes after pain had resolved (Chowdhury et al., 2022). As such, if we had used an alternative design with blocks of warm stimuli intermixed with blocks of painful stimuli, the warm stimuli blocks would not serve as a valid non-painful baseline”

Chowdhury, N. S., Chang, W. J., Millard, S. K., Skippen, P., Bilska, K., Seminowicz, D. A., & Schabrun, S. M. (2022). The Effect of Acute and Sustained Pain on Corticomotor Excitability: A Systematic Review and Meta-Analysis of Group and Individual Level Data. The Journal of Pain, 23(10), 1680-1696.

Please explain, and provide evidence that stimulation of people with predetermined temperatures is able to create warm/pain/warm sensations, without entraining pain in the last warm stimulation.

A previous study by Dube et al. (2011) used sequences of warm (36°C), painful and neutral (32° C) and found that participants did not experience pain at any time when the temperature was at a warm temperature of 36°C. We have now cited this study:

“Based on a previous study (Dubé & Mercier, 2011) which also used sequences of painful (50ºC) and warm (36°C) thermal stimuli, we did not anticipate that the stimulus in the pain block would entrain pain in the post-pain block”

Dubé, J. A., & Mercier, C. (2011). Effect of pain and pain expectation on primary motor cortex excitability. Clinical neurophysiology, 122(11), 2318-2323.

METHODS

It is not clear if participants with chronic pain, present in 20% of the general population, were excluded. If they were, please provide "how" in methods.

We excluded participants with a history or presence of acute/chronic pain. This has now been clarified:

“Participants were excluded if they had a history of chronic pain condition or any current acute pain”

Line 489: the definition of warm detection threshold is unusual, please provide a reference.

We used an identical method to Furman et al., (2020). We have made the reference to this clearer: “Warmth, cold and pain thresholds were assessed in line with a previous study (Furman et al., 2020)”

Furman, A. J., Prokhorenko, M., Keaser, M. L., Zhang, J., Chen, S., Mazaheri, A., & Seminowicz, D. A. (2020). Sensorimotor peak alpha frequency is a reliable biomarker of prolonged pain sensitivity. Cerebral Cortex, 30(12), 6069-6082.

In Experiment 2, please explain how the lack of randomisation between "pre-pain" and "pain" may have influenced results.

Given we tried to replicate Experiment 1’s methodology as close as possible (to isolate the source of the effect from Experiment 1) we chose to repeat the same sequence of blocks as Experiment 1: pre-pain followed by pain.

Given there was no conclusive evidence for a difference in N45 amplitude between pre-pain and post-pain conditions of Experiment 1 (Supplementary Figure 1), it is unlikely that the effect of pain was an order effect i.e., the explanation that successive thermal stimuli applied to the skin results an increase in the N45, regardless of whether the stimuli are painful or not.

We now discuss the issue of randomization:

“Lastly, future research should consider replicating our experiment using intermixed pain and no pain blocks, as opposed to fixed pre-pain and pain blocks, to control for order effects i.e. the explanation that successive thermal stimuli applied to the skin results an increase in the N45 peak, regardless of whether the stimuli are painful or not. However, we note that there was no conclusive evidence for a difference in N45 peak amplitude between pre-pain and post-pain conditions of Experiment 1 (Supplementary Figure 1), suggesting it is unlikely that the observed effects were an artefact of time”

Also, in Methods in general, disclose how pain intensity was assessed, and how.

Pain intensity was assessed using a verbal rating scale (0 = no pain, and 10 = most pain imaginable). We have provided more detail:

“During each 40 second thermal stimulus, TMS pulses were manually delivered, with a verbal pain rating score (0 = no pain, and 10 = worst pain imaginable) obtained between pulses. To avoid contamination of TEPs by verbal ratings, the subsequent TMS pulse was not delivered until the verbal rating was complete, and the participant was cued by the experimenter to provide the pain rating after each pulse”

Please explain how auditory masking was made during data collection.

Auditory masking noise was not played through the headphones, given that Experiment 2 controlled for auditory evoked potentials. We have made this clearer:

“Auditory masking was not used. Instead, auditory evoked potentials resulting from the TMS click sound were controlled for in Experiment 2”

Please explain if online TEP monitoring was used during data collection

Online TEP monitoring was not available with our EEG software. We have made this clearer in the manuscript:

“Online TEP monitoring was not available with the EEG software”

Line 499: what is subthreshold TMS here? You are measuring TEPs, and not MEPs initially, so you may have a threshold for MEPs and TEPs, which are not the same.

The intensity was calibrated relative to the MEP response (rather than TEP response) - this has now been clarified:

“… and the inclusion of a subthreshold TMS (90% of resting motor threshold) condition intermixed within both the pre-pain and pain blocks.”

Please provide a reference and a figure to illustrate the electric stimulation used in the sham procedure in Study 2

The apparatus for the electrical stimulation is shown in Figure 7A, and was based on previous papers using electrical stimulation over motor cortex to simulate the somatosensory aspect of real TMS (Chowdhury et al., 2022; Gordon et al., 2022; Rocchi et al., 2021). We have made this clearer:

“Electrical stimulation was based on previous studies attempting to simulate the somatosensory component of active TMS (Chowdhury et al., 2022; Gordon et al., 2022; Rocchi et al., 2021)”

Gordon, P. C., Jovellar, D. B., Song, Y., Zrenner, C., Belardinelli, P., Siebner, H. R., & Ziemann, U. (2021). Recording brain responses to TMS of primary motor cortex by EEG–utility of an optimized sham procedure. Neuroimage, 245, 118708.

Chowdhury, N. S., Rogasch, N. C., Chiang, A. K., Millard, S. K., Skippen, P., Chang, W. J., ... & Schabrun, S. M. (2022). The influence of sensory potentials on transcranial magnetic stimulation–Electroencephalography recordings. Clinical Neurophysiology, 140, 98-109.

Rocchi, L., Di Santo, A., Brown, K., Ibánez, J., Casula, E., Rawji, V., ... & Rothwell, J. (2021). Disentangling EEG responses to TMS due to cortical and peripheral activations. Brain stimulation, 14(1), 4-18.

It is not so common to use active electrodes for TMS-EEG. Please confirm the electrodes used and if they are c-ring TMS compatible and provide reference if otherwise (or actual papers recommending active ones)

To be more specific about the electrode type we have indicated:

“Signals were recorded from 63 TMS-compatible active electrodes (6mm height, 13mm width), embedded in an elastic cap (ActiCap, Brain Products, Germany), in line with the international 10-10 system”

A paper directly comparing TEPs between active and passive electrodes found no difference between the two and concluded TEPs can be reliably obtained using active electrodes (Mancuso et al., 2021). There is also evidence that active electrodes have better signal quality than passive electrodes at higher impedance levels (Laszlo et al., 2014).

This information has now been added to the paper:

“Active electrodes result in similar TEPs (both magnitude and peaks) to more commonly used passive electrodes (Mancuso et al., 2021). There is also evidence that active electrodes have higher signal quality than passive electrodes at higher impedance levels (Laszlo, Ruiz-Blondet, Khalifian, Chu, & Jin, 2014).”

There is a growing literature showing that monophonic pulses are not reliable for TEPs when compared to biphasic ones, please provide references. https://doi.org/10.1016/j.brs.2023.02.009

The reference provided by the reviewer states that biphasic and monophasic pulses both have advantages and disadvantages, rather than stating “monophonic pulses are not reliable for TEPs”. While there is some evidence that the artefacts resulting from monophasic pulses are larger than biphasic pulses, the EEG signal still returns to baseline levels within 5ms of the TMS pulse (Rogasch et al., 2013). Moreover, one paper (Casula et al. 2018) found that the resultant TEPs evoked by monophasic pulses are larger than those resulting from biphasic pulses. The authors postulated that monophasic pulses are more effective at activating widespread cortical areas than biphasic pulses. Ultimately the reference provided by the reviewer concludes that “effect of pulse shape on TEPs has not been systematically investigated and more studies are needed”.

Rogasch, N. C., Thomson, R. H., Daskalakis, Z. J., & Fitzgerald, P. B. (2013). Short-latency artifacts associated with concurrent TMS–EEG. Brain stimulation, 6(6), 868-876.

Casula, E. P., Rocchi, L., Hannah, R., & Rothwell, J. C. (2018). Effects of pulse width, waveform and current direction in the cortex: A combined cTMS-EEG study. Brain stimulation, 11(5), 1063-1070.

In most heads, a pulse in the PA direction is not obtained by a coil oriented 45o to the midline. The later induced later-medial pulses, good to obtain MEPs

We followed previous studies measuring MEPs from the ECRB elbow muscle (Schabrun et al., 2016; de Martino et al., 2019) whereby the TMS coil handle was angled at 45 degrees relative to the midline in order to induce a posterior-anterior current. We are not aware of literature that shows that the 45 degrees orientation does not induce a posterior anterior current in most heads.

Schabrun, S. M., Christensen, S. W., Mrachacz-Kersting, N., & Graven-Nielsen, T. (2016). Motor cortex reorganization and impaired function in the transition to sustained muscle pain. Cerebral Cortex, 26(5), 1878-1890.

De Martino, E., Seminowicz, D. A., Schabrun, S. M., Petrini, L., & Graven-Nielsen, T. (2019). High frequency repetitive transcranial magnetic stimulation to the left dorsolateral prefrontal cortex modulates sensorimotor cortex function in the transition to sustained muscle pain. Neuroimage, 186, 93-102.

The definition of RMT is (very) unusual. RMT provides small 50microV MEPs in 50% of times. If you obtain MEPs at 50microV you are supra threshold!

The TMS motor threshold assessment tool calculates threshold in the same manner as other threshold tools – it calculates the intensity that elicits an MEP of 50 microvolts, 50% of the time. We have made this clearer:

“The RMT was determined using the TMS motor thresholding assessment tool, which estimates the TMS intensity required to induce an MEP of 50 microvolts with a 50% probability using maximum likelihood parametric estimation by sequential testing (Awiszus and Borckardt, 2011). This method has been shown to achieve the accuracy of methods such as the Rossini-Rothwell method (Rossini et al., 1994; Rothwell et al., 1999) but with fewer pulses (Qi et al., 2011; Silbert et al., 2013).”

Please inform the inter TMS pulse interval used of TEPs and whether they were randomly generated.

The pulses were delivered manually – the interval was not randomly generated – as stated:

“As TMS was delivered manually, there was no set interpulse interval. However, the 40 second stimulus duration allowed for 11 pulses for each heat stimulus …. (~ 4 seconds in between …)”

Why have you stimulated suprathreshold on M1 when assessing TEP´s? The whole idea is that large TEPs can be obtained at lower intensities below real RMT and that prevents re-entering loops of somatosensory and joint movement inputs that insert "noise" to the TEPs.

The suprathreshold intensity was used to concurrently measure MEPs during pre-pain, pain and post-pain blocks.

We have made this clearer:

“The test stimulus intensity was set at 110% RMT to concurrently measure MEPs and TEPs during pre-pain, pain and post-pain blocks.”

The influence of re-afferent muscle activity was controlled for in Experiment 3.

Did you assess pain intensity after each of the TEP pulses? Please discuss how such a cognitive task may have influenced results

Pain intensity was assessed after each TMS pulse, as stated:

“TMS pulses were manually delivered, with a verbal pain rating score (0 = no pain, and 10 = most pain imaginable) obtained between pulses”

Reviewer 3 also brought up a concern of whether the verbal rating task might have influenced the TEPs. However, it is unlikely that the task contaminated the TEP response as the subsequent TMS pulse was not delivered until the verbal rating was complete and given that each participant was cued by the experimenter to provide the pain rating after each pulse (rather than the participant giving the rating at any time). We have made this clearer where we state:

“To avoid contamination of TEPs by verbal ratings, the subsequent TMS pulse was not delivered until the verbal rating was complete, and the participant was cued by the experimenter to provide the pain rating after each pulse”

The QST approach is unusual. Please confirm the sequence of CDT, WDT and HPT were not randomised and that no interval beyond 6sec were used. Proper references are welcome.

In line with a previous study (Furman et al., 2020), the sequence of the CPT, WDT and HPT were not randomized, and the interval was not more than 6 seconds.

We have made this clearer:

“A total of three trials was conducted for each test to obtain an average, with an interstimulus interval of six seconds. The sequence of cold, warmth and pain threshold was the same for all participants (Furman et al. 2020)”

Performing 60 pulses for TEPs is unusual, and against the minimum number in recommendations

Please explain and comment.https://doi.org/10.1016/j.brs.2023.02.009

Please refer to our previous response to this concern in the public reviews.

Line 578: when you refer to "heat" the reader may confound warm/heat with heat meaning suprathreshold. Please revise the wording.

We have now replaced the word heat stimulus with thermal stimulus.

Why were Bayesian statistics used instead as frequentist ones?

We have made this clearer:

“Given we were interested in determining the evidence for pain altering TEP peaks in certain conditions (e.g., active TMS) and pain not altering TEP peaks in other conditions (sham TMS), we used a Bayesian approach as opposed to a frequentist approach, which considers the strength of the evidence for the alternative vs. null hypothesis”

RESULTS

There is a huge response with high power after 100ms- Please discuss if you believe auditory potentials may have influenced it.

It is indeed possible that auditory potentials were present at 100ms. We now state:

“Indeed, the signal at ~100ms post-TMS from Experiment 1 may reflect an auditory N100 response”

The presence of auditory contamination does not impact the main conclusions of the paper given this was controlled for in Experiment 2.

Please discuss how pain ranging from 3-10 may have influenced results in the "PAIN" situation,

It is anticipated that the fixed thermal stimulus intensity approach would lead to large variations in pain ratings (Adamczyk et al., 2022). This is a recommended approach when the aim of the research is to determine relationships between neurophysiological measures and individual differences in pain sensitivity (Adamczyk et al., 2022). Indeed, we were interested in whether alterations in neurophysiological measures were associated with pain intensity, and we found that higher pain ratings were associated with smaller reductions in MEP amplitude and larger increases in N45 amplitude.

Adamczyk, W. M., Szikszay, T. M., Nahman-Averbuch, H., Skalski, J., Nastaj, J., Gouverneur, P., & Luedtke, K. (2022). To calibrate or not to calibrate? A methodological dilemma in experimental pain research. The Journal of Pain, 23(11), 1823-1832.

Please indicate if any participants offered pain after warm stimulation ( possible given secondary hyperalgesia after so many plateaux of heat stimulation).

As stated in the results “All participants reported 0/10 pain during the pre-pain and post-pain blocks”.

Please discuss the potential effects of having around 10% of "bad channels) In average per experiment per participants, its impacts in source localisation and in TEP measurement. Same for >5 epochs excluded by participant.

The number of bad channels has been incorrectly stated by the reviewer as being 10% on average per experiment per participant, whereas the correct number of reported bad channels was 3%, 4.7% and 9.8% for Experiment 1, 2 and 3 respectively (see supplementary material). These numbers are below the accepted number of bad channels to interpolate (10%) in EEG pipelines (e.g., Debnath et al., 2020; Kayhan et al., 2022), so it is unlikely that our channel exclusions significantly influenced the quality of our source localization an TEP data.

Debnath, R., Buzzell, G. A., Morales, S., Bowers, M. E., Leach, S. C., & Fox, N. A. (2020). The Maryland analysis of developmental EEG (MADE) pipeline. Psychophysiology, 57(6), e13580.

Kayhan, E., Matthes, D., Haresign, I. M., Bánki, A., Michel, C., Langeloh, M., ... & Hoehl, S. (2022). DEEP: A dual EEG pipeline for developmental hyperscanning studies. Developmental cognitive neuroscience, 54, 101104.

The number of excluded epochs is unlikely to have influenced the results given there was evidence for no difference in the number of rejected epochs between conditions (E1 BF10 = 0.145, E2 BF10 = 0.27, E3 BF10 = 0.169 – these BFs have now been reported in the supplementary material), and given the reliability of the N45 was high (see response to previous comment on the number of trials per condition).

HPT of 42.9 {plus minus} 2.5{degree sign}C means many participants had HPT close to 46oC. Please discuss

While some participants did indeed have pain thresholds close to 46 degrees, they nonetheless reported pain during the test blocks. While such participants may have reported less pain compared to others, we aimed for larger variations in pain ratings, given one of the research questions was to determine why pain intensity differs between individuals (given the same noxious stimulus). Indeed, we showed that this variation was meaningful (pain intensity was related to alterations in N45 and MEP amplitude).

Please explain the sentence : line 139 "As such, if we had used an alternative design with blocks of warm stimuli intermixed with blocks of painful stimuli, the warm stimuli blocks would not serve as a valid non-painful baseline." I cannot see why.

Please refer to our previous point on why the fixed sequence was included.

And on the top of that heat was not individualised according to HPT.

Please refer to our previous point on why we used a fixed stimulus approach.

Sequences of warm/heat were not randomised. Please refer to our previous point on the why the sequence of blocks was not randomized.

Line 197: "However, as this is the first study investigating the effects of experimental pain on TEPsamplitude, there were no a priori regions or timepoints of interest to compare betweenconditions". This is not clear. It means you have not measured the activity (size of the N45) under the electrode closest to the TMS coil? The TEP is supposed to by higher under the stimulated target/respective corresponding electrode…

We are not aware of any current recommendations that state that the region of interest should be based on the site of stimulation. The advantage of TMS-EEG is that it allows characterisation of cortical excitability changes throughout the brain, not just the site of stimulation. We based our region of interest on a cluster-based permutation analysis, as recommended by Frömer, Maier, & Abdel Rahman, (2018)

Frömer, R., Maier, M., & Abdel Rahman, R. (2018). Group-level EEG-processing pipeline for flexible single trial-based analyses including linear mixed models. Frontiers in neuroscience, 12, 48.

Please explain where N45 values came from.

The N45 was calculated using the TESA peak function (Rogasch et al., 2017) which identifies a data point which is larger/smaller than +/- 5 data points within a specified time window (e,g, 40-70ms post-TMS as in the present study). Where multiple peaks are found, the amplitude of the largest peak is returned. Where no peak is found, the amplitude at the specified latency is returned.

Rogasch, N. C., Sullivan, C., Thomson, R. H., Rose, N. S., Bailey, N. W., Fitzgerald, P. B., ... & Hernandez-Pavon, J. C. (2017). Analysing concurrent transcranial magnetic stimulation and electroencephalographic data: A review and introduction to the open-source TESA software. Neuroimage, 147, 934-951.

If only the cluster assessment was made please provide the comparison between P45 from the target TMS channel location in pre pain vs pain.

We assume the reviewer is referring to the N45 rather than P45, and that by “target” TMS channel they are referring to the stimulated region.

We first clarify that there is no “target” channel given the motor hotspot differs between individuals and so the channel that is closest to the site of stimulation will always differ.

Secondly, as stated above, we are not aware of any current recommendations in TMS-EEG research that states that the region of interest for TEP analysis should be based on the site of stimulation. The advantage of TMS-EEG is that it allows characterisation of cortical excitability throughout the brain, not just the site of stimulation. If we based our ROI on the target channel only, we would lose valuable information about excitability changes occurring in other brain regions.

Lastly, the N45 was localized at frontocentral electrodes, which is also where the cluster differences emerged. As such, we do not believe it would be informative to compare N45 peak amplitude at the region of stimulation.

Also explain how correction for multiple comparisons was made

Please refer to our response to the public review related to this issue.

And report data from pain vs post-pain.

The pain vs. post-pain comparisons are now reported in the Supplementary material.

There is a strong possibility the response at N85 is an auditory /muscle signal. Please provide the location of this response.

We have opted not to include the topography at 85ms in the main paper as it would introduce too much clutter into the figures (which are already very dense), and because the topography was very similar to the topography at 100ms. As an example, for the reviewer, in Author response image 1 we have shown the topography for the pre-pain condition of Experiment 1.

Author response image 1.

Experiment 2: I have a strong impression both active TEPs and sham TEPs were contaminated by auditory (and muscle) noise. Please explain.

While it possible that auditory noise may have influenced TEPs in the active and sham groups, it does not impact the main conclusions of the paper, given that the purpose of the sham condition was to control for auditory and somatosensory stimulation resulting from TMS.

While muscle activity may also affect have influenced the TEPs in active and sham conditions, we used fastICA in all conditions to suppress muscle activity. The fastICA algorithm (Rogasch et al., 2017) runs an independent component analysis on the data, and classifies components as neural, TMS-evoked muscle, eye movements and electrode noise, based on a set of heuristic thresholding rules (e.g., amplitude, frequency and topography of the components). Components classified as TMS-evoked muscle/other muscle artefacts are then removed. In the supplementary material, we further report that the number of components removed did not differ between conditions, suggesting the impact of muscle artefacts are not larger in some conditions vs. others.

Rogasch, N. C., Sullivan, C., Thomson, R. H., Rose, N. S., Bailey, N. W., Fitzgerald, P. B., ... & Hernandez-Pavon, J. C. (2017). Analysing concurrent transcranial magnetic stimulation and electroencephalographic data: A review and introduction to the open-source TESA software. Neuroimage, 147, 934-951.

Experiment 3: One interpretation can be that both supra and sub-threshold TMS were leading to somatosensory re-afferent responses, based on the way RMT was calculated, which hyper estimate the RMT and delivers in reality 2 types of supra-threshold stimulations. Please discuss

Please refer to our response to the public review related to this issue.

Please provide correlation between N45 size and MEPs amplitudes.

This has now been included:

“There was no conclusive evidence of any relationship between alterations in MEP amplitude during pain, and alterations in N100, N45 and P60 amplitude during pain (see supplementary material).”<br /> The supporting statistics for these analyses have been included in the supplementary material.

DISCUSSION

Line 303: " The present study determined whether acute experimental pain induces alterations in cortical inhibitory and/or facilitatory activity observed in TMS-evoked potentials".

Well, no. The study assessed the N45, and was based on it. It did not really explore other metrics in a systematic fashion. P60 and N100 changes were not replicated in experiments 2 and 3..

We assume the reviewer is stating that we did not assess other TEP peaks (such as the N15, P30 and P180). However, we did indeed assess these peaks in a systematic fashion. First, we identified the ROI by using a cluster-based analysis. This is a recommended approach when the ROI is unclear (Frömer, Maier, & Abdel Rahman, 2018). We then analysed the TEP representing the mean voltage across the electrodes within the cluster, and then identified any differences in all peaks between conditions (not just the N45). This has been made clearer in the manuscript.

This has now been included:

“For all experiments, the mean TEP waveform of any identified clusters from Experiment 1 were plotted, and peaks (e.g., N15, P30, N45, P60, N100) were identified using the TESA peak function (Rogasch et al., 2017)”

Frömer, R., Maier, M., & Abdel Rahman, R. (2018). Group-level EEG-processing pipeline for flexible single trial-based analyses including linear mixed models. Frontiers in neuroscience, 12, 48.

And the N45 is not related to facilitatory or inhibitory activity, it is a measure of an evoked response indicating excitability

Evidence suggests the N45 is mediated by GABAAergic neurotransmission (inhibitory activity), as drugs which increase GABAA receptor activity increase the amplitude of the N45 (Premoli et al., 2014) and drugs which decrease GABAA receptor activity decrease the amplitude of the N45 (Darmani et al., 2016). As such, we and various other empirical papers (e.g., Bellardinelli et al., 2021; Noda et al., 2021; Opie at 2019 ) and review papers (Farzan & Bortoletto, 2022; Tremblay et al., 2019) have interpreted changes in the N45 peak as reflecting changes in cortical inhibitory/GABAA mediated activity.

Premoli, I., Castellanos, N., Rivolta, D., Belardinelli, P., Bajo, R., Zipser, C., ... & Ziemann, U. (2014). TMS-EEG signatures of GABAergic neurotransmission in the human cortex. Journal of Neuroscience, 34(16), 5603-5612.

Belardinelli, P., König, F., Liang, C., Premoli, I., Desideri, D., Müller-Dahlhaus, F., ... & Ziemann, U. (2021). TMS-EEG signatures of glutamatergic neurotransmission in human cortex. Scientific reports, 11(1), 8159.

Darmani, G., Zipser, C. M., Böhmer, G. M., Deschet, K., Müller-Dahlhaus, F., Belardinelli, P., ... & Ziemann, U. (2016). Effects of the selective α5-GABAAR antagonist S44819 on excitability in the human brain: a TMS–EMG and TMS–EEG phase I study. Journal of Neuroscience, 36(49), 12312-12320.

Noda, Y., Barr, M. S., Zomorrodi, R., Cash, R. F., Lioumis, P., Chen, R., ... & Blumberger, D. M. (2021). Single-pulse transcranial magnetic stimulation-evoked potential amplitudes and latencies in the motor and dorsolateral prefrontal cortex among young, older healthy participants, and schizophrenia patients. Journal of Personalized Medicine, 11(1), 54.

Farzan, F., & Bortoletto, M. (2022). Identification and verification of a'true'TMS evoked potential in TMS-EEG. Journal of neuroscience methods, 378, 109651.

Opie, G. M., Foo, N., Killington, M., Ridding, M. C., & Semmler, J. G. (2019). Transcranial magnetic stimulation-electroencephalography measures of cortical neuroplasticity are altered after mild traumatic brain injury. Journal of Neurotrauma, 36(19), 2774-2784.

Tremblay, S., Rogasch, N. C., Premoli, I., Blumberger, D. M., Casarotto, S., Chen, R., ... & Daskalakis, Z. J. (2019). Clinical utility and prospective of TMS–EEG. Clinical Neurophysiology, 130(5), 802-844.

Line 321: why have you not measured SEPs in experiment 3?

It is not possible to directly measure the somatosensory evoked potentials resulting from a TMS pulse, given that the TMS pulse produces a range of signals including cortical activity, muscle/eye blink responses, auditory responses, somatosensory responses and other artefacts. While some researchers attempt to isolate the SEP from TMS using pre-processing methods such as ICA, others use control conditions such as sensory sham conditions (to control for the “tapping” artefact) or subthreshold intensity conditions (to control for reafferent muscle activity), as we have done in Experiment 2 and 3 of our study.

We have now stated this in the manuscript:

“As it is extremely challenging to isolate and filter these auditory and somatosensory evoked potentials using pre-processing pipelines, masking methods have been used to suppress these sensory inputs, (Ilmoniemi and Kičić, 2010; Massimini et al., 2005). However recent studies have shown that even when these methods are used, sensory contamination of TEPs is still present, as shown by commonalities in the signal between active and sensory sham conditions that mimic the auditory/somatosensory aspects of real TMS (Biabani et al., 2019; Conde et al., 2019; Rocchi et al., 2021). This has led many leading authors (Biabani et al., 2019; Conde et al., 2019) to recommend the use of sham conditions to control for sensory contamination”

Line 365: SICI is dependent on GABAa activity. But the way the text is written if conveys the idea that TMS pulses "activate" GABA receptors, which is weird...Please rephrase.

This has now been reworded.

“SICI refers to the reduction in MEP amplitude to a TMS pulse that is preceded 1-5ms by a subthreshold pulse, with this reduction believed to be mediated by GABAA neurotransmission (Chowdhury et al., 2022)”

Reviewer #3 (Recommendations For The Authors):

-Key references Ye et al., 2022 and Che et al., 2019 need to be included in the reference list.

These references have now been included in the reference list.

-Heat pain stimuli and TMS stimuli are applied simultaneously. Sometimes the term "stimulus" is used without specifying whether it refers to TMS pulses or heat pain stimuli. Clarifying this whenever the word "stimulus" is used would enhance clarity for the reader.

We have now clarified the use of the word “stimulus” throughout the paper.

-Panels A-D in Figure 6 should be correctly labeled in the text and the figure legend.

Figure 6 Panel labels have now been amended.

eLife assessment

This valuable study provides convincing evidence that acute experimental pain induces changes of cortical excitability. Although the modality specificity of the findings is not fully clear, the findings will be of interest to researchers interested in the brain mechanisms of pain.

Reviewer #1 (Public Review):

The objective of this investigation was to determine whether experimental pain could induce alterations in cortical inhibitory / facilitatory activity observed in TMS-evoked potentials (TEPs). Previous TMS investigations of pain perception had focused on motor evoked potentials (MEPs), which reflect a combination of cortical, spinal, and peripheral activity, as well as restricting the focus to M1. The main strength of this investigation is the combined use of TMS and EEG in the context of experimental pain. More specifically, Experiment 1 investigated whether acute pain altered cortical excitability, reflected in the modulation of TEPs. The main outcome of this study is that relative to non-painful warm stimuli, painful thermal stimuli led to an increase on the amplitude of the TEP N45, with a larger increase associated with higher pain ratings. Because it has been argued that a significant portion of TEPs could reflect auditory potentials elicited by the sound (click) of the TMS, Experiment 2 constituted a control study that aimed to disentangle the cortical response related to TMS and auditory activity. Finally, Experiment 3 aimed to disentangle the cortical response to TMS and reafferent feedback from muscular activity elicited by suprathreshold TMS applied over M1. The fact that the authors accompanied their main experiment with two control experiments strengthens the conclusion that the N45 TEP peak could be implicated in the perception of painful stimuli. Perhaps, the addition of a highly salient but non-painful stimulus (i.e. from another modality) would have further ruled out that the effects on the N45 are not predominantly related to intensity / saliency of the stimulus rather than to pain per se.

Reviewer #3 (Public Review):

The present study aims to investigate whether pain influences cortical excitability. To this end, heat pain stimuli are applied to healthy human participants. Simultaneously, TMS pulses are applied to M1 and TMS-evoked potentials (TEPs) and pain ratings are assessed after each TMS pulse. TEPs are used as measures of cortical excitability. The results show that TEP amplitudes at 45 msec (N45) after TMS pulses are higher during painful stimulation than during non-painful warm stimulation. Control experiments indicate that auditory, somatosensory, or proprioceptive effects cannot explain this effect. Considering that the N45 might reflect GABAergic activity, the results suggest that pain changes GABAergic activity. The authors conclude that TEP indices of GABAergic transmission might be useful as biomarkers of pain sensitivity.

Pain-induced cortical excitability changes is an interesting, timely, and potentially clinically relevant topic. The paradigm and the analysis are sound, the results are convincing, and the interpretation is adequate. The findings will be of interest to researchers interested in the brain mechanisms of pain.

eLife assessment

This study presents valuable findings on the presence of 6mA in the Drosophila genome and challenges previous findings regarding the role of TET in 6mA modification. The evidence supporting the claims is solid, and the paper has the potential to stimulate re-evaluations of the significance and regulatory mechanisms of 6mA DNA modifications in Drosophila.

Reviewer #1 (Public Review):

This work challenges previously published results regarding the presence and abundance of 6mA in the Drosophila genome, as well as the claim that the TET or DMAD enzyme serves as the "eraser" of this DNA methylation mark and its roles in development. This information is needed to clarify these questions in the field. I am less familiar with the biochemical approaches in this work, so my comments are mainly on the genetic analyses. Generally speaking, the methods for fly husbandry and treatment seem to be in accordance with those established in the field.

Reviewer #2 (Public Review):

DNA adenine methylation (6mA) is a rediscovered modification that has been described in a wide range of eukaryotes. However, 6mA presence in eukaryote remains controversial due to the low abundance of its modification in eukaryotic genome. In this manuscript, Boulet et al. re-investigate 6mA presence in drosophila using axenic or conventional fly to avoid contaminants from feeding bacteria. By using these flies, they find that 6mA is rare but present in the drosophila genome by performing LC/MS/MS. They also find that the loss of TET (also known as DMAD) does not impact 6mA levels in drosophila, contrary to previous studies. In addition, the authors find that TET is required for fly development in its enzymatic activity-independent manner.

The strength of this study is, that compared to previous studies of 6mA in drosophila, the authors employed axenic or conventional fly for 6mA analysis. These fly strains make it possible to analyze 6mA presence in drosophila without bacterial contaminant. Therefore, showing data of 6mA abundance in drosophila by performing LC-MS/MS in this manuscript is more convincing as compared with previous studies. Intriguingly, the authors find that the conserved iron-binding motif required for the catalytic activity of TET is dispensable for its function. This finding could be important to reveal TET function in organisms whose genomic 5mC levels are very low.

The manuscript in this paper is well written but some aspects of data analysis and discussion need to be clarified and extended.<br /> 1) It is convincing that an increase in 6mA levels is not observed in TETnull presented in Fig1. But it seems 6mA levels are altered in Ax.TET1/2 compared with Ax.TETwt and Ax.TETnull presented in Fig1f (and also WT vs TET1/2 presented in Fig1g). Is it sure that no statistically significant were not observed between Ax.TET1/2 and Ax.TETwt?<br /> 2) The representing data of in vitro demethylation assay presented in Fig.3 is convincing, but it is not well discussed and analyzed why these results are contrary to previous reports (Yao et al., 2018 and Zhang et al., 2015).

eLife assessment

This fundamental study advances our understanding of how Notch signaling activates transcription by analyzing the dynamics of the Mastermind transcriptional co-activator and its role in the activation complex. The evidence is compelling with precise quantitative measurements. The evidence presented by the authors features methods, data, and analyses that are currently state-of-the-art but have not previously been applied to how transcription is regulated in the Notch pathway.

Reviewer #1 (Public Review):

In this manuscript by DeHaro-Arbona et al., the authors wish to understand how a signaling pathway (Notch) is dynamically decoded to elicit a specific transcriptional output. In particular, they investigate the kinetic properties of Notch-responsive nuclear complexes (the DNA binding factor CSL and its co-activator Mastermind (mam) along with several candidate interacting partners). Their experimental model is the polytene chromosome of the Drosophila salivary gland, in which the naturally inactive Notch can be artificially induced through the expression of a constitutively active form of Notch.

The authors develop a series of CRISPR and transgenic lines enabling the live imaging of these complexes at a specific locus and in various backgrounds (genetic perturbations/drug treatments). This quantitative live imaging data suggests that Notch nuclear complexes form hubs and the authors characterize their binding dynamics. Interestingly, they elegantly demonstrate that the content of these hubs and their kinetic properties can evolve, even within Notch ON cells. Hence, they propose the existence of distinct hubs, distinguishing an open (CSL), engaged (CSK-Mam), or active (CSL-Mam-Med-PolII) configuration in Notch ON cells and an inactive hub (in Notch OFF having previously been exposed to Notch) state, that would explain the surprising transcriptional memory that the authors observe hours after Notch withdrawal.

Reviewer #2 (Public Review):

The manuscript from deHaro-Arbona et al, entitled "Dynamic modes of Notch transcription hubs conferring memory and stochastic activation revealed by live imaging the co-activator Mastermind", uses single molecule microscopy imaging in live tissues to understand the dynamics and molecular determinants of transcription factor recruitment to the E(spl)-C locus in Drosophila salivary gland cells under Notch-ON and -OFF conditions. Previous studies have identified the major players that are involved in transcription regulation in the Notch pathway, as well as the importance of general transcriptional coregulators, such as CBP/P300 and the Mediator CDK module, but the detailed steps and dynamics involved in these processes are poorly defined. The authors present a wealth of single molecule data that provides significant insights into Notch pathway activation, including:

1. Activation complexes, containing CSL and Mam, have slower dynamics than the repressor complexes, containing CSL and Hairless.

2. Contribution of CSL, NICD, and Mam IDRs to recruitment.

3. CSL-Mam slow-diffusing complexes are recruited and form a hub of high protein concentrations around the target locus in Notch-ON conditions.

4. Mam recruitment is not dependent on transcription initiation or RNA production.

5. CBP/P300 or its associated HAT activity is not required for Mam recruitment.

6. Mediator CDK module and CDK8 activity are required for Mam recruitment, and vice-versa, but not CSL recruitment.

7. Mam is not required for chromatin accessibility but is dependent on CSL and NICD.

8. CSL recruitment and increased chromatin accessibility persist after NICD removal and loss of Mam, which confers a memory state that enables rapid re-activation in response to subsequent Notch activation.

9. Differences in the proportions of nuclei with both Pol II and with Mam enrichment, which results in transcription being probabilistic/stochastic. These data demonstrate that the presence of Mam-complexes is not sufficient to drive all the steps required for transcription in every Notch-ON nucleus.

10. The switch from more stochastic to robust transcription initiation was elicited when ecdysone was added.

Overall, the manuscript is well written, concise, and clear, and makes significant contributions to the Notch field, which are also important for a general understanding of transcription factor regulation and behavior in the nucleus. I recommend that the authors address my relatively minor criticisms detailed below.

Page 7, bottom. The authors speculate, "It is possible therefore that, once recruited, Mam can be retained at target loci independently of CSL by interactions with other factors so that it resides for longer." Is it possible that another interpretation of that data is that Mam is a limiting factor?

Page 9. The authors write, "A very low level of enrichment was evident for... for the CSL C-terminus..". The recruitment of CSL ct IDR does not appear to be statistically significant or there is no apparent difference (Figure S2C), suggesting the CSL ct IDR does not play a role in enrichment.

Page 9. The authors write, "Notably, MamnIDR::GFP fusion was present in droplets, suggesting it can self-associate when present in a high local concentration (Figure S2B)." Is this result only valid for Mam nIDR or does full-length Mam also localize into droplets, as has been previously observed for full-length mammalian Maml1 in transfected cells?

Previous studies in mammalian cells suggest that Maml1 is a high-confidence target for phosphorylation by CDK8, see Poss et al 2016 Cell Reports https://doi.org/10.1016/j.celrep.2016.03.030. By sequence comparison, does fly Mam have similar potential phosphorylation sites, and might these be critical for Mam/CDK module recruitment?

Page 11: The authors write, "The differences in the effects on Mam and CSL imply that the CDK module is specifically involved in retaining Mam in the hub, and that in its absence other CSL complexes "win-out", either because the altered conditions favour them and/or because they are the more abundant." Are the "other" complexes the authors are referring to Hairless-containing complexes? With the reagents the authors have in hand couldn't this be explicitly shown for CSL-complexes rather than speculated upon?

Page 12/13: The authors write, "Based on these results we propose that, after Notch activity decays, the locus remains accessible because when Mam-containing complexes are lost they are replaced by other CSL complexes (e.g. co-repressor complexes)." Again, why not actually test this hypothesis rather than speculate? The dynamics of Hairless complexes following the removal of Notch would be very interesting and build upon previously published results from the Bray lab.

Page 13: The authors write, "As Notch removal leads to a loss of Mam, but not CSL, from the hub, it should recapitulate the effects of MamDN." While the data in Figure 5B seem to support this hypothesis, it's not clear to me that the loss of Mam and MamDN should phenocopy each other, bc in the case of MamDN, NICD would still be present.

The temporal dynamics for Mam recruitment using the temperature- and optogenetic-paradigms are quite different. For example, in the optogenetic time course experiments, the preactivated cells are in the dark for 4 hours, while in the temperature-controlled experiments, there is still considerable enrichment of Mam at 4 hours. For the preactivated optogenetic experiments, how sure are the authors that Mam is completely gone from the locus, and alternatively, can the optogenetic experimental results be replicated in the temperature-controlled assays? My concern is whether the putative "memory" observation is just due to incomplete Mam removal from the previous activation event.

Reviewer #3 (Public Review):

Summary:<br /> DeHaro-Arbona and colleagues investigate the in vivo dynamics of Notch-dependent transcriptional activation with a focus on the role of the Mastermind (MAM) transcriptional co-activator. They use GFP and HALO-tagged versions of the CSL DNA-binding protein and MAM to visualize the complex, and Int/ParB to visualize the site of Notch-dependent E(Spl)-C transcription. They make several conclusions. First, MAM accumulates at E(Spl)-C when Notch signaling is active, just like CSL. Second, MAM recruits the CDK module of Mediator but does not initiate chromatin accessibility. Third, after signaling is turned off, MAM leaves the site quickly but CSL and chromatin accessibility are retained. Fourth, RNA pol II recruitment, Mediator recruitment, and active transcription were similar and stochastic. Fifth, ecdysone enhances the probability of transcriptional initiation.

Strengths:<br /> The conclusions are well supported by multiple lines of extensive data that are carefully executed and controlled. A major strength is the strategic combination of Drosophila genetics, imaging, and quantitative analyses to conduct compelling and easily interpretable experiments. A second major strength is the focus on MAM to gain insights into the dynamics of transcriptional activation specifically.

Weaknesses:<br /> Weaknesses are minor. There were no p-values reported for data presented in Figure S1D and no indication of how variable measurements were. In addition, the discussion of stochasticity was not integrated optimally with relevant literature.

eLife assessment

This valuable work on the paleovegetation history of Iceland has implications for the field of paleoecology, and the deglaciation history of Iceland and additional localities in Northern America and Europe via woody shrub colonization. The study uses a sedimentary ancient DNA metabarcoding approach to study this historic process. The strength of evidence is solid, with the methods (analysis of sedimentary DNA) and data analyses broadly supporting the claims.

Joint Public Review:

The revised version of the manuscript "Delayed postglacial colonization of Betula in Iceland and the circum North Atlantic" by Harning et al. investigates the colonization of shrubs during the Late Pleistocene/Holocene in Northern America and Europe by comparing published sedimentary ancient DNA (sedaDNA) records (and pollen data) with a new sedaDNA record from Island. The manuscript aims to identify shrub colonization patterns, discusses their drivers and evaluates the importance of shrubification under future warming.

The revised version improved the clarity of methods and discussion and results presented are more convincing.

However, parts of the methods (e.g. assessment of blanks and data filtering) and results (e.g. visualization of plant community data) could still be polished, and the figures should be improved to increase the clarity of the manuscript.

eLife assessment

This study presents initial findings in the generation of 3D cell constructs from endometrial cell mixtures seeded in Matrigel scaffold and treated with hormones as a proof of concept. While the study findings are valuable, functional validation to demonstrate its robustness is lacking, and therefore the strength of evidence is incomplete. The term organoids might not be appropriate to describe this in vitro model.

Reviewer #1 (Public Review):

Summary:

This study generated 3D cell constructs from endometrial cell mixtures that were seeded in the Matrigel scaffold. The cell assemblies were treated with hormones to induce a "window of implantation" (WOI) state. Although many bioinformatic analyses point in this direction, there are major concerns that must be addressed.

Strengths:

The addition of 3 hormones to enhance the WOI state (although not clearly supported in comparison to the secretory state).

Weaknesses:

First of all, the term organoid must be discarded. The authors just seed the endometrial cell mixture which assembles and aggregates into a 3D structure which is then immediately used for analysis. Organoids grow from tissue stem cells and must be passage-able (see their own description in lines 69-71). So, the term organoid must be removed everywhere, to not confuse the organoid field. It is not shown that the whole 3D assembly is passageable, which would be very surprising given the fact that immune and stromal cells do not grow in Matrigel because of the unfavorable growing conditions (which are targeted to epithelial cell growth).

Second, the study remains fully descriptive, bombing the reader with a mass of bioinformatic analyses without clear descriptions and take-home messages. The paper is very dense, meaning readers may give up. Moreover, functional validation, except for morphological and immunostaining analyses (which are posed as "functional" but actually are only again expression) is missing, such as in vivo functionality (after transplantation e.g.) and embryo interaction. Importantly, the 3D structure misses the right architecture with a lining luminal epithelium which is present in the receptive endometrium in vivo and needed as the first contact site with the embryo. So, in contrast to what the authors claim, this is not the best model to study embryo interaction, or the closest model to the in vivo state (line 318, line 326).

Third, receptive endometrial organoids (assembloids; Rawlings et al., eLife 2021) and receptive organoid-derived "open-faced endometrial layer" (Kagawa et al., Nature 2022) have already been described, which is in contrast to what the authors claim in several places that "they are the first" (e.g. lines 87-88, 316-319, etc). These studies used real organoids to achieve their model (and even showed embryo interaction), while in the present study, different cell types are just seeded and assembled. Hence, logically, immune cells are present which are never found in real organoid models. The only original aspect in the present study is the use of hormones to enhance the WOI phenotype. However, crucial information on this original aspect is missing such as concentration of the hormones, refreshment schedule, all 3 hormones added together or separately, and all 3 required?

Moreover, it is not a "robust" model at all as the authors claim, given the variability of the initial cell mixture (varying from patient to patient). Actually, the reproducibility is not shown. The proportions of the different cell types seeded in the Matrigel droplet will be different with every endometrial biopsy. It would be much better to recombine epithelial (passageable) organoids with stromal and immune cells in a quantified, standardized manner to establish a "robust" model.

Reviewer #2 (Public Review):

A wide variety of assays are used to describe the new culture system and compare it both with those previously described and with the endometrial tissue itself. The three different cultures they used are control organoids (CTRL) cultured with described expansion media, secretory organoids (SEC, cultured with E2, MPA and cAMP inducing secretory phase as previously reported) and WOI organoids (cultured with E2, MPA, cAMP, prolactin (PRL), human chorionic gonadotropin (hCG) and human placental lactogen (hPL)). First, they performed morphological characterization of cultures using different antibodies, showing the presence of epithelial glandular cells and stromal cells, as well as their proliferation and absence of apoptosis. Glycogen secretion and progesterone receptor expression complete organoid characterization at the functional and hormone response levels respectively.

Then, they performed single-cell transcriptomics to analyse its composition in terms of cell type, comparing with different databases, but with an unknown "n". They detect stromal, epithelial, and immune cells (also by microscopy), and analyse gene expression and transcription regulation, showing similarities between WOI organoids and mid-secretory endometrium. With endometrial receptivity analysis, they suggest a successful formation of the implantation window in vitro, but this result is difficult to interpret.

Analyzing transcriptome and proteome information of WOI organoids, authors demonstrate a strong response to estrogen and progesterone, but some comparisons are made with CTRL and SEC, and others only with CTRL, which limits the power of some results. In the same way, some genes related to Cilia and pinopodes appear dominant in WOI organoids, but the comparison by electron microscopy is made only against CTRL organoids.

In subsequent analysis, WOI organoids showed a marked differentiation from proliferative to secretory epithelium, and from proliferative epithelium to EMT-derived stromal cells than SEC organoids. These statements are based on their upregulation of monocarboxylic acid and lipid metabolism, their enhanced peptide metabolism and mitochondrial energy metabolism, or their pseudotime trajectories. However, other analyses (such as the accumulation of secretory epithelium or decreased proliferative epithelium, the increased ciliated epithelium after hormonal treatment, or the presence of EMT-derived stromal cells) show only small differences between SEC and WOI organoids.

In summary, the development of an endometrial organoid culture methodology that allows targeting the endometrial situation in the window of implantation could change the experimental approaches of many studies, but more evidence is needed, and above all, more approaches on how different WOI organoids are from SEC organoids, to be sure if it is worth using them in implantation.

eLife assessment

The current manuscript presents a cryo-EM structure of a tripartite ATP-independent periplasmic (TRAP) transporter that contributes to Haemophilus influenzae virulence. Convincing biophysical and cryo-EM experiments yield a valuable molecular model, but evidence to support some of the mechanistic conclusions is currently incomplete.

Reviewer #1 (Public Review):

Summary:<br /> TRAP transporters are an unusual class of secondary active transporters that utilize periplasmic binding proteins to deliver their substrates. This paper contributes a new 3 Å structure of the Haemophilus influenzae TRAP transporter. The structure joins two other recent cryo-EM structures of TRAP transporters, including a lower-resolution structure of the same H. influenzae protein (overall 4.7 Å), and a ~3 Å structure of a homologue from P. profundum. In addition to reporting a higher resolution cryo-EM structure, the authors also recapitulate protein activity in a reconstituted system, investigate protein oligomerization using analytic ultracentrifugation, and evaluate interactions and function in "mix and match" configurations with periplasmic subunits from other homologues.

Strengths:<br /> The strength of the paper is that the better resolution cryo-EM data permits sidechain assignment, the identification of bound lipids, and the identification of sodium ions. It is important to get this structure out there since the resolution passes an important threshold for model-building accuracy. The current structure nicely explains a lot of prior mutagenesis data on the H. influenzae TRAP. This is also the first structure of a TRAP protein to be solved without a fiducial, although the overall structure is not very different from those solved with fiducials.

Weaknesses:<br /> The experiments examining the monomer/dimer equilibrium appear somewhat preliminary. The biological or mechanistic importance of oligomerization is not established, so these experiments are inherently of limited scope. Moreover, cryo-EM datasets exhibit both parallel and antiparallel dimers, the latter of which are clearly not biologically relevant. It is probably impossible to distinguish these in the AUC experiments, which makes interpretation of these experiments more difficult.

Similarly, the importance of the lipid binding sites observed in cryo-EM isn't experimentally established (for example by mutating the binding site) and it thus seems too preliminary to infer that they are important for function.

Reviewer #2 (Public Review):

Summary:<br /> In this manuscript, the membrane component of the sialic acid-specific TRAP transporter, SiaQM (HiSiaQM), from H. influenzae, is structurally characterized. TRAP transporters are substrate binding protein (SBP)-dependent secondary-active transporters, and HiSiaQM is the most comprehensively studied member of this family. While all previous work on fused TRAP transporter membrane proteins suggests that they are monomeric (including the previous structural characterization of HiSiaQM by a different group), a surprising finding from this work is the observation that HiSiaQM can form higher oligomers, consistent with it being a dimer. These higher oligomeric states were initially observed after extraction of the protein with LMNG detergent but were also observed in DDM detergent, amphipol and nanodiscs using analytical ultracentrifugation (AUC). Structural characterization of dimeric HiSiaQM revealed 2 arrangements, parallel and antiparallel arrangements, the latter of which is unlikely to be physiologically relevant.

The higher resolution of this new structure of HiSiaQM (2.2-2.7 Å compared to 4.7 Å for the previous structure) facilitated the assignment of bound lipids at the dimer interface and a lipid molecule embedded in each of the protomers; allowed for a clearer refinement of the Na+ and putative substrate binding sites, which differ slightly from the previous structure; and produced better-modelled side chains for the residues involved in the SBP:HiSiaQM interaction. The authors developed a useful AUC-based assay to determine the affinity for this interaction revealing an affinity of 65 µM. Finally, the authors make the very interesting observation that a sialic acid-specific SBP from a different TRAP transporter can utilize HiSiaQM for transport, contrary to previous observations, revealing for the first time that TRAP membrane components can recognize multiple SBPs.

Overall, this is a well-written and presented manuscript detailing some interesting new observations about this interesting protein family. One of the main findings, that the protein can form a dimer, is supported by data, but the physiological relevance of this is questionable, and the possibility that this is artefactual has not been ruled out. Conclusions regarding the mechanistic importance of the lipid-binding sites are not currently supported by the data.

Strengths:<br /> The main strength of this work is the increased resolution of HiSiaQM, which allows for a much more precise assignment of side chains and their orientation. This will be of importance for subsequent mechanistic studies on the contributions of these residues to Na+ and sialic acid binding and conformational changes.

The observation of the lipids, especially the lipid embedded near the fusion helix, is an intriguing observation, which lays the groundwork for future work to understand the lipid-dependence of these transporters. The development of the AUC-based approach to measure SBP affinity for the membrane component will likely prove useful to future studies.

Weaknesses:<br /> One of the main results from this work is the observation that HiSiaQM can form a dimer. Two arrangements were observed, parallel and antiparallel, the latter of which is almost certainly physiologically irrelevant as it would preclude essential interactions with the extracytoplasmic substrate-binding protein. As acknowledged by the author, this non-physiological arrangement is likely a consequence of protein preparation (overexpression, extraction, purification, etc.). However, if one dismisses the antiparallel arrangement as non-relevant and an artefact of protein preparation, it is difficult for the parallel arrangement to maintain its credibility, as it was also processed in the same way. This is especially true when one considers that there is only 100 Å2 buried surface area in the parallel arrangement that does not involve any sidechains; it is difficult to envisage this as a specific interaction, e.g. compared to related proteins that have ~2000 Å2 buried surface area. Unless this dimerization is observed in a bacterial membrane at physiological protein concentrations, it is difficult to rule out the possibility that the observed dimerization is merely an artefact caused by the expression, purification and concentration of the protein.

The manuscript contains some excellent structural analysis of this protein, whose higher resolution reveals some new and interesting insights. However, a weakness of the current work is a lack of validation of these observations using other approaches. For example, lipid interactions are observed in the structure that the authors claim is mechanistically important. However, without disrupting these interactions to look at the effect on transport, this conclusion is not supported. Similarly, the authors use their structure to predict residues that are important for the SBP:membrane protein interaction, and they develop an AUC-based binding assay to study this interaction, but they do not test their predictions using this approach.

Reviewer #3 (Public Review):

Summary:<br /> The manuscript reports new molecular characterization of the Haemophilus influenza tripartite ATP-independent periplasmic (TRAP) transporter of N-acetylneuraminate (Neu5Ac). This membrane transporter is important for the virulence of the pathogen. H. influenza lacks Neu5Ac biosynthetic pathway and utilizes the TRAP transporter to import it. Neu5Ac is used as a nutrient source but also as a protection from the human immune response. The transporter is composed of two fused membrane subunits, HiSiaQM, and one soluble, periplasmic subunit HiSiaP. HiSiaP, by binding to the substrate Neu5Ac, changes its conformation, allowing its binding to HiSiaQM, followed by Neu5Ac and Na+ transport to the cytoplasm. The combination of structural, biophysical and biochemical approaches provides a solid basis for describing the functioning of the Haemophilus influenza Neu5Ac TRAP transporter, which is essential for the pathogen virulence.

Strengths:<br /> The paper describes the electron microscopy structure of HiSiaQM, thanks to its solubilization in L-MNG followed by the exchange to amphipol or nanodisc. In these conditions, HiSiaQM consists of a mixture of monomers and dimers, as characterized by analytical ultracentrifugation. The cryo-EM analysis shows two types of dimers: one in an antiparallel configuration, which is artifactual, and a parallel one, which may be physiologically relevant. Cryo-EM on the dimers allows high-resolution (≈ 3 Å) structure determination. The structure is the first one of a fused SiaQM, and is the first obtained without megabody. The work highlights structural elements (fusion helix, lipids) that could modulate transport. The authors checked the functionality of the purified HiSiaQM, which, after reconstitution in liposome, displays a significantly larger Neu5Ac transport activity compared to the non-fused PpSiaQM homolog. The work identifies Na+ binding sites, and the putative Neu5Ac binding site. From analytical ultracentrifugation using fluorescently labelled HiSiaP, the authors show that HiSiaP is able to interact with HiSiaQM monomer and dimer, with a low but physiologically relevant affinity. HiSiaP interaction with HiSiaQM was modelled using AlphaFold2, and discussed in view of published activity on mutants, and new transport activity assays using SiaQM and SiaP from different organisms. In conclusion, the combination of structural, biophysical and biochemical approaches provides a solid basis for describing the functioning of this TRAP fused transporter.

Weakness:<br /> This work evidences in vitro a HiSiaQM dimer, whose in vivo relevance is not ascertained. However, the authors are very careful, not to over-interpret their data, and their conclusions regarding the transporter structure and function are valid irrespective of its state of association.

eLife assessment

This study provides the fundamental insight that TGN46, a single-pass membrane protein, acts as a cargo receptor for proteins at the Trans-Golgi Network that are destined for secretion. Compelling evidence shows that the luminal domain of TGN46 is crucial for the incorporation of the soluble secretory protein PAUF into CARTS. The clear effect but partial block of secretion after depletion of TGN46 points to the need for further exploration of the process.

Author Response

We thank the reviewers and the editorial team for their assessment and valuable feedback on our manuscript. Their supporting comments reinforce the significance of our findings.

Regarding the specific point raised about the partial effects observed in the TGN46 KO cell line, we acknowledge the importance of addressing this issue in more detail in the revised version of our manuscript. The partial effects observed when using the TGN46 KO cell line are likely caused by several factors:

1) It is important to consider the phenomenon of cell adaptation/compensation, which is documented to occur in gene knockout cell lines. Cells often respond to genetic perturbations by adapting to compensate the loss of a specific gene. These compensatory effects could potentially mitigate the full impact of TGN46 depletion and might explain the partial effects observed.

2) Our data indicate that the absence of TGN46 reduces PAUF secretion, but does not completely block its export. These results align with our proposed role TGN46 in cargo sorting. In its absence, the secretory proteins likely exit the TGN via alternative routes/mechanisms, such as "bulk flow" or by entering other transport carriers in an uncontrolled manner. The partial redistribution of the TGN46-∆lum mutant into VSVG carriers (Figure 4D) supports this likelihood. Importantly, similar situations are observed when unrelated sorting factors are depleted from the Golgi membranes. For example, when the cofilin/SPCA1/Cab45 sorting pathway is genetically disrupted, the secretion of this pathway's clients is inhibited but not completely halted (e.g., von Blume et al. Dev. Cell 2011; J. Cell Biol. 2012).

3) As suggested by the reviewers, it remains possible that TGN46 is not the sole player for cargo sorting. The existence of redundant or alternative mechanisms cannot be ruled out.

In our revised manuscript, we will provide a more in-depth discussion of these factors and their potential contributions to the observed partial effects in TGN46 KO cells. We believe that a comprehensive exploration of these possibilities will improve our understanding of the role(s) of TGN46 in cargo sorting and TGN export.

eLife assessment

This valuable study of Iceland's paleovegetation history has implications for the field of paleoecology, as well as shrubification in the Arctic. It presents solid evidence that postglacial colonisation by birch was later than willow in Iceland and nearby areas, based on a new analysis with multiple lines of existing evidence, including one new site with sedimentary ancient DNA. The study would benefit from a clearer description of key methods and results, and more critical reflection on the assessment of colonisation lags and the predictive use of paleo datasets.

Author Response

The following is the authors’ response to the original reviews.

Reviewer #1 (Public Review):

This work provides a new dataset of 71,688 images of different ape species across a variety of environmental and behavioral conditions, along with pose annotations per image. The authors demonstrate the value of their dataset by training pose estimation networks (HRNet-W48) on both their own dataset and other primate datasets (OpenMonkeyPose for monkeys, COCO for humans), ultimately showing that the model trained on their dataset had the best performance (performance measured by PCK and AUC). In addition to their ablation studies where they train pose estimation models with either specific species removed or a certain percentage of the images removed, they provide solid evidence that their large, specialized dataset is uniquely positioned to aid in the task of pose estimation for ape species.

The diversity and size of the dataset make it particularly useful, as it covers a wide range of ape species and poses, making it particularly suitable for training off-the-shelf pose estimation networks or for contributing to the training of a large foundational pose estimation model. In conjunction with new tools focused on extracting behavioral dynamics from pose, this dataset can be especially useful in understanding the basis of ape behaviors using pose.

We thank the reviewer for the kind comments.

Since the dataset provided is the first large, public dataset of its kind exclusively for ape species, more details should be provided on how the data were annotated, as well as summaries of the dataset statistics. In addition, the authors should provide the full list of hyperparameters for each model that was used for evaluation (e.g., mmpose config files, textual descriptions of augmentation/optimization parameters).

We have added more details on the annotation process and have included the list of instructions sent to the annotators. We have also included mmpose configs with the code provided. The following files include the relevant details:

File including the list of instructions sent to the annotators: OpenMonkeyWild Photograph Rubric.pdf

Mmpose configs:

i) TopDownOAPDataset.py

ii) animal_oap_dataset.py

iii) init.py

iv) hrnet_w48_oap_256x192_full.py

Anaconda environment files:

i) OpenApePose.yml

ii) requirements.txt

Overall this work is a terrific contribution to the field and is likely to have a significant impact on both computer vision and animal behavior.

Strengths:

• Open source dataset with excellent annotations on the format, as well as example code provided for working with it.

• Properties of the dataset are mostly well described.

• Comparison to pose estimation models trained on humans vs monkeys, finding that models trained on human data generalized better to apes than the ones trained on monkeys, in accordance with phylogenetic similarity. This provides evidence for an important consideration in the field: how well can we expect pose estimation models to generalize to new species when using data from closely or distantly related ones? - Sample efficiency experiments reflect an important property of pose estimation systems, which indicates how much data would be necessary to generate similar datasets in other species, as well as how much data may be required for fine-tuning these types of models (also characterized via ablation experiments where some species are left out).

• The sample efficiency experiments also reveal important insights about scaling properties of different model architectures, finding that HRNet saturates in performance improvements as a function of dataset size sooner than other architectures like CPMs (even though HRNets still perform better overall).

We thank the reviewer for the kind comments.

Weaknesses:

• More details on training hyperparameters used (preferably full config if trained via mmpose).

We have now included mmpose configs and anaconda environment files that allow researchers to use the dataset with specific versions of mmpose and other packages we trained our models with. The list of files is provided above.

• Should include dataset datasheet, as described in Gebru et al 2021 (arXiv:1803.09010).

We have included a datasheet for our dataset in the appendix lines 621-764.

• Should include crowdsourced annotation datasheet, as described in Diaz et al 2022 (arXiv:2206.08931). Alternatively, the specific instructions that were provided to Hive/annotators would be highly relevant to convey what annotation protocols were employed here.

We have included the list of instructions sent to the Hive annotators in the supplementary materials. File: OpenMonkeyWild Photograph Rubric.pdf

• Should include model cards, as described in Mitchell et al (arXiv:1810.03993).

We have included a model card for the included model in the results section line 359. See Author response image 1.

Author response image 1.  

• It would be useful to include more information on the source of the data as they are collected from many different sites and from many different individuals, some of which may introduce structural biases such as lighting conditions due to geography and time of year.

We agree that the source could introduce structural biases. This is why we included images from so many different sources and captured images at different times from the same source—in hopes that a large variety of background and lighting conditions are represented. However, doing so limits our ability to document each source background and lighting condition separately.

• Is there a reason not to use OKS? This incorporates several factors such as landmark visibility, scale, and landmark type-specific annotation variability as in Ronchi & Perona 2017 (arXiv:1707.05388). The latter (variability) could use the human pose values (for landmarks types that are shared), the least variable keypoint class in humans (eyes) as a conservative estimate of accuracy, or leverage a unique aspect of this work (crowdsourced annotations) which affords the ability to estimate these values empirically.

The focus of this work is on overall keypoint localization accuracy and hence we wanted a metric that is easy to interpret and implement, in this case we made use of PCK (Percentage of Correct Keypoints). PCK is a simple and widely used metric that measures the percentage of correctly localized keypoints within a certain distance threshold from their corresponding groundtruth keypoints.

• A reporting of the scales present in the dataset would be useful (e.g., histogram of unnormalized bounding boxes) and would align well with existing pose dataset papers such as MS-COCO (arXiv:1405.0312) which reports the distribution of instance sizes and instance density per image.

RESPONSE: We have now included a histogram of unnormalized bounding boxes in the manuscript, Author response image 2.

Author response image 2.

Reviewer #2 (Public Review):

The authors present the OpenApePose database constituting a collection of over 70000 ape images which will be important for many applications within primatology and the behavioural sciences. The authors have also rigorously tested the utility of this database in comparison to available Pose image databases for monkeys and humans to clearly demonstrate its solid potential.

We thank the reviewer for the kind comments.

However, the variation in the database with regards to individuals, background, source/setting is not clearly articulated and would be beneficial information for those wishing to make use of this resource in the future. At present, there is also a lack of clarity as to how this image database can be extrapolated to aid video data analyses which would be highly beneficial as well.

I have two major concerns with regard to the manuscript as it currently stands which I think if addressed would aid the clarity and utility of this database for readers.

1) Human annotators are mentioned as doing the 16 landmarks manually for all images but there is no assessment of inter-observer reliability or the such. I think something to this end is currently missing, along with how many annotators there were. This will be essential for others to know who may want to use this database in the future.

We thank the reviewer for pointing this out. Inter-observer reliability is important for ensuring the quality of the annotations. We first used Amazon MTurk to crowd source annotations and found that the inter-observer reliability and the annotation quality was poor. This was the reason for choosing a commercial service such as Hive AI. As the crowd sourcing and quality control are managed by Hive through their internal procedures, we do not have access to data that can allow us to assess inter-observer reliability. However, the annotation quality was assessed by first author ND through manual inspections of the annotations visualized on all of the images the database. Additionally, our ablation experiments with high out of sample performances further vaildate the quality of the annotations.

Relevant to this comment, in your description of the database, a table or such could be included, providing the number of images from each source/setting per species and/or number of individuals. Something to give a brief overview of the variation beyond species. (subspecies would also be of benefit for example).

Our goal was to obtain as many images as possible from the most commonly studied ape species. In order to ensure a large enough database, we focused only on the species and combined images from as many sources as possible to reach our goal of ~10,000 images per species. With the wide range of people involved in obtaining the images, we could not ensure that all the photographers had the necessary expertise to differentiate individuals and subspecies of the subjects they were photographing. We could only ensure that the right species was being photographed. Hence, we cannot include more detailed information.

2) You mention around line 195 that you used a specific function for splitting up the dataset into training, validation, and test but there is no information given as to whether this was simply random or if an attempt to balance across species, individuals, background/source was made. I would actually think that a balanced approach would be more appropriate/useful here so whether or not this was done, and the reasoning behind that must be justified.

This is especially relevant given that in one test you report balancing across species (for the sample size subsampling procedure).

We created the training set to reflect the species composition of the whole dataset, but used test sets balanced by species. This was done to give a sense of the performance of a model that could be trained with the entire dataset, that does not have the species fully balanced. We believe that researchers interested in training models using this dataset for behavior tracking applications would use the entire dataset to fully leverage the variation in the dataset. However, for those interested in training models with balanced species, we provide an annotation file with all the images included, which would allow researchers to create their own training and test sets that meet their specific needs. We have added this justification in the manuscript to guide the other users with different needs. Lines 530-534: “We did not balance our training set for the species as we wanted to utilize the full variation in the dataset and assess models trained with the proportion of species as reflected in the dataset. We provide annotations including the entire dataset to allow others to make create their own training/validation/test sets that suit their needs.”

And another perhaps major concern that I think should also be addressed somewhere is the fact that this is an image database tested on images while the abstract and manuscript mention the importance of pose estimation for video datasets, yet the current manuscript does not provide any clear test of video datasets nor engage with the practicalities associated with using this image-based database for applications to video datasets. Somewhere this needs to be added to clarify its practical utility.

We thank the reviewer for this important suggestion. Since we can separate a video into its constituent frames, one can indeed use the provided model or other models trained using this dataset for inference on the frames, thus allowing video tracking applications. We now include a short video clip of a chimpanzee with inferences from the provided model visualized in the supplementary materials.

Reviewer #1 (Recommendations For The Authors):

• Please provide a more thorough description of the annotation procedure (i.e., the instructions given to crowd workers)! See public review for reference on dataset annotation reporting cards.

We have included the list of instructions for Hive annotators in the supplementary materials.

• An estimate of the crowd worker accuracy and variability would be super valuable!

While we agree that this is useful, we do not have access to Hive internal data on crowd worker IDs that could allow us to estimate these metrics. Furthermore, we assessed each image manually to ensure good annotation quality.

• In the methods section it is reported that images were discarded because they were either too blurry, small, or highly occluded. Further quantification could be provided. How many images were discarded per species?

It’s not really clear to us why this is interesting or important. We used a large number of photographers and annotators, some of whom gave a high ratio of great images; some of whom gave a poor ratio. But it’s not clear what those ratios tell us.

• Placing the numerical values at the end of the bars would make the graphs more readable in Figures 4 and 5.

We thank the reviewer for this suggestion. While we agree that this can help, we do not have space to include the number in a font size that would be readable. Smaller font sizes that are likely to fit may not be readable for all readers. We have included the numerical values in the main text in the results section for those interested and hope that the figures provide a qualitative sense of the results to the readers.

eLife assessment

The OpenApePose database presented in this manuscript will be important for many applications within primatology and the behavioural sciences, and a beneficial resource for developing additional tools using computer-vision based methods. The authors have rigorously tested the utility of this database to clearly demonstrate its convincing potential, especially in relation to current alternatives. The transparent and open nature of this work will surely be beneficial to advancing automated methods for pose estimation both in captive and wild settings, and for image and video processing.

Reviewer #2 (Public Review):

The authors present the OpenApePose database constituting a collection of over 70000 ape images which will be important for many applications within primatology and the behavioural sciences. The authors have also rigorously tested the utility of this database in comparison to available Pose image databases for monkeys and humans to clearly demonstrate its solid potential. However, the variation in the database with regards to individuals, background, source/setting is not clearly articulated and would be beneficial information for those wishing to make use of this resource in the future.

Author Response

eLife assessment

Building on their own prior work, the authors present valuable findings that add to our understanding of cortical astrocytes, which respond to synaptic activity with calcium release in subcellular domains that can proceed to larger calcium waves. The proposed concept of a spatial "threshold" is based on solid evidence from in vivo and ex vivo imaging data and the use of mutant mice. However, details of the specific threshold should be taken with caution and appear incomplete unless supported by additional experiments with higher resolution in space and time.

We thank the reviewers and editors for the positive assessment of our work as containing valuable findings that add to our understanding of cortical astrocytes. We also appreciate their positive appraisal of the proposed concept of a spatial threshold supported by solid evidence.

Regarding their specific comments, we truly appreciate them because they have helped to clarify issues and to improve the study. Provisional point-by-point responses to these comments are provided below. Regarding the general comment on the spatial and temporal resolution of our study, we would like to clarify that the spatial and temporal resolution used in the current study (i.e., 2 - 5 Hz framerate using a 25x objective with 1.7x digital zoom with pixels on the order of 1 µm2) is within the norm in the field, does not compromise the results, nor diminish the main conceptual advancement of the study, namely the existence of a spatial threshold for astrocyte calcium surge.

We respect the thoughtfulness of the reviewers and editors and look forward to improving the paper to fully answer both public and private comments with a revised manuscript.

Reviewer #1 (Public Review):

Lines et al., provide evidence for a sequence of events in vivo in adult anesthetized mice that begin with a footshock driving activation of neural projections into layer 2/3 somatosensory cortex, which in turn triggers a rise in calcium in astrocytes within "domains" of their "arbor". The authors segment the astrocyte morphology based on SR101 signal and show that the timing of "arbor" Ca2+ activation precedes somatic activation and that somatic activation only occurs if at least {greater than or equal to}22.6% of the total segmented astrocyte "arbor" area is active. Thus, the authors frame this {greater than or equal to}22.6% activation as a spatial property (spatial threshold) with certain temporal characteristics - i.e., must occur before soma and global activation. The authors then elaborate on this spatial threshold by providing evidence for its intrinsic nature - is not set by the level of neuronal stimulus and is dependent on whether IP3R2, which drives Ca2+ release from the endoplasmic reticulum (ER) in astrocytes, is expressed. Lastly, the authors suggest a potential physiologic role for this spatial threshold by showing ex vivo how exogenous activation of layer 2/3 astrocytes by ATP application can gate glutamate gliotransmission to layer 2/3 cortical neurons - with a strong correlation between the number of active astrocyte Ca2+ domains and the slow inward current (SIC) frequency recorded from nearby neurons as a readout of glutamatergic gliotransmission. This is interesting and would potentially be of great interest to readers within and outside the glia research community, especially in how the authors have tried to systematically deconstruct some of the steps underlying signal integration and propagation in astrocytes. Many of the conclusions posited by the authors are potentially important but we think their approach needs experimental/analytical refinement and elaboration.

We thank the reviewer for her/his positive appraisal and comments that has helped us to improve the study. In response to their insights, we aim to address the key points raised below:

1. Sequence of Events: We acknowledge the reviewer's interest in our findings regarding the sequence of events. We will provide a more detailed description of the methods and results to clarify the temporal relationships between neural activation, astrocyte calcium dynamics, and astrocyte morphology segmentation.

2. Spatial Threshold: The reviewer accurately identifies our characterization of a spatial threshold (≥22.6% activation) with temporal characteristics as a crucial aspect of our study. We will expand upon this concept by offering a clearer illustration of how this threshold relates to somatic and global activation.

3. Intrinsic Nature of Spatial Threshold: The reviewer's insightful observation regarding the inherent quality of the spatial threshold, regardless of its dependence on neuronal stimuli is noteworthy. We will provide additional details to substantiate this claim, shedding more light on the fundamental nature of this phenomenon.

4. Physiological Implications: The reviewer rightly highlights the potential physiological significance of our findings, particularly in relation to gliotransmission in cortical neurons. We will enhance our discussion by elaborating on the implications of these observations.

The primary issue for us, and which we would encourage the authors to address, relates to the low spatialtemporal resolution of their approach. This issue does not necessarily compromise the concept of a spatial threshold, but more refined observations and analyses are likely to provide more reliable quantitative parameters and a more comprehensive view of the mode of Ca2+ signal integration in astrocytes.

We agree with the reviewer that our spatial-temporal resolution (2 – 5 Hz framerate using a 25x objective and 1.7x digital zoom with pixels on the order of 1 µm) does not compromise the proposed concept of the existence of a spatial threshold for the intracellular calcium expansion.

For this reason, and because their observations might be perceived as both a conceptual and numerical standard in the field, we believe that the authors should proceed with both experimental and analytical refinement. Notably, we have difficulty with the reported mean delays of astrocyte Ca2+ elevations upon sensory stimulation. The 11s delay for response onset in "arbor" and 13s in the soma are extremely long, and we do not think they represent a true physiologic latency for astrocyte responses to the sensory activity. Indeed, such delays appear to be slower even than those reported in the initial studies of sensory stimulation in anesthetized mice with limited spatial-temporal resolution (Wang et al. Nat Neurosci., 2006) - not to say of more recent and refined ones in awake mice (Stobart et al. Neuron, 2018) that identified even sub-second astrocyte Ca2+ responses, largely preserved in IP3R2KO mice. Thus, we are inclined to believe that the slowness of responses reported here is an indicator of experimental/analytical issues. There can be several explanations of such slowness that the authors may want to consider for improving their approach: (a) The authors apparently use low zoom imaging for acquiring signals from several astrocytes present in the FOV: do all of these astrocytes respond homogeneously in terms of delay from sensory stimulus? Perhaps some are faster responders than others and only this population is directly activated by the stimulus. Others could be slower in activation because they respond secondarily to stimuli. In this case, the authors could focus their analysis specifically on the "fast-responding population". (b) By focusing on individual astrocytes and using higher zoom, the authors could unmask more subtle Ca2+ elevations that precede those reported in the current manuscript. These signals have been reported to occur mainly in regions of the astrocyte that are GCaMP6-positive but SR101-negative and constitute a large percentage of its volume (Bindocci et al., 2017). By restricting analysis to the SR101-positive part of the astrocyte, the authors might miss the fastest components of the astrocyte Ca2+ response likely representing the primary signals triggered by synaptic activity. It would be important if they could identify such signals in their records, and establish if none/few/many of them propagate to the SR-101-positive part of the astrocyte. In other words, if there is only a single spatial threshold, the one the authors reported, or two or more of them along the path of signal propagation towards the cell soma that leads eventually to the transformation of the signal into a global astrocyte Ca2+ surge.

We thank the reviewer for these excellent and important comments. The qualm with the mean delays of astrocyte activation is indeed a result of averaging together astrocyte responses to a 20 second stimulus. Indeed, astrocyte responses are heterogeneous and many astrocytes respond much quicker, as can be seen in example traces in Figs. 1D, 1G, and 3C. Indeed, with any biological system variability exists, however here we take the averaged responses in order to identify a general property of astrocyte calcium dynamics: the existence of the concept of a spatial threshold for astrocyte calcium surge.

Further, we used a lower stimulus frequency (2Hz) than Stobart et al. (90 Hz) to assess subthreshold activities. We found that stronger stimuli decreased response delays and will include this result in the revised manuscript. Interestingly, from Fig 4F, higher stimulus did not significantly alter the spatial threshold. In the revised version of the manuscript, we will provide a more detailed analysis and the consequent discussion of this analysis.

In this context, there is another concept that we encourage the authors to better clarify: whether the spatial threshold that they describe is constituted by the enlargement of a continuous wavefront of Ca2+ elevation, e.g. in a single process, that eventually reaches 22.6% of the segmented astrocyte, or can it also be constituted by several distinct Ca2+ elevations occurring in separate domains of the arbor, but overall totaling 22.6% of the segmented surface? Mechanistically, the latter would suggest the presence of a general excitability threshold of the astrocyte, whereas the former would identify a driving force threshold for the centripetal wavefront. In light of the above points, we think the authors should use caution in presenting and interpreting the experiments in which they use SIC as a readout. Their results might lead some readers to bluntly interpret the 22.6% spatial threshold as the threshold required for the astrocyte to evoke gliotransmitter release. Indeed, SIC are robust signals recorded somatically from a single neuron and likely integrate activation of many synapses all belonging to that neuron. On the other hand, an astrocyte impinges in a myriad of synapses belonging to several distinct neurons. In our opinion, it is quite possible that more local gliotransmission occurs at lower Ca2+ signal thresholds (see above) that may not be efficiently detected by using SIC as a readout; a more sensitive approach, such as the use of a gliotransmitter sensor expressed all along the astrocyte plasma-membrane could be tested to this aim.

The reviewer raised an excellent point. Whether the spatial threshold of 22.6% occur in the segmented astrocyte or may be reached occurring in separate domains of the arbor, is an important question and we aim to address this by novel analysis that will be provided in the revised version of the manuscript.

Regarding comments on SIC, we fully agree with the reviewer. In the revised version of the manuscript, we will include text in the discussion to ensure the correct interpretation of the results, i.e., the observed 22.6% spatial threshold for the SIC does not necessarily indicates an intrinsic property of gliotransmitter release; rather, since SICs have been shown to be calcium-dependent, it is not surprising that their presence, monitored at the whole-cell soma, matches the threshold for the intracellular calcium extension.

Additional considerations are that the authors propose an event sequence as follows: stimulus - synaptic drive to L2/3 - arbor activation - spatial threshold - soma activation - post soma activation - gliotransmission. This seems reminiscent of the sequence underlying neuronal spike propagation - from dendrite to soma to axon, and the resulting vesicular release. However, there is no consensus within the glial field about an analogous framework for astrocytes. Thus, "arbor activation", "soma activation", and "post soma activation" are not established `terms-of-art´. Similarly, the way the authors use the term "domain" contrasts with how others have (Agarwal et al., 2017; Shigetomi et al., 2013; Di Castro et al., 2011; Grosche et al., 1999) and may produce some confusion. The authors could adopt a more flexible nomenclature or clarify that their terms do not have a defined structural-functional basis, being just constructs that they justifiably adapted to deal with the spatial complexity of astrocytes in line with their past studies (Lines et al., 2020; Lines et al., 2021).

We agree there is no consensus within the glial field about this event sequence. One major difference between this sequence of events and neuronal spike propagation is directionality from dendrite to soma to axon. It is unknown whether directionality of the calcium signal exists in astrocytes. The term “microdomain” is used in the references above to define distal subcellular domains in contact with synapses, and in order to dissociate from this term we adopt the nomenclature “domain” to define all subcellular domains in the astrocyte arborization. These items will be discussed and clarified in the revised version of the manuscript.

Our previous points suggest that the paper would be significantly strengthened by new experimental observations focusing on single astrocytes and using acquisitions at higher spatial and temporal resolution. If the authors will not pursue this option, we encourage them to at least improve their analysis, and at the same time recognize in the text some limitations of their experimental approach as discussed above. We indicate here several levels of possible analytical refinement.

We believe our spatial (25x objective and 1.7x digital zoom with pixels on the order of 1µm) and temporal (2 – 5 Hz framerate) resolution is within the range used in the glial field. In any case the existence of a spatial threshold for astrocyte calcium surge is not compromised with the use of this imaging resolution.

The first relates to the selection of astrocytes being analyzed, and the need to focus on a much narrower subpopulation than (for example) 987 astrocytes used for the core data. This selection would take into greater consideration the aspects of structure and latency. With the structural and latency-based criteria for selection, the number of astrocytes to analyze might be reduced by 10-fold or more, making our second analytical recommendation much more feasible.

We agree that individual differences exist, however, establishing a general concept requires the sampling of many astrocytes. Nevertheless, we aim to further address this issue in the revised version of the manuscript by analyzing the calcium dynamics in individual domains.

For structure-based selection - Genetically-encoded Ca2+ indicators such as GCaMP6 are in principle expressed throughout an astrocyte, even in regions that are not labelled by SR101. Moreover, astrocytes form independent 3D territories, so one can safely assume that the GCaMP6 signal within an astrocyte volume belongs to that specific astrocyte (this is particularly evident if the neighboring astrocytes are GCaMP6negative). Therefore, authors could extend their analysis of Ca2+ signals in individual astrocytes to the regions that are SR101-negative and try to better integrate fast signals in their spatial threshold concept. Even if they decided to be conservative on their methods, and stick to the astrocyte segmentation based on the SR-101 signal, they should acknowledge that SR101 dye staining quality can vary considerably between individual astrocytes within a FOV - some astrocytes will have much greater structural visibility in the distal processes than others. This means that some astrocytes may have segmented domains extending more distally than others and we think that authors should privilege such astrocytes for analysis. However, cases like the representative astrocytes shown in Figure 4A or Figure S1B, have segmented domains localized only to proximal processes near the soma. Accordingly, given the reported timing differences between "arbor" and "soma" activation, one might expect there to be comparable timing differences between domains that are distal vs proximal to the soma as well. Fast signals in peripheral regions of astrocytes in contact with synapses are largely IP3R2-independent (Stobart et al., 2018). However, the quality of SR101 staining has implications for interpreting the IP3R2 KO data. There is evidence IP3R2 KO may preferentially impact activity near the soma (Srinivasan et al., 2015). Thus, astrocytes with insufficient staining - visible only in the soma and proximal domains - might show a biased effect for IP3R2 KO. While not necessarily disrupting the core conclusions made by the authors based on their analysis of SR101-segmented astrocytes, we think results would be strengthened if astrocytes with sufficient SR101 staining - i.e. more consistent with previous reports of L2/3 astrocyte area (Lanjakornsiripan et al., 2018) - were only included. This could be achieved by using max or cumulative projections of individual astrocytes in combination with SR101 staining to construct more holistic structural maps (Bindocci et al., 2017).

We agree with the ideas concerning SR101, and indeed there could be variability in the origins of the astrocyte calcium signal. Astrocyte territory boundaries can be difficult to discern when both astrocytes express GCaMP6. Here we take a conservative approach to constrain ROIs to SR101-positive astrocyte territory outlines without invading neighboring cells in order to reduce error in the estimate of a spatial threshold. The effect of IP3R2 KO preferentially impacting activity near the soma is interesting, and in line with our conclusions. We agree that the findings from SR101-negative pixels would not necessarily disrupt the core conclusions of the study, and the additional analysis suggested would further strengthen results.

For latency-based selection - The authors record calcium activity within a FOV containing at least 20+ astrocytes over a period of 60s, during which a 2Hz hindpaw stimulation at 2mA is applied for 20s. As discussed above, presumably some astrocytes in a FOV are the first to respond to the stimulus series, while others likely respond with longer latency to the stimulus. For the shorter-latency responders <3s, it is easier to attribute their calcium increases as "following the sensory information" projecting to L2/3. In other cases, when "arbor" responses occur at 10s or later, only after 20 stimulus events (at 2Hz), it is likely they are being activated by a more complex and recurrent circuit containing several rounds of neuron-glia crosstalk etc., which would be mechanistically distinct from astrocytes responding earlier. We suggest that authors focus more on the shorter latency response astrocytes, as they are more likely to have activity corresponding to the stimulus itself.

We agree that different times of astrocyte calcium increases may be due to different mechanisms outside of the astrocyte. We believe the spatial threshold will be intrinsic to these external variables; yet we believe that longer latency responses are physiological and may carry important information to determining the astrocyte calcium responses.

The second level of analysis refinement we suggest relates specifically to the issue of propagation and timing for the activity within "arbor", "soma" and "post-soma". Currently, the authors use an ROI-based approach that segments the "arbor" into domains. We suggest that this approach could be supplemented by a more robust temporal analysis. This could for example involve starting with temporal maps that take pixels above a certain amplitude and plot their timing relative to the stimulus-onset, or (better) the first active pixel of the astrocyte. This type of approach has become increasingly used (Bindocci et al., 2017; Wang et al., 2019; Ruprecht et al., 2022) and we think its use can greatly help clarify both the proposed sequence and better characterize the spatial threshold. We think this analysis should specifically address several important points:

We agree that the creation of temporal maps from our own data will be interesting. We will provide the results of the suggested analysis in the revised version of the manuscript.

1) Where/when does the astrocyte activation begin? Understanding the beginning is very important, particularly because another potential spatial threshold - preceding the one the authors describe in the paper - could gate the initial activation of more distal processes, as discussed above. This sequentially earlier spatial threshold could (for example) rely on microdomain interaction with synaptic elements and (in contrast) be IP3R2 independent (Srinivasan et al., 2015, Stobart et al., 2018). We would be interested to know whether, in a subset of astrocytes that meet the structure and latency criteria proposed above and can produce global activation, there is an initial local GCaMP6f response of a minimal size that must occur before propagation towards the soma begins. The data associated with varying stimulus parameters could potentially be useful here and reveal stimulus intensity/duration-dependent differences.

This is a very important point. It is difficult to pinpoint the beginning of the signal, which is why we rely on the average of responses.

2) Whether the propagation in the authors' experimental model is centripetal? This is implied throughout the manuscript but never shown. We think establishing whether (or not) the calcium dynamics are centripetal is important because it would clarify whether spatially adjacent domains within the "arbor" need to be sequentially active before reaching the threshold and then reaching the soma. More broadly, visualizing propagation will help to better visualize summation, which is presumably how the threshold is first reached (and overcome). The alternative hypothesis of a general excitability threshold, as discussed above, would be challenged here and possibly rejected, thereby clarifying the nature of the Ca2+ process that needs to reach a threshold for further expansion to the soma and other parts of the astrocyte.

We agree that our view is centripetal. Indeed, we have found arborization activity precedes soma activity. However, whether this is intrinsic or due to the fact that synapses are more likely to occur in the periphery requires further studies.

3) In complement to the previous point: we understand that the spatial threshold does not per se have a location, but is there some spatial logic underlying the organization of active domains before the soma response occurs? One can easily imagine multiple scenarios of sparse heterogeneous GCaMP6f signal distributions that correspond to {greater than or equal to}22.6% of the arborization, but that would not be expected to trigger soma activation. For example, the diagram in Figure 4C showing the astrocyte response to 2Hz stim (which lacks a soma response) underscores this point. It looks like it has {greater than or equal to}22.6% activation that is sparsely localized throughout the arborization. If an alternative spatial distribution for this activity occurred, such that it localized primarily to a specific process within the arbor, would it be more likely to trigger a soma response?

This is an interesting point and an analysis of spatial clustering on pre-soma domain activation may be useful to answer it.

4) Does "pre-soma" activation predict the location and onset time of "post-soma" activation? For example, are arbor domains that were part of the "pre-soma" response the first to exhibit GCaMP6f signal in the "post-soma" response?

This is another interesting analysis that can be done with a spatial clustering analysis.

Reviewer #2 (Public Review):

Lines et al investigated the integration of calcium signals in astrocytes of the primary somatosensory cortex. Their goal was to better characterize the mechanisms that govern the spatial characteristics of calcium signals in astrocytes. In line with previous reports in the field, they found that most events originated and stayed localized within microdomains in distal astrocyte processes, occasionally coinciding with larger events in the soma, referred to as calcium surges. As a single astrocyte communicates with hundreds of thousands of synapses simultaneously, understanding the spatial integration of calcium signals in astrocytes and the mechanisms governing the latter is of tremendous importance to deepen our understanding of signal processing in the central nervous system. The authors thus aimed to unveil the properties governing the emergence of calcium surges. The main claim of this manuscript is that there would be a spatial threshold of ~23% of microdomain activation above which a calcium surge, i.e. a calcium signal that spreads to the soma, is observed. Although the study provides data that is highly valuable for the community, the conclusions of the current version of the manuscript seem a little too assertive and general compared with what can be deduced from the data and methods used.

The major strength of this study is the experimental approach that allowed the authors to obtain numerous and informative calcium recordings in vivo in the somatosensory cortex in mice in response to sensory stimuli as well as in situ. Notably, they developed an interesting approach to modulating the number of active domains in peripheral astrocyte processes by varying the intensity of peripheral stimulation (its amplitude, frequency, or duration).

We thank the reviewer for their kind and thoughtful review of our study.

The major weakness of the manuscript is the method used to analyze and quantify calcium activity, which mostly relies on the analysis of averaged data and overlooks the variability of the signals measured. As a result, the main claims from the manuscript seem to be incompletely supported by the data. The choice of the use of a custom-made semi-automatic ROI-based calcium event detection algorithm rather than established state-of-the-art software, such as the event-based calcium event detection software AQuA (DOI: 10.1038/s41593-019-0492-2), is insufficiently discussed and may bias the analysis. Some references on this matter include: Semyanov et al, Nature Rev Neuro, 2020 (DOI: 10.1038/s41583-020-0361-8); Covelo et al 2022, J Mol Neurosci (DOI: 10.1007/s12031-022-02006-w) & Wang et al, 2019, Nat Neuroscience (DOI: 10.1038/s41593-019-0492-2). Moreover, the ROIs used to quantify calcium activity are based on structural imaging of astrocytes, which may not be functionally relevant.

Unfortunately, there is no general consensus for calcium analysis in the astrocyte or neuronal field, and many groups use custom made software made in lab or custom software such as GECIquant or AQuA. While AQuA is an event-based calcium event detection software, it may be that not including inactive domains that are SR101 positive could underestimate the spatial threshold for calcium surge. Our data is not based on the functional events but is based on calcium with structural constraints within a single astrocyte. This is crucial to properly determine the ratio of active vs inactive pixels within a single astrocyte.

For the reasons listed above, the manuscript would probably benefit from some rephrasing of the conclusions and a discussion highlighting the advantages and limitations of the methodological approach. The question investigated by this study is of great importance in the field of neuroscience as the mechanisms dictating the spatio-temporal properties of calcium signals in astrocytes are poorly characterized, yet are essential to understand their involvement in the modulation of signal integration within neural circuits.

We thank the reviewer for their suggestions to benefit the conclusions and discussion.

Reviewer #3 (Public Review):

Summary:

The study aims to elucidate the spatial dynamics of subcellular astrocytic calcium signaling. Specifically, they elucidate how subdomain activity above a certain spatial threshold (~23% of domains being active) heralds a calcium surge that also affects the astrocytic soma. Moreover, they demonstrate that processes on average are included earlier than the soma and that IP3R2 is necessary for calcium surges to occur. Finally, they associate calcium surges with slow inward currents.

Strengths:

The study addresses an interesting topic that is only partially understood. The study uses multiple methods including in vivo two-photon microscopy, acute brain slices, electrophysiology, pharmacology, and knockout models. The conclusions are strengthened by the same findings in both in vivo anesthetized mice and in brain slices.

We thank the reviewer for the positive assessment of the study and his/her comments.

Weaknesses:

The method that has been used to quantify astrocytic calcium signals only analyzes what seems to be a small proportion of the total astrocytic domain on the example micrographs, where a structure is visible in the SR101 channel (see for instance Reeves et al. J. Neurosci. 2011, demonstrating to what extent SR101 outlines an astrocyte). This would potentially heavily bias the results: from the example illustrations presented it is clear that the calcium increases in what is putatively the same astrocyte goes well beyond what is outlined with automatically placed small ROIs. The smallest astrocytic processes are an order of magnitude smaller than the resolution of optical imaging and would not be outlined by either SR101 or with the segmentation method judged by the ROIs presented in the figures. Completely ignoring these very large parts of the spatial domain of an astrocyte, in particular when making claims about a spatial threshold, seems inappropriate. Several recent methods published use pixel-by-pixel event-based approaches to define calcium signals. The data should have been analyzed using such a method within a complete astrocyte spatial domain in addition to the analyses presented. Also, the authors do not discuss how two-dimensional sampling of calcium signals from an astrocyte that has processes in three dimensions (see Bindocci et al, Science 2017) may affect the results: if subdomain activation is not homogeneously distributed in the three-dimensional space within the astrocyte territory, the assumptions and findings between a correlation between subdomain activation and somatic activation may be affected.

In order to reduce noise from individual pixels, we chose to segment astrocyte arborizations into domains of several pixels. As pointed out previously, including pixels outside of the SR101-positive territory runs the risk of including a pixel that may be from a neighboring cell, and we chose to avoid this source of error. We agree that the results have limitations from being acquired in 2D instead of 3D, but it is likely to assume the 3D astrocyte is homogeneously distributed and that the 2D plane is representative of the whole astrocyte. Indeed, no dimensional effects were reported in Bindocci et al, Science 2017. We plan to include a paragraph in the discussion to address this limitation in our study.

The experiments are performed either in anesthetized mice, or in slices. The study would have come across as much more solid and interesting if at least a small set of experiments were performed also in awake mice (for instance during spontaneous behavior), given the profound effect of anesthesia on astrocytic calcium signaling and the highly invasive nature of preparing acute brain slices. The authors mention the caveat of studying anesthetized mice but claim that the intracellular machinery should remain the same. This explanation appears a bit dismissive as the response of an astrocyte not only depends on the internal machinery of the astrocyte, but also on how the astrocyte is stimulated: for instance synaptic stimulation or sensory input likely would be dependent on brain state and concurrent neuromodulatory signaling which is absent in both experimental paradigms. The discussion would have been more balanced if these aspects were dealt with more thoroughly.

Yes, we agree that this is a limitation, and we will acknowledge this is in the discussion.

The study uses a heaviside step function to define a spatial 'threshold' for somata either being included or not in a calcium signal. However, Fig 4E and 5D showing how the method separates the signal provide little understanding for the reader. The most informative figure that could support the main finding of the study, namely a ~23% spatial threshold for astrocyte calcium surges reaching the soma, is Fig. 4G, showing the relationship between the percentage of arborizations active and the soma calcium signal. A similar plot should have been presented in Fig 5 as well. Looking at this distribution, though, it is not clear why ~23% would be a clear threshold to separate soma involvement, one can only speculate how the threshold for a soma event would influence this number. Even if the analyses in Fig. 4H and the fact that the same threshold appears in two experimental paradigms strengthen the case, the results would have been more convincing if several types of statistical modeling describing the continuous distribution of values presented in Fig. 4E (in addition to the heaviside step function) were presented.

We agree with the reviewer that we should add to the paper a discussion for our justification on the use of the Heaviside step function, and plan to include this. We chose the Heaviside step function to represent the on/off situation that we observed in the data. We agree with the reviewer that Fig. 4G is informative and demonstrates that under 23% most of the soma fluorescence values are clustered at baseline. We agree that a similar graph should be included in Fig. 5 as well. We agree that a different statistical model describing the data would be more convincing and also confirmed the spatial threshold with the use of a confidence interval in the text.

The description of methods should have been considerably more thorough throughout. For instance which temperature the acute slice experiments were performed at, and whether slices were prepared in ice-cold solution, are crucial to know as these parameters heavily influence both astrocyte morphology and signaling. Moreover, no monitoring of physiological parameters (oxygen level, CO2, arterial blood gas analyses, temperature etc) of the in vivo anesthetized mice is mentioned. These aspects are critical to control for when working with acute in vivo two-photon microscopy of mice; the physiological parameters rapidly decay within a few hours with anesthesia and following surgery.

We will increase the thoroughness of our methods section. Especially including that body temperature and respiration were indeed monitored throughout anesthesia.

eLife assessment

Building on their own prior work, the authors present valuable findings that add to our understanding of cortical astrocytes, which respond to synaptic activity with calcium release in subcellular domains that can proceed to larger calcium waves. The proposed concept of a spatial "threshold" is based on solid evidence from in vivo and ex vivo imaging data and the use of mutant mice. However, details of the specific threshold should be taken with caution and appear incomplete unless supported by additional experiments with higher resolution in space and time.

Reviewer #1 (Public Review):

Lines et al., provide evidence for a sequence of events in vivo in adult anesthetized mice that begin with a foot-shock driving activation of neural projections into layer 2/3 somatosensory cortex, which in turn triggers a rise in calcium in astrocytes within "domains" of their "arbor". The authors segment the astrocyte morphology based on SR101 signal and show that the timing of "arbor" Ca2+ activation precedes somatic activation and that somatic activation only occurs if at least {greater than or equal to}22.6% of the total segmented astrocyte "arbor" area is active. Thus, the authors frame this {greater than or equal to}22.6% activation as a spatial property (spatial threshold) with certain temporal characteristics - i.e., must occur before soma and global activation. The authors then elaborate on this spatial threshold by providing evidence for its intrinsic nature - is not set by the level of neuronal stimulus and is dependent on whether IP3R2, which drives Ca2+ release from the endoplasmic reticulum (ER) in astrocytes, is expressed. Lastly, the authors suggest a potential physiologic role for this spatial threshold by showing ex vivo how exogenous activation of layer 2/3 astrocytes by ATP application can gate glutamate gliotransmission to layer 2/3 cortical neurons - with a strong correlation between the number of active astrocyte Ca2+ domains and the slow inward current (SIC) frequency recorded from nearby neurons as a readout of glutamatergic gliotransmission. This is interesting and would potentially be of great interest to readers within and outside the glia research community, especially in how the authors have tried to systematically deconstruct some of the steps underlying signal integration and propagation in astrocytes. Many of the conclusions posited by the authors are potentially important but we think their approach needs experimental/analytical refinement and elaboration.

The primary issue for us, and which we would encourage the authors to address, relates to the low spatial-temporal resolution of their approach. This issue does not necessarily compromise the concept of a spatial threshold, but more refined observations and analyses are likely to provide more reliable quantitative parameters and a more comprehensive view of the mode of Ca2+ signal integration in astrocytes. For this reason, and because their observations might be perceived as both a conceptual and numerical standard in the field, we believe that the authors should proceed with both experimental and analytical refinement. Notably, we have difficulty with the reported mean delays of astrocyte Ca2+ elevations upon sensory stimulation. The 11s delay for response onset in "arbor" and 13s in the soma are extremely long, and we do not think they represent a true physiologic latency for astrocyte responses to the sensory activity. Indeed, such delays appear to be slower even than those reported in the initial studies of sensory stimulation in anesthetized mice with limited spatial-temporal resolution (Wang et al. Nat Neurosci., 2006) - not to say of more recent and refined ones in awake mice (Stobart et al. Neuron, 2018) that identified even sub-second astrocyte Ca2+ responses, largely preserved in IP3R2KO mice. Thus, we are inclined to believe that the slowness of responses reported here is an indicator of experimental/analytical issues. There can be several explanations of such slowness that the authors may want to consider for improving their approach: (a) The authors apparently use low zoom imaging for acquiring signals from several astrocytes present in the FOV: do all of these astrocytes respond homogeneously in terms of delay from sensory stimulus? Perhaps some are faster responders than others and only this population is directly activated by the stimulus. Others could be slower in activation because they respond secondarily to stimuli. In this case, the authors could focus their analysis specifically on the "fast-responding population". (b) By focusing on individual astrocytes and using higher zoom, the authors could unmask more subtle Ca2+ elevations that precede those reported in the current manuscript. These signals have been reported to occur mainly in regions of the astrocyte that are GCaMP6-positive but SR101-negative and constitute a large percentage of its volume (Bindocci et al., 2017). By restricting analysis to the SR101-positive part of the astrocyte, the authors might miss the fastest components of the astrocyte Ca2+ response likely representing the primary signals triggered by synaptic activity. It would be important if they could identify such signals in their records, and establish if none/few/many of them propagate to the SR-101-positive part of the astrocyte. In other words, if there is only a single spatial threshold, the one the authors reported, or two or more of them along the path of signal propagation towards the cell soma that leads eventually to the transformation of the signal into a global astrocyte Ca2+ surge. In this context, there is another concept that we encourage the authors to better clarify: whether the spatial threshold that they describe is constituted by the enlargement of a continuous wavefront of Ca2+ elevation, e.g. in a single process, that eventually reaches 22.6% of the segmented astrocyte, or can it also be constituted by several distinct Ca2+ elevations occurring in separate domains of the arbor, but overall totaling 22.6% of the segmented surface? Mechanistically, the latter would suggest the presence of a general excitability threshold of the astrocyte, whereas the former would identify a driving force threshold for the centripetal wavefront. In light of the above points, we think the authors should use caution in presenting and interpreting the experiments in which they use SIC as a readout. Their results might lead some readers to bluntly interpret the 22.6% spatial threshold as the threshold required for the astrocyte to evoke gliotransmitter release. Indeed, SIC are robust signals recorded somatically from a single neuron and likely integrate activation of many synapses all belonging to that neuron. On the other hand, an astrocyte impinges in a myriad of synapses belonging to several distinct neurons. In our opinion, it is quite possible that more local gliotransmission occurs at lower Ca2+ signal thresholds (see above) that may not be efficiently detected by using SIC as a readout; a more sensitive approach, such as the use of a gliotransmitter sensor expressed all along the astrocyte plasma-membrane could be tested to this aim.

Additional considerations are that the authors propose an event sequence as follows: stimulus - synaptic drive to L2/3 - arbor activation - spatial threshold - soma activation - post soma activation - gliotransmission. This seems reminiscent of the sequence underlying neuronal spike propagation - from dendrite to soma to axon, and the resulting vesicular release. However, there is no consensus within the glial field about an analogous framework for astrocytes. Thus, "arbor activation", "soma activation", and "post soma activation" are not established `terms-of-art´. Similarly, the way the authors use the term "domain" contrasts with how others have (Agarwal et al., 2017; Shigetomi et al., 2013; Di Castro et al., 2011; Grosche et al., 1999) and may produce some confusion. The authors could adopt a more flexible nomenclature or clarify that their terms do not have a defined structural-functional basis, being just constructs that they justifiably adapted to deal with the spatial complexity of astrocytes in line with their past studies (Lines et al., 2020; Lines et al., 2021).

Our previous points suggest that the paper would be significantly strengthened by new experimental observations focusing on single astrocytes and using acquisitions at higher spatial and temporal resolution. If the authors will not pursue this option, we encourage them to at least improve their analysis, and at the same time recognize in the text some limitations of their experimental approach as discussed above. We indicate here several levels of possible analytical refinement.

The first relates to the selection of astrocytes being analyzed, and the need to focus on a much narrower subpopulation than (for example) 987 astrocytes used for the core data. This selection would take into greater consideration the aspects of structure and latency. With the structural and latency-based criteria for selection, the number of astrocytes to analyze might be reduced by 10-fold or more, making our second analytical recommendation much more feasible.

For structure-based selection - Genetically-encoded Ca2+ indicators such as GCaMP6 are in principle expressed throughout an astrocyte, even in regions that are not labelled by SR101. Moreover, astrocytes form independent 3D territories, so one can safely assume that the GCaMP6 signal within an astrocyte volume belongs to that specific astrocyte (this is particularly evident if the neighboring astrocytes are GCaMP6-negative). Therefore, authors could extend their analysis of Ca2+ signals in individual astrocytes to the regions that are SR101-negative and try to better integrate fast signals in their spatial threshold concept. Even if they decided to be conservative on their methods, and stick to the astrocyte segmentation based on the SR-101 signal, they should acknowledge that SR101 dye staining quality can vary considerably between individual astrocytes within a FOV - some astrocytes will have much greater structural visibility in the distal processes than others. This means that some astrocytes may have segmented domains extending more distally than others and we think that authors should privilege such astrocytes for analysis. However, cases like the representative astrocytes shown in Figure 4A or Figure S1B, have segmented domains localized only to proximal processes near the soma. Accordingly, given the reported timing differences between "arbor" and "soma" activation, one might expect there to be comparable timing differences between domains that are distal vs proximal to the soma as well. Fast signals in peripheral regions of astrocytes in contact with synapses are largely IP3R2-independent (Stobart et al., 2018). However, the quality of SR101 staining has implications for interpreting the IP3R2 KO data. There is evidence IP3R2 KO may preferentially impact activity near the soma (Srinivasan et al., 2015). Thus, astrocytes with insufficient staining - visible only in the soma and proximal domains - might show a biased effect for IP3R2 KO. While not necessarily disrupting the core conclusions made by the authors based on their analysis of SR101-segmented astrocytes, we think results would be strengthened if astrocytes with sufficient SR101 staining - i.e. more consistent with previous reports of L2/3 astrocyte area (Lanjakornsiripan et al., 2018) - were only included. This could be achieved by using max or cumulative projections of individual astrocytes in combination with SR101 staining to construct more holistic structural maps (Bindocci et al., 2017).

For latency-based selection - The authors record calcium activity within a FOV containing at least 20+ astrocytes over a period of 60s, during which a 2Hz hindpaw stimulation at 2mA is applied for 20s. As discussed above, presumably some astrocytes in a FOV are the first to respond to the stimulus series, while others likely respond with longer latency to the stimulus. For the shorter-latency responders <3s, it is easier to attribute their calcium increases as "following the sensory information" projecting to L2/3. In other cases, when "arbor" responses occur at 10s or later, only after 20 stimulus events (at 2Hz), it is likely they are being activated by a more complex and recurrent circuit containing several rounds of neuron-glia crosstalk etc., which would be mechanistically distinct from astrocytes responding earlier. We suggest that authors focus more on the shorter latency response astrocytes, as they are more likely to have activity corresponding to the stimulus itself.

The second level of analysis refinement we suggest relates specifically to the issue of propagation and timing for the activity within "arbor", "soma" and "post-soma". Currently, the authors use an ROI-based approach that segments the "arbor" into domains. We suggest that this approach could be supplemented by a more robust temporal analysis. This could for example involve starting with temporal maps that take pixels above a certain amplitude and plot their timing relative to the stimulus-onset, or (better) the first active pixel of the astrocyte. This type of approach has become increasingly used (Bindocci et al., 2017; Wang et al., 2019; Ruprecht et al., 2022) and we think its use can greatly help clarify both the proposed sequence and better characterize the spatial threshold. We think this analysis should specifically address several important points:

1. Where/when does the astrocyte activation begin? Understanding the beginning is very important, particularly because another potential spatial threshold - preceding the one the authors describe in the paper - could gate the initial activation of more distal processes, as discussed above. This sequentially earlier spatial threshold could (for example) rely on microdomain interaction with synaptic elements and (in contrast) be IP3R2 independent (Srinivasan et al., 2015, Stobart et al., 2018). We would be interested to know whether, in a subset of astrocytes that meet the structure and latency criteria proposed above and can produce global activation, there is an initial local GCaMP6f response of a minimal size that must occur before propagation towards the soma begins. The data associated with varying stimulus parameters could potentially be useful here and reveal stimulus intensity/duration-dependent differences.

2. Whether the propagation in the authors' experimental model is centripetal? This is implied throughout the manuscript but never shown. We think establishing whether (or not) the calcium dynamics are centripetal is important because it would clarify whether spatially adjacent domains within the "arbor" need to be sequentially active before reaching the threshold and then reaching the soma. More broadly, visualizing propagation will help to better visualize summation, which is presumably how the threshold is first reached (and overcome). The alternative hypothesis of a general excitability threshold, as discussed above, would be challenged here and possibly rejected, thereby clarifying the nature of the Ca2+ process that needs to reach a threshold for further expansion to the soma and other parts of the astrocyte.

3. In complement to the previous point: we understand that the spatial threshold does not per se have a location, but is there some spatial logic underlying the organization of active domains before the soma response occurs? One can easily imagine multiple scenarios of sparse heterogeneous GCaMP6f signal distributions that correspond to {greater than or equal to}22.6% of the arborization, but that would not be expected to trigger soma activation. For example, the diagram in Figure 4C showing the astrocyte response to 2Hz stim (which lacks a soma response) underscores this point. It looks like it has {greater than or equal to}22.6% activation that is sparsely localized throughout the arborization. If an alternative spatial distribution for this activity occurred, such that it localized primarily to a specific process within the arbor, would it be more likely to trigger a soma response?

4. Does "pre-soma" activation predict the location and onset time of "post-soma" activation? For example, are arbor domains that were part of the "pre-soma" response the first to exhibit GCaMP6f signal in the "post-soma" response?

Reviewer #2 (Public Review):

Lines et al investigated the integration of calcium signals in astrocytes of the primary somatosensory cortex. Their goal was to better characterize the mechanisms that govern the spatial characteristics of calcium signals in astrocytes. In line with previous reports in the field, they found that most events originated and stayed localized within microdomains in distal astrocyte processes, occasionally coinciding with larger events in the soma, referred to as calcium surges. As a single astrocyte communicates with hundreds of thousands of synapses simultaneously, understanding the spatial integration of calcium signals in astrocytes and the mechanisms governing the latter is of tremendous importance to deepen our understanding of signal processing in the central nervous system. The authors thus aimed to unveil the properties governing the emergence of calcium surges. The main claim of this manuscript is that there would be a spatial threshold of ~23% of microdomain activation above which a calcium surge, i.e. a calcium signal that spreads to the soma, is observed. Although the study provides data that is highly valuable for the community, the conclusions of the current version of the manuscript seem a little too assertive and general compared with what can be deduced from the data and methods used.

The major strength of this study is the experimental approach that allowed the authors to obtain numerous and informative calcium recordings in vivo in the somatosensory cortex in mice in response to sensory stimuli as well as in situ. Notably, they developed an interesting approach to modulating the number of active domains in peripheral astrocyte processes by varying the intensity of peripheral stimulation (its amplitude, frequency, or duration).

The major weakness of the manuscript is the method used to analyze and quantify calcium activity, which mostly relies on the analysis of averaged data and overlooks the variability of the signals measured. As a result, the main claims from the manuscript seem to be incompletely supported by the data. The choice of the use of a custom-made semi-automatic ROI-based calcium event detection algorithm rather than established state-of-the-art software, such as the event-based calcium event detection software AQuA (DOI: 10.1038/s41593-019-0492-2), is insufficiently discussed and may bias the analysis. Some references on this matter include: Semyanov et al, Nature Rev Neuro, 2020 (DOI: 10.1038/s41583-020-0361-8); Covelo et al 2022, J Mol Neurosci (DOI: 10.1007/s12031-022-02006-w) & Wang et al, 2019, Nat Neuroscience (DOI: 10.1038/s41593-019-0492-2). Moreover, the ROIs used to quantify calcium activity are based on structural imaging of astrocytes, which may not be functionally relevant.

For the reasons listed above, the manuscript would probably benefit from some rephrasing of the conclusions and a discussion highlighting the advantages and limitations of the methodological approach. The question investigated by this study is of great importance in the field of neuroscience as the mechanisms dictating the spatio-temporal properties of calcium signals in astrocytes are poorly characterized, yet are essential to understand their involvement in the modulation of signal integration within neural circuits.

Reviewer #3 (Public Review):

Summary:<br /> The study aims to elucidate the spatial dynamics of subcellular astrocytic calcium signaling. Specifically, they elucidate how subdomain activity above a certain spatial threshold (~23% of domains being active) heralds a calcium surge that also affects the astrocytic soma. Moreover, they demonstrate that processes on average are included earlier than the soma and that IP3R2 is necessary for calcium surges to occur. Finally, they associate calcium surges with slow inward currents.

Strengths:<br /> The study addresses an interesting topic that is only partially understood. The study uses multiple methods including in vivo two-photon microscopy, acute brain slices, electrophysiology, pharmacology, and knockout models. The conclusions are strengthened by the same findings in both in vivo anesthetized mice and in brain slices.

Weaknesses:<br /> The method that has been used to quantify astrocytic calcium signals only analyzes what seems to be a small proportion of the total astrocytic domain on the example micrographs, where a structure is visible in the SR101 channel (see for instance Reeves et al. J. Neurosci. 2011, demonstrating to what extent SR101 outlines an astrocyte). This would potentially heavily bias the results: from the example illustrations presented it is clear that the calcium increases in what is putatively the same astrocyte goes well beyond what is outlined with automatically placed small ROIs. The smallest astrocytic processes are an order of magnitude smaller than the resolution of optical imaging and would not be outlined by either SR101 or with the segmentation method judged by the ROIs presented in the figures. Completely ignoring these very large parts of the spatial domain of an astrocyte, in particular when making claims about a spatial threshold, seems inappropriate. Several recent methods published use pixel-by-pixel event-based approaches to define calcium signals. The data should have been analyzed using such a method within a complete astrocyte spatial domain in addition to the analyses presented. Also, the authors do not discuss how two-dimensional sampling of calcium signals from an astrocyte that has processes in three dimensions (see Bindocci et al, Science 2017) may affect the results: if subdomain activation is not homogeneously distributed in the three-dimensional space within the astrocyte territory, the assumptions and findings between a correlation between subdomain activation and somatic activation may be affected.

The experiments are performed either in anesthetized mice, or in slices. The study would have come across as much more solid and interesting if at least a small set of experiments were performed also in awake mice (for instance during spontaneous behavior), given the profound effect of anesthesia on astrocytic calcium signaling and the highly invasive nature of preparing acute brain slices. The authors mention the caveat of studying anesthetized mice but claim that the intracellular machinery should remain the same. This explanation appears a bit dismissive as the response of an astrocyte not only depends on the internal machinery of the astrocyte, but also on how the astrocyte is stimulated: for instance synaptic stimulation or sensory input likely would be dependent on brain state and concurrent neuromodulatory signaling which is absent in both experimental paradigms. The discussion would have been more balanced if these aspects were dealt with more thoroughly.

The study uses a heaviside step function to define a spatial 'threshold' for somata either being included or not in a calcium signal. However, Fig 4E and 5D showing how the method separates the signal provide little understanding for the reader. The most informative figure that could support the main finding of the study, namely a ~23% spatial threshold for astrocyte calcium surges reaching the soma, is Fig. 4G, showing the relationship between the percentage of arborizations active and the soma calcium signal. A similar plot should have been presented in Fig 5 as well. Looking at this distribution, though, it is not clear why ~23% would be a clear threshold to separate soma involvement, one can only speculate how the threshold for a soma event would influence this number. Even if the analyses in Fig. 4H and the fact that the same threshold appears in two experimental paradigms strengthen the case, the results would have been more convincing if several types of statistical modeling describing the continuous distribution of values presented in Fig. 4E (in addition to the heaviside step function) were presented.

The description of methods should have been considerably more thorough throughout. For instance which temperature the acute slice experiments were performed at, and whether slices were prepared in ice-cold solution, are crucial to know as these parameters heavily influence both astrocyte morphology and signaling. Moreover, no monitoring of physiological parameters (oxygen level, CO2, arterial blood gas analyses, temperature etc) of the in vivo anesthetized mice is mentioned. These aspects are critical to control for when working with acute in vivo two-photon microscopy of mice; the physiological parameters rapidly decay within a few hours with anesthesia and following surgery.

Author Response

We are grateful to the reviewers for recognizing the importance of our work on transcription-independent early recovery of proteasome activity. We also thank them for their thoughtful criticisms and suggested improvements, which we will address in the revised version as described below.

The reviewers and editors asked for data to support the model that early recovery of proteasome activity is due to accelerated proteasome assembly. This model is backed by published data that proteasome assembly intermediates increase dramatically in cells treated with proteasome inhibitors (Fig. 6 in Ref. 46 of the revised manuscript). We will expand the discussion of this paper in a paragraph that describes our model. Another key experiment to confirm this model would be to determine what fraction of nascent polypeptides is degraded within minutes after synthesis, which is not trivial, and Ibtisam ran out of time to conduct these experiments because she had to graduate in spring before the expiration of her visa. This type of experiment usually uses metabolic labeling by a heavy or radioactive amino acid that always includes a prior depletion of a non-labeled amino acid. However, the fundamental flaw of this approach, which is not recognized by the scientific community, is that depletion of an amino acid stresses cells and reduces the rate of protein synthesis, especially if this amino acid is methionine. Thus, this model is not easily to test, and should be considered a speculation. We will therefore move the description of this model, together with Fig. 4, into a separate "Ideas and Speculation" section and remove this model's description from the abstract.

Reviewer 1 raised the possibility that a background band detected on the western blot of DDI2 KO cells could be a highly homologous protease DDI1. This is highly unlikely because, according to Protein Atlas, DDI1 is selectively expressed in the testis and is not expressed in the cell lines we used. Reviewer 1 also suggested that we should base our conclusion on Nrf1 KD, which we de-facto did because we confirmed that DDI2 KD blocks Nrf1 activation (Fig. 1d).

In response to Reviewer 1 critiques regarding the presentation of proteasome subunits stability data in Fig. 4 (Ref. 45 of the revised manusript), we will remove PSMB8 and replace chaperons with the subunits of the 26S base. We will change color palettes, symbols, and axis scales to improve clarity.

We will acknowledge in the discussion that our work did not exclude DDI2 role in the recovery of proteasome after repeated pulse treatments, as suggested by Reviewer 1.

We agree with Reviewer 2 that using proteasome levels is inaccurate when describing our activity measurement data. However, in the manuscript, we use "levels" only when discussing data in the literature. We believe measuring activity and not the total levels is more important because not all proteasomes are active, e.g., latent 20S proteasome core particles.

Reviewer 3 expressed concern that our conclusions were based on data in HAP1 cells, which are haploid, and appear not very sensitive to proteasome inhibitors. This is why we used DDI2 KD in MDA-MB-231 and SUM149 cells, which are highly sensitive to proteasome inhibitors (Weyburne et al., Ref. 11). In our experience, full extent of proteasome inhibitor cytotoxicity is not revealed until 48hr after treatments, and viability determined at 12hr and 24hr as on Fig. 1c should not be used to determine sesnsitivity (it was used for activity assay normalization). We will add a new supplementary figure showing that HAP1 cells are as sensitive to proteasome inhibitors as MDA-MD-231 cells when cell viability is assayed 48hr after treatment (new Fig. S2). Another panel on this new figure will demonstrate that the baseline proteasome activity is very similar in HAP1, MD-MB-231 and SUM149 cells. We will also add data demonstrating that inactivativion of DDI2 by mutation does not change the recovery of proteasome activity in HCT-116 cells (new Fig. 1g). Recovery in MDA-MB-231, SUM149, and HCT-116 cells was measured at 18hr, which is still within the 12 – 24hr window when other investigators observed partially DDI2-dependent recovery.

We have conducted an experiment in which we followed activity recovery for up to 72hr. We found that activity plateaued at 24hr and opted against the repeat because there were no changes. We feel that the manuscript should not include one biological replicate data. The fact that the recovery is incomplete and that cells seem to survive with lower levels of proteasome activity is interesting; however, investigating the molecular basis for this phenomenon is beyond the scope of the current project.

We were not disputing the conclusions of previous studies that DDI2/Nrf1 is responsible for enhanced expression of proteasomal mRNA in cells continuously treated with proteasome inhibitors. In fact, we confirmed that pulse-treatment causes similar increase (Fig. 2b). As for papers that measured activity recovery after pulse treatment, we objectively discuss our results in the context of these papers.

We will also respond to Reviewers' recommendations and minor points:

• We will review the revised version carefully to eliminate spelling and grammatical errors and typos.

• We will no longer refer to DDI2 as a novel protease, as suggested by Reviewer 1.

• We agree with Reviewer 2 that our CHX results do not necessarily mean that recovery involves translation of proteasomal mRNAs, and we will now conclude that proteasome recovery requires protein synthesis.

• We will revise Fig. 1c, 3a and 4a to improve clarity.

• We have stated in the caption that data in Fig. 4a comes from Table S4 in Ref. 45.

• We will accept an excellent suggestion of Reviewer 3 to change "recovery" to "early recovery" in the title.

• Regarding Reviewer 3 request to assay activity recovery at additional time points before 12hr, this was done in the cycloheximide experiment in Fig. 3A.

• Even if we assume that the differences in the observed recovery activity in MDA-MB-231 cells (Fig. 1f) are statistically significant, which may implicate DDI2 involvement in the activity recovery, the percentage is still small, suggesting that most activity recovery is DDI2-independent.

• We will tone down the statement "the present findings suggest that DDI2 desensitizes cells to PI by a different mechanism," replacing "suggest" with "raise a possibility."

• We will indicate that only Bortezomib is approved for mantle cell lymphoma.

• We will change the description of clinical dosing as suggested by Reviewer 3. We will add a reference on PK of subcutaneous bortezomib (Ref. 9), even though the review we quoted (Ref. 7) discussed subcutaneous dosing.

eLife assessment

This study presents important findings on a transcription-independent component of the early recovery of proteasome activity from a short pulse of proteasome inhibitor treatment. While the evidence supporting this conclusion is solid, experimental support forthe proposal of an alternative regulatory process operative at the level of proteasome assembly, is incomplete. Lacking is experimental quantification of the effect of proteasome inhibition on the rate at which newly synthesized subunits are assembled into proteasomes.

Reviewer #1 (Public Review):

Summary:<br /> There has been substantial prior work trying to understand the transcriptional control of proteasome expression as an adaptive response to proteasome inhibition. This field has been mired by fierce debates over the role of the protease Ddi2 in activating the transcription factor Nrf1/NFE2L1. As the authors of this manuscript point out, most of the previous research centers on the continuous treatment of cells with proteasome inhibitors rather than a brief pulse of inhibition that better models the situation when these drugs are used clinically. The authors find that the initial recovery of proteasome activity is independent of Ddi2 and involves a mechanism distinct from transcription. The authors intriguingly point to a model in which the assembly of proteasomes is regulated. If true, this would be a significant finding, but for now, this model remains more speculative.

Strengths:<br /> The pulsed treatment of proteasome inhibitors is a strength of this lab that few others use. It better mimics the clinical use of these inhibitors and allows for a more detailed analysis of the initial response to inhibition. The authors have used multiple different clones of Ddi2 knockouts and siRNA against Ddi2 to rule out the necessity of Ddi2 in the early production of proteasomes when cells are inhibited with proteasomes. establishing a thorough knockout approach while also avoiding compensatory mutations. These experiments are well controlled showing both the levels of Ddi2 upon knockout or knockdown and demonstration that cleavage of Nrf1, one of two known targets of Ddi2, is impaired. However, it should be noted that even in the knockout residual bands for Ddi2 remain. Since these HAP1 cells only have one copy of the Ddi2 gene, it is possible that this other band could be Ddi1, a very similar paralogue. If so the conclusions of Ddi2-like activity with Ddi1 must be tempered and rely more on the data with Nrf1 knockdowns.

This article sensitively monitors the recovery of proteasome function with the β5 activity assay and for the production of new proteasome transcripts by Q-PCR. This precision coupled with detailed analysis of the timing are strengths that pointed to a more rapid recovery than transcription alone.

Weaknesses:<br /> This paper's major weakness is the difficulty in establishing the authors' model that assembly is regulating this process. They do a convincing job demonstrating that activity recovers before transcription. The evidence that translation is not affected depends entirely on the polysome RNA profiling from two replicates. Clearer and orthogonal data would help establish this finding. The stability of subunits is interesting and important in its own right. However, the clustering of proteins is somewhat unusual. The authors include PSMB8, an immuno-proteasome subunit that is not regulated by Nrf1. The proteins highlighted in green are an unusual assortment of alternative activators (PSME1-3), a ubiquitin-binding protein (ADRM1), and proteasome chaperones (PSMG1-2). Similarly, the purple proteins are not just proteins in the 19S regulatory particle but also assembly chaperones. However, these labeling issues do not detract from the conclusions of this figure.

In short, the authors establish that Ddi2 is not necessary for the initial, non-transcriptional, recovery of proteasome activity after a pulse of proteasome inhibition.

It is not clear what clinical impact this work will have. Although it models the pulse of proteasome inhibition more perfectly, it only looks at a single pulse rather than multiple treatments. Thus, ruling Ddi2's importance out for clinical benefit may be premature. More significantly this work suggests that the assembly of proteasomes might be a regulated process worth substantial follow-up that will be interesting to follow.

Reviewer #2 (Public Review):

Summary:<br /> In this work, Ibtisam and Kisselev explore the role of DDI2 in proteasome function recovery after a clinically relevant pulse dosing using different proteasome inhibitors and their corresponding PK properties. The authors report that despite the lack of NRF1 activation by DDI2 there was no difference in recovery from pulsed proteasome inhibition observed in DDI2 KO cells as compared to WT controls suggesting that DDI2 is not required for recovery in this system. They further show that transcription of the proteasome subunits is initiated only after partial recovery of proteasome activity is already observed suggesting that non-transcriptional mechanisms might be also involved. The authors further show that translation inhibition blocked the recovery from proteasome inhibitors.

Strengths:<br /> Overall, it is very important and informative to use a pulse treatment type approach (mimicking the PK properties of the drugs) to explore the biology of PIs as used in this study. The authors also provide convincing data that DDI2 is not required for proteasome activity recovery post-PI pulse treatment in the systems they explored.

Weaknesses:<br /> Many of the other conclusions are not supported by the data in the current form of the manuscript and are too speculative and ignore the major findings in the field that can present alternative mechanisms. In particular, the authors discuss the "levels" of the proteasomes post-PI treatment without measuring the actual protein level of the individual subunits or the different assembled proteasome complexes.

Reviewer #3 (Public Review):

Summary:<br /> In their manuscript "Recovery of proteasome activity in cells pulse-treated with proteasome inhibitors is independent of DDI2", Ibtisam and Kisselev investigate proteasome recovery in HAP1 cells either WT or DDI2 KO upon inhibition of proteasome via bortezomib or carfilzomib. The authors argue that proteasome recovery is independent of DDI2 as it is independent of the novo proteasome subunit synthesis. They argue recovery is dependent on the assembly of already synthesized proteasome subunits.

Strengths:<br /> The findings are important as they provide insight into a transcriptionally-independent proteasome stress recovery that is likely applicable across distinct cellular subtypes. Comparable proteasome recovery early on (<12 hours) from proteasomal inhibition in DDI2 KO cell lines was already noted in other manuscripts, including Chen et al, suggesting that this phenomenon is applicable to other histotypes.

Weaknesses:<br /> Some of the conclusions are not adequately supported by the data and how generalizable these findings are is unclear. In particular, there is concern regarding the status of the ubiqutin-proteasome-system in the HAP1 cell line that was used for these studies. In a previously published model system, a dependency on DDI2 and NRF1 was clearly demonstrated and this pathway was critical for late (12-24 hours) proteasome recovery as well as cell viability. The model system used here (HAP1 cells) seems completely independent of DDI2 both for proteasome recovery and viability as curves are substantially overlapping. It would be important to assess how the baseline proteasome activity in HAP1 cells compare to other cell lines and model system as these cells may be largely independent of proteasome degradation and their synthetic load on the pathway very modest.

It would also be relevant to look at later time points of proteasome recovery as one would expect DDI2 to play a role later on in the recovery of proteasome. the authors may have missed that time point as cells do not appear to recover close to 100% proteasome activity by 24 hours not even when the smallest concentration of carfilzomib is used.

A critical experiment to look at de novo proteasome assembly was not carried out, leaving the data hypothetical.

Finally, the authors leverage HAP1 cells for their work and should be mindful of not generalizing findings or disputing other author's conclusions in the absence of adequate experiments to support their hypothesis.

eLife assessment

This manuscript used the sci-Plex system for screening compounds to improve the Ascl1-induced reprogramming from Müller glia to bipolar neurons in vitro, followed by in vivo characterization of two promising compounds in mice. The findings are valuable for future studies to develop cell replacement strategies for treatment of retinal degeneration. The strength of evidence is solid, featuring a scalable drug screening design, albeit with limited mechanistic insights.

Reviewer #1 (Public Review):

Summary:<br /> The study used the sci-Plex system to perform in vitro screen of chemicals and found that 2 compounds improved the reprogramming efficiency in Ascl1-overexpressed MG (Muller glia), and in addition, administration of the identified compounds in the previously established in vivo model (Ascl1, NMDA, TSA) showed that DBZ and metformin increased Otx2+ cells for improved neurogenesis.

Strengths: The overall study was straightforward and well designed. The method in the study could be potentially useful for large-scale in vitro screens for compounds to further improve reprogramming efficiency. The data and results of the study are of good quality.

Weaknesses: The findings may not generate significant interest for two main reasons. One, the compounds only increased the population of bipolar neurons but did not generate new retinal neuronal types compared to the earlier methods, and the reprogramming efficiency may not be as high as other earlier strategies such as overexpression of Ascl1 plus Atoh1 reported from the same group. Two, the overall study produced some interesting initial discoveries but was quite descriptive overall, was weak on performing more in-depth analysis and weak on mechanistic examinations.

Reviewer #2 (Public Review):

Summary:

In the current manuscript, Tresenrider et al., present their recent study focusing on screening of small molecules to enhance the conversion from Müller cells (MG) to retina neurons induced by ectopic Ascl1 expression.

Strengths:

To analyze results from multiple treatment conditions in a single experiment, the authors employed a method called sci-Plex to perform scRNA-seq on mixed samples to investigate the effects of different durations of Ascl1 expression and screen for potential small molecules to promote reprogramming. Ultimately, they identified two compounds with intended activities on mouse retina. The findings may aid in future development of a cell replacement strategy for treating retinal degeneration.

Weaknesses:

The mechanistic insights are limited. Certain claims are confusing or superficial at this point, as detailed in issues/concerns.

eLife assessment

This study shows that unlike prior reports, cortical spreading depression does not lead to spontaneous activity in the majority of meningeal afferent sensory neurons but that it increases sensitivity to mechanical deformation of the meninges. This has important implications for headache disorders like migraine where cortical spreading depression is thought to contribute to the pathology, but how this occurs is unclear. The current work uses convincing methods in awake and freely moving animals compared to prior studies of this nature that used recordings in anesthetized animals.

Reviewer #1 (Public Review):

Summary:<br /> Herein, Blaeser et al. explored the impact of migraine-related cortical spreading depression (CSD) on the calcium dynamics of meningeal afferents that are considered the putative source of migraine-related pain. Critically previous studies have identified widespread activation of these meningeal afferents following CSD; however, most studies of this kind have been performed in anesthetized rodents. By conducting a series of technically challenging calcium imaging experiments in conscious head fixed mice they find in contrast that a much smaller proportion of meningeal afferents are persistently activated following CSD. Instead, they identify that post-CSD responses are differentially altered across a wide array of afferents, including increased and decreased responses to mechanical meningeal deformations and activation of previously non-responsive afferents following CSD. Given that migraine is characterized by worsening head pain in response to movement, the findings offer a potential mechanism that may explain this clinical phenomenon.

Strengths:<br /> Using head fixed conscious mice overcomes the limitations of anesthetized preps and the potential impact of anaesthesia on meningeal afferent function which facilitated novel results when compared to previous anesthetized studies. Further, the authors used a closed cranial window preparation to maximize normal physiological states during recording, although the introduction of a needle prick to induce CSD will have generated a small opening in the cranial preparation, rendering it not fully closed as suggested.

Weaknesses:<br /> Although this is a well conducted technically challenging study that has added valuable knowledge on the response of meningeal afferents the study would have benefited from the inclusion of more female mice. Migraine is a female dominant condition and an attempt to compare potential sex-differences in afferent responses would undoubtedly have improved the outcome.

The authors imply that the current method shows clear differences when compared to older anaesthetized studies; however, many of these were conducted in rats and relied on recording from the trigeminal ganglion. Inclusion of a subgroup of anesthetized mice in the current preparation may have helped to answer these outstanding questions, being is this species dependent or as a result of the different technical approaches.

The authors discuss meningeal deformations as a result of locomotion; however, despite referring to their previous work (Blaeser et al., 2022), the exact method of how these deformations were measured could be clearer. It is challenging to imaging that simple locomotion would induce such deformations and the one reference in the introduction refers to straining, such as cough that may induce intracranial hypertension, which is likely a more powerful stimulus than locomotion.

Reviewer #2 (Public Review):

This is an interesting study examining the question of whether CSD sensitizes meningeal afferent sensory neurons leading to spontaneous activity or whether CSD sensitizes these neurons to mechanical stimulation related to locomotion. Using two-photon in vivo calcium imaging based on viral expression of GCaMP6 in the TG, awake mice on a running wheel were imaged following CSD induction by cortical pinprick. The CSD wave evoked a rise in intracellular calcium in many sensory neurons during the propagation of the wave but several patterns of afferent activity developed after the CSD. The minority of recorded neurons (10%) showed spontaneous activity while slightly larger numbers (20%) showed depression of activity, the latter pattern developed earlier than the former. The vast majority of neurons (70%) were unaffected by the CSD. CSD decreased the time spent running and the numbers of bouts per minute but each bout was unaffected by CSD. There also was no influence of CSD on the parameters referred to as meningeal deformation including scale, shear, and Z-shift. Using GLM, the authors then determine that there there is an increase in locomotion/deformation-related afferent activity in 51% of neurons, a decrease in 12% of neurons, and no change in 37%. GLM coefficients were increased for deformation related activity but not locomotion related activity after CSD. There also was an increase in afferents responsive to locomotion/deformation following CSD that were previously silent. This study shows that unlike prior reports, CSD does not lead to spontaneous activity in the majority of sensory neurons but that it increases sensitivity to mechanical deformation of the meninges. This has important implications for headache disorders like migraine where CSD is thought to contribute to the pathology in unclear ways with this new study suggesting that it may lead to increased mechanical sensitivity characteristic of migraine attacks.

Reviewer #3 (Public Review):

Summary:<br /> Blaeser et al. set out to explore the link between CSD and headache pain. How does an electrochemical wave in the brain parenchyma, which lacks nociceptors, result in pain and allodynia in the V1-3 distribution? Prior work had established that CSD increased the firing rate of trigeminal neurons, measured electrophysiologically at the level of the peripheral ganglion. Here, Blaeser et al. focus on the fine afferent processes of the trigeminal neurons, resolving Ca2+ activity of individual fibers within the meninges. To accomplish these experiments, the authors injected AAV encoding the Ca2+ sensitive fluorophore GCamp6s into the trigeminal ganglion, and 8 weeks later imaged fluorescence signals from the afferent terminals within the meninges through a closed cranial window. They captured activity patterns at rest, with locomotion, and in response to CSD. They found that mechanical forces due to meningeal deformations during locomotion (shearing, scaling, and Z-shifts) drove non-spreading Ca2+ signals throughout the imaging field, whereas CSD caused propagating Ca2+ signals in the trigeminal afferent fibers, moving at the expected speed of CSD (3.8 mm/min). Following CSD, there were variable changes in basal GCamp6s signals: these signals decreased in the majority of fibers, signals increased (after a 25 min delay) in other fibers, and signals remained unchanged in the remainder of fibers. Bouts of locomotion were less frequent following CSD, but when they did occur, they elicited more robust GCamp6s signals than pre-CSD. These findings advance the field, suggesting that headache pain following CSD can be explained on the basis of peripheral cranial nerve activity, without invoking central sensitization at the brain stem/thalamic level. This insight could open new pathways for targeting the parenchymal-meningeal interface to develop novel abortive or preventive migraine treatments.

Strengths:<br /> The manuscript is well-written. The studies are broadly relevant to neuroscientists and physiologists, as well as neurologists, pain clinicians, and patients with migraine with aura and acephalgic migraine. The studies are well-conceived and appear to be technically well-executed.

Weaknesses:<br /> 1) Lack of anatomic confirmation that the dura were intact in these studies: it is notoriously challenging to create a cranial window in mouse skull without disrupting or even removing the dura. It was unclear which meningeal layers were captured in the imaging plane. Did the visualized trigeminal afferents terminate in the dura, subarachnoid space, or pia (as suggested by Supplemental Fig 1, capturing a pial artery in the imaging plane)? Were z-stacks obtained, to maintain the imaging plane, or to follow visualized afferents when they migrated out of the imaging plane during meningeal deformations?<br /> 2) Findings here, from mice with chronic closed cranial windows, failed to fully replicate prior findings from rats with acute open cranial windows. While the species, differing levels of inflammation and intracranial pressure in these two preparations may contribute, as the authors suggested, the modality of measuring neuronal activity could also contribute to the discrepancy. In the present study, conclusions are based entirely on fluorescence signals from GCamp6s, whereas prior rat studies relied upon multiunit recordings/local field potentials from tungsten electrodes inserted in the trigeminal ganglion. As a family, GCamp6 fluorophores are strongly pH dependent, with decreased signal at acidic pH values (at matched Ca2+ concentration). CSD induces an impressive acidosis transient, at least in the brain parenchyma, so one wonders whether the suppression of activity reported in the wake of CSD (Figure 2) in fact reflects decreased sensitivity of the GCamp6 reporter, rather than decreased activity in the fibers. If intracellular pH in trigeminal afferent fibers acidifies in the wake of CSD, GCamp6s fluorescence may underestimate the actual neuronal activity.

eLife assessment

The findings of this study are valuable as they provide new insights into the role of acetylcholine in modulating sensory processing in the auditory cortex. This paper reports a systematic measurement of cell activity in the auditory cortex before and after applying ACh during an oddball and cascade sequence of auditory stimuli in anesthetized rats. The results presented are solid given the rigorous experimental design and statistical analysis. The conclusions are provocative and will interest researchers in auditory neuroscience and neuromodulation, as well as clinicians and individuals with auditory processing disorders. However, the findings support multiple interpretations, beyond that offered by the authors.

Reviewer #1 (Public Review):

Summary:<br /> This study examined the impact of exogenous microapplication of acetylcholine (Ach) on metrics of novelty detection in the anesthetized rat auditory cortex. The authors found that the majority of units showed some degree of modulation of novelty detection, with roughly similar numbers showing enhanced novelty detection, suppressed novelty detection, or no change. Enhanced novelty responses were driven by increases in repetition suppression. Suppressed novelty responses were driven by deviance suppression. There were no compelling differences seen between auditory cortical subfields or layers, though there was heterogeneity in the Ach effects within subfields. Overall, these findings are important because they suggest that fluctuations in cortical Ach, which are known to occur during changes in arousal or attentional states, will likely influence the capacity of individual auditory cortical neurons to respond to novel stimuli.

Strengths:<br /> The work addresses an important problem in auditory neuroscience. The main strengths of the study are that the work was systematically done with appropriate controls (cascaded stimuli) and utilizes a classical approach that ensures that drug application is isolated to the micro-environment of the recorded neuron. In addition, the authors do not isolate their study to only the primary auditory cortex, but examine the impact of Ach across all known auditory cortical subfields.

Weaknesses:<br /> 1. As acknowledged by the authors, this study explicitly examines a phenomenon of high relevance to active listening but is done in anesthetized animals, limiting its applicability to the waking state.<br /> 2. The authors do not make any attempt to determine, by spike shape/duration, if their units are excitatory or inhibitory, which may explain some of the variance of the data.<br /> 3. The application of exogenous Ach, potentially in supra-physiological amounts, makes this study hard to extrapolate to a behaving animal. A more compelling design would be to block Ach, particularly at particular receptor types, to determine the effect of endogenous Ach.

Reviewer #2 (Public Review):

Summary:<br /> In this study, the authors investigate the effect of ACh on neuronal responses in the auditory cortex of anesthetized rats during an auditory oddball task. The paradigm consisted of two pure tones (selected from the frequency responses at each recording site) presented in a pseudo-random sequence. One tone was presented frequently (the "standard" tone) and the other infrequently (the "deviant" tone). The authors found that ACh enhances the detection of unexpected stimuli in the auditory environment by increasing or decreasing the neuronal responses to deviant and standard tones.

Strengths:<br /> The study includes the use of appropriate and validated methodology in line with the current state-of-the-art, rigorous statistical analysis, and the demonstration of the effects of acetylcholine on auditory processing.

Weaknesses:<br /> The study was conducted in anesthetized rats, and further research is needed to determine the behavioral relevance of these findings.

eLife assessment

This manuscript reports a useful computational study of the effects of axon de-myelination and re-myelination on models of working memory, with potential applications implications in disorders such as multiple sclerosis. In its present form, the provided evidence is partly incomplete due to certain modeling choices and lack of clarity on model details, but these shortcomings could be addressed.

Reviewer #1 (Public Review):

Summary: The authors study the effects of myelin alterations in working memory via the complementary use of two computational approaches: one based on the de- and re-myelination in multicompartmental models of pyramidal neurons, and one based on synaptic changes in a spiking bump attractor model for spatial working memory. The first model provides the most precise angle (biophysically speaking) of the different effects (loss of myelin lamella or segments, remyelination with thinner and shorter nodes, etc), while the second model allows to infer the consequences of myelin alterations in working memory performance, including memory stability, duration, and bump diffusion. The results indicate (i) a slowing down and failure of propagation of spikes with demyelination and partial recovery with remyelination, with detailed predictions on the role of nodes and myelina lamella, and (ii) a decrease in memory duration and an increase in memory drift as a function of the demyelination, in agreement with multiple experimental studies.

Strengths: Overall, the work offers a very interesting approach of a topic which is hard to accomplish experimentally --therefore the computational take is entirely justified and extremely useful. The authors carefully designed the computational experiments to shed light into the demyelination effects on working memory from multiple levels of description, increasing the reliability of their conclusions. I think this work is solid and has the potential to be influential in future studies of myelin alterations (and related disorders such as multiple sclerosis).

Weaknesses: In its current form, the study still presents several issues which prevent it from achieving a higher potential impact. These can be summarized in two main items. First, the manuscript is missing some important details about how demyelination and remyelination are incorporated in both models (and what is the connection between both implementations). For example, it is unclear whether an unperturbed axon and a fully remyelinated axon would be mathematically equivalent in the multicompartment model, or how the changes in the number of nodes, myelin lamella, etc, are implemented in the spiking neural network model. Second, it is unclear whether some of the conclusions are strong computational predictions or just a consequence of the model chosen. For example, the lack of effect of decreasing the conduction velocity on working memory performance could be due to the choice of considering a certain type of working memory model (continuous attractor), and therefore be absent under other valid assumptions (i.e. a silent working memory model, which has a higher dependence on temporal synaptic dynamics).

With additional simulations to address these issues, I consider that the present study would become a convincing milestone in the computational modeling of myelin-related models, and an important study in the field of working memory.

Reviewer #2 (Public Review):

This paper analyzes the effect of axon de-myelination and re-myelination on action potential speed, and propagation failure. Next, the findings are then incorporated in a standard spiking ring attractor model of working memory.

I think the results are not very surprising or solid and there are issues with method and presentation.<br /> The authors did many simulations with random parameters, then averaged the result, and found for instance that the Conduction Velocity drops in demyelination. It gives the reader little insight into what is really going on. My personal preference is for a well understood simple model rather than a poorly understood complex model. The link between the model outcome of WM and data remains qualitative, and is further weakened by the existence of known other age-related effects in PFC circuits.

* Both for the de/re myelination the spatial patterns are fully random. Why is this justified?<br /> * Similarly, to model the myelin parameters where drawn from uniform distributions, Table 1 (I guess). Again, why is this reasonable?

* The focus of most analysis is on the conduction velocity but in the end, this has no effect on WM, so the discussion of CV remains sterile.

* The more important effect of de/re myelination is on failure.<br /> However, the failure is, AFAIK, just characterized by a constant current injection of 380pA.<br /> From Fig 2 it seems however that the first spike is particularly susceptible to failure.<br /> In other words, it has not been justified that it is fine to use the failure rates from this artificial protocol in the I&F model. I would expect the temporal current trace to affect whether the propagation fails or not.<br /> I don't know if there are many axon-collaterals in the WM circuits and or distance dependence in the connectivity, but if so, then the current implementation of failure would be questionable.<br /> I would also advise against thresholding at 75% failure in Fig3C. Why don't the authors not simply plot the failure rate?

Regarding the presentation, there are a number of dead-end results that are not used further on. The paper is rather extensive, and it would be clearer if written up in half the space. In addition, much information is really supplementary. The issue of the CV I already mentioned, also the Lasso regression for instance remains unused.

eLife assessment

Schnell et al report important differences between the strategies used by rodents and humans when discriminating different visual objects. The evidence supporting these findings is convincing, showing that rat performance was influenced far more by low-level cues compared to humans. It is, however, unclear to what extent these differences can be explained by the lower visual acuity of rats. This work will be of general interest to vision and cognition researchers, particularly those studying object vision.

Reviewer #1 (Public Review):

Schnell et al. performed two extensive behavioral experiments concerning the processing of objects in rats and humans. To this aim, they designed a set of objects parametrically varying along alignment and concavity and then they used activations from a pretrained deep convolutional neural network to select stimuli that would require one of two different discrimination strategies, i.e. relying on either low- or high-level processing exclusively. The results show that rodents rely more on low-level processing than humans.

Strengths:

1. The results are challenging and call for a different interpretation of previous evidence. Indeed, this work shows that common assumptions about task complexity and visual processing are probably biased by our personal intuitions and are not equivalent in rodents, which instead tend to rely more on low-level properties.<br /> 2. This is an innovative (and assumption-free) approach that will prove useful to many visual neuroscientists. Personally, I second the authors' excitement about the proposed approach, and its potential to overcome the limits of experimenters' creativity and intuitions. In general, the claims seem well supported and the effects sufficiently clear.<br /> 3. This work provides an insightful link between rodent and human literature on object processing. Given the increasing number of studies on visual perception involving rodents, these kinds of comparisons are becoming crucial.<br /> 4. The paper raises several novel questions that will prompt more research in this direction.

Weaknesses:<br /> 1. The choice of alignment and concavity as baseline properties of the stimuli is not properly discussed.<br /> 2. From the low-correlations I got the feeling that AlexNet is not the best baseline model for rat visual processing.

Reviewer #2 (Public Review):

Schnell and colleagues trained rats on a two-alternative forced choice visual discrimination task. They used object pairs that differed in their concavity and the alignment of features. They found that rats could discriminate objects across various image transformations. Rat performance correlated best with late convolutional layers of an artificial neural network and was partially explained by factors of brightness and pixel-level similarity. In contrast, human performance showed the strongest correlation with higher, fully connected layers, indicating that rats employed simpler strategies to accomplish this task as compared to humans.

Strengths:<br /> 1. This is a methodologically rigorous study. The authors tested a substantial number of rats across a large variety of stimuli.<br /> 2. The innovative use of neural networks to generate stimuli with varying levels of complexity is a compelling approach that motivates principled experimental design.<br /> 3. The study provides important data points for cross-species comparisons of object discrimination behavior<br /> 4. The data strongly support the authors' conclusion that rats and humans rely on different visual features for discrimination tasks.<br /> 5. This is a valuable study that provides novel, important insights into the visual capabilities of rats.

Weaknesses:<br /> 1. The impact of rat visual acuity (~1cycle/degree) on the discriminability of stimuli could be more directly modeled and taken into consideration when comparing rat behavior to humans, who possess substantially higher acuity.<br /> 2. The distinction between low- and high-level visual behavior is coarse, and it remains uncertain which specific features rats utilized for discrimination. The correlations with brightness and pixel-level similarity do provide some insight.<br /> 3. The relatively weak correspondence between rat behavior and AlexNet raises the question of which network architecture, whether computational or biological, might better capture rat behavior, particularly to the level of cross-rat consistency.

eLife assessment

This study provides a valuable contribution to the study of eye-movements in reading, revealing that attention-weights from a deep neural network show a statistically reliable fit to the word-level reading patterns of humans. Its evidence is convincing and strengthens a line of research arguing that attention in reading reflects task optimization. The work would be of interest to psychologists, neuroscientists, and machine learning researchers.

Reviewer #1 (Public Review):

This manuscript describes a set of four passage-reading experiments which are paired with computational modeling to evaluate how task-optimization might modulate attention during reading. Broadly, participants show faster reading and modulated eye-movement patterns of short passages when given a preview of a question they will be asked. The attention weights of a Transformer-based neural network (BERT and variants) show a statistically reliable fit to these reading patterns above-and-beyond text- and semantic-similarity baseline metrics, as well as a recurrent-network-based baseline. Reading strategies are modulated when questions are not previewed, and when participants are L1 versus L2 readers, and these patterns are also statistically tracked by the same transformer-based network.

Strengths:

- Task-optimization is a key notion in current models of reading and the current effort provides a computationally rigorous account of how such task effects might be modeled<br /> - Multiple experiments provide reasonable effort towards generalization across readers and different reading scenarios<br /> - Use of RNN-based baseline, text-based features, and semantic features provides a useful baseline for comparing Transformer-based models like BERT

Weaknesses:

- Generalization across neural network models may be limited (models differ in size, training data etc.); it is thus not always clear which specific model characteristics support their fit to human reading patterns.

Reviewer #2 (Public Review):

In this study, researchers aim to understand the computational principles behind attention allocation in goal-directed reading tasks. They explore how deep neural networks (DNNs) optimized for reading tasks can predict reading time and attention distribution. The findings show that attention weights in transformer-based DNNs predict reading time for each word. Eye tracking reveals that readers focus on basic text features and question-relevant information during initial reading and rereading, respectively. Attention weights in shallow and deep DNN layers are separately influenced by text features and question relevance. Additionally, when readers read without a specific question in mind, DNNs optimized for word prediction tasks can predict their reading time. Based on these findings, the authors suggests that attention in real-world reading can be understood as a result of task optimization.

Strengths of the Methods and Results:<br /> The present study employed stimuli consisting of paragraphs read by middle and high school students, covering a wide range of diverse topics. This choice ensured that the reading experience for participants remained natural, ultimately enhancing the ecological validity of the findings and conclusions.

In Experiments 1-3, participants were instructed to read questions before the text, while in Experiment 4 participants were instructed to read questions after the text. This deliberate manipulation allowed the paper to assess how different reading task conditions influence reading and eye movements.

Weaknesses of the Methods and Results:

While the study benefits from several strengths, it is important to acknowledge its limitations. Notably, recent months have seen significant advancements in Deep Neural Network (DNN) models, including the development of models such as GPT-3.5 and GPT-4, which have demonstrated remarkable capabilities in tasks resembling human cognition, like Theory of Mind. However, as the code for these cutting-edge models was not publicly accessible, they were unable to evaluate whether the attention mechanisms in the most up-to-date DNN models could provide improved predictions for human eye-movement data. This constraint represents a limitation in the investigation.

The methods and data presented in this study are valuable for gaining insights into the psychological mechanisms of reading. Moreover, the data provided in this paper may prove instrumental in enhancing the performance of future DNN models.

Reviewer #3 (Public Review):

This paper presents several eyetracking experiments measuring task-directed reading behavior where subjects read texts and answered questions. It then models the measured reading times using attention patterns derived from deep-neural network models from the natural language processing literature. Results are taken to support the theoretical claim that human reading reflects task-optimized attention allocation.

Strengths:

(1) The paper leverages modern machine learning to model a high-level behavioral task (reading comprehension). While the claim that human attention reflects optimal behavior is not new, the paper considers a substantially more high-level task in comparison to prior work. The paper leverages recent models from the NLP literature which are known to provide strong performance on such question-answering tasks, and is methodologically well grounded in the NLP literature.

(2) The modeling uses text- and question-based features in addition to DNNs, specifically evaluates relevant effects, and compares vanilla pretrained and task-finetuned models. This makes the results more transparent and helps assess the contributions of task optimization. In particular, besides fine-tuned DNNs, the role of the task is further established by directly modeling the question relevance of each word. Specifically, the claim that human reading is predicted better by task-optimized attention distributions rests on (i) a role of question relevance in influencing reading in Expts 1-2 but not 4, and (ii) the fact that fine-tuned DNNs improve prediction of gaze in Expts 1-2 but not 4.

(3) The paper conducts experiments on both L2 and L1 speakers.

Weaknesses:

(1) Under the hypothesis advanced, human reading should adapt rationally to task demands. Indeed, Experiment 1 tests questions from different types in blocks (local and global), and the paper provides evidence that this encourages the development of question-type-specific reading strategies -- indeed, this specifically motivates Experiment 2, and is confirmed indirectly in the comparison of the effects found in the two experiments ("all these results indicated that the readers developed question-type-specific strategies in Experiment 1"). On the other hand, finetuning the model on one of the two types does not seem to reproduce this differential behavior, in the sense that fit to reading data is not improved. In this sense, the model seems to have limited abilities in reproducing the observed task dependence of human reading.

The results support the conclusions well, with the weakness described above a limitation of the modeling approach chosen.

The data are likely to be useful as a benchmark in further modeling of eye-movements, an area of interest to computational research on psycholinguistics.<br /> The modeling results contribute to theoretical understanding of human reading behavior, and strengthens a line of research arguing that it reflects task-adaptive behavior.

eLife assessment

The authors provide an important series of metabolic measurements characterizing group dynamics in fish, rationalizing that schooling behavior presents several benefits. The strength of evidence supporting this conclusion is solid, but the methodological and analytical approaches taken could be strengthened. Similarly, numerous test-based clarifications and improved statistical analyses could significantly strengthen the study and underlying interpretations.

Reviewer #1 (Public Review):

Summary:<br /> In the presented manuscript the authors aim at quantifying the costs of locomotion in schooling versus solitary fish across a considerable range of speeds. Specifically, they quantify the possible reduction in the cost of locomotion in fish due to schooling behavior. The main novelty appears to be the direct measurement of absolute swimming costs and total energy expenditure, including the anaerobic costs at higher swimming speeds.

In addition to metabolic parameters, the authors also recorded some basic kinematic parameters such as average distances or school elongation. They find both for solitary and schooling fish, similar optimal swimming speeds of around 1BL/s, and a significant reduction in costs of locomotion due to schooling at high speeds, in particular at ~5-8 BL/s.

Given the lack of experimental data and the direct measurements across a wide range of speeds comparing solitary and schooling fish, this appears indeed like a potentially important contribution of interest to a broader audience beyond the specific field of fish physiology, in particular for researchers working broadly on collective (fish) behavior.

Strengths:<br /> The manuscript is for the most part well written, and the figures are of good quality. The experimental method and protocols are very thorough and of high quality. The results are quite compelling and interesting. What is particularly interesting, in light of previous literature on the topic, is that the authors conclude that based on their results, specific fixed relative positions or kinematic features (tail beat phase locking) do not seem to be required for energetic savings. They also provide a review of potential different mechanisms that could play a role in the energetic savings.

Weaknesses:<br /> A weakness is the actual lack of critical discussion of the different mechanisms as well as the discussion on the conjecture that relative positions and kinematic features do not matter. I found the overall discussion on this rather unsatisfactory, lacking some critical reflections as well as different relevant statements or explanations being scattered across the discussion section. Here I would suggest a revision of the discussion section.

Also, there is a statement that Danio regularly move within the school and do not maintain inter-individual positions. However, there is no quantitative data shown supporting this statement, quantifying the time scales of neighbor switches. This should be addressed as core conclusions appear to rest on this statement and the authors have 3d tracks of the fish.

Further, there is a fundamental question on the comparison of schooling in a flow (like a stream or here flow channel) versus schooling in still water. While it is clear that from a pure physics point of view that the situation for individual fish is equivalent. as it is about maintaining a certain relative velocity to the fluid, I do think that it makes a huge qualitative difference from a biological point of view in the context of collective swimming. In a flow, individual fish have to align with the external flow to ensure that they remain stationary and do not fall back, which then leads to highly polarized schools. However, this high polarization is induced also for completely non-interacting fish. At high speeds, also the capability of individuals to control their relative position in the school is likely very restricted, simply by being forced to put most of their afford into maintaining a stationary position in the flow. This appears to me fundamentally different from schooling in still water, where the alignment (high polarization) has to come purely from social interactions. Here, relative positioning with respect to others is much more controlled by the movement decisions of individuals. Thus, I see clearly how this work is relevant for natural behavior in flows and that it provides some insights on the fundamental physiology, but I at least have some doubts about how far it extends actually to "voluntary" highly ordered schooling under still water conditions. Here, I would wish at least some more critical reflection and or explanation.

Related to this, the reported increase in the elongation of the school at a higher speed could have also different explanations. The authors speculate briefly it could be related to the optimal structure of the school, but it could be simply inter-individual performance differences, with slower individuals simply falling back with respect to faster ones. Did the authors test for certain fish being predominantly at the front or back? Did they test for individual swimming performance before testing them in groups together? Again this should be at least critically reflected somewhere.

Reviewer #2 (Public Review):

Summary:<br /> This paper tests the idea that schooling can provide an energetic advantage over solitary swimming. The present study measures oxygen consumption over a wide range of speeds, to determine the differences in aerobic and anaerobic cost of swimming, providing a potentially valuable addition to the literature related to the advantages of group living.

Strengths:<br /> The strength of this paper is related to providing direct measurements of the energetics (oxygen consumption) of fish while swimming in a group vs solitary. The energetic advantages of schooling have been claimed to be one of the major advantages of schooling and therefore a direct energetic assessment is a useful result.

Weaknesses:<br /> The manuscript suffers from a number of weaknesses which are summarised below:

1) The possibility that fish in a school show lower oxygen consumption may also be due to a calming effect. While the authors show that there is no difference at low speed, one cannot rule out that calming effects play a more important role at higher speed, i.e. in a more stressful situation.

2) The ratio of fish volume to water volume in the respirometer is much higher than that recommended by the methodological paper by Svendsen et al (J Fish Biol 2016)

3) Because the same swimming tunnel was used for schools and solitary fish, schooling fish may end up swimming closer to the wall (because of less volume per fish) than solitary fish. Distances to the wall of schooling fish are not given, and they could provide an advantage to schooling fish.

4) The statistical analysis has a number of problems. The values of MO2 of each school are the result of the oxygen consumption of each fish, and therefore the test is comparing 5 individuals (i.e. an individual is the statistical unit) vs 5 schools (a school made out of 8 fish is the statistical unit). Therefore the test is comparing two different statistical units. One can see from the graphs that schooling MO2 tends to have a smaller SD than solitary data. This may well be due to the fact that schooling data are based on 5 points (five schools) and each point is the result of the MO2 of five fish, thereby reducing the variability compared to solitary fish. Other issues are related to data (for example Tail beat frequency) not being independent in schooling fish.

Reviewer #3 (Public Review):

Summary:<br /> Zhang and Lauder characterized both aerobic and anaerobic metabolic energy contributions in schools and solitary fishes in the Giant danio (Devario aequipinnatus) over a wide range of water velocities. By using a highly sophisticated respirometer system, the authors measure the aerobic metabolisms by oxygen uptake rate and the non-aerobic oxygen cost as excess post-exercise oxygen consumption (EPOC). With these data, the authors model the bioenergetic cost of schools and solitary fishes. The authors found that fish schools have a J-shaped metabolism-speed curve, with reduced total energy expenditure per tail beat compared to solitary fish. Fish in schools also recovered from exercise faster than solitary fish. Finally, the authors conclude that these energetic savings may underlie the prevalence of coordinated group locomotion in fish.

The conclusions of this paper are mostly well supported by data, but some aspects of methods and data acquisition need to be clarified and extended.

Strengths:<br /> This work aims to understand whether animals moving through fluids (water in this case) exhibit highly coordinated group movement to reduce the cost of locomotion. By calculating the aerobic and anaerobic metabolic rates of school and solitary fishes, the authors provide direct energetic measurements that demonstrate the energy-saving benefits of coordinated group locomotion in fishes. The results of this paper show that fish schools save anaerobic energy and reduce the recovery time after peak swimming performance, suggesting that fishes can apport more energy to other fitness-related activities whether they move collectively through water.

Weaknesses:<br /> Although the paper does have strengths in principle, the weakness of the paper is the method section. There is too much irrelevant information in the methods that sometimes is hard to follow for a researcher unfamiliar with the research topic. In addition, it was hard to imagine the experimental (respirometer) system used by the authors in the experiments; therefore, it would be beneficial for the article to include a diagram/scheme of that respiratory system.

eLife assessment

This valuable study reports comprehensive multi-omic data on the changes induced in young and aged male mouse tail fibroblasts after treatment with chemical reprogramming factors. The authors claim that chemical reprogramming factors induce changes consistent with a reduction of cellular 'biological' age (e.g., correlations with established aging markers in whole tissues). However, the study relies on previously identified aging markers (instead of aging in the tail fibroblast system itself), and thus, at this stage, the evidence in support of the observed molecular changes truly reflecting changes in biological age in the study system is still incomplete.

Reviewer #1 (Public Review):

Summary:<br /> The investigators employed multi-omics approach to show the functional impact of partial chemical reprogramming in fibroblasts from young and aged mice.

Strengths:<br /> Multi-omics data was collected, including epigenome, transcriptome, proteome, phosphoproteome, and metabolome. Different analyses were conducted accordingly, including differential expression analysis, gene set enrichment analysis, transcriptomic and epigenetic clock-based analyses. The impact of partial chemical reprogramming on aging was supported by these multi-source results.

Weaknesses:<br /> More experimental data may be needed to further validate current findings.

Reviewer #2 (Public Review):

The short-term administration of reprogramming factors to partially reprogram cells has gained traction in recent years as a potential strategy to reverse aging in cells and organisms. Early studies used Yamanaka factors in transgenic mice to reverse aging phenotypes, but chemical cocktails could present a more feasible approach for in vivo delivery. In this study, Mitchell et al sought to determine the effects that short-term administration of chemical reprogramming cocktails have on biological age and function. To address this question, they treated young and old mouse fibroblasts with chemical reprogramming cocktails and performed transcriptome, proteome, metabolome, and DNA methylation profiling pre- and post-treatment. For each of these datasets, they identified changes associated with treatment, showing downregulation of some previously identified molecular signatures of aging in both young and old cells. From these data, the authors conclude that partial chemical reprogramming can rejuvenate both young and old fibroblasts.

The main strength of this study is the comprehensive profiling of cells pre- and post-treatment with the reprogramming cocktails, which will be a valuable resource for better understanding the molecular changes induced by chemical reprogramming. The authors highlighted consistent changes across the different datasets that are thought to be associated with aging phenotypes, showing reduction of age-associated signatures previously identified in various tissues. However, from the findings, it remains unclear which changes are functionally relevant in the specific fibroblast system being used. Specifically:

1) The 4 month and 20 month mouse fibroblasts are designated "young" vs "old" in this study. An important analysis that was not shown for each of the profiled modalities was a comparison of untreated young vs old fibroblasts to determine age-associated molecular changes in this specific model of aging. Then, rather than using aging signatures defined in other tissues, it would be more appropriate to determine whether the chemical cocktails reverted old fibroblasts to a younger state based on the age-associated changes identified in this comparison.<br /> 2) Across all datasets, it appears that the global profiles of young vs old mouse fibroblasts are fairly similar compared to treated fibroblasts, suggesting that the chemical cocktails are not reverting the fibroblasts to a younger state but instead driving them to a different cell state. Similarly, in most cases where specific age-related processes/genes are being compared across untreated and treated samples, no significant differences are observed between young and old fibroblasts.<br /> 3) Functional validation experiments to confirm that specific changes observed after partial reprogramming are indeed reducing biological age is limited.<br /> 4) Partial reprogramming appears to substantially reduce biological age of the young (4 month) fibroblasts based on the aging signatures used. It is unclear how this result should be interpreted.

eLife assessment

This valuable study investigates evolutionary aspects around a single amino acid polymorphism in an immune peptide of Drosophila melanogaster,. This polymorphism is known to be under long-term balancing selection. Using alleles with different substitutions, the investigators found that one allele provides better survival after systemic infections by a bacterial pathogen, but that the alternative allele endows its carriers with a longer lifespan under certain conditions. The authors suggest that these contrasting fitness effects of the two alleles contribute to balancing their long-term evolutionary fate. The strength of the provided evidence is still incomplete and would benefit from more rigorous approaches.

Reviewer #1 (Public Review):

Summary:<br /> In this manuscript, Unckless and colleagues address the issue of the maintenance of genetic diversity of the gene diptericin A, which encodes an antimicrobial peptide in the model organism Drosophila melanogaster.

Strengths:<br /> The data indicate that flies homozygous for the dptA S69 allele are better protected against some bacteria. By contrast, male flies homozygous for the R69 allele better resist starvation than flies homozygous for the S69 allele.

Weaknesses:<br /> -I am surprised by the inconsistency between the data presented in Fig. 1A and Fig. S2A for the survival of male flies after infection with P. rettgeri. I am not convinced that the data presented support the claim that females have lower survival rates than males when infected with P. rettgeri (lines 176-182).

-The data in Fig. 2 do not seem to support the claim that female flies with either the dptA S69 or the R69 alleles have a longer lifespan than males (lines 211-215). A comment on the [delta] dpt line, which is one of the CRISPR edited lines, would be welcome.

-The data in Fig. 2B show that male flies with the dptA S69 or R69 alleles have the same lifespan when poly-associated with L. plantarum and A. tropicalis, which contradicts the claim of the authors (lines 256-260).

Reviewer #2 (Public Review):

Summary: In this study, the authors delve into the mechanisms responsible for the maintenance of two diptericin alleles within Drosophila populations. Diptericin is a significant antimicrobial peptide that plays a dual role in fly defense against systemic bacterial infections and in shaping the gut bacterial community, contributing to gut homeostasis.

Strengths: The study unquestionably demonstrates the distinct functions of these two diptericin alleles in responding to systemic infections caused by specific bacteria and in regulating gut homeostasis and fly physiology. Notably, these effects vary between male and female flies.

Weaknesses: Although the findings are highly intriguing and shed light on crucial mechanisms contributing to the preservation of both diptericin alleles in fly populations, a more comprehensive investigation is warranted to dissect the selection mechanisms at play, particularly concerning diptericin's roles in systemic infection and gut homeostasis. Unfortunately, the results from the association study conducted on wild-caught flies lack conclusive evidence.

Major Concerns:

Lines 120-134: The second hypothesis is not adequately defined or articulated. Please revise it to provide more clarity. Additionally, it should be explicitly stated that the first part of the first hypothesis (pathogen specificity), i.e., the superior survival of the S allele in Providencia infections compared to the R allele, has been previously investigated and supported by the results in the Unkless et al. 2016 paper. The current study aims to additionally investigate the opposite scenario: whether the R allele exhibits better survival in a different infection. Please consider revising to emphasize this point.

Figures and statistical analyses: It is essential to present the results of significant differences from the statistical analyses within Figures 1B, 2B, and 3. Additionally, please include detailed descriptions of the statistical analysis methods in the figure legends. Specify whether the error bars represent standard error or standard deviation, particularly in Figure 3, where assays were conducted with as few as 3 flies.

Lines 317-318 (as well as 320-328): The data related to P. rettgeri appear somewhat incomplete, and the authors acknowledge that bacterial load varies significantly, and this bacterium establishes poorly in the gut. These data may introduce more noise than clarity to the study. Please consider revising these sections by either providing more data, refining the presentation, or possibly removing them altogether.

Lines 335-387 and Figure 4: Although these results are intriguing and suggest interactions between functional diptericin and fly physiology, some mediated by the gut microbiome, they remain descriptive and do not significantly contribute to our understanding of the mechanism that maintains the diptericin alleles.

Lines 399-400: The contrast between this result and statement and the highly reproducible data presented in Figures 2-4 should be discussed.

Lines 422-429 and Figure 5D: The conclusion regarding an association between diptericin alleles and Morganellaceae bacteria is not clearly supported by Figure 5D and lacks statistical evidence.